Citation
Multi-Level regulation of phenylalanine hydroxylase in Pseudomonas aeruginosa

Material Information

Title:
Multi-Level regulation of phenylalanine hydroxylase in Pseudomonas aeruginosa
Creator:
Song, Jian, 1963-
Publication Date:
Language:
English
Physical Description:
viii, 98 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Amino acids ( jstor )
Boxes ( jstor )
DNA ( jstor )
Enzymes ( jstor )
Gels ( jstor )
Operator regions ( jstor )
Operon ( jstor )
Plasmids ( jstor )
Promoter regions ( jstor )
Pseudomonas aeruginosa ( jstor )
Bacterial genetics ( lcsh )
Dissertations, Academic -- Microbiology and Cell Science -- UF ( lcsh )
Microbiology and Cell Science thesis, Ph. D ( lcsh )
Phenylalanine ( lcsh )
Pseudomonas aeruginosa ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 89-96).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Jian Song.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
027621514 ( ALEPH )
37231976 ( OCLC )

Downloads

This item has the following downloads:


Full Text













MULTI-LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa




















By

JIAN SONG



















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1997




































Copyright 1997

by

Jian Song
































Dedicated to my father and my mother,

whose love, care, and encouragement make it possible

for me to complete this dissertation















ACKNOWLEDGMENTS


I wish to express my deep and sincere gratitude to Dr.

Roy A. Jensen, chairman of my supervisory committee, whose

invaluable guidance, constant encouragement, endless ideas,

critical input, and financial support made the fulfillment of

this study possible.

I would also like to thank Dr. Dean W. Gabriel, Dr.

Lonnie 0. Ingram, Dr. James F. Preston, and Dr. Keelnatham T.

Shanmugam for their help, encouragement, advice, and critical

review of the dissertation.

My special thanks are also extended to Dr. Carol Bonner,

Dr. Tianhui Xia, and Wei Gu for their great help in all

aspects of my study, particularly helping me get started

during my first year in the lab.

I am also very thankful to Dr. Randy Fischer, Dr. Prem

Subramaniam, and Gary Xie for their help during this study.

I am indebted to my family, especially to my parents, to

whom this dissertation is dedicated. Without their love,

support, and encouragement, this study could not have been

accomplished. I am also indebted to my brother and sister-in-

law for helping me in taking care of my parents.

Finally, but not least, I wish to express my sincere

appreciation to my wife, Tao Sun, for her love, support,

iv













patience, and encouragement during these years of study, and

to my son, Peter, and my daughter, Kerry, for filling the

family with great joy and happiness.


















































v
















TABLE OF CONTENTS



ACKNOWLEDGMENTS. ............... .......................... iv

ABSTRACT ................................................. vii

CHAPTERS

1 LITERATURE REVIEW ................................. 1

Phenylalanine Hydroxylase in Nature ............... 1
The Pterin-Recycling Enzymes ...................... 4
Regulation of Phenylalanine Hydroxylase ........... 8

2 PhhR, A DIVERGENETLY TRANSCRIBED ACTIVATOR OF
PHENYLALANINE HYDROXYLASE GENE CLUSTER
OF Pseudomonas aeruginosa.......................... 11

Introduction ...................................... 11
Materials and Methods ............................. 13
Results . .......................................... 23
Discussion. ....................................... 42

3 BIFUNCTIONAL PhhB REGULATES THE EXPRESSION OF
PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa.......................... 53

Introduction ...................................... 53
Materials and Methods............................... 55
Results ........................................... 64
Discussion. ....................................... 84

REFERENCES ............... ............................... 89

BIOGRAPHICAL SKETCH. ..................................... 97











vi















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

MULTI-LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa

By

Jian Song

May, 1997


Chairperson: Roy A. Jensen
Major Department: Microbiology and Cell Science

Pseudomonas aeruginosa was recently found to possess a

cluster of genes (phh operon) encoding phenylalanine

hydroxylase (PhhA), 4a-carbinolamine dehydratase (PhhB), and

aromatic aminotransferase (PhhC). In the flanking upstream

region of the phh operon, a divergently transcribed gene

(phhR) that encodes an activator protein was identified.

Inactivation of phhR markedly reduced expression of the three

structural genes. PhhR belongs to the large prokaryote family

of a54 enhancer-binding proteins, and activation of the phh

operon by PhhR in P. aeruginosa required rpoN. P. aeruginosa

PhhR was able to replace E. coli TyrR as a repressor of the

aroF-tyrA operon (but not as an activator of mtr) in the

heterologous E. coli system. The phh operon was strongly

induced in fructose- or glucose-based minimal medium by L-

phenylalanine and L-tyrosine, and less by L-tryptophan.

vii









Inactivation of phhR in P. aeruginosa abolished ability to

utilize either L-phenylalanine and L-tyrosine as a sole source

of carbon for growth.

PhhB is a bifunctional protein. It was shown to have 4a-

carbinolamine dehydratase activity as well as regulatory

activity. The expression of phhA was activated by the presence

of phhB in both E. coli and P. aeruginosa. Transcriptional

and translational fusion analysis showed that the regulatory

effect of PhhB on the expression of phhA is at the post-

transcriptional level.

An insertionally inactivated phhB mutant failed to grow

on L-phenylalanine or L-tyrosine as a sole carbon source.

Expression of PhhA in the absence of PhhB causes strong growth

inhibition in E. coli. The inhibitory effect is probably

caused by 7-tetrahydrobiopterin, which is known to be formed

in the absence of PhhB. Since 7-tetrahydrobiopterin is a

potent inhibitor of phenylalanine hydroxylase, this could

account for the inability of phhA in the absence of phhB to

complement E. coli tyrosine auxotrophy. The general inhibition

of growth may be due to inhibition of some unidentified

essential pterin-dependent enzymes.












viii















CHAPTER 1
LITERATURE REVIEW


Phenylalanine Hydroxylase in Nature


Phenylalanine hydroxylase (phenylalanine hydroxylase 4-

monooxygenase; EC 1.14.16.1) catalyzes the irreversible

conversion of L-phenylalanine to L-tyrosine (Kaufman, 1987).

In mammals this enzyme catalyzes the initial, obligatory, and

rate-limiting step in the complete catabolism of serum

phenylalanine to CO2 and H20 (Kaufman, 1986) . A deficiency of

this enzyme causes accumulation of serum phenylalanine,

leading to hyperphenylalanemia. Because metabolism of

phenylalanine is restricted to alterations in the alanyl side

chain of phenylalanine, in the absence of phenylalanine

hydroxylase, the formation and excretion in the urine of

compounds such as phenylpyruvate and phenyllactate occurs.

This condition is called phenylketonuria, a genetic disorder

associated with severe mental retardation in untreated

children (Dilella et al., 1986). Many mutations at the

phenylalanine hydroxylase locus have been identified (Guldberg

et al., 1996).

Phenylalanine hydroxylase has been intensively studied in

mammals for many years. It is a member of a family of enzymes

that also includes tryptophan hydroxylases (EC 1.14.16.4) and

1









2

tyrosine hydroxylases (EC 1.14.16.2). All three enzymes

utilize a tetrahydrobiopterin cofactor and molecular oxygen to

hydroxylate their respective aromatic amino acid substrates

(Kaufman and Fisher, 1974). Phenylalanine hydroxylase has

been purified from rat liver where it is an oligomeric protein

(predominantly homotetramers) composed of 52-kDa subunits

(Davis et al., 1996). It has non-heme iron as the active-site

metal. The rat liver hydroxylase was also expressed in E.

coli and purified to homogeneity (Kappock et al., 1995). The

homotetrameric recombinant rat hepatic phenylalanine

hydroxylase is highly active and is identical to the native

enzyme in many properties.

Although mammalian phenylalanine hydroxylase has been

intensively studied, few studies on bacterial phenylalanine

hydroxylase have been done. Phenylalanine hydroxylase has

generally been considered to be of rare occurrence in

prokaryotes, where scattered reports of its existence have

been limited to one phylogenetic division of gram-negative

bacteria. They include Pseudomonas acidovorans (previously

known as Pseudomonas sp. ATCC 11299a) (Guroff & Ito, 1963).

P. facilis (Decicco & Umbreit, 1964), Alcaligenes eutrophus

(Friedrich & Schlegel, 1972), and Chromobacterium violaceum

(Letendre et al., 1974). Of the three pterin-dependent and

metal-containing hydroxylases, only phenylalanine hydroxylase

from Pseudomonas acidovorans (Letendre et al, 1975) and C.

violaceum (Nakata et al., 1979; Pember et al., 1986) has been









3

purified and characterized. The C. violaceum phenylalanine

hydroxylase gene was the first one to be cloned and sequenced

from a bacterium (Onishi et al., 1991). High identity of the

deduced amino acid sequence with those deduced for the

mammalian hydroxylase gene family was found and showed that

the microbial hydroxylase and the mammalian hydroxylases are

homologous. Although C. violaceum phenylalanine hydroxylase is

a pterin-dependent enzyme, it differs from the mammalian

enzymes in its smaller subunit size (lacking the N-terminal

domain responsible for the complex regulation in the mammalian

enzymes), its existence as a monomer (rather than a

homotetramer), and binding of copper (instead of iron) at its

active site. However, the surprising claim has been advanced

that C. violaceum phenylalanine hydroxylase does not require

any redox active metal for its activity (Carr & Benkovic,

1993; Carr et al., 1995).

P. aeruginosa belongs to a different superfamily of gram-

negative prokaryotes than do the aforementioned organisms. It

was found to possess homologues of mammalian phenylalanine

hydroxylase, 4a-carbinolamine dehydratase/DCoH, and aromatic

aminotransferase as part of a three-component gene cluster

(Zhao et al., 1994). These three genes are phhA, phhB, and

phhC, respectively. The P. aeruginosa phenylalanine

hydroxylase contains iron and is pterin-dependent. Unlike the

multimeric mammalian hydroxylase, the native P. aeruginosa

hydroxylase is a monomer.









4

The Pterin-Recycling Enzymes


Phenylalanine hydroxylase catalyzes the conversion of L-

phenylalanine to L-tyrosine, using tetrahydrobiopterin as a

reducing agent and relying upon molecular oxygen as an

oxidizing agent (Kaufman, 1987). During this hydroxylation

reaction, the tetrahydrobiopterin cofactor is

stoichiometrically oxidized to a carbinolamine, 4a-

hydroxytetrahydrobiopterin. Two essential enzymes, 4a-

carbinolamine dehydratase and dihydropteridine reductase, are

involved in regenerating the pterin cofactor in two steps.

4a-Hydroxytetrabiopterin is first converted by 4a-

carbinolamine dehydratase to quinonoid dihydrobiopterin, and

the latter compound is then reduced back to

tetrahydrobiopterin by NADH-dependent dihydropteridine

reductase (Kaufman, 1987).


4a-Carbinolamine Dehydratase/DCoH


4a-Carbinolamine dehydratase was first purified from rat

liver as a fraction called "phenylalanine hydroxylase

stimulator", which could stimulate the hydroxylation reaction

at pH 8.2 to 8.4 (Huang et al., 1973). It was later found to

be an enzyme that catalyzes the conversion of 4a-

hydroxytetrabiopterin to the quinonoid dihydropterin (Lazarus

et al., 1983). 4a-Hydroxytetrabiopterin is also known to be

unstable, breaking down nonenzymatically to the corresponding

quinonoid dihydropterin (Kaufman, 1975). However, in the









5

absence of 4a-carbinolamine dehydratase, the dehydration of

the 4a-carbinolamine becomes rate-limiting for the

hydroxylation of phenylalanine. The consequent accumulation

of 4a-carbinolamine results in a small percentage of

rearrangement to the 7-tetrahydrobiopterin isomer (Curtius et

al., 1990). The latter 7-isomer was shown to be a potent

inhibitor of the phenylalanine hydroxylase (Davis et al.,

1992). Under conditions where 4a-carbinolamine and the 7-

isomer are generated, the addition of 4a-carbinolamine

dehydratase markedly inhibits the rate of formation of the 7-

isomer by diverting a greater fraction of the 4a-carbinolamine

to the quinonoid dihydropterin (Davis et al., 1991). Thus, the

dehydratase not only directly catalyzes the dehydration of the

carbinolamine, but also indirectly prevents isomerization to

the inhibitory 7-isomer (Kaufman et al., 1993).

4a-Carbinolamine dehydratase from rat liver has been

cloned and sequenced (Citron et al., 1992). It then became

apparent that this dehydratase is identical to DCoH, a protein

that facilitates the dimerization of hepatic nuclear factor 1

alpha (HNF-la), a homeodomain transcription factor. DCoH was

found to display a restricted tissue distribution and did not

bind directly to DNA. The formation of a stable tetrameric

DCoH-HNF-la complex does not change the DNA-binding

characteristics of HNF-1a, but does enhance the

transcriptional activity of HNF-1 (Mendel et al., 1991). X-

ray crystallography has revealed DCoH to form a tetramer









6

containing two saddle-shaped grooves that comprise likely

macromolecular binding sites (Endrizzi et al., 1995).

Structural similarities between the DCoH and nucleic acid-

binding proteins imply that the saddle motif has evolved to

bind diverse ligands or that DCoH may bind nucleic acid

according to Endrizzi et al. (1995).

DCoH homologues have been identified in Xenopus (XDCoH)

(Pogge-yon-Strandmann & Ryffel, 1995) and P. aeruginosa (PhhB)

(Zhao et al., 1994) . XDCoH was found to be a maternal factor.

The amount of XDCoH increases dramatically following

neurulation, when the formation of liver, pronephros, and

other organs takes place. The tissue distribution of XDCoH

during embryogenesis suggests that XDCoH is involved in

determination and differentiation of various unrelated cell

types. The interaction with XDCoH was found to be essential

for the function of several tissue-specific transcription

factors (Pogge-yon-Strandman & Ryffel, 1995) . In P. aeruginosa

expression of phhA (encoding phenylalanine hydroxylase) was

reported to require phhB (encoding 4a-carbinolamine

dehydratase), suggesting that PhhB may have a positive

regulatory role. If so, this would be an intriguing parallel

with the dual catalytic and regulatory roles of the

corresponding mammalian homolog (Zhao et al., 1994).









7

Dihydropteridine Reductase


Dihydropteridine reductase (DHPR; EC 1.6.99.7) is one of

the two essential enzymes involved in recycling the pterin

cofactor for aromatic amino acid hydroxylases. It catalyzes

the reduction of quinonoid dihydropterin to

tetrahydrobiopterin, using NADH as a cofactor. DHPR is an

ubiquitous enzyme in animals, being found in all tissues that

contain the aromatic amino acid hydroxylases (Armarego et al.,

1984). Close correlation between levels of 4a-carbinolamine

dehydratase and dihydropterine reductase in liver during human

fetal development strongly suggests a physiologically

significant role for both enzymes in tetrahydrobiopterin

regeneration. Genetic defects in DHPR cause malignant

phenylketonuria. A concomitant deficiency of

neurotransmitters such as 3,4-dihydroxyphenylalanine (DOPA)

and 5-hydroxytryptophan reflects the essential coupling of

DHPR to tyrosine hydroxylase and tryptophan hydroxylase as

well (Gudinchet et al., 1992).

DHPR is also found in bacteria. DHPR has been purified

from Pseudomonas acidovorans (Williams et al., 1976) and E.

coli (Vasudevan et al., 1988). In P. acidovorans, both DHPR

and phenylalanine hydroxylase activities were found to be

higher in cells adapted to a medium containing L-phenylalanine

or L-tyrosine as the sole carbon source than in those grown in

L-asparagine (Williams et al., 1976). Interestingly, DHPR has

also been found in E. coli even though no aromatic amino acid









8

hydroxylases or 4a-carbinolamine dehydratase have ever been

detected (Vasudevan et al., 1988). Unlike other

dihydropteridine reductases that have been studied, the E.

coli DHPR possesses an FAD prosthetic group, and has

dihydrofolate reductase and pterin-independent oxidoreductase

activities (Vasudevan et al., 1992).


Regulation of Phenylalanine Hydroxylase


Phenylalanine hydroxylase in mammals is tightly regulated

at different levels. At the protein level, it is

allosterically regulated by phenylalanine (Kaufman, 1987).

The activity of phenylalanine hydroxylase increases at least

20-fold after incubation with phenylalanine (Tourian, 1971).

It is also activated through phosphorylation by a cAMP-

dependent kinase both in vivo and in vitro (Abita et al,

1976). At the DNA level, expression of the phenylalanine

hydroxylase gene in liver and kidney tissues of mice is

enhanced at birth and is induced by glucocorticoids and cAMP

in liver (Faust et al., 1996). Regulatory elements including

a tissue-specific and hormone-inducible enhancer in the

upstream region have been characterized. The enhancer region

contains separate protein-binding sites for the glucocorticoid

receptor and the hepatocyte-enriched transcription factor,

hepatocyte nuclear factor 1 (HNF1) (Faust et al., 1996). HNF1

is a transcriptional activator of many hepatic genes including

albumin, c-antitrypsin, and u- or 3-fibrinogen. Mice lacking









9

HNF1 die with a marked liver enlargement. The gene coding for

phenylalanine hydroxylase is totally silent, thus giving rise

to phenylketonuria (Pontoglio et al., 1996).

Little information is available about the regulation of

phenylalanine hydroxylase in bacteria. However, some evidence

has indicated that the bacterial phenylalanine hydroxylase is

also regulated. In Pseudomonas acidovorans, a higher level of

phenylalanine hydroxylase was found after growth in the

presence of phenylalanine (Willams et al, 1976). Induction of

both phenylalanine hydroxylase and tryptophan hydroxylase in

the presence of their substrates was also reported in C.

violaceum (Letendre et al, 1974).

The most extensively characterized microbial

phenylalanine system is that of P. aeruginosa. Whether this

system is subject to any regulatory controls has not been

studied prior to this work. The initial report of Zhao et al.

(1994) provided a strong basis for anticipation that the phh

operon would be subject to regulation for the following

reasons. (i) The closely spaced organization of the three

structural genes (phhABC) in an apparent operon implies

regulation. (ii) Analysis of effects of the presence or

absence of regions immediately flanking the phh operon upon

expression of phenylalanine hydroxylase indicated the likely

location there of one or more regulatory genes. (iii) The

reported lack of phhA expression in the absence of phhB









10

suggested a positive regulatory role of phhB in addition to

its catalytic function.

The major objectives of this study have been to elucidate

the physiological conditions under which regulation occurs, to

identify and characterize at the molecular-genetic level any

regulatory genes which control the phh operon, and to

determine the nature of the apparent positive regulatory

action of phhB.















CHAPTER 2
PHHR, A DIVERGENTLY TRANSCRIBED ACTIVATOR OF
THE PHENYLALANINE HYDROXYLASE GENE CLUSTER
OF Pseudomonas aeruginosa


Introduction


A recent report (Zhao et al., 1994) revealed that

Pseudomonas aeruginosa possesses a tetrahydrobiopterin (BH4) -

dependent monooxygenase that is capable of catalyzing the

phenylalanine hydroxylase reaction. It is encoded by the

proximal member (phhA) of a three-gene cluster. The second

gene, phhB, encodes carbinolamine dehydratase, a key enzyme

within the cycle regenerating BH4. phhC encodes an aromatic

aminotransferase and belongs to the Family-I aminotransferases

(Gu and Jensen, 1996) . The reactions, as they are known to

function for the mammalian homologs in the catabolism of L-

phenylalanine, are shown in Fig. 2-1.

The physiological function of phenylalanine hydroxylase

in P. aeruginosa has not been obvious. A primary role in L-

tyrosine biosynthesis seems unlikely because of the

established presence for this purpose of a cyclohexadienyl

dehydrogenase that is widely distributed in gram-negative

bacteria and which is highly sensitive to feedback inhibition

by L-tyrosine (Xia and Jensen, 1990). Although function as an

initial step of L-phenylalanine catabolism has precedent in

11








12







14-Hydroxyphenylpyruvate


[PhhC]


Tyrosine

4a-Carbinolamine [PhhBl

[PhhA] o02 Dihydropteridine
[DHPR
TetrahydrobiopterinD-HP

Phenylalanine NAD+ NADH+H+



FIG. 2-1. Initial reactions of phenylalanine catabolism in
mammals. The three structural genes of the phh operon encode
enzymes catalyzing three of the four steps shown. The
abbreviations: PhhA, phenylalanine hydroxylase; PhhB, 4a-
carbinolamine dehydratase; PhhC, aromatic aminotransferase;
DHPR, dihydropteridine reductase. 4a-Carbinolamine is an
alternative designation for 4a-hydroxytetrahydrobiopterin.









13

mammalian metabolism, the literature encompassing the widely

studied catabolism of aromatic compounds in pseudomonad

bacteria (indeed, in prokaryotes) does not include the

phenylalanine hydroxylase step. Furthermore, L-phenylalanine

(substrate of PhhA) is an extremely poor source of carbon for

growth of P. aeruginosa, whereas L-tyrosine (product of PhhA)

is an excellent carbon source.

Zhao et al. (1994) had previously noted that subclones

lacking the flanking regions around the phh operon possessed

20-fold greater activity for phenylalanine hydroxylase. This

suggested the presence of a regulatory gene. Since an

understanding of the regulation governing the phh operon

should provide important physiological clues about function,

I have analyzed the flanking regions and now report the

characteristics of a regulatory gene, denoted phhR.


Materials and Methods


Materials


The bacterial strains and plasmids used in this study are

listed in Table 1. The LB and M9 formulations (Sambrook et

al., 1989) were used as growth media for E. coli and P.

aeruginosa. Pseudomonas isolation agar (Difco) was used for

isolating Pseudomonas "knockout" mutants. Additions of

ampicillin (100 Ag/ml), chloramphenicol (40 ig/ml), kanamycin

(50 tg/ml), tetracycline (25 ig/ml), mercuric chloride (15

pig/ml), L-phenylalanine (50 Ag/ml), and thiamine (17 Ag/ml)









14

TABLE 2-1. Bacterial strains and plasmids

Strain or Relevant genotype or Source or
plasmid description reference

E. coli

BL21(DE3) F- ompT hsdSB (rB-mB-) gal dcm; Novagen
with DE3, a X prophage carrying
the T7 RNA polymerase gene

BW545 A(lacU)169 rpsL Rosentel et al.

DH5a F-AlacU169 q80dlacZAM15 hsdRl7 GIBCO/BRL
recAl endAl gyrA96 thi-1 relAl
supE44

LE392 F-el4- (McrA-) hsdR514 (rk-mk�) Sambrook et al.
supE44 supF58 lacYl or d (lacIZY)6
galK galT22 metBl trpR55

JP2255 aroF363 pheA361 phe0352 tyrA382 Baldwin &
thi-1 strR712 lacYl xyl-15 Davidson

S17-1 [RP4-2(Tc:Mu) (Km:Tn7)Tra(incP)] Simon et al.
pro hsdR recA Tpr Smr

SP1312 zah-735:TnlO A(argF-lac)U169 Heatwole &
Somerville
SP1312 SP1312 0(mtr'-lacZ�) Heatwole &
(XSLW20) Somerville

SP1313 SP1312, A(tyrR) Heatwole &
Somerville
SP1313 SP1313 0(mtr'-lacZ�) Heatwole &
(XSLW20) Somerville,


P. aeruginosa

PA103 Prototroph Totten et al.

PA103NG rpoN Totten et al.

PAO-1 Prototroph Holloway

JSI01 PAO-1 phhA, Hgr This study

JS102 PAO-1 phhR, Hgr This study









15

Table 2-1. (continued)


Plasmids

pUC18 Ampr lac'IPOZ' Yanisch-
Perron et al.

pUC19 Ampr lac'IPOZ' Yanisch-
perron et al.
pACYC184 P15A replicon, Cmr TCr Chang & Cohen

pETllb T71ac promoter, lacI� Apr Novagen

pRS1274 lacZY fusion vector Simons et al.

Z1918 Promoterless lacZ, Apr Schweizer

pJZ9 phhRABC, Apr Zhao et al.

pJZ9-3a phhAB,Apr Zhao et al.

pJS7 phhRABC, Apr This study

pJS60 phhABC, Apr This study

pJS61Z phhRA'-lacZ transcriptional This study
fusion, Apr

pJS62Z phhA'-lacZ transcriptional This study
fusion, Apr

pJS88 pETllb carrying phhR translational This study
fusion at the ATG start site

pJS91 pACYC184 carrying phhR , Cmr This study

pJS102 pRS1274 carrying phhR'-lacZY This study
transcriptional fusion

pCRII Ampr Kanr lacZa Invitrogen

pDG106 Hgr Kmr P15A replicon Gambill &
Summers
pJSl01 PstI-SmaI fragment of pDG106 This study
inserted into pUC18

pUFR004 ColEl Cmr Mob' mob(P) DeFeyter et al.









16

were made as appropriate. Agar was added at 20 g/liter for

preparation of solid medium. Restriction enzymes, T4 DNA

ligase, DNA-modifying enzymes (New England Biolab or Promaga)

and Taq DNA polymerase (Perkin-Elmer) were used as recommended

by the suppliers. Other biochemicals were purchased from Sigma

Chemical Co. Inorganic chemicals (analytical grade) were from

Fisher Scientific.


Phenylalanine Hydroxylase Assay


Cultures of E. coli JP2255 carrying the various plasmids

specified were grown in 500 ml of LB broth supplemented with

ampicillin (100 Aug/ml) at 370C and harvested at late

exponential phase of growth. The cell pellets were

resuspended in 10 ml of 10 mM potassium phosphate buffer (pH

7.4) containing 1 mM dithiothreitol and were disrupted by

sonication for 30 s at 40C using a Lab-Line Ultratip Labsonic

System (Lab-Line Instruments, Inc., Melrose Park, IL) . The

resulting extracts were centrifuged at 150,000 x g for 1 hr at

40C. The supernatant (crude extract) was desalted using

Sephadex G-25 and used for enzyme assay. PhhA was assayed by

following tyrosine formation (Nakata et al., 1979).


Recombinant DNA Techniques


Molecular cloning and DNA manipulation, including plasmid

purification, restriction enzyme digestion, ligation, and

transformation were conducted by standard methods (Sambrook et









17

al., 1989). DNA fragments were purified from agarose gel with

a "Geneclean" kit (BiolOl) . Electroporation (Invitrogen) was

used for simultaneous transformation of E. coli with two

compatible plasmids.



Construction of PhhR Expression Vectors


For expression of PhhR protein in E. coli, the T7

expression system (Novagen) was employed. The phhR coding

region was cloned into a translational fusion vector pETllb.

Polymerase chain reaction (PCR) was used to amplify the phhR

gene. The upper primer (5' -ATACATATGCGTATCAAAGTGCACTGC-3')

was made with a built-in NdeI restriction site (underlined)

which allows fusion of phhR at the translational start site

(ATG in bold). The lower primer (5'-CCTCCACCGTTTCTTTCCCAGCCT-

3') was chosen at a position 48 bases downstream of the

translational stop codon. PhhR protein made from this PCR

fragment was designed to be a native protein, not a fusion

protein. The PCR fragment was cloned into a TA cloning

vector, pCRII. The phhR gene was excised from pCRII with NdeI

and EcoRI. The NdeI-EcoRI fragment was first ligated with

EcoRI-BamHI adaptor to create a NdeI-BamHI fragment which was

then ligated with pETllb digested with NdeI and BamHI to

create the PhhR expression plasmid, pJS88 (Fig.2-3A).

For construction of a PhhR constitutive expression

plasmid, pACYC184 was chosen as the expression vector. The

pACYC184 vector has a P15A origin of replication which is









18

compatible with most commonly used plasmids using a ColEI

origin of replication, and it has low-copy number (about 20

copies/cell) . High level of PhhR produced from a high-copy

number plasmid was found to be toxic to the host cells. The

BglII-BamHI fragment carrying phhR gene was excised from the

expression plasmid pJS88 and cloned into the BamHI site of

pACYC184, thereby interrupting the tetracycline resistance

gene (Tcr) (Fig.2-3A).


Evaluation of Sensitivity/Resistance to m-Fluoro-tyrosine


Three E. coli strains, SP1312 (tyrR') , SP1313 (tyrR-)

carrying pJS91 (phhR'), and SP1313 (tyrR-) carrying pACYC184

(phhR-), were compared for sensitivity to m-fluoro-tyrosine

(MFT). All three strains were first grown in M9 medium with

appropriate antibiotics up to late-exponential phase of growth

and then used to swab M9 agar plates containing appropriate

antibiotics. A sterile Difco concentration disk (0.6 cm) was

positioned at the center of each plate, and 10 il of 50 jig/ml

m-fluoro-tyrosine was applied onto the disks. The plates were

then incubated at 370C for 24 hours.


Construction of phhA'-lacZ and phhR'-lacZ Transcriptional
Fusions


To compare levels of phhA transcription in both pJZ9 and

pJZ9-3a, plasmids pJS61Z and pJS62Z were constructed,

respectively. These have a promoterless lacZ gene (from









19

plasmid Z1918) fused at the BamH I site within phhA to form

phhA'-lacZ transcriptional fusions. Plasmid pJS61Z has the

same upstream sequence as plasmid pJZ9, and plasmid pJS62Z has

the same upstream sequence as pJZ9-3a. Hence, the phhA'-lacZ

fusions in pJS61Z and pJS62Z should represent the phhA

transcriptional levels in pJZ9 and pJZ9-3a, respectively.

To study regulation of the phhR promoter, the HincII-

BamHI fragment (phhR') was cloned into the pRS1274 lacZY

transcriptional fusion vector at the BamHI-SmaI site to create

pJS102(phhR'-lacZ) .


3-Galactosidase Assay


O-Galactosidase activity was assayed under conditions of

proportionality as described by Miller (1972), and specific

activities are expressed in Miller units. The data are the

results of at least two independent assays.


Gene Inactivation


P. aeruginosa is well known for its relatively high

resistance to most antibiotics, which complicates attempts to

use most of available antibiotic-resistance genes as selective

markers for gene replacement. Mercury resistance (Hgr) was

used as a selective marker since P. aeruginosa has been shown

to be sensitive to mercury (Essar et al., 1990; Gambill and

Summers, 1985). Insertional inactivation technique described

by Sophien et al. (1992) utilizes a mobilizable suicide vector









20

containing a truncated gene fragment (at both 5' and 3' ends)

and Hgr-cassette, and this suicide plasmid was integrated into

the chromosome by a single homologous recombination event. PCR

was used to generate truncated fragments. To generate a 'phhR'

(601bp) fragment, the upper primer 5'-

CCGTGTAGGCATCCTCCGCGACAT-3', and the lower primer 5'-

CTGGAAGATACTGTCGAAGCCACG-3' were used; to generate the 'phhA'

(639bp) fragment, we used the upper primer 5'-

ACGACAACGGTTTCATCCACTATC-3' and the lower primer 5'-

GGACGAAATAGAGCGGTTGCAGGA-3'. The PCR-generated fragments were

cloned into pCRII (a TA cloning vector) and then excised with

EcoRI. The EcoRI fragments were subsequently cloned into the

EcoRI site of pUFR004 (a mobilizable suicide vector) to create

pUFR/'phhA', and Hgr HindIII-cassette from pJS101 was inserted

into the HindIII site of pUFR/'phhA' to create

pUFR/'phhA'/Hgr. pUFR/'phhR'/Hgr was created in a similar

fashion. These plasmids were then used to transform E. coli

strain S17-1 (a mobilizing strain). Strain S17-1 harboring

either pUFR/'phhA' /Hgr or pUFR/'phhR' /Hgr was used as the donor

in biparental mating with P. aeruginosa performed as described

by Simon et al. (1983). Donor and recipient cells were grown

in LB broth to an OD600 of about 1.0 (E. coli S17-1 at 370C and

P. aeruginosa PAO-1 at 420C), mixed (0.5 ml volume of each) in

a 1.5-ml microcentrifuge tube, and pelleted by centrifugation.

The mating mixture was carefully resuspended in 0.2 ml of LB

broth and spread onto a sterile nitrocellulose filter (0.45-Am









21

pore size) resting on a prewarmed LB agar plate. The plates

were incubated for 16-24 hours at 370C, and then cells were

removed from the filter by an inoculation loop and resuspended

by vortexing into 0.5 ml of LB broth. Aliquots of 10-, 20-,

50-, and 100-A1 volume of the cell suspension were spread onto

Pseudomonas isolation agar plates containing 15 pg of HgCl2.

The plates were incubated overnight and Hgr colonies were

isolated.


Preparation of PhhA-specific Polyclonal Antiserum


PhhA was partially purified by anion-exchange and gel-

filtration chromatography following the methods described by

Zhao et al., (1994). The partially purified PhhA was subject

to SDS-PAGE (12%) and the gel was stained with Commassie blue

R-250. The PhhA band was cut from the gel and used for the

production of polyclonal antiserum in rabbits (Cocalico

Biologicals, Inc., Reamstown, PA). Antiserum was purified by

using an Econo-Pac protein A column (Bio Rad) and further

absorbed with a total cell extract from the PhhA-deficient

mutant JS101.


SDS-PAGE and Western Blot Analysis


SDS-PAGE (12%) was performed with the Mini-PROTEAN II

Cell (Bio-Rad) by the method of Laemmli (1970). Samples of

exponential-phase cells were collected by centrifugation, and

the cell pellets were suspended in gel-loading buffer and









22

heated at 1000C for 10 min. Samples of 5-10 Jil were loaded

onto two SDS-acrylamide gels. After separation of the proteins

by electrophoresis, one gel was stained with Coomassie blue R-

250 and the other gel was used for blotting. When crude

extracts were used, equivalent amounts of protein were loaded

in each lane. Western blots were performed according to

Towbin et al. (1979). The proteins were eletrophoretically

transferred onto nitrocellulose membranes and reacted with

polyclonal antibodies raised against PhhA in a rabbit.


N-Terminal Amino Acid Sequencing


PhhR protein produced in E. coli BL21(DE3)/pJS88

following induction by 1 mM IPTG for 3 hours was first

separated from the whole lysate by SDS-PAGE. The proteins

were then blotted to a polyvinylidene difluoride membrane

(Bio-Rad) and were stained with Coomassie brilliant blue R-250

(Sigma). The band corresponding to PhhR was excised from the

membrane and used for sequencing by using an Applied

Biosystems model 407A protein sequencer with an on-line 120A

phenylthiohydantoin analyzer in the Protein Core Facility of

the ICBR at the University of Florida.


DNA Sequencing and Data Analysis


Sequencing of phhR region was performed by the DNA

Sequencing Core Laboratory of the University of Florida. The

nucleotide sequence and the deduced amino acid sequence were









23

analyzed by using the updated version of sequence analysis

software package offered by the Genetics Computer Group (GCG)

of University of Wisconsin (Devereux et al., 1984).


Nucleotide Sequence Accession Number


The nucleotide sequence reported in this work has been

assigned Genbank accession number U62581.


Results


Evidence for a Flanking Regulatory Region


The original clone (pJZ9) isolated by Zhao et al. (1994)

produced markedly less phenylalanine hydroxylase activity than

did subclone pJZ9-3a (Fig. 2-2A). A possible explanation was

the presence of a negatively-acting regulatory gene in either

the upstream or downstream flanking region. Plasmid pJS60, in

correlation with its absence of upstream DNA but presence of

downstream DNA, expressed a very high level of activity. Thus,

the upstream region appeared to be responsible for decreased

expression of phhA in E. coli.

Transcriptional fusions were constructed using lacZ as a

reporter gene, as diagrammed in Fig. 2-2B. The results

indicate that the negative effect conferred by upstream DNA

occurs at the transcriptional level.









24





(A)
M M eH H M > H M M
- a I I PhhA activity
phhA phhB phhC
1 kb
pJZ9 2.7
pJZ9-3a 94.0
pJS7 3.5
pJS60 69.7

(B) P-Galactosidase
activity
(Mder Unas)
pJS61Z [::-::-:-:-:-:-:-- -: c :::: : 17
pJS62Z --[ : -: I 1250





FIG. 2-2. Localization of a regulatory region upstream of the
phh operon. (A) On the right phenylalanine hydroxylase (PhhA)
activities are shown in E. coli JP2255 harboring different
plasmids shown on the left. (B) On the right 6-galactosidase
activities are shown in BW545 harboring the phhA'-lacZ
transcriptional fusions diagrammed on the left.









25

Identification of phhR


A large open reading frame (Fig. 2-3) located upstream of

the phh structural genes appeared likely to be functional on

the criterion of GCG codon preference analysis. The gene,

denoted phhR, produces a deduced protein having 518 residues,

an anhydrous molecular weight of 56,855, and an isoelectric

point of 7.17. It contains a single tryptophan residue.

Regions corresponding to a possible a70 promoter region

and a factor-independent transcription terminator are marked

in Fig.2-3. A strong ribosome-binding site was not apparent.

Bases that are complementary to P. aeruginosa 16S rRNA at the

3' terminus are marked. Perhaps the "A-richness" of the

initiator region enhances ribosome binding (Ivey-Hoyle and

Steege, 1992).

A physical map is given in Fig. 2-4 of the 5874-bp DNA

segment containing the structural genes of the phh operon, the

divergently transcribed regulatory gene phhR, and a gene

(pbpG) downstream of the phh operon which encodes a

penicillin-binding protein (Song and Jensen, unpublished

data).


Homology of PhhR with E. coli TyrR


The closest homolog of PhhR was found to be E. coli TyrR.

The pairwise GAP alignment (GCG) is shown in Fig. 2-5. TyrR

belongs to a family of modular proteins which usually have

three functional domains. The alignment showed high level of












26











-350 -10 iN ,--- P
GT5O OCTTGTG -"(-TTO 120
CGCrCrACOSG 77CCCGG1G CrC X 240
C Q N I V G0 I L RDN I L N L L V D0Y GV I N V N R G N V G G0D Q G N A 0 Y L L C P 46

CGAAnxTGATC3LAC C_ T G GTGGG 360

C0G

SS V D Al PV RA NRINA P L ID A PLRV VL IL L 24166
OSZHD3S ALAG V LTL*,VGQLPIDS 2 106








0 C P R N V G S D B N V V L D V R V I C A V 0 V I U S N . C A P 0 B P I 0 I L Y 166
c7 1320

S~ 1640



XA TCCaUXACA.....T*CX T 1680

(?TBGGCOmA*AGJUTcQGGaGJUXK'(n^orGCuAXTTCmS 1800










FIG. 2-3. Nucleotide sequence of the phhR region. The numbers
at the right indicate nucleotide and amino acid positions. The
putative promoter region and ribosome-binding site (RBS) are
indicated with bold print. RBS bases that are complementary to
P. aeruginosa 16S rRNA are overlined. The translational start
site is indicated by a bent arrow and the stop codon by an
asterisk. Nucleotides forming the complementary stems of the
putative transcriptional terminator are marked with tandem
arrowheads. Restriction endonuclease recognition sites are
marked above the nucleotide sequence.









27












5" a = > B a :


phhR phhA phhB phhC pbpG







FIG. 2-4. Physical map of the DNA fragment containing phh
structural genes, the divergently-transcribed regulatory gene
phhR, and the downstream penicillin-binding protein gene
(pbpG) in pJZ9. Terminators downstream of phhC and phhR are
indicated. The shaded bars at both ends are portions of the
multiple cloning site of the pUC18 vector. The location of
restriction sites is shown.
















FIG. 2-5. Pairwise alignment (GAP program of GCG) of amino acid sequences corresponding
to E. coli TyrR and P. aeruginosa PhhR. The similarity is 66.3% and identity is 45.7%.
The three functional domains are indicated at the right, the central domain also being
shaded. Domain boundaries are based upon those formulated by Morrett and Segovia (1993).
Alernative boundaries based upon domain segments surviving partial hydrolysis by trypsin
(Cui and Somerville, 1993) are residues 1-190, 191-467, and 468-513 for the three
domains of Eco-TyrR, respectively. In the N-terminal domain, the region between amino
acids 2 and 19 that have a critical role in activation of the expression of E. coli tyrP
and mtr (Pittard, 1996) is double-underlined, and mutations at residues marked in
boldface type abolish TyrR-mediated activation without affecting repression; a second
region between amino acids 92 and 103 which may play a subsidiary role in activation is
also double-underlined; mutations at the residues in boldface type resulted in loss of
function (Pittard, 1996). In the central domain, two ATP-binding sites and a leucine-
zipper motif are underlined. Mutations altering ATP-binding site A and mutations at the
highly conserved residues E-274, G-285, and E-302 abolish TyrR-mediated repression. In
the C-terminal domain, a helix-turn-helix DNA-binding motif is identified with the helix
regions underlined and critical residues in bold print.














TyrR-Eco MRLEVFCEDRLGLTRELLDLLVLRGIDLRGIEI..DPIGRIYLNFAELEFESFSSLMAEIRRIAGVTDVR 68
7 7 1 ! : : ! : 1 11.! 1:....: 1. . I I I ! I . : . : :I :
PhhR-Pae MRIKVHCQNRVGILRDILNLLVDYGINVNRGEVGGDQGNAIYLLCPNMINLQLQSLRPKLEAVPGVFGVK 70

TyrR-Eco TVPWMPSEREHLALSALLEALPEPVLSVDMKSKVDMANPASCQLFGQKLDRLRNHTAAQLINGFNFLRWL 138
I : I I III I .I . I I . I i ! I ! :.: . -. | : ::! : .. . ... :::::: : : N -Terminal
PhhR-Pae RVGLMPSERRHLELNALLAALDFPVLSVDMGGQIVAANRAAAQLLGVRVDEVPGIPLSRYVEDLDLPELV 140 Domain

TyrR-Eco ESEPQDSHNEHWINGQNFLMEITPVYLQDENDQ .HVLTGAVVMLRSTIRMGRQLQNVAAQDVSAFSQIV 207
�:-I II 1: . II1- :1 : .. I*III : I:.. 1:1 .: :I !::.:!. 1.
PhhR-Pae RANKARINGLRVKVKGDVFLADIAP..LQSEHDESEALAGAVLTLHRADRVGERIYHVRKQELRGFDSIF 208
ATP-binding Site A
TyrR-Eco AVSPKMKHVEQAQKLAMLSAPLLITGDTGTGKDLFAYACHQASPRAGKPYLALNCASIPEDAVESELFG 277

PhhR-Pae QSSRVMAAVVREARRMAPLAPLLIEGETTGKLAACHSP F NC E TELF 278
Unique Gap ATP-binding Site B
TyrR-Eco HA ........PEGKKGFFEQANGGSVLLDEIGEMSPRMQAKLLRFLNDGTFRRVGEDHEVHVDVRVICAT 339
M. : .1 Central
PhhR- Pae YGPGAFEGARPEGKLGLLELTAGGTL LDGVGEMSPRLQAKLLRFLQDGCRVD VYLDVRVICAT 348 Domain

TyrR-Eco QKNLVELVQKGMFREDLYYRLNVLTLNLPPLRDCPQDIMPLTELFVARFADEQGVPRPKLAADLNTVLTR 409
j . .1 Hf - H . : 1 i I.:-i. 1 -- 11i.: ;
i * ! i - �: t ;j ii:i - :: iii i i i i:: ::'.[ :fii: <.i: :-:i F 1:i : i :i*r :i::: i{:;-:i: i- i:: ;;!:s^?;*?:;i g::^;; iiii ii ;� l ::::i:ii-
PhhR-Pae QVDLSELCAKGEFRQDLYHRLNLSLHIPPLECLDGLAPLA LQ 318
Leucine-zipper Motif
TyrR-Eco YAWPGNVRQLKNAIYRALTQLDGYELRPQDILLPDYDAATVAVGEDAMEGSLDEITSRFERSVLTQLYRN 486
, ,1. 1I . 1. , - - - 1 : I . I .
PhhR-Pae YHWPGNVRQLENVLFQAVSLCEGGTVKAEHIRLPDYGAPQ.PLGDFSLEGDSTHRRA.LREGVLERLFRE 486
Helix-Trum-Helix Motif C-Terminal
TyrR-Eco YPSTRKLAKRLGVSHTAIANKLREYGLSQKKNEE* .................................. 514 Domain
�-I I I . l . I I I I I I I l :r . l . . . .5-- .
PhhR-Pae HPSTRQLGKRLGVSHTTAANKLRQHGVGQSEG* .................................... 519









30

conservation throughout the entire length of PhhR and TyrR,

and 45.7% of the deduced residues were identical. The N-

terminal domain mediates regulatory modulations, and in TyrR

it binds all three aromatic amino acids. A central domain,

highly conserved throughout the entire family of a54 enhancer-

binding proteins, exhibits two established motifs that reflect

the binding of ATP (Pittard, 1996). Site A corresponds to the

ATP-binding pocket motif and site B corresponds to segment 3

of adenylate kinase. In this region a perfect leucine-zipper

motif is apparent in P. aeruginosa PhhR, whereas E. coli

displays an imperfect motif. Residues E-274, G-285, and E-302

were found to be important for TyrR-mediated repression of

aroF-tyrA in E. coli (Yang et al., 1993; Kwok et al., 1995),

and these residues are all conserved in P. aeruginosa PhhR.

The C-terminal domain possesses a helix-turn-helix motif which

is responsible for DNA binding. The absolute conservation of

residues shown to be critical in E. coli (Pittard, 1996)

strongly indicates that PhhR and TyrR might target to a

similar DNA sequences.

Similar to E. coli TyrR, the two aspartate residues and

the lysine residue conserved in the amino-terminal domain of

all response regulator proteins (Stock et al., 1989) were not

found.









31

Overproduction of PhhR


PhhR protein was overexpressed in E. coli BL21(DE3) as

detailed under Materials and methods by use of the T7

overexpression system; the construct is illustrated in Fig.2-

6. The initial use of overexpression vectors containing phhR

on the BamHI-SphI fragment (see Fig. 2-4) of pJZ9 failed. This

is probably due to autogenous regulation of phhR, judging from

the precedent set by tyrR in E. coli (Argaet et al., 1994).

Accordingly, overexpression was achieved through excision of

DNA upstream of phhR. PCR methodology was used to generate an

intact phhR gene which was fused with the T7 translational

start codon at a NdeI restriction site to create

overexpression plasmid, pJS88. E. coli BL21(DE3) that had been

transformed with pJS88 was induced with 1 mM IPTG for 3 hours

to express PhhR. Whole-cell lysates obtained before and after

IPTG induction were analyzed by SDS-PAGE, as shown in Fig. 2-

6B. Overproduction of a 56-kDa protein was observed, and N-

terminal amino acid sequencing confirmed its synonymy with

PhhR.

Initial attempts to express phhR in E. coli under

physiological conditions indicated that expression of phhR is

highly toxic. The BglII-EcoRI fragment from pJS88 was cloned

into the BamHI-EcoRI site of pUC19 behind a lac promoter. When

transformed into E. coli DH5a, transformants achieved only

pinpoint colony size and eventual survivor cells inevitably

had lost the plasmid. Success was finally achieved by use of









32


(A)


phhR pACYC184







pJS88 pJS91 a Ih
(pET11blphhR) U, (pACYC184phhR)hR










(B)
123

97.4 - -.
66.2
45.0 $ i
31.01 --
21.5 -
14.5




FIG. 2-6. Overproduction of PhhR protein. (A) Map of PhhR
overexpression plasmid pJS88 (left), and low-copy number,
constitutive PhhR expression plasmid pJS91 (right). (B) SDS-
PAGE analysis of whole-cell lysate of E. coli BL21(DE3)
harboring pJS88. The gel was stained with Coomassie blue. Lane
1, molecular-weight markers; lane 2, before IPTG induction;
lane 3, induced by 1 mM IPTG for 3 h.









33

pACYC184, a low copy-number plasmid, to create pJS91 which

carried the BglII-BamHI fragment of pJS88 ligated into the

BamHI site of pACYC184 (Fig. 2-6A). Analysis of 11 plasmids

isolated showed that the orientation of phhR in each case was

opposite to that of the Tcr gene. Presumably, the higher level

of expression expected when driven by the Tcr promoter still

confers an intolerable level of toxicity.


Functional Replacement of E. coli tyrR with phhR


A simple test was used to see whether phhR could

substitute for tyrR as a repressor of the aroF-tyrA operon.

Mutants deficient in TyrR exhibit resistance to m-

fluorotyrosine (Fig. 2-7, middle) whereas tyrR* strains

exhibit sensitivity to growth inhibitory effects of the analog

(Fig. 2-7, left) . pJS91 (phhR�) was used to transform an E.

coli tyrR-deficient background (strain SP1313) . The ability of

PhhR to replace TyrR is qualitatively apparent (Fig. 2-7,

right) by inspection of the halo of growth inhibition on a

bacterial lawn surrounding a disc containing m-fluorotyrosine

in SP1313 (tyrR- phhR) .

We also examined the ability of PhhR to replace TyrR as

an activator of mtr, encoding a component of a tryptophan-

specific transport system. The phhR* plasmid pJS91 was

transformed into two E. coli X lysogens (Heatwole and

Somerville, 1991) which carried mtr'-lacZ transcriptional

fusions integrated in the chromosome as single-copy fusions.









34








A B C

















FIG. 2-7. Functional replacement of TyrR by P. aeruginosa
PhhR in E. coli, as monitored by sensitivity to growth
inhibition by m-fluoro-tyrosine (MFT). (A) E. coli tyrR�
(wildtype) strain SP1312 is very sensitive to MFT present on
a central disc, exhibiting a large zone of growth inhibition;
(B) E. coli tyrR- strain SP1313 is insensitive to MFT, showing
no zone of growth inhibition; (C) P. aeruginosa phhR(pJS91) in
trans complements E. coli tyrR- and restores the sensitivity
to MFT, as visualized by a zone of growth inhibition.









35

Strain SP1312 (tyrR*) exhibited the expected elevation of

S-galactosidase activity following growth in the presence of

tyrosine, phenylalanine, or both. However, strain SP1313

(tyrR-) carrying pJS91 (phhR�) produced the control level of

9-galactosidase activity, regardless of the presence or

absence of aromatic amino acids (data not shown). Thus, PhhR

appears to be incapable of replacing TyrR as an activator of

E. coli mtr.


Autogenous Regulation of phhR


The BamHI-HincII fragment containing the 5' coding

regions of phhA and phhR and the intervening region (see Fig.

2-4) was fused to lacZ to give the reporter-gene construct

pJS102 (phhR'-lacZ) . This plasmid construct was introduced

into the tyrR-negative background of strain SP1313 in the

presence or absence of pACYC184 possessing a phhR' insert. The

results (Table 2-2) demonstrated a repressive effect of phhR'

upon PhhR levels as monitored by measurement of 3-

galactosidase activity. Since the copy number of pJS91 (the

source of PhhR molecules) in this experiment is lower than the

number of repressor target sites provided by the high-copy

number pJS102 and since TyrR boxes are present within seven

other transcriptional units of E. coli, auto-regulation is

undoubtedly grossly under-estimated due to titration of

available PhhR molecules in the system.









36

Table 2-2. Autoregulation of P. aeruginosa phhR in E. coli
SP1313(tyrR-) containing pJSlO2 (phhR'-lacZ)

Second B-Galactosidase levelsb in cells grown in:
plasmida M9c M9 + F M9 + Y

pACYC184 550 510 589

pACYC184(phhR�) 362 376 372
a pACYC184 (phhR') is denoted pJS91 in Table 1.
b '-Galactosidase levels are reported in Miller Units.
cM9 minimal medium was supplemented with 1 mM thiamine-HC1
and, where indicated, 1 mM phenylalanine (F), or 1 mM
tyrosine (Y).









37

PhhR as A Positive Regulator


PhhR and TyrR form a cluster within the larger family of

a54 enhancer-binding proteins, as illustrated by Fig. 2-8. A

rpoN mutant of P. aeruginosa was assayed by Western analysis

for PhhA levels of expression in order to determine whether

expression of the phh operon is dependent upon a54 like most

family members, or whether it is o54-independent like tyrR and

luxO. Only low basal levels of PhhA were present in the rpoN

mutant, indicating expression to be largely o54-dependent.

This, in turn, implied that phhR might function as an

activator protein for phhABC transcription. phhR was

inactivated as described under Materials and Methods, and

Western analysis of the effect upon PhhA level was carried

out. The results (Fig. 2-9) indicated that phhR encodes an

activator, the absence of which allows only a low basal level

of activity.

The small molecules, L-phenylalanine and L-tyrosine, was

found to function as an inducer (Fig. 2-10). Western analysis

of PhhA showed no detectable band in minimal medium and a

barely detectable band when L-tryptophan was present, compared

to prominent bands when L-phenylalanine or L-tyrosine was

additionally present. Carbon-source levels of L-phenylalanine

or L-tyrosine were not required for induction. It is probable

that L-phenylalanine or L-tyrosine is a co-activator moiety

which, in combination with PhhR, forms the holo-activator

moiety. It is perhaps relevant that for those transcriptional

















FIG. 2-8. Homology relationships of the central domain of P.
aeruginosa PhhR with the central domain of other members of
the a54-dependent family of transcriptional regulators. The
dendrogram was generated with amino acid sequences of the
central domain as defined by Morrett and Segovia (1993) by
using the PILEUP program of GCG. The top three proteins form
a cluster designated as subfamily u, and the remaining
proteins form a larger cluster designated as subfamily 3. Due
to their high degree of similarity, only one of the ortholog
sequences of NifA, NtrC and HydG proteins is shown. The six
paralogs from E. coli and the three paralogs from P.
aeruginosa are designated with * and A, respectively.
Abbreviations: Eco, Escherichia coli; Avi, Azotobacter
vinelandii; Hin, Haemophilus influenzae; Pae, Pseudomonas
aeruginosa; Vha, Vibrio harveyi. Functions controlled by the
following regulators are given parenthetically: PhhR
(phenylalanine hydroxylase), TyrR (aromatic amino acid
biosynthesis and transport), VnfA (nitrogen fixation,
nitrogenase-2), Anf (nitrogen fixation, nitrogenase-3), NifA
(nitrogen fixation, nitrogenase-1) , HydG (hydrogen oxidation) ,
NtrC (nitrogen assimilation), PilR (synthesis of Type IV
pili), AlgB (alginate production), LuxO (luminescence), FhlA
(formate metabolism), YfhA (possible control of glnB), PspF
(phage shock protein).







39



SPae-PhhR-

*Eco-TyrR (

Hin-TyrR_

*Pae-PilR-

*Eco-HydG

*Eco-YfhA

Avi -VnfA

Avi -AnfA

Avi-Ni fA

*Eco-FhlA

*Eco-NtrC

*Pae-AlgB

Vha-LuxO

*Eco-PspF_









40

















PAO-1 PA103
WT phhA phhR WT rpoN

















FIG. 2-9. Western blot analysis of phhA expression in mutant
derivatives of P. aeruginosa strains PAO-1 and PA103. The
proteins in crude extracts prepared from cultures grown in LB
medium were separated by SDS-PAGE, and equal amounts of
protein (50Ag) were applied to each lane.









41





(A)
Fructose
- +Phe +Trp +Tyr




PhhA*- d







(B) Glucose+Phe Fructose+Phe

z z
















FIG. 2-10. Western blot analysis of phhA expression. (A)
Examination of aromatic amino acids as inducers of phhA
expression. P. aeruginosa PAO-1 was grown in minimal salts-
glucose or minimal salts-fructose medium with or without
addition of one of the three aromatic amino acids at a final
concentration of 100 Ag/ml. (B) Phenylalanine induction of
phhA expression in different P. aeruginosa strains. Bacteria
were grown in minimal-glucose or minimal-fructose media
containing 100 gg/ml phenylalanine.









42

units where TyrR functions as an activator, L-phenylalanine

functions as an essential co-activator (Pittard and Davidson,

1991)


Discussion


Anomalous Repression of the phh Operon by PhhR in E. coli


PhhR can mimic the ability of TyrR to repress the aroF-

tyrA operon at a U70 promoter. This indicates that PhhR can

recognize TyrR boxes and is consistent with the high

similarity of the helix-turn-helix, DNA-binding domain within

the carboxy-terminal segments of TyrR and PhhR. However, PhhR

was unable to activate the phh operon in the heterologous E.

coli background, suggesting an incompatibility between the E.

coli RpoN and the P. aeruginosa a" -dependent system. The

expression of PhhA from a promoter recognized by E. coli

upstream of the native a"4 promoter was in fact severely

depressed in constructs containing phhR, even in the presence

of added co-activator (L-phenylalanine) . In the presence of P.

aeruginosa PhhR, an aberrant complex apparently blocks

transcription initiated upstream of the a54 promoter.


Emerging Subfamilies within the a"4 Enhancer-Binding Protein
Family


P. aeruginosa PhhR belongs to an outlying subgroup (which

we denote subfamily c in Fig. 2-8) of the a"4 enhancer-binding

protein family. All members of the family possess in common a









43

homologous central domain, but the amino-terminal and carboxy-

terminal domains may vary considerably within the family.

Thus, this exemplifies a complex multi-domain protein family

in which family membership is defined by a common ancestral

central domain. Future subdivisions within what is termed

subfamily 3 in Fig. 2-8 could likely be defined on the

criterion of homology for the remaining two domains. For

example, Eco-NtrC and Eco-FhlA belong to different mechanistic

subgroups: the two-component regulatory system and direct

response-to-small-molecules, respectively (reviewed by

Shingler, 1996).

Figure 2-8 highlights the emerging homology relationships

of selected paralog and ortholog proteins, with respect to the

central domain. E. coli possesses at least six paralogs, some

of which diverged in a common ancestor that existed prior to

speciation events which generated orthologs. Thus, the

divergence of Eco-NtrC and Pae-PilR was a more recent event

than was the divergence of Eco-NtrC and Eco-PspF. In contrast

to the ancient duplication events which generated all of the

E. coli paralogs (or the P. aeruginosa paralogs) are the

relatively recent duplication events generating the three

paralogs which regulate three distinctly separate nitrogenase

systems in Azotobacter (Joerger et al., 1989).

P. aeruginosa PhhR and E. coli TyrR exhibit homology in

all three domains: 36% identity, amino-terminal; 52% identity,

central; and 47% identity, carboxy-terminal). Curiously, the









44

amino-terminal domain of H. influenzae TyrR appears to be

absent. It is not known whether sequencing errors might

account for this, or whether the equivalent of the amino-

terminal domain might exist separately as a different protein.

A multiple alignment of the central-domain modules of

subfamilies a and 3 was shown in Fig. 2-11. In addition to the

many residues that are absolutely conserved throughout the

family, some residues which may prove to be uniquely conserved

within subfamily a are apparent, e.g., APLL corresponding to

residues 29-32 of Hin-TyrR.

Both Eco-TyrR and Rca-NtrC exhibit deletions in the

"unique-gap region" of the central domain (Fig. 2-12) in

correlation with their regulation of "70 promoters, rather than

a54 promoters. This observation led to the suggestion (Morrett

and Segovia, 1993) that this region of the central domain

might be critical for functional interfacing with a"4. Since

this DNA segment of Pae-PhhR is intact with absolute retention

of highly conserved residues, the foregoing hypothesis is

consistent with the successful interaction of PhhR with a o54

promoter. Hin TyrR, on the other hand, is likely to be

deficient in interaction with oa" (like E. coli TyrR), owing

to a 6-residue deletion in this region.


Intervening Region of Divergent Transcription


Since the DNA-binding region of the carboxy terminus of

PhhR is identical at all important residues with E. coli TyrR,





















FIG. 2-11. A comparison of the amino acid sequences in the
central domain of the PhhR protein and 13 other homologs. The
sequences were aligned by using the PILEUP program of GCG. The
numbering of amino acid residues is given on the left. Percent
identity of PhhR with its homologs is given at the lower
right. Amino acid residues conserved in all 14 sequences are
in double-lined boxes. Amino acid residues conserved in 13 of
14 sequences are shaded. Conserved residues which are confined
to either the top cluster (subfamily a) or the bottom cluster
(subfamily 3) are in single-lined boxes. Two ATP-binding
motifs are indicated above the consensus sequences in boldface
type. See the legend for Fig. 8 for abbreviations.













46







ATP-binding Motif A
a G a t a GeI
PhhR-Pa F PQSS V AAV AAVMR BAR R APL DAPLLXGK EL G UTRA 3CHLADPRQS 257
TyrR-Rco VAV8 KKHV V EQA QI(LAML SIA P L L GITI T KDLIA A |PRAGK 256
TyrM-Hin IVOSXAlS AA R FANMRA DAPLLQ S G S G K D L AAKA HYQ L LRRDK 65

PilR-Pae L G S P P RAL RNQ I G LARS A PVY S S GSGK LV ARL I HEQOPRI ER 186
HydG-Bco VGKSPAU HL LSE IALVAPS 3ATVL HGDS GTGr LVARA I ASBARSEK 191
YfhA-Hco VTR PL LRL L8QARLVAQS DVSVL^NGQ S GTG I PAQ I NA PRNSK 187
VB(A-Avi IGNSKP LIV YQLIERVVRT RTTVL iLGlS GV LV A GA I HYN PAAKG 260
AnlA-Avi IGNSKP Q V YBLIHKVAST KATVL�LGIS GVG LV A I YNSPNAEA 269
NifA-Avi VG TPT PRRV FDQIRRVAKW NSTVLVLG S VG ILI ASA I YKSPRAHR 261
fhIA-Sco IGRS 3A YSV LIKOVEMVAQS DSTVL LG3 T GTG LIARA I HNL GRNNR 431
NtrC-Eco IG AP A DV FRI IGRLSRS S I SVL.NG S GT KILV HA L RHK PRAKA 190
AlgB-P&a ISHSPA AAV LF TARQVAAT DANI LRLG S SGKGA LAR I HTWGKRAKK 197
LuxO-Vha I SgSQ Q QV YRTIDBAASS KA I P TG S GTG EVCAeA I HAA KRGDK 183
PSpP-Bco LGZANIS L V LIQVSHLAPL DKPVL^ IGBR G G LIA S R L YL SRWQG 58



PhhR-Pa Pr F nL A G L PB S HA Tl L YG PG-AfClA G A RPBGKL GL L LTAG TLFLrT 307
TyrR-Ico PY LIA L NA SI PI HA ..... P E G KKG P FB Q N G S V LL D 298
TyR-Hlin PI v AAGL PD B DA M GRVGD ....SeT IGPFJ Y N KTV L LD 109
TyrR . H n K F I A A GL PTDS D A E Y N T V L L 109
Pill-Pa. PPVPV oAI PS BL N FP GHW K GK T G6 . I D KQGL Q A SGdTL LD 235
HydO-Bco PL TL N AAL LLLL L GH KGAFTGA DKRRIGR V B" DG TL LD 240
YfhA-Bco pF IA INGAL P.QLL L f L GP H HARGAFPT GA VS N FR GL Q A G 0T L IL D 235
VnfA-Avi PPVKI N ASL PBISVI i LK GH s KGSPTG . IG LRKGR BE ] A G i T IILD 309
AnA-Avi ALVTSAPL PBLA L GB G . LTHGC QiDGTILD 318
NHiA-Avi PFVRL AAL PBTLL t L G]H KGA T VITRKO G7 0DGFT LD 310
PFhA-co RMV KMNAAM PAGLL L . SAQ I P LIDKS LD 480
NtrC-.co P F I A AAI P KD L G KGAPTG T R P QAlDGT TLD 239
AlgB-Pa PQVTIN PSL TA LM LI1 LP GHSRGAFTG . TBSTLORV S QA D GBTLLD 246
LuxO-Vha PTIAIA AAI PADLI Ii LF GHVKGA TG . A DRN GAA LiDG TL LD 232
PpF-Bco P1F I S LN AAL NBNLL LW . GHBAGA T G QK HPGRF R DG TL_ L D 107

ATP-binding Motif B
.kLLCVLqL
PhhK-Pae GV V Gln1P R L 0 R la l CI PRRIVGSODBl VYLDVV I W Iv D1S BRLIcA 357
TyrR-Bco B I RBMSP ALLR L IDG TIHRRIVG DH B VHV RV I CC t 0 VKH VL L V0 348

PillR-Pa l A D L P V A KLLRAI K AVRAViG G E VAV DLRI LC H K D AA VG 285
HydG-Bco BI GDISP MMQ V ILILIR A 1Q 0 EVQRVGSNQI I SVDVR LIA tH RD AAEVN 290
YthA-co I IG 0 MP A P L Q V KLILl V L 0 R K KVR S N RD I D I N V I I St HR D 0 P KAMA 285
VnfA-Avi R VVG 1SLTTQ AKLLRVLOa R SFERVGGNTT IHVDLRVIAA T RN AHMVA 359
AnfA-Avi VG L S P T VQ AKLVRVLQ R T P VGG S KP V V D VRI I A A NRN VEV MV 368
NitA-Avi jIGI IS PMFQ Al LLRVLQ 8 G I FERVGGNQT VRVNVR I VA StNRD E S EVE 360
FhlA-Hco VOD PLIL PKLLRVLQQ FERLG SNK I I 0TDV RL I A HNRD KKMVA 530
NtrC-Eco IGDMP LDVQ TRLLRVLADG QPYRVGGYAP VKVDVRI I A A H N HQRVQ 289
AlgB-PaC I GD P LTLQ P LLR I QDK BY ERVGD PVT RADVR I LAIt RNRDIGAMVA 296
LuxO-Vha ILCBN LDL Q T RLLrPI TG TFIKVGSo S K KSVDVRFVCA RnRDPWKEVQ 282
PWpF-Eco LATAPMMVQ EKLLRVI BYG LERVGGSQP LQVNVRLVCA RLAD PAMVN 157



PhhR-Pae KS S S VRi# QD L Y H 5 SLHIP p CL LA TL HLS iL 0I0DLA P H DQ SRQIGCGL .MP 407
TyrR-Eco M i D L YY RL NI T LTLN P DCPQPIM DL E LL F A DEIQGVPR . G V 398
TyrR-Hin Q KV ADLFH RL N #MLjT INV AL DRMAI [P L A Q GJL Q S E L KIAK P 209

PilR-pae A R 0 D L Y Y RL NVI R LRV P LlERR I P LLAB R I LKR AGDTGLPA.A 335
HydG-Eco AG I QDLYY RL N VAI8V SLiORREBI LLGHFLQRF A RNRKAV . 340
YfhA-Bco R S DLYY RLN VS L K IP AL DGT B P LL NH LL LR A ADGHKP FV.R 334
VnfA-Avi D T ';#A BDLYY RL N P ITI P L ERGS I I TL DH FVSR SRBMG I V N 409
AnfA-Avi 0 T Ri DDLYY RL PITP PL ERGS V VI AL DHFVSAP S RENGKNV K 418
NifA-Avi K K s DLYY R LN MA I R IP PL tERTAI P E L FLLG K I GRQGR P L .T 409
FhlA-co Di KE S DLYY RLN L P ER P EIP L L K A T F K I ARR LGRN I .D 580
NtrC-co K IF BD L FH RL I RVH L P P RR L R H P L QVA AR LGV E A .K 339
AlgB-Pae Q Q EDLLY RILN IVL L P PL RRA E IL G L E R FLAR VKDYGR PA.R 345
LuxO-Vha BR : DLYY RL I PLHL PL E RGKDV I I Y S LLGYM S HE BGKS F V 332
PpF-ECO E T AD L LD RL A DVVQ L PLB E RE S I M LM EY FA I QN CRE I KLP L F P 208


% Identity
PhhR- Pae KI LSAQAL RL I RY-lH ni N Vt i V QL VL QiV L G 442 -
TyrR-Eco KA LAAD LNTVL T R YA ;GNV QL N A I Y RIAIL TQ L DG 4Y 33 54
TyrR-Hin TFD KD FLLYL Q K[YJD it GNVIt 2LYNTLY R AC S L V D N 244 47

PilR-Pae RLTGDAQBKL KNYRWP GNVG 8LENML RRAY TLCEDD 370 421
HydO-Eco GPTPQAMNDLL I GYD W. GI IR EL BNAV5RAV VLLTGB 375 431
YfhA-co AFSTDAMKRL MTAS Gti#GNVK QLVNVI RQCV ALTSSP 260 451
VntA-Avi R I STPRLNML OSYQ G NV R BL ENVI BRAM LLSEDG 444 475
AnfA-Avi R I STPALNML MS YH GIN VK E L NVME RAV I LSDDD 452 451
NifA-Avi . VTD S A I R LL M S HR : G NV L C L R S A I M S E DG 444 48%
PhlA-Eco S I PA TLRTL SNME^ G NV R EL9NV I 5RAV LLTRGN 615 44
NtrC-9co LLHPRTBAAL TRLA .IGNVI Q L 8NTTCRWLT VMAAGQ 374 44%
AlgB-Pae G FS I AARBAM RQY P K IGN Vtf E LRN V VI RAS I I CNQ E 381 45�
LuxO-Vha R FA DV I R F NS YE i N VR Q LQN VLRN IV VLNNG K 367 45%
PspP-Eco G FTERARBTL LNYR ! =GN IN B NK VVE R S V YRHGTS 243 41%














Eco-TyrR P E D A V E SE L F H A ....... . P E G K KG F F E Q A N G T S V LID 298
Bin-TyrR P D E D A E S E M F G R K V G D ... . . . S E T I GO F F E Y A N K GTVL L D 109
Pae-PhhR P E S M A E T :E L Fi YG P GAF EYGA R P E G K L G L L E L T A G G T L F L D 307

Pao-PilR P S E L M E S E F F G H K K G S F TG A . I E D K Q G L F Q A A S G G T L F L D 235
Eco-HydO NE S L L E S ELF G H E K GAF TIG A D K R R EGR F V E A D GGT L F D 240
Eco-YfhA P E Q L L E S i LF H A RiG A F TIG A V S N R E G L F Q A A E G G T L F LD:: 235
Avi-VnfA P E S V I E S E L G H E KG S F TGA . I G L R K GR F E E A A G G T I Fll D 309
Avi-AnfA P E N L A E S E L F G H E KIG S F T G A L T M H KiG C F E Q A D G G T I F I D 318
Avi-NifA P E T L L E SELF H E K G A F TG A V K Q R KGR F E Q A D G G T L FL D 310
Eco-FhlA PAG L L ES L H E RGAFTGA . SAQRIGRFE LADK S S L F L D 480
Eco-NtrC P K D L I E S E LF G H E K GA F TaG A N T I R Q G R F E Q A D G G T L F L D 239
Pas-AlgB T A E L M E S E LF |G H S RGA F TGA . T E S T L G RRVS QADGGT L F L D 246
Vha-LuxO P K D L I E S E L F GHVKIGAFT GA A ANDRQGAAE LADG GT L F L D 232
Eco-PspF N E N L L D SELF H E AAFTG AGAFTG . Q K R H P GRFE R A D GGTLF L D 107

Rca-NtrC L G A D G P S S LL . . . . . . . . . . . . . . . . .. . A R R C G R L V V F 222




FIG. 2-12. Alignment of the unique-gap region in the central domains of TyrR
proteins with selected homologs. Amino acid residues conserved in all of the 15
sequences that include both subfamily a (top cluster) and subfamily ( (lower
cluster) are in shaded boxes. Amino acid residues conserved within the gap region
are shown in open boxes. Rca, Rhodobacter capsulatus; see the legend of Fig. 8 for
other abbreviations used.









48

it is likely that PhhR binds to the same binding sites for E.

coli, which are referred as "TyrR boxes" (consensus:

TGTAAAN6TTTACA). This conclusion is also supported by the

ability of PhhR to replace TyrR as a repressor of the aroF-

tyrA transcriptional unit. The location of two "PhhR boxes"

which match the consensus for "TyrR boxes" was shown in Fig.

2-13. PhhR Box 1 is a strong box (with more conserved-symmetry

and higher affinity for TyrR) that overlaps the putative -10

region of the phhR promoter. TyrR boxes in E. coli occur in

tandem with variable spacing (Pittard, 1996), and a TyrR

hexameric molecule is thought to bind both a strong box and a

weak box with DNA looping in between. PhhR Box 2 is a weak box

located in the middle of the intervening region. It seems

probable that by analogy with autorepression of tyrR in E.

coli, both phhR boxes participate in the autogenous repression

of phhR by PhhR, probably with tyrosine as a corepressor.

In the opposite direction of transcription, the a54

promoter for phhABC requires an upstream activator site (UAS).

PhhR Box 1 may be the most likely UAS, although perhaps both

boxes participate in activation of phhA. L-Phenylalanine and

L-tyrosine, potent inducers of phenylalanine hydroxylase,

presumably are the effector molecules. Since a rpoN mutant

retained low basal level of PhhA, another promoter that is

independent of a"4 might be present.

No motif for binding of integration host factor (IHF)

(Friedman, 1988) was located in the intervening region.













49
















Banrm I
CCTAGGCGAGCACCCCSTCCGGCTCGACAACGTCSGCAGCTCCATAA3ACGCGCGG CAGCAA120
P I R E H P L G L Q E I G D L Y E C A R G E I V K L R T I L T N V H E T
ACOGACTCCATGCCTC GATCGA 240
GGCCTTCACCTACTGGCAAGCAGCCCGACCGCCCGGTCTGACGCAGCAAAAGTATGCCTGgaTACTCGC rCACCGGa
E P Y H I F G N D D P Q R A V Y Q T T KM EF U
pA -- ----12

PhhR Box 2
TACCGCGTCGGGACCGTCCGAGCCCCGGATAGGGACCGGGCAACGGGGAGGAATCGGCGT CGTA TACG AAT TGT CATIGGTCAGGCT 360
ATaGCGCAGCCCTGCCAGGCArCGTGGGCCTATCCCIGGCCCGTGCCCCTCCTTAGCCGCAAAA GCATrTCAAKAAGGAOC TTAACCGGACCCAGCGGACAAGTAACCCAGTCCGTA
-24
Tyrr Box 1
-- phhX
-35 -0 ES M R I K V H
TGTTCAGTCTAACTcGTAATTCACATATcTrGACG GAAATTCCGGCCGGGAGTrAAAAAACCGGCGCGAGCCCATCAGHCi5CACACCGGGCCACGCCATGCTATCAAAGTGC 480
ACAAECGACCACTCADATTG1 ACAGTATAAGAACTGC 7AAAGGCCGGCC ..............
Tri-itor ?
C 0 N R V G I L R D I L N L L V D Y G I N V N R G E V G G D G N A I Y L L C P
ACTGCCAGAACCGTGTAGGCATCCTCCGCGACATCCTCAACCTGCTGGTCGACTACGGCATGTTG GTCAC 600















FIG. 2-13. The intervening sequence between the divergently
transcribed phhA and phhR genes. The number at the end of each
line indicates the nucleotide position. The ribosome binding
sites and putative promoter sites (-12/-24 promoter for phhA,
and -10/-35 promoter for phhR) are indicated. The
translational start sites are indicated by arrows. Two PhhR
boxes are identified. A stem-loop structure is shaded.
Restriction endonuclease recognition sites are marked.









50

Therefore, this region may possess intrinsic DNA-bending

capabilities.


Function of the phh Operon


The primary function of the phh operon is clearly not to

accommodate tyrosine biosynthesis since the feedback-inhibited

cyclohexadienyl dehydrogenase which is widely distributed in

gram-negative bacteria exists for this purpose. However, the

phh operon probably provides a fortuitous backup capability

for tyrosine biosynthesis. "Reluctant auxotrophy" for tyrosine

(Patel et al., 1978) can be explained as follows. Mutational

deficiency of cyclohexadienyl dehydrogenase would lead to

accumulation of prephenate, a potent product inhibitor of

chorismate mutase. The subsequent backup of chorismate,

enhanced by lack of early-pathway control in the absence of L-

tyrosine, results in passage of chorismate to the periplasm

where chorismate mutase-F (Gu and Jensen, unpublished data)

and cyclohexadienyl dehydratase (Zhao et al., 1993) generate

L-phenylalanine. Subsequent induction of phenylalanine

hydroxylase completes the alternative circuit to L-tyrosine.

The established function of phenylalanine hydroxylase in

mammals is for catabolism of L-phenylalanine as a carbon

source. We have found that phenylalanine hydroxylase is indeed

essential for use of L-phenylalanine as a sole carbon source

in P. aeruginosa. Thus, inactivation of phhA resulted in

inability to use L-phenylalanine as a sole source of carbon









51

(data not shown). However, induction of the phh operon under

conditions where better carbon sources (such as glucose)

coexist, suggests that the phh operon might be dedicated to

provision of some specialized compound from L-phenylalanine.

Inactivation of phhR resulted not only in the inability

to use L-phenylalanine as a carbon source, but also in an

inability to use L-tyrosine as a carbon source. Since TyrR

regulates aromatic amino acid permeases in E. coli, we

considered the possibility that the phhR mutant might fail to

grow on L-tyrosine because of a permease deficiency. Since MFT

is likely to be transported by the same system as L-tyrosine,

a permease-deficient phenotype should be resistance to growth

inhibition by MFT. However, the phhR- mutant has a MFT-

sensitive phenotype on fructase-based medium (data not shown).

Therefore, PhhR might regulate steps of tyrosine catabolism.


Regulation of Multiple Transcriptional Units by PhhR?


TyrR represses or activates eight transcriptional units

in E. coli (Pittard, 1996). Similarly organized

transcriptional units are absent or unknown in P. aeruginosa.

However, the counterpart of the aroF-tyrA operon in P.

aeruginosa would be genes encoding tyrosine-sensitive DAHP

synthase and cyclohexadienyl dehydrogenase. Physiological

manipulations in our laboratory have never revealed repression

control of these apparently constitutive enzymes. Consistent

with this, PhhR exhibits no regulatory control of either of









52

these enzymes, on the criterion of assessment of specific

activities determined in comparison of tyrR' and tyrR-

backgrounds (data not shown).















CHAPTER 3
BIFUNCTIONAL PhhB REGULATES THE EXPRESSION OF
PHENYLALANINE HYDROXYLASE IN Pseudomonas aeruginosa


Introduction


Mammalian 4a-carbinolamine dehydratase was initially

known for its catalytic activity of converting 4a-

carbinolamine to quinonoid dihydrobiopterin in regenerating

the tetrahydrobiopterin for phenylalanine hydroxylase (Fig. 3-

1B). Later, it was found to be synonymous with DCoH, the

dimerization cofactor for hepatic nuclear factor 1 alpha (HNF-

la) (Citron et al., 1992).

A homolog of the mammalian DCoH, PhhB, was found in

Pseudomonas aeruginosa by Zhao et al. (1994) . The PhhB

protein is encoded by the second structural gene, phhB, of the

phh operon (Fig. 3-1A) . Zhao et al. (1994) reported that phhB

is required for the expression of phenylalanine hydroxylase,

encoded by the first structural gene, phhA. In the absence of

the phhB gene, phhA by itself not only failed to complement E.

coli tyrosine auxotrophy, but was not expressed in E. coli as

indicated by SDS-PAGE. Dual catalytic and regulatory roles of

PhhB are an intriguing possibility in the context of the fact

that DCoH, the mammalian counterpart of PhhB, is a

bifunctional protein with enzymatic activity as 4a-


53









54



(A)




p phhR phhA phhB phhC




(B)

0 PHE H H

HN3o N o O
H 0 H
t2' H H
i, , 8 IPhhiAI H,N N N
H H
TETRAHYDROBIOPTERIN 4a-CARBINOLAMINE


|DHPR|
NAD* ' PRI [PhhB

NADH+H� N H t10


HN N N
H
QUINOID DIHYDROBIOPTERIN




FIG. 3-1. (A) Physical map of the phh operon in Pseudomonas
aeruginosa. The endonuclease restriction sites are shown at
the top. The arrows indicate the position of the genes and the
directions of transcription. Putative transcriptional
terminators (t inside a circle) are indicated. The proteins
encoded by the genes are as follows: phhR, a"u transcriptional
activator of the phh operon; phhA, phenylalanine hydroxylase;
phhB, 4a-carbinolamine dehydratase; and phhC, aromatic
aminotransferase. (B) Regeneration of the pterin cofactor for
phenylalanine hydroxylase. The enzymes involved are indicated
as follows: PhhA, phenylalanine hydroxylase; PhhB, 4a-
carbinolamine dehydratase; and DHPR, dihydropteridine
reductase.









55

carbinolamine dehydratase and regulatory activity as the

dimerization cofactor of HNFla. In this chapter, I report the

results of studies aimed at elucidation of the extent and

nature of the regulatory function of PhhB protein.


Materials and Methods


Bacterial Strains, Plasmids, Phage, and Media


The bacterial strains, plasmids, and phage used in this

study are listed in Table 3-1. The LB and M9 formulations

(Sambrook et al., 1989) were used as growth media for E. coli

and P. aeruginosa. Pseudomonas isolation agar (Difco) was

used for isolating P. aeruginosa knockout mutants. Additions

of ampicillin (100 Ag/ml), chloramphenicol (40 Aug/ml),

kanamycin (50 Ag/ml), mercuric chloride (15 Ag/ml), L-

phenylalanine (50 Ag/ml), and thiamine (17 .g/ml) were made as

indicated. Agar was added at a final concentration of 2%

(w/v) for preparation of solid medium.


Recombinant DNA Techniques


Molecular cloning and DNA manipulation including plasmid

purification, restriction enzyme digestion, ligation, and

transformation were conducted by standard methods (Sambrook et

al., 1989). DNA fragments were purified from agarose gel with

a Geneclean kit (Bio 101). Electroporation (Invitrogen) was

used for simutaneous transformation of E. coli with two

compatible plasmids. Restriction enzymes, T4 DNA ligase, DNA-









56

Table 3-1. Bacterial strains, plasmids, and phages used in
this study

Strain or Relevant genotype Source or
plasmid or description reference

E. coli

BL21(DE3) F- ompT hsdSB (rB-mg-) gal dcm; Novagen
with DE3, a X prophage carrying
the T7 RNA polymerase gene

DH5u F-AlacU169 480dlacZAM15 hsdRl7 GIBCO/BRL
recAl endAl gyrA96 thi-1 relAl
supE44

LE392 F-e14- (McrA-)hsdR514 (rk-mk�) Sambrook et al.
supE44 supF58 lacYl or A (lacIZY) 6
galK galT22 metBl trpR55

JP2255 aroF363 pheA361 phe0352 tyrA382 Baldwin &
thi-1 strR712 lacYl xyl-15 Davidson

JSI SP1313i(phhA'-lacZ) This study

S17-1 [RP4-2(Tc:Mu) (Km:Tn7)Tra(incP)1 Simon et al.
pro hsdR recA Tpr Smr

SP1313 zah-735:Tn10A(argF-lac) U169A (tyrR) Heatwole &
Somerville

P. aeruginosa

PAO-1 Prototroph Holloway

JS101 PAO-1 phhA, Hgr Song & Jensen

JS102 PAO-1 phhR, Hgr Song & Jensen

JS103 PAO-1 phhB, Hgr This study

JS104 PAO-1 phhC, Hgr

Plasmids

pUC18 Ampr lac'IPOZ' Yanisch-
Perron et al.
pUC19 Ampr lac'IPOZ' Yanisch-
Perron et al.
pACYC177 P15A replicon, Apr Kmr Chang & Cohen









57

Table 3-1. (continued)

pETllb T71ac promoter, lacI' Apr Novagen

pET23 T71ac promoter, lacI+ Apr Novagen

pGEM-3Z T7 promoter, Apr Promega

pGST-DCoH In-frame protein fusion of
glutathione S-transferase and DCoH Citron et al.

pJS10 phhAB, 2.5-kb HincII fragment This study
cloned into pGEM-3Z behind
the T7 promoter

pJSll phhAB', 1.44-kb HincII-EcoRV This study
fragment cloned into pACYC177

pJS12 phhAB, 2.5-kb HincII fragment This study
cloned into pACYC177

pJS51 HincII-BamHI fragment containing This study
truncated phhA' cloned into
pACYC177

pJS51Z phhA'-lacZ transcriptional fusion This study
in pACYC177

pJS63 phh'ABC, BamHI-HindIII fragment This study
cloned into pGEM-3Z behind
the T7 promoter

pJS72 phhA, PCR-generated fragment This study
containing the native ribosome
-binding site and PhhA-coding
region cloned into pET23 behind
T7lac promoter

pJS95 PhhA overexpression vector; This study
PhhA-coding region fused with T7
translational initiation signal
at NdeI site of pETllb

pJS96 PhhA overexpression vector; This study
phhA fused with T7 translational
initiation signal cloned into pUC19
behind lac promoter to constitutively
overexpress PhhA









58

Table 3-1. (continued)

pJS97 PhhA overexpression vector; This study
phhA fused with T7 translational
signal cloned into pTrc99A behind
trc promoter

pJSl01 Hg'-cassette, Apr Song & Jensen

pJS105 HincII-BamHI PCR fragment This study
containing phhA' with a frameshift

pJS105Z phhA'-'lacZ protein fusion cloned This study
into pACYC177

pJZ9 phhRABC, Apr Zhao et al.

pJZ9-3a phhAB, Apr Zhao et al.

pJZ9-4 phh'ABC', Apr Zhao et al.

pJZ9-5 phhAB', Apr Zhao et al.

pMC1871 lacZ protein fusion vector Pharmacia

pTrc99A Trc promoter, lacId Apr Pharmacia

pUFR004 ColEl replicon, Cmr Mob* mobP, DeFeyter et al.
lacZa'

Z1918 Promoterless lacZ, Apr Schweizer


Phages

XRZ5 X'bla 'lacZ lackY Resental et al.

XJSI X)(phhA'-lacZ) lacrY 'bla This study









59

modifying enzymes (New England Biolab or Promega), Taq DNA

polymerase (Perkin-Elmer), and Vent DNA polymerase (New

England Biolab) were used as recommended by the suppliers.


Phenylalanine Hydroxylase Assay


E. coli JP2255(pJZ9-3a) was grown at 370C in 500 ml of LB

broth supplemented with ampicillin (100 ig/ml), and harvested

at the late-exponential phase of growth. Cell pellets were

resuspended into 8 ml of 10 mM potassium phosphate buffer (pH

7.4) containing 1 mM dithiothreitol (DTT), and the cells were

disrupted by sonication. The resulting extract was

centrifuged at 150,000 x g for 1 hr at 40C. The supernatant

was desalted using Sephadex G-25 and used as crude extract for

enzyme assay. Phenylalanine hydroxylase (PhhA) was assayed by

following tyrosine formation (Nakata et al., 1979).


Phenylalanine Hydroxylase Stimulation Assay


4a-Carbinolamine dehydratase activity in E. coli(pJZ9-4)

was assayed indirectly using the phenylalanine hydroxylase

stimulation assay (Citron et al., 1992). Reaction mixtures

containing 30 mM potassium phosphate (pH 8.3), catalase (1

mg/ml), 100 MM NADH, 1 mM phenylalanine, 20 MUg

dihydropteridine reductase, 14.4 Mg rat liver phenylalanine

hydroxylase, and 2.9 AM 6,7-dimethyltetrahydropterin were

incubated at 250C. Approximately 1 min after the reaction was

started, either buffer (control) or 15 Mg of the E. coli









60

crude extract containing PhhB or GST-DCoH was added. The

reaction was monitored at 340 nm for the oxidation of NADH by

dihydropteridine reductase as quinonoid dihydropterin was

recycled to tetrahydropterin.


Construction of phhA:lacZ Transcriptional and Translational
Fusions


For the transcriptional fusion (phhA'-lacZ), the HincII-

BamHI fragment containing the upstream region of phhA was

first cloned into pACYC177 that had been digested with HincII

and BamHI, creating pJS51. A BamHI-cassette of a promoterless

lacZ gene from the plasmid Z1918 was then inserted at the

BamHI site of pJS51 in the same orientation as phhA to create

pJS51Z, which was used as a low-copy phhA'-lacZ fusion. A

single-copy fusion X(phhA'-lacZ) was obtained by transferring

the phhA'-lacZ fusion from pJS51Z into XRZ5 following the

procedure described by Yu and Reznikoff (1984).

For the translational fusion (phhA'-'lacZ), the HincII-

BamHI fragment containing the upstream region of phhA was

generated by PCR with the upper primer 5'-

GACAGAGCAGGTAGATGGCGTT-3', and the lower primer

5'GGGATCCGGCTCGTGGGGCAGGCCGA-3' (BamHI site underlined). An

extra guanine nucleotide (G in bold) was added in the lower

primer to create the frameshift needed for an in-frame fusion

at the BamHI site to generate phhA'-'lacZ. The HincII-BamHI

fragment with the frameshift was inserted into the HincII-

BamHI site of pACYC177 to create pJS105, and a BamHI-cassette









61

of truncated 'lacZ from pMC1871 was inserted into pJS105 to

create the translational fusion plasmid, pJS105Z.


13-Galactosidase Assay


O-Galactosidase activity was assayed under conditions of

proportionality as described by Miller (1972), and specific

activities are expressed in Miller units. The data are the

results of at least two independent assays.


Construction of PhhA and PhhB Expression Vectors


To express PhhA protein, expression plasmids pJS72 and

pJS95 were constructed. A PCR fragment containing the

complete coding region of phhA and the native ribosome-binding

site (RBS) was amplified with the upper primer 5'-

CATGGAGTCCGTATGAAAACGACGCA-3' (RBS underlined; ATG start codon

in bold) and the downstream primer 5'-

CTTGGTTGTCGCATGTGGGAGCGGCG-3', and cloned into pET23 behind

the T71ac promoter to create pJS72. pJS95 was constructed by

inserting the coding region of phhA into the translational

fusion vector pETllb. The coding region was amplified by PCR

with the upstream primer 5'-CCATATGAAAACGACGCAGTACGTG-3' and

the downstream primer 5'-CAAGTCTGGTTGTCGCATGTGGGAGCGGCG-3'.

The upper primer was made with a built-in NdeI site

(underlined) which allows fusion of phhA at the translational

start site (ATG in bold) with the T7 translational initiation

signals. To constitutively express the PhhA protein, the









62

phhA-coding region together with the upstream T7 translational

start signals were excised from pJS95 as a XbaI fragment and

cloned into pUC18 downstream of a lac promoter to create

pJS96. The XbaI fragment was also cloned into pTrc99A

downstream of the inducible trc promoter to create pJS97.

Two similar plasmids, pJS10 and pJS63, were constructed

to express PhhB. The HincII fragment containing both phhA and

phhB gene was inserted into pGEM-3Z to create pJS10, and the

BamHI-HindIII fragment containing both phhB and phhC was

inserted into pGEM-3Z to create pJS63. The phhB gene was

under the control of a T7 promoter in both plasmids.


Preparation of PhhB-specific Polyclonal Antiserum

PhhB was partially purified by anion-exchange and gel-

filtration chromatography following the methods described by

Zhao et al., (1994). The partially purified PhhB was subject

to SDS-PAGE (12%) and the gel was stained with Commassie blue

R-250. The PhhB band was cut from the gel and used for the

production of polyclonal antiserum in rabbits (Cocalico

Biologicals, Inc., Reamstown, PA). Antiserum was purified by

using an Econo-Pac protein A column (Bio Rad) and further

absorbed with a total cell extract from the PhhB-deficient

mutant JS103.


SDS-PAGE and Western Blot Analysis


SDS-PAGE (12% gel) was performed with the Mini-PROTEAN II

cell (Bio-Rad) by the method of Laemmli (1970). Samples of









63

exponential-phase cells were collected by centrifugation, and

the cell pellets were suspended in gel-loading buffer and

heated at 1000C for 10 min. Samples of 5-10 1l were loaded

onto two SDS-polyacrylamide gels. After separation of the

proteins by electrophoresis, one gel was stained with

Coomassie blue and the other gel was used for blotting. When

crude extracts were used, equivalent amounts of protein were

loaded into each lane. Western blots were performed according

to Towbin et al. (1979). The proteins were

electrophoretically transferred onto nitrocellulose membranes

and reacted with the polyclonal antiserum at a dilution of

1:1000. Membranes were then incubated with secondary alkaline

phosphatase-labelled anti-rabbit antibodies at a dilution

(1:30,000) and developed by adding NBT and BICP as chromogenic

substrates (Bibco BRL) for alkaline phosphatase.


Gene Inactivation


Both phhB and phhC were inactivated following the method

described by Song and Jensen (1996). To generate the

truncated 'phhB' fragment (308bp), the upper primer 5'-

ACCCAAGCCCATTGCGAAGCCTGCCG-3', and the lower primer 5'-

GTGCGCGCCGCCATGATGAAATCGTT-3' were used. To generate the

truncated 'phhC' fragment (652bp), the upper primer 5'-

GTCGAGCAGGAAACCACCAAGA-3', and the lower primer 5'-

GTTGGCTACGCAGGTCGGTGAG-3' were used. Interruption of the phhB









64

or phhC gene in a Hgr isolate was confirmed by Southern

hybridization.


Southern Hybridization


Genomic DNA was extracted from the P. aeruginosa phhB-

strain by the method described by Silhavy et al. (1984).

Southern hybridization was performed as described by Sambrook

et al. (1989). The DNA was completely digested with EcoRI,

separated by electrophoresis in 1% agarose gel, and

transferred to a nylon membrane (Bio-Rad). The DNA was fixed

by baking the membrane under vacuum at 800C for 2 hr and

hybridized at 420C overnight with the truncated 'phhB' (the

same as used for gene inactivation) probes that had been

labeled with biotin-14-dATP using a BioNick labelling system

(GIBCO/BRL). The membrane was washed in 2X SSC (300 mM NaCl,

30 mM sodium citrate, pH 7.0) plus 0.1% SDS (twice for 3 min

each time at room temperature), in 0.2X SSC plus 0.1% SDS

(twice for 3 min each time at room temperature), and in 0.16X

SSC plus 0.1% SDS (twice for 15 min each time at 500C). The

probes were detected with the BluGene nonradioactive nucleic

acid detection system (GIBCO/BRL).


RESULTS


PhhB Has 4a-Carbinolamine Dehydratase Activity


PhhB is a homologue of an established 4a-carbinolamine

dehydratase. To confirm that PhhB catalyzes the 4a-









65

carbinolamine dehydratase reaction, I used the phenylalanine

hydroxylase stimulation assay where the utilization of 4a-

carbinolamine limits the rate of the hydroxylation (Huang et

al., 1973; Citron et al., 1992). Either PhhB or DCoH was able

to stimulate the phenylalanine hydroxylase reaction in E. coli

crude extracts where DCoH was used as a postive control (Fig.

3-2), indicating that PhhB protein has 4a-carbinolamine

dehydratase activity. Furthermore, using the expression

construct pJS63, PhhB protein was purified in the laboratory

of Dr. June E. Ayling at the University of South Alabama and

the 4a-carbinolamine dehydratase activity of PhhB was

confirmed by direct assay (personal communication).



Complementation of Tyrosine Auxotrophy by phhA and phhB in
trans


Both phhA and phhB are needed for functional

complementation of E. coli tyrosine auxotrophy (Zhao et al.,

1994) . If phhB functions as both a structural gene and a

regulatory gene in a fashion that parallels the mammalian

homologue, it would be expected to complement in the trans

comfiguration with respect to phhA. A trans-complementation

study was done in which phhA and phhB (or DCoH) were inserted

into two compatible plasmids, pJSll1 and pJZ9-4A (or pGST-

DCoH), respectively. The results (Table 3-2) did indeed show

that phhB was able to complement E. coli tyrosine auxotrophy

in trans with respect to phhA and that mammalian DCoH was able









66







1000




800

\ Control

0600-
S\ \ GST-DCoH


400 -




200 - PhhB



0 I II I I I I I

2 4 6 8 10

Time (min)



FIG. 3-2. Stimulation of phenylalanine hydroxylase activity by
the addition (at the arrow) of a crude extract of E. coli
JP2255 containing PhhB or GST-DCoH fusion protein.
Approximately 1 min after the reaction was started, either
buffer, or 15 Ag of the crude extract containing PhhB, or GST-
DCoH was added. The reaction was monitored at 340 nm for the
oxidation of NADH by dihydropteridine reductase as quinonoid
dihydropterin was recycled to a tetrabiohydropterin (see Fig.
3-1).









67

Table 3-2. Complementation of an E. coli Tyr-a mutant by phhA
and either phhB or DCoH gene in trans

Plasmid(s) Relevent Ability to complementb
genotype E. coli Tyr- mutant

pJS11 phhA' No

pJS12 (phhAB)� Yes

pJZ9-4 phhBt No

pJSll + pJZ9-4 phhA' phhBt Yes

pGST-DCoH DCoH* No

pJSll+pGST-DCoH phhA� DCoIH Yes

a E. coli JP2255 (Phe-Tyr-) mutant was used as the host strain
for this complementation study.
b E. coli JP2255 harboring various plasmids was plated on M9
+ phenylalanine plates supplemented with appropriate
antibiotics as selective agents.









68

to replace PhhB in the bacterial system. The results also

ruled out any possible cis effect of phhB on the expression of

phhA.


Expression of phhA in the Presence or Absence of phhB


The trans-complementation study confirmed that

complementation of E. coli tyrosine auxotrophy by phhA

requires the presence of phhB. To understand whether the

requirement of phhB for the complementation was due to

increased expression of phhA in the presence of PhhB, the

expression of phhA in the presence or absence of phhB in E.

coli JP2255 was studied through Western analysis (Fig. 3-3).

A substantial level of PhhA was still detected when only phhA

was present. The presence of either PhhB or DCoH in trans

indeed increased the expression of phhA, but only by about 2-

fold. Although this result indicated that PhhB may regulate

the expression of phhA, much more than the relatively modest

2-3 fold reduction in PhhA level observed in the phhB- mutant

was expected. This was because little or no PhhA had been

detected in the absence of phhB (pJZ9-5) in E. coli, whereas

a very high level of PhhA was produced in the presence of phhB

(pJZ9-3a) (Zhao et al., 1994). The inconsistency between the

results of this study and that obtained by Zhao et al. (1994)

was found to be due to the incorrectly reported orientation of

the phhA insert in pJZ9-5. Sequencing the DNA insert in pJZ9-5

revealed that the orientation of the phhA gene was opposite to









69
















- ^ - I- PhhA










FIG. 3-3. Western blot analysis of PhhA expression in E. coli
JP2255. Proteins in the whole cell lysates of JP2255 carrying
various plasmids were separated by SDS-PAGE and reacted with
rabbit anti-PhhA polyclonal antibodies. Plasmids containing
the gene(s) shown above are as follows: phhA, pJSl1; phhB,
pJZ9-4; DCoH, pGST-DCoH. pUC18 was used as the control
plasmid.









70

the lac promoter, rather than in the same orientation as

reported by Zhao, et al. (1994).


PhhB Regulates Expression at the Post-transcriptional Level


To understand the regulatory role of PhhB in the

expression of phhA, I constructed phhA'-lacZ transcriptional

fusions in E. coli in both multi-copy form (pJS51Z) and a

single-copy form [X(phhA'-lacZ)] . In both cases, the presence

of PhhB (pJZ9-4) or DCoH (pGST-DCoH) on a second plasmid did

not result in a higher level of 0-galactosidase as compared to

the control (pUC18) (Table 3-3), indicating that neither PhhB

nor DCoH functions at the transcriptional level. I then

constructed the translational fusion phhA'-'lacZ (pJS105Z)

(Table 3-3), and PhhB or DCoH was again provided in trans on

a second plasmid. f-Galactosidase activity increased about

two-fold, indicating that PhhB regulates the expression of

phhA at the translational or post-transcriptional level. This

level of activation by PhhB is consistent with the results

obtained from Western-blot analysis in E. coli, where

similarly modest levels of activation in phhA expression were

observed.


Induction of The phh Operon by Phenylalanine in P. aeruginosa


Expression of both phhA and phhB were coordinately

induced by phenylalanine in P. aeruginosa when grown on

minimal fructose medium (Fig. 3-4). The induction process is









71

Table 3-3. Levels of phhA expression by phhB and DCoH in
trans .

phhB or DCoH 6-Galactosidase Activityb
in trans pJS51Z X(phhA'-lacZ) PJS105Z

pJZ9-4 (PhhB*) 180 13.2 13.1

pGST-DCoH (DCoH�) 191 14.1 10.4

pUC18 (control) 182 15.7 7.3

a Regulation of phhA expression was studied using lacZ as the
reporter gene and P-galactosidase activities in
transcriptional fusions pJS51Z (phhA'-lacZ) (multicopy) and
X(phhA'-lacZ) (single copy), and translational fusion
pJS105Z (phhA'-'lacZ) (multicopy) was assayed in absence
(pUC18) or presence of PhhB (pJZ9-4) and DCoH (pGST-DCoH).
b O-Galactosidase activities are reported in Miller units.









72












1 2 3 4 5 6 7




SPhhA


9.. - PhhB








FIG. 3-4. Induction of the phh operon by phenylalanine
hydroxylase in P. aeruginosa. Proteins in the whole cell
lysates were separated by SDS-PAGE. Lane 1, E. coli JP2255
harboring pJSll and pJZ9-4; lanes 2-7, samples taken after
elapsed times of 0, 10, 30, 60, 90, 120 min, respectively,
following addition of 100 /Lg/ml phenylalanine at zero time.









73

quite protracted and requires 90 minutes or more reach a

maximum. A basal level of PhhB was expressed under non-

inducing conditions as seen in lane 2 on the Western blot. It

is unknown whether this is due to expression from a promoter

upstream of phhA or from an internal promoter within the

coding sequence of phhA. Although the latter interpretation

comes to mind because no PhhA band is visualized at zero time,

this could be due to differing sensitivities of the antibody

probes.


Effect of phhB Knockout in P. aeruginosa


The phhB gene in P. aeruginosa was inactivated by

chromosomal insertion of the suicide plasmid pUFR/'phhB'/Hgr

through a single homologous crossover event (Fig. 3-5A). The

insertional inactivation of the phhB gene in the resulting

mutant was confirmed by Southern blot analysis (Fig. 3-5B).

The expression of PhhA in a phhB- background was examined

by Western-blot methodology under fully induced conditions

fostered by growth on LB media (Fig. 3-6). The PhhA level

observed in the phhB- mutant was reduced about 2-3 fold

compared to the wildtype parent. No PhhA was detected in

phhA- mutant (negative control) . The PhhB level was also

checked at the same time. A low basal level of PhhB was

detected in phhA- mutant when compared with that in the

wildtype. No PhhB was detected in phhB- mutant (negative

control).





















FIG. 3-5. Inactivation of phhB in P. aeruginosa. (A) Schematic
representation for insertional inactivation of the chromosomal
phhB gene by the integration of the suicide plasmid
pUFR/'phhB'/Hgr through a single homologous crossover. The
resulting Hgr mutant does not contain a complete copy of phhB
gene, but instead has two truncated copies of phhB gene. (B)
Southern-blot analysis of chromosomal DNA from P. aeruginosa
PAO-1 wildtype (lane 1) and mutant JS103 (lane 2). Chromosomal
DNA was completely digested with EcoRI and probed at high
stringency with the truncated 'phhB' fragment of
pUFR/'phhB'/Hgr.







75



(A) phhB P. aeruginosa

X chromosome




lacla

Onr
Mofb phhHg Suicide plasmid







SphhB' H r 'phhBphh-
mutant

(B) 1 2









76


















PhhA -o .












PhhB






FIG. 3-6. Western blot analysis of PhhA and PhhB expression
in P. aeruginosa PAO strains. Proteins in whole cell lysates
were separated by SDS-PAGE and probed with rabbit anti-PhhA or
anti-PhhB polyclonal antibodies.









77

The physiological effect of the phhB knockout mutant in

P. aeruginosa was examined. Inactivation of phhB abolished the

ability to grow on either phenylalanine or tyrosine as the

sole carbon source. However, interpretation of this result is

complicated by results obtained with a phhC knockout mutant

which was also not able to grow on either phenylalanine or

tyrosine as the sole carbon source. Since insertion of the

suicide plasmid into the chromosome of P. aeruginosa is

expected to create polar effects on the downstream genes in

the operon (as indeed seen in phhA knockout where amount of

PhhB expressed was dramatically decreased), it seems probable

that the physiological effect observed in phhB knockout mutant

is due to the polar effect on the expression of phhC.



Overexpression of PhhA and PhhB Proteins


Although PhhA was expressed to detectable levels in both

E. coli (Fig. 3-3) and P. aeruginosa (Fig. 3-6) in the absence

of phhB, initial attempts to express PhhA at high levels in

the absence of PhhB in E. coli were unsuccessful. I then

employed a T7 overexpression system (see Methods) to express

PhhA in E. coli BL21(DE3) under induction conditions not

requiring growth (Fig. 3-7A). When phhA was expressed from a

native ribosomal binding site in pJS72, high PhhA levels were

produced after IPTG induction for 3 h (Lane 3, Fig. 3-7B).

When phhA was expressed from 10 translational signals

(pJS95), higher levels of PhhA was made after IPTG induction







78





(A)
RBS -- PhhA
pJS72 FP[- CATIGGAG TCCGTATG AAAACG...
Native translational signal M K T


RBS I - PhhA
pJS95 [P0 lGAAGGAG ATATACATATG AAAACG...
T7 translational signal M K T


(B)
1 2 3 4 5
97.4 -
66.2
45.0 - --
31.0 --
21.5 -
14.5 -





FIG. 3-7. Expression of PhhA using T7 expression system in E.
coli. (A) Construction of PhhA expression plasmids, pJS72
(with the native translational signals) and pJS95 (with 010
translational signals). (B) SDS-PAGE analysis of expressed
PhhA protein. Proteins in whole cell lysate were separated by
SDS-PAGE and stained with Coomassie blue. Lane 1, molecular
weight markers; lanes 2&3, BL21(DE3) harboring pJS72, before
and after 1 mM IPTG induction for 3 h at 300C, respectively;
lanes 4&5, BL21(DE3) harboring pJS95, before and after 1 mM
IPTG induction for 3 h at 300C, respectively.









79

for 3 h (Lane 5, Fig. 3-7B). PhhA produced using this T7

overexpression system in E. coli is active. Phenylalanine

hydroxylase activity was assayed in crude extract of E. coli

BL21(DE3) harboring either pJS72 or pJS95 after IPTG

induction, and compared with activity present in E. coli

JP2255 (pJS9-3a). Expression of phhA from 010 translational

signals resulted in over a five-fold increase in phenylalanine

hydroxylase activity (Table 3-4).

PhhB protein was expressed very well from a lac promoter

in pJZ9-4 by Zhao et al.(1994). We obtained an even higher

level of PhhB by using the T7 overexpression system (Fig. 3-

8). The pJS10 construct used contains both phhA and phhB.

After IPTG induction, only PhhB was overexpressed and little

PhhA was made (Lane 3, Fig.3-8) . A similar phenomenon was

observed with pJS63 construct containing both phhB and phhC

where only PhhB protein was overproduced (Lane 5, Fig. 3-8).

These results indicated that phhB has stronger translational

signals and, therefore, is preferably translated over phhA and

phhC. PhhB protein expressed is fully active, and has been

purified and characterized from the latter construct by Ayling

(unpublished results).


Expression of PhhA without PhhB in E. coli Has Growth
Inhibitory Effects


Even though a low level of PhhA was produced from pJS11

in the absence of PhhB, it did not complement the E. coli

tyrosine auxotrophy in JP2255. A possible explanation was that









80

Table 3-4. Phenylalanine hydroxylase activities in different
expression clones'

Expression Specific Activity
clones (nanomoles/min/mg)

JP2255/pJZ9-3a 91.7

BL21(DE3)/pJS72 72.6

BL21(DE3)/pJS95 379.2

a Cells of E. coli JP2255 harboring pJZ9-3a were grown in LB
broth at 370C and harvested at late exponential phase; cells
of BL21(DE3) harboring pJS72 or pJS95 were grown in LB broth
at 370C to O.D=1 and induced for 3 hr by addition of 1 mM
IPTG. Crude extracts were used as the enzyme sources.









81










1 2 3 4 5

97.4
66.2

45.0

31.0

21.5 a
14.5 M*4 PhhB








FIG. 3-8. Expression of PhhB using T7 expression system in E.
coli. Proteins in whole cell lysates were separated by SDS-
PAGE and stained with Coomassie blue. Lane 1, molecular weight
markers; Lanes 2&3, BL21(DE3)/pJSlO, before and after 1 mM
IPTG induction for 3 h at 300C, respectively. Lanes 4&5,
BL21(DE3)/pJS63, before and after 1 mM IPTG induction for 3 h
at 300C, respectively.









82

the PhhA produced from pJSll was not high enough to result in

complementation. Thus, attempts were made to construct clones

from which higher levels of PhhA could be produced to see

whether complementation of E. coli tyrosine auxotrophy would

then occur. Two constructs pJS96 and pJS97 were made with a

XbaI fragment from pJS95 carrying phhA gene (fused with 010

translational signals) cloned into pUC18 behind a lac promoter

or into pTrc99A behind a trc promoter, respectively (Fig. 3-

9). Both plasmids were found to be unstable and cells tended

to lose the plasmid, especially with pJS96 where phhA is

constitutively expressed at a very high level from the lac

promoter. An elapsed time of several days was required to see

pinpoint colonies with pJS96. E. coli carrying pJS97 was able

to grow to a pinpoint colony overnight at 370C without IPTG

induction, while it took two or more days to see the pinpoint

colonies on the plate with IPTG induction. These results

indicated that high level of phhA expression in the absence of

phhB triggers potent growth inhibition. Although the

expression of phhA from the trc promoter on pJS97 has to be

induced because of the presence of a laclI gene on the

plasmid, a high level of PhhA is still produced without IPTG

induction because the trc promoter is very strong and leaky.

Because PhhA was constitutively expressed from pJS96 at

high level, the cells were not able to maintain the plasmid.

Therefore, a high level of PhhA could not be overproduced in

E. coli (pJS97) (Lane 1, Fig. 3-9B). Since a lower level of








83




(A)
"uS I- PhhA
pJS96 lac IGAAGGA GATATACATATGAAAACG...
T7 translational signal M K T


RBS F--PhhA
pJS97 J]lG AAGGAGATATACATATGAAAACG...
T7 translational signal M K T

(B) 1 2 3 4 5

97.4
66.2

W 145.0


S ....... 21.5

___ 14.5





FIG. 3-9. Expression of PhhA in E. coli JP2255. (A)
Construction of PhhA expression plasmids, pJS96 and pJS97.
(B) SDS-PAGE analysis of PhhA expression. Proteins in whole
cell lysates were separated by SDS-PAGE and stained with
Coomassie blue. Lane 1, JP2255/pJS96; lane 2, JP2255/pTrc99A
(control) ; lanes 3&4, JP2255/pJS97, before and after 1 mM IPTG
induction for 3 h, respectively; lane 5, molecular weight
markers.









84

PhhA was produced from pJS97 due to the presence of lacIq gene

one the plasmid, the cells carrying pJS97 were better able to

maitain the plasmid, thus being able to overproduce PhhA after

IPTG induction (Lane 4, Fig. 3-9B) . Neither pJS96 nor pJS97 by

itself was able to complement E. coli tyrosine auxotrophy.

However, they were able to complement the auxotrophy when phhB

was provided in trans on pJZ9-4 (data not shown). This result

indicates that PhhB was able to remove the inhibitory effect

imposed by overproduction of PhhA on the host cells. To assure

that apparent growth inhibition was not due to excessive

conversion of phenylalanine to tyrosine in the phenylalanine

auxotrophy background of strain JP2255, the plasmids were

moved to a prototrophic background. E. coli DH5a carrying

either plasmid was found to develop only pinpoint colonies on

LB + Amp plates without IPTG induction, confirming that

expression of phhA at higher levels created a general

inhibitory effect on the growth of E. coli.



Discussion


Regulatory Role of PhhB?


The mammalian PhhB homolog, DCoH, has both catalytic and

regulatory functions. It was initially thought that P.

aeruginosa phhB exerted an essential positive role in

expression of phhA since constructs lacking phhB did not

express phhA, as monitored by SDS-PAGE (Zhao et al., 1994).









85

Although phhA is in fact expressed in the absence of phhB,

phhB does appear to exercise a positive regulatory role in the

expression of phhA with a relatively modest effect of perhaps

2-3 fold. The transcriptional fusion approach seems to have

eliminated the possibility of this regulation being at the

transcriptional level, with the reservation that the

experiments were performed in an E. coli background. I have

shown that PhhR, the positive regulator of the phh operon in

P. aeruginosa, does not interact properly with the E. coli o54

machinery to activate phhA expression. But it is still

possible that PhhB interacts with PhhR as a co-activator

entity in the native P. aeruginosa organism. This would amount

to a parallel with the mammalian system where DCoH is a co-

activator for HNFIa (an upstream enhancer element that can be

considered comparable to PhhR).

However, I have presented results to show that PhhB does

enhance expression of PhhA at the post-transcriptonal level in

E. coli. The 2-3 fold enhancement effects obtained correspond

with the similar 2-3 fold magnitude of effect seen in P.

aeruginosa, when comparing Western blots of PhhA in PhhB' and

PhhB- backgrounds. If PhhB regulates at a post-transcriptional

level, several possibilities envisioned include: (i) the

sencondary structure of phhA mRNA might mask the ribosomal

binding site. A stem-loop structure has been located in this

area. Binding of PhhB in this region might disrupt the

secondary mRNA structure and enhance translational initiation.









86

(ii) PhhB might bind phhA mRNA in the 3'region and protect

against nuclease-catalyzed mRNA degradation. (iii) PhhB may

complex with PhhA. This complex may protect PhhA from

proteolysis. I have in fact obtained preliminary evidence for

a PhhA-PhhB complex.


Rationale for Positive Regulatory Role of PhhB


When PhhA is highly expressed in the absence of PhhB, the

E. coli host cells become subject to drastic growth

inhibition. The reasonable explanation for this effect of

PhhB is that removal of the inhibitory effect generated by

PhhA reaction is direct result of 4a-carbinolamine

dehydratase activity. It is known that in the absence of 4a-

carbinolamine dehydratase activity, a 7-isomer of

tetrahydrobiopterin is generated during the reaction catalyzed

by phenylalanine hydroxylase (Davis et al., 1992), and this 7-

isomer is a potent inhibitor of phenylalanine hydrxoylase in

the mammalian sytem. Since potent growth inhibition persists

in wildtype E. coli backgrounds where overexpressed

phenylalanine hydroxylase has no purpose, inhibition cannot be

attributed to inhibition of phenylalanine hydroxylase by the

7-isomer. Two explanations accounting for general growth

inhibition come to mind. (i) The reduced pterin cofactor of E.

coli may be depleted because PhhB is not present for

recycling. Consequently, some pterin-dependent enzymes that

are essential for growth may become limiting. (ii)









87

Alternatively, the 7-isomer may be a potent inhibitor of a

pterin-dependent enzyme needed for growth. Possible targets of

inhibition could be dihydropteridine reductase or

dihydrofolate reductase. The E. coli dihydropteridine

reductase has been reported to possess broad specificity for

pteridine compounds (Vasudevan et al.,1992).

When phhA was expressed at relatively low levels in E.

coli no growth inhibition, or at least no severe growth

inhibition occured. However, this PhhA was evidently not

functional in vivo ( in the absence of PhhB) because

complementation of tyrosine auxotrophy in the presence of

exogenous L-phenylalanine was unsuccessful. On the other hand,

the joint presence of PhhA and PhhB readily allowed functional

complementation. I concluded that PhhA is an essential target

of the 7-isomer. At the low levels of PhhA expression in the

absence of PhhB, the 7-isomer is generated and inhibits PhhA

function--but not enough 7-isomer is produced to cause general

growth inhibiton. At high levels of PhhA expression in the

absence of PhhB, sufficient 7-isomer is produced to inhibit

one or more enzymes essential for the growth.

With the above background, a rationale to explain a basis

for selection of regulation of PhhA by PhhB is apparent. If it

is correct that the 7-isomer generated from the carbinolamine

pterin product of the PhhA reaction has general antimetabolite

properties, then the significance of PhhB goes beyond its









88

catalytic capability. It also diverts the carbinolamine

substrate from an undesirable nonenzymatic fate.















REFERENCES


Abita, J. P., Milstien, S., Chang, N., and Kaufman, S. 1976.
In vitro activation of rat liver phenylalanine
hydroxylase by phosphorylation. J. Biol. Chem. 251:5310-
5314.

Argaet, V.P., Wilson, T.J., and Davidson, B.E. 1994.
Purification of the Escherichia coli regulatory protein
TyrR and analysis of its interactions with ATP, tyrosine,
phenylalanine and Tryptophan. J. Biol. Chem. 269:5171-
5178.

Armarego, W. L., Randles, D., and Waring, P. 1984.
Dihydropteridine reductase (DHPR), its cofactors, and its
mode of action. Med. Res. Rev. 43:267-321.

Bailey, S. W., Boerth, S. R., Dillard, S. B., and Ayling, J.
E. 1993. The mechanism of cofactor regeneration during
phenylalanine hydroxylation. Adv. Exp. Med. Biol. 338:47-
54.

Baldwin, G.S., and Davidson, B.E. 1981. A kinetic and
structural comparison of chorismate mutase/prephenate
dehydratase from mutant strains of Escherichia coli K-12
defective in the pheA gene. Arch. Biochem. Biophys.
211:66-75.

Carr, R. T. and Benkovic, S. J. 1993. An examination of the
copper requirement of phenylalanine hydroxylase from
Chromobacterium violaceum. Biochemistry. 32:14132-14138.

Carr, R. T., Balasubramanian, S. Hawkins, P. C., and
Benkovic, S. J. 1995. Mechanism of metal-independent
hydroxylation by Chromobacterium violaceum phenylalanine
hdyroxylase. Biochemistry. 34:7525-7532.

Chang, A.C.Y., and Cohen, S.N. 1978. Construction and
characterization of amplifiable multicopy DNA cloning
vehicles derived from the pl5A cryptic miniplasmid. J.
Bacteriol. 134:1141-1156.





89









90

Citron, B. A., Davis, M. D., Milstien, S., Gutierrez, J.,
Mendel, D. B., Crabtree, G. R., and Kaufman, S. 1992.
Identity of 4a-carbinolamine dehydratase, a component of
the phenylalanine hydroxylation system, and DCoH, a
transregulator of homeodomain proteins. Proc. Natl. Acad.
Sci. USA. 89:11891-11894.

Cui, J., and Somerville, R.L. 1993. The TyrR protein of
Escherichia coli: analysis by limited proteolysis of
domain structural and ligand-mediated conformational
changes. J. Biol. Chem. 268:5040-5047.

Curtius, H. C., Adler, C., Rebrin, I., Heizmann, C., and
Ghisla, S. 1990. 7-Substituted pterins: formation during
phenylalanine hydroxylation in the absence of
dehydratase. Biochem. Biophys. Res. Commun. 172:1060-
1066.

Davis, M. D. and Kaufman, S. 1989. Evidence for the formation
of the 4a-carbinolamine during the tyrosine-dependent
oxidation of tetrahydrobiopterin by rat liver
phenylalanine hydroxylase. J. Biol. Chem. 264:8585-8596.

Davis, M. D. and Kaufman, S. 1991. Studies on the partially
uncoupled oxidation of tetrahydropterins by phenylalanine
hydroxylase. Neurochem. Res. 16:813-819.

Davis, M. D., P. Ribeiro, J. Tipper, and S. Kaufman. 1992.
"7-tetrahydrobiopterin," a naturally occourring analogue
of tetrahydrobiopterin, is a cofactor for and a potential
inhibitor of the aromatic amino acid hydroxylases. Proc.
Natl. Acad. Sci. USA. 89:10109-10113.

Davis, M. D., Parniak, M. A., Kaufman, S., and Kemper, E.
1996. Structure-function relationships of phenylalanine
hydroxylase revealed by rediation target analysis. Arch.
Biochem. Biophys. 325:235-241.

Decicco, B. T. and Umbreit, W. W. 1964. Utilization of
aromatic amino acids by Hydrogenomonas facilis. J.
Bacteriol. 83:1590-1594.

DeFeyter, R., Kado, C. I., and Gabriel, D. W. 1990. Small,
stable shuttle vectors for use in Xanthomonas. Gene
88:65-72.

Devereux, J., P. Haeberli, and 0. Smithies. 1984. A
comprehensive set of sequence analysis programs for the
VAX. Nucleic Acids Res. 12:387-395.









91

Dilella, A. G., Kwok, S. C. M., Ledley, F. D., Marvit, J., and
Woo, S. L. C. 1986. Molecular structure and polymorphic
map of the human phenylalanine hydroxylase gene.
Biochemistry. 25:743749.

Endrizzi, J. A., Cronk, J. D., Wang, W., Crabtree, G. R.,
and Alber, T. 1995. Crystal structure of DCoH, a
bifunctional, protein-binding transcriptional
coactivator. Science. 268:556-559.

Essar, D. W., Eberly, L., Han, C.-Y., and Crawford, I. P.
1990. DNA sequences and characterization of four early
genes of the tryptophan pathway in Pseudomonas
aeruginosa. J. Bacteriol. 172:853-866.

Faust, D. M., Catherin, A. M., Barbaux, S., Belkadi, L.,
Imaizumi-Sherrer, T., and Weiss, M. C. 1996. The activity
of the highly inducible mouse phenylalanine hydroxylase
gene promoter is dependent upon a tissue-specific,
hormone-inducible enhancer. Mol. Cell. Biol. 16:3125-
3137.

Friedman, D.I. 1988. Integration host factor: a protein for
all reasons. Cell 55:545-554.

Friedrich, B. and H. G. Schlegel. 1972. Die hydroxylierung
von phenyalanin durch Hydrogenomonas eutrophus H16. Arch.
Mikrobiol. 83:17-31.

Gambill, D. B., and Summers, A. 0. 1985. Versatile mercury-
resistant cloning and expression vectors. Gene 39:293-297

Gu, W., and Jensen, R.A. 1996. Evolutionary recruitment of
biochemically specialized subdivisions of Family I within
the protein Superfamily of aminotransferases. J.
Bacteriol. 178:2161-2171.

Gudinchet, F., Maeder, P., Meuli, R. A., Deonna, T., and
Mathieu, J. M. 1992. Cranial CT and MRI in malignant
phenylketonuria. Pediatr. Radiol. 22:223-224.

Guldberg, P., H. L. Levy, W. B. Hanley, R. Kock, R. Matalon,
B. M. Rouse, F. Trefz, F. de la Cruz, K. F. Henriksen,
and F. Guttler. 1996. Phenylalanine hydroxylase gene
mutations in the United States: report from the Maternal
PKU Collaborative Study. Am. J. Hum. Genet. 59:84-94.

Guroff, G. and T. Ito. 1963. Induced soluble
phenylalanine hydroxylase from Pseudomonas sp. grown on
phenylalanine or tyrosine. Biochim. Biophys. Acta
77:159-161.









92

Haldenwang, W. G. 1995. The sigma factors of Bacillus
subtilis. Microbiol. Rev. 59:1-30.

Heatwole, V.M., and Somerville, R.L. 1991. The tryptophan-
specific permease gene, mtr, is differentially regulated
by the tryptophan and tyrosine repressors in Escherichia
coli K-12. J. Bacteriol. 173:3601-3604.

Holloway, B.W. (1955) Genetic recombination in Pseudomonas
aeruginosa. J. Gen. Microbiol. 13:572-581.

Huang, C. Y., Max, E. E., and Kaufman, S. 1973. Purification
and characterization of phenylalanine hydroxylase-
stimulating protein from rat liver. J. Biol. Chem.
248:4235-4241.

Ivey-Hoyle, M., Steege, D.A. 1992. Mutational analysis of an
inherently defective translation initiation site. J. Mol.
Biol. 224:1039-1054.

Jakoby, G. A. 1964. The induction and repression of amino
acid oxidation in Pseudomonas fluorescens. Biochem. J.
92:1-8.

Joerger, R.D., Jacobson, M.R., and Bishop, P.E. 1989. Two
nifA-like genes required for expression of alternative
nitrogenases by Azotobacter vinelandii. J. Bacteriol.
171:3258-3267.

Kamoun, S., Kamdar, H. V., Tola, E., and Kado, C. I. 1992.
Incompatible interactions between crucifers and
Xanthomonas campestris invole a vascular hypersensitive
response: role of the hrpX locus. Mol. Plant-Microbe
Interact. 5:22-23.

Kappock, T. J., Hartins, P. C., Friedenberg, S., and
Caradonna, J. P. 1995. Spectroscopic properties of
unphosphorylated rat hepatic phenylalanine hydroxylase
expressed in Escherichia coli. Comparison of resting and
activated states. J. Biol. Chem. 270:30532-30544.

Kaufman, S. 1975. Studies on the mechanism of phenylalanine
hydroxylase:detection of an intermediate, p.495-523. In
W. Pfleidere (ed.), Chemistry and biology of pteridines.
Walter de Gruyter, Berlin.

Kaufman, S. 1986. Regulation of the activity of hepatic
phenylalanine hydroxylase. Adv. Enz. Regulation. 25:37-
64.




Full Text
88
catalytic capability. It also diverts the carbinolamine
substrate from an undesirable nonenzymatic fate.


42
units where TyrR functions as an activator, L-phenylalanine
functions as an essential co-activator (Pittard and Davidson,
1991).
Discussion
Anomalous Repression of the nhh Operon by PhhR in E. coli
PhhR can mimic the ability of TyrR to repress the aroF-
tyrA operon at a a70 promoter. This indicates that PhhR can
recognize TyrR boxes and is consistent with the high
similarity of the helix-turn-helix, DNA-binding domain within
the carboxy-terminal segments of TyrR and PhhR. However, PhhR
was unable to activate the phh operon in the heterologous E.
coli background, suggesting an incompatibility between the E.
coli RpoN and the P. aeruginosa a54-dependent system. The
expression of PhhA from a promoter recognized by E. coli
upstream of the native a54 promoter was in fact severely
depressed in constructs containing phhR, even in the presence
of added co-activator (L-phenylalanine). In the presence of P.
aeruginosa PhhR, an aberrant complex apparently blocks
transcription initiated upstream of the a54 promoter.
Emerging Subfamilies within the a54 Enhancer-Binding Protein
Family
P. aeruginosa PhhR belongs to an outlying subgroup (which
we denote subfamily a in Fig. 2-8) of the a54 enhancer-binding
protein family. All members of the family possess in common a


34
FIG. 2-7. Functional replacement of TyrR by P. aeruginosa
PhhR in E. coli, as monitored by sensitivity to growth
inhibition by m-fluoro-tyrosine (MFT) (A) E. coli tyrR+
(wildtype) strain SP1312 is very sensitive to MFT present on
a central disc, exhibiting a large zone of growth inhibition;
(B) E. coli tyrR~ strain SP1313 is insensitive to MFT, showing
no zone of growth inhibition; (C) P. aeruginosa phhR(pJS91) in
trans complements E. coli tyrR~ and restores the sensitivity
to MFT, as visualized by a zone of growth inhibition.


60
crude extract containing PhhB or GST-DCoH was added. The
reaction was monitored at 340 nm for the oxidation of NADH by
dihydropteridine reductase as quinonoid dihydropterin was
recycled to tetrahydropterin.
Construction of phhA:lacZ Transcriptional and Translational
Fusions
For the transcriptional fusion (phhA'-lacZ) the HincII-
BamHI fragment containing the upstream region of phhA was
first cloned into pACYC177 that had been digested with Hindi
and BamHI, creating pJS51. A BamHI-cassette of a promoterless
lacZ gene from the plasmid Z1918 was then inserted at the
BamHI site of pJS51 in the same orientation as phhA to create
pJS51Z, which was used as a low-copy phhA'-lacZ fusion. A
single-copy fusion X(phhA'-lacZ) was obtained by transferring
the phhA'-lacZ fusion from pJS51Z into ARZ5 following the
procedure described by Yu and Reznikoff (1984) .
For the translational fusion (phhA'-'lacZ) the Hindi-
Bam HI fragment containing the upstream region of phhA was
generated by PCR with the upper primer 5'-
GACAGAGCAGGTAGATGGCGTT-3' and the lower primer
5'GGGATCCGGCTCGTGGGGCAGGCCGA-3' (BamHI site underlined). An
extra guanine nucleotide (G in bold) was added in the lower
primer to create the frameshift needed for an in-frame fusion
at the BamHI site to generate phhA' lacZ. The Hindi-BamHI
fragment with the frameshift was inserted into the HincII-
BamHI site of pACYC177 to create pJS105, and a BamHI-cassette


27
0 -
a PC
a: 3
a a
X
5! a;
I I!
phhA phhB
%
phhC
H
8
5
o
pbpG
B
?
Si
-f ^ 1
FIG. 2-4. Physical map of the DNA fragment containing phh
structural genes, the divergently-transcribed regulatory gene
phhR, and the downstream penicillin-binding protein gene
(pbpG) in pJZ9. Terminators downstream of phhC and phhR are
indicated. The shaded bars at both ends are portions of the
multiple cloning site of the pUC18 vector. The location of
restriction sites is shown.


54
(A)
a -
1 =
I
l l
¡U 5:
phhR
-c -J-
phhA phhB phhC
_ P
I
3 *
(B)
TETRAHYDROBIOPTERIN
4a-CARBINOLAMINE
QUTNOID DTHYDROBIOPTERTN
FIG. 3-1. (A) Physical map of the phh operon in Pseudomonas
aeruginosa. The endonuclease restriction sites are shown at
the top. The arrows indicate the position of the genes and the
directions of transcription. Putative transcriptional
terminators (t inside a circle) are indicated. The proteins
encoded by the genes are as follows: phhR, a54 transcriptional
activator of the phh operon; phhA, phenylalanine hydroxylase;
phhB, 4a-carbinolamine dehydratase; and phhC, aromatic
aminotransferase. (B) Regeneration of the pterin cofactor for
phenylalanine hydroxylase. The enzymes involved are indicated
as follows: PhhA, phenylalanine hydroxylase; PhhB, 4a-
carbinolamine dehydratase; and DHPR, dihydropteridine
reductase.


83
RBS
PhhA
pJS96|P/flcl|GAAGGAG|TATACATATGAAAACG...
T7 translational signal MKT
RBS
'PhhA
p.IS97lPfr?l|GAAGGAOA|TATACATATOAAAACG...
T7 translational signal
M K
(B) 1 2 3 4 5
97.4
66.2
45.0
31.0
21.5
14.5
FIG. 3-9. Expression of PhhA in E. coli JP2255. (A)
Construction of PhhA expression plasmids, pJS96 and pJS97.
(B) SDS-PAGE analysis of PhhA expression. Proteins in whole
cell lysates were separated by SDS-PAGE and stained with
Coomassie blue. Lane 1, JP2255/pJS96; lane 2, JP2255/pTrc99A
(control); lanes 3&4, JP2255/pJS97, before and after 1 mM IPTG
induction for 3 h, respectively; lane 5, molecular weight
markers.


9
HNF1 die with a marked liver enlargement. The gene coding for
phenylalanine hydroxylase is totally silent, thus giving rise
to phenylketonuria (Pontoglio et al., 1996).
Little information is available about the regulation of
phenylalanine hydroxylase in bacteria. However, some evidence
has indicated that the bacterial phenylalanine hydroxylase is
also regulated. In Pseudomonas acidovorans, a higher level of
phenylalanine hydroxylase was found after growth in the
presence of phenylalanine (Wiliams et al, 1976) Induction of
both phenylalanine hydroxylase and tryptophan hydroxylase in
the presence of their substrates was also reported in C.
violaceum (Letendre et al, 1974).
The most extensively characterized microbial
phenylalanine system is that of P. aeruginosa. Whether this
system is subject to any regulatory controls has not been
studied prior to this work. The initial report of Zhao et al.
(1994) provided a strong basis for anticipation that the phh
operon would be subject to regulation for the following
reasons, (i) The closely spaced organization of the three
structural genes (phhABC) in an apparent operon implies
regulation. (ii) Analysis of effects of the presence or
absence of regions immediately flanking the phh operon upon
expression of phenylalanine hydroxylase indicated the likely
location there of one or more regulatory genes. (iii) The
reported lack of phhA expression in the absence of phhB


10
suggested a positive regulatory role of phhB in addition to
its catalytic function.
The major objectives of this study have been to elucidate
the physiological conditions under which regulation occurs, to
identify and characterize at the molecular-genetic level any
regulatory genes which control the phh operon, and to
determine the nature of the apparent positive regulatory
action of phhB.


61
of truncated 'lacZ from pMC1871 was inserted into pJS105 to
create the translational fusion plasmid, pJS105Z.
ff-Galactosidase Assay
/3-Galactosidase activity was assayed under conditions of
proportionality as described by Miller (1972), and specific
activities are expressed in Miller units. The data are the
results of at least two independent assays.
Construction of PhhA and PhhB Expression Vectors
To express PhhA protein, expression plasmids pJS72 and
pJS95 were constructed. A PCR fragment containing the
complete coding region of phhA and the native ribosome-binding
site (RBS) was amplified with the upper primer 5'-
CATGGAGTCCGTATGAAAACGACGCA-3' (RBS underlined; ATG start codon
in bold) and the downstream primer 5'-
CTTGGTTGTCGCATGTGGGAGCGGCG-3', and cloned into pET23 behind
the T7lac promoter to create pJS72. pJS95 was constructed by
inserting the coding region of phhA into the translational
fusion vector pETllb. The coding region was amplified by PCR
with the upstream primer 5'-CCATATGAAAACGACGCAGTACGTG-3' and
the downstream primer 5'-CAAGTCTGGTTGTCGCATGTGGGAGCGGCG-3'.
The upper primer was made with a built-in Ndel site
(underlined) which allows fusion of phhA at the translational
start site (ATG in bold) with the T7 translational initiation
signals. To constitutively express the PhhA protein, the


18
compatible with most commonly used plasmids using a ColEI
origin of replication, and it has low-copy number (about 20
copies/cell) High level of PhhR produced from a high-copy
number plasmid was found to be toxic to the host cells. The
Bglll-BairiEI fragment carrying phhR gene was excised from the
expression plasmid pJS88 and cloned into the BamHI site of
pACYC184, thereby interrupting the tetracycline resistance
gene (Tcr) (Fig.2-3A).
Evaluation of Sensitivitv/Resistance to m-FIuoro-tyrosine
Three E. coli strains, SP1312 (tyrR+) SP1313 (tyrR~)
carrying pJS91 [phhR*), and SP1313 (tyrR~) carrying pACYC184
(phhR") were compared for sensitivity to m-fluoro-tyrosine
(MFT). All three strains were first grown in M9 medium with
appropriate antibiotics up to late-exponential phase of growth
and then used to swab M9 agar plates containing appropriate
antibiotics. A sterile Difco concentration disk (0.6 cm) was
positioned at the center of each plate, and 10 n 1 of 50 fig/ml
m-fluoro-tyrosine was applied onto the disks. The plates were
then incubated at 37C for 24 hours.
Construction of vhhA'-lacZ and vhhR'-lacZ Transcriptional
Fusions
To compare levels of phhA transcription in both pJZ9 and
pJZ9-3a, plasmids pJS61Z and pJS62Z were constructed,
respectively. These have a promoterless lacZ gene (from


82
the PhhA produced from pJSll was not high enough to result in
complementation. Thus, attempts were made to construct clones
from which higher levels of PhhA could be produced to see
whether complementation of E. coli tyrosine auxotrophy would
then occur. Two constructs pJS96 and pJS97 were made with a
Xbal fragment from pJS95 carrying phhA gene (fused with 010
translational signals) cloned into pUC18 behind a lac promoter
or into pTrc99A behind a trc promoter, respectively (Fig. 3-
9). Both plasmids were found to be unstable and cells tended
to lose the plasmid, especially with pJS96 where phhA is
constitutively expressed at a very high level from the lac
promoter. An elapsed time of several days was required to see
pinpoint colonies with pJS96. E. coli carrying pJS97 was able
to grow to a pinpoint colony overnight at 37C without IPTG
induction, while it took two or more days to see the pinpoint
colonies on the plate with IPTG induction. These results
indicated that high level of phhA expression in the absence of
phhB triggers potent growth inhibition. Although the
expression of phhA from the trc promoter on pJS97 has to be
induced because of the presence of a lacT1 gene on the
plasmid, a high level of PhhA is still produced without IPTG
induction because the trc promoter is very strong and leaky.
Because PhhA was constitutively expressed from pJS96 at
high level, the cells were not able to maintain the plasmid.
Therefore, a high level of PhhA could not be overproduced in
E. coli (pJS97) (Lane 1, Fig. 3-9B). Since a lower level of


96
Williams, C. D., Dickens, G. Letendre, C. H., Guroff, G.,
Haines, C., and Shiota, T. 1976. Isolation and
characterization of dihydropteridine reductase from
Pseudomonas species. J. Bacteriol. 127:1197-1207.
Xia, T., and Jensen, R.A. 1990. A single cyclohexadienyl
dehydrogenase specifies the prephenate dehydrogenase and
arogenate dehydrogenase components of the dual pathways
to L-tyrosine in Pseudomonas aeruginosa. J. Biol. Chem.
265:200333-20036.
Yang, J. Ganesan, S., Sarsero, J. and Pittard, A.J. 1993.
A genetic analysis of various functions of the TyrR
protein of Escherichia coli. J. Bacteriol. 175:1767-1776.
Yanisch-Perron, C., Vieira, J., and Messing, J. 1985. Improved
M13 phage cloning vectors and host strains: nucleotide
sequences of the M13mpl8 and pUC vectors. Gene 33:103-
119 .
Yu, X.-M. and Reznikoff, W. S. 1984. Deletion analysis of the
CAP-cAMP binding site of the Escherichia coli lactose
promoter. Nucleic Acids Res. 12:5449-5464.
Zhao, G. Xia, T., Aldrich, H. and Jensen, R.A. 1993.
Cyclohexadienyl dehydratase from Pseudomonas aeruginosa
is a periplasmic protein. J. Gen. Microbiol. 139:807-813.
Zhao, G., Xia, T., Song, J., and Jensen, R. A. 1994.
Pseudomonas aeruginosa possesses homologues of mammalian
phenylalanine hydroxylase and 4a-carbinolamine
dehydratase/DCoH as part of a three-component gene
cluster. Proc. Natl. Acad. Sci. USA 91:1366-1370.


MULTI-LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa
By
JIAN SONG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1997


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E2SDS6NJ0_DS22NS INGEST_TIME 2015-02-19T20:52:49Z PACKAGE AA00028759_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


MULTI-LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa
By
JIAN SONG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1997

Copyright 1997
by
Jian Song

Dedicated to my father and my mother,
whose love, care, and encouragement make it possible
for me to complete this dissertation

ACKNOWLEDGMENTS
I wish to express my deep and sincere gratitude to Dr.
Roy A. Jensen, chairman of my supervisory committee, whose
invaluable guidance, constant encouragement, endless ideas,
critical input, and financial support made the fulfillment of
this study possible.
I would also like to thank Dr. Dean W. Gabriel, Dr.
Lonnie 0. Ingram, Dr. James F. Preston, and Dr. Keelnatham T.
Shanmugam for their help, encouragement, advice, and critical
review of the dissertation.
My special thanks are also extended to Dr. Carol Bonner,
Dr. Tianhui Xia, and Wei Gu for their great help in all
aspects of my study, particularly helping me get started
during my first year in the lab.
I am also very thankful to Dr. Randy Fischer, Dr. Prem
Subramaniam, and Gary Xie for their help during this study.
I am indebted to my family, especially to my parents, to
whom this dissertation is dedicated. Without their love,
support, and encouragement, this study could not have been
accomplished. I am also indebted to my brother and sister-in-
law for helping me in taking care of my parents.
Finally, but not least, I wish to express my sincere
appreciation to my wife, Tao Sun, for her love, support,
IV

patience, and encouragement during these years of study, and
to my son, Peter, and my daughter, Kerry, for filling the
family with great joy and happiness.
v

TABLE OF CONTENTS
ACKNOWLEDGMENTS iv
ABSTRACT vii
CHAPTERS
1 LITERATURE REVIEW 1
Phenylalanine Hydroxylase in Nature 1
The Pterin-Recycling Enzymes 4
Regulation of Phenylalanine Hydroxylase 8
2 PhhR, A DIVERGENETLY TRANSCRIBED ACTIVATOR OF
PHENYLALANINE HYDROXYLASE GENE CLUSTER
OF Pseudomonas aeruginosa 11
Introduction 11
Materials and Methods 13
Results 23
Discussion 42
3 BIFUNCTIONAL PhhB REGULATES THE EXPRESSION OF
PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa 53
Introduction 53
Materials and Methods 55
Results 64
Discussion 84
REFERENCES 8 9
BIOGRAPHICAL SKETCH 97
vi

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
MULTI-LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa
By
Jian Song
May, 1997
Chairperson: Roy A. Jensen
Major Department: Microbiology and Cell Science
Pseudomonas aeruginosa was recently found to possess a
cluster of genes (phh operon) encoding phenylalanine
hydroxylase (PhhA), 4a-carbinolamine dehydratase (PhhB), and
aromatic aminotransferase (PhhC). In the flanking upstream
region of the phh operon, a divergently transcribed gene
(phhR) that encodes an activator protein was identified.
Inactivation of phhR markedly reduced expression of the three
structural genes. PhhR belongs to the large prokaryote family
of a54 enhancer-binding proteins, and activation of the phh
operon by PhhR in P. aeruginosa required rpoN. P. aeruginosa
PhhR was able to replace E. coli TyrR as a repressor of the
aroF-tyrA operon (but not as an activator of mtr) in the
heterologous E. coli system. The phh operon was strongly
induced in fructose- or glucose-based minimal medium by L-
phenylalanine and L-tyrosine, and less by L-tryptophan.
Vll

Inactivation of phhR in P. aeruginosa abolished ability to
utilize either L-phenylalanine and L-tyrosine as a sole source
of carbon for growth.
PhhB is a bifunctional protein. It was shown to have 4a-
carbinolamine dehydratase activity as well as regulatory
activity. The expression of phhA was activated by the presence
of phhB in both E. coli and P. aeruginosa. Transcriptional
and translational fusion analysis showed that the regulatory
effect of PhhB on the expression of phhA is at the post-
transcriptional level.
An insertionally inactivated phhB mutant failed to grow
on L-phenylalanine or L-tyrosine as a sole carbon source.
Expression of PhhA in the absence of PhhB causes strong growth
inhibition in E. coli. The inhibitory effect is probably
caused by 7-tetrahydrobiopterin, which is known to be formed
in the absence of PhhB. Since 7-tetrahydrobiopterin is a
potent inhibitor of phenylalanine hydroxylase, this could
account for the inability of phhA in the absence of phhB to
complement E. coli tyrosine auxotrophy. The general inhibition
of growth may be due to inhibition of some unidentified
essential pterin-dependent enzymes.
vi 11

CHAPTER 1
LITERATURE REVIEW
Phenylalanine Hydroxylase in Nature
Phenylalanine hydroxylase (phenylalanine hydroxylase 4-
monooxygenase; EC 1.14.16.1) catalyzes the irreversible
conversion of L-phenylalanine to L-tyrosine (Kaufman, 1987).
In mammals this enzyme catalyzes the initial, obligatory, and
rate-limiting step in the complete catabolism of serum
phenylalanine to C02 and H20 (Kaufman, 1986) A deficiency of
this enzyme causes accumulation of serum phenylalanine,
leading to hyperphenylalanemia. Because metabolism of
phenylalanine is restricted to alterations in the alanyl side
chain of phenylalanine, in the absence of phenylalanine
hydroxylase, the formation and excretion in the urine of
compounds such as phenylpyruvate and phenyllactate occurs.
This condition is called phenylketonuria, a genetic disorder
associated with severe mental retardation in untreated
children (Dilella et al. 1986) Many mutations at the
phenylalanine hydroxylase locus have been identified (Guldberg
et al., 1996).
Phenylalanine hydroxylase has been intensively studied in
mammals for many years. It is a member of a family of enzymes
that also includes tryptophan hydroxylases (EC 1.14.16.4) and
1

2
tyrosine hydroxylases (EC 1.14.16.2). All three enzymes
utilize a tetrahydrobiopterin cofactor and molecular oxygen to
hydroxylate their respective aromatic amino acid substrates
(Kaufman and Fisher, 1974). Phenylalanine hydroxylase has
been purified from rat liver where it is an oligomeric protein
(predominantly homotetramers) composed of 52-kDa subunits
(Davis et al. 1996) It has non-heme iron as the active-site
metal. The rat liver hydroxylase was also expressed in E.
coli and purified to homogeneity (Kappock et al., 1995) The
homotetrameric recombinant rat hepatic phenylalanine
hydroxylase is highly active and is identical to the native
enzyme in many properties.
Although mammalian phenylalanine hydroxylase has been
intensively studied, few studies on bacterial phenylalanine
hydroxylase have been done. Phenylalanine hydroxylase has
generally been considered to be of rare occurrence in
prokaryotes, where scattered reports of its existence have
been limited to one phylogenetic division of gram-negative
bacteria. They include Pseudomonas acidovorans (previously
known as Pseudomonas sp. ATCC 11299a) (Guroff & Ito, 1963) .
P. facilis (Decicco & Umbreit, 1964), Alcaligenes eutrophus
(Friedrich & Schlegel, 1972), and Chromobacterium violaceum
(Letendre et al., 1974). Of the three pterin-dependent and
metal-containing hydroxylases, only phenylalanine hydroxylase
from Pseudomonas acidovorans (Letendre et al, 1975) and C.
violaceum (Nakata et al., 1979; Pember et al., 1986) has been

3
purified and characterized. The C. violaceum phenylalanine
hydroxylase gene was the first one to be cloned and sequenced
from a bacterium (Onishi et al., 1991). High identity of the
deduced amino acid sequence with those deduced for the
mammalian hydroxylase gene family was found and showed that
the microbial hydroxylase and the mammalian hydroxylases are
homologous. Although C. violaceum phenylalanine hydroxylase is
a pterin-dependent enzyme, it differs from the mammalian
enzymes in its smaller subunit size (lacking the N-terminal
domain responsible for the complex regulation in the mammalian
enzymes), its existence as a monomer (rather than a
homotetramer), and binding of copper (instead of iron) at its
active site. However, the surprising claim has been advanced
that C. violaceum phenylalanine hydroxylase does not require
any redox active metal for its activity (Carr & Benkovic,
1993; Carr et al., 1995).
P. aeruginosa belongs to a different superfamily of gram
negative prokaryotes than do the aforementioned organisms. It
was found to possess homologues of mammalian phenylalanine
hydroxylase, 4a-carbinolamine dehydratase/DCoH, and aromatic
aminotransferase as part of a three-component gene cluster
(Zhao et al., 1994). These three genes are phhA, phhB, and
phhC, respectively. The P. aeruginosa phenylalanine
hydroxylase contains iron and is pterin-dependent. Unlike the
multimeric mammalian hydroxylase, the native P. aeruginosa
hydroxylase is a monomer.

4
The Pterin-Recycling Enzymes
Phenylalanine hydroxylase catalyzes the conversion of L-
phenylalanine to L-tyrosine, using tetrahydrobiopterin as a
reducing agent and relying upon molecular oxygen as an
oxidizing agent (Kaufman, 1987). During this hydroxylation
reaction, the tetrahydrobiopterin cofactor is
stoichiometrically oxidized to a carbinolamine, 4a-
hydroxytetrahydrobiopterin. Two essential enzymes, 4a-
carbinolamine dehydratase and dihydropteridine reductase, are
involved in regenerating the pterin cofactor in two steps.
4a-Hydroxytetrabiopterin is first converted by 4a-
carbinolamine dehydratase to quinonoid dihydrobiopterin, and
the latter compound is then reduced back to
tetrahydrobiopterin by NADH-dependent dihydropteridine
reductase (Kaufman, 1987).
4a-Carbinolamine Dehydratase/DCoH
4a-Carbinolamine dehydratase was first purified from rat
liver as a fraction called "phenylalanine hydroxylase
stimulator", which could stimulate the hydroxylation reaction
at pH 8.2 to 8.4 (Huang et al., 1973). It was later found to
be an enzyme that catalyzes the conversion of 4a-
hydroxytetrabiopterin to the quinonoid dihydropterin (Lazarus
et al., 1983). 4a-Hydroxytetrabiopterin is also known to be
unstable, breaking down nonenzymatically to the corresponding
quinonoid dihydropterin (Kaufman, 1975). However, in the

5
absence of 4a-carbinolamine dehydratase, the dehydration of
the 4a-carbinolamine becomes rate-limiting for the
hydroxylation of phenylalanine. The consequent accumulation
of 4a-carbinolamine results in a small percentage of
rearrangement to the 7-tetrahydrobiopterin isomer (Curtius et
al. 1990) The latter 7-isomer was shown to be a potent
inhibitor of the phenylalanine hydroxylase (Davis et al.,
1992) Under conditions where 4a-carbinolamine and the 7-
isomer are generated, the addition of 4a-carbinolamine
dehydratase markedly inhibits the rate of formation of the 7-
isomer by diverting a greater fraction of the 4a-carbinolamine
to the quinonoid dihydropterin (Davis et al. 1991) Thus, the
dehydratase not only directly catalyzes the dehydration of the
carbinolamine, but also indirectly prevents isomerization to
the inhibitory 7-isomer (Kaufman et al., 1993).
4a-Carbinolamine dehydratase from rat liver has been
cloned and sequenced (Citron et al. 1992). It then became
apparent that this dehydratase is identical to DCoH, a protein
that facilitates the dimerization of hepatic nuclear factor 1
alpha (HNF-la), a homeodomain transcription factor. DCoH was
found to display a restricted tissue distribution and did not
bind directly to DNA. The formation of a stable tetrameric
DCoH-HNF-la complex does not change the DNA-binding
characteristics of HNF-la, but does enhance the
transcriptional activity of HNF-la (Mendel et al. 1991). X-
ray crystallography has revealed DCoH to form a tetramer

6
containing two saddle-shaped grooves that comprise likely
macromolecular binding sites (Endrizzi et al. 1995) .
Structural similarities between the DCoH and nucleic acid
binding proteins imply that the saddle motif has evolved to
bind diverse ligands or that DCoH may bind nucleic acid
according to Endrizzi et al. (1995) .
DCoH homologues have been identified in Xenopus (XDCoH)
(Pogge-yon-Strandmann & Ryffel, 1995) and P. aeruginosa (PhhB)
(Zhao et al., 1994) XDCoH was found to be a maternal factor.
The amount of XDCoH increases dramatically following
neurulation, when the formation of liver, pronephros, and
other organs takes place. The tissue distribution of XDCoH
during embryogenesis suggests that XDCoH is involved in
determination and differentiation of various unrelated cell
types. The interaction with XDCoH was found to be essential
for the function of several tissue-specific transcription
factors (Pogge-yon-Strandman & Ryffel, 1995) In P. aeruginosa
expression of phhA (encoding phenylalanine hydroxylase) was
reported to require phhB (encoding 4a-carbinolamine
dehydratase) suggesting that PhhB may have a positive
regulatory role. If so, this would be an intriguing parallel
with the dual catalytic and regulatory roles of the
corresponding mammalian homolog (Zhao et al., 1994).

7
Dihvdropteridine Reductase
Dihydropteridine reductase (DHPR; EC 1.6.99.7) is one of
the two essential enzymes involved in recycling the pterin
cofactor for aromatic amino acid hydroxylases. It catalyzes
the reduction of quinonoid dihydropterin to
tetrahydrobiopterin, using NADH as a cofactor. DHPR is an
ubiquitous enzyme in animals, being found in all tissues that
contain the aromatic amino acid hydroxylases (Armarego et al. ,
1984). Close correlation between levels of 4a-carbinolamine
dehydratase and dihydropterine reductase in liver during human
fetal development strongly suggests a physiologically
significant role for both enzymes in tetrahydrobiopterin
regeneration. Genetic defects in DHPR cause malignant
phenylketonuria. A concomitant deficiency of
neurotransmitters such as 3,4-dihydroxyphenylalanine (DOPA)
and 5-hydroxytryptophan reflects the essential coupling of
DHPR to tyrosine hydroxylase and tryptophan hydroxylase as
well (Gudinchet et al., 1992) .
DHPR is also found in bacteria. DHPR has been purified
from Pseudomonas acidovorans (Williams et al. 1976) and E.
coli (Vasudevan et al., 1988). In P. acidovorans, both DHPR
and phenylalanine hydroxylase activities were found to be
higher in cells adapted to a medium containing L-phenylalanine
or L-tyrosine as the sole carbon source than in those grown in
L-asparagine (Williams et al. 1976). Interestingly, DHPR has
also been found in E. coli even though no aromatic amino acid

8
hydroxylases or 4a-carbinolamine dehydratase have ever been
detected (Vasudevan et al. 1988). Unlike other
dihydropteridine reductases that have been studied, the E.
coli DHPR possesses an FAD prosthetic group, and has
dihydrofolate reductase and pterin-independent oxidoreductase
activities (Vasudevan et al. 1992) .
Regulation of Phenylalanine Hydroxylase
Phenylalanine hydroxylase in mammals is tightly regulated
at different levels. At the protein level, it is
allosterically regulated by phenylalanine (Kaufman, 1987).
The activity of phenylalanine hydroxylase increases at least
20-fold after incubation with phenylalanine (Tourian, 1971).
It is also activated through phosphorylation by a cAMP-
dependent kinase both in vivo and in vitro (Abita et al,
1976) At the DNA level, expression of the phenylalanine
hydroxylase gene in liver and kidney tissues of mice is
enhanced at birth and is induced by glucocorticoids and cAMP
in liver (Faust et al. 1996) Regulatory elements including
a tissue-specific and hormone-inducible enhancer in the
upstream region have been characterized. The enhancer region
contains separate protein-binding sites for the glucocorticoid
receptor and the hepatocyte-enriched transcription factor,
hepatocyte nuclear factor 1 (HNF1) (Faust et al., 1996) HNF1
is a transcriptional activator of many hepatic genes including
albumin, a-antitrypsin, and a- or /3-fibrinogen. Mice lacking

9
HNF1 die with a marked liver enlargement. The gene coding for
phenylalanine hydroxylase is totally silent, thus giving rise
to phenylketonuria (Pontoglio et al., 1996).
Little information is available about the regulation of
phenylalanine hydroxylase in bacteria. However, some evidence
has indicated that the bacterial phenylalanine hydroxylase is
also regulated. In Pseudomonas acidovorans, a higher level of
phenylalanine hydroxylase was found after growth in the
presence of phenylalanine (Wiliams et al, 1976) Induction of
both phenylalanine hydroxylase and tryptophan hydroxylase in
the presence of their substrates was also reported in C.
violaceum (Letendre et al, 1974).
The most extensively characterized microbial
phenylalanine system is that of P. aeruginosa. Whether this
system is subject to any regulatory controls has not been
studied prior to this work. The initial report of Zhao et al.
(1994) provided a strong basis for anticipation that the phh
operon would be subject to regulation for the following
reasons, (i) The closely spaced organization of the three
structural genes (phhABC) in an apparent operon implies
regulation. (ii) Analysis of effects of the presence or
absence of regions immediately flanking the phh operon upon
expression of phenylalanine hydroxylase indicated the likely
location there of one or more regulatory genes. (iii) The
reported lack of phhA expression in the absence of phhB

10
suggested a positive regulatory role of phhB in addition to
its catalytic function.
The major objectives of this study have been to elucidate
the physiological conditions under which regulation occurs, to
identify and characterize at the molecular-genetic level any
regulatory genes which control the phh operon, and to
determine the nature of the apparent positive regulatory
action of phhB.

CHAPTER 2
PHHR, A DIVERGENTLY TRANSCRIBED ACTIVATOR OF
THE PHENYLALANINE HYDROXYLASE GENE CLUSTER
OF Pseudomonas aeruginosa
Introduction
A recent report (Zhao et al., 1994) revealed that
Pseudomonas aeruginosa possesses a tetrahydrobiopterin (BH4) -
dependent monooxygenase that is capable of catalyzing the
phenylalanine hydroxylase reaction. It is encoded by the
proximal member (phhA) of a three-gene cluster. The second
gene, phhB, encodes carbinolamine dehydratase, a key enzyme
within the cycle regenerating BH4. phhC encodes an aromatic
aminotransferase and belongs to the Family-I aminotransferases
(Gu and Jensen, 1996) The reactions, as they are known to
function for the mammalian homologs in the catabolism of L-
phenylalanine, are shown in Fig. 2-1.
The physiological function of phenylalanine hydroxylase
in P. aeruginosa has not been obvious. A primary role in L-
tyrosine biosynthesis seems unlikely because of the
established presence for this purpose of a cyclohexadienyl
dehydrogenase that is widely distributed in gram-negative
bacteria and which is highly sensitive to feedback inhibition
by L-tyrosine (Xia and Jensen, 1990). Although function as an
initial step of L-phenylalanine catabolism has precedent in
11

12
4-Hydroxyphenylpyruvate
[PhhC]^
Tyrosine
[PhhA]
4a-Carbinolamine -5^5!I
+
-T etrahy drobi opterin
h2o
Dihydropteridine
[DHPR
Phenylalanine
NAD+ NADH+H+
FIG. 2-1. Initial reactions of phenylalanine catabolism in
mammals. The three structural genes of the phh operon encode
enzymes catalyzing three of the four steps shown. The
abbreviations: PhhA, phenylalanine hydroxylase; PhhB, 4a-
carbinolamine dehydratase; PhhC, aromatic aminotransferase;
DHPR, dihydropteridine reductase. 4a-Carbinolamine is an
alternative designation for 4a-hydroxytetrahydrobiopterin.

13
mammalian metabolism, the literature encompassing the widely
studied catabolism of aromatic compounds in pseudomonad
bacteria (indeed, in prokaryotes) does not include the
phenylalanine hydroxylase step. Furthermore, L-phenylalanine
(substrate of PhhA) is an extremely poor source of carbon for
growth of P. aeruginosa, whereas L-tyrosine (product of PhhA)
is an excellent carbon source.
Zhao et al. (1994) had previously noted that subclones
lacking the flanking regions around the phh operon possessed
20-fold greater activity for phenylalanine hydroxylase. This
suggested the presence of a regulatory gene. Since an
understanding of the regulation governing the phh operon
should provide important physiological clues about function,
I have analyzed the flanking regions and now report the
characteristics of a regulatory gene, denoted phhR.
Materials and Methods
Materials
The bacterial strains and plasmids used in this study are
listed in Table 1. The LB and M9 formulations (Sambrook et
al. 1989) were used as growth media for E. coli and P.
aeruginosa. Pseudomonas isolation agar (Difco) was used for
isolating Pseudomonas "knockout" mutants. Additions of
ampicillin (100 /xg/ml) chloramphenicol (40 /xg/ml) kanamycin
(50 /xg/ml), tetracycline (25 /xg/ml), mercuric chloride (15
/xg/ml) L-phenylalanine (50 /xg/ml) and thiamine (17 /xg/ml)

14
TABLE 2-1
. Bacterial strains and plasmids
Strain or
Relevant genotype or
Source or
plasmid
description
reference
E. coli
BL21(DE3)
F- ompT hsdSB (rBmB) gal dcm;
with DE3, a X prophage carrying
the T7 RNA polymerase gene
Novagen
BW545
A (lacU) 169 rpsL
Rosentel et al
DH5a
F~AlacU169 4>80dlacZAM15 hsdR17
recAl endAl gyrA96 thi-1 relAl
supE44
GIBCO/BRL
LE392
F_el4~ (McrA~) hsdR514 (rk~mk+)
supE44 supF58 lacYl or a (lacIZY) 6
galK galT22 metBl trpR55
Sambrook et al
JP2255
aroF363 pheA361 phe0352 tyrA382
thi-1 strR712 lacYl xyl-15
Baldwin &
Davidson
S17-1
[RP4-2(Tc:Mu)(Km:Tn7)Tra(incP)]
pro hsdR recA Tpr Smr
Simon et al.
SP1312
zah-735:Tnl0 A(argF-lac) U169
Heatwole &
Somerville
SP1312
(XSLW20)
SP1312 0 {mtr -lacZ+)
Heatwole &
Somerville
SP1313
SP1312, A(tyrR)
Heatwole &
Somerville
SP1313 SP1313 0 (mtr -lacZ*)
(XSLW20)
P. aeruginosa
Heatwole &
Somerville,
PA103
Prototroph
Totten et al.
PA103NG
rpoN
Totten et al.
PAO -1
Prototroph
Holloway
JS101
PAO-1 phhA, Hgr
This study
JS102
PAO-1 phhR, Hgr
This study

15
Table 2-1
. (continued)
Plasmids
pUC18
Ampr lac'IPOZ'
Yanisch-
Perron et al.
pUC19
Ampr lac'IPOZ'
Yanisch-
perron et al.
pACYC184
P15A replicn, Cmr Tcr
Chang & Cohen
pETllb
T7lac promoter,lacl" Apr
Novagen
pRS1274
lacZY fusion vector
Simons et al.
Z1918
Promoterless lacZ, Apr
Schweizer
pJZ9
phhRABC, Apr
Zhao et al.
pJZ9-3a
phhAB,Apr
Zhao et al.
pJS7
phhRABC, Apr
This study
pJS6 0
phhAB C, Apr
This study
pJS61Z
phhRA'-lacZ transcriptional
fusion, Apr
This study
pJS62Z
phhA'-lacZ transcriptional
fusion, Apr
This study
pJS8 8
pETllb carrying phhR translational
fusion at the ATG start site
This study
pJS91
pACYC184 carrying phhR* Cmr
This study
pJS102
pRS1274 carrying phhR '-lacZY
transcriptional fusion
This study
pCRII
Ampr Kanr lacZa
Invitrogen
pDG106
Hgr Kmr P15A replicn
Gambill &
Summers
pJSlOl
Pstl-Smal fragment of pDG106
inserted into pUC18
This study
pUFRO 04
ColEl Cmr Mob+ mob{ P)
DeFeyter et al

16
were made as appropriate. Agar was added at 20 g/liter for
preparation of solid medium. Restriction enzymes, T4 DNA
ligase, DNA-modifying enzymes (New England Biolab or Promaga)
and Taq DNA polymerase (Perkin-Elmer) were used as recommended
by the suppliers. Other biochemicals were purchased from Sigma
Chemical Co. Inorganic chemicals (analytical grade) were from
Fisher Scientific.
Phenylalanine Hydroxylase Assay
Cultures of E. coli JP2255 carrying the various plasmids
specified were grown in 500 ml of LB broth supplemented with
ampicillin (100 xg/ml) at 37C and harvested at late
exponential phase of growth. The cell pellets were
resuspended in 10 ml of 10 mM potassium phosphate buffer (pH
7.4) containing 1 mM dithiothreitol and were disrupted by
sonication for 30 s at 4C using a Lab-Line Ultratip Labsonic
System (Lab-Line Instruments, Inc., Melrose Park, IL) The
resulting extracts were centrifuged at 150,000 x g for 1 hr at
4 C. The supernatant (crude extract) was desalted using
Sephadex G-25 and used for enzyme assay. PhhA was assayed by
following tyrosine formation (Nakata et al., 1979).
Recombinant DNA Techniques
Molecular cloning and DNA manipulation, including plasmid
purification, restriction enzyme digestion, ligation, and
transformation were conducted by standard methods (Sambrook et

17
al., 1989) DNA fragments were purified from agarose gel with
a "Geneclean" kit (BiolOl). Electroporation (Invitrogen) was
used for simultaneous transformation of E. coli with two
compatible plasmids.
Construction of PhhR Expression Vectors
For expression of PhhR protein in E. coli, the T7
expression system (Novagen) was employed. The phhR coding
region was cloned into a translational fusion vector pETllb.
Polymerase chain reaction (PCR) was used to amplify the phhR
gene. The upper primer (5'-ATACATATGCGTATCAAAGTGCACTGC-3')
was made with a built-in NdeI restriction site (underlined)
which allows fusion of phhR at the translational start site
(ATG in bold). The lower primer (5'-CCTCCACCGTTTCTTTCCCAGCCT-
3') was chosen at a position 48 bases downstream of the
translational stop codon. PhhR protein made from this PCR
fragment was designed to be a native protein, not a fusion
protein. The PCR fragment was cloned into a TA cloning
vector, pCRII. The phhR gene was excised from pCRII with Nde I
and EcoRI. The Ndel-EcoRI fragment was first ligated with
EcoRI-BamHI adaptor to create a Ndel-BamHI fragment which was
then ligated with pETllb digested with Nde 1 and BamHI to
create the PhhR expression plasmid, pJS88 (Fig.2-3A).
For construction of a PhhR constitutive expression
plasmid, pACYC184 was chosen as the expression vector. The
pACYC184 vector has a P15A origin of replication which is

18
compatible with most commonly used plasmids using a ColEI
origin of replication, and it has low-copy number (about 20
copies/cell) High level of PhhR produced from a high-copy
number plasmid was found to be toxic to the host cells. The
Bglll-BairiEI fragment carrying phhR gene was excised from the
expression plasmid pJS88 and cloned into the BamHI site of
pACYC184, thereby interrupting the tetracycline resistance
gene (Tcr) (Fig.2-3A).
Evaluation of Sensitivitv/Resistance to m-FIuoro-tyrosine
Three E. coli strains, SP1312 (tyrR+) SP1313 (tyrR~)
carrying pJS91 [phhR*), and SP1313 (tyrR~) carrying pACYC184
(phhR") were compared for sensitivity to m-fluoro-tyrosine
(MFT). All three strains were first grown in M9 medium with
appropriate antibiotics up to late-exponential phase of growth
and then used to swab M9 agar plates containing appropriate
antibiotics. A sterile Difco concentration disk (0.6 cm) was
positioned at the center of each plate, and 10 n 1 of 50 fig/ml
m-fluoro-tyrosine was applied onto the disks. The plates were
then incubated at 37C for 24 hours.
Construction of vhhA'-lacZ and vhhR'-lacZ Transcriptional
Fusions
To compare levels of phhA transcription in both pJZ9 and
pJZ9-3a, plasmids pJS61Z and pJS62Z were constructed,
respectively. These have a promoterless lacZ gene (from

19
plasmid Z1918) fused at the BarnU I site within phhA to form
phhA'-lacZ transcriptional fusions. Plasmid pJS61Z has the
same upstream sequence as plasmid pJZ9, and plasmid pJS62Z has
the same upstream sequence as pJZ9-3a. Hence, the phhA'-lacZ
fusions in pJS61Z and pJS62Z should represent the phhA
transcriptional levels in pJZ9 and pJZ9-3a, respectively.
To study regulation of the phhR promoter, the HincII-
BanMI fragment (phhR') was cloned into the pRS1274 lacZY
transcriptional fusion vector at the BarriRI Smal site to create
pJS102(phhR'-lacZ).
3-Galactosidase Assay
/?-Galactosidase activity was assayed under conditions of
proportionality as described by Miller (1972), and specific
activities are expressed in Miller units. The data are the
results of at least two independent assays.
Gene Inactivation
P. aeruginosa is well known for its relatively high
resistance to most antibiotics, which complicates attempts to
use most of available antibiotic-resistance genes as selective
markers for gene replacement. Mercury resistance (Hgr) was
used as a selective marker since P. aeruginosa has been shown
to be sensitive to mercury (Essar et al., 1990; Gambill and
Summers, 1985). Insertional inactivation technique described
by Sophien et al. (1992) utilizes a mobilizable suicide vector

20
containing a truncated gene fragment (at both 5' and 3' ends)
and Hgr-cassette, and this suicide plasmid was integrated into
the chromosome by a single homologous recombination event. PCR
was used to generate truncated fragments. To generate a 'phhR'
(6 0lbp) fragment, the upper primer 5'-
CCGTGTAGGCATCCTCCGCGACAT-3', and the lower primer 5'-
CTGGAAGATACTGTCGAAGCCACG-3' were used; to generate the 'phhA'
(639bp) fragment, we used the upper primer 5'-
ACGACAACGGTTTCATCCACTATC-3' and the lower primer 5'-
GGACGAAATAGAGCGGTTGCAGGA-3'. The PCR-generated fragments were
cloned into pCRII (a TA cloning vector) and then excised with
EcoRI. The EcoRI fragments were subsequently cloned into the
EcoRI site of pUFR004 (a mobilizable suicide vector) to create
pUFR/'phhA', and Hgr HindiII-cassette from pJSlOl was inserted
into the Hindlll site of pUFR/'phhA' to create
pUFR/'phhA'/Hgr. pUFR/'phhR'/Hgr was created in a similar
fashion. These plasmids were then used to transform E. coli
strain S17-1 (a mobilizing strain). Strain S17-1 harboring
either pUFR/ 'phhA' /Hgr or pUFR/ 'phhR' /Hgr was used as the donor
in biparental mating with P. aeruginosa performed as described
by Simon et al. (1983) Donor and recipient cells were grown
in LB broth to an OD600 of about 1.0 (E. coli S17-1 at 37C and
P. aeruginosa PAO-1 at 42C) mixed (0.5 ml volume of each) in
a 1.5-ml microcentrifuge tube, and pelleted by centrifugation.
The mating mixture was carefully resuspended in 0.2 ml of LB
broth and spread onto a sterile nitrocellulose filter (0.45-^tm

21
pore size) resting on a prewarmed LB agar plate. The plates
were incubated for 16-24 hours at 37C, and then cells were
removed from the filter by an inoculation loop and resuspended
by vortexing into 0.5 ml of LB broth. Aliquots of 10-, 20-,
50-, and 100-fil volume of the cell suspension were spread onto
Pseudomonas isolation agar plates containing 15 /g of HgCl2.
The plates were incubated overnight and Hgr colonies were
isolated.
Preparation of PhhA-specific Polyclonal Antiserum
PhhA was partially purified by anion-exchange and gel-
filtration chromatography following the methods described by
Zhao et al., (1994). The partially purified PhhA was subject
to SDS-PAGE (12%) and the gel was stained with Commassie blue
R-250. The PhhA band was cut from the gel and used for the
production of polyclonal antiserum in rabbits (Cocalico
Biologicals, Inc., Reamstown, PA) Antiserum was purified by
using an Econo-Pac protein A column (Bio Rad) and further
absorbed with a total cell extract from the PhhA-deficient
mutant JS101.
SDS-PAGE and Western Blot Analysis
SDS-PAGE (12%) was performed with the Mini-PROTEAN II
Cell (Bio-Rad) by the method of Laemmli (1970) Samples of
exponential-phase cells were collected by centrifugation, and
the cell pellets were suspended in gel-loading buffer and

22
heated at 100C for 10 min. Samples of 5-10 /xl were loaded
onto two SDS-acrylamide gels. After separation of the proteins
by electrophoresis, one gel was stained with Coomassie blue R-
250 and the other gel was used for blotting. When crude
extracts were used, equivalent amounts of protein were loaded
in each lane. Western blots were performed according to
Towbin et al. (1979) The proteins were eletrophoretically
transferred onto nitrocellulose membranes and reacted with
polyclonal antibodies raised against PhhA in a rabbit.
N-Terminal Amino Acid Sequencing
PhhR protein produced in E. coli BL21(DE3)/pJS88
following induction by 1 mM IPTG for 3 hours was first
separated from the whole lysate by SDS-PAGE. The proteins
were then blotted to a polyvinylidene difluoride membrane
(Bio-Rad) and were stained with Coomassie brilliant blue R-250
(Sigma). The band corresponding to PhhR was excised from the
membrane and used for sequencing by using an Applied
Biosystems model 407A protein sequencer with an on-line 120A
phenylthiohydantoin analyzer in the Protein Core Facility of
the ICBR at the University of Florida.
DNA Sequencing and Data Analysis
Sequencing of phhR region was performed by the DNA
Sequencing Core Laboratory of the University of Florida. The
nucleotide sequence and the deduced amino acid sequence were

23
analyzed by using the updated version of sequence analysis
software package offered by the Genetics Computer Group (GCG)
of University of Wisconsin (Devereux et al., 1984) .
Nucleotide Sequence Accession Number
The nucleotide sequence reported in this work has been
assigned Genbank accession number U62581.
Results
Evidence for a Flanking Regulatory Region
The original clone (pJZ9) isolated by Zhao et al. (1994)
produced markedly less phenylalanine hydroxylase activity than
did subclone pJZ9-3a (Fig. 2-2A). A possible explanation was
the presence of a negatively-acting regulatory gene in either
the upstream or downstream flanking region. Plasmid pJS60, in
correlation with its absence of upstream DNA but presence of
downstream DNA, expressed a very high level of activity. Thus,
the upstream region appeared to be responsible for decreased
expression of phhA in E. coli.
Transcriptional fusions were constructed using lacZ as a
reporter gene, as diagrammed in Fig. 2-2B. The results
indicate that the negative effect conferred by upstream DNA
occurs at the transcriptional level.

24
(A)
M M
M M H H H M
HHH HHH > H HM
3 a
1 1
a a i s*
1 1 r-J
S .-5
1 Lrf-w
I 3 3
II i 1 i
PhhA activity
phhA phhB
phhC
(nmol/min/mg)
1 kb
pJZ9 2.7
pJZ9-3a 94.0
pJS7 3.5
pJS60 69.7
(B)
pJS61Z 1 tacW
pJS62Z 1 lacZ
P-Galactosidase
activity
(Miller Units)
17
1250
FIG. 2-2. Localization of a regulatory region upstream of the
phh operon. (A) On the right phenylalanine hydroxylase (PhhA)
activities are shown in E. coli JP2255 harboring different
plasmids shown on the left. (B) On the right /6-galactosidase
activities are shown in BW545 harboring the phhA'-lacZ
transcriptional fusions diagrammed on the left.

25
Identification of phhR
A large open reading frame (Fig. 2-3) located upstream of
the phh structural genes appeared likely to be functional on
the criterion of GCG codon preference analysis. The gene,
denoted phhR, produces a deduced protein having 518 residues,
an anhydrous molecular weight of 56,855, and an isoelectric
point of 7.17. It contains a single tryptophan residue.
Regions corresponding to a possible a70 promoter region
and a factor-independent transcription terminator are marked
in Fig.2-3. A strong ribosome-binding site was not apparent.
Bases that are complementary to P. aeruginosa 16S rRNA at the
3' terminus are marked. Perhaps the "A-richness" of the
initiator region enhances ribosome binding (Ivey-Hoyle and
Steege, 1992).
A physical map is given in Fig. 2-4 of the 5874-bp DNA
segment containing the structural genes of the phh operon, the
divergently transcribed regulatory gene phhR, and a gene
(pbpG) downstream of the phh operon which encodes a
penicillin-binding protein (Song and Jensen, unpublished
data).
Homology of PhhR with E. coli TvrR
The closest homolog of PhhR was found to be E. coli TyrR.
The pairwise GAP alignment (GCG) is shown in Fig. 2-5. TyrR
belongs to a family of modular proteins which usually have
three functional domains. The alignment showed high level of

26
-35 -10 IBS phhX
TGTTOCTOGTGAGTCTAACTGTCACATATTCTTGACGGAAATTTCCGGCCGGGAGTTAAAAAACCGGCCGCGAGCCCATCAGJUlCaiLCiACACCGGGCCACGCCATGCGTATCAAAGTGC 120
M R I K V H 6
Hindi
ACTGCCAGAACCGTGTAGGCATCCTCCX3CGACATCCTCAACCTGCTGGTCGACTACGGCATCAACGTCAATCGCGGCGAAGTCGGCGGCGACCAGGGCAACGCCATCTACCTGCTCTGTC 240
CQNRVGILRDILNLLVDYGINVNRGBVGGDQGNAIYLLCP 46
NMINLQLQSLRPKLBAVPGVPGVKR
GLMPSBRRHLBLNA
HindI
CXJCrGCTCGCCGCCCTGGACTTCCCGGTGCTCTCGGTGGACATGGGCGGGCAGATCGTCGCCGCCAACCGTGCCGCCGCGCAGTTGCTCGGCGTGCGCGTCGACGAGGTGCCGGGGATTC 480
LLAALDPPVLSVDMGGQIVAANRAAAQLLGVRVDBVPGIP 126
BHDBSBALAGA
LTLHRADRVGBRIYH
RKQBLRGFDS 206
RVMAAVVRBARRMAPLDAPLLIBGBTGTGKBLLAR 246
GCPRRVGSDBBVYLDVRVICATQ
ATCACCGTCTCAATGTGCTCTCGCT3CACATCCCGCCGCTGCGCGAGTGCCTGGATGGCCTGGCGCCGCTCGCCGAGCATTTCCTCGACCAGGCCAGCCGGCAGATCGGCTGCGGCCTX3C 1320
HRLNVLSLHIPPLRBCLDGLAPLAEHPLDQASRQIGCGLP 406
1440
446
CCGAGCACATCCGCCTGCCGGATTACGGCGCGCCGCAGCCXjCTGGGGGATnTTCCCTGGAAGGAGACTCGACGCATCGTCGGGCGCTTCGAGAAGGGGTGCTGGAGCGCCTGTTCCGCG 1560
BHIRLPDYGAPQPLGDPSLBGDSTHRRALRBGVLBRLPRB 486
AACATCCGAGCACCCGCCAGTTGGGCAAGCGCCTCGGCGTTTCGCATACCACGGCGGCGAACAAGCTGCGCCAGCATC
HPSTRQLGK
H T T A A N
LRQHGVGQSBG
1680
518
GTTGCAGGCTGGGAAAGAAACGGTGGAGGCGTCGGCCCTTGTGAAAGGGCGGCAGCACGATGAGGCGTGGGGAAGACGGCCAGGATGTTCGCCGAGGGGCGGGATTGTCGCCGCCGAGTA 1800
FIG. 2-3. Nucleotide sequence of the phhR region. The numbers
at the right indicate nucleotide and amino acid positions. The
putative promoter region and ribosome-binding site (RBS) are
indicated with bold print. RBS bases that are complementary to
P. aeruginosa 16S rRNA are overlined. The translational start
site is indicated by a bent arrow and the stop codon by an
asterisk. Nucleotides forming the complementary stems of the
putative transcriptional terminator are marked with tandem
arrowheads. Restriction endonuclease recognition sites are
marked above the nucleotide sequence.

27
0 -
a PC
a: 3
a a
X
5! a;
I I!
phhA phhB
%
phhC
H
8
5
o
pbpG
B
?
Si
-f ^ 1
FIG. 2-4. Physical map of the DNA fragment containing phh
structural genes, the divergently-transcribed regulatory gene
phhR, and the downstream penicillin-binding protein gene
(pbpG) in pJZ9. Terminators downstream of phhC and phhR are
indicated. The shaded bars at both ends are portions of the
multiple cloning site of the pUC18 vector. The location of
restriction sites is shown.

FIG. 2-5. Pairwise alignment (GAP program of GCG) of amino acid sequences corresponding
to E. coli TyrR and P. aeruginosa PhhR. The similarity is 66.3% and identity is 45.7%.
The three functional domains are indicated at the right, the central domain also being
shaded. Domain boundaries are based upon those formulated by Morrett and Segovia (1993) .
Alernative boundaries based upon domain segments surviving partial hydrolysis by trypsin
(Cui and Somerville, 1993) are residues 1-190, 191-467, and 468-513 for the three
domains of Eco-TyrR, respectively. In the N-terminal domain, the region between amino
acids 2 and 19 that have a critical role in activation of the expression of E. coli tyrP
and mtr (Pittard, 1996) is double-underlined, and mutations at residues marked in
boldface type abolish TyrR-mediated activation without affecting repression; a second
region between amino acids 92 and 103 which may play a subsidiary role in activation is
also double-underlined; mutations at the residues in boldface type resulted in loss of
function (Pittard, 1996). In the central domain, two ATP-binding sites and a leucine-
zipper motif are underlined. Mutations altering ATP-binding site A and mutations at the
highly conserved residues E-274, G-285, and E-302 abolish TyrR-mediated repression. In
the C-terminal domain, a helix-turn-helix DNA-binding motif is identified with the helix
regions underlined and critical residues in bold print.

TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
MRLEVFCEDRLGLTRELLDLLVLRGIDLRGIEI..DPIGRIYLNFAELEFESFSSLMAEIRRIAGVTDVR
ITl I i . i l I . I Ill II.. I I III ... .11.. ..II I .
I I I I I I I I I I I M l l ill ll ll l
MR IKVHCQNRVGILRDILNLLVDYGINVNRGEVGGDQGNAIYLLCPNMINLQLQSLRPKLEAVPGVFGVK
TVPWMPSEREHLALSALLEALPEPVLSVDMKSKVDMANPASCOLFGOKLDRLRNHTAAOLINGFNFLRWL
l .Mill II I ill ll I I I II I I:: ill I l l . l .
I I I I I I I I l I II I I I l l ll I I I I I I I I I
RVGLMPSERRHLELNALLAALDFPVLSVDMGGQIVAANRAAAQLLGVRVDEVPGIPLSRYVEDLDLPELV
ESEPQDSHNEHWINGQNFLMEITPVYLQDENDQ. HVLTGAWMLRSTIRMGRQLQNVAAQDVSAFSQIV
.1 I ll .1 I ll I I I ill. I i i .1 i . .1 I si:
i M ll I I I I i I I l l I I Ml M I M I W
RANKARINGLRVKVKGDVFLADIAP. .LQS EHDES EALAGAVLTLHRADRVGERIYHVRKQELRGFDSIF
ATP-binding Site A
AVSPKMKHWEOAOKLAMLSAPLLITGDTGTGKDLFAYACHOASPRAGKPYLALNCASIPEDAVESELFG
¡ 11 :!:-! l-liMMlMMiMMl Ml 11 i M - M M 111 M M M -MS 111
QSSRVMAAWREARRMAPLDAPLLIEGETGTGKELLARACHLASPRGQSPFMALNCAGLPESMAETELFG
Unique Gap ATP-binding Site B
HA PEGKKGFFEOANGGSVLLDEIGEMSPRMOAKLLRFLNDGTFRRVGEDHEVHVDVRVICAT
MSI MM -1 M M M M 1 H 1 ^TTTTTTTMTI M 11 M I -1 M M 1111 M I
YGPGAFEGARPEGKLGLLELTAGGTLFLDGVGEMSPRLQAKLLRFLQDGCFRRVGSDEEVYLDVRVICAT
QKNLVELVQKGMFREDLYYRLNVLTLNLPPLRDCPQDIMPLTELFVARFADEQGVPRPKLAADLNTVLTR
1
I
t
I
I M 11--1! 1M11M1IMMMM11M1 = = = I Ml M-- : M- SIMM Mi
QVDLSELCAKGEFRODLYHRLNVLSLHIPPLRECLDGLAPLAEHFLDOASROIGCGLPKLSAOALERLER
Leucine-Tipper Motif
YAWPGNVRQLKNAIYRALTQLDGYELRPQDILLPDYDAATVAVGEDAMEGSLDEITSRFERSVLTQLYRN
i l ll i I ll i l . I .1 .... I ll ll I . I mi mi i i .
I I 11 II M I I I -I I I I I I I I I I II I I
YHWPGNVRQLENVLFQAVSLCEGGTVKAEHIRLPDYGAPQ.PLGDFSLEGDSTHRRA.LREGVLERLFRE
Helix-Tum-Helix Motif
YPSTRKLAKRLGVSHTAIANKLREYGLSOKKNEE*
rm[71 I I II II I IT II r I..I
II I I I II I I I I i I I I II I I I
HPSTRQLGKRLGVSHTTAANKLRQHGVGQSEG*
68
70
138
140
207
208
277
278
339
348
409
318
486
486
514
519
N-Terminal
Domain
Central
Domain
C-Terminal
Domain

30
conservation throughout the entire length of PhhR and TyrR,
and 45.7% of the deduced residues were identical. The N-
terminal domain mediates regulatory modulations, and in TyrR
it binds all three aromatic amino acids. A central domain,
highly conserved throughout the entire family of a54 enhancer
binding proteins, exhibits two established motifs that reflect
the binding of ATP (Pittard, 1996). Site A corresponds to the
ATP-binding pocket motif and site B corresponds to segment 3
of adenylate kinase. In this region a perfect leucine-zipper
motif is apparent in P. aeruginosa PhhR, whereas E. coli
displays an imperfect motif. Residues E-274, G-285, and E-302
were found to be important for TyrR-mediated repression of
aroF-tyrA in E. coli (Yang et al., 1993; Kwok et al., 1995),
and these residues are all conserved in P. aeruginosa PhhR.
The C-terminal domain possesses a helix-turn-helix motif which
is responsible for DNA binding. The absolute conservation of
residues shown to be critical in E. coli (Pittard, 1996)
strongly indicates that PhhR and TyrR might target to a
similar DNA sequences.
Similar to E. coli TyrR, the two aspartate residues and
the lysine residue conserved in the amino-terminal domain of
all response regulator proteins (Stock et al., 1989) were not
found.

31
Overproduction of PhhR
PhhR protein was overexpressed in E. coli BL21(DE3) as
detailed under Materials and methods by use of the T7
overexpression system; the construct is illustrated in Fig.2-
6. The initial use of overexpression vectors containing phhR
on the BairiRl-Sphl fragment (see Fig. 2-4) of pJZ9 failed. This
is probably due to autogenous regulation of phhR, judging from
the precedent set by tyrR in E. coli (Argaet et al., 1994).
Accordingly, overexpression was achieved through excision of
DNA upstream of phhR. PCR methodology was used to generate an
intact phhR gene which was fused with the T7 translational
start codon at a Nde I restriction site to create
overexpression plasmid, pJS88. E. coli BL21(DE3) that had been
transformed with pJS88 was induced with 1 mM IPTG for 3 hours
to express PhhR. Whole-cell lysates obtained before and after
IPTG induction were analyzed by SDS-PAGE, as shown in Fig. 2-
6B. Overproduction of a 56-kDa protein was observed, and N-
terminal amino acid sequencing confirmed its synonymy with
PhhR.
Initial attempts to express phhR in E. coli under
physiological conditions indicated that expression of phhR is
highly toxic. The Bglll-EcoRl fragment from pJS88 was cloned
into the BamHI-EcoRI site of pUC19 behind a lac promoter. When
transformed into E. coli DH5q;, transformants achieved only
pinpoint colony size and eventual survivor cells inevitably
had lost the plasmid. Success was finally achieved by use of

32
(B)
12 3
97.4
66.2
45.0
31.0
21.5
14.5
FIG. 2-6. Overproduction of PhhR protein. (A) Map of PhhR
overexpression plasmid pJS88 (left), and low-copy number,
constitutive PhhR expression plasmid pJS91 (right). (B) SDS-
PAGE analysis of whole-cell lysate of E. coli BL21(DE3)
harboring pJS88. The gel was stained with Coomassie blue. Lane
1, molecular-weight markers; lane 2, before IPTG induction;
lane 3, induced by 1 mM IPTG for 3 h.

33
pACYCl84, a low copy-number plasmid, to create pJS91 which
carried the Bglll-BamHI fragment of pJS88 ligated into the
BamHI site of pACYC184 (Fig. 2-6A). Analysis of 11 plasmids
isolated showed that the orientation of phhR in each case was
opposite to that of the Tcr gene. Presumably, the higher level
of expression expected when driven by the Tcr promoter still
confers an intolerable level of toxicity.
Functional Replacement of E. coli tvrR with phhR
A simple test was used to see whether phhR could
substitute for tyrR as a repressor of the aroF-tyrA operon.
Mutants deficient in TyrR exhibit resistance to m-
fluorotyrosine (Fig. 2-7, middle) whereas tyrR* strains
exhibit sensitivity to growth inhibitory effects of the analog
(Fig. 2-7, left) pJS91 {phhR*) was used to transform an E.
coli tyrR-deficient background (strain SP1313). The ability of
PhhR to replace TyrR is qualitatively apparent (Fig. 2-7,
right) by inspection of the halo of growth inhibition on a
bacterial lawn surrounding a disc containing m-fluorotyrosine
in SP1313 (tyrR' phhR*).
We also examined the ability of PhhR to replace TyrR as
an activator of mtr, encoding a component of a tryptophan-
specific transport system. The phhR* plasmid pJS91 was
transformed into two E. coli X lysogens (Heatwole and
Somerville, 1991) which carried mtr'-lacZ transcriptional
fusions integrated in the chromosome as single-copy fusions.

34
FIG. 2-7. Functional replacement of TyrR by P. aeruginosa
PhhR in E. coli, as monitored by sensitivity to growth
inhibition by m-fluoro-tyrosine (MFT) (A) E. coli tyrR+
(wildtype) strain SP1312 is very sensitive to MFT present on
a central disc, exhibiting a large zone of growth inhibition;
(B) E. coli tyrR~ strain SP1313 is insensitive to MFT, showing
no zone of growth inhibition; (C) P. aeruginosa phhR(pJS91) in
trans complements E. coli tyrR~ and restores the sensitivity
to MFT, as visualized by a zone of growth inhibition.

35
Strain SP1312 (tyrR+) exhibited the expected elevation of
S-galactosidase activity following growth in the presence of
tyrosine, phenylalanine, or both. However, strain SP1313
(tyrR~) carrying pJS91 (phhR+) produced the control level of
S-galactosidase activity, regardless of the presence or
absence of aromatic amino acids (data not shown). Thus, PhhR
appears to be incapable of replacing TyrR as an activator of
E. coli mtr.
Autogenous Regulation of vhhR
The BamHI-Hind I fragment containing the 5' coding
regions of phhA and phhR and the intervening region (see Fig.
2-4) was fused to lacZ to give the reporter-gene construct
pJS102 (phhR' lacZ) This plasmid construct was introduced
into the tyrR-negative background of strain SP1313 in the
presence or absence of pACYC184 possessing a phhR+ insert. The
results (Table 2-2) demonstrated a repressive effect of phhR*
upon PhhR levels as monitored by measurement of /3-
galactosidase activity. Since the copy number of pJS91 (the
source of PhhR molecules) in this experiment is lower than the
number of repressor target sites provided by the high-copy
number pJS102 and since TyrR boxes are present within seven
other transcriptional units of E. coli, auto-regulation is
undoubtedly grossly under-estimated due to titration of
available PhhR molecules in the system.

36
Table 2-2. Autoregulation of P. aeruginosa phhR in E. coli
SP1313(tyrR') containing pJS102{phhR1 -lacZ)
Second
/3-Galactosidase
levelsb in cells
crown in:
plasmid9
M9C
M9 + F M9
+ Y
pACYC184
550
510
589
pACYC184 (phhR*)
362
376
372
a pACYC184 {phhR*)
is denoted pJS91
in Table 1.
b (S-Galactosidase levels are reported in Miller Units.
c M9 minimal medium was supplemented with 1 mM thiamine-HCl
and, where indicated, 1 mM phenylalanine (F), or 1 mM
tyrosine (Y).

37
PhhR as A Positive Regulator
PhhR and TyrR form a cluster within the larger family of
a54 enhancer-binding proteins, as illustrated by Fig. 2-8. A
rpoN mutant of P. aeruginosa was assayed by Western analysis
for PhhA levels of expression in order to determine whether
expression of the phh operon is dependent upon a54 like most
family members, or whether it is a54-independent like tyrR and
luxO. Only low basal levels of PhhA were present in the rpoN
mutant, indicating expression to be largely a54-dependent.
This, in turn, implied that phhR might function as an
activator protein for phhABC transcription. phhR was
inactivated as described under Materials and Methods, and
Western analysis of the effect upon PhhA level was carried
out. The results (Fig. 2-9) indicated that phhR encodes an
activator, the absence of which allows only a low basal level
of activity.
The small molecules, L-phenylalanine and L-tyrosine, was
found to function as an inducer (Fig. 2-10). Western analysis
of PhhA showed no detectable band in minimal medium and a
barely detectable band when L-tryptophan was present, compared
to prominent bands when L-phenylalanine or L-tyrosine was
additionally present. Carbon-source levels of L-phenylalanine
or L-tyrosine were not required for induction. It is probable
that L-phenylalanine or L-tyrosine is a co-activator moiety
which, in combination with PhhR, forms the holo-activator
moiety. It is perhaps relevant that for those transcriptional

FIG. 2-8. Homology relationships of the central domain of P.
aeruginosa PhhR with the central domain of other members of
the a54-dependent family of transcriptional regulators. The
dendrogram was generated with amino acid sequences of the
central domain as defined by Morrett and Segovia (1993) by
using the PILEUP program of GCG. The top three proteins form
a cluster designated as subfamily a, and the remaining
proteins form a larger cluster designated as subfamily 13. Due
to their high degree of similarity, only one of the ortholog
sequences of NifA, NtrC and HydG proteins is shown. The six
paralogs from E. coli and the three paralogs from P.
aeruginosa are designated with and 4*, respectively.
Abbreviations: Eco, Escherichia coli; Avi, Azotobacter
vinelandii; Hin, Haemophilus influenzae; Pae, Pseudomonas
aeruginosa; Vha, Vibrio harveyi. Functions controlled by the
following regulators are given parenthetically: PhhR
(phenylalanine hydroxylase), TyrR (aromatic amino acid
biosynthesis and transport), VnfA (nitrogen fixation,
nitrogenase-2), Anf (nitrogen fixation, nitrogenase-3), NifA
(nitrogen fixation, nitrogenase-1), HydG (hydrogen oxidation),
NtrC (nitrogen assimilation) PilR (synthesis of Type IV
pili), AlgB (alginate production), LuxO (luminescence), FhlA
(formate metabolism), YfhA (possible control of glnB), PspF
(phage shock protein).

39
*Pae-PhhR
*Eco-TyrR
Hin-TyrR_
*Pae-PilR
*Eco-HydG
*Eco-YfhA
Avi-VnfA
Avi-AnfA
Avi-NifA
*Eco-FhlA
*Eco-NtrC
*Pae-AlgB
Vha-LuxO
*Eco-PspF_

40
PAO-1 PA 103
WT phM. phhK WT rpoN *'
FIG. 2-9. Western blot analysis of phhA expression in mutant
derivatives of P. aeruginosa strains PAO-1 and PA103. The
proteins in crude extracts prepared from cultures grown in LB
medium were separated by SDS-PAGE, and equal amounts of
protein (50/xg) were applied to each lane.

41
(A)
Fructose
+Phe +Trp +Tyr
(B)
Glucose+Phe Fructose+Phe
ii
I
Z
£
O
z
Z
o
oc
<
<
H
£
e-
o
E
fifi
§
S'
z
w
w
c*
o
C/3
s
<
S
<
1
o
<
o
fH
C/3
O
*5
a.
ft-
0,
ft-
FIG. 2-10. Western blot analysis of phhA expression. (A)
Examination of aromatic amino acids as inducers of phhA
expression. P. aeruginosa PAO-1 was grown in minimal salts-
glucose or minimal salts-fructose medium with or without
addition of one of the three aromatic amino acids at a final
concentration of 100 fig/ml. (B) Phenylalanine induction of
phhA expression in different P. aeruginosa strains. Bacteria
were grown in minimal-glucose or minimal-fructose media
containing 100 g/ml phenylalanine.

42
units where TyrR functions as an activator, L-phenylalanine
functions as an essential co-activator (Pittard and Davidson,
1991).
Discussion
Anomalous Repression of the nhh Operon by PhhR in E. coli
PhhR can mimic the ability of TyrR to repress the aroF-
tyrA operon at a a70 promoter. This indicates that PhhR can
recognize TyrR boxes and is consistent with the high
similarity of the helix-turn-helix, DNA-binding domain within
the carboxy-terminal segments of TyrR and PhhR. However, PhhR
was unable to activate the phh operon in the heterologous E.
coli background, suggesting an incompatibility between the E.
coli RpoN and the P. aeruginosa a54-dependent system. The
expression of PhhA from a promoter recognized by E. coli
upstream of the native a54 promoter was in fact severely
depressed in constructs containing phhR, even in the presence
of added co-activator (L-phenylalanine). In the presence of P.
aeruginosa PhhR, an aberrant complex apparently blocks
transcription initiated upstream of the a54 promoter.
Emerging Subfamilies within the a54 Enhancer-Binding Protein
Family
P. aeruginosa PhhR belongs to an outlying subgroup (which
we denote subfamily a in Fig. 2-8) of the a54 enhancer-binding
protein family. All members of the family possess in common a

43
homologous central domain, but the amino-terminal and carboxy-
terminal domains may vary considerably within the family.
Thus, this exemplifies a complex multi-domain protein family
in which family membership is defined by a common ancestral
central domain. Future subdivisions within what is termed
subfamily ¡3 in Fig. 2-8 could likely be defined on the
criterion of homology for the remaining two domains. For
example, Eco-NtrC and Eco-FhlA belong to different mechanistic
subgroups: the two-component regulatory system and direct
response-to-small-molecules, respectively (reviewed by
Shingler, 1996).
Figure 2-8 highlights the emerging homology relationships
of selected paralog and ortholog proteins, with respect to the
central domain. E. coli possesses at least six paralogs, some
of which diverged in a common ancestor that existed prior to
speciation events which generated orthologs. Thus, the
divergence of Eco-NtrC and Pae-PilR was a more recent event
than was the divergence of Eco-NtrC and Eco-PspF. In contrast
to the ancient duplication events which generated all of the
E. coli paralogs (or the P. aeruginosa paralogs) are the
relatively recent duplication events generating the three
paralogs which regulate three distinctly separate nitrogenase
systems in Azotobacter (Joerger et al., 1989).
P. aeruginosa PhhR and E. coli TyrR exhibit homology in
all three domains: 36% identity, amino-terminal; 52% identity,
central; and 47% identity, carboxy-terminal). Curiously, the

44
amino-terminal domain of H. influenzae TyrR appears to be
absent. It is not known whether sequencing errors might
account for this, or whether the equivalent of the amino-
terminal domain might exist separately as a different protein.
A multiple alignment of the central-domain modules of
subfamilies a and (3 was shown in Fig. 2-11. In addition to the
many residues that are absolutely conserved throughout the
family, some residues which may prove to be uniquely conserved
within subfamily a are apparent, e.g., APLL corresponding to
residues 29-32 of Hin-TyrR.
Both Eco-TyrR and Rca-NtrC exhibit deletions in the
"unique-gap region" of the central domain (Fig. 2-12) in
correlation with their regulation of a70 promoters, rather than
a54 promoters. This observation led to the suggestion (Morrett
and Segovia, 1993) that this region of the central domain
might be critical for functional interfacing with a54. Since
this DNA segment of Pae-PhhR is intact with absolute retention
of highly conserved residues, the foregoing hypothesis is
consistent with the successful interaction of PhhR with a a54
promoter. Hin TyrR, on the other hand, is likely to be
deficient in interaction with a54 (like E. coli TyrR) owing
to a 6-residue deletion in this region.
Intervening Region of Divergent Transcription
Since the DNA-binding region of the carboxy terminus of
PhhR is identical at all important residues with E. coli TyrR,

FIG. 2-11. A comparison of the amino acid sequences in the
central domain of the PhhR protein and 13 other homologs. The
sequences were aligned by using the PILEUP program of GCG. The
numbering of amino acid residues is given on the left. Percent
identity of PhhR with its homologs is given at the lower
right. Amino acid residues conserved in all 14 sequences are
in double-lined boxes. Amino acid residues conserved in 13 of
14 sequences are shaded. Conserved residues which are confined
to either the top cluster (subfamily a) or the bottom cluster
(subfamily ¡3) are in single-lined boxes. Two ATP-binding
motifs are indicated above the consensus sequences in boldface
type. See the legend for Fig. 8 for abbreviations.

46
PhhR-Pae F Q S S
TyrR-Bco V A V S
TyrR-Hin I V Q S
PilR-Pae L G B S
HydG-Bco V G K S
YfhA-Bco V T R S
VnfA-Avi I G N S
AnfA-Avi I G N S
NifA-Avi V G H T
PhlA-Bco I G R S
NtrC-Bco I G B A
AlgB-Pae B S H S
LuxO-Vha I G S S
PspF-Bco L G B A
R V H A A V [vl
P K M K H V V
B A M K S A |vj
P P M R A L R
PA8QHL L
P L H L R L L
K P M L B V Y
K P W Q B V Y
P T M R R V F
B A Y S V L
P A H Q D V P
P A M A A V L
QTQQ V Y
N S f L B V L
R BaIr R M A P L
B Q A Q K L A M L
b n[aJk R P A M F
NQIGKLARS
SBIALVAPS
EQARLVAQS
QLIERVVRT
BLIHKVAST
DQIRRVAKW
KQVBMVAQS
RIIGRLSRS
BTARQVAAT
RTIDSAASS
BQVSHLAPL
ATP-binding Motif A
A
P
L
L
*
E
G
E
T\
A
P
L
L
X
T
G
D
T
A
P
L
L
X
Q
G
E
T
A
P
V
Y
X
s
G
E
s
A
T
V
L
x
H
G
D
s
V
S
V
L
t
N
G
0
s
T
T
V
L

L
G
E
s
A
T
V
L
i
L
G
E
s
S
T
V
L
V
L
G
E
s
S
T
V
L

L
G
E
T
I
S
V
L
*
N
0
E
s
A
N
I
L
X
L
G
E
S
A
S
I
F
X
T
G
E
s
K
P
V
L
t
1
in
E
R
£ P R G Q S
B P R A G K
S L R R D K
SPRIER
B A R S E K
$ P R N S K
$ P A A K G
S P N A E A
5 P R A H R
SGRNNR
6 P R A K A
B K R A K K
5 K R G D K
3 S R W Q G
257
256
65
186
191
187
260
269
261
431
190
197
183
58
PhhR-Pae
TyrR-Bco
TyrR-Hin
PilR-Pae
HydG-Bco
YfhA-Bco
VnfA-Avi
AnfA-Avi
NifA-Avi
PhlA-Bco
NtrC-Bco
AlgB-Pae
LuxO-Vha
PspP-Bco
PhhR-Pae
TyrR-Bco
TyrR-Hin
PilR-Pae
HydG-Bco
YfhA-Bco
VnfA-Avi
AnfA-Avi
NifA-Avi
FhlA-Eco
NtrC-Bco
AlgB-Pae
LuxO-Vha
PspF-Bco
P
P
K
F m[a]l n C
Y L A L N C
F I [aJ V N e
P F V P V N C
P L V T L N
P f I A I N i
PFVKPN
A L V T S N t
P P V R L N C
R M V K M N C
P F I A L N i
PQ VTIH
P F I A I N G
P F I S L N ej
G V G B M S P
B I G B M S P
G I A B L S L
E V A D L P M
E I G D I S P
E I G D M P A
B V G B M S L
B V G E L S P
B I G B I S P
B V G D M P L
B I G D M P L
B I G D P P L
E L C B M D L
E L A T A P M
A G L
ASI
A G L
G A I
A A L
GAL
A S L
A P L
A A L
A A M
A A I
P S L
A A I
A A L
P E S M A
P B D A V
P D B D A
P S B L M
N B S L L
P B Q L L
P B S V I
P B N L A
P B T L L
P A G L L
P K D L I
T A B L M
P K D L I
N B N L L
* T 3 L
B t L
i a a m
|y G P gI A fFI BIG Al
H A
R K V G D
R P
B
G
K
L
G
L
L
L T
A
G
S
T
L
P
L D
307
. P
E
G
K
K
G
F
F
E
Q A
N
G
G
S
V
L
L D
298
S
E
T
I
G
F
u
Y ft
N
K
a
T
V
L
L
D
109
. I
B
D
K
Q
G
L
F
Q
A ft
S
G
0
T
L
F
L
D
235
. D
K
R
R
E
G
R
F
V
B A
D
G

T
L
F
L
D
240
. V
S
N
R
B
G
L
F
Q
A A
B
G
9
T
L
F
L
D
235
. I
G
L
R
K
G
R
F
E
E ft
A
G
0
T
I
F
L
D
309
. L
T
M
H
K
G
C
F
E
Q ft
D
G
G
T
I
F
L
D
318
. V
K
Q
R
K
G
R
F
E
Q D
G
G
T
L
F
L
D
310
. S
A
Q
R
I
G
R
F
B
B *
D
K
S
S
L
F
L
D
480
. N
T
I
R
Q
G
R
F
E
0 B
G
i
T
L
F
L
D
239
. T
E
S
T
L
G
R
V
S
Q ft
D
G
G
T
L
F
L
D
246
. A
N
D
R
Q
G
A
A
B
L ft
D
G
0
T
L
F
L
D
232
Q
K
R
H
P|[Gj
R
F
B
R ft
D
G
0
T
L
P
L D
107
ATP-binding Motif B
C F R R VMG S
T F R R V G B
S F R R V G E
A V R A V G G
B V Q R V G S
K V R P L G S
S F E R V G G
T F E R V G G
E F B R V G G
E F B R L GIIS
Q F Y R V G G
B Y B R V G D
T F Q K VlG|S
B L E R v||G||G
D
E
E
V
Y
L
D
V
R
V
I
C
A
X
0
V
D
h
s
B
L
c
A
357
D
H
E
V
H
V
D
V
R
V
I
C
A
T
Q
K
N
h
V
E
L
V
Q
348
E
K
E
H
Y
A
N
V
R
V
I
C
T
3
Q
V
P
b
H
B
B
V
E
159
Q
Q
E
V
A
V
D
L
R
I
L
C
i
H
K
D
L
A
A
E
V
G
285
N
Q
I
I
S
V
D
V
R
L
I
A
A
?
H
R
D
L
A
A
E
V
N
290
N
R
D
I
D
I
N
V
R
I
I
S
*
T
H
R
D
b
P
K
A
M
A
285
N
T
T
I
H
V
D
L
R
V
I
A
A
T
N
R
N
L
A
E
M
V
A
359
S
K
P
V
K
V
D
V
R
I
I
A
A
T
N
R
N
L
V
E
M
V
E
368
N
Q
T
V
R
V
N
V
R
I
V
A
A
T
N
R
D
L
E
S
E
V
E
360
N
K
I
I
Q
T
D
V
R
L
I
A
a
t
N
R
D
t
K
K
M
V
A
530
Y
A
P
V
K
V
D
V
R
I
I
A
A
T
H
0
N
h
E
Q
R
V
Q
289
P
V
T
R
R
A
D
V
R
I
L
A
A
T
N
R
D
L
G
A
M
V
A
296
S
K
M
K
S
V
D
V
R
F
V
C
A
?
N
R
D
P
W
K
E
V
Q
282
S
Q
P
L
Q
V
N
V
R
L
V
C
A
*
N
A
D
b
P
A
M
V
N
157
PhhR-Pae
TyrR-Bco
TyrR-Hin
PilR-Pae
HydG-Bco
YfhA-Bco
VnfA-Avi
AnfA-Avi
NifA-Avi
FhlA-Bco
NtrC-Bco
AlgB-Pae
LuxO-Vha
PspF-Eco
k s b tm Q D L Y
K 5 M *;!& B D L Y
Q G K V a A D L F
a j; r q d l y
A G R ISX Q D L Y
RGB B D L Y
D S T # A B D L Y
06T am B D L Y
K d K am E D L Y
D R B S D L Y
B Q K B D L P
Q G Q 'M B D L L
E G R B D L Y
B G T itm A |d l| l
h [FTIn V[l1s
Y R L N V L T
H R L N V[lJt
Y R L N V I E
Y R L N V V A
Y R L N V V S
Y R L N V F P
Y R L N f F P
Y R L N V M A
Y R L N V F P
H R L N V I R
Y R L N V I V
Y R L Y V I P
D R L A ft D V
L H
L N
I N
L R
I B
L K
I T
I T
I R
I H
V H
L N
L H
V Q
P [l] 3 E C L
P L R D C P
A L ft D R M
plSerr
S L ft Q R R
A L A D G T
P L A E R G
P L R E R G
P L R E R T
P L ft E R P
P L R E R R
P L ft E R A
P L ft E R G
P ij ft B R E
D O L A [p|l
Q D I M PL
A? I B |_p|l
E P I P L L
E P I P L L
E P I P L L
S I I T L
S d V I A L
A Q I P EL
E & I P L L
E p I P R L
E P I L G L
K P V I El
S P I M L M
A E H[F]L D Q A
T B LEV A R F
A Q G[f|L Q E I
Aerilkrl
Aghflqrf
Anhllrqa
ftDHPVSRF
ADHFVSAF
ftEFLLGKI
Akaptfki
Arhflqva
Aerflarf
Aysllgym
Aeyfaiqm
SRQIGCGL .
ADB-QGVPR .
SBBLKIAK .
AGDTGLPA A 335
AERNRKAV K 340
ADGHKPFV R 334
SREMGIEV N 409
SRENGKNV K 418
GRQQGRPL T 409
ARRLGRNI D 580
ARELGVBA K 339
VKDYGRPA R 345
SHEEGKSF.V 332
CREIKLPLFP 208
P 407
P 398
P 209
PhhR-Pae K L S
TyrR-Bco K L A
TyrR-Hin T F D
PilR-Pae R L T
HydG-Bco G F T
YfhA-Bco APS
VnfA-Avi R I S
AnfA-Avi R I S
NifA-Avi V T
FhlA-Bco SIP
NtrC-Eco L L H
AlgB-Pae G F S
LuxO-Vha R F A
PspF-Eco G F T
A Q A L B R L
A D L N T V L
K D F L L Y L
G D A Q B K L
P Q A M D L L
T D A M K R L
T P R L N M L
T P A L N M L
D S A I R L L
A E T L R T L
P B T E A A L
B A A R B A M
Q D V I B R F
E R A R E T L
E
R
Y
H
G
Hi
V
it
0
L
E
1
V
L
F
Q
A
V
s
L
C
E
G
G
442
% Identity
T
R
Y
a am
G
N
V
R
Q
L
K
N
A
I
Y
R
A
L
T
Q
L
D
G
Y
433
54%
Q
*
[yJ
d r*
G

v
R
E
L
Y
N
T
L
Y
R
A
C
S
L
V
Q
D
N
244
47%
K
N
Y
R F ft
G
N
V
G
B
L
E
N
M
L
E
R
A
Y
T
L
C
E
D
D
370
42%
I
H
Y
D P
G
N
I
R
E
L
E
N
A
V
E
R
A
V
V
L
L
T
G
B
375
43%
M
T
A
s w p
G
N
V
ft
Q
L
V
N
V
I
E
Q
C
V
A
L
T
s
s
P
260
45%
Q
S
Y
Q WP
G
N
V
ft
E
L
E
N
V
I
E
R
A
M
L
L
S
E
D
G
444
47%
M
s
Y
H
G
N
V
ft
E
L
E
N
V
M
E
R
A
V
I
L
S
D
D
D
452
45%
M
S
H
R W :P
G
N
V
S
E
L
E
N
c
L
E
R
s
A
I
M
S
E
D
G
444
48%
S
N
M
E M P
G
N
V
8
E
L
E
N
V
I
E
R
A
V
L
L
T
R
G
N
615
44%
T
R
L
a am
G
N
V
R
Q
L
E
N
T
C
R
W
L
T
V
M
A
A
G
Q
374
44%
R
Q
Y
p
G
N
V
R
E
L
R
N
V
I
E
R
A
S
I
I
C
N
Q
E
381
45%
N
s
Y
B P
G
N
V
R
Q
L
Q
N
V
L
R
N
I
V
V
L
N
N
G
K
367
45%
L
N
Y
E K P
G N
I
ft
E
L
K
N
V
V
E
R
S
V
Y
R
H
G
T
S
243
41%

Eco-TyrR
P
E
D
A
V
E
S
E
L
F
G
H
A
. P
E
G
K
K
G
F
F
E
Q
A
N
G
G
S
V
L
L
D
Hin-TyrR
P
D
E
D
A
E
S
E
M
F
G
R
K
V
G
D
.
.
, .
S
E
T
I
G
F
F
E
Y
A
N
K
G
T
V
L
L
D
Pae-PhhR
P
E
S
M
A
E
T
E
L
F
G
Y
G
P
G
A
F
E
G
A
R P
E
G
K
L,
G
Li
L
E
L
T
A
G
G
T
L
F
L
D
Pae-PilR
P
S
E
L
M
E
S
E
F
F
G
H
K
K
G
S
F
T
G
A
. I
E
D
K
Q
G
L
F
Q
A
A
S
G
G
T
L
F
L
D
Eco-HydG
N
E
S
L
L
E
S
E
L
F
G
H
E
K
G
A
F
T
G
A
. D
K
R
R
E
G
R
F
V
E
A
D
G
G
T
L
F
L
D
Eco-YfhA
P
E
Q
L
L
E
s
E
L
F
G
H
A
R
G
A
F
T
G
A
. V
S
N
R
E
G
L
F
Q
A
A
E
G
G
T
L
F
L
D
Avi-VnfA
P
E
s
V
I
E
s
E
L
F
G
H
E
K
G
S
F
T
G
A
. I
G
L
R
K
G
R
F
E
E
A
A
G
G
T
I
F
L
D
Avi-AnfA
P
E
N
L
A
E
s
E
L
F
G
H
E
K
G
S
F
T
G
A
. L
T
M
H
K
G
C
F
E
Q
A
D
G
G
T
I
F
L
D
Avi-NifA
P
E
T
L
L
E
s
E
L
F
G
H
E
K
G
A
F
T
G
A
. V
K
Q
R
K
G
R
F
E
Q
A
D
G
G
T
L
F
L
D
Eco-FhlA
P
A
G
L
L
E
s
D
L
F
G
H
E
R
G
A
F
T
G
A
. S
A
Q
R
I
G
R
F
E
L
A
D
K
S
S
L
F
L
D
Eco-NtrC
P
K
D
L
I
E
s
E
L
F
G
H
E
K
G
A
F
T
G
A
. N
T
I
R
Q
G
R
F
E
Q
A
D
G
G
T
L
F
L
D
Pae-AlgB
T
A
E
L
M
E
s
E
L
F
G
H
S
R
G
A
F
T
G
A
. T
E
s
T
L
G
R
V
S
Q
A
D
G
G
T
L
F
L
D
Vha-LuxO
P
K
D
L
I
E
s
E
L
F
G
H
V
K
G
A
F
T
G
A
. A
N
D
R
Q
G
A
A
E
L
A
D
G
G
T
L
F
L
D
Eco-PspF
N
E
N
L
L
D
s
E
L
F
G
H
E
A
G
A
_F
T
G
A
Q
K
R
H
P
G
R
F
E
R
A
D
G
G
T
L
F
L
D
298
109
307
235
240
235
309
318
310
480
239
246
232
107
Rca-NtrC
L G A D G
P S S L L
A RRCGRLVV f[d]222
-J
FIG. 2-12. Alignment of the unique-gap region in the central domains of TyrR
proteins with selected homologs. Amino acid residues conserved in all of the 15
sequences that include both subfamily a (top cluster) and subfamily /3 (lower
cluster) are in shaded boxes. Amino acid residues conserved within the gap region
are shown in open boxes. Rea, Rhodobacter capsulatus; see the legend of Fig. 8 for
other abbreviations used.

48
it is likely that PhhR binds to the same binding sites for E.
coli, which are referred as "TyrR boxes" (consensus:
TGTAAAN6TTTACA) This conclusion is also supported by the
ability of PhhR to replace TyrR as a repressor of the aroF-
tyrA transcriptional unit. The location of two "PhhR boxes"
which match the consensus for "TyrR boxes" was shown in Fig.
2-13. PhhR Box 1 is a strong box (with more conserved-symmetry
and higher affinity for TyrR) that overlaps the putative -10
region of the phhR promoter. TyrR boxes in E. coli occur in
tandem with variable spacing (Pittard, 1996), and a TyrR
hexameric molecule is thought to bind both a strong box and a
weak box with DNA looping in between. PhhR Box 2 is a weak box
located in the middle of the intervening region. It seems
probable that by analogy with autorepression of tyrR in E.
coli, both phhR boxes participate in the autogenous repression
of phhR by PhhR, probably with tyrosine as a corepressor.
In the opposite direction of transcription, the a54
promoter for phhABC requires an upstream activator site (UAS) .
PhhR Box 1 may be the most likely UAS, although perhaps both
boxes participate in activation of phhA. L-Phenylalanine and
L-tyrosine, potent inducers of phenylalanine hydroxylase,
presumably are the effector molecules. Since a rpoN mutant
retained low basal level of PhhA, another promoter that is
independent of a54 might be present.
No motif for binding of integration host factor (IHF)
(Friedman, 1988) was located in the intervening region.

49
BajnH I
CCTAGGCGAGCACCCCGTCCGGCTCGACAACGTACGGCAGCTCCATAAGGACTGTCCGCGCCGGAAGCTAGTGGAAGTCAACGGCCCACTAGTCCCATAAGGTCTGGACCACGAGCCAAA 120
PIREHPLGLQEIGDLYEQCARGEIVKLQRTILTNWVQHET
GGCCTATCACCTACTTTGGCAACAGCAGCCCGACCGCCCGGTGCATGACGCAGCAAAAGTATGCCTOAOGTACTCGCCGAACACCGGCGCCCCAGAAACAACAACAGCAACGGAATAGCT
BPYHI FGNDDPQRAVYQTTKM RBS- U
phhA 4 1 -12
240
PhhR Box 2
TACCGCGTCGGGACGGTCCGTGCAGCCCCGGATAGGGACCGGGCAACGGGGAGGAATCGGCGTTTT CGTAAAGTTTTCCTTACG AATTGGCCTGGGTCGCCTGTTCATTGGGTCAGGCAT
ATGGCGCAGCCCTGCCAGGCACGTCGGGGCCTATCCCTGGCCCGTTGCCCCTCCTTAGCCGCAAAA GCATTTCAAAAGGAATGC TTAACCGGACCCAGCGGACAAGTAACCCAGTCCGTA
TGTTGCTGGTGAGTCTAAC '
ACAACGACCACTCAGATTG .
1 GAAATTTCCGGCCGGGAGTTAAAAAACCGGCCGCGAGCCCATCAGAACGACXACACCGGGCCACGCCATGCGTATCAAAGTGC 4 80
j CTTTAAAGGCCGGCCCTO^TTCTTGOCCGkS33GTt?35tW3TCPKK3tCTTQ3GGGOQ6nQGGG!rO(3
Terminator ?
CQNRVGILRDILNLLVDYGINVNRGEVGGDQGNAIYLLCP
ACTGCCAGAACCGTGTAGGCATCCTCCGCGACATCCTCAACCTGCTGGTCGACTACGGCATCAACGTCAATCGCGGCGAAGTCGGCGGCGACCAGGGCAACGCCATCTACCTGCTCTGTC
Hindi
FIG. 2-13. The intervening sequence between the divergently
transcribed phhA and phhR genes. The number at the end of each
line indicates the nucleotide position. The ribosome binding
sites and putative promoter sites (-12/-24 promoter for phhA,
and -10/-35 promoter for phhR) are indicated. The
translational start sites are indicated by arrows. Two PhhR
boxes are identified. A stem-loop structure is shaded.
Restriction endonuclease recognition sites are marked.

50
Therefore, this region may possess intrinsic DNA-bending
capabilities.
Function of the phh Qperon
The primary function of the phh operon is clearly not to
accommodate tyrosine biosynthesis since the feedback-inhibited
cyclohexadienyl dehydrogenase which is widely distributed in
gram-negative bacteria exists for this purpose. However, the
phh operon probably provides a fortuitous backup capability
for tyrosine biosynthesis. "Reluctant auxotrophy" for tyrosine
(Patel et al., 1978) can be explained as follows. Mutational
deficiency of cyclohexadienyl dehydrogenase would lead to
accumulation of prephenate, a potent product inhibitor of
chorismate mutase. The subsequent backup of chorismate,
enhanced by lack of early-pathway control in the absence of L-
tyrosine, results in passage of chorismate to the periplasm
where chorismate mutase-F (Gu and Jensen, unpublished data)
and cyclohexadienyl dehydratase (Zhao et al., 1993) generate
L-phenylalanine. Subsequent induction of phenylalanine
hydroxylase completes the alternative circuit to L-tyrosine.
The established function of phenylalanine hydroxylase in
mammals is for catabolism of L-phenylalanine as a carbon
source. We have found that phenylalanine hydroxylase is indeed
essential for use of L-phenylalanine as a sole carbon source
in P. aeruginosa. Thus, inactivation of phhA resulted in
inability to use L-phenylalanine as a sole source of carbon

51
(data not shown). However, induction of the phh operon under
conditions where better carbon sources (such as glucose)
coexist, suggests that the phh operon might be dedicated to
provision of some specialized compound from L-phenylalanine.
Inactivation of phhR resulted not only in the inability
to use L-phenylalanine as a carbon source, but also in an
inability to use L-tyrosine as a carbon source. Since TyrR
regulates aromatic amino acid permeases in E. coli, we
considered the possibility that the phhR mutant might fail to
grow on L-tyrosine because of a permease deficiency. Since MFT
is likely to be transported by the same system as L-tyrosine,
a permease-deficient phenotype should be resistance to growth
inhibition by MFT. However, the phhR" mutant has a MFT-
sensitive phenotype on fructase-based medium (data not shown).
Therefore, PhhR might regulate steps of tyrosine catabolism.
Regulation of Multiple Transcriptional Units by PhhR?
TyrR represses or activates eight transcriptional units
in E. coli (Pittard, 1996) Similarly organized
transcriptional units are absent or unknown in P. aeruginosa.
However, the counterpart of the aroF-tyrA operon in P.
aeruginosa would be genes encoding tyrosine-sensitive DAHP
synthase and cyclohexadienyl dehydrogenase. Physiological
manipulations in our laboratory have never revealed repression
control of these apparently constitutive enzymes. Consistent
with this, PhhR exhibits no regulatory control of either of

52
these enzymes, on the criterion of assessment of specific
activities determined in comparison of tyrR* and t yrR~
backgrounds (data not shown).

CHAPTER 3
BIFUNCTIONAL PhhB REGULATES THE EXPRESSION OF
PHENYLALANINE HYDROXYLASE IN Pseudomonas aeruginosa
Introduction
Mammalian 4a-carbinolamine dehydratase was initially
known for its catalytic activity of converting 4a-
carbinolamine to quinonoid dihydrobiopterin in regenerating
the tetrahydrobiopterin for phenylalanine hydroxylase (Fig. 3-
1B) Later, it was found to be synonymous with DCoH, the
dimerization cofactor for hepatic nuclear factor 1 alpha (HNF-
la) (Citron et al., 1992) .
A homolog of the mammalian DCoH, PhhB, was found in
Pseudomonas aeruginosa by Zhao et al. (19 94) The PhhB
protein is encoded by the second structural gene, phhB, of the
phh operon (Fig. 3-1A) Zhao et al (1994) reported that phhB
is required for the expression of phenylalanine hydroxylase,
encoded by the first structural gene, phhA. In the absence of
the phhB gene, phhA by itself not only failed to complement E.
coli tyrosine auxotrophy, but was not expressed in E. coli as
indicated by SDS-PAGE. Dual catalytic and regulatory roles of
PhhB are an intriguing possibility in the context of the fact
that DCoH, the mammalian counterpart of PhhB, is a
bifunctional protein with enzymatic activity as 4a-
53

54
(A)
a -
1 =
I
l l
¡U 5:
phhR
-c -J-
phhA phhB phhC
_ P
I
3 *
(B)
TETRAHYDROBIOPTERIN
4a-CARBINOLAMINE
QUTNOID DTHYDROBIOPTERTN
FIG. 3-1. (A) Physical map of the phh operon in Pseudomonas
aeruginosa. The endonuclease restriction sites are shown at
the top. The arrows indicate the position of the genes and the
directions of transcription. Putative transcriptional
terminators (t inside a circle) are indicated. The proteins
encoded by the genes are as follows: phhR, a54 transcriptional
activator of the phh operon; phhA, phenylalanine hydroxylase;
phhB, 4a-carbinolamine dehydratase; and phhC, aromatic
aminotransferase. (B) Regeneration of the pterin cofactor for
phenylalanine hydroxylase. The enzymes involved are indicated
as follows: PhhA, phenylalanine hydroxylase; PhhB, 4a-
carbinolamine dehydratase; and DHPR, dihydropteridine
reductase.

55
carbinolamine dehydratase and regulatory activity as the
dimerization cofactor of HNFlu. In this chapter, I report the
results of studies aimed at elucidation of the extent and
nature of the regulatory function of PhhB protein.
Materials and Methods
Bacterial Strains, Plasmids, Phage, and Media
The bacterial strains, plasmids, and phage used in this
study are listed in Table 3-1. The LB and M9 formulations
(Sambrook et al., 1989) were used as growth media for E. coli
and P. aeruginosa. Pseudomonas isolation agar (Difco) was
used for isolating P. aeruginosa knockout mutants. Additions
of ampicillin (100 xg/ml), chloramphenicol (40 fjig/ml) ,
kanamycin (50 /xg/ml) mercuric chloride (15 /xg/ml) L-
phenylalanine (50 xg/ml) and thiamine (17 /g/ml) were made as
indicated. Agar was added at a final concentration of 2%
(w/v) for preparation of solid medium.
Recombinant DNA Techniques
Molecular cloning and DNA manipulation including plasmid
purification, restriction enzyme digestion, ligation, and
transformation were conducted by standard methods (Sambrook et
al., 1989) DNA fragments were purified from agarose gel with
a Geneclean kit (Bio 101). Electroporation (Invitrogen) was
used for simutaneous transformation of E. coli with two
compatible plasmids. Restriction enzymes, T4 DNA ligase, DNA-

56
Table 3-1. Bacterial strains, plasmids, and phages used in
this study
Strain or
plasmid
Relevant genotype
or description
Source or
reference
E. coli
BL21(DE3)
F" ompT hsdSB (rB~mB~) gal dcm;
with DE3, a X prophage carrying
the T7 RNA polymerase gene
Novagen
DH50!
F'AlacU169 4>80dlacZAM15 hsdR17
recAl endAl gyrA96 thi-1 relAl
supE44
GIBCO/BRL
LE392
F~el4~ (McrA~) hsdR514 (rkmk)
supE44 supF58 lacYl or a (lacIZY) 6
galK galT22 metBl trpR55
Sambrook et al
JP2255
aroF363 pheA361 phe0352 tyrA382
thi-1 strR712 lacYl xyl-15
Baldwin &
Davidson
JS1
SP1313§(phhA'-lacZ)
This study
S17-1
[RP4-2(Tc:Mu)(Km:Tn7)Tra(incP)]
pro hsdR recA Tpr Smr
Simon et al.
SP1313
zah-735: TnlOA (argF-lac) U1691A (tyrR)
Heatwole &
Somerville
P. aeruginosa
PAO -1
Prototroph
Holloway
JS101
PAO-1 phhA, Hgr
Song & Jensen
JS102
PAO-1 phhR, Hgr
Song & Jensen
JS103
PAO-1 phhB, Hgr
This study
JS104
PAO-1 phhC, Hgr
Plasmids
pUC18
pUC19
pACYC177
Ampr lac'IPOZ'
Ampr lac'IPOZ'
P15A replicn, Apr Kmr
Yanisch-
Perron et al.
Yanisch-
Perron et al.
Chang & Cohen

57
Table 3-
-1. (continued)
pETllb
T7lac promoter, lacI+ Apr Novagen
pET23
Tllac promoter, lad+ Apr Novagen
pGEM-3Z
T7 promoter, Apr Promega
pGST-DCoH In-frame protein fusion of
glutathione S-transferase and DCoH Citron et al.
pJSlO
phhAB, 2.5-kb Hindi fragment This study-
cloned into pGEM-3Z behind
the T7 promoter
pJSll
phhAB', 1.44-kb Hindi-EcoRV This study
fragment cloned into pACYC177
pJS12
phhAB, 2.5-kb Hindi fragment This study
cloned into pACYC177
pJS51
Hindi-BauMI fragment containing This study
truncated phhA' cloned into
pACYC177
pJS51Z
phhA'-lacZ transcriptional fusion This study
in pACYC177
pJS63
phh'ABC, BamHI-Hindi! fragment This study
cloned into pGEM-3Z behind
the T7 promoter
p JS72
phhA, PCR-generated fragment This study
containing the native ribosome
-binding site and PhhA-coding
region cloned into pET23 behind
T7lac promoter
pJS95
PhhA overexpression vector; This study
PhhA-coding region fused with T7
translational initiation signal
at Ndel site of pETllb
p JS96
PhhA overexpression vector; This study
phhA fused with T7 translational
initiation signal cloned into pUC19
behind lac promoter to constitutively
overexpress PhhA

58
Table 3-1. (continued)
pJS97
PhhA overexpression vector;
phhA fused with T7 translational
signal cloned into pTrc99A behind
trc promoter
This study
pJSlOl
Hgr-cassette, Apr
Song & Jensen
pJS105
Hindi-Ba/nHI PCR fragment
containing phhA' with a frameshift
This study
pJS105Z
phhA'-'lacZ protein fusion cloned
into pACYC177
This study
pJZ9
phhRABC, Apr
Zhao et al.
pJZ9-3a
phhAB, Apr
Zhao et al.
pJZ9-4
phh'ABC', Apr
Zhao et al.
pJZ9- 5
phhAB', Apr
Zhao et al.
pMC1871
lacZ protein fusion vector
Pharmacia
pTrc99A
Trc promoter, lad* Apr
Pharmacia
pUFRO 04
ColEl replicn, Cmr Mob+ mobP,
lacZa*
DeFeyter et
al.
Z1918
Promoterless lacZ, Apr
Schweizer
Phages
XRZ5
X'hla 'lacZ lacY*
Resental et
al.
XJS1
X$ {phhA'-lacZ) lacY* 'hla
This study

59
modifying enzymes (New England Biolab or Promega), Taq DNA
polymerase (Perkin-Elmer), and Vent DNA polymerase (New
England Biolab) were used as recommended by the suppliers.
Phenylalanine Hydroxylase Assay
E. coli JP2255 (pJZ9-3a) was grown at 37C in 500 ml of LB
broth supplemented with ampicillin (100 /xg/ml) and harvested
at the late-exponential phase of growth. Cell pellets were
resuspended into 8 ml of 10 mM potassium phosphate buffer (pH
7.4) containing 1 mM dithiothreitol (DTT), and the cells were
disrupted by sonication. The resulting extract was
centrifuged at 150,000 x g for 1 hr at 4C. The supernatant
was desalted using Sephadex G-25 and used as crude extract for
enzyme assay. Phenylalanine hydroxylase (PhhA) was assayed by
following tyrosine formation (Nakata et al., 1979).
Phenylalanine Hydroxylase Stimulation Assay
4a-Carbinolamine dehydratase activity in E. coli(pJZ9-4)
was assayed indirectly using the phenylalanine hydroxylase
stimulation assay (Citron et al. 1992). Reaction mixtures
containing 30 mM potassium phosphate (pH 8.3), catalase (1
mg/ml), 100 /xM NADH, 1 mM phenylalanine, 20 /xg
dihydropteridine reductase, 14.4 /xg rat liver phenylalanine
hydroxylase, and 2.9 /xM 6,7-dimethyltetrahydropterin were
incubated at 25C. Approximately 1 min after the reaction was
started, either buffer (control) or 15 /xg of the E. coli

60
crude extract containing PhhB or GST-DCoH was added. The
reaction was monitored at 340 nm for the oxidation of NADH by
dihydropteridine reductase as quinonoid dihydropterin was
recycled to tetrahydropterin.
Construction of phhA:lacZ Transcriptional and Translational
Fusions
For the transcriptional fusion (phhA'-lacZ) the HincII-
BamHI fragment containing the upstream region of phhA was
first cloned into pACYC177 that had been digested with Hindi
and BamHI, creating pJS51. A BamHI-cassette of a promoterless
lacZ gene from the plasmid Z1918 was then inserted at the
BamHI site of pJS51 in the same orientation as phhA to create
pJS51Z, which was used as a low-copy phhA'-lacZ fusion. A
single-copy fusion X(phhA'-lacZ) was obtained by transferring
the phhA'-lacZ fusion from pJS51Z into ARZ5 following the
procedure described by Yu and Reznikoff (1984) .
For the translational fusion (phhA'-'lacZ) the Hindi-
Bam HI fragment containing the upstream region of phhA was
generated by PCR with the upper primer 5'-
GACAGAGCAGGTAGATGGCGTT-3' and the lower primer
5'GGGATCCGGCTCGTGGGGCAGGCCGA-3' (BamHI site underlined). An
extra guanine nucleotide (G in bold) was added in the lower
primer to create the frameshift needed for an in-frame fusion
at the BamHI site to generate phhA' lacZ. The Hindi-BamHI
fragment with the frameshift was inserted into the HincII-
BamHI site of pACYC177 to create pJS105, and a BamHI-cassette

61
of truncated 'lacZ from pMC1871 was inserted into pJS105 to
create the translational fusion plasmid, pJS105Z.
ff-Galactosidase Assay
/3-Galactosidase activity was assayed under conditions of
proportionality as described by Miller (1972), and specific
activities are expressed in Miller units. The data are the
results of at least two independent assays.
Construction of PhhA and PhhB Expression Vectors
To express PhhA protein, expression plasmids pJS72 and
pJS95 were constructed. A PCR fragment containing the
complete coding region of phhA and the native ribosome-binding
site (RBS) was amplified with the upper primer 5'-
CATGGAGTCCGTATGAAAACGACGCA-3' (RBS underlined; ATG start codon
in bold) and the downstream primer 5'-
CTTGGTTGTCGCATGTGGGAGCGGCG-3', and cloned into pET23 behind
the T7lac promoter to create pJS72. pJS95 was constructed by
inserting the coding region of phhA into the translational
fusion vector pETllb. The coding region was amplified by PCR
with the upstream primer 5'-CCATATGAAAACGACGCAGTACGTG-3' and
the downstream primer 5'-CAAGTCTGGTTGTCGCATGTGGGAGCGGCG-3'.
The upper primer was made with a built-in Ndel site
(underlined) which allows fusion of phhA at the translational
start site (ATG in bold) with the T7 translational initiation
signals. To constitutively express the PhhA protein, the

62
phhA-coding region together with the upstream T7 translational
start signals were excised from pJS95 as a XJbal fragment and
cloned into pUC18 downstream of a lac promoter to create
pJS96. The Xbal fragment was also cloned into pTrc99A
downstream of the inducible trc promoter to create pJS97.
Two similar plasmids, pJSlO and pJS63, were constructed
to express PhhB. The Hindi fragment containing both phhA and
phhB gene was inserted into pGEM-3Z to create pJSlO, and the
BawHl-Hindlll fragment containing both phhB and phhC was
inserted into pGEM-3Z to create pJS63. The phhB gene was
under the control of a T7 promoter in both plasmids.
Preparation of PhhB-specific Polyclonal Antiserum
PhhB was partially purified by anion-exchange and gel-
filtration chromatography following the methods described by
Zhao et al., (1994). The partially purified PhhB was subject
to SDS-PAGE (12%) and the gel was stained with Commassie blue
R-250. The PhhB band was cut from the gel and used for the
production of polyclonal antiserum in rabbits (Cocalico
Biologicals, Inc., Reamstown, PA) Antiserum was purified by
using an Econo-Pac protein A column (Bio Rad) and further
absorbed with a total cell extract from the PhhB-deficient
mutant JS103.
SDS-PAGE and Western Blot Analysis
SDS-PAGE (12% gel) was performed with the Mini-PROTEAN II
cell (Bio-Rad) by the method of Laemmli (1970). Samples of

63
exponential-phase cells were collected by centrifugation, and
the cell pellets were suspended in gel-loading buffer and
heated at 100C for 10 min. Samples of 5-10 /x 1 were loaded
onto two SDS-polyacrylamide gels. After separation of the
proteins by electrophoresis, one gel was stained with
Coomassie blue and the other gel was used for blotting. When
crude extracts were used, equivalent amounts of protein were
loaded into each lane. Western blots were performed according
to Towbin et al. (1979). The proteins were
electrophoretically transferred onto nitrocellulose membranes
and reacted with the polyclonal antiserum at a dilution of
1:1000. Membranes were then incubated with secondary alkaline
phosphatase-labelled anti-rabbit antibodies at a dilution
(1:30,000) and developed by adding NBT and BICP as chromogenic
substrates (Bibco BRL) for alkaline phosphatase.
Gene Inactivation
Both phhB and phhC were inactivated following the method
described by Song and Jensen (1996) To generate the
truncated 'phhB' fragment (308bp), the upper primer 5'-
ACCCAAGCCCATTGCGAAGCCTGCCG-3', and the lower primer 5'-
GTGCGCGCCGCCATGATGAAATCGTT-3' were used. To generate the
truncated 'phhC' fragment (652bp), the upper primer 5'-
GTCGAGCAGGAAACCACCAAGA-3', and the lower primer 5'-
GTTGGCTACGCAGGTCGGTGAG-3' were used. Interruption of the phhB

64
or phhC gene in a Hgr isolate was confirmed by Southern
hybridization.
Southern Hybridization
Genomic DNA was extracted from the P. aeruginosa phhB'
strain by the method described by Silhavy et al. (1984).
Southern hybridization was performed as described by Sambrook
et al. (1989). The DNA was completely digested with EcoRI,
separated by electrophoresis in 1% agarose gel, and
transferred to a nylon membrane (Bio-Rad). The DNA was fixed
by baking the membrane under vacuum at 80C for 2 hr and
hybridized at 42C overnight with the truncated 'phhB' (the
same as used for gene inactivation) probes that had been
labeled with biotin-14-dATP using a BioNick labelling system
(GIBCO/BRL). The membrane was washed in 2X SSC (300 mM NaCl,
30 mM sodium citrate, pH 7.0) plus 0.1% SDS (twice for 3 min
each time at room temperature), in 0.2X SSC plus 0.1% SDS
(twice for 3 min each time at room temperature), and in 0.16X
SSC plus 0.1% SDS (twice for 15 min each time at 50C). The
probes were detected with the BluGene nonradioactive nucleic
acid detection system (GIBCO/BRL).
RESULTS
PhhB Has 4a-Carbinolamine Dehydratase Activity
PhhB is a homologue of an established 4a-carbinolamine
dehydratase. To confirm that PhhB catalyzes the 4a-

65
carbinolamine dehydratase reaction, I used the phenylalanine
hydroxylase stimulation assay where the utilization of 4a-
carbinolamine limits the rate of the hydroxylation (Huang et
al. 1973; Citron et al., 1992) Either PhhB or DCoH was able
to stimulate the phenylalanine hydroxylase reaction in E. coli
crude extracts where DCoH was used as a postive control (Fig.
3-2), indicating that PhhB protein has 4a-carbinolamine
dehydratase activity. Furthermore, using the expression
construct pJS63, PhhB protein was purified in the laboratory
of Dr. June E. Ayling at the University of South Alabama and
the 4a-carbinolamine dehydratase activity of PhhB was
confirmed by direct assay (personal communication).
Complementation of Tyrosine Auxotrophy by yhhA and phhB in
trans
Both phhA and phhB are needed for functional
complementation of E. coli tyrosine auxotrophy (Zhao et al.,
1994) If phhB functions as both a structural gene and a
regulatory gene in a fashion that parallels the mammalian
homologue, it would be expected to complement in the trans
comfiguration with respect to phhA. A trans-complementation
study was done in which phhA and phhB (or DCoH) were inserted
into two compatible plasmids, pJSll and pJZ9-4A (or pGST-
DCoH), respectively. The results (Table 3-2) did indeed show
that phhB was able to complement E. coli tyrosine auxotrophy
in trans with respect to phhA and that mammalian DCoH was able

66
Time (min)
FIG. 3-2. Stimulation of phenylalanine hydroxylase activity by
the addition (at the arrow) of a crude extract of E. coli
JP2255 containing PhhB or GST-DCoH fusion protein.
Approximately 1 min after the reaction was started, either
buffer, or 15 pig of the crude extract containing PhhB, or GST-
DCoH was added. The reaction was monitored at 340 nm for the
oxidation of NADH by dihydropteridine reductase as quinonoid
dihydropterin was recycled to a tetrabiohydropterin (see Fig.
3-1) .

67
Table 3-2. Complementation of an E. coli Tyr a mutant by phhA
and either phhB or DCoH gene in trans
Plasmid(s)
Relevent
genotype
Ability to complement15
E. coli Tyr- mutant
pJSll
phhA*
No
pJS12
(phhAB)*
Yes
pJZ9-4
phhB*
No
pJSll + pJZ9-4
phhA* phhB+
Yes
pGST-DCoH
DCoH*
No
pJSll+pGST-DCoH
phhA* DCoH*
Yes
a E. coli JP2255 (Phe Tyr ) mutant was used as the host strain
for this complementation study.
b E. coli JP2255 harboring various plasmids was plated on M9
+ phenylalanine plates supplemented with appropriate
antibiotics as selective agents.

68
to replace PhhB in the bacterial system. The results also
ruled out any possible cis effect of phhB on the expression of
phhA.
Expression of phhA in the Presence or Absence of phhB
The trans-complementation study confirmed that
complementation of E. coli tyrosine auxotrophy by phhA
requires the presence of phhB. To understand whether the
requirement of phhB for the complementation was due to
increased expression of phhA in the presence of PhhB, the
expression of phhA in the presence or absence of phhB in E.
coli JP2255 was studied through Western analysis (Fig. 3-3).
A substantial level of PhhA was still detected when only phhA
was present. The presence of either PhhB or DCoH in trans
indeed increased the expression of phhA, but only by about 2-
fold. Although this result indicated that PhhB may regulate
the expression of phhA, much more than the relatively modest
2-3 fold reduction in PhhA level observed in the phhB~ mutant
was expected. This was because little or no PhhA had been
detected in the absence of phhB (pJZ9-5) in E. coli, whereas
a very high level of PhhA was produced in the presence of phhB
(pJZ9-3a) (Zhao et al., 1994). The inconsistency between the
results of this study and that obtained by Zhao et al. (1994)
was found to be due to the incorrectly reported orientation of
the phhA insert in pJZ9-5. Sequencing the DNA insert in pJZ9-5
revealed that the orientation of the phhA gene was opposite to

PhhA
FIG. 3-3. Western blot analysis of PhhA expression in E. coli
JP2255. Proteins in the whole cell lysates of JP2255 carrying
various plasmids were separated by SDS-PAGE and reacted with
rabbit anti-PhhA polyclonal antibodies. Plasmids containing
the gene(s) shown above are as follows: phhA, pJSll; phhB,
pJZ9-4; DCoH, pGST-DCoH. pUC18 was used as the control
plasmid.

70
the lac promoter, rather than in the same orientation as
reported by Zhao, et al. (1994) .
PhhB Regulates Expression at the Post-transcriotional Level
To understand the regulatory role of PhhB in the
expression of phhA, I constructed phhA'-lacZ transcriptional
fusions in E. coli in both multi-copy form (pJS51Z) and a
single-copy form [X[phhA'-lacZ)] In both cases, the presence
of PhhB (pJZ9-4) or DCoH (pGST-DCoH) on a second plasmid did
not result in a higher level of (S-galactosidase as compared to
the control (pUC18) (Table 3-3), indicating that neither PhhB
nor DCoH functions at the transcriptional level. I then
constructed the translational fusion phhA'-'lacZ (pJS105Z)
(Table 3-3), and PhhB or DCoH was again provided in trans on
a second plasmid. (S-Galactosidase activity increased about
two-fold, indicating that PhhB regulates the expression of
phhA at the translational or post-transcriptional level. This
level of activation by PhhB is consistent with the results
obtained from Western-blot analysis in E. coli, where
similarly modest levels of activation in phhA expression were
observed.
Induction of The vhh Ooeron by Phenylalanine in P. aeruginosa
Expression of both phhA and phhB were coordinately
induced by phenylalanine in P. aeruginosa when grown on
minimal fructose medium (Fig. 3-4). The induction process is

71
Table 3-3. Levels of phhA expression by phhB and DCoH in
transa.
phhB or DCoH
in trans
(S-Galactosidase Activityb
pJS51Z
X {phhA' -lacZ)
PJS105Z
pJZ9 -4 (PhhB+)
180
13.2
13.1
pGST-DCoH (DCoH+)
191
14.1
10.4
pUC18 (control)
182
15.7
7.3
a Regulation of phhA expression was studied using lacZ as the
reporter gene and /3-galactosidase activities in
transcriptional fusions pJS51Z (phhA'-lacZ) (multicopy) and
X{phhA'-lacZ) (single copy), and translational fusion
pJS105Z {phhA'lacZ) (multicopy) was assayed in absence
(pUC18) or presence of PhhB (pJZ9-4) and DCoH (pGST-DCoH).
b /?-Galactosidase activities are reported in Miller units.

72
1 2 3 4 5 6 7
PhhA
PhhB
FIG. 3-4. Induction of the phh operon by phenylalanine
hydroxylase in P. aeruginosa. Proteins in the whole cell
lysates were separated by SDS-PAGE. Lane 1, E. coli JP2255
harboring pJSll and pJZ9-4; lanes 2-7, samples taken after
elapsed times of 0, 10, 30, 60, 90, 120 min, respectively,
following addition of 100 pig/ml phenylalanine at zero time.

73
quite protracted and requires 90 minutes or more reach a
maximum. A basal level of PhhB was expressed under non
inducing conditions as seen in lane 2 on the Western blot. It
is unknown whether this is due to expression from a promoter
upstream of phhA or from an internal promoter within the
coding sequence of phhA. Although the latter interpretation
comes to mind because no PhhA band is visualized at zero time,
this could be due to differing sensitivities of the antibody-
probes .
Effect of phhB Knockout in P. aerucrinosa
The phhB gene in P. aeruginosa was inactivated by
chromosomal insertion of the suicide plasmid pUFR/'phhB'/Hgr
through a single homologous crossover event (Fig. 3-5A). The
insertional inactivation of the phhB gene in the resulting
mutant was confirmed by Southern blot analysis (Fig. 3-5B).
The expression of PhhA in a phhB~ background was examined
by Western-blot methodology under fully induced conditions
fostered by growth on LB media (Fig. 3-6) The PhhA level
observed in the phhB~ mutant was reduced about 2-3 fold
compared to the wildtype parent. No PhhA was detected in
phhA~ mutant (negative control) The PhhB level was also
checked at the same time. A low basal level of PhhB was
detected in phhA~ mutant when compared with that in the
wildtype. No PhhB was detected in phhB~ mutant (negative
control).

FIG. 3-5. Inactivation of phhB in P. aeruginosa. (A) Schematic
representation for insertional inactivation of the chromosomal
phhB gene by the integration of the suicide plasmid
pUFR/'phhB'/Hgr through a single homologous crossover. The
resulting Hgr mutant does not contain a complete copy of phhB
gene, but instead has two truncated copies of phhB gene. (B)
Southern-blot analysis of chromosomal DNA from P. aeruginosa
PAO -1 wildtype (lane 1) and mutant JS103 (lane 2) Chromosomal
DNA was completely digested with EcoRI and probed at high
stringency with the truncated 'phhB' fragment of
pUFR/' phhB /Hgr.

75
P. aeruginosa
chromosome
Suicide plasmid
phhB'
mutant
(B)
1 2

PhhA
PhhB
FIG. 3-6. Western blot analysis of PhhA and PhhB expression
in P. aeruginosa PAO strains. Proteins in whole cell lysates
were separated by SDS-PAGE and probed with rabbit anti-PhhA or
anti-PhhB polyclonal antibodies.

77
The physiological effect of the phhB knockout mutant in
P. aeruginosa was examined. Inactivation of phhB abolished the
ability to grow on either phenylalanine or tyrosine as the
sole carbon source. However, interpretation of this result is
complicated by results obtained with a phhC knockout mutant
which was also not able to grow on either phenylalanine or
tyrosine as the sole carbon source. Since insertion of the
suicide plasmid into the chromosome of P. aeruginosa is
expected to create polar effects on the downstream genes in
the operon (as indeed seen in phhA knockout where amount of
PhhB expressed was dramatically decreased), it seems probable
that the physiological effect observed in phhB knockout mutant
is due to the polar effect on the expression of phhC.
Overexpression of PhhA and PhhB Proteins
Although PhhA was expressed to detectable levels in both
E. coli (Fig. 3-3) and P. aeruginosa (Fig. 3-6) in the absence
of phhB, initial attempts to express PhhA at high levels in
the absence of PhhB in E. coli were unsuccessful. I then
employed a T7 overexpression system (see Methods) to express
PhhA in E. coli BL21(DE3) under induction conditions not
requiring growth (Fig. 3-7A). When phhA was expressed from a
native ribosomal binding site in pJS72, high PhhA levels were
produced after IPTG induction for 3 h (Lane 3, Fig. 3-7B).
When phhA was expressed from $10 translational signals
(pJS95), higher levels of PhhA was made after IPTG induction

78
(A)
pJS72 [P0
RBS
PhhA
C A T|G GAGlTCCGTATG AAAACG...
Native translational signal MKT
RBS
PhhA
pJS95 lP0||GAAGGAGAlTATACATATG AAAACG
T7 translational signal M K T
1 2 3 4 5
FIG. 3-7. Expression of PhhA using T7 expression system in E.
coli. (A) Construction of PhhA expression plasmids, pJS72
(with the native translational signals) and pJS95 (with 010
translational signals). (B) SDS-PAGE analysis of expressed
PhhA protein. Proteins in whole cell lysate were separated by
SDS-PAGE and stained with Coomassie blue. Lane 1, molecular
weight markers; lanes 2&3, BL21(DE3) harboring pJS72, before
and after 1 mM IPTG induction for 3 h at 30C, respectively;
lanes 4&5, BL21(DE3) harboring pJS95, before and after 1 mM
IPTG induction for 3 h at 30C, respectively.

79
for 3 h (Lane 5, Fig. 3-7B) PhhA produced using this T7
overexpression system in E. coli is active. Phenylalanine
hydroxylase activity was assayed in crude extract of E. coli
BL21(DE3) harboring either pJS72 or pJS95 after IPTG
induction, and compared with activity present in E. coli
JP2255 (pJS9-3a). Expression of phhA from 010 translational
signals resulted in over a five-fold increase in phenylalanine
hydroxylase activity (Table 3-4).
PhhB protein was expressed very well from a lac promoter
in pJZ9-4 by Zhao et al. (1994) We obtained an even higher
level of PhhB by using the T7 overexpression system (Fig. 3-
8) The pJSlO construct used contains both phhA and phhB.
After IPTG induction, only PhhB was overexpressed and little
PhhA was made (Lane 3, Fig. 3-8) A similar phenomenon was
observed with pJS63 construct containing both phhB and phhC
where only PhhB protein was overproduced (Lane 5, Fig. 3-8).
These results indicated that phhB has stronger translational
signals and, therefore, is preferably translated over phhA and
phhC. PhhB protein expressed is fully active, and has been
purified and characterized from the latter construct by Ayling
(unpublished results).
Expression of PhhA without PhhB in E. coli Has Growth
Inhibitory Effects
Even though a low level of PhhA was produced from pJSll
in the absence of PhhB, it did not complement the E. coli
tyrosine auxotrophy in JP2255. A possible explanation was that

80
Table 3-4. Phenylalanine hydroxylase activities in different
expression clones3
Expression
clones
Specific Activity
(nanomoles/min/mg)
JP2255/pJZ9-3a
91.7
BL21(DE3)/pJS72
72.6
BL21(DE3)/pJS95
379.2
3 Cells of E. coli JP2255 harboring pJZ9-3a were grown in LB
broth at 37C and harvested at late exponential phase; cells
of BL21(DE3) harboring pJS72 or pJS95 were grown in LB broth
at 37C to 0.D=1 and induced for 3 hr by addition of 1 mM
IPTG. Crude extracts were used as the enzyme sources.

81
FIG. 3-8. Expression of PhhB using T7 expression system in E.
coli. Proteins in whole cell lysates were separated by SDS-
PAGE and stained with Coomassie blue. Lane 1, molecular weight
markers; Lanes 2&3, BL21(DE3)/pJSlO, before and after 1 mM
IPTG induction for 3 h at 30C, respectively. Lanes 4&5,
BL21(DE3)/pJS63, before and after 1 mM IPTG induction for 3 h
at 30C, respectively.

82
the PhhA produced from pJSll was not high enough to result in
complementation. Thus, attempts were made to construct clones
from which higher levels of PhhA could be produced to see
whether complementation of E. coli tyrosine auxotrophy would
then occur. Two constructs pJS96 and pJS97 were made with a
Xbal fragment from pJS95 carrying phhA gene (fused with 010
translational signals) cloned into pUC18 behind a lac promoter
or into pTrc99A behind a trc promoter, respectively (Fig. 3-
9). Both plasmids were found to be unstable and cells tended
to lose the plasmid, especially with pJS96 where phhA is
constitutively expressed at a very high level from the lac
promoter. An elapsed time of several days was required to see
pinpoint colonies with pJS96. E. coli carrying pJS97 was able
to grow to a pinpoint colony overnight at 37C without IPTG
induction, while it took two or more days to see the pinpoint
colonies on the plate with IPTG induction. These results
indicated that high level of phhA expression in the absence of
phhB triggers potent growth inhibition. Although the
expression of phhA from the trc promoter on pJS97 has to be
induced because of the presence of a lacT1 gene on the
plasmid, a high level of PhhA is still produced without IPTG
induction because the trc promoter is very strong and leaky.
Because PhhA was constitutively expressed from pJS96 at
high level, the cells were not able to maintain the plasmid.
Therefore, a high level of PhhA could not be overproduced in
E. coli (pJS97) (Lane 1, Fig. 3-9B). Since a lower level of

83
RBS
PhhA
pJS96|P/flcl|GAAGGAG|TATACATATGAAAACG...
T7 translational signal MKT
RBS
'PhhA
p.IS97lPfr?l|GAAGGAOA|TATACATATOAAAACG...
T7 translational signal
M K
(B) 1 2 3 4 5
97.4
66.2
45.0
31.0
21.5
14.5
FIG. 3-9. Expression of PhhA in E. coli JP2255. (A)
Construction of PhhA expression plasmids, pJS96 and pJS97.
(B) SDS-PAGE analysis of PhhA expression. Proteins in whole
cell lysates were separated by SDS-PAGE and stained with
Coomassie blue. Lane 1, JP2255/pJS96; lane 2, JP2255/pTrc99A
(control); lanes 3&4, JP2255/pJS97, before and after 1 mM IPTG
induction for 3 h, respectively; lane 5, molecular weight
markers.

84
PhhA was produced from pJS97 due to the presence of laclq gene
one the plasmid, the cells carrying pJS97 were better able to
maitain the plasmid, thus being able to overproduce PhhA after
IPTG induction (Lane 4, Fig. 3-9B). Neither pJS96 nor pJS97 by
itself was able to complement E. coli tyrosine auxotrophy.
However, they were able to complement the auxotrophy when phhB
was provided in trans on pJZ9-4 (data not shown). This result
indicates that PhhB was able to remove the inhibitory effect
imposed by overproduction of PhhA on the host cells. To assure
that apparent growth inhibition was not due to excessive
conversion of phenylalanine to tyrosine in the phenylalanine
auxotrophy background of strain JP2255, the plasmids were
moved to a prototrophic background. E. coli DH5a carrying
either plasmid was found to develop only pinpoint colonies on
LB + Amp plates without IPTG induction, confirming that
expression of phhA at higher levels created a general
inhibitory effect on the growth of E. coli.
Discussion
Regulatory Role of PhhB?
The mammalian PhhB homolog, DCoH, has both catalytic and
regulatory functions. It was initially thought that P.
aeruginosa phhB exerted an essential positive role in
expression of phhA since constructs lacking phhB did not
express phhA, as monitored by SDS-PAGE (Zhao et al 1994).

85
Although phhA is in fact expressed in the absence of phhB,
phhB does appear to exercise a positive regulatory role in the
expression of phhA with a relatively modest effect of perhaps
2-3 fold. The transcriptional fusion approach seems to have
eliminated the possibility of this regulation being at the
transcriptional level, with the reservation that the
experiments were performed in an E. coli background. I have
shown that PhhR, the positive regulator of the phh operon in
P. aeruginosa, does not interact properly with the E. coli a54
machinery to activate phhA expression. But it is still
possible that PhhB interacts with PhhR as a co-activator
entity in the native P. aeruginosa organism. This would amount
to a parallel with the mammalian system where DCoH is a co
activator for HNFlo' (an upstream enhancer element that can be
considered comparable to PhhR).
However, I have presented results to show that PhhB does
enhance expression of PhhA at the post -transcriptonal level in
E. coli. The 2-3 fold enhancement effects obtained correspond
with the similar 2-3 fold magnitude of effect seen in P.
aeruginosa, when comparing Western blots of PhhA in PhhB+ and
PhhB- backgrounds. If PhhB regulates at a post-transcriptional
level, several possibilities envisioned include: (i) the
sencondary structure of phhA mRNA might mask the ribosomal
binding site. A stem-loop structure has been located in this
area. Binding of PhhB in this region might disrupt the
secondary mRNA structure and enhance translational initiation.

86
(ii) PhhB might bind phhA mRNA in the 3' region and protect
against nuclease-catalyzed mRNA degradation, (iii) PhhB may
complex with PhhA. This complex may protect PhhA from
proteolysis. I have in fact obtained preliminary evidence for
a PhhA-PhhB complex.
Rationale for Positive Regulatory Role of PhhB
When PhhA is highly expressed in the absence of PhhB, the
E. coli host cells become subject to drastic growth
inhibition. The reasonable explanation for this effect of
PhhB is that removal of the inhibitory effect generated by
PhhA reaction is direct result of 4a-carbinolamine
dehydratase activity. It is known that in the absence of 4a-
carbinolamine dehydratase activity, a 7-isomer of
tetrahydrobiopterin is generated during the reaction catalyzed
by phenylalanine hydroxylase (Davis et al. 1992) and this 7-
isomer is a potent inhibitor of phenylalanine hydrxoylase in
the mammalian sytem. Since potent growth inhibition persists
in wildtype E. coli backgrounds where overexpressed
phenylalanine hydroxylase has no purpose, inhibition cannot be
attributed to inhibition of phenylalanine hydroxylase by the
7-isomer. Two explanations accounting for general growth
inhibition come to mind, (i) The reduced pterin cofactor of E.
coli may be depleted because PhhB is not present for
recycling. Consequently, some pterin-dependent enzymes that
are essential for growth may become limiting. (ii)

87
Alternatively, the 7-isomer may be a potent inhibitor of a
pterin-dependent enzyme needed for growth. Possible targets of
inhibition could be dihydropteridine reductase or
dihydrofolate reductase. The E. coli dihydropteridine
reductase has been reported to possess broad specificity for
pteridine compounds (Vasudevan et al. 1992).
When phhA was expressed at relatively low levels in E.
coli no growth inhibition, or at least no severe growth
inhibition occured. However, this PhhA was evidently not
functional in vivo ( in the absence of PhhB) because
complementation of tyrosine auxotrophy in the presence of
exogenous L-phenylalanine was unsuccessful. On the other hand,
the joint presence of PhhA and PhhB readily allowed functional
complementation. I concluded that PhhA is an essential target
of the 7-isomer. At the low levels of PhhA expression in the
absence of PhhB, the 7-isomer is generated and inhibits PhhA
function--but not enough 7-isomer is produced to cause general
growth inhibiton. At high levels of PhhA expression in the
absence of PhhB, sufficient 7-isomer is produced to inhibit
one or more enzymes essential for the growth.
With the above background, a rationale to explain a basis
for selection of regulation of PhhA by PhhB is apparent. If it
is correct that the 7-isomer generated from the carbinolamine
pterin product of the PhhA reaction has general antimetabolite
properties, then the significance of PhhB goes beyond its

88
catalytic capability. It also diverts the carbinolamine
substrate from an undesirable nonenzymatic fate.

REFERENCES
Abita, J. P., Milstien, S., Chang, N., and Kaufman, S. 1976.
In vitro activation of rat liver phenylalanine
hydroxylase by phosphorylation. J. Biol. Chem. 251:5310-
5314 .
Argaet, V.P., Wilson, T.J., and Davidson, B.E. 1994.
Purification of the Escherichia coli regulatory protein
TyrR and analysis of its interactions with ATP, tyrosine,
phenylalanine and Tryptophan. J. Biol. Chem. 269:5171-
5178 .
Armarego, W. L., Randles, D., and Waring, P. 1984.
Dihydropteridine reductase (DHPR), its cofactors, and its
mode of action. Med. Res. Rev. 43:267-321.
Bailey, S. W., Boerth, S. R., Dillard, S. B., and Ayling, J.
E. 1993. The mechanism of cofactor regeneration during
phenylalanine hydroxylation. Adv. Exp. Med. Biol. 338:47-
54 .
Baldwin, G.S., and Davidson, B.E. 1981. A kinetic and
structural comparison of chorismate mutase/prephenate
dehydratase from mutant strains of Escherichia coli K-12
defective in the pheA gene. Arch. Biochem. Biophys.
211:66-75 .
Carr, R. T. and Benkovic, S. J. 1993. An examination of the
copper requirement of phenylalanine hydroxylase from
Chromobacterium violaceum. Biochemistry. 32:14132-14138.
Carr, R. T., Balasubramanian, S. Hawkins, P. C., and
Benkovic, S. J. 1995. Mechanism of metal-independent
hydroxylation by Chromobacterium violaceum phenylalanine
hdyroxylase. Biochemistry. 34:7525-7532.
Chang, A.C.Y., and Cohen, S.N. 1978. Construction and
characterization of amplifiable multicopy DNA cloning
vehicles derived from the pl5A cryptic miniplasmid. J.
Bacteriol. 134:1141-1156.
89

90
Citron, B. A., Davis, M. D., Milstien, S., Gutierrez, J. ,
Mendel, D. B., Crabtree, G. R., and Kaufman, S. 1992.
Identity of 4a-carbinolamine dehydratase, a component of
the phenylalanine hydroxylation system, and DCoH, a
transregulator of homeodomain proteins. Proc. Natl. Acad.
Sci. USA. 89:11891-11894.
Cui, J., and Somerville, R.L. 1993. The TyrR protein of
Escherichia coli: analysis by limited proteolysis of
domain structural and ligand-mediated conformational
changes. J. Biol. Chem. 268:5040-5047.
Curtius, H. C., Adler, C., Rebrin, I., Heizmann, C., and
Ghisla, S. 1990. 7-Substituted pterins: formation during
phenylalanine hydroxylation in the absence of
dehydratase. Biochem. Biophys. Res. Commun. 172:1060-
1066 .
Davis, M. D. and Kaufman, S. 1989. Evidence for the formation
of the 4a-carbinolamine during the tyrosine-dependent
oxidation of tetrahydrobiopterin by rat liver
phenylalanine hydroxylase. J. Biol. Chem. 264:8585-8596.
Davis, M. D. and Kaufman, S. 1991. Studies on the partially
uncoupled oxidation of tetrahydropterins by phenylalanine
hydroxylase. Neurochem. Res. 16:813-819.
Davis, M. D., P. Ribeiro, J. Tipper, and S. Kaufman. 1992.
"7-tetrahydrobiopterin," a naturally occourring analogue
of tetrahydrobiopterin, is a cofactor for and a potential
inhibitor of the aromatic amino acid hydroxylases. Proc.
Natl. Acad. Sci. USA. 89:10109-10113.
Davis, M. D., Parniak, M. A., Kaufman, S., and Kemper, E.
1996. Structure-function relationships of phenylalanine
hydroxylase revealed by rediation target analysis. Arch.
Biochem. Biophys. 325:235-241.
Decicco, B. T. and Umbreit, W. W. 1964. Utilization of
aromatic amino acids by Hydrogenomonas facilis. J.
Bacteriol. 83:1590-1594.
DeFeyter, R., Kado, C. I., and Gabriel, D. W. 1990. Small,
stable shuttle vectors for use in Xanthomonas. Gene
88:65-72.
Devereux, J. P. Haeberli, and 0. Smithies. 1984. A
comprehensive set of sequence analysis programs for the
VAX. Nucleic Acids Res. 12:387-395.

91
Dilella, A. G., Kwok, S. C. M., Ledley, F. D., Marvit, J.( and
Woo, S. L. C. 1986. Molecular structure and polymorphic
map of the human phenylalanine hydroxylase gene.
Biochemistry. 25:743749.
Endrizzi, J. A., Cronk, J. D., Wang, W. Crabtree, G. R. ,
and Alber, T. 1995. Crystal structure of DCoH, a
bifunctional, protein-binding transcriptional
coactivator. Science. 268:556-559.
Essar, D. W. Eberly, L., Han, C.-Y., and Crawford, I. P.
1990. DNA sequences and characterization of four early
genes of the tryptophan pathway in Pseudomonas
aeruginosa. J. Bacteriol. 172:853-866.
Faust, D. M., Catherin, A. M., Barbaux, S., Belkadi, L.,
Imaizumi-Sherrer, T., and Weiss, M. C. 1996. The activity
of the highly inducible mouse phenylalanine hydroxylase
gene promoter is dependent upon a tissue-specific,
hormone-inducible enhancer. Mol. Cell. Biol. 16:3125-
3137 .
Friedman, D.I. 1988. Integration host factor: a protein for
all reasons. Cell 55:545-554.
Friedrich, B. and H. G. Schlegel. 1972. Die hydroxylierung
von phenyalanin durch Hydrogenomonas eutrophus H16 Arch.
Mikrobiol. 83:17-31.
Gambill, D. B., and Summers, A. O. 1985. Versatile mercury-
resistant cloning and expression vectors. Gene 39:293-297
Gu, W. and Jensen, R.A. 1996. Evolutionary recruitment of
biochemically specialized subdivisions of Family I within
the protein Superfamily of aminotransferases. J.
Bacteriol. 178:2161-2171.
Gudinchet, F., Maeder, P., Meuli, R. A., Deonna, T., and
Mathieu, J. M. 1992. Cranial CT and MRI in malignant
phenylketonuria. Pediatr. Radiol. 22:223-224.
Guldberg, P., H. L. Levy, W. B. Hanley, R. Kock, R. Mataln,
B. M. Rouse, F. Trefz, F. de la Cruz, K. F. Henriksen,
and F. Guttler. 1996. Phenylalanine hydroxylase gene
mutations in the United States: report from the Maternal
PKU Collaborative Study. Am. J. Hum. Genet. 59:84-94.
Guroff, G. and T. Ito. 1963. Induced soluble
phenylalanine hydroxylase from Pseudomonas sp. grown on
phenylalanine or tyrosine. Biochim. Biophys. Acta
77:159-161.

92
Haldenwang, W. G. 1995. The sigma factors of Bacillus
subtilis. Microbiol. Rev. 59:1-30.
Heatwole, V.M., and Somerville, R.L. 1991. The tryptophan-
specific permease gene, mtr, is differentially regulated
by the tryptophan and tyrosine repressors in Escherichia
coli K-12. J. Bacteriol. 173:3601-3604.
Holloway, B.W. (1955) Genetic recombination in Pseudomonas
aeruginosa. J. Gen. Microbiol. 13:572-581.
Huang, C. Y., Max, E. E., and Kaufman, S. 1973. Purification
and characterization of phenylalanine hydroxylase-
stimulating protein from rat liver. J. Biol. Chem.
248 :4235-4241.
Ivey-Hoyle, M., Steege, D.A. 1992. Mutational analysis of an
inherently defective translation initiation site. J. Mol.
Biol. 224:1039-1054.
Jakoby, G. A. 1964. The induction and repression of amino
acid oxidation in Pseudomonas fluorescens. Biochem. J.
92:1-8.
Joerger, R.D., Jacobson, M.R., and Bishop, P.E. 1989. Two
nifA-like genes required for expression of alternative
nitrogenases by Azotobacter vinelandii. J. Bacteriol.
171:3258-3267.
Kamoun, S., Kamdar, H. V. Tola, E., and Kado, C. I. 1992.
Incompatible interactions between crucifers and
Xanthomonas campestris invole a vascular hypersensitive
response: role of the hrpX locus. Mol. Plant-Microbe
Interact. 5:22-23.
Kappock, T. J., Hartins, P. C., Friedenberg, S., and
Caradonna, J. P. 1995. Spectroscopic properties of
unphosphorylated rat hepatic phenylalanine hydroxylase
expressed in Escherichia coli. Comparison of resting and
activated states. J. Biol. Chem. 270:30532-30544.
Kaufman, S. 1975. Studies on the mechanism of phenylalanine
hydroxylase:detection of an intermediate, p.495-523. In
W. Pfleidere (ed.), Chemistry and biology of pteridines.
Walter de Gruyter, Berlin.
Kaufman, S. 1986. Regulation of the activity of hepatic
phenylalanine hydroxylase. Adv. Enz. Regulation. 25:37-
64 .

93
Kaufman, S. 1987. Phenylalanine 4-monooxygenase from rat
liver. Meth. in Enzymol. 142:3-27.
Kaufman, S., Citron, B. A., Davis, M. and Milstien, S.
1993. The isolation and characterization of clones of 4a-
hydroxytetrahydrobiopterin dehydratase. Adv. Exp. Med.
Biol. 338:97-102.
Kaufman, S and Fisher, D. B. 1970. Purification and some
physical properties of phenylalanine hydroxylase from rat
liver. J. Biol. Chem. 245:4745-4750.
Kwok, T., Yang, J. Pittard, A.J., Wilson, T.J., and Davidson,
B.E. 1995. Analysis of an Escherichia coli mutant TyrR
protein with impaired capacity for tyrosine-mediated
repression, but still able to activate at a70 promoters.
Mol. Microbiol. 17:471-481.
Laemmli, U.K. 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature
(London) 227:680-685.
Lazarus, R. A., S. J. Benkovic and S. Kaufman. 1983.
Phenylalanine hydroxylase stimulation protein is a 4a-
carbinolamine dehydratase. J. Biol. Chem. 258:10960-
10962 .
Letendre, C. H. G. Dickens, and G. Guroff. 1974. The
tryptophan hydroxylase of Chromobacterium violaceum. J.
Biol. Chem. 249:7186-7191.
Letendre. C. H. G. Dickens, and G. Guroff. 1975.
Phenylalanine hydroxylase from Pseudomonas sp. (ATCC
11299a) : purification, molecular weight, and influence of
tyrosine metabolites on activation and hydroxylation. J.
Biol. Chem. 254:1829-1833.
Mendel, D. B. and Crabtree, G. R. 1991. HNF-1, a member of a
novel class of dimerizing homeodomain proteins. J. Biol.
Chem. 266:677-680.
Miller, J. H. 1972. Experiments in molecular genetics. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, N. Y.
Morrett, E., and Segovia, L. (1993) The bacterial enhances-
binding protein family: mechanism of action and
phylogenetic relationship of their functional domains. J.
Bacteriol. 175:6067-6074.
Nakata, H. Yamauchi, T. and Fujisawa, H. 1979. Phenylalanine
hydroxylase from Chromobacterium violaceum: purification
and characterization. J. Biol. Chem. 266:18454-18459.

94
Onishi, A., L. J. Liotta, and S. J. Benkovic. 1991.
Cloning and expression of Chromobacterium violaceum
phenylalanine hydroxylase in Escherichia coli and
comparison of aromatic amino acid sequence with mammalian
aromatic amino acid hydroxylases. J. Biol. Chem.
266 :18454-18459.
Patel, N., Stemark-Cox, S., and Jensen, R.A. 1978.
Enzymological basis of reluctant auxotrophy for
phenylalanine and tyrosine in Pseudomonas aeruginosa. J.
Biol. Chem. 253:2972-2978.
Pember, S. 0. J. Villafranca, and S. J. Benkovic. 1986.
Phenylalanine hydroxylase from Chromobacterium violaceum
is a copper-containing monooxygenase: Kinetics of the
reductive activation of the enzyme. Biochemistry
25:6611-6619 .
Pittard, A.J. 1996. Biosynthesis of the aromatic amino acids.
In Escherichia coli and Salmonella, pp 458-484. (F,C.
Neidhardt, ed.) ASM Publications. Washington, D. C.
Pittard, A.J., and Davidson, B.E. 1991. TyrR protein of
Escherichia coli and its role as repressor and activator.
Mol. Microbiol. 5:1585-1592.
Pogge-yon-Strandmann, E. and Ryffel, G. U. 1995.
Developmental expression of the maternal protein XDCoH,
the dimerization cofactor of the homeoprotein LFB1
(HNF1). Development. 121:1217-1226.
Pontoglio, M., Barra, J., Hadchouel, M., Doyen, A., Kress,
C., Bach, J. P., Babinet, C., and Yaniv, M. 1996.
Hepatocyte nuclear factor 1 inactivation results in
hepatic dysfunction, phenylketonuria, and renal Fanconi
syndrome. Cell 84:575-585.
Rosentel, J.K., Healy, F., Maupin-Furlow, J.A., Lee, J.H., and
Shanmugam, K.T. 1995. Molybdate and regulation of mod
(molybdate transport), fdhF, and hyc (formate
hydrogenlyase) operons in Escherichia coli. J.
Bacteriol. 177:4857-4864.
Sambrook, J., Fritsch, E. F., and Maniatis, T. 1989.
Molecular Cloning: A Laboratory Manual, 2nd ed. Cold
Spring Harbor, N.Y. : Cold Spring Harbor Laboratory Press.
Schweizer, H.P. 1993. Two plasmids, X1918 and Z1918, for easy
recovery of the xylE and lacZ reporter genes. Gene
134 : 89-91.

95
Shingler, V. 1996. Signal sensing by cr54-dependent reulators:
derepression as a control mechanism. Mol. Microbiol.
19:409-416.
Silhavy, T. J. Berman, M. L., and Enquist, L. W. 184. DNA
extraction from bacterial cells, p.137-139. In
Experiments with gene fusions. Cold Spring Harbor
Laboratory, Cold Spring Harbor, N.Y.
Simon, R., Priefer, U., and Phler, A. 1983. A broad host
range mobilization system for in vivo genetic
engineering:transposon mutagenesis in gram negative
bacteria. Bio/Technology 1:784-791.
Simons, R.W., Houman, F., and Kleckner, N. 1987. Improved
single and multicopy lac-based cloning vectors for
protein and operon fusions. Gene 53:85-96.
Song, J. and Jensen, R. A. 1996. PhhR, a divergently
trancribed activator of the phenylalanine hydroxylase
gene cluster of Pseudomonas aeruginosa. Mol. Microbiol.
22:000-000 (in press).
Stock, J.B., Ninfa, A.J., and Stock, A.M. 1989. Protein
phosphorylation and regulation of adaptive responses in
bacteria. Microbiol. Rev. 53:450-490.
Totten, P.A., Lara, J.C., and Lory, S. 1990. The rpoN gene
product of Pseudomonas aeruginosa is required for
expression of diverse genes, including the flagellin
gene. J. Bacteriol. 172:389-396.
Tourian, A. 1971. Activation of phenylalanine hydroxylase by
phenylalanine. Biochim. Biophys. Acta. 242:345-354.
Towbin, H, Staehelin, T., and Gordon, J. 1979.
Electrophoretic transfer of proteins from polyacrylamide
gels to nitrocellulose sheets: procedure and some
applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.
Vasudevan, S. G., Shaw, D. C., and Armarego, W. L. F., 1988.
Dihydropteridine reductase from Escherichia coli.
Biochem. J. 255:581-588.
Vasudevan, S. G., Paal, B. and Armarego, W. L. F. 19 92.
Dihydropteridine reductase from Escherichia coli exhibits
dihydrofolate reductase activity. Bio. Chem. Hoppe-
Seyler. 373:1067-1073.
Wang, W. Carey, M., and Gralla, J. D. 1992. Polymerase II
promoter activation: closed complex formation and ATP-
driven start site opening. Science 255:450-453.

96
Williams, C. D., Dickens, G. Letendre, C. H., Guroff, G.,
Haines, C., and Shiota, T. 1976. Isolation and
characterization of dihydropteridine reductase from
Pseudomonas species. J. Bacteriol. 127:1197-1207.
Xia, T., and Jensen, R.A. 1990. A single cyclohexadienyl
dehydrogenase specifies the prephenate dehydrogenase and
arogenate dehydrogenase components of the dual pathways
to L-tyrosine in Pseudomonas aeruginosa. J. Biol. Chem.
265:200333-20036.
Yang, J. Ganesan, S., Sarsero, J. and Pittard, A.J. 1993.
A genetic analysis of various functions of the TyrR
protein of Escherichia coli. J. Bacteriol. 175:1767-1776.
Yanisch-Perron, C., Vieira, J., and Messing, J. 1985. Improved
M13 phage cloning vectors and host strains: nucleotide
sequences of the M13mpl8 and pUC vectors. Gene 33:103-
119 .
Yu, X.-M. and Reznikoff, W. S. 1984. Deletion analysis of the
CAP-cAMP binding site of the Escherichia coli lactose
promoter. Nucleic Acids Res. 12:5449-5464.
Zhao, G. Xia, T., Aldrich, H. and Jensen, R.A. 1993.
Cyclohexadienyl dehydratase from Pseudomonas aeruginosa
is a periplasmic protein. J. Gen. Microbiol. 139:807-813.
Zhao, G., Xia, T., Song, J., and Jensen, R. A. 1994.
Pseudomonas aeruginosa possesses homologues of mammalian
phenylalanine hydroxylase and 4a-carbinolamine
dehydratase/DCoH as part of a three-component gene
cluster. Proc. Natl. Acad. Sci. USA 91:1366-1370.

BIOGRAPHICAL SKETCH
Jian Song was born on November 2, 1963, in Hebei
Province, China. He completed his elementary and high school
education in Xinhe County, Hebei Province. In 1980, he
attended the Agricultural University of Hebei, where he
majored in plant protection. He received his B.S. degree in
1984, then worked at the Institute of Plant Protection, Hebei
Academy of Agricultural and Forestry Sciences until 1987. He
was awarded a scholarship by Hebei Academy of Agricultural and
Forestry Sciences to come to the Department of Entomology and
Plant Pathology at the University of Tennessee, Knoxville for
graduate study in 1988. He studied the interactions among the
plant, aphid, and parasitoid under the supervision of Dr.
Charles D. Pless. He received his M.S. degree in entomology
in May, 1990. He then went to the Department of Entomology
and Nematology at the University of Florida to continue
graduate study toward a Ph.D. in entomology. He worked on
insect toxicology under the supervision of Dr. Simon Yu until
October, 1991. He then joined Dr. Roy A. Jensen's group at
the Department of Microbiology and Cell Science in January,
1992 and studied regulation of phenylalanine hydroxylase
system in Pseudomonas aeruginosa, and obtained his Ph.D.
degree in May, 1997. He has accepted a postdoctoral position
97

98
in Dr. Vojo Deretic's Laboratory in the Department of
Microbiology and Immunology at the University of Michigan, and
will join the laboratory in January, 1997 to work on
Mycobacterium tuberculosis.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
l-xft-M -N jir
Roy A. Jipsen, lichair
Professor of Microbiolgy and
Cell Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree^of Doctor of Philosophy.
Dean W. Gabriel
Professor of Plant Pathology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Lonnie 0. Ingram
Professor of Microbiolgy and
Cell Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degrees of Doctor of .philosophy.
imes F. Preston
'Professor of Microbiolgy and
Cell Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Keelnatham T. Shanmugam
Professor of Microbiolgy and
Cell Science

This dissertation was submitted to the Graduate Faculty
of the College of Agriculture and to the Graduate School and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
May, 1997
, ^College of ^gr
Dean
griculture
Dean, Graduate School



CHAPTER 3
BIFUNCTIONAL PhhB REGULATES THE EXPRESSION OF
PHENYLALANINE HYDROXYLASE IN Pseudomonas aeruginosa
Introduction
Mammalian 4a-carbinolamine dehydratase was initially
known for its catalytic activity of converting 4a-
carbinolamine to quinonoid dihydrobiopterin in regenerating
the tetrahydrobiopterin for phenylalanine hydroxylase (Fig. 3-
1B) Later, it was found to be synonymous with DCoH, the
dimerization cofactor for hepatic nuclear factor 1 alpha (HNF-
la) (Citron et al., 1992) .
A homolog of the mammalian DCoH, PhhB, was found in
Pseudomonas aeruginosa by Zhao et al. (19 94) The PhhB
protein is encoded by the second structural gene, phhB, of the
phh operon (Fig. 3-1A) Zhao et al (1994) reported that phhB
is required for the expression of phenylalanine hydroxylase,
encoded by the first structural gene, phhA. In the absence of
the phhB gene, phhA by itself not only failed to complement E.
coli tyrosine auxotrophy, but was not expressed in E. coli as
indicated by SDS-PAGE. Dual catalytic and regulatory roles of
PhhB are an intriguing possibility in the context of the fact
that DCoH, the mammalian counterpart of PhhB, is a
bifunctional protein with enzymatic activity as 4a-
53


79
for 3 h (Lane 5, Fig. 3-7B) PhhA produced using this T7
overexpression system in E. coli is active. Phenylalanine
hydroxylase activity was assayed in crude extract of E. coli
BL21(DE3) harboring either pJS72 or pJS95 after IPTG
induction, and compared with activity present in E. coli
JP2255 (pJS9-3a). Expression of phhA from 010 translational
signals resulted in over a five-fold increase in phenylalanine
hydroxylase activity (Table 3-4).
PhhB protein was expressed very well from a lac promoter
in pJZ9-4 by Zhao et al. (1994) We obtained an even higher
level of PhhB by using the T7 overexpression system (Fig. 3-
8) The pJSlO construct used contains both phhA and phhB.
After IPTG induction, only PhhB was overexpressed and little
PhhA was made (Lane 3, Fig. 3-8) A similar phenomenon was
observed with pJS63 construct containing both phhB and phhC
where only PhhB protein was overproduced (Lane 5, Fig. 3-8).
These results indicated that phhB has stronger translational
signals and, therefore, is preferably translated over phhA and
phhC. PhhB protein expressed is fully active, and has been
purified and characterized from the latter construct by Ayling
(unpublished results).
Expression of PhhA without PhhB in E. coli Has Growth
Inhibitory Effects
Even though a low level of PhhA was produced from pJSll
in the absence of PhhB, it did not complement the E. coli
tyrosine auxotrophy in JP2255. A possible explanation was that


72
1 2 3 4 5 6 7
PhhA
PhhB
FIG. 3-4. Induction of the phh operon by phenylalanine
hydroxylase in P. aeruginosa. Proteins in the whole cell
lysates were separated by SDS-PAGE. Lane 1, E. coli JP2255
harboring pJSll and pJZ9-4; lanes 2-7, samples taken after
elapsed times of 0, 10, 30, 60, 90, 120 min, respectively,
following addition of 100 pig/ml phenylalanine at zero time.


51
(data not shown). However, induction of the phh operon under
conditions where better carbon sources (such as glucose)
coexist, suggests that the phh operon might be dedicated to
provision of some specialized compound from L-phenylalanine.
Inactivation of phhR resulted not only in the inability
to use L-phenylalanine as a carbon source, but also in an
inability to use L-tyrosine as a carbon source. Since TyrR
regulates aromatic amino acid permeases in E. coli, we
considered the possibility that the phhR mutant might fail to
grow on L-tyrosine because of a permease deficiency. Since MFT
is likely to be transported by the same system as L-tyrosine,
a permease-deficient phenotype should be resistance to growth
inhibition by MFT. However, the phhR" mutant has a MFT-
sensitive phenotype on fructase-based medium (data not shown).
Therefore, PhhR might regulate steps of tyrosine catabolism.
Regulation of Multiple Transcriptional Units by PhhR?
TyrR represses or activates eight transcriptional units
in E. coli (Pittard, 1996) Similarly organized
transcriptional units are absent or unknown in P. aeruginosa.
However, the counterpart of the aroF-tyrA operon in P.
aeruginosa would be genes encoding tyrosine-sensitive DAHP
synthase and cyclohexadienyl dehydrogenase. Physiological
manipulations in our laboratory have never revealed repression
control of these apparently constitutive enzymes. Consistent
with this, PhhR exhibits no regulatory control of either of


3
purified and characterized. The C. violaceum phenylalanine
hydroxylase gene was the first one to be cloned and sequenced
from a bacterium (Onishi et al., 1991). High identity of the
deduced amino acid sequence with those deduced for the
mammalian hydroxylase gene family was found and showed that
the microbial hydroxylase and the mammalian hydroxylases are
homologous. Although C. violaceum phenylalanine hydroxylase is
a pterin-dependent enzyme, it differs from the mammalian
enzymes in its smaller subunit size (lacking the N-terminal
domain responsible for the complex regulation in the mammalian
enzymes), its existence as a monomer (rather than a
homotetramer), and binding of copper (instead of iron) at its
active site. However, the surprising claim has been advanced
that C. violaceum phenylalanine hydroxylase does not require
any redox active metal for its activity (Carr & Benkovic,
1993; Carr et al., 1995).
P. aeruginosa belongs to a different superfamily of gram
negative prokaryotes than do the aforementioned organisms. It
was found to possess homologues of mammalian phenylalanine
hydroxylase, 4a-carbinolamine dehydratase/DCoH, and aromatic
aminotransferase as part of a three-component gene cluster
(Zhao et al., 1994). These three genes are phhA, phhB, and
phhC, respectively. The P. aeruginosa phenylalanine
hydroxylase contains iron and is pterin-dependent. Unlike the
multimeric mammalian hydroxylase, the native P. aeruginosa
hydroxylase is a monomer.


TABLE OF CONTENTS
ACKNOWLEDGMENTS iv
ABSTRACT vii
CHAPTERS
1 LITERATURE REVIEW 1
Phenylalanine Hydroxylase in Nature 1
The Pterin-Recycling Enzymes 4
Regulation of Phenylalanine Hydroxylase 8
2 PhhR, A DIVERGENETLY TRANSCRIBED ACTIVATOR OF
PHENYLALANINE HYDROXYLASE GENE CLUSTER
OF Pseudomonas aeruginosa 11
Introduction 11
Materials and Methods 13
Results 23
Discussion 42
3 BIFUNCTIONAL PhhB REGULATES THE EXPRESSION OF
PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa 53
Introduction 53
Materials and Methods 55
Results 64
Discussion 84
REFERENCES 8 9
BIOGRAPHICAL SKETCH 97
vi


6
containing two saddle-shaped grooves that comprise likely
macromolecular binding sites (Endrizzi et al. 1995) .
Structural similarities between the DCoH and nucleic acid
binding proteins imply that the saddle motif has evolved to
bind diverse ligands or that DCoH may bind nucleic acid
according to Endrizzi et al. (1995) .
DCoH homologues have been identified in Xenopus (XDCoH)
(Pogge-yon-Strandmann & Ryffel, 1995) and P. aeruginosa (PhhB)
(Zhao et al., 1994) XDCoH was found to be a maternal factor.
The amount of XDCoH increases dramatically following
neurulation, when the formation of liver, pronephros, and
other organs takes place. The tissue distribution of XDCoH
during embryogenesis suggests that XDCoH is involved in
determination and differentiation of various unrelated cell
types. The interaction with XDCoH was found to be essential
for the function of several tissue-specific transcription
factors (Pogge-yon-Strandman & Ryffel, 1995) In P. aeruginosa
expression of phhA (encoding phenylalanine hydroxylase) was
reported to require phhB (encoding 4a-carbinolamine
dehydratase) suggesting that PhhB may have a positive
regulatory role. If so, this would be an intriguing parallel
with the dual catalytic and regulatory roles of the
corresponding mammalian homolog (Zhao et al., 1994).


FIG. 2-11. A comparison of the amino acid sequences in the
central domain of the PhhR protein and 13 other homologs. The
sequences were aligned by using the PILEUP program of GCG. The
numbering of amino acid residues is given on the left. Percent
identity of PhhR with its homologs is given at the lower
right. Amino acid residues conserved in all 14 sequences are
in double-lined boxes. Amino acid residues conserved in 13 of
14 sequences are shaded. Conserved residues which are confined
to either the top cluster (subfamily a) or the bottom cluster
(subfamily ¡3) are in single-lined boxes. Two ATP-binding
motifs are indicated above the consensus sequences in boldface
type. See the legend for Fig. 8 for abbreviations.


66
Time (min)
FIG. 3-2. Stimulation of phenylalanine hydroxylase activity by
the addition (at the arrow) of a crude extract of E. coli
JP2255 containing PhhB or GST-DCoH fusion protein.
Approximately 1 min after the reaction was started, either
buffer, or 15 pig of the crude extract containing PhhB, or GST-
DCoH was added. The reaction was monitored at 340 nm for the
oxidation of NADH by dihydropteridine reductase as quinonoid
dihydropterin was recycled to a tetrabiohydropterin (see Fig.
3-1) .


90
Citron, B. A., Davis, M. D., Milstien, S., Gutierrez, J. ,
Mendel, D. B., Crabtree, G. R., and Kaufman, S. 1992.
Identity of 4a-carbinolamine dehydratase, a component of
the phenylalanine hydroxylation system, and DCoH, a
transregulator of homeodomain proteins. Proc. Natl. Acad.
Sci. USA. 89:11891-11894.
Cui, J., and Somerville, R.L. 1993. The TyrR protein of
Escherichia coli: analysis by limited proteolysis of
domain structural and ligand-mediated conformational
changes. J. Biol. Chem. 268:5040-5047.
Curtius, H. C., Adler, C., Rebrin, I., Heizmann, C., and
Ghisla, S. 1990. 7-Substituted pterins: formation during
phenylalanine hydroxylation in the absence of
dehydratase. Biochem. Biophys. Res. Commun. 172:1060-
1066 .
Davis, M. D. and Kaufman, S. 1989. Evidence for the formation
of the 4a-carbinolamine during the tyrosine-dependent
oxidation of tetrahydrobiopterin by rat liver
phenylalanine hydroxylase. J. Biol. Chem. 264:8585-8596.
Davis, M. D. and Kaufman, S. 1991. Studies on the partially
uncoupled oxidation of tetrahydropterins by phenylalanine
hydroxylase. Neurochem. Res. 16:813-819.
Davis, M. D., P. Ribeiro, J. Tipper, and S. Kaufman. 1992.
"7-tetrahydrobiopterin," a naturally occourring analogue
of tetrahydrobiopterin, is a cofactor for and a potential
inhibitor of the aromatic amino acid hydroxylases. Proc.
Natl. Acad. Sci. USA. 89:10109-10113.
Davis, M. D., Parniak, M. A., Kaufman, S., and Kemper, E.
1996. Structure-function relationships of phenylalanine
hydroxylase revealed by rediation target analysis. Arch.
Biochem. Biophys. 325:235-241.
Decicco, B. T. and Umbreit, W. W. 1964. Utilization of
aromatic amino acids by Hydrogenomonas facilis. J.
Bacteriol. 83:1590-1594.
DeFeyter, R., Kado, C. I., and Gabriel, D. W. 1990. Small,
stable shuttle vectors for use in Xanthomonas. Gene
88:65-72.
Devereux, J. P. Haeberli, and 0. Smithies. 1984. A
comprehensive set of sequence analysis programs for the
VAX. Nucleic Acids Res. 12:387-395.


32
(B)
12 3
97.4
66.2
45.0
31.0
21.5
14.5
FIG. 2-6. Overproduction of PhhR protein. (A) Map of PhhR
overexpression plasmid pJS88 (left), and low-copy number,
constitutive PhhR expression plasmid pJS91 (right). (B) SDS-
PAGE analysis of whole-cell lysate of E. coli BL21(DE3)
harboring pJS88. The gel was stained with Coomassie blue. Lane
1, molecular-weight markers; lane 2, before IPTG induction;
lane 3, induced by 1 mM IPTG for 3 h.


4
The Pterin-Recycling Enzymes
Phenylalanine hydroxylase catalyzes the conversion of L-
phenylalanine to L-tyrosine, using tetrahydrobiopterin as a
reducing agent and relying upon molecular oxygen as an
oxidizing agent (Kaufman, 1987). During this hydroxylation
reaction, the tetrahydrobiopterin cofactor is
stoichiometrically oxidized to a carbinolamine, 4a-
hydroxytetrahydrobiopterin. Two essential enzymes, 4a-
carbinolamine dehydratase and dihydropteridine reductase, are
involved in regenerating the pterin cofactor in two steps.
4a-Hydroxytetrabiopterin is first converted by 4a-
carbinolamine dehydratase to quinonoid dihydrobiopterin, and
the latter compound is then reduced back to
tetrahydrobiopterin by NADH-dependent dihydropteridine
reductase (Kaufman, 1987).
4a-Carbinolamine Dehydratase/DCoH
4a-Carbinolamine dehydratase was first purified from rat
liver as a fraction called "phenylalanine hydroxylase
stimulator", which could stimulate the hydroxylation reaction
at pH 8.2 to 8.4 (Huang et al., 1973). It was later found to
be an enzyme that catalyzes the conversion of 4a-
hydroxytetrabiopterin to the quinonoid dihydropterin (Lazarus
et al., 1983). 4a-Hydroxytetrabiopterin is also known to be
unstable, breaking down nonenzymatically to the corresponding
quinonoid dihydropterin (Kaufman, 1975). However, in the


75
P. aeruginosa
chromosome
Suicide plasmid
phhB'
mutant
(B)
1 2


BIOGRAPHICAL SKETCH
Jian Song was born on November 2, 1963, in Hebei
Province, China. He completed his elementary and high school
education in Xinhe County, Hebei Province. In 1980, he
attended the Agricultural University of Hebei, where he
majored in plant protection. He received his B.S. degree in
1984, then worked at the Institute of Plant Protection, Hebei
Academy of Agricultural and Forestry Sciences until 1987. He
was awarded a scholarship by Hebei Academy of Agricultural and
Forestry Sciences to come to the Department of Entomology and
Plant Pathology at the University of Tennessee, Knoxville for
graduate study in 1988. He studied the interactions among the
plant, aphid, and parasitoid under the supervision of Dr.
Charles D. Pless. He received his M.S. degree in entomology
in May, 1990. He then went to the Department of Entomology
and Nematology at the University of Florida to continue
graduate study toward a Ph.D. in entomology. He worked on
insect toxicology under the supervision of Dr. Simon Yu until
October, 1991. He then joined Dr. Roy A. Jensen's group at
the Department of Microbiology and Cell Science in January,
1992 and studied regulation of phenylalanine hydroxylase
system in Pseudomonas aeruginosa, and obtained his Ph.D.
degree in May, 1997. He has accepted a postdoctoral position
97


98
in Dr. Vojo Deretic's Laboratory in the Department of
Microbiology and Immunology at the University of Michigan, and
will join the laboratory in January, 1997 to work on
Mycobacterium tuberculosis.


92
Haldenwang, W. G. 1995. The sigma factors of Bacillus
subtilis. Microbiol. Rev. 59:1-30.
Heatwole, V.M., and Somerville, R.L. 1991. The tryptophan-
specific permease gene, mtr, is differentially regulated
by the tryptophan and tyrosine repressors in Escherichia
coli K-12. J. Bacteriol. 173:3601-3604.
Holloway, B.W. (1955) Genetic recombination in Pseudomonas
aeruginosa. J. Gen. Microbiol. 13:572-581.
Huang, C. Y., Max, E. E., and Kaufman, S. 1973. Purification
and characterization of phenylalanine hydroxylase-
stimulating protein from rat liver. J. Biol. Chem.
248 :4235-4241.
Ivey-Hoyle, M., Steege, D.A. 1992. Mutational analysis of an
inherently defective translation initiation site. J. Mol.
Biol. 224:1039-1054.
Jakoby, G. A. 1964. The induction and repression of amino
acid oxidation in Pseudomonas fluorescens. Biochem. J.
92:1-8.
Joerger, R.D., Jacobson, M.R., and Bishop, P.E. 1989. Two
nifA-like genes required for expression of alternative
nitrogenases by Azotobacter vinelandii. J. Bacteriol.
171:3258-3267.
Kamoun, S., Kamdar, H. V. Tola, E., and Kado, C. I. 1992.
Incompatible interactions between crucifers and
Xanthomonas campestris invole a vascular hypersensitive
response: role of the hrpX locus. Mol. Plant-Microbe
Interact. 5:22-23.
Kappock, T. J., Hartins, P. C., Friedenberg, S., and
Caradonna, J. P. 1995. Spectroscopic properties of
unphosphorylated rat hepatic phenylalanine hydroxylase
expressed in Escherichia coli. Comparison of resting and
activated states. J. Biol. Chem. 270:30532-30544.
Kaufman, S. 1975. Studies on the mechanism of phenylalanine
hydroxylase:detection of an intermediate, p.495-523. In
W. Pfleidere (ed.), Chemistry and biology of pteridines.
Walter de Gruyter, Berlin.
Kaufman, S. 1986. Regulation of the activity of hepatic
phenylalanine hydroxylase. Adv. Enz. Regulation. 25:37-
64 .


TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
TyrR-Eco
PhhR-Pae
MRLEVFCEDRLGLTRELLDLLVLRGIDLRGIEI..DPIGRIYLNFAELEFESFSSLMAEIRRIAGVTDVR
ITl I i . i l I . I Ill II.. I I III ... .11.. ..II I .
I I I I I I I I I I I M l l ill ll ll l
MR IKVHCQNRVGILRDILNLLVDYGINVNRGEVGGDQGNAIYLLCPNMINLQLQSLRPKLEAVPGVFGVK
TVPWMPSEREHLALSALLEALPEPVLSVDMKSKVDMANPASCOLFGOKLDRLRNHTAAOLINGFNFLRWL
l .Mill II I ill ll I I I II I I:: ill I l l . l .
I I I I I I I I l I II I I I l l ll I I I I I I I I I
RVGLMPSERRHLELNALLAALDFPVLSVDMGGQIVAANRAAAQLLGVRVDEVPGIPLSRYVEDLDLPELV
ESEPQDSHNEHWINGQNFLMEITPVYLQDENDQ. HVLTGAWMLRSTIRMGRQLQNVAAQDVSAFSQIV
.1 I ll .1 I ll I I I ill. I i i .1 i . .1 I si:
i M ll I I I I i I I l l I I Ml M I M I W
RANKARINGLRVKVKGDVFLADIAP. .LQS EHDES EALAGAVLTLHRADRVGERIYHVRKQELRGFDSIF
ATP-binding Site A
AVSPKMKHWEOAOKLAMLSAPLLITGDTGTGKDLFAYACHOASPRAGKPYLALNCASIPEDAVESELFG
¡ 11 :!:-! l-liMMlMMiMMl Ml 11 i M - M M 111 M M M -MS 111
QSSRVMAAWREARRMAPLDAPLLIEGETGTGKELLARACHLASPRGQSPFMALNCAGLPESMAETELFG
Unique Gap ATP-binding Site B
HA PEGKKGFFEOANGGSVLLDEIGEMSPRMOAKLLRFLNDGTFRRVGEDHEVHVDVRVICAT
MSI MM -1 M M M M 1 H 1 ^TTTTTTTMTI M 11 M I -1 M M 1111 M I
YGPGAFEGARPEGKLGLLELTAGGTLFLDGVGEMSPRLQAKLLRFLQDGCFRRVGSDEEVYLDVRVICAT
QKNLVELVQKGMFREDLYYRLNVLTLNLPPLRDCPQDIMPLTELFVARFADEQGVPRPKLAADLNTVLTR
1
I
t
I
I M 11--1! 1M11M1IMMMM11M1 = = = I Ml M-- : M- SIMM Mi
QVDLSELCAKGEFRODLYHRLNVLSLHIPPLRECLDGLAPLAEHFLDOASROIGCGLPKLSAOALERLER
Leucine-Tipper Motif
YAWPGNVRQLKNAIYRALTQLDGYELRPQDILLPDYDAATVAVGEDAMEGSLDEITSRFERSVLTQLYRN
i l ll i I ll i l . I .1 .... I ll ll I . I mi mi i i .
I I 11 II M I I I -I I I I I I I I I I II I I
YHWPGNVRQLENVLFQAVSLCEGGTVKAEHIRLPDYGAPQ.PLGDFSLEGDSTHRRA.LREGVLERLFRE
Helix-Tum-Helix Motif
YPSTRKLAKRLGVSHTAIANKLREYGLSOKKNEE*
rm[71 I I II II I IT II r I..I
II I I I II I I I I i I I I II I I I
HPSTRQLGKRLGVSHTTAANKLRQHGVGQSEG*
68
70
138
140
207
208
277
278
339
348
409
318
486
486
514
519
N-Terminal
Domain
Central
Domain
C-Terminal
Domain


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
MULTI-LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE
IN Pseudomonas aeruginosa
By
Jian Song
May, 1997
Chairperson: Roy A. Jensen
Major Department: Microbiology and Cell Science
Pseudomonas aeruginosa was recently found to possess a
cluster of genes (phh operon) encoding phenylalanine
hydroxylase (PhhA), 4a-carbinolamine dehydratase (PhhB), and
aromatic aminotransferase (PhhC). In the flanking upstream
region of the phh operon, a divergently transcribed gene
(phhR) that encodes an activator protein was identified.
Inactivation of phhR markedly reduced expression of the three
structural genes. PhhR belongs to the large prokaryote family
of a54 enhancer-binding proteins, and activation of the phh
operon by PhhR in P. aeruginosa required rpoN. P. aeruginosa
PhhR was able to replace E. coli TyrR as a repressor of the
aroF-tyrA operon (but not as an activator of mtr) in the
heterologous E. coli system. The phh operon was strongly
induced in fructose- or glucose-based minimal medium by L-
phenylalanine and L-tyrosine, and less by L-tryptophan.
Vll


91
Dilella, A. G., Kwok, S. C. M., Ledley, F. D., Marvit, J.( and
Woo, S. L. C. 1986. Molecular structure and polymorphic
map of the human phenylalanine hydroxylase gene.
Biochemistry. 25:743749.
Endrizzi, J. A., Cronk, J. D., Wang, W. Crabtree, G. R. ,
and Alber, T. 1995. Crystal structure of DCoH, a
bifunctional, protein-binding transcriptional
coactivator. Science. 268:556-559.
Essar, D. W. Eberly, L., Han, C.-Y., and Crawford, I. P.
1990. DNA sequences and characterization of four early
genes of the tryptophan pathway in Pseudomonas
aeruginosa. J. Bacteriol. 172:853-866.
Faust, D. M., Catherin, A. M., Barbaux, S., Belkadi, L.,
Imaizumi-Sherrer, T., and Weiss, M. C. 1996. The activity
of the highly inducible mouse phenylalanine hydroxylase
gene promoter is dependent upon a tissue-specific,
hormone-inducible enhancer. Mol. Cell. Biol. 16:3125-
3137 .
Friedman, D.I. 1988. Integration host factor: a protein for
all reasons. Cell 55:545-554.
Friedrich, B. and H. G. Schlegel. 1972. Die hydroxylierung
von phenyalanin durch Hydrogenomonas eutrophus H16 Arch.
Mikrobiol. 83:17-31.
Gambill, D. B., and Summers, A. O. 1985. Versatile mercury-
resistant cloning and expression vectors. Gene 39:293-297
Gu, W. and Jensen, R.A. 1996. Evolutionary recruitment of
biochemically specialized subdivisions of Family I within
the protein Superfamily of aminotransferases. J.
Bacteriol. 178:2161-2171.
Gudinchet, F., Maeder, P., Meuli, R. A., Deonna, T., and
Mathieu, J. M. 1992. Cranial CT and MRI in malignant
phenylketonuria. Pediatr. Radiol. 22:223-224.
Guldberg, P., H. L. Levy, W. B. Hanley, R. Kock, R. Mataln,
B. M. Rouse, F. Trefz, F. de la Cruz, K. F. Henriksen,
and F. Guttler. 1996. Phenylalanine hydroxylase gene
mutations in the United States: report from the Maternal
PKU Collaborative Study. Am. J. Hum. Genet. 59:84-94.
Guroff, G. and T. Ito. 1963. Induced soluble
phenylalanine hydroxylase from Pseudomonas sp. grown on
phenylalanine or tyrosine. Biochim. Biophys. Acta
77:159-161.


77
The physiological effect of the phhB knockout mutant in
P. aeruginosa was examined. Inactivation of phhB abolished the
ability to grow on either phenylalanine or tyrosine as the
sole carbon source. However, interpretation of this result is
complicated by results obtained with a phhC knockout mutant
which was also not able to grow on either phenylalanine or
tyrosine as the sole carbon source. Since insertion of the
suicide plasmid into the chromosome of P. aeruginosa is
expected to create polar effects on the downstream genes in
the operon (as indeed seen in phhA knockout where amount of
PhhB expressed was dramatically decreased), it seems probable
that the physiological effect observed in phhB knockout mutant
is due to the polar effect on the expression of phhC.
Overexpression of PhhA and PhhB Proteins
Although PhhA was expressed to detectable levels in both
E. coli (Fig. 3-3) and P. aeruginosa (Fig. 3-6) in the absence
of phhB, initial attempts to express PhhA at high levels in
the absence of PhhB in E. coli were unsuccessful. I then
employed a T7 overexpression system (see Methods) to express
PhhA in E. coli BL21(DE3) under induction conditions not
requiring growth (Fig. 3-7A). When phhA was expressed from a
native ribosomal binding site in pJS72, high PhhA levels were
produced after IPTG induction for 3 h (Lane 3, Fig. 3-7B).
When phhA was expressed from $10 translational signals
(pJS95), higher levels of PhhA was made after IPTG induction



PAGE 1

MULT I -LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE IN Pseudomonas aeruginosa By JIAN SONG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1997

PAGE 2

Copyright 19 by Jian Song

PAGE 3

Dedicated to my father and my mother, whose love, care, and encouragement make it possible for me to complete this dissertation

PAGE 4

ACKNOWLEDGMENTS I wish to express my deep and sincere gratitude to Dr. Roy A. Jensen, chairman of my supervisory committee, whose invaluable guidance, constant encouragement, endless ideas, critical input, and financial support made the fulfillment of this study possible. I would also like to thank Dr. Dean W. Gabriel, Dr. Lonnie 0. Ingram, Dr. James F. Preston, and Dr. Keelnatham T. Shanmugam for their help, encouragement, advice, and critical review of the dissertation. My special thanks are also extended to Dr. Carol Bonner, Dr. Tianhui Xia, and Wei Gu for their great help in all aspects of my study, particularly helping me get started during my first year in the lab. I am also very thankful to Dr. Randy Fischer, Dr. Prem Subramaniam, and Gary Xie for their help during this study. I am indebted to my family, especially to my parents, to whom this dissertation is dedicated. Without their love, support, and encouragement, this study could not have been accomplished. I am also indebted to my brother and sister-inlaw for helping me in taking care of my parents. Finally, but not least, I wish to express my sincere appreciation to my wife, Tao Sun, for her love, support, iv

PAGE 5

patience, and encouragement during these years of study, and to my son, Peter, and my daughter, Kerry, for filling the family with great joy and happiness. v

PAGE 6

TABLE OF CONTENTS ACKNOWLEDGMENTS ABSTRACT CHAPTERS 1 LITERATURE REVIEW 1 Phenylalanine Hydroxylase in Nature 1 The Pterin-Recycling Enzymes 4 Regulation of Phenylalanine Hydroxylase 8 2 PhhR, A DIVERGENETLY TRANSCRIBED ACTIVATOR OF PHENYLALANINE HYDROXYLASE GENE CLUSTER OF Pseudowonas aeruginosa Introduction Materials and Methods Results Discussion 3 B I FUNCT I ONAL PhhB REGULATES THE EXPRESSION OF PHENYLALANINE HYDROXYLASE IN Pseudomonas aeruginosa 53 Introduction 53 Materials and Methods 55 Results 64 Discussion 84 REFERENCES 8 9 BIOGRAPHICAL SKETCH 97 11 11 13 23 42 vi

PAGE 7

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MULT I -LEVEL REGULATION OF PHENYLALANINE HYDROXYLASE IN Pseudomonas aeruginosa By Jian Song May, 1997 Chairperson: Roy A. Jensen Major Department: Microbiology and Cell Science Pseudomonas aeruginosa was recently found to possess a cluster of genes (phh operon) encoding phenylalanine hydroxylase (PhhA) , 4a-carbinolamine dehydratase (PhhB) , and aromatic aminotransferase (PhhC) . In the flanking upstream region of the phh operon, a divergently transcribed gene (phhR) that encodes an activator protein was identified. Inactivation of phhR markedly reduced expression of the three structural genes. PhhR belongs to the large prokaryote family of a 54 enhancer-binding proteins, and activation of the phh operon by PhhR in P. aeruginosa required rpoN. P. aeruginosa PhhR was able to replace E . coli TyrR as a repressor of the aroF-tyrA operon (but not as an activator of mtr) in the heterologous E. coli system. The phh operon was strongly induced in fructoseor glucose-based minimal medium by Lphenylalanine and Ltyrosine, and less by Ltryptophan . vii

PAGE 8

Inactivation of phhR in P. aeruginosa abolished ability to utilize either L-phenylalanine and L-tyrosine as a sole source of carbon for growth. PhhB is a bifunctional protein. It was shown to have 4acarbinolamine dehydratase activity as well as regulatory activity. The expression of phhA was activated by the presence of phhB in both E . coli and P. aeruginosa . Transcriptional and translational fusion analysis showed that the regulatory effect of PhhB on the expression of phhA is at the posttranscript ional level. An insertionally inactivated phhB mutant failed to grow on L-phenylalanine or L-tyrosine as a sole carbon source. Expression of PhhA in the absence of PhhB causes strong growth inhibition in E. coli. The inhibitory effect is probably caused by 7-tetrahydrobiopterin, which is known to be formed in the absence of PhhB. Since 7-tetrahydrobiopterin is a potent inhibitor of phenylalanine hydroxylase, this could account for the inability of phhA in the absence of phhB to complement E. coli tyrosine auxotrophy. The general inhibition of growth may be due to inhibition of some unidentified essential pterin-dependent enzymes. viii

PAGE 9

CHAPTER 1 LITERATURE REVIEW Phenylalanine Hydroxylase in Nature Phenylalanine hydroxylase (phenylalanine hydroxylase 4monooxygenase ; EC 1.14.16.1) catalyzes the irreversible conversion of L-phenylalanine to L-tyrosine (Kaufman, 1987) . In mammals this enzyme catalyzes the initial, obligatory, and rate-limiting step in the complete catabolism of serum phenylalanine to C0 2 and H 2 0 (Kaufman, 1986) . A deficiency of this enzyme causes accumulation of serum phenylalanine, leading to hyperphenylalanemia . Because metabolism of phenylalanine is restricted to alterations in the alanyl side chain of phenylalanine, in the absence of phenylalanine hydroxylase, the formation and excretion in the urine of compounds such as phenylpyruvate and phenyllactate occurs. This condition is called phenylketonuria, a genetic disorder associated with severe mental retardation in untreated children (Dilella et al . , 1986). Many mutations at the phenylalanine hydroxylase locus have been identified (Guldberg et al. , 1996) . Phenylalanine hydroxylase has been intensively studied in mammals for many years. It is a member of a family of enzymes that also includes tryptophan hydroxylases (EC 1.14.16.4) and 1

PAGE 10

2 tyrosine hydroxylases (EC 1.14.16.2) . All three enzymes utilize a tetrahydrobiopterin cofactor and molecular oxygen to hydroxylate their respective aromatic amino acid substrates (Kaufman and Fisher, 1974) . Phenylalanine hydroxylase has been purified from rat liver where it is an oligomeric protein (predominantly homotetramers ) composed of 52-kDa subunits (Davis et al . , 1996) . It has non-heme iron as the active-site metal. The rat liver hydroxylase was also expressed in E. coli and purified to homogeneity (Kappock et al . , 1995) . The homotetrameric recombinant rat hepatic phenylalanine hydroxylase is highly active and is identical to the native enzyme in many properties. Although mammalian phenylalanine hydroxylase has been intensively studied, few studies on bacterial phenylalanine hydroxylase have been done . Phenylalanine hydroxylase has generally been considered to be of rare occurrence in prokaryotes, where scattered reports of its existence have been limited to one phylogenetic division of gram-negative bacteria. They include Pseudomonas acidovorans (previously known as Pseudomonas sp . ATCC 11299a) (Guroff & Ito, 1963) . P. facilis (Decicco & Umbreit, 1964) , Alcaligenes eutrophus (Friedrich & Schlegel, 1972), and Chromobacterium violaceum (Letendre et al . , 1974). Of the three pterin-dependent and metal -containing hydroxylases, only phenylalanine hydroxylase from Pseudomonas acidovorans (Letendre et al, 1975) and C. violaceum (Nakata et al . , 1979; Pember et al . , 1986) has been

PAGE 11

3 purified and characterized. The C. violaceum phenylalanine hydroxylase gene was the first one to be cloned and sequenced from a bacterium (Onishi et al . , 1991). High identity of the deduced amino acid sequence with those deduced for the mammalian hydroxylase gene family was found and showed that the microbial hydroxylase and the mammalian hydroxylases are homologous. Although C. violaceum phenylalanine hydroxylase is a pterin-dependent enzyme, it differs from the mammalian enzymes in its smaller subunit size (lacking the N-terminal domain responsible for the complex regulation in the mammalian enzymes) , its existence as a monomer (rather than a homotetramer ) , and binding of copper (instead of iron) at its active site. However, the surprising claim has been advanced that C. violaceum phenylalanine hydroxylase does not require any redox active metal for its activity (Carr & Benkovic, 1993; Carr et al . , 1995). P. aeruginosa belongs to a different superfamily of gramnegative prokaryotes than do the aforementioned organisms. It was found to possess homologues of mammalian phenylalanine hydroxylase, 4a-carbinolamine dehydratase/DCoH, and aromatic aminotransferase as part of a three -component gene cluster (Zhao et al . , 1994). These three genes are phhA, phhB, and phhC, respectively. The P. aeruginosa phenylalanine hydroxylase contains iron and is pterin-dependent. Unlike the multimeric mammalian hydroxylase, the native P. aeruginosa hydroxylase is a monomer.

PAGE 12

4 The Pterin-Recycling Enzymes Phenylalanine hydroxylase catalyzes the conversion of Lphenylalanine to Ltyrosine, using tetrahydrobiopterin as a reducing agent and relying upon molecular oxygen as an oxidizing agent (Kaufman, 1987) . During this hydroxylat ion reaction, the tetrahydrobiopterin cofactor is stoichiometrically oxidized to a carbinolamine , 4ahydroxytetrahydrobiopterin . Two essential enzymes, 4acarbinolamine dehydratase and dihydropteridine reductase, are involved in regenerating the pterin cofactor in two steps. 4a-Hydroxytetrabiopterin is first converted by 4acarbinolamine dehydratase to quinonoid dihydrobiopterin , and the latter compound is then reduced back to tetrahydrobiopterin by NADHdependent dihydropteridine reductase (Kaufman, 1987) . 4a -Carbinolamine Dehydratase /DCoH 4a-Carbinolamine dehydratase was first purified from rat liver as a fraction called "phenylalanine hydroxylase stimulator", which could stimulate the hydroxylat ion reaction at pH 8.2 to 8.4 (Huang et al . , 1973). It was later found to be an enzyme that catalyzes the conversion of 4ahydroxytetrabiopterin to the quinonoid dihydropterin (Lazarus et al . , 1983). 4a-Hydroxytetrabiopterin is also known to be unstable, breaking down nonenzymat ically to the corresponding quinonoid dihydropterin (Kaufman, 1975) . However, in the

PAGE 13

5 absence of 4a-carbinolamine dehydratase, the dehydration of the 4a-carbinolamine becomes rate-limiting for the hydroxylat ion of phenylalanine. The consequent accumulation of 4a-carbinolamine results in a small percentage of rearrangement to the 7-tetrahydrobiopterin isomer (Curtius et al . , 1990) . The latter 7-isomer was shown to be a potent inhibitor of the phenylalanine hydroxylase (Davis et al . , 1992) . Under conditions where 4a-carbinolamine and the 7isomer are generated, the addition of 4a-carbinolamine dehydratase markedly inhibits the rate of formation of the 7isomer by diverting a greater fraction of the 4a-carbinolamine to the quinonoid dihydropterin (Davis et al . , 1991) . Thus, the dehydratase not only directly catalyzes the dehydration of the carbinolamine , but also indirectly prevents isomerizat ion to the inhibitory 7-isomer (Kaufman et al . , 1993). 4a-Carbinolamine dehydratase from rat liver has been cloned and sequenced (Citron et al . , 1992) . It then became apparent that this dehydratase is identical to DCoH, a protein that facilitates the dimerization of hepatic nuclear factor 1 alpha (HNF-lcif) , a homeodomain transcription factor. DCoH was found to display a restricted tissue distribution and did not bind directly to DNA. The formation of a stable tetrameric DCoH-HNF-la complex does not change the DNA-binding characteristics of HNF-la, but does enhance the transcriptional activity of HNF-la (Mendel et al . , 1991) . Xray crystallography has revealed DCoH to form a tetramer

PAGE 14

6 containing two saddle-shaped grooves that comprise likely macromolecular binding sites (Endrizzi et al . , 1995) . Structural similarities between the DCoH and nucleic acidbinding proteins imply that the saddle motif has evolved to bind diverse ligands or that DCoH may bind nucleic acid according to Endrizzi et al . (1995) . DCoH homologues have been identified in Xenopus (XDCoH) (Pogge-yon-Strandmann & Ryffel, 1995) and P. aeruginosa (PhhB) (Zhao et al . , 1994) . XDCoH was found to be a maternal factor. The amount of XDCoH increases dramatically following neurulation, when the formation of liver, pronephros, and other organs takes place. The tissue distribution of XDCoH during embryogenesis suggests that XDCoH is involved in determination and differentiation of various unrelated cell types. The interaction with XDCoH was found to be essential for the function of several tissue-specific transcription factors (Pogge-yon-Strandman & Ryffel, 1995) . In P. aeruginosa expression of phhA (encoding phenylalanine hydroxylase) was reported to require phhB (encoding 4a-carbinolamine dehydratase) , suggesting that PhhB may have a positive regulatory role. If so, this would be an intriguing parallel with the dual catalytic and regulatory roles of the corresponding mammalian homolog (Zhao et al . , 1994) .

PAGE 15

7 Dihydropteridine Reductase Dihydropteridine reductase (DHPR; EC 1.6.99.7) is one of the two essential enzymes involved in recycling the pterin cof actor for aromatic amino acid hydroxylases. It catalyzes the reduction of quinonoid dihydropter in to tetrahydrobiopterin, using NADH as a cof actor. DHPR is an ubiquitous enzyme in animals, being found in all tissues that contain the aromatic amino acid hydroxylases (Armarego et al . , 1984) . Close correlation between levels of 4a-carbinolamine dehydratase and dihydropterine reductase in liver during human fetal development strongly suggests a physiologically significant role for both enzymes in tetrahydrobiopterin regeneration. Genetic defects in DHPR cause malignant phenylketonuria. A concomitant deficiency of neurotransmitters such as 3 , 4 -dihydroxyphenylalanine (DOPA) and 5-hydroxytryptophan reflects the essential coupling of DHPR to tyrosine hydroxylase and tryptophan hydroxylase as well (Gudinchet et al . , 1992). DHPR is also found in bacteria. DHPR has been purified from Pseudomonas acidovorans (Williams et al . , 1976) and E . coli (Vasudevan et al . , 1988). In P. acidovorans, both DHPR and phenylalanine hydroxylase activities were found to be higher in cells adapted to a medium containing L-phenylalanine or L-tyrosine as the sole carbon source than in those grown in L-asparagine (Williams et al . , 1976). Interestingly, DHPR has also been found in E. coli even though no aromatic amino acid

PAGE 16

8 hydroxylases or 4a-carbinolamine dehydratase have ever been detected (Vasudevan et al . , 1988). Unlike other dihydropteridine reductases that have been studied, the E . coli DHPR possesses an FAD prosthetic group, and has dihydrof olate reductase and pterinindependent oxidoreductase activities (Vasudevan et al . , 1992) . Regulation of Phenylalanine Hydroxylase Phenylalanine hydroxylase in mammals is tightly regulated at different levels. At the protein level, it is allosterically regulated by phenylalanine (Kaufman, 1987) . The activity of phenylalanine hydroxylase increases at least 20-fold after incubation with phenylalanine (Tourian, 1971) . It is also activated through phosphorylation by a cAMPdependent kinase both in vivo and in vitro (Abita et al , 1976) . At the DNA level, expression of the phenylalanine hydroxylase gene in liver and kidney tissues of mice is enhanced at birth and is induced by glucocorticoids and cAMP in liver (Faust et al . , 1996). Regulatory elements including a tissue-specific and hormone -inducible enhancer in the upstream region have been characterized. The enhancer region contains separate protein-binding sites for the glucocorticoid receptor and the hepatocyte-enriched transcription factor, hepatocyte nuclear factor 1 (HNF1) (Faust et al . , 1996) . HNF1 is a transcriptional activator of many hepatic genes including albumin, a -anti trypsin, and aor 0fibrinogen . Mice lacking

PAGE 17

9 HNF1 die with a marked liver enlargement. The gene coding for phenylalanine hydroxylase is totally silent, thus giving rise to phenylketonuria (Pontoglio et al . , 1996). Little information is available about the regulation of phenylalanine hydroxylase in bacteria. However, some evidence has indicated that the bacterial phenylalanine hydroxylase is also regulated. In Pseudomonas acidovorans , a higher level of phenylalanine hydroxylase was found after growth in the presence of phenylalanine (Willams et al , 1976) . Induction of both phenylalanine hydroxylase and tryptophan hydroxylase in the presence of their substrates was also reported in C. violaceum (Letendre et al , 1974) . The most extensively characterized microbial phenylalanine system is that of P. aeruginosa. Whether this system is subject to any regulatory controls has not been studied prior to this work. The initial report of Zhao et al . (1994) provided a strong basis for anticipation that the phh operon would be subject to regulation for the following reasons, (i) The closely spaced organization of the three structural genes {phhABC) in an apparent operon implies regulation. (ii) Analysis of effects of the presence or absence of regions immediately flanking the phh operon upon expression of phenylalanine hydroxylase indicated the likely location there of one or more regulatory genes. (iii) The reported lack of phhA expression in the absence of phhB

PAGE 18

10 suggested a positive regulatory role of phhB in addition to its catalytic function. The major objectives of this study have been to elucidate the physiological conditions under which regulation occurs, to identify and characterize at the molecular-genetic level any regulatory genes which control the phh operon, and to determine the nature of the apparent positive regulatory action of phhB .

PAGE 19

CHAPTER 2 PHHR , A DIVERGENTLY TRANSCRIBED ACTIVATOR OF THE PHENYLALANINE HYDROXYLASE GENE CLUSTER OF Pseudomonas aeruginosa Introduction A recent report (Zhao et al . , 1994) revealed that Pseudomonas aeruginosa possesses a tetrahydrobiopterin (BH 4 ) dependent monooxygenase that is capable of catalyzing the phenylalanine hydroxylase reaction. It is encoded by the proximal member (phhA) of a three -gene cluster. The second gene, phhB, encodes carbinolamine dehydratase, a key enzyme within the cycle regenerating BH 4 . phhC encodes an aromatic aminotransferase and belongs to the FamilyI aminotransferases (Gu and Jensen, 1996) . The reactions, as they are known to function for the mammalian homologs in the catabolism of Lphenylalanine , are shown in Fig. 2-1. The physiological function of phenylalanine hydroxylase in P. aeruginosa has not been obvious. A primary role in Ltyrosine biosynthesis seems unlikely because of the established presence for this purpose of a cyclohexadienyl dehydrogenase that is widely distributed in gram-negative bacteria and which is highly sensitive to feedback inhibition by Ltyrosine (Xia and Jensen, 1990) . Although function as an initial step of L-phenylalanine catabolism has precedent in 11

PAGE 20

12 4-Hydroxyphenylpyruvate [PhhC]^ Tyrosine 4a-Carbinolamine [PhhA] + 0, -Tetrahydrobiopterin H 2 0 Dihydropteridine [DHPRJ Phenylalanine NAD + NADH+H + FIG. 2-1. Initial reactions of phenylalanine catabolism in mammals. The three structural genes of the phh operon encode enzymes catalyzing three of the four steps shown. The abbreviations: PhhA, phenylalanine hydroxylase; PhhB, 4acarbinolamine dehydratase; PhhC, aromatic aminotransferase; DHPR, dihydropteridine reductase. 4a-Carbinolamine is an alternative designation for 4a-hydroxytetrahydrobiopterin .

PAGE 21

13 mammalian metabolism, the literature encompassing the widely studied catabolism of aromatic compounds in pseudomonad bacteria (indeed, in prokaryotes) does not include the phenylalanine hydroxylase step. Furthermore, L-phenylalanine (substrate of PhhA) is an extremely poor source of carbon for growth of P. aeruginosa , whereas Ltyrosine (product of PhhA) is an excellent carbon source. Zhao et al . (1994) had previously noted that subclones lacking the flanking regions around the phh operon possessed 20 -fold greater activity for phenylalanine hydroxylase. This suggested the presence of a regulatory gene. Since an understanding of the regulation governing the phh operon should provide important physiological clues about function, I have analyzed the flanking regions and now report the characteristics of a regulatory gene, denoted phhR. Materials and Methods Materials The bacterial strains and plasmids used in this study are listed in Table 1. The LB and M9 formulations (Sambrook et al . , 1989) were used as growth media for E. coli and P. aeruginosa. Pseudomonas isolation agar (Difco) was used for isolating Pseudomonas "knockout" mutants. Additions of ampicillin (100 /xg/ml) , chloramphenicol (40 /xg/ml) , kanamycin (50 /xg/ml) , tetracycline (25 /xg/ml) , mercuric chloride (15 /xg/ml) , L-phenylalanine (50 /xg/ml) , and thiamine (17 /xg/ml)

PAGE 22

14 TABLE 2-1. Bacterial strains and plasmids Strain or plasmid Relevant genotype or description Source or reference E. coli BL21 (DE3) BW54 5 DH5a LE392 JP2255 S17-1 SP1312 SP1312 (XSLW2 0) SP1313 SP1313 (XSLW2 0) F" ompT hsdS B (r B ~m B ~) gal dcm; with DE3 , a X prophage carrying the T 7 RNA polymerase gene A (lacU)169 rpsL F'AlacU169 $80dlacZAM15 hsdRU recAl endAl gyrA96 thi-1 relAl supE44 F'el4(McrA~ ) hsdR514 (r k "m k + ) supE44 supF58 lacYl or a (lacIZY) 6 galK galT22 metBl trpR55 aroF363 pheA361 phe0352 tyrA382 thi-1 strR712 lacYl xyl-15 [RP4-2 (Tc :Mu) ( Km : Tn7 ) Tra ( incP) ] pro hsdR recA Tp r Sm r zah-735:TnlO MargF-lac) U169 SP1312 (jntr ' -lacZ + ) SP1312, A(tyrR) SP1313 4>(mtr ' -lacZ*) Novagen Rosentel et al GIBCO/BRL Sambrook et al Baldwin & Davidson Simon et al Heatwole & Somerville Heatwole & Somerville Heatwole & Somerville Heatwole & Somerville, P. aeruginosa PA103 Prototroph PA10 3NG PAO-1 JS101 JS102 rpoN Prototroph PAO-1 phhA, Hg r PAO-1 phhR, Hg r Totten et al Totten et al Holloway This study This study

PAGE 23

15 Table 2-1. (continued) Plasmids pUC18 Amp r lac'IPOZ' pUC19 Amp r lac'IPOZ' pACYC184 P15A replicon, Cm r Tc r pETllb T7lac promoter , lacl + Ap r pRS1274 lacZY fusion vector Z1918 Promoterless la.cZ, Ap r pJZ9 phhRABC, Ap r pJZ9-3a phhAB, Ap r pJS7 phhRABC, Ap r pJS60 phhABC, Ap r pJS61Z phhRA' -lacZ transcriptional fusion, Ap r pJS62Z phhA' -lacZ transcriptional fusion, Ap r pJS88 pETllb carrying phhR translat ional fusion at the ATG start site pJS91 pACYC184 carrying phhR* , Cm r pJS102 pRS1274 carrying phhR ' -lacZY transcriptional fusion pCRII Amp r Kan r lacZa pDG106 Hg r Km r P15A replicon pJSlOl Pstl-Smal fragment of pDG106 inserted into pUC18 YanischPerron et al . Yanischperron et al . Chang & Cohen Novagen Simons et al . Schweizer Zhao et al . Zhao et al . This study This study This study This study This study This study This study Invitrogen Gambill & Summers This study pUFR004 ColEl Cm r Mob + mob(P) DeFeyter et al

PAGE 24

were made as appropriate. Agar was added at 2 0 g/liter for preparation of solid medium. Restriction enzymes, T4 DNA ligase, DNA-modif ying enzymes (New England Biolab or Promaga) and Taq DNA polymerase ( Perkin-Elmer ) were used as recommended by the suppliers. Other biochemicals were purchased from Sigma Chemical Co. Inorganic chemicals (analytical grade) were from Fisher Scientific. Phenylalanine Hydroxylase Assay Cultures of E. coli JP2255 carrying the various plasmids specified were grown in 500 ml of LB broth supplemented with ampicillin (100 /xg/ml) at 37°C and harvested at late exponential phase of growth. The cell pellets were resuspended in 10 ml of 10 mM potassium phosphate buffer (pH 7.4) containing 1 mM dithiothreitol and were disrupted by sonication for 30 s at 4°C using a Lab-Line Ultratip Labsonic System (Lab-Line Instruments, Inc., Melrose Park, IL) . The resulting extracts were centrifuged at 150,000 x g for 1 hr at 4°C. The supernatant (crude extract) was desalted using Sephadex G-2 5 and used for enzyme assay. PhhA was assayed by following tyrosine formation (Nakata et al . , 1979). Recombinant DNA Techniques Molecular cloning and DNA manipulation, including plasmid purification, restriction enzyme digestion, ligation, and transformation were conducted by standard methods (Sambrook et

PAGE 25

17 al . , 1989) . DNA fragments were purified from agarose gel with a "Geneclean" kit (BiolOl) . Electroporat ion (Invitrogen) was used for simultaneous transformation of E. coli with two compatible plasmids . Construction of PhhR Expression Vectors For expression of PhhR protein in E. coli, the T7 expression system (Novagen) was employed. The phhR coding region was cloned into a translat ional fusion vector pETllb. Polymerase chain reaction (PCR) was used to amplify the phhR gene. The upper primer ( 5 ' -ATACATATGCGTATCAAAGTGCACTGC3 ' ) was made with a built-in Ndel restriction site (underlined) which allows fusion of phhR at the translational start site (ATG in bold) . The lower primer (5 ' -CCTCCACCGTTTCTTTCCCAGCCT3') was chosen at a position 48 bases downstream of the translational stop codon. PhhR protein made from this PCR fragment was designed to be a native protein, not a fusion protein. The PCR fragment was cloned into a TA cloning vector, pCRII. The phhR gene was excised from pCRII with Ndel and EcoRI . The Ndel-EcoRI fragment was first ligated with EcoRI -BamHI adaptor to create a Ndel-BanMI fragment which was then ligated with pETllb digested with Wdel and BarrMI to create the PhhR expression plasmid, pJS88 (Fig.2-3A). For construction of a PhhR constitutive expression plasmid, pACYC184 was chosen as the expression vector. The pACYC184 vector has a P15A origin of replication which is

PAGE 26

18 compatible with most commonly used plasmids using a ColEI origin of replication, and it has low-copy number (about 20 copies/cell). High level of PhhR produced from a high-copy number plasmid was found to be toxic to the host cells. The Bglll BarriH.1 fragment carrying phhR gene was excised from the expression plasmid pJS88 and cloned into the BawRI site of pACYC184, thereby interrupting the tetracycline resistance gene (Tc r ) (Fig.2-3A). Evaluation of Sensitivity/Resistance to m-Fluorotyrosine Three E . coli strains, SP1312 ( tyrR + ) , SP1313 {tyrR') carrying pJS91 (phhR*), and SP1313 (tyrR~) carrying pACYC184 {phhR') , were compared for sensitivity to mfluorotyrosine (MFT) . All three strains were first grown in M9 medium with appropriate antibiotics up to late-exponential phase of growth and then used to swab M9 agar plates containing appropriate antibiotics. A sterile Difco concentration disk (0.6 cm) was positioned at the center of each plate, and 10 [xl of 50 iiq/ml mfluorotyrosine was applied onto the disks. The plates were then incubated at 37°C for 24 hours. Construction of phhA ' -lacZ and phhR' -lacZ Transcriptional Fusions To compare levels of phhA transcription in both pJZ9 and pJZ9-3a, plasmids pJS61Z and pJS62Z were constructed, respectively. These have a promoterless lacZ gene (from

PAGE 27

19 plasmid Z1918) fused at the BairiH I site within phhA to form phhA' -lacZ transcriptional fusions. Plasmid pJS61Z has the same upstream sequence as plasmid pJZ9, and plasmid pJS62Z has the same upstream sequence as pJZ9-3a. Hence, the phhA' -lacZ fusions in pJS61Z and pJS62Z should represent the phhA transcriptional levels in pJZ9 and pJZ9-3a, respectively. To study regulation of the phhR promoter, the Hindi BamHI fragment (phhR') was cloned into the pRS12 74 lacZY transcriptional fusion vector at the BairiRI Smal site to create pJS102 (phhR' -lacZ) . <3-Galactosidase Assay i6-Galactosidase activity was assayed under conditions of proportionality as described by Miller (1972) , and specific activities are expressed in Miller units. The data are the results of at least two independent assays. Gene Inactivation P. aeruginosa is well known for its relatively high resistance to most antibiotics, which complicates attempts to use most of available antibiotic-resistance genes as selective markers for gene replacement. Mercury resistance (Hg r ) was used as a selective marker since P. aeruginosa has been shown to be sensitive to mercury (Essar et al . , 1990; Gambill and Summers, 1985) . Insertional inactivation technique described by Sophien et al . (1992) utilizes a mobilizable suicide vector

PAGE 28

20 containing a truncated gene fragment (at both 5' and 3' ends) and Hg r -cassette , and this suicide plasmid was integrated into the chromosome by a single homologous recombination event. PCR was used to generate truncated fragments. To generate a * phhR' (601bp) fragment, the upper primer 5'CCGTGTAGGCATCCTCCGCGACAT3 ' , and the lower primer 5'CTGGAAGATACTGTCGAAGCCACG3 ' were used; to generate the ' phhA' (639bp) fragment, we used the upper primer 5'ACGACAACGGTTTCATCCACTATC3 ' and the lower primer 5'GGACGAAATAGAGCGGTTGCAGGA3 ' . The PCR-generated fragments were cloned into pCRII (a TA cloning vector) and then excised with EcoRI . The EcoRI fragments were subsequently cloned into the EcoRI site of pUFR004 (a mobilizable suicide vector) to create pUFR/ 'phhA' , and Hg r Hindi I I -cassette from pJSlOl was inserted into the Hindi I I site of pUFR/'phhA' to create pUFR/ 'phhA' /Hg r . pUFR/ ' phhR' /Hg r was created in a similar fashion. These plasmids were then used to transform E . coli strain S17-1 (a mobilizing strain) . Strain S17-1 harboring either pUFR/ ' phhA' /Hg r or pUFR / * phhR' /Hg r was used as the donor in biparental mating with P. aeruginosa performed as described by Simon et al . (1983) . Donor and recipient cells were grown in LB broth to an OD 600 of about 1.0 {E . coli S17-1 at 37°C and P. aeruginosa PAO-1 at 42°C), mixed (0.5 ml volume of each) in a 1.5-ml microcentrifuge tube, and pelleted by centrif ugation . The mating mixture was carefully resuspended in 0.2 ml of LB broth and spread onto a sterile nitrocellulose filter (0.45-^tm

PAGE 29

pore size) resting on a prewarmed LB agar plate. The plates were incubated for 16-24 hours at 37°C, and then cells were removed from the filter by an inoculation loop and resuspended by vortexing into 0.5 ml of LB broth. Aliquots of 10-, 20-, 50-, and 100-/il volume of the cell suspension were spread onto Pseudomonas isolation agar plates containing 15 /xg of HgCl 2 . The plates were incubated overnight and Hg r colonies were isolated . Preparation of PhhAspecif ic Polyclonal Antiserum PhhA was partially purified by anion-exchange and gelfiltration chromatography following the methods described by Zhao et al . , (1994). The partially purified PhhA was subject to SDS-PAGE (12%) and the gel was stained with Commassie blue R-250. The PhhA band was cut from the gel and used for the production of polyclonal antiserum in rabbits (Cocalico Biologicals, Inc., Reamstown, PA). Antiserum was purified by using an Econo-Pac protein A column (Bio Rad) and further absorbed with a total cell extract from the PhhA-def icient mutant JS101. SDS-PAGE and Western Blot Analysis SDS-PAGE (12%) was performed with the Mini-PROTEAN II Cell (Bio-Rad) by the method of Laemmli (1970) . Samples of exponential -phase cells were collected by centrif ugation, and the cell pellets were suspended in gel-loading buffer and

PAGE 30

22 heated at 100°C for 10 min. Samples of 5-10 fxl were loaded onto two SDS-acrylamide gels. After separation of the proteins by electrophoresis, one gel was stained with Coomassie blue R250 and the other gel was used for blotting. When crude extracts were used, equivalent amounts of protein were loaded in each lane. Western blots were performed according to Towbin et al . (1979) . The proteins were eletrophoret ically transferred onto nitrocellulose membranes and reacted with polyclonal antibodies raised against PhhA in a rabbit . N-Terminal Amino Acid Sequencing PhhR protein produced in E. coli BL21 (DE3 ) /pJS88 following induction by 1 mM IPTG for 3 hours was first separated from the whole lysate by SDS-PAGE. The proteins were then blotted to a polyvinyl idene difluoride membrane (Bio-Rad) and were stained with Coomassie brilliant blue R-250 (Sigma) . The band corresponding to PhhR was excised from the membrane and used for sequencing by using an Applied Biosystems model 407A protein sequencer with an on-line 120A phenylthiohydantoin analyzer in the Protein Core Facility of the ICBR at the University of Florida. DNA Sequencing and Data Analysis Sequencing of phhR region was performed by the DNA Sequencing Core Laboratory of the University of Florida. The nucleotide sequence and the deduced amino acid sequence were

PAGE 31

23 analyzed by using the updated version of sequence analysis software package offered by the Genetics Computer Group (GCG) of University of Wisconsin (Devereux et al . , 1984) . Nucleotide Sequence Accession Number The nucleotide sequence reported in this work has been assigned Genbank accession number U62581. Results Evidence for a Flanking Regulatory Region The original clone (pJZ9) isolated by Zhao et al . (1994) produced markedly less phenylalanine hydroxylase activity than did subclone pJZ9-3a (Fig. 2-2A) . A possible explanation was the presence of a negatively-acting regulatory gene in either the upstream or downstream flanking region. Plasmid pJS60, in correlation with its absence of upstream DNA but presence of downstream DNA, expressed a very high level of activity. Thus, the upstream region appeared to be responsible for decreased expression of phhA in E . coli. Transcriptional fusions were constructed using lacZ as a reporter gene, as diagrammed in Fig. 2-2B. The results indicate that the negative effect conferred by upstream DNA occurs at the transcriptional level.

PAGE 32

24 (A) pJZ9 pJZ9-3a pJS7 pJS60 si i i 1 kb a a j i_ phhA phhB phhC i PhhA activity (nmol/min/mg) 2.7 94.0 3.5 69.7 (B) pJS61Z pJS62Z tacZ. tacZ P-Galactosidase activity (Miller Units) 17 1250 FIG. 2-2. Localization of a regulatory region upstream of the phh operon. (A) On the right phenylalanine hydroxylase (PhhA) activities are shown in E. coli JP2255 harboring different plasmids shown on the left. (B) On the right /S-galactosidase activities are shown in BW545 harboring the phhA' -lacZ transcriptional fusions diagrammed on the left .

PAGE 33

25 Identification of phhR A large open reading frame (Fig. 2-3) located upstream of the phh structural genes appeared likely to be functional on the criterion of GCG codon preference analysis. The gene, denoted phhR, produces a deduced protein having 518 residues, an anhydrous molecular weight of 56,855, and an isoelectric point of 7.17. It contains a single tryptophan residue. Regions corresponding to a possible a 70 promoter region and a factorindependent transcription terminator are marked in Fig. 2-3. A strong ribosome-binding site was not apparent. Bases that are complementary to P. aeruginosa 16S rRNA at the 3' terminus are marked. Perhaps the "A-richness" of the initiator region enhances ribosome binding (Ivey-Hoyle and Steege, 1992) . A physical map is given in Fig. 2-4 of the 5874 -bp DNA segment containing the structural genes of the phh operon, the divergently transcribed regulatory gene phhR, and a gene (pbpG) downstream of the phh operon which encodes a penicillin-binding protein (Song and Jensen, unpublished data) . Homology of PhhR with E. coli TyrR The closest homolog of PhhR was found to be E . coli TyrR. The pairwise GAP alignment (GCG) is shown in Fig. 2-5. TyrR belongs to a family of modular proteins which usually have three functional domains. The alignment showed high level of

PAGE 34

26 JS -10 Ui i • phhX TGTTQCTOGTGAGTCrAACTGTCACATATKTITXMCGGAAATTrc 120 M R I K V H 6 Hindi ACTGCCAGAACCGTGTAGGCATCCTCOXX^CATCCTC^ 2*0 C Q N R V G I L R D I LNLLVDYGINVNKGEVGGDQGNAIYLLCP 46 rCAAGCGCGTCGGCCTGATGCCCAGCGAGOXXXXX* 360 KRVGLMPSBRRHLBLNA 86 H:r,i: : : CGCTGCTOXX^^CTGGACTTCCCGGTGCTCTCG^ 4B0 LLAALDPPVLSVDMGGQI VAANRAAAQLLGVRVDEVPG I P 126 TGCAATCCGAGCACGACGAGAGCGAGGCGCTGGCTGGAGCGGTG^ 720 QS BHDBSBALAGAVLTLHRADRVGBRIYHVRKQSLRGPDS 206 GCGCCCGCCCGGAGGGCAAGCTCGGCXnGCTGGAGC^ 1060 ARPBGKLGLLBLTAGGTLPLDGVGEMSFRLQAKLLRPLQD 326 Kcofil ACGGCTGTTTCCGCCt>3GTCGG<^GCGACGAGGAGGTGTAC 1200 GCPRRVGSDBBVYLDVRVICATQVDLSBLCAKGEPRQDLY 366 Hindi 1 1 CGAAGfTTTITXTXIGCAGGCGCTGGAAOGTCTCGAAGGCTACCACTQGC^ 1440 KLSAQALBRLHRYHNPGHVROLBNVLPQAVSLCBGGTVKA 446 ^CCTTGTGAAAGGGCGGCAGCACGATGAGGCGTGGGGAAGACGGCCAGGATGTTC^ 1800 FIG. 2-3. Nucleotide sequence of the phhR region. The numbers at the right indicate nucleotide and amino acid positions. The putative promoter region and ribosome -binding site (RBS) are indicated with bold print . RBS bases that are complementary to P. aeruginosa 16S rRNA are overlined. The translat ional start site is indicated by a bent arrow and the stop codon by an asterisk. Nucleotides forming the complementary stems of the putative transcriptional terminator are marked with tandem arrowheads. Restriction endonuclease recognition sites are marked above the nucleotide sequence.

PAGE 35

27 phhR phhA phhB phhC pbpG FIG. 2-4. Physical map of the DNA fragment containing phh structural genes, the divergentlytranscribed regulatory gene phhR, and the downstream penicillin-binding protein gene (pbpG) in pJZ9. Terminators downstream of phhC and phhR are indicated. The shaded bars at both ends are portions of the multiple cloning site of the pUC18 vector. The location of restriction sites is shown.

PAGE 36

tn • tn S3 o\» c -H f> -H TS CU S3 IT) 43 0 a 0 CQ CQ CO (U -H M (4 0 u 0) 4J -H 4J c Q) U T) Tl a -h CO tr a 0) cC CQ rrt oV> T) ro rd ^ 43 4J CD 0 o o u o 4-1 0 4-1 >^43 4J tn -H -H u u rd rH cu H 43 e -u -H c cu . — -H CU m cq i-t cn CX43 CTl S 4-1 rH rH ' 4-> cu -H 43 CU i? 4-1 CO ^ O ^ d) rH Sh CO rd cu -H ^ > 1 H H > CTl S4 H CQ . CO 4J rd CD re o 43 &ft a -H < co cu O 0 >4 -"cm g cu ro 44 CQ -U cu c CQ CU O . tn 4_> CU CQ _ CQ CU C w 3 O fit 3 CO CQ TJ O cu xs CQ rd S3 43 O 0 ^ S3 s3 H -h rH ^ C n >H S8| 0 « CQ ^44 § « °^ 0 a ' H CU CQ ^ rH CQ CU J1 CU ^ rd 43 a X CQ CU S3 2 ^ 4-> TJ O § H g ^ CO o X) C -H S3 -H 4-> (0 £ > . CU H T3 1 y £ S3 * -H 43 M CD 4-1 CU ^ H CU CQ 4-1 I H O CU CQ -H 3 tn 0 H rd ft ft rH cd O -H 4J u u >i C -H 4-1 •H . M CU in 0 cu I U rH CM 43 • 4-1 • Kl O cu m O 4^ [l, 4J H CU CO m CQ CQ ^ CU CO "J •H 43 H rH CO CQ T3 CU CU ^ rH rH 0 rS C r-l C ^ CU O E rC 43 O T) CO m C !h -H cfl 0 3 4lHU CQ i (0 43 ^ u 13 CU -H o > 4-1 TS H -H U U CU a co CQ CU CU Sh > cn CO CTl &^ CO o •H CQ CQ CU U 0) M a H 4J U 0) 4-1 4-1 rd 4-1 o 4^ -U -H 3 C o H 4J re > •H 4J O rti (U 4-) rd H TS 0) E i •X u >i H 4^ CQ rd CO a O CQ U •H O 4-1 rH CO > a H -H 4J OtJrrJ CO CU fH 4J a rH H 3 111 „, m cu cu CU ^ 2 CU m rri a H (0 j-, CQ 4_| 44 cu a x 43 M -H 4-1 rH • CU 4-j a 4^ co o •H CU , — i CQ CQ 43 C CQ 4-1 o CU H ,4 f ^ & -H CO CU d -H 4^ 43 TJ 3 rH CQ O rrt ^ rd fO rH CO a cu r^ P CQ CU ^ 43 "5 4J e ro g CQ S3 O 43 -H 4J 4-1 rd C 4-> M 31 6 • U , 4J W H 44 g ^ O ro X) S3 rd CN (Tl CQ U O S3 H S3 "H H rH P. u xs CU r4 S3 rd . ft c a £j S3 Q rH ^ X CQ cd -no) §s s| rd r-j d ^-^^ S CM X CJ . i "H -H Ti W rH 4-1 ^ CU "rH £ CQ 43 rH S § CO ° CU T! ^ S3 H 01 ^ CO T3 E -H TJ O U S3 T3 (0 (0 S3 0) cu 3 CU C4 4J rH CU 43 43 3S O S3 TS O •H 0 tn cq CU rH rH Cd 3 4-1 I -H y 5 Tl H U S! rd CO > ^ 7 1 SH CO rH 44 CU d cu H CQ -h XS 4J C E c O O 14 3 E cj cu jj CQ U >, , S3 CU rH (J O SO.43 -H S3j cn cu tn H -H 43 cu N 43 4J rH

PAGE 37

29 CO U3 o H — I 10 « q £3 Si 5« It) •H QJ O O u w I ft u >l {-I 0) id ft ft A A ft CO o rco tro >* O O t~ H CN CN CN • > > mm* a • J M M fa H O CO j ft CO • Q W fa • J fa — fa co • Q < o w fa • J CO • BJ > o • Q > •• J < • W Q •• W Q M • :S a — a W a o • OS H o o w ft H a) ft i ft a. o u w ft a) ft i ft ft O u w ft ft ft ft O U w I ft 0) ft ft X! X! ft • a s > o-a 0 0) 0 o u w 1 ft w ft « ft u M X! fn ft ft ft o V w I ft M N f-< ft i ft ft

PAGE 38

30 conservation throughout the entire length of PhhR and TyrR, and 45.7% of the deduced residues were identical. The Nterminal domain mediates regulatory modulations, and in TyrR it binds all three aromatic amino acids. A central domain, highly conserved throughout the entire family of a 54 enhancerbinding proteins, exhibits two established motifs that reflect the binding of ATP (Pittard, 1996) . Site A corresponds to the ATP-binding pocket motif and site B corresponds to segment 3 of adenylate kinase. In this region a perfect leucinezipper motif is apparent in P. aeruginosa PhhR, whereas E. coli displays an imperfect motif. Residues E-274, G-285, and E-302 were found to be important for TyrR-mediated repression of aroF-tyrA in E . coli (Yang et al . , 1993; Kwok et al . , 1995), and these residues are all conserved in P. aeruginosa PhhR. The C-terminal domain possesses a helixturn-helix motif which is responsible for DNA binding. The absolute conservation of residues shown to be critical in E. coli (Pittard, 1996) strongly indicates that PhhR and TyrR might target to a similar DNA sequences. Similar to E. coli TyrR, the two aspartate residues and the lysine residue conserved in the aminoterminal domain of all response regulator proteins (Stock et al . , 1989) were not found .

PAGE 39

31 Overproduction of PhhR PhhR protein was overexpressed in E. coli BL21(DE3) as detailed under Materials and methods by use of the T7 overexpression system; the construct is illustrated in Fig.26. The initial use of overexpression vectors containing phhR on the BamEI-SphI fragment (see Fig. 2-4) of pJZ9 failed. This is probably due to autogenous regulation of phhR, judging from the precedent set by tyrR in E . coli (Argaet et al . , 1994) . Accordingly, overexpression was achieved through excision of DNA upstream of phhR. PCR methodology was used to generate an intact phhR gene which was fused with the T7 translational start codon at a JVdel restriction site to create overexpression plasmid, pJS88. E. coli BL21(DE3) that had been transformed with pJS88 was induced with 1 mM IPTG for 3 hours to express PhhR. Whole-cell lysates obtained before and after IPTG induction were analyzed by SDS-PAGE, as shown in Fig. 26B . Overproduction of a 56-kDa protein was observed, and Nterminal amino acid sequencing confirmed its synonymy with PhhR. Initial attempts to express phhR in E. coli under physiological conditions indicated that expression of phhR is highly toxic. The Bglll-EcoRI fragment from pJS88 was cloned into the BamHI -EcoRI site of pUC19 behind a lac promoter. When transformed into E . coli DH5a, transf ormants achieved only pinpoint colony size and eventual survivor cells inevitably had lost the plasmid. Success was finally achieved by use of

PAGE 40

32 (B) 12 3 FIG. 2-6. Overproduction of PhhR protein. (A) Map of PhhR overexpression plasmid pJS88 (left) , and low-copy number, constitutive PhhR expression plasmid pJS91 (right) . (B) SDSPAGE analysis of whole-cell lysate of E. coli BL21(DE3) harboring pJS88. The gel was stained with Coomassie blue. Lane 1, molecular-weight markers; lane 2, before IPTG induction; lane 3, induced by 1 mM IPTG for 3 h.

PAGE 41

33 pACYC184, a low copy-number plasmid, to create pJS91 which carried the Bglll-BanMI fragment of pJS88 ligated into the BamRI site of pACYC184 (Fig. 2-6A) . Analysis of 11 plasmids isolated showed that the orientation of phhR in each case was opposite to that of the Tc r gene. Presumably, the higher level of expression expected when driven by the Tc r promoter still confers an intolerable level of toxicity. Functional Replacement of E. coli tyrR with phhR A simple test was used to see whether phhR could substitute for tyri? as a repressor of the aroF-tyrA operon. Mutants deficient in TyrR exhibit resistance to :nf luorotyrosine (Fig. 2-7, middle) whereas tyri?* strains exhibit sensitivity to growth inhibitory effects of the analog (Fig. 2-7, left) . pJS91 {phhR + ) was used to transform an E . coli tyri?deficient background (strain SP1313) . The ability of PhhR to replace TyrR is qualitatively apparent (Fig. 2-7, right) by inspection of the halo of growth inhibition on a bacterial lawn surrounding a disc containing m-t luorotyrosine in SP1313 (tyri?" phhR*) . We also examined the ability of PhhR to replace TyrR as an activator of mtr, encoding a component of a tryptophanspecific transport system. The phhR* plasmid pJS91 was transformed into two E. coli X lysogens (Heatwole and Somerville, 1991) which carried mtr'-lacZ transcriptional fusions integrated in the chromosome as single-copy fusions.

PAGE 42

34 ABC FIG. 2-7. Functional replacement of TyrR by P. aeruginosa PhhR in E. coli, as monitored by sensitivity to growth inhibition by mfluorotyrosine (MFT) . (A) E . coli tyrR* (wildtype) strain SP1312 is very sensitive to MFT present on a central disc, exhibiting a large zone of growth inhibition; (B) E . coli tyrR' strain SP1313 is insensitive to MFT, showing no zone of growth inhibition; (C) P. aeruginosa phhP(pJS91) in trans complements E. coli tyrR' and restores the sensitivity to MFT, as visualized by a zone of growth inhibition.

PAGE 43

35 Strain SP1312 (tyrR*) exhibited the expected elevation of S-galactosidase activity following growth in the presence of tyrosine, phenylalanine, or both. However, strain SP1313 (tyrR~) carrying pJS91 {phhR*) produced the control level of S-galactosidase activity, regardless of the presence or absence of aromatic amino acids (data not shown) . Thus, PhhR appears to be incapable of replacing TyrR as an activator of E. coli mtr. Autogenous Regulation of phhR The BairiHI -Hindi fragment containing the 5' coding regions of phhA and phhR and the intervening region (see Fig. 2-4) was fused to lacZ to give the reporter-gene construct pJS102 {phhR' lacZ) . This plasmid construct was introduced into the tyrJ?-negative background of strain SP1313 in the presence or absence of pACYC184 possessing a phhR* insert. The results (Table 2-2) demonstrated a repressive effect of phhR* upon PhhR levels as monitored by measurement of (3galactosidase activity. Since the copy number of pJS91 (the source of PhhR molecules) in this experiment is lower than the number of repressor target sites provided by the high-copy number pJS102 and since TyrR boxes are present within seven other transcriptional units of E. coli, auto-regulation is undoubtedly grossly under-estimated due to titration of available PhhR molecules in the system.

PAGE 44

36 Table 2-2. Autoregulation of P. aeruginosa phhR in E. coli SP1313 (tyriT) containing pJS102 {phhR ' lacZ) Second /3-Galactosidase levels b in cells qrown in : plasmid a M9 C M9 + F M9 + Y pACYC184 550 510 589 pACYC184 {phhR*) 362 376 372 a pACYC184 {phhR*) is denoted pJS91 in Table 1. b /3-Galactosidase levels are reported in Miller Units. c M9 minimal medium was supplemented with 1 mM thiamine-HCl and, where indicated, 1 mM phenylalanine (F) , or 1 mM tyrosine (Y) .

PAGE 45

37 PhhR as A Positive Regulator PhhR and TyrR form a cluster within the larger family of a 54 enhancer-binding proteins, as illustrated by Fig. 2-8. A rpoN mutant of P. aeruginosa was assayed by Western analysis for PhhA levels of expression in order to determine whether expression of the phh operon is dependent upon a 54 like most family members, or whether it is a 54 -independent like tyrR and luxO. Only low basal levels of PhhA were present in the rpoN mutant, indicating expression to be largely a 54 -dependent . This, in turn, implied that phhR might function as an activator protein for phhABC transcription. phhR was inactivated as described under Materials and Methods, and Western analysis of the effect upon PhhA level was carried out. The results (Fig. 2-9) indicated that phhR encodes an activator, the absence of which allows only a low basal level of activity. The small molecules, L-phenylalanine and Ltyrosine, was found to function as an inducer (Fig. 2-10) . Western analysis of PhhA showed no detectable band in minimal medium and a barely detectable band when Ltryptophan was present, compared to prominent bands when L-phenylalanine or Ltyrosine was additionally present. Carbon-source levels of L-phenylalanine or Ltyrosine were not required for induction. It is probable that L-phenylalanine or L-tyrosine is a co-activator moiety which, in combination with PhhR, forms the holo-act ivator moiety. It is perhaps relevant that for those transcriptional

PAGE 46

FIG. 2-8. Homology relationships of the central domain of P. aeruginosa PhhR with the central domain of other members of the a 54 -dependent family of transcriptional regulators. The dendrogram was generated with amino acid sequences of the central domain as defined by Morrett and Segovia (1993) by using the PILEUP program of GCG . The top three proteins form a cluster designated as subfamily a, and the remaining proteins form a larger cluster designated as subfamily /3 . Due to their high degree of similarity, only one of the ortholog sequences of NifA, NtrC and HydG proteins is shown. The six paralogs from E. coli and the three paralogs from P. aeruginosa are designated with * and +, respectively. Abbreviations: Eco, Escherichia coli; Avi , Azotobacter vinelandii; Hin, Haemophilus influenzae; Pae, Pseudomonas aeruginosa; Vha, Vibrio harveyi . Functions controlled by the following regulators are given parenthetically: PhhR (phenylalanine hydroxylase) , TyrR (aromatic amino acid biosynthesis and transport), VnfA (nitrogen fixation, nitrogenase-2) , Anf (nitrogen fixation, nitrogenase-3 ) , NifA (nitrogen fixation, nitrogenase1 ) , HydG (hydrogen oxidation) , NtrC (nitrogen assimilation), PilR (synthesis of Type IV pili) , AlgB (alginate production) , LuxO (luminescence) , FhlA (formate metabolism) , YfhA (possible control of glnB) , PspF (phage shock protein) .

PAGE 47

*Pae -PhhR" *Eco -TyrR Hin -TyrR_ *Pae -PilR" *Eco -HydG *Eco -YfhA Avi -VnfA Avi -AnfA Avi -NifA *Eco -FhlA *Eco -NtrC *Pae -AlgB Vha -LuxO *Eco -PspF__

PAGE 48

40 PAO-1 PA103 WT pWiA~ phhR " WT rpoW FIG. 2-9. Western blot analysis of phhA expression in mutant derivatives of P. aeruginosa strains PAO-1 and PA103 . The proteins in crude extracts prepared from cultures grown in LB medium were separated by SDS-PAGE, and equal amounts of protein (50/xg) were applied to each lane.

PAGE 49

41 (A) Fructose — +Phe +Trp +Tyr PhhA* (B) z Glucose+Phe Fructose+Phe i ii ~ z § | O w w z 8 S S S £ a< Bu a. OS P £ NG O-l
PAGE 50

42 units where TyrR functions as an activator, L-phenylalanine functions as an essential co-activator (Pittard and Davidson, 1991) . Discussion Anomalous Repression of the phh Operon by PhhR in E. coli PhhR can mimic the ability of TyrR to repress the aroFtyrA operon at a a 70 promoter. This indicates that PhhR can recognize TyrR boxes and is consistent with the high similarity of the helixturn-helix, DNA-binding domain within the carboxyterminal segments of TyrR and PhhR. However, PhhR was unable to activate the phh operon in the heterologous E. coli background, suggesting an incompatibility between the E. coli RpoN and the P. aeruginosa a 54 -dependent system. The expression of PhhA from a promoter recognized by E. coli upstream of the native a 54 promoter was in fact severely depressed in constructs containing phhR, even in the presence of added co-activator (L-phenylalanine) . In the presence of P. aeruginosa PhhR, an aberrant complex apparently blocks transcription initiated upstream of the ct 54 promoter. Emerging Subfamilies within the a 54 Enhancer-Binding Protein Family P. aeruginosa PhhR belongs to an outlying subgroup (which we denote subfamily a in Fig. 2-8) of the a 54 enhancer-binding protein family. All members of the family possess in common a

PAGE 51

43 homologous central domain, but the aminoterminal and carboxyterminal domains may vary considerably within the family. Thus, this exemplifies a complex multi-domain protein family in which family membership is defined by a common ancestral central domain. Future subdivisions within what is termed subfamily f3 in Fig. 2-8 could likely be defined on the criterion of homology for the remaining two domains. For example, Eco-NtrC and Eco-FhlA belong to different mechanistic subgroups: the two-component regulatory system and direct response-to-small-molecules, respectively (reviewed by Shingler, 1996) . Figure 2-8 highlights the emerging homology relationships of selected paralog and ortholog proteins, with respect to the central domain. E. coli possesses at least six paralogs, some of which diverged in a common ancestor that existed prior to speciation events which generated orthologs . Thus, the divergence of Eco-NtrC and Pae-PilR was a more recent event than was the divergence of Eco-NtrC and Eco-PspF. In contrast to the ancient duplication events which generated all of the E. coli paralogs (or the P. aeruginosa paralogs) are the relatively recent duplication events generating the three paralogs which regulate three distinctly separate nitrogenase systems in Azotobacter (Joerger et al . , 1989). P. aeruginosa PhhR and E . coli TyrR exhibit homology in all three domains: 36% identity, aminoterminal ; 52% identity, central; and 47% identity, carboxyterminal ) . Curiously, the

PAGE 52

44 amino-terminal domain of H . influenzae TyrR appears to be absent. It is not known whether sequencing errors might account for this, or whether the equivalent of the aminoterminal domain might exist separately as a different protein. A multiple alignment of the central -domain modules of subfamilies a and /3 was shown in Fig. 2-11. In addition to the many residues that are absolutely conserved throughout the family, some residues which may prove to be uniquely conserved within subfamily a are apparent, e.g., APLL corresponding to residues 29-32 of Hin-TyrR. Both Eco-TyrR and Rca-NtrC exhibit deletions in the "unique-gap region" of the central domain (Fig. 2-12) in correlation with their regulation of a 70 promoters, rather than a 54 promoters. This observation led to the suggestion (Morrett and Segovia, 1993) that this region of the central domain might be critical for functional interfacing with a 54 . Since this DNA segment of Pae-PhhR is intact with absolute retention of highly conserved residues, the foregoing hypothesis is consistent with the successful interaction of PhhR with a a 54 promoter. Hin TyrR, on the other hand, is likely to be deficient in interaction with a 54 (like E. coli TyrR) , owing to a 6-residue deletion in this region. Intervening Region of Divergent Transcription Since the DNA-binding region of the carboxy terminus of PhhR is identical at all important residues with E . coli TyrR,

PAGE 53

FIG. 2-11. A comparison of the amino acid sequences in the central domain of the PhhR protein and 13 other homologs . The sequences were aligned by using the PILEUP program of GCG. The numbering of amino acid residues is given on the left. Percent identity of PhhR with its homologs is given at the lower right. Amino acid residues conserved in all 14 sequences are in double-lined boxes. Amino acid residues conserved in 13 of 14 sequences are shaded. Conserved residues which are confined to either the top cluster (subfamily a) or the bottom cluster (subfamily /8) are in single-lined boxes. Two ATP-binding motifs are indicated above the consensus sequences in boldface type. See the legend for Fig. 8 for abbreviations.

PAGE 54

46 ATP-binding Motif A G E a a_ t G K PhhR-Pae F 0 S S R V H A A V R R M A P L D A P L L * E G E T G T G K E L L A R A C H LA6PRGQS 257 TyrR-Bco V A V S P K 1 K H V I " i 1 K L A M L S A P L L X T G D T G T G K D L F A "i A c H QASPRAGK 256 TyrR-Hin I V 0 S E A K H K S A R P A M F D A P L L x Q G E T a S G K D L_ L A K A c H YQ5LRRDK 65 PilR-Pae L G E s P P R A L R N Q I G K L A R S Q A P V Y X S G E S a S G K E L V A R L I H BQOPRIBR 186 HydG-Bco V G K s P A « Q H L L S E I A L V A P s E A T V L X H G D s a T G K E L V A R A I H ASBARSEK 191 YfhA-Bco V T R s P L L R L L E 0 A R L V A a 8 D V S V L x N G Q s a T G K E I P A 0 A I H NA3PRNSK 187 VnfA-Avi I G N s K P j L E V Y Q L I E R V V R T R T T V I t L G E s a V G K E L V A G A I H ynspaakg 260 AnfA-Avi I G N s K P M Q E V Y E L I H K V A S T K A T V L i L G E s G V G K E L V A N A I H YNSPNAEA 269 NifA-Avi V G H T P T M R R V F D Q I R R V A K W N S T V L V L a E s G T G K E L I A S A I H YKSPRAHR 261 FhlA-Bco I G R S E A 1 Y S V L K Q V E M V A 0 S D S T V L i L G E T G T G K E L I A R A I H NLSGRNNR 431 NtrC-Bco I G E A P A « Q D V P R I I G R I S R s S I S V L t N G E S G T G I E L V A H A L H RHSPRAKA 190 AlgB-Pae E S H S P A H A A V L B T A R Q V A A T D A N I L X L G B s G S G K G E L. A R A I H TWBKRAKK 197 LuxO-Vha I G S s Q T 1 0 Q V Y R T I D S A A S S K A S I P X T G E s G T G K E V c A E A I H AABKRGDK 183 PspP-Eeo L G E A N S r L E V L B 0 V S H L A P L D K P V L x I G E R G T G K L I A S R L H YL8SRWQG 58 PhhR-Pae p F M A e A G L P E S M A B T 8 P Y G p[g]a|T E G~A R P E G K L E TyrR-Bco P Y L A A S I P B D A V B 8 8 a A . P E G X \ F E TyrR-Hin K F I _A s A G L P D E D A b a 8 M P K V G D S E T i j F _E PllR-Pae P F V P V N c G A I P S E L M B a 8 P P G ir K K "g S P T G A . I E D K Q P a HydG-Bco P L V I L N e A A L N E S Li L b a 8 L P a H E K G A F T G A . D K R R E G R F V YfhA-Bco P P I A I N e G A L P E 0 L L B S 8 L F G H A R G A P T G A . V S N R E G L F Q VnfA-Avi P F V K P N e A S L P E s V I B S * L P G H E K G S P T G A . I G L R K G R P E AnfA-Avi A L V T S N C A P L P E N L A 8 B B L F G H E K G S F T G A . L T M H K G C F E NifA-Avi P F V R L N c A A L P E T L L 8 £ B L P G H E K G A F T G A . V K a R K G R P E PhlA-Bco R M V K M N e A A M P A G L L « 8 t> L P G H E R G A P T G A . s A 0 R I G R F E NtrC-Bco P F I A L N H A A I P K D L I S 8 8 H E K G A P T G A . H T i R Q G R P E AlgB-Pae P Q V T I N e P S L T A E L M b a 8 1 S R G A P T G A . T E s T L l V S LuxO-Vha P F I A I N C A A I P K D I I b a 8 I I V K G A P T G A . A N D R j A E PapP-Bco P P I S L N s A A L N B N L L D S 8 E A G A F T G A . 0 K R H F E ATP-binding Motif B UL « V L q a rIf i PilR-Pae HydG-Bco YfhA-Bco VnfA-Avi AnfA-Avi NifA-Avi PhlA-Bco NtrC-Bco AlgB-Pae LuxO-Vha PspP-Bco VADLPMAH IGDISPMM IGDMPAPL VGEMSLTT VGELSPTV IGBZSPMP P L B L P L D V P L T L LCEMDLDL LATAPMMV2 V G D M I G D M I G D P ltagstlf qAngssyl YjiiNKBTVL A A S G G T E A D G S T A A E G G T E A A G G T Q I D G G T Q I D G G T L ft D K S S Q | D G $ T Q ft D G ft T L A D G G T R A D G 8 T V I c A I 0 V D L S E L C A 357 V I c A T a K N L. V E L V Q 348 V I c T 3 0 V P L H L L V E 159 thkdlaae thrdlaae thrdlpka Tnrn£aem tnrnlvem tnrdlese tnrdlkkm THQNLEQR TNRDLGAM TNRDFWKB TNADLPAM 285 290 285 359 368 360 530 289 296 282 157 PhhR-Pae k a B t:« Q D L Y H R L N V L S L H p L R E C TyrR-Bco K G M * : 8 B D L Y Y R L N * L T L N I p L R D C TyrR-Hin Q | K V'% A D L F H R L N L T I N V A L R D R PilR-Pae A 1 R ? S Q D L Y Y R L N S I E L R V p P L R E R HydG-Bco A G r *::S o D L Y Y R L N * V A I E V p S L R 0 R YfhA-Bco R e E V S B D L Y Y R L N » V S L K I p A L A D G VnfA-Avi d a T W'i E D L Y Y R L N • F P I T I p P L s E R AnfA-Avi Q S T fc» B D L Y Y R L N * P P I T V p P L R E R NifA-Avi k a K F R B D L Y Y R L N V M A I R p P L R R FhlA-Bco D S B Ft! S D L Y Y R L N V F P I H P L r E R NtrC-Bco E G k *m b D L F H R L N V I R V H P L R E R AlgB-Pae Q G o »;:* b D L L Y R L N V I V L N i L R B R LuxO-Vha B G r *:* e D L Y Y R L Y « I P L H \ L R E R PspP-Eco E B t i-i a D L L D R L A t D V V 0 L R B R L d a L A P L A B O 0 A S R a I G C G L . P 407 P Q D I M P L T E i I s A R P A D s Q G V P R P 398 M A D I B _P_ L A 0 Q E I S E E L K I A K . _P_ 209 R E 0 I P L L A E R I L K R L A G D T G L P A . A 335 R E B I P L L A 6 H P L Q R F A E R N R K A V . K 340 T E C I P L L A N H L L R 0 A A D G H K P F V . R 334 G S P I I T L A D H P V S R F S R E M G I E V . N 409 G S 0 V I A L A D H F V S A F s R E N G K N V . K 418 T A B I P E L A E F L L G K I 0 R a Q G R P L T 409 P E 0 I P L L A K A F T F K I A R R L G R N I . D 580 R E P I P R L A R H F L Q V A A R E L G V E A . K 339 A B D I L G L A E R F L A R F V D Y G R P A . R 345 G K B V I I A Y S L L G Y M s H E E G K S F . V 332 E S D I M L H A E Y F A I Q M c R E I K L P L F P 208 % Identity PhhR-Pae K L S A 0 A L E R L E W t G N V R 0 L E N V L F 0 A V s L C E G G 442 TyrR-Bco K L A A D L N T V L T \ I I N P G N V R 0 L K N A I Y R A L T Q L D G Y 433 54( TyrR-Hin T F D K D F L L Y L Q H X G N V R E L Y N T L Y R _A C s L V 0 D N 244 47* PilR-Pae R L T G D A Q E K L K N Y R t P G N V G E L E N M L B R A Y T L c E D D 370 42% HydG-Bco G F T P Q A M D L L I H Y D N P G N I R E L E N A V E R A V V L L T G E 375 43% YfhA-Bco A F S T D A M K R L M T A S h e G N V R 0 L V N V I E 0 c V A L T s S P 260 45% VnfA-Avi R I S T P R L N M L Q S Y Q * s G N V R E L E N V I B R A M L Ii S E D G 444 47% AnfA-Avi R I S T P A L N M L M s Y H X P G N V R E L E N V M B R A V I L S D D D 452 45% NifA-Avi V T D S A I R L L M s H R w p G N V i E L E N C L E R S A I M s E D G 444 48% PhlA-Bco S I P A E T L R T L S N M E m f G N V S E L E N V I E R A V L L T R G N 615 44% NtrC-Bco L L H P E T B A A L T R L A h r G N V R Q L E N T C R W L T V M A A G Q 374 44% AlgB-Pae G F s E A A R E A M R 0 Y P X P G N V 8 E L R N V I E R A E I I C N 0 E 381 45% LuxO-Vha R P A Q D V I E R F N s Y E H P G N V R Q L 0 N V L R N I V V L N N G K 367 45% PspP-Eco G F T E R A R E T L L N Y R H P G N I R E L K N V V E R S V Y R H G T S 243 41%

PAGE 55

47 00 rin O in CT1 CO o o CN tCN a\ O o ro ro o oo ro ro o CN CN rH ro (N CN CN ro ro ro «# > > J M H J fa fa fa fa fa > Eh Eh Eh Eh Eh EH Eh Eh CO Eh Eh Eh Eh fa \.r o o O O o O o o CO O o o o os o fa O O o O 0 o 0 0 o o u 2 2 < to a u < Q a Q Q Q Q a u < < En *! < < < < < < < < < DS O >H < w w a o< fa a fa DS fa fa w a > a fa fa fa fa w CO fa u < fa fa fa fa fa fa fa fa fa fa fa > < fa fa fa J J DS u DS DS DS DS < DS O O o 0 C5 o o o o 0 o o o o fa H fa CX w w t>s H OI fa a CM fa fa fa OS DS X os DS DS Eh DS X o fa 0 a Pi 2 s a a H CO a fa w W u w fa to 0 Eh < Eh fa 2 fa CM Cm M a > H > CO 2 < DS < < < < < < > Cm fa fa PS « hS DS « DS fa < < fa o fa w < fa fa fa fa fa CO > fa X OS >H K fa fa 33 X X X fa > o 0 O o O o o o o o 0 d> fa Ex* fa fa fa fa fa fa fa fa fa fa fa s fa fa fa J J fa J fa J fa fa I" w W fa w W fa fa fa Q fa fa fa w w to EH w to to W CO to CO CO CO CO CO CO w w fa fa w w w w W w fa fa fa Q CM > < £ fa M < i — i t — 1 H E M o 4 a s fa J J > fa fa J fa fa Q Q w w w CQ o CO 2 a Q w Q 2 < U Q w to fa w fa fa fa < « < a fa o a. fa fa CM 2 fa Cm Cm CM CM CM H Cm 2 fa a a « o * < «! CQ s fa CJ H n H 0 *4-J H u 01 O u Pi >1 0 M 0 > 0 0 u i 0 o £ j> O 0 w w Cm Cm w w w w m PS SH H CU >i 3 E-i i CU -H -H 3 E E ro -H M-l C • A 3 CQ 3 cn » CU O X rH X! fa O fa I o o ^ cu fa^T3 « fa ^ CU ^ rH 6 m u c CU rn 0 o -H U-l Cn CU CO U a cn ro H Cn^ x; o X 01 -H 1-1 5 cu X > CD -H i — ! CO §« 'O CQ ^ CU ^ ° "5 rc P 0 m ro CU u ^ . 0 rJ X! . ^ CD « (U x 3 «5S X CQ c fa X "5 I! X! -H J-) • 3 CM CQ H CQ CU 1 G U ^-h a CU CU fa 3 o cr M M CU Qj CQ o H fa _j cu cu a > h -H CU rO M a X — 3 X w 0 crj cu x fa CQ H CQ CU 3 cu X rH U 4J U rO 0

PAGE 56

48 it is likely that PhhR binds to the same binding sites for E . coli, which are referred as "TyrR boxes" (consensus: TGTAAAN 6 TTTACA ) . This conclusion is also supported by the ability of PhhR to replace TyrR as a repressor of the aroFtyrA transcriptional unit. The location of two "PhhR boxes" which match the consensus for "TyrR boxes" was shown in Fig. 2-13. PhhR Box 1 is a strong box (with more conservedsymmetry and higher affinity for TyrR) that overlaps the putative -10 region of the phhR promoter. TyrR boxes in E. coli occur in tandem with variable spacing (Pittard, 1996), and a TyrR hexameric molecule is thought to bind both a strong box and a weak box with DNA looping in between. PhhR Box 2 is a weak box located in the middle of the intervening region. It seems probable that by analogy with autorepression of tyrR in E. coli, both phhR boxes participate in the autogenous repression of phhR by PhhR, probably with tyrosine as a corepressor. In the opposite direction of transcription, the a 54 promoter for phhABC requires an upstream activator site (UAS) . PhhR Box 1 may be the most likely UAS, although perhaps both boxes participate in activation of phhA. L-Phenylalanine and Ltyrosine, potent inducers of phenylalanine hydroxylase, presumably are the effector molecules. Since a rpoN mutant retained low basal level of PhhA, another promoter that is independent of a 54 might be present. No motif for binding of integration host factor (IHF) (Friedman, 1988) was located in the intervening region.

PAGE 57

49 Bam i C CTAGGCGAGCACCC CGTCCGGCTCGACAACGTACGGCAG CTCCATAAGGACTGTCCGCGC CGGAAGCTAGTGGAAGTCAACGGCCCACTAGTCCCATAAGGTCTGGACCACGAGCC^AA PIREHPLGLQEIGDLYEQCARGEIVKLQRTILTKHVQHBT ACGGACTCCATGAGCGGCITGTGGCCGCGGGGTCTTTGTTGTTGTCGTTGCC^ GGCCTATCACCTACTTTGGCAACAGCAGCCCGACCGCCCGGTGCATGACGCAGCAAAAGTATGCC TQA^ EPYH I FGNDDPQRAVYQTTKM 553" U phhA i ' -12 PhhR Box 2 T ACCGCGTCGGG A CGGTCCGTGCAGCC C CGGATAGGGACCGGGCAACGGGGAGG AATCGGCGTTTT CGTAAAGTTTTCCTTACG AATTGG CCTGGGTCGCCTGTTCATTGGGTCAGG CAT 3 60 ATG
PAGE 58

50 Therefore, this region may possess intrinsic DNA-bending capabilities . Function of the phh Operon The primary function of the phh operon is clearly not to accommodate tyrosine biosynthesis since the feedbackinhibited cyclohexadienyl dehydrogenase which is widely distributed in gram-negative bacteria exists for this purpose. However, the phh operon probably provides a fortuitous backup capability for tyrosine biosynthesis. "Reluctant auxotrophy" for tyrosine (Patel et al . , 1978) can be explained as follows. Mutational deficiency of cyclohexadienyl dehydrogenase would lead to accumulation of prephenate, a potent product inhibitor of chorismate mutase . The subsequent backup of chorismate, enhanced by lack of early-pathway control in the absence of Ltyrosine, results in passage of chorismate to the periplasm where chorismate mutase-F (Gu and Jensen, unpublished data) and cyclohexadienyl dehydratase (Zhao et al . , 1993) generate L-phenylalanine . Subsequent induction of phenylalanine hydroxylase completes the alternative circuit to L-tyrosine. The established function of phenylalanine hydroxylase in mammals is for catabolism of L-phenylalanine as a carbon source. We have found that phenylalanine hydroxylase is indeed essential for use of L-phenylalanine as a sole carbon source in P. aeruginosa. Thus, inactivation of phhA resulted in inability to use L-phenylalanine as a sole source of carbon

PAGE 59

51 (data not shown) . However, induction of the phh operon under conditions where better carbon sources (such as glucose) coexist, suggests that the phh operon might be dedicated to provision of some specialized compound from L-phenylalanine . Inactivation of phhR resulted not only in the inability to use L-phenylalanine as a carbon source, but also in an inability to use L-tyrosine as a carbon source. Since TyrR regulates aromatic amino acid permeases in E. coli, we considered the possibility that the phhR mutant might fail to grow on L-tyrosine because of a permease deficiency. Since MFT is likely to be transported by the same system as L-tyrosine, a permease-def icient phenotype should be resistance to growth inhibition by MFT. However, the phhR" mutant has a MFTsensitive phenotype on f ructase-based medium (data not shown) . Therefore, PhhR might regulate steps of tyrosine catabolism. Regulation of Multiple Transcriptional Units by PhhR? TyrR represses or activates eight transcriptional units in E. coli (Pittard, 1996) . Similarly organized transcriptional units are absent or unknown in P. aeruginosa . However, the counterpart of the aroF-tyrA operon in P. aeruginosa would be genes encoding tyrosine-sensitive DAHP synthase and cyclohexadienyl dehydrogenase. Physiological manipulations in our laboratory have never revealed repression control of these apparently constitutive enzymes. Consistent with this, PhhR exhibits no regulatory control of either of

PAGE 60

52 these enzymes, on the criterion of assessment of specific activities determined in comparison of tyrR* and tyrR' backgrounds (data not shown) .

PAGE 61

CHAPTER 3 B I FUNCTIONAL PhhB REGULATES THE EXPRESSION OF PHENYLALANINE HYDROXYLASE IN Pseudomonas aeruginosa Introduction Mammalian 4a-carbinolamine dehydratase was initially known for its catalytic activity of converting 4acarbinolamine to quinonoid dihydrobiopterin in regenerating the tetrahydrobiopterin for phenylalanine hydroxylase (Fig. 31B) . Later, it was found to be synonymous with DCoH, the dimerization cofactor for hepatic nuclear factor 1 alpha (HNFlot) (Citron et al . , 1992). A homolog of the mammalian DCoH, PhhB, was found in Pseudomonas aeruginosa by Zhao et al . (1994) . The PhhB protein is encoded by the second structural gene, phhB, of the phh operon (Fig. 3-1A) . Zhao et al . (1994) reported that phhB is required for the expression of phenylalanine hydroxylase, encoded by the first structural gene, phhA. In the absence of the phhB gene, phhA by itself not only failed to complement E. coli tyrosine auxotrophy, but was not expressed in E. coli as indicated by SDS-PAGE. Dual catalytic and regulatory roles of PhhB are an intriguing possibility in the context of the fact that DCoH, the mammalian counterpart of PhhB, is a bifunctional protein with enzymatic activity as 4a53

PAGE 62

54 FIG. 3-1. (A) Physical map of the phh operon in Pseudomonas aeruginosa . The endonuclease restriction sites are shown at the top. The arrows indicate the position of the genes and the directions of transcription. Putative transcriptional terminators (t inside a circle) are indicated. The proteins encoded by the genes are as follows: phhR, a 54 transcriptional activator of the phh operon; phhA, phenylalanine hydroxylase; phhB, 4a-carbinolamine dehydratase; and phhC , aromatic aminotransferase. (B) Regeneration of the pterin cof actor for phenylalanine hydroxylase. The enzymes involved are indicated as follows: PhhA, phenylalanine hydroxylase; PhhB, 4acarbinolamine dehydratase; and DHPR, dihydropteridine reductase .

PAGE 63

55 carbinolaraine dehydratase and regulatory activity as the dimerization cof actor of HNFla. In this chapter, I report the results of studies aimed at elucidation of the extent and nature of the regulatory function of PhhB protein. Materials and Methods Bacterial Strains, Plasmids, Phage, and Media The bacterial strains, plasmids, and phage used in this study are listed in Table 3-1. The LB and M9 formulations (Sambrook et al . , 1989) were used as growth media for E. coli and P. aeruginosa. Pseudomonas isolation agar (Difco) was used for isolating P. aeruginosa knockout mutants. Additions of ampicillin (100 /ig/ml), chloramphenicol (40 fig/ml) , kanamycin (50 /xg/ml) , mercuric chloride (15 /xg/ml) , Lphenylalanine (50 /xg/ml) , and thiamine (17 /ig/ml) were made as indicated. Agar was added at a final concentration of 2% (w/v) for preparation of solid medium. Recombinant DNA Techniques Molecular cloning and DNA manipulation including plasmid purification, restriction enzyme digestion, ligation, and transformation were conducted by standard methods (Sambrook et al . , 1989) . DNA fragments were purified from agarose gel with a Geneclean kit (Bio 101) . Electroporat ion (Invitrogen) was used for simutaneous transformation of E. coli with two compatible plasmids. Restriction enzymes, T4 DNA ligase, DNA-

PAGE 64

56 Table 3-1. Bacterial strains, plasmids, and phages used in this study Strain or plasmid Relevant genotype or description Source or reference E . coli BL21 (DE3) F" ompT hsdS B (r B ~m B ~) gal dcm; wit~h DE3 a X cronhaop rarrvina the T7 RNA polymerase gene Novagen recAl endAl gyrA96 thi-1 relAl O L-i iyj JJj nTDfn /DDT, LE392 F"el4" (McrA~ / >hsd.R514 (r^m^) sur>E44 sur>F58 lacYl or a (larT7,Y)f> galK galT22 metBl trpR55 Sambrook et al JP2255 aroF3 63 cheA3 61 r>he03^2 tvrATBP thi-1 stri?712 lacY"! xyl-15 R^l Hwi n £1 l CL _L ' — A W ±11 OC Davidson JS1 SP1313* {phhA' -lacZ) This study S17-1 [RP4-2 (Tc :Mu) (Km : Tn7 ) Tra ( incP) ] pro hsdR recA Tp r Sm r Simon et al . SP1313 zah-735 :TnlOA {argF-lac) U169& ( tyrR) ~\-J (=± fT A 7r~i 1 £. n.ccx w
PAGE 65

57 Table 3-1. (continued) pETllb T7 lac promoter, lad* Ap r Novagen pET23 lilac promoter, lacl + Ap r Novagen pGEM-3Z T7 promoter, Ap r Promega pGST-DCoH In-frame protein fusion of glutathione S-transf erase and DCoH Citron et al pJSlO phhAB, 2.5-kb Hindi fragment This study cloned into pGEM-3Z behind the T7 promoter pJSll phhAB' , 1.44-kb Hind I -EcoRV This study fragment cloned into pACYC177 pJS12 phhAB, 2.5-kb Hindi fragment This study cloned into pACYC177 pJS51 Hindi -BamHI fragment containing This study truncated phhA' cloned into pACYC177 pJS51Z phhA' -lacZ transcriptional fusion This study in pACYC177 pJS63 phh'ABC, BairiH.1 -Hindi 1 1 fragment This study cloned into pGEM-3Z behind the T7 promoter pJS72 phhA, PCR-generated fragment This study containing the native ribosome -binding site and PhhA-coding region cloned into pET23 behind T7lac promoter pJS95 PhhA overexpression vector; This study PhhA-coding region fused with T7 translational initiation signal at Ndel site of pETllb pJS96 PhhA overexpression vector; This study phhA fused with T7 translational initiation signal cloned into pUC19 behind lac promoter to const itutively overexpress PhhA

PAGE 66

58 Table 3-1. (continued) pJS97 PhhA overexpression vector; phhA fused with T7 translational signal cloned into pTrc99A behind trc promoter pJSlOl Hg r -cassette, Ap r pJS105 Hindi -BamH.1 PCR fragment containing phhA' with a frameshift pJS105Z phhA'-'lacZ protein fusion cloned into pACYC177 pJZ9 phhRABC, Ap r pJZ9-3a phhAB, Ap r pJZ9-4 phh'ABC, Ap r pJZ9-5 phhAB' , Ap r pMC1871 lacZ protein fusion vector pTrc99A Trc promoter, lacV Ap r pUFR004 ColEl replicon, Cm r Mob + mobP , lacZa* Z1918 Promoterless lacZ, Ap r This study Song & Jensen This study This study Zhao et al . Zhao et al . Zhao et al . Zhao et al . Pharmacia Pharmacia DeFeyter et al Schweizer Phages XRZ5 XJS1 X'bla 'lacZ lacT \${phhA' -lacZ) lacT 'bla Resental et al This study

PAGE 67

59 modifying enzymes (New England Biolab or Promega) , Taq DNA polymerase (Perkin-Elmer) , and Vent DNA polymerase (New England Biolab) were used as recommended by the suppliers. Phenylalanine Hydroxylase Assay E. coli JP2255 (pJZ9-3a) was grown at 37°C in 500 ml of LB broth supplemented with ampicillin (100 /xg/ml) , and harvested at the late-exponential phase of growth. Cell pellets were resuspended into 8 ml of 10 mM potassium phosphate buffer (pH 7.4) containing 1 mM dithiothreitol (DTT) , and the cells were disrupted by sonication. The resulting extract was centrifuged at 150,000 x g for 1 hr at 4°C. The supernatant was desalted using Sephadex G-25 and used as crude extract for enzyme assay. Phenylalanine hydroxylase (PhhA) was assayed by following tyrosine formation (Nakata et al . , 1979). Phenylalanine Hydroxylase Stimulation Assay 4a-Carbinolamine dehydratase activity in E. coli (pJZ9-4) was assayed indirectly using the phenylalanine hydroxylase stimulation assay (Citron et al . , 1992). Reaction mixtures containing 30 mM potassium phosphate (pH 8.3), catalase (1 mg/ml) , 10 0 /xM NADH, 1 mM phenylalanine, 2 0 /xg dihydropteridine reductase, 14.4 /xg rat liver phenylalanine hydroxylase, and 2.9 /xM 6 , 7 -dimethyltetrahydropterin were incubated at 25°C. Approximately 1 min after the reaction was started, either buffer (control) or 15 /xg of the E . coli

PAGE 68

60 crude extract containing PhhB or GST-DCoH was added. The reaction was monitored at 34 0 nm for the oxidation of NADH by dihydropteridine reductase as quinonoid dihydropterin was recycled to tetrahydropterin . Construction of phhA: lacZ Transcriptional and Translat ional Fusions For the transcriptional fusion (phhA' -lacZ) , the HincIIBamHI fragment containing the upstream region of phhA was first cloned into pACYC177 that had been digested with Hindi and BarriKI , creating pJS51. A BamEI -cassette of a promoterless lacZ gene from the plasmid Z1918 was then inserted at the BarnH.1 site of pJS51 in the same orientation as phhA to create pJS51Z, which was used as a low-copy phhA' -lacZ fusion. A single-copy fusion \(phhA' -lacZ) was obtained by transferring the phhA' -lacZ fusion from pJS51Z into XRZ5 following the procedure described by Yu and Reznikoff (1984) . For the translational fusion (phhA' ' lacZ) , the HincIIBamHI fragment containing the upstream region of phhA was generated by PCR with the upper primer 5'GACAGAGCAGGTAGATGGCGTT 3 ' , and the lower primer 5 ' GGGATCCGGCTCGTGGGGCAGGCCGA3 ' (BarriRI site underlined). An extra guanine nucleotide (G in bold) was added in the lower primer to create the frameshift needed for an inframe fusion at the BamHI site to generate phhA'-'lacZ. The HincII-BamHI fragment with the frameshift was inserted into the HincIIBamHI site of pACYC177 to create pJS105, and a BamHI -cassette

PAGE 69

61 of truncated ' lacZ from pMC1871 was inserted into pJS105 to create the translat ional fusion plasmid, pJS105Z. 16-Galactosidase Assay jS-Galactosidase activity was assayed under conditions of proportionality as described by Miller (1972), and specific activities are expressed in Miller units. The data are the results of at least two independent assays. Construction of PhhA and PhhB Expression Vectors To express PhhA protein, expression plasmids pJS72 and pJS95 were constructed. A PCR fragment containing the complete coding region of phhA and the native ribosome-binding site (RBS) was amplified with the upper primer 5'CATGGAGTCCGTATGAAAACGACGCA-3 ' (RBS underlined; ATG start codon in bold) and the downstream primer 5'CTTGGTTGTCGCATGTGGGAGCGGCG3 ' , and cloned into pET23 behind the Tllac promoter to create pJS72 . pJS95 was constructed by inserting the coding region of phhA into the translational fusion vector pETllb. The coding region was amplified by PCR with the upstream primer 5 ' CCATATGAAAACGACGCAGTACGTG 3 ' and the downstream primer 5 ' -CAAGTCTGGTTGTCGCATGTGGGAGCGGCG-3 ' . The upper primer was made with a built-in JVdel site (underlined) which allows fusion of phhA at the translational start site (ATG in bold) with the T7 translational initiation signals. To constitut ively express the PhhA protein, the

PAGE 70

62 phfrAcoding region together with the upstream T7 translat ional start signals were excised from pJS95 as a Xbal fragment and cloned into pUC18 downstream of a lac promoter to create pJS96 . The Xbal fragment was also cloned into pTrc99A downstream of the inducible trc promoter to create pJS97. Two similar plasmids, pJSlO and pJS63, were constructed to express PhhB . The Hindi fragment containing both phhA and phhB gene was inserted into pGEM-3Z to create pJSlO, and the BanMI-Hindlll fragment containing both phhB and phhC was inserted into pGEM-3Z to create pJS63 . The phhB gene was under the control of a T7 promoter in both plasmids. Preparation of PhhB-specif ic Polyclonal Antiserum PhhB was partially purified by anionexchange and gelfiltration chromatography following the methods described by Zhao et al . , (1994). The partially purified PhhB was subject to SDS-PAGE (12%) and the gel was stained with Commassie blue R-250. The PhhB band was cut from the gel and used for the production of polyclonal antiserum in rabbits (Cocalico Biologicals, Inc., Reamstown, PA). Antiserum was purified by using an Econo-Pac protein A column (Bio Rad) and further absorbed with a total cell extract from the PhhB-def icient mutant JS103 . SDS-PAGE and Western Blot Analysis SDS-PAGE (12% gel) was performed with the Mini-PROTEAN II cell (Bio-Rad) by the method of Laemmli (1970) . Samples of

PAGE 71

63 exponent ial -phase cells were collected by centrif ugation, and the cell pellets were suspended in gel-loading buffer and heated at 100°C for 10 min. Samples of 5-10 fil were loaded onto two SDS-polyacrylamide gels. After separation of the proteins by electrophoresis, one gel was stained with Coomassie blue and the other gel was used for blotting. When crude extracts were used, equivalent amounts of protein were loaded into each lane. Western blots were performed according to Towbin et al . (1979) . The proteins were electrophoretically transferred onto nitrocellulose membranes and reacted with the polyclonal antiserum at a dilution of 1:1000. Membranes were then incubated with secondary alkaline phosphatase-labelled anti-rabbit antibodies at a dilution (1:30,000) and developed by adding NBT and BICP as chromogenic substrates (Bibco BRL) for alkaline phosphatase. Gene Inactivation Both phhB and phhC were inactivated following the method described by Song and Jensen (1996) . To generate the truncated 'phhB' fragment (308bp), the upper primer 5'ACCCAAGCCCATTGCGAAGCCTGCCG3 ' , and the lower primer 5'GTGCGCGCCGCCATGATGAAATCGTT3 ' were used. To generate the truncated 'phhC fragment (652bp), the upper primer 5'GTCGAGCAGGAAACCACCAAGA3 ' , and the lower primer 5'GTTGGCTACGCAGGTCGGTGAG3 ' were used. Interruption of the phhB

PAGE 72

64 or phhC gene in a Hg r isolate was confirmed by Southern hybridization . Southern Hybridization Genomic DNA was extracted from the P. aeruginosa phhB~ strain by the method described by Silhavy et al . (1984) . Southern hybridization was performed as described by Sambrook et al . (1989). The DNA was completely digested with EcoRI , separated by electrophoresis in 1% agarose gel, and transferred to a nylon membrane (Bio-Rad) . The DNA was fixed by baking the membrane under vacuum at 80 °C for 2 hr and hybridized at 42 °C overnight with the truncated 'phhB' (the same as used for gene inact ivat ion) probes that had been labeled with biotin-14-dATP using a BioNick labelling system (GIBCO/BRL) . The membrane was washed in 2X SSC (300 mM NaCl , 30 mM sodium citrate, pH 7.0) plus 0.1% SDS (twice for 3 min each time at room temperature), in 0 . 2X SSC plus 0.1% SDS (twice for 3 min each time at room temperature), and in 0.16X SSC plus 0.1% SDS (twice for 15 min each time at 50°C) . The probes were detected with the BluGene nonradioactive nucleic acid detection system (GIBCO/BRL) . RESULTS PhhB Has 4a-Carbinolamine Dehydratase Activity PhhB is a homologue of an established 4a-carbinolamine dehydratase. To confirm that PhhB catalyzes the 4a-

PAGE 73

65 carbinolamine dehydratase reaction, I used the phenylalanine hydroxylase stimulation assay where the utilization of 4acarbinolamine limits the rate of the hydroxylation (Huang et al . , 1973; Citron et al . , 1992) . Either PhhB or DCoH was able to stimulate the phenylalanine hydroxylase reaction in E. coli crude extracts where DCoH was used as a postive control (Fig. 3-2) , indicating that PhhB protein has 4a-carbinolamine dehydratase activity. Furthermore, using the expression construct pJS63, PhhB protein was purified in the laboratory of Dr. June E. Ayling at the University of South Alabama and the 4a-carbinolamine dehydratase activity of PhhB was confirmed by direct assay (personal communication) . Complementation of Tyrosine Auxotrophy by phhA and phhB in trans Both phhA and phhB are needed for functional complementation of E. coli tyrosine auxotrophy (Zhao et al . , 1994) . If phhB functions as both a structural gene and a regulatory gene in a fashion that parallels the mammalian homologue, it would be expected to complement in the trans comf igurat ion with respect to phhA. A trans-complementation study was done in which phhA and phhB (or DCoH) were inserted into two compatible plasmids, pJSll and pJZ9-4A (or pGSTDCoH) , respectively. The results (Table 3-2) did indeed show that phhB was able to complement E. coli tyrosine auxotrophy in trans with respect to phhA and that mammalian DCoH was able

PAGE 74

66 2 4 6 8 10 Time (min) FIG. 3-2. Stimulation of phenylalanine hydroxylase activity by the addition (at the arrow) of a crude extract of E. coli JP2255 containing PhhB or GST-DCoH fusion protein. Approximately 1 min after the reaction was started, either buffer, or 15 fig of the crude extract containing PhhB, or GSTDCoH was added. The reaction was monitored at 34 0 nm for the oxidation of NADH by dihydropteridine reductase as quinonoid dihydropterin was recycled to a tetrabiohydropterin (see Fig. 3-1) .

PAGE 75

67 Table 3-2. Complementation of an E . coli Tyr a mutant by phhA and either phhB or DCoH gene in trans Plasmid ( s ) Relevpnt genotype E. coli Tyr~ mutant pJSll phhA + No pJS12 iphhAB) * Yes pJZ9-4 phhB + No pJSll + pJZ9-4 phhA* phhB + Yes pGST-DCoH DColT No pJSll+pGST-DCoH phhA* DColT Yes a E. coli JP2255 (Phe Tyr ) mutant was used as the host strain for this complementation study. b E. coli JP2255 harboring various plasmids was plated on M9 + phenylalanine plates supplemented with appropriate antibiotics as selective agents.

PAGE 76

68 to replace PhhB in the bacterial system. The results also ruled out any possible cis effect of phhB on the expression of phhA. Expression of phhA in the Presence or Absence of vhhB The trans-complementation study confirmed that complementation of E. coli tyrosine auxotrophy by phhA requires the presence of phhB. To understand whether the requirement of phhB for the complementation was due to increased expression of phhA in the presence of PhhB, the expression of phhA in the presence or absence of phhB in E. coli JP2255 was studied through Western analysis (Fig. 3-3). A substantial level of PhhA was still detected when only phhA was present. The presence of either PhhB or DCoH in trans indeed increased the expression of phhA, but only by about 2fold. Although this result indicated that PhhB may regulate the expression of phhA, much more than the relatively modest 2-3 fold reduction in PhhA level observed in the phhB~ mutant was expected. This was because little or no PhhA had been detected in the absence of phhB (pJZ9-5) in E. coli, whereas a very high level of PhhA was produced in the presence of phhB (pJZ9-3a) (Zhao et al . , 1994). The inconsistency between the results of this study and that obtained by Zhao et al . (1994) was found to be due to the incorrectly reported orientation of the phhA insert in pJZ9-5. Sequencing the DNA insert in pJZ9-5 revealed that the orientation of the phhA gene was opposite to

PAGE 77

69 * * * *> ^> N^f S^> cf PhhA FIG. 3-3. Western blot analysis of PhhA expression in E. coli JP2255. Proteins in the whole cell lysates of JP2255 carrying various plasmids were separated by SDS-PAGE and reacted with rabbit anti-PhhA polyclonal antibodies. Plasmids containing the gene(s) shown above are as follows: phhA, pJSll; phhB, pJZ9-4; DCoH, pGST-DCoH. pUC18 was used as the control plasmid .

PAGE 78

70 the lac promoter, rather than in the same orientation as reported by Zhao, et al . (1994) . PhhB Regulates Expression at the Posttranscriptional Level To understand the regulatory role of PhhB in the expression of phhA, I constructed phhA' -lacZ transcriptional fusions in E. coli in both multi-copy form (pJS51Z) and a single-copy form [X {phhA' -lacZ) ] . In both cases, the presence of PhhB (pJZ9-4) or DCoH (pGST-DCoH) on a second plasmid did not result in a higher level of (S-galactosidase as compared to the control (pUC18) (Table 3-3), indicating that neither PhhB nor DCoH functions at the transcriptional level. I then constructed the translat ional fusion phhA'-'lacZ (pJS105Z) (Table 3-3) , and PhhB or DCoH was again provided in trans on a second plasmid. /3-Galactosidase activity increased about two-fold, indicating that PhhB regulates the expression of phhA at the translat ional or post-transcriptional level. This level of activation by PhhB is consistent with the results obtained from Western-blot analysis in E. coli, where similarly modest levels of activation in phhA expression were observed . Induction of The phh Operon by Phenylalanine in P. aerucrinosa Expression of both phhA and phhB were coordinately induced by phenylalanine in P. aeruginosa when grown on minimal fructose medium (Fig. 3-4) . The induction process is

PAGE 79

71 Table 3-3. Levels of phhA expression by phhB and DCoH in trans* . phhB or DCoH in trans -Galactosidase Activity^ pJS51Z X (phhA' -lacZ) PJS105Z pJZ9-4 (PhhB*) 180 13 . 2 13 . 1 pGST-DCoH (DCoH + ) 191 14 . 1 10 . 4 pUC18 (control) 182 15 . 7 7 . 3 a Regulation of phhA expression was studied using lacZ as the reporter gene and /3-galactosidase activities in transcriptional fusions pJS51Z (phhA' -lacZ) (multicopy) and X (phhA' -lacZ) (single copy), and translational fusion pJS105Z (phhA' ' lacZ) (multicopy) was assayed in absence (pUC18) or presence of PhhB (pJZ9-4) and DCoH (pGST-DCoH) . b iS-Galactosidase activities are reported in Miller units.

PAGE 80

72 1 2 3 4 5 6 7 FIG. 3-4. Induction of the phh operon by phenylalanine hydroxylase in P. aeruginosa. Proteins in the whole cell lysates were separated by SDS-PAGE. Lane 1, E . coli JP2255 harboring pJSll and pJZ9-4; lanes 2-7, samples taken after elapsed times of 0, 10, 30, 60, 90, 120 min, respectively, following addition of 100 ptg/ml phenylalanine at zero time.

PAGE 81

73 quite protracted and requires 90 minutes or more reach a maximum. A basal level of PhhB was expressed under noninducing conditions as seen in lane 2 on the Western blot. It is unknown whether this is due to expression from a promoter upstream of phhA or from an internal promoter within the coding sequence of phhA. Although the latter interpretation comes to mind because no PhhA band is visualized at zero time, this could be due to differing sensitivities of the antibodyprobes . Effect of phhB Knockout in P. aeruginosa. The phhB gene in P. aeruginosa was inactivated by chromosomal insertion of the suicide plasmid pUFR/ 'phhB ' /Hg r through a single homologous crossover event (Fig. 3-5A) . The insertional inactivation of the phhB gene in the resulting mutant was confirmed by Southern blot analysis (Fig. 3-5B) . The expression of PhhA in a phhB~ background was examined by Western-blot methodology under fully induced conditions fostered by growth on LB media (Fig. 3-6) . The PhhA level observed in the phhB~ mutant was reduced about 2-3 fold compared to the wildtype parent. No PhhA was detected in phhA~ mutant (negative control) . The PhhB level was also checked at the same time. A low basal level of PhhB was detected in phhA~ mutant when compared with that in the wildtype. No PhhB was detected in phhB~ mutant (negative control) .

PAGE 82

FIG. 3-5. Inactivation of phhB in P. aeruginosa . (A) Schematic representation for insertional inactivation of the chromosomal phhB gene by the integration of the suicide plasmid pUFR/ ' phhB' /Hg r through a single homologous crossover. The resulting Hg r mutant does not contain a complete copy of phhB gene, but instead has two truncated copies of phhB gene. (B) Southern-blot analysis of chromosomal DNA from P. aeruginosa PAO-1 wildtype (lane 1) and mutant JS103 (lane 2) . Chromosomal DNA was completely digested with EcoRI and probed at high stringency with the truncated 'phhB' fragment of pUFR/' phhB' /Hg r .

PAGE 83

75 mutant (B) 1 2

PAGE 84

76 * //// PhhA PhhB FIG. 3-6. Western blot analysis of PhhA and PhhB expression in P. aeruginosa PAO strains. Proteins in whole cell lysates were separated by SDS-PAGE and probed with rabbit anti-PhhA or anti-PhhB polyclonal antibodies.

PAGE 85

77 The physiological effect of the phhB knockout mutant in P. aeruginosa was examined. Inactivation of phhB abolished the ability to grow on either phenylalanine or tyrosine as the sole carbon source. However, interpretation of this result is complicated by results obtained with a phhC knockout mutant which was also not able to grow on either phenylalanine or tyrosine as the sole carbon source. Since insertion of the suicide plasmid into the chromosome of P. aeruginosa is expected to create polar effects on the downstream genes in the operon (as indeed seen in phhA knockout where amount of PhhB expressed was dramatically decreased) , it seems probable that the physiological effect observed in phhB knockout mutant is due to the polar effect on the expression of phhC. Overexpression of PhhA and PhhB Proteins Although PhhA was expressed to detectable levels in both E. coli (Fig. 3-3) and P. aeruginosa (Fig. 3-6) in the absence of phhB, initial attempts to express PhhA at high levels in the absence of PhhB in E. coli were unsuccessful. I then employed a T7 overexpression system (see Methods) to express PhhA in E. coli BL21(DE3) under induction conditions not requiring growth (Fig. 3-7A) . When phhA was expressed from a native ribosomal binding site in pJS72, high PhhA levels were produced after IPTG induction for 3 h (Lane 3, Fig. 3-7B). When phhA was expressed from $10 translational signals (pJS95) , higher levels of PhhA was made after IPTG induction

PAGE 86

78 (A) pJS72 RBS PhhA CATGGAG TCCGTATG AAAACG Native translational signal M K RBS PhhA pJS95 |P0 || GAAGGAGA| TATACAT ATG AAAACG T7 translational signal MKT (B) 97.4 66.2 45.0 31.0 21.5 14.5 1 2 3 4 5 FIG. 3-7. Expression of PhhA using T7 expression system in E. coli. (A) Construction of PhhA expression plasmids, pJS72 (with the native translational signals) and pJS95 (with 010 translational signals) . (B) SDS-PAGE analysis of expressed PhhA protein. Proteins in whole cell lysate were separated by SDS-PAGE and stained with Coomassie blue. Lane 1, molecular weight markers; lanes 2&3, BL21(DE3) harboring pJS72, before and after 1 mM IPTG induction for 3 h at 30°C, respectively; lanes 4&5, BL21(DE3) harboring pJS95, before and after 1 mM IPTG induction for 3 h at 30°C, respectively.

PAGE 87

79 for 3 h (Lane 5, Fig. 3-7B) . PhhA produced using this T7 overexpression system in E . coli is active. Phenylalanine hydroxylase activity was assayed in crude extract of E. coli BL2KDE3) harboring either pJS72 or pJS95 after IPTG induction, and compared with activity present in E. coli JP2255 (pJS9-3a) . Expression of phhA from 010 translat ional signals resulted in over a fivefold increase in phenylalanine hydroxylase activity (Table 3-4) . PhhB protein was expressed very well from a lac promoter in pJZ9-4 by Zhao et al . (1994) . We obtained an even higher level of PhhB by using the T7 overexpression system (Fig. 38) . The pJSlO construct used contains both phhA and phhB . After IPTG induction, only PhhB was overexpressed and little PhhA was made (Lane 3, Fig. 3-8) . A similar phenomenon was observed with pJS63 construct containing both phhB and phhC where only PhhB protein was overproduced (Lane 5, Fig. 3-8) . These results indicated that phhB has stronger translational signals and, therefore, is preferably translated over phhA and phhC. PhhB protein expressed is fully active, and has been purified and characterized from the latter construct by Ayling (unpublished results) . Expression of PhhA without PhhB in E. coli Has Growth Inhibitory Effects Even though a low level of PhhA was produced from pJSll in the absence of PhhB, it did not complement the E . coli tyrosine auxotrophy in JP2255. A possible explanation was that

PAGE 88

80 Table 3-4. Phenylalanine hydroxylase activities in different expression clones 3 Expression Specific Activity clones (nanomoles/min/mg) JP2255/pJZ9-3a 91.7 BL21 (DE3) /pJS72 72.6 BL21 (DE3) /pJS95 379.2 a Cells of E. coli JP2255 harboring pJZ9-3a were grown in LB broth at 37°C and harvested at late exponential phase; cells of BL21(DE3) harboring pJS72 or pJS95 were grown in LB broth at 37°C to O.D=l and induced for 3 hr by addition of 1 mM IPTG. Crude extracts were used as the enzyme sources.

PAGE 89

81 FIG. 3-8. Expression of PhhB using T7 expression system in E. coli. Proteins in whole cell lysates were separated by SDSPAGE and stained with Coomassie blue. Lane 1, molecular weight markers; Lanes 2&3, BL21 (DE3 ) /pJSlO , before and after 1 mM IPTG induction for 3 h at 30°C, respectively. Lanes 4&5, BL21 (DE3 ) /pJS63 , before and after 1 mM IPTG induction for 3 h at 30°C, respectively.

PAGE 90

82 the PhhA produced from pJSll was not high enough to result in complementation. Thus, attempts were made to construct clones from which higher levels of PhhA could be produced to see whether complementation of E. coli tyrosine auxotrophy would then occur. Two constructs pJS96 and pJS97 were made with a Xbal fragment from pJS95 carrying phhA gene (fused with 010 translational signals) cloned into pUC18 behind a lac promoter or into pTrc99A behind a trc promoter, respectively (Fig. 39) . Both plasmids were found to be unstable and cells tended to lose the plasmid, especially with pJS96 where phhA is const itutively expressed at a very high level from the lac promoter. An elapsed time of several days was required to see pinpoint colonies with pJS96. E. coli carrying pJS97 was able to grow to a pinpoint colony overnight at 37°C without IPTG induction, while it took two or more days to see the pinpoint colonies on the plate with IPTG induction. These results indicated that high level of phhA expression in the absence of phhB triggers potent growth inhibition. Although the expression of phhA from the trc promoter on pJS97 has to be induced because of the presence of a lacT 1 gene on the plasmid, a high level of PhhA is still produced without IPTG induction because the trc promoter is very strong and leaky. Because PhhA was const itutively expressed from pJS96 at high level, the cells were not able to maintain the plasmid. Therefore, a high level of PhhA could not be overproduced in E . coli (pJS97) (Lane 1, Fig. 3-9B) . Since a lower level of

PAGE 91

83 (A) RBS PhhA P JS96 \Plac ||G A A G G A G A| T A T A C A T ATGAAAACG T7 translational signal MKT rbs | — PhhA pJS97 \Ptrc ||G A A G G A G A| T ATACAT ATGAAAACG T7 translational signal M K (B) 1 2 3 4 5 97.4 66.2 45.0 31.0 21.5 14.5 FIG. 3-9. Expression of PhhA in E. coli JP2255. (A) Construction of PhhA expression plasmids, pJS96 and pJS97. (B) SDS-PAGE analysis of PhhA expression. Proteins in whole cell lysates were separated by SDS-PAGE and stained with Coomassie blue. Lane 1, JP2255/pJS96 ; lane 2, JP2255/pTrc99A (control) ; lanes 3&4, JP2255/pJS97 , before and after 1 mM IPTG induction for 3 h, respectively; lane 5, molecular weight markers .

PAGE 92

84 PhhA was produced from pJS97 due to the presence of lacl q gene one the plasmid, the cells carrying pJS97 were better able to maitain the plasmid, thus being able to overproduce PhhA after IPTG induction (Lane 4, Fig. 3-9B) . Neither pJS96 nor pJS97 by itself was able to complement E. coli tyrosine auxotrophy. However, they were able to complement the auxotrophy when phhB was provided in trans on pJZ9-4 (data not shown) . This result indicates that PhhB was able to remove the inhibitory effect imposed by overproduction of PhhA on the host cells. To assure that apparent growth inhibition was not due to excessive conversion of phenylalanine to tyrosine in the phenylalanine auxotrophy background of strain JP2255, the plasmids were moved to a prototrophic background. E. coli DH5a carrying either plasmid was found to develop only pinpoint colonies on LB + Amp plates without IPTG induction, confirming that expression of phhA at higher levels created a general inhibitory effect on the growth of E. coli. Discussion Regulatory Role of PhhB? The mammalian PhhB homolog, DCoH, has both catalytic and regulatory functions. It was initially thought that P. aeruginosa phhB exerted an essential positive role in expression of phhA since constructs lacking phhB did not express phhA, as monitored by SDS-PAGE (Zhao et al . , 1994).

PAGE 93

85 Although phhA is in fact expressed in the absence of phhB, phhB does appear to exercise a positive regulatory role in the expression of phhA with a relatively modest effect of perhaps 2-3 fold. The transcriptional fusion approach seems to have eliminated the possibility of this regulation being at the transcriptional level, with the reservation that the experiments were performed in an E . coli background. I have shown that PhhR, the positive regulator of the phh operon in P. aeruginosa , does not interact properly with the E. coli a 54 machinery to activate phhA expression. But it is still possible that PhhB interacts with PhhR as a co-activator entity in the native P. aeruginosa organism. This would amount to a parallel with the mammalian system where DCoH is a coactivator for HNFla (an upstream enhancer element that can be considered comparable to PhhR) . However, I have presented results to show that PhhB does enhance expression of PhhA at the post transcriptonal level in E. coli. The 2-3 fold enhancement effects obtained correspond with the similar 2-3 fold magnitude of effect seen in P. aeruginosa , when comparing Western blots of PhhA in PhhB + and PhhB" backgrounds. If PhhB regulates at a post-transcriptional level, several possibilities envisioned include: (i) the sencondary structure of phhA mRNA might mask the ribosomal binding site. A stemloop structure has been located in this area. Binding of PhhB in this region might disrupt the secondary mRNA structure and enhance translat ional initiation.

PAGE 94

86 (ii) PhhB might bind phhA mRNA in the 3 ' region and protect against nuclease-catalyzed mRNA degradation, (iii) PhhB may complex with PhhA. This complex may protect PhhA from proteolysis. I have in fact obtained preliminary evidence for a PhhAPhhB complex. Rationale for Positive Regulatory Role of PhhB When PhhA is highly expressed in the absence of PhhB, the E. coli host cells become subject to drastic growth inhibition. The reasonable explanation for this effect of PhhB is that removal of the inhibitory effect generated by PhhA reaction is direct result of 4a-carbinolamine dehydratase activity. It is known that in the absence of 4acarbinolamine dehydratase activity, a 7isomer of tetrahydrobiopterin is generated during the reaction catalyzed by phenylalanine hydroxylase (Davis et al . , 1992) , and this 7isomer is a potent inhibitor of phenylalanine hydrxoylase in the mammalian sytem. Since potent growth inhibition persists in wildtype E. coli backgrounds where overexpressed phenylalanine hydroxylase has no purpose, inhibition cannot be attributed to inhibition of phenylalanine hydroxylase by the 7-isomer. Two explanations accounting for general growth inhibition come to mind, (i) The reduced pterin cof actor of E. coli may be depleted because PhhB is not present for recycling. Consequently, some pterin-dependent enzymes that are essential for growth may become limiting. (ii)

PAGE 95

87 Alternatively, the 7-isomer may be a potent inhibitor of a pterin-dependent enzyme needed for growth. Possible targets of inhibition could be dihydropteridine reductase or dihydrof olate reductase. The E. coli dihydropteridine reductase has been reported to possess broad specificity for pteridine compounds (Vasudevan et al., 1992) . When phhA was expressed at relatively low levels in E. coli no growth inhibition, or at least no severe growth inhibition occured. However, this PhhA was evidently not functional in vivo ( in the absence of PhhB) because complementation of tyrosine auxotrophy in the presence of exogenous L-phenylalanine was unsuccessful. On the other hand, the joint presence of PhhA and PhhB readily allowed functional complementation. I concluded that PhhA is an essential target of the 7-isomer. At the low levels of PhhA expression in the absence of PhhB, the 7-isomer is generated and inhibits PhhA function-but not enough 7-isomer is produced to cause general growth inhibiton. At high levels of PhhA expression in the absence of PhhB, sufficient 7-isomer is produced to inhibit one or more enzymes essential for the growth. With the above background, a rationale to explain a basis for selection of regulation of PhhA by PhhB is apparent. If it is correct that the 7-isomer generated from the carbinolamine pterin product of the PhhA reaction has general antimetabolite properties, then the significance of PhhB goes beyond its

PAGE 96

88 catalytic capability. It also diverts the carbinolamine substrate from an undesirable nonenzymatic fate.

PAGE 97

REFERENCES Abita, J. P., Milstien, S., Chang, N . , and Kaufman, S. 1976. In vitro activation of rat liver phenylalanine hydroxylase by phosphorylation . J. Biol. Chem. 251:53105314 . Argaet, V.P., Wilson, T.J., and Davidson, B.E. 1994. Purification of the Escherichia coli regulatory protein TyrR and analysis of its interactions with ATP, tyrosine, phenylalanine and Tryptophan. J. Biol. Chem. 269:51715178 . Armarego, W. L . , Randies, D., and Waring, P. 1984. Dihydropteridine reductase (DHPR) , its cof actors, and its mode of action. Med. Res. Rev. 43:267-321. Bailey, S. W. , Boerth, S. R. , Dillard, S. B., and Ayling, J. E. 1993. The mechanism of cofactor regeneration during phenylalanine hydroxylat ion . Adv. Exp. Med. Biol. 338:4754 . Baldwin, G.S., and Davidson, B.E. 1981. A kinetic and structural comparison of chorismate mutase/prephenate dehydratase from mutant strains of Escherichia coli K-12 defective in the pheA gene. Arch. Biochem. Biophys . 211 : 66-75 . Carr, R. T. and Benkovic, S. J. 1993. An examination of the copper requirement of phenylalanine hydroxylase from Chrowobacterium violaceum. Biochemistry. 32:14132-14138. Carr, R. T., Balasubramanian, S. Hawkins, P. C, and Benkovic, S. J. 1995. Mechanism of metal independent hydroxylation by Chromobacterium violaceum phenylalanine hdyroxylase. Biochemistry. 34:7525-7532. Chang, A.C.Y., and Cohen, S.N. 1978. Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the pl5A cryptic miniplasmid. J. Bacteriol. 134:1141-1156. 89

PAGE 98

90 Citron, B. A., Davis, M. D., Milstien, S., Gutierrez, J., Mendel, D. B., Crabtree, G. R., and Kaufman, S. 1992. Identity of 4a-carbinolamine dehydratase, a component of the phenylalanine hydroxylation system, and DCoH, a transregulator of homeodomain proteins . Proc . Natl. Acad. Sci. USA. 89:11891-11894. Cui, J., and Somerville, R.L. 1993. The TyrR protein of Escherichia coli: analysis by limited proteolysis of domain structural and ligand-mediated conformational changes. J. Biol. Chem. 268:5040-5047. Curtius, H. C., Adler, C., Rebrin, I., Heizmann, C., and Ghisla, S. 1990. 7 -Substituted pterins: formation during phenylalanine hydroxylation in the absence of dehydratase. Biochem. Biophys . Res. Commun. 172:10601066 . Davis, M. D. and Kaufman, S. 198 9. Evidence for the formation of the 4a-carbinolamine during the tyrosine-dependent oxidation of tetrahydrobiopterin by rat liver phenylalanine hydroxylase. J. Biol. Chem. 264:8585-8596. Davis, M. D. and Kaufman, S. 1991. Studies on the partially uncoupled oxidation of tetrahydropterins by phenylalanine hydroxylase. Neurochem. Res. 16:813-819. Davis, M. D., P. Ribeiro, J. Tipper, and S. Kaufman. 1992. " 7 -tetrahydrobiopterin, 11 a naturally occourring analogue of tetrahydrobiopterin, is a cofactor for and a potential inhibitor of the aromatic amino acid hydroxylases. Proc. Natl. Acad. Sci. USA. 89:10109-10113. Davis, M. D. , Parniak, M. A., Kaufman, S., and Kemper, E. 1996. Structure-function relationships of phenylalanine hydroxylase revealed by rediation target analysis. Arch. Biochem. Biophys. 325:235-241. Decicco, B. T. and Umbreit, W. W. 1964. Utilization of aromatic amino acids by Hydrogenomonas facilis. J. Bacteriol. 83:1590-1594.' DeFeyter, R., Kado, C. I., and Gabriel, D. W. 1990. Small, stable shuttle vectors for use in Xanthomonas . Gene 88 : 65-72 . Devereux, J., P. Haeberli, and 0. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.

PAGE 99

91 Dilella, A. G . , Kwok, S. C. M. , Ledley, F. D., Marvit, J., and Woo, S. L. C. 1986. Molecular structure and polymorphic map of the human phenylalanine hydroxylase gene. Biochemistry. 25:743749. Endrizzi, J. A., Cronk, J. D., Wang, W., Crabtree, G. R., and Alber, T. 1995. Crystal structure of DCoH, a bif unct ional , protein-binding transcriptional coactivator. Science. 268:556-559. Essar, D. W. , Eberly, L. ( Han, C.-Y., and Crawford, I. P. 1990. DNA sequences and characterization of four early genes of the tryptophan pathway in Pseudomonas aeruginosa. J. Bacteriol . 172:853-866. Faust, D. M . , Catherin, A. M . , Barbaux, S., Belkadi, L., Imaizumi-Sherrer , T., and Weiss, M. C. 1996. The activity of the highly inducible mouse phenylalanine hydroxylase gene promoter is dependent upon a tissue-specific, hormoneinducible enhancer. Mol . Cell. Biol. 16:31253137 . Friedman, D.I. 1988. Integration host factor: a protein for all reasons. Cell 55:545-554. Friedrich, B. and H. G. Schlegel . 1972. Die hydroxylierung von phenyalanin durch Hydrogenomonas eutrophus H16 . Arch. Mikrobiol. 83:17-31. Gambill, D. B., and Summers, A. O. 1985. Versatile mercuryresistant cloning and expression vectors. Gene 39:293-297 Gu, W. , and Jensen, R.A. 1996. Evolutionary recruitment of biochemically specialized subdivisions of Family I within the protein Superfamily of aminotransferases. J. Bacteriol. 178:2161-2171. Gudinchet, F., Maeder, P., Meuli, R. A., Deonna, T., and Mathieu, J. M. 1992. Cranial CT and MR I in malignant phenylketonuria. Pediatr. Radiol. 22:223-224. Guldberg, P., H. L. Levy, W. B. Hanley, R. Kock, R. Matalon, B. M. Rouse, F. Trefz, F. de la Cruz, K. F. Henriksen, and F. Guttler. 1996. Phenylalanine hydroxylase gene mutations in the United States: report from the Maternal PKU Collaborative Study. Am. J. Hum. Genet. 59:84-94. Guroff, G. and T. Ito. 1963. Induced soluble phenylalanine hydroxylase from Pseudomonas sp . grown on phenylalanine or tyrosine. Biochim. Biophys . Acta 77 : 159-161 .

PAGE 100

92 Haldenwang, W. G. 1995. The sigma factors of Bacillus subtilis. Microbiol. Rev. 59:1-30. Heatwole, V.M., and Somerville, R.L. 1991. The tryptophan specific permease gene, mtr , is differentially regulated by the tryptophan and tyrosine repressors in Escherichia coli K-12. J. Bacterid. 173:3601-3604. Holloway, B.W. (1955) Genetic recombination in Pseudomonas aeruginosa. J. Gen. Microbiol. 13:572-581. Huang, C. Y., Max, E. E., and Kaufman, S. 1973. Purification and characterization of phenylalanine hydroxylasestimulating protein from rat liver. J. Biol. Chem. 248 : 4235-4241 . Ivey-Hoyle, M., Steege, D.A. 1992. Mutational analysis of an inherently defective translation initiation site. J. Mol . Biol. 224:1039-1054. Jakoby, G. A. 1964. The induction and repression of amino acid oxidation in Pseudomonas fluorescens . Biochem. J. 92 : 1-8 . Joerger, R.D., Jacobson, M.R., and Bishop, P.E. 198 9. Two nifA-like genes required for expression of alternative nitrogenases by Azotobacter vinelandii. J. Bacterid. 171:3258-3267. Kamoun, S., Kamdar, H. V., Tola, E., and Kado, C. I. 1992. Incompatible interactions between crucifers and Xanthomonas campestris invole a vascular hypersensitive response: role of the hrpX locus. Mol. Plant-Microbe Interact. 5:22-23. Kappock, T. J., Hartins, P. C, Friedenberg, S., and Caradonna, J. P. 1995. Spectroscopic properties of unphosphorylated rat hepatic phenylalanine hydroxylase expressed in Escherichia coli. Comparison of resting and activated states. J. Biol. Chem. 270:30532-30544. Kaufman, S. 1975. Studies on the mechanism of phenylalanine hydroxylase : detection of an intermediate, p. 495-523. In W. Pfleidere (ed.) , Chemistry and biology of pteridines . Walter de Gruyter, Berlin. Kaufman, S. 1986. Regulation of the activity of hepatic phenylalanine hydroxylase. Adv. Enz . Regulation. 25:3764 .

PAGE 101

93 Kaufman, S. 1987. Phenylalanine 4 -monooxygenase from rat liver. Meth. in Enzymol . 142:3-27. Kaufman, S., Citron, B. A., Davis, M. , and Milstien, S. 1993 . The isolation and characterization of clones of 4ahydroxytetrahydrobiopterin dehydratase. Adv. Exp. Med. Biol. 338:97-102. Kaufman, S and Fisher, D. B. 1970. Purification and some physical properties of phenylalanine hydroxylase from rat liver. J. Biol. Chem. 245:4745-4750. Kwok, T., Yang, J. Pittard, A.J., Wilson, T.J., and Davidson, B.E. 1995. Analysis of an Escherichia coli mutant TyrR protein with impaired capacity for tyrosine-mediated repression, but still able to activate at a 70 promoters. Mol. Microbiol. 17:471-481. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4 . Nature (London) 227:680-685. Lazarus, R. A., S. J. Benkovic and S. Kaufman. 1983. Phenylalanine hydroxylase stimulation protein is a 4acarbinolamine dehydratase. J. Biol. Chem. 258:1096010962 . Letendre, C. H., G. Dickens, and G. Guroff. 1974. The tryptophan hydroxylase of Chromobacterium violaceum. J. Biol. Chem. 249:7186-7191. Letendre. C. H., G. Dickens, and G. Guroff. 1975. Phenylalanine hydroxylase from Pseudomonas sp . (ATCC 11299a) : purification, molecular weight, and influence of tyrosine metabolites on activation and hydroxylation . J. Biol. Chem. 254:1829-1833. Mendel, D. B. and Crabtree, G. R. 1991. HNF-1, a member of a novel class of dimerizing homeodomain proteins. J. Biol. Chem. 266:677-680. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N. Y. Morrett, E., and Segovia, L. (1993) The bacterial enhancesbinding protein family: mechanism of action and phylogenetic relationship of their functional domains. J. Bacteriol. 175:6067-6074. Nakata, H., Yamauchi, T. and Fujisawa, H. 1979. Phenylalanine hydroxylase from Chromobacterium violaceum: purification and characterization. J. Biol. Chem. 266:18454-18459.

PAGE 102

94 Onishi, A., L. J. Liotta, and S. J. Benkovic . 1991. Cloning and expression of Chromobacterium violaceum phenylalanine hydroxylase in Escherichia coli and comparison of aromatic amino acid sequence with mammalian aromatic amino acid hydroxylases. J. Biol. Chem. 266 : 18454-18459 . Patel, N . , Stemark-Cox, S., and Jensen, R.A. 1978. Enzymological basis of reluctant auxotrophy for phenylalanine and tyrosine in Pseudomonas aeruginosa . J. Biol. Chem. 253:2972-2978. Pember, S. 0. J. Villafranca, and S. J. Benkovic. 1986. Phenylalanine hydroxylase from Chromobacterium violaceum is a copper-containing monooxygenase : Kinetics of the reductive activation of the enzyme. Biochemistry 25 : 6611-6619 . Pittard, A.J. 1996. Biosynthesis of the aromatic amino acids. In Escherichia coli and Salmonella, pp 458-484. (F,C. Neidhardt, ed.) ASM Publications. Washington, D. C. Pittard, A.J., and Davidson, B.E. 1991. TyrR protein of Escherichia coli and its role as repressor and activator. Mol. Microbiol. 5:1585-1592. Pogge-yon-Strandmann, E. and Ryffel, G. U. 1995. Developmental expression of the maternal protein XDCoH, the dimerization cofactor of the homeoprotein LFB1 (HNF1) . Development. 121:1217-1226. Pontoglio, M., Barra, J., Hadchouel, M., Doyen, A., Kress, C, Bach, J. P., Babinet, C, and Yaniv, M. 1996. Hepatocyte nuclear factor 1 inactivation results in hepatic dysfunction, phenylketonuria, and renal Fanconi syndrome. Cell 84:575-585. Rosentel, J.K., Healy, F., Maupin-Furlow, J. A., Lee, J.H., and Shanmugam, K.T. 1995. Molybdate and regulation of mod (molybdate transport) , fdhF, and hyc (formate hydrogenlyase) operons in Escherichia coli. J. Bacteriol. 177:4857-4864. Sambrook, J., Fritsch, E. F., and Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual, 2 nd ed. Cold Spring Harbor, N.Y. : Cold Spring Harbor Laboratory Press. Schweizer, H.P. 1993. Two plasmids, X1918 and Z1918, for easy recovery of the xylE and lacZ reporter genes. Gene 134 : 89-91 .

PAGE 103

95 Shingler, V. 1996. Signal sensing by cr 54 -dependent reulators : derepression as a control mechanism. Mol . Microbiol. 19:409-416 . Silhavy, T. J., Berman, M. L . , and Enquist, L. W. 184. DNA extraction from bacterial cells, p. 137-139. In Experiments with gene fusions. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Simon, R., Priefer, U., and Piihler, A. 1983. A broad host range mobilization system for in vivo genetic engineering : transposon mutagenesis in gram negative bacteria. Bio/Technology 1:784-791. Simons, R.W. , Houman, F., and Kleckner, N. 1987. Improved single and multicopy lac-based cloning vectors for protein and operon fusions. Gene 53:85-96. Song, J. and Jensen, R. A. 1996. PhhR, a divergently trancribed activator of the phenylalanine hydroxylase gene cluster of Pseudomonas aeruginosa. Mol. Microbiol. 22:000-000 (in press). Stock, J.B., Ninfa, A.J., and Stock, A.M. 1989. Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol. Rev. 53:450-490. Tot ten, P. A., Lara, J.C., and Lory, S. 199 0. The rpoN gene product of Pseudomonas aeruginosa is required for expression of diverse genes, including the flagellin gene. J. Bacteriol . 172:389-396. Tourian, A. 1971. Activation of phenylalanine hydroxylase by phenylalanine. Biochim. Biophys . Acta. 242:345-354. Towbin, H, Staehelin, T. , and Gordon, J. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets : procedure and some applications. Proc . Natl. Acad. Sci . USA 76:4350-4354. Vasudevan, S. G., Shaw, D. C, and Armarego, W. L. F., 1988. Dihydropteridine reductase from Escherichia coli. Biochem. J. 255:581-588. Vasudevan, S. G., Paal, B. and Armarego, W. L. F. 19 92. Dihydropteridine reductase from Escherichia coli exhibits dihydrof olate reductase activity. Bio. Chem. HoppeSeyler. 373:1067-1073. Wang, W., Carey, M., and Gralla, J. D. 1992. Polymerase II promoter activation: closed complex formation and ATPdriven start site opening. Science 255:450-453.

PAGE 104

96 Williams, C. D., Dickens, G. , Letendre, C. H., Guroff, G., Haines, C., and Shiota, T. 1976. Isolation and characterization of dihydropteridine reductase from Pseudomonas species. J. Bacteriol . 127:1197-1207. Xia, T., and Jensen, R.A. 1990. A single cyclohexadienyl dehydrogenase specifies the prephenate dehydrogenase and arogenate dehydrogenase components of the dual pathways to L-tyrosine in Pseudomonas aeruginosa. J. Biol. Chem. 265 :200333-20036 . Yang, J., Ganesan, S., Sarsero, J., and Pittard, A.J. 1993. A genetic analysis of various functions of the TyrR protein of Escherichia coli. J. Bacteriol. 175:1767-1776. Yanisch-Perron, C., Vieira, J., and Messing, J. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC vectors. Gene 33:103119 . Yu, X.-M. and Reznikoff, W. S. 1984. Deletion analysis of the CAP-cAMP binding site of the Escherichia coli lactose promoter. Nucleic Acids Res. 12:5449-5464. Zhao, G. , Xia, T., Aldrich, H., and Jensen, R.A. 1993. Cyclohexadienyl dehydratase from Pseudomonas aeruginosa is a periplasmic protein. J. Gen. Microbiol. 139:807-813. Zhao, G., Xia, T., Song, J., and Jensen, R. A. 1994. Pseudomonas aeruginosa possesses homologues of mammalian phenylalanine hydroxylase and 4a-carbinolamine dehydratase/DCoH as part of a three -component gene cluster. Proc. Natl. Acad. Sci . USA 91:1366-1370.

PAGE 105

BIOGRAPHICAL SKETCH Jian Song was born on November 2, 1963, in Hebei Province, China. He completed his elementary and high school education in Xinhe County, Hebei Province. In 1980, he attended the Agricultural University of Hebei, where he majored in plant protection. He received his B.S. degree in 1984, then worked at the Institute of Plant Protection, Hebei Academy of Agricultural and Forestry Sciences until 1987. He was awarded a scholarship by Hebei Academy of Agricultural and Forestry Sciences to come to the Department of Entomology and Plant Pathology at the University of Tennessee, Knoxville for graduate study in 1988. He studied the interactions among the plant, aphid, and parasitoid under the supervision of Dr. Charles D. Pless . He received his M.S. degree in entomology in May, 1990. He then went to the Department of Entomology and Nematology at the University of Florida to continue graduate study toward a Ph.D. in entomology. He worked on insect toxicology under the supervision of Dr. Simon Yu until October, 1991. He then joined Dr. Roy A. Jensen's group at the Department of Microbiology and Cell Science in January, 1992 and studied regulation of phenylalanine hydroxylase system in Pseudomonas aeruginosa, and obtained his Ph.D. degree in May, 1997. He has accepted a postdoctoral position 97

PAGE 106

98 in Dr. Vojo Deretic's Laboratory in the Department of Microbiology and Immunology at the University of Michigan, and will join the laboratory in January, 1997 to work on Mycobacterium tuberculosis .

PAGE 107

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Roy A. Jensen, I/Chair Professor of Microbiolgy and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree/-of Doctor of Philosophy. Dean W. Gabriel Professor of Plant Pathology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Lonnie 0. Ingram Professor of Microbiolgy and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree) of Doctor of .^Philosophy . mes F. Preston rofessor of Microbiolgy and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Keelnatham T. Shanmugam Professor of Microbiolgy and Cell Science

PAGE 108

This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May, 1997 _ Dean. "College of A&riculture ., ^College of A"gricultu] Dean, Graduate School


67
Table 3-2. Complementation of an E. coli Tyr a mutant by phhA
and either phhB or DCoH gene in trans
Plasmid(s)
Relevent
genotype
Ability to complement15
E. coli Tyr- mutant
pJSll
phhA*
No
pJS12
(phhAB)*
Yes
pJZ9-4
phhB*
No
pJSll + pJZ9-4
phhA* phhB+
Yes
pGST-DCoH
DCoH*
No
pJSll+pGST-DCoH
phhA* DCoH*
Yes
a E. coli JP2255 (Phe Tyr ) mutant was used as the host strain
for this complementation study.
b E. coli JP2255 harboring various plasmids was plated on M9
+ phenylalanine plates supplemented with appropriate
antibiotics as selective agents.


55
carbinolamine dehydratase and regulatory activity as the
dimerization cofactor of HNFlu. In this chapter, I report the
results of studies aimed at elucidation of the extent and
nature of the regulatory function of PhhB protein.
Materials and Methods
Bacterial Strains, Plasmids, Phage, and Media
The bacterial strains, plasmids, and phage used in this
study are listed in Table 3-1. The LB and M9 formulations
(Sambrook et al., 1989) were used as growth media for E. coli
and P. aeruginosa. Pseudomonas isolation agar (Difco) was
used for isolating P. aeruginosa knockout mutants. Additions
of ampicillin (100 xg/ml), chloramphenicol (40 fjig/ml) ,
kanamycin (50 /xg/ml) mercuric chloride (15 /xg/ml) L-
phenylalanine (50 xg/ml) and thiamine (17 /g/ml) were made as
indicated. Agar was added at a final concentration of 2%
(w/v) for preparation of solid medium.
Recombinant DNA Techniques
Molecular cloning and DNA manipulation including plasmid
purification, restriction enzyme digestion, ligation, and
transformation were conducted by standard methods (Sambrook et
al., 1989) DNA fragments were purified from agarose gel with
a Geneclean kit (Bio 101). Electroporation (Invitrogen) was
used for simutaneous transformation of E. coli with two
compatible plasmids. Restriction enzymes, T4 DNA ligase, DNA-


37
PhhR as A Positive Regulator
PhhR and TyrR form a cluster within the larger family of
a54 enhancer-binding proteins, as illustrated by Fig. 2-8. A
rpoN mutant of P. aeruginosa was assayed by Western analysis
for PhhA levels of expression in order to determine whether
expression of the phh operon is dependent upon a54 like most
family members, or whether it is a54-independent like tyrR and
luxO. Only low basal levels of PhhA were present in the rpoN
mutant, indicating expression to be largely a54-dependent.
This, in turn, implied that phhR might function as an
activator protein for phhABC transcription. phhR was
inactivated as described under Materials and Methods, and
Western analysis of the effect upon PhhA level was carried
out. The results (Fig. 2-9) indicated that phhR encodes an
activator, the absence of which allows only a low basal level
of activity.
The small molecules, L-phenylalanine and L-tyrosine, was
found to function as an inducer (Fig. 2-10). Western analysis
of PhhA showed no detectable band in minimal medium and a
barely detectable band when L-tryptophan was present, compared
to prominent bands when L-phenylalanine or L-tyrosine was
additionally present. Carbon-source levels of L-phenylalanine
or L-tyrosine were not required for induction. It is probable
that L-phenylalanine or L-tyrosine is a co-activator moiety
which, in combination with PhhR, forms the holo-activator
moiety. It is perhaps relevant that for those transcriptional


95
Shingler, V. 1996. Signal sensing by cr54-dependent reulators:
derepression as a control mechanism. Mol. Microbiol.
19:409-416.
Silhavy, T. J. Berman, M. L., and Enquist, L. W. 184. DNA
extraction from bacterial cells, p.137-139. In
Experiments with gene fusions. Cold Spring Harbor
Laboratory, Cold Spring Harbor, N.Y.
Simon, R., Priefer, U., and Phler, A. 1983. A broad host
range mobilization system for in vivo genetic
engineering:transposon mutagenesis in gram negative
bacteria. Bio/Technology 1:784-791.
Simons, R.W., Houman, F., and Kleckner, N. 1987. Improved
single and multicopy lac-based cloning vectors for
protein and operon fusions. Gene 53:85-96.
Song, J. and Jensen, R. A. 1996. PhhR, a divergently
trancribed activator of the phenylalanine hydroxylase
gene cluster of Pseudomonas aeruginosa. Mol. Microbiol.
22:000-000 (in press).
Stock, J.B., Ninfa, A.J., and Stock, A.M. 1989. Protein
phosphorylation and regulation of adaptive responses in
bacteria. Microbiol. Rev. 53:450-490.
Totten, P.A., Lara, J.C., and Lory, S. 1990. The rpoN gene
product of Pseudomonas aeruginosa is required for
expression of diverse genes, including the flagellin
gene. J. Bacteriol. 172:389-396.
Tourian, A. 1971. Activation of phenylalanine hydroxylase by
phenylalanine. Biochim. Biophys. Acta. 242:345-354.
Towbin, H, Staehelin, T., and Gordon, J. 1979.
Electrophoretic transfer of proteins from polyacrylamide
gels to nitrocellulose sheets: procedure and some
applications. Proc. Natl. Acad. Sci. USA 76:4350-4354.
Vasudevan, S. G., Shaw, D. C., and Armarego, W. L. F., 1988.
Dihydropteridine reductase from Escherichia coli.
Biochem. J. 255:581-588.
Vasudevan, S. G., Paal, B. and Armarego, W. L. F. 19 92.
Dihydropteridine reductase from Escherichia coli exhibits
dihydrofolate reductase activity. Bio. Chem. Hoppe-
Seyler. 373:1067-1073.
Wang, W. Carey, M., and Gralla, J. D. 1992. Polymerase II
promoter activation: closed complex formation and ATP-
driven start site opening. Science 255:450-453.


FIG. 2-8. Homology relationships of the central domain of P.
aeruginosa PhhR with the central domain of other members of
the a54-dependent family of transcriptional regulators. The
dendrogram was generated with amino acid sequences of the
central domain as defined by Morrett and Segovia (1993) by
using the PILEUP program of GCG. The top three proteins form
a cluster designated as subfamily a, and the remaining
proteins form a larger cluster designated as subfamily 13. Due
to their high degree of similarity, only one of the ortholog
sequences of NifA, NtrC and HydG proteins is shown. The six
paralogs from E. coli and the three paralogs from P.
aeruginosa are designated with and 4*, respectively.
Abbreviations: Eco, Escherichia coli; Avi, Azotobacter
vinelandii; Hin, Haemophilus influenzae; Pae, Pseudomonas
aeruginosa; Vha, Vibrio harveyi. Functions controlled by the
following regulators are given parenthetically: PhhR
(phenylalanine hydroxylase), TyrR (aromatic amino acid
biosynthesis and transport), VnfA (nitrogen fixation,
nitrogenase-2), Anf (nitrogen fixation, nitrogenase-3), NifA
(nitrogen fixation, nitrogenase-1), HydG (hydrogen oxidation),
NtrC (nitrogen assimilation) PilR (synthesis of Type IV
pili), AlgB (alginate production), LuxO (luminescence), FhlA
(formate metabolism), YfhA (possible control of glnB), PspF
(phage shock protein).


21
pore size) resting on a prewarmed LB agar plate. The plates
were incubated for 16-24 hours at 37C, and then cells were
removed from the filter by an inoculation loop and resuspended
by vortexing into 0.5 ml of LB broth. Aliquots of 10-, 20-,
50-, and 100-fil volume of the cell suspension were spread onto
Pseudomonas isolation agar plates containing 15 /g of HgCl2.
The plates were incubated overnight and Hgr colonies were
isolated.
Preparation of PhhA-specific Polyclonal Antiserum
PhhA was partially purified by anion-exchange and gel-
filtration chromatography following the methods described by
Zhao et al., (1994). The partially purified PhhA was subject
to SDS-PAGE (12%) and the gel was stained with Commassie blue
R-250. The PhhA band was cut from the gel and used for the
production of polyclonal antiserum in rabbits (Cocalico
Biologicals, Inc., Reamstown, PA) Antiserum was purified by
using an Econo-Pac protein A column (Bio Rad) and further
absorbed with a total cell extract from the PhhA-deficient
mutant JS101.
SDS-PAGE and Western Blot Analysis
SDS-PAGE (12%) was performed with the Mini-PROTEAN II
Cell (Bio-Rad) by the method of Laemmli (1970) Samples of
exponential-phase cells were collected by centrifugation, and
the cell pellets were suspended in gel-loading buffer and


17
al., 1989) DNA fragments were purified from agarose gel with
a "Geneclean" kit (BiolOl). Electroporation (Invitrogen) was
used for simultaneous transformation of E. coli with two
compatible plasmids.
Construction of PhhR Expression Vectors
For expression of PhhR protein in E. coli, the T7
expression system (Novagen) was employed. The phhR coding
region was cloned into a translational fusion vector pETllb.
Polymerase chain reaction (PCR) was used to amplify the phhR
gene. The upper primer (5'-ATACATATGCGTATCAAAGTGCACTGC-3')
was made with a built-in NdeI restriction site (underlined)
which allows fusion of phhR at the translational start site
(ATG in bold). The lower primer (5'-CCTCCACCGTTTCTTTCCCAGCCT-
3') was chosen at a position 48 bases downstream of the
translational stop codon. PhhR protein made from this PCR
fragment was designed to be a native protein, not a fusion
protein. The PCR fragment was cloned into a TA cloning
vector, pCRII. The phhR gene was excised from pCRII with Nde I
and EcoRI. The Ndel-EcoRI fragment was first ligated with
EcoRI-BamHI adaptor to create a Ndel-BamHI fragment which was
then ligated with pETllb digested with Nde 1 and BamHI to
create the PhhR expression plasmid, pJS88 (Fig.2-3A).
For construction of a PhhR constitutive expression
plasmid, pACYC184 was chosen as the expression vector. The
pACYC184 vector has a P15A origin of replication which is


15
Table 2-1
. (continued)
Plasmids
pUC18
Ampr lac'IPOZ'
Yanisch-
Perron et al.
pUC19
Ampr lac'IPOZ'
Yanisch-
perron et al.
pACYC184
P15A replicn, Cmr Tcr
Chang & Cohen
pETllb
T7lac promoter,lacl" Apr
Novagen
pRS1274
lacZY fusion vector
Simons et al.
Z1918
Promoterless lacZ, Apr
Schweizer
pJZ9
phhRABC, Apr
Zhao et al.
pJZ9-3a
phhAB,Apr
Zhao et al.
pJS7
phhRABC, Apr
This study
pJS6 0
phhAB C, Apr
This study
pJS61Z
phhRA'-lacZ transcriptional
fusion, Apr
This study
pJS62Z
phhA'-lacZ transcriptional
fusion, Apr
This study
pJS8 8
pETllb carrying phhR translational
fusion at the ATG start site
This study
pJS91
pACYC184 carrying phhR* Cmr
This study
pJS102
pRS1274 carrying phhR '-lacZY
transcriptional fusion
This study
pCRII
Ampr Kanr lacZa
Invitrogen
pDG106
Hgr Kmr P15A replicn
Gambill &
Summers
pJSlOl
Pstl-Smal fragment of pDG106
inserted into pUC18
This study
pUFRO 04
ColEl Cmr Mob+ mob{ P)
DeFeyter et al


REFERENCES
Abita, J. P., Milstien, S., Chang, N., and Kaufman, S. 1976.
In vitro activation of rat liver phenylalanine
hydroxylase by phosphorylation. J. Biol. Chem. 251:5310-
5314 .
Argaet, V.P., Wilson, T.J., and Davidson, B.E. 1994.
Purification of the Escherichia coli regulatory protein
TyrR and analysis of its interactions with ATP, tyrosine,
phenylalanine and Tryptophan. J. Biol. Chem. 269:5171-
5178 .
Armarego, W. L., Randles, D., and Waring, P. 1984.
Dihydropteridine reductase (DHPR), its cofactors, and its
mode of action. Med. Res. Rev. 43:267-321.
Bailey, S. W., Boerth, S. R., Dillard, S. B., and Ayling, J.
E. 1993. The mechanism of cofactor regeneration during
phenylalanine hydroxylation. Adv. Exp. Med. Biol. 338:47-
54 .
Baldwin, G.S., and Davidson, B.E. 1981. A kinetic and
structural comparison of chorismate mutase/prephenate
dehydratase from mutant strains of Escherichia coli K-12
defective in the pheA gene. Arch. Biochem. Biophys.
211:66-75 .
Carr, R. T. and Benkovic, S. J. 1993. An examination of the
copper requirement of phenylalanine hydroxylase from
Chromobacterium violaceum. Biochemistry. 32:14132-14138.
Carr, R. T., Balasubramanian, S. Hawkins, P. C., and
Benkovic, S. J. 1995. Mechanism of metal-independent
hydroxylation by Chromobacterium violaceum phenylalanine
hdyroxylase. Biochemistry. 34:7525-7532.
Chang, A.C.Y., and Cohen, S.N. 1978. Construction and
characterization of amplifiable multicopy DNA cloning
vehicles derived from the pl5A cryptic miniplasmid. J.
Bacteriol. 134:1141-1156.
89


65
carbinolamine dehydratase reaction, I used the phenylalanine
hydroxylase stimulation assay where the utilization of 4a-
carbinolamine limits the rate of the hydroxylation (Huang et
al. 1973; Citron et al., 1992) Either PhhB or DCoH was able
to stimulate the phenylalanine hydroxylase reaction in E. coli
crude extracts where DCoH was used as a postive control (Fig.
3-2), indicating that PhhB protein has 4a-carbinolamine
dehydratase activity. Furthermore, using the expression
construct pJS63, PhhB protein was purified in the laboratory
of Dr. June E. Ayling at the University of South Alabama and
the 4a-carbinolamine dehydratase activity of PhhB was
confirmed by direct assay (personal communication).
Complementation of Tyrosine Auxotrophy by yhhA and phhB in
trans
Both phhA and phhB are needed for functional
complementation of E. coli tyrosine auxotrophy (Zhao et al.,
1994) If phhB functions as both a structural gene and a
regulatory gene in a fashion that parallels the mammalian
homologue, it would be expected to complement in the trans
comfiguration with respect to phhA. A trans-complementation
study was done in which phhA and phhB (or DCoH) were inserted
into two compatible plasmids, pJSll and pJZ9-4A (or pGST-
DCoH), respectively. The results (Table 3-2) did indeed show
that phhB was able to complement E. coli tyrosine auxotrophy
in trans with respect to phhA and that mammalian DCoH was able


46
PhhR-Pae F Q S S
TyrR-Bco V A V S
TyrR-Hin I V Q S
PilR-Pae L G B S
HydG-Bco V G K S
YfhA-Bco V T R S
VnfA-Avi I G N S
AnfA-Avi I G N S
NifA-Avi V G H T
PhlA-Bco I G R S
NtrC-Bco I G B A
AlgB-Pae B S H S
LuxO-Vha I G S S
PspF-Bco L G B A
R V H A A V [vl
P K M K H V V
B A M K S A |vj
P P M R A L R
PA8QHL L
P L H L R L L
K P M L B V Y
K P W Q B V Y
P T M R R V F
B A Y S V L
P A H Q D V P
P A M A A V L
QTQQ V Y
N S f L B V L
R BaIr R M A P L
B Q A Q K L A M L
b n[aJk R P A M F
NQIGKLARS
SBIALVAPS
EQARLVAQS
QLIERVVRT
BLIHKVAST
DQIRRVAKW
KQVBMVAQS
RIIGRLSRS
BTARQVAAT
RTIDSAASS
BQVSHLAPL
ATP-binding Motif A
A
P
L
L
*
E
G
E
T\
A
P
L
L
X
T
G
D
T
A
P
L
L
X
Q
G
E
T
A
P
V
Y
X
s
G
E
s
A
T
V
L
x
H
G
D
s
V
S
V
L
t
N
G
0
s
T
T
V
L

L
G
E
s
A
T
V
L
i
L
G
E
s
S
T
V
L
V
L
G
E
s
S
T
V
L

L
G
E
T
I
S
V
L
*
N
0
E
s
A
N
I
L
X
L
G
E
S
A
S
I
F
X
T
G
E
s
K
P
V
L
t
1
in
E
R
£ P R G Q S
B P R A G K
S L R R D K
SPRIER
B A R S E K
$ P R N S K
$ P A A K G
S P N A E A
5 P R A H R
SGRNNR
6 P R A K A
B K R A K K
5 K R G D K
3 S R W Q G
257
256
65
186
191
187
260
269
261
431
190
197
183
58
PhhR-Pae
TyrR-Bco
TyrR-Hin
PilR-Pae
HydG-Bco
YfhA-Bco
VnfA-Avi
AnfA-Avi
NifA-Avi
PhlA-Bco
NtrC-Bco
AlgB-Pae
LuxO-Vha
PspP-Bco
PhhR-Pae
TyrR-Bco
TyrR-Hin
PilR-Pae
HydG-Bco
YfhA-Bco
VnfA-Avi
AnfA-Avi
NifA-Avi
FhlA-Eco
NtrC-Bco
AlgB-Pae
LuxO-Vha
PspF-Bco
P
P
K
F m[a]l n C
Y L A L N C
F I [aJ V N e
P F V P V N C
P L V T L N
P f I A I N i
PFVKPN
A L V T S N t
P P V R L N C
R M V K M N C
P F I A L N i
PQ VTIH
P F I A I N G
P F I S L N ej
G V G B M S P
B I G B M S P
G I A B L S L
E V A D L P M
E I G D I S P
E I G D M P A
B V G B M S L
B V G E L S P
B I G B I S P
B V G D M P L
B I G D M P L
B I G D P P L
E L C B M D L
E L A T A P M
A G L
ASI
A G L
G A I
A A L
GAL
A S L
A P L
A A L
A A M
A A I
P S L
A A I
A A L
P E S M A
P B D A V
P D B D A
P S B L M
N B S L L
P B Q L L
P B S V I
P B N L A
P B T L L
P A G L L
P K D L I
T A B L M
P K D L I
N B N L L
* T 3 L
B t L
i a a m
|y G P gI A fFI BIG Al
H A
R K V G D
R P
B
G
K
L
G
L
L
L T
A
G
S
T
L
P
L D
307
. P
E
G
K
K
G
F
F
E
Q A
N
G
G
S
V
L
L D
298
S
E
T
I
G
F
u
Y ft
N
K
a
T
V
L
L
D
109
. I
B
D
K
Q
G
L
F
Q
A ft
S
G
0
T
L
F
L
D
235
. D
K
R
R
E
G
R
F
V
B A
D
G

T
L
F
L
D
240
. V
S
N
R
B
G
L
F
Q
A A
B
G
9
T
L
F
L
D
235
. I
G
L
R
K
G
R
F
E
E ft
A
G
0
T
I
F
L
D
309
. L
T
M
H
K
G
C
F
E
Q ft
D
G
G
T
I
F
L
D
318
. V
K
Q
R
K
G
R
F
E
Q D
G
G
T
L
F
L
D
310
. S
A
Q
R
I
G
R
F
B
B *
D
K
S
S
L
F
L
D
480
. N
T
I
R
Q
G
R
F
E
0 B
G
i
T
L
F
L
D
239
. T
E
S
T
L
G
R
V
S
Q ft
D
G
G
T
L
F
L
D
246
. A
N
D
R
Q
G
A
A
B
L ft
D
G
0
T
L
F
L
D
232
Q
K
R
H
P|[Gj
R
F
B
R ft
D
G
0
T
L
P
L D
107
ATP-binding Motif B
C F R R VMG S
T F R R V G B
S F R R V G E
A V R A V G G
B V Q R V G S
K V R P L G S
S F E R V G G
T F E R V G G
E F B R V G G
E F B R L GIIS
Q F Y R V G G
B Y B R V G D
T F Q K VlG|S
B L E R v||G||G
D
E
E
V
Y
L
D
V
R
V
I
C
A
X
0
V
D
h
s
B
L
c
A
357
D
H
E
V
H
V
D
V
R
V
I
C
A
T
Q
K
N
h
V
E
L
V
Q
348
E
K
E
H
Y
A
N
V
R
V
I
C
T
3
Q
V
P
b
H
B
B
V
E
159
Q
Q
E
V
A
V
D
L
R
I
L
C
i
H
K
D
L
A
A
E
V
G
285
N
Q
I
I
S
V
D
V
R
L
I
A
A
?
H
R
D
L
A
A
E
V
N
290
N
R
D
I
D
I
N
V
R
I
I
S
*
T
H
R
D
b
P
K
A
M
A
285
N
T
T
I
H
V
D
L
R
V
I
A
A
T
N
R
N
L
A
E
M
V
A
359
S
K
P
V
K
V
D
V
R
I
I
A
A
T
N
R
N
L
V
E
M
V
E
368
N
Q
T
V
R
V
N
V
R
I
V
A
A
T
N
R
D
L
E
S
E
V
E
360
N
K
I
I
Q
T
D
V
R
L
I
A
a
t
N
R
D
t
K
K
M
V
A
530
Y
A
P
V
K
V
D
V
R
I
I
A
A
T
H
0
N
h
E
Q
R
V
Q
289
P
V
T
R
R
A
D
V
R
I
L
A
A
T
N
R
D
L
G
A
M
V
A
296
S
K
M
K
S
V
D
V
R
F
V
C
A
?
N
R
D
P
W
K
E
V
Q
282
S
Q
P
L
Q
V
N
V
R
L
V
C
A
*
N
A
D
b
P
A
M
V
N
157
PhhR-Pae
TyrR-Bco
TyrR-Hin
PilR-Pae
HydG-Bco
YfhA-Bco
VnfA-Avi
AnfA-Avi
NifA-Avi
FhlA-Bco
NtrC-Bco
AlgB-Pae
LuxO-Vha
PspF-Eco
k s b tm Q D L Y
K 5 M *;!& B D L Y
Q G K V a A D L F
a j; r q d l y
A G R ISX Q D L Y
RGB B D L Y
D S T # A B D L Y
06T am B D L Y
K d K am E D L Y
D R B S D L Y
B Q K B D L P
Q G Q 'M B D L L
E G R B D L Y
B G T itm A |d l| l
h [FTIn V[l1s
Y R L N V L T
H R L N V[lJt
Y R L N V I E
Y R L N V V A
Y R L N V V S
Y R L N V F P
Y R L N f F P
Y R L N V M A
Y R L N V F P
H R L N V I R
Y R L N V I V
Y R L Y V I P
D R L A ft D V
L H
L N
I N
L R
I B
L K
I T
I T
I R
I H
V H
L N
L H
V Q
P [l] 3 E C L
P L R D C P
A L ft D R M
plSerr
S L ft Q R R
A L A D G T
P L A E R G
P L R E R G
P L R E R T
P L ft E R P
P L R E R R
P L ft E R A
P L ft E R G
P ij ft B R E
D O L A [p|l
Q D I M PL
A? I B |_p|l
E P I P L L
E P I P L L
E P I P L L
S I I T L
S d V I A L
A Q I P EL
E & I P L L
E p I P R L
E P I L G L
K P V I El
S P I M L M
A E H[F]L D Q A
T B LEV A R F
A Q G[f|L Q E I
Aerilkrl
Aghflqrf
Anhllrqa
ftDHPVSRF
ADHFVSAF
ftEFLLGKI
Akaptfki
Arhflqva
Aerflarf
Aysllgym
Aeyfaiqm
SRQIGCGL .
ADB-QGVPR .
SBBLKIAK .
AGDTGLPA A 335
AERNRKAV K 340
ADGHKPFV R 334
SREMGIEV N 409
SRENGKNV K 418
GRQQGRPL T 409
ARRLGRNI D 580
ARELGVBA K 339
VKDYGRPA R 345
SHEEGKSF.V 332
CREIKLPLFP 208
P 407
P 398
P 209
PhhR-Pae K L S
TyrR-Bco K L A
TyrR-Hin T F D
PilR-Pae R L T
HydG-Bco G F T
YfhA-Bco APS
VnfA-Avi R I S
AnfA-Avi R I S
NifA-Avi V T
FhlA-Bco SIP
NtrC-Eco L L H
AlgB-Pae G F S
LuxO-Vha R F A
PspF-Eco G F T
A Q A L B R L
A D L N T V L
K D F L L Y L
G D A Q B K L
P Q A M D L L
T D A M K R L
T P R L N M L
T P A L N M L
D S A I R L L
A E T L R T L
P B T E A A L
B A A R B A M
Q D V I B R F
E R A R E T L
E
R
Y
H
G
Hi
V
it
0
L
E
1
V
L
F
Q
A
V
s
L
C
E
G
G
442
% Identity
T
R
Y
a am
G
N
V
R
Q
L
K
N
A
I
Y
R
A
L
T
Q
L
D
G
Y
433
54%
Q
*
[yJ
d r*
G

v
R
E
L
Y
N
T
L
Y
R
A
C
S
L
V
Q
D
N
244
47%
K
N
Y
R F ft
G
N
V
G
B
L
E
N
M
L
E
R
A
Y
T
L
C
E
D
D
370
42%
I
H
Y
D P
G
N
I
R
E
L
E
N
A
V
E
R
A
V
V
L
L
T
G
B
375
43%
M
T
A
s w p
G
N
V
ft
Q
L
V
N
V
I
E
Q
C
V
A
L
T
s
s
P
260
45%
Q
S
Y
Q WP
G
N
V
ft
E
L
E
N
V
I
E
R
A
M
L
L
S
E
D
G
444
47%
M
s
Y
H
G
N
V
ft
E
L
E
N
V
M
E
R
A
V
I
L
S
D
D
D
452
45%
M
S
H
R W :P
G
N
V
S
E
L
E
N
c
L
E
R
s
A
I
M
S
E
D
G
444
48%
S
N
M
E M P
G
N
V
8
E
L
E
N
V
I
E
R
A
V
L
L
T
R
G
N
615
44%
T
R
L
a am
G
N
V
R
Q
L
E
N
T
C
R
W
L
T
V
M
A
A
G
Q
374
44%
R
Q
Y
p
G
N
V
R
E
L
R
N
V
I
E
R
A
S
I
I
C
N
Q
E
381
45%
N
s
Y
B P
G
N
V
R
Q
L
Q
N
V
L
R
N
I
V
V
L
N
N
G
K
367
45%
L
N
Y
E K P
G N
I
ft
E
L
K
N
V
V
E
R
S
V
Y
R
H
G
T
S
243
41%


41
(A)
Fructose
+Phe +Trp +Tyr
(B)
Glucose+Phe Fructose+Phe
ii
I
Z
£
O
z
Z
o
oc
<
<
H
£
e-
o
E
fifi
§
S'
z
w
w
c*
o
C/3
s
<
S
<
1
o
<
o
fH
C/3
O
*5
a.
ft-
0,
ft-
FIG. 2-10. Western blot analysis of phhA expression. (A)
Examination of aromatic amino acids as inducers of phhA
expression. P. aeruginosa PAO-1 was grown in minimal salts-
glucose or minimal salts-fructose medium with or without
addition of one of the three aromatic amino acids at a final
concentration of 100 fig/ml. (B) Phenylalanine induction of
phhA expression in different P. aeruginosa strains. Bacteria
were grown in minimal-glucose or minimal-fructose media
containing 100 g/ml phenylalanine.


94
Onishi, A., L. J. Liotta, and S. J. Benkovic. 1991.
Cloning and expression of Chromobacterium violaceum
phenylalanine hydroxylase in Escherichia coli and
comparison of aromatic amino acid sequence with mammalian
aromatic amino acid hydroxylases. J. Biol. Chem.
266 :18454-18459.
Patel, N., Stemark-Cox, S., and Jensen, R.A. 1978.
Enzymological basis of reluctant auxotrophy for
phenylalanine and tyrosine in Pseudomonas aeruginosa. J.
Biol. Chem. 253:2972-2978.
Pember, S. 0. J. Villafranca, and S. J. Benkovic. 1986.
Phenylalanine hydroxylase from Chromobacterium violaceum
is a copper-containing monooxygenase: Kinetics of the
reductive activation of the enzyme. Biochemistry
25:6611-6619 .
Pittard, A.J. 1996. Biosynthesis of the aromatic amino acids.
In Escherichia coli and Salmonella, pp 458-484. (F,C.
Neidhardt, ed.) ASM Publications. Washington, D. C.
Pittard, A.J., and Davidson, B.E. 1991. TyrR protein of
Escherichia coli and its role as repressor and activator.
Mol. Microbiol. 5:1585-1592.
Pogge-yon-Strandmann, E. and Ryffel, G. U. 1995.
Developmental expression of the maternal protein XDCoH,
the dimerization cofactor of the homeoprotein LFB1
(HNF1). Development. 121:1217-1226.
Pontoglio, M., Barra, J., Hadchouel, M., Doyen, A., Kress,
C., Bach, J. P., Babinet, C., and Yaniv, M. 1996.
Hepatocyte nuclear factor 1 inactivation results in
hepatic dysfunction, phenylketonuria, and renal Fanconi
syndrome. Cell 84:575-585.
Rosentel, J.K., Healy, F., Maupin-Furlow, J.A., Lee, J.H., and
Shanmugam, K.T. 1995. Molybdate and regulation of mod
(molybdate transport), fdhF, and hyc (formate
hydrogenlyase) operons in Escherichia coli. J.
Bacteriol. 177:4857-4864.
Sambrook, J., Fritsch, E. F., and Maniatis, T. 1989.
Molecular Cloning: A Laboratory Manual, 2nd ed. Cold
Spring Harbor, N.Y. : Cold Spring Harbor Laboratory Press.
Schweizer, H.P. 1993. Two plasmids, X1918 and Z1918, for easy
recovery of the xylE and lacZ reporter genes. Gene
134 : 89-91.


Eco-TyrR
P
E
D
A
V
E
S
E
L
F
G
H
A
. P
E
G
K
K
G
F
F
E
Q
A
N
G
G
S
V
L
L
D
Hin-TyrR
P
D
E
D
A
E
S
E
M
F
G
R
K
V
G
D
.
.
, .
S
E
T
I
G
F
F
E
Y
A
N
K
G
T
V
L
L
D
Pae-PhhR
P
E
S
M
A
E
T
E
L
F
G
Y
G
P
G
A
F
E
G
A
R P
E
G
K
L,
G
Li
L
E
L
T
A
G
G
T
L
F
L
D
Pae-PilR
P
S
E
L
M
E
S
E
F
F
G
H
K
K
G
S
F
T
G
A
. I
E
D
K
Q
G
L
F
Q
A
A
S
G
G
T
L
F
L
D
Eco-HydG
N
E
S
L
L
E
S
E
L
F
G
H
E
K
G
A
F
T
G
A
. D
K
R
R
E
G
R
F
V
E
A
D
G
G
T
L
F
L
D
Eco-YfhA
P
E
Q
L
L
E
s
E
L
F
G
H
A
R
G
A
F
T
G
A
. V
S
N
R
E
G
L
F
Q
A
A
E
G
G
T
L
F
L
D
Avi-VnfA
P
E
s
V
I
E
s
E
L
F
G
H
E
K
G
S
F
T
G
A
. I
G
L
R
K
G
R
F
E
E
A
A
G
G
T
I
F
L
D
Avi-AnfA
P
E
N
L
A
E
s
E
L
F
G
H
E
K
G
S
F
T
G
A
. L
T
M
H
K
G
C
F
E
Q
A
D
G
G
T
I
F
L
D
Avi-NifA
P
E
T
L
L
E
s
E
L
F
G
H
E
K
G
A
F
T
G
A
. V
K
Q
R
K
G
R
F
E
Q
A
D
G
G
T
L
F
L
D
Eco-FhlA
P
A
G
L
L
E
s
D
L
F
G
H
E
R
G
A
F
T
G
A
. S
A
Q
R
I
G
R
F
E
L
A
D
K
S
S
L
F
L
D
Eco-NtrC
P
K
D
L
I
E
s
E
L
F
G
H
E
K
G
A
F
T
G
A
. N
T
I
R
Q
G
R
F
E
Q
A
D
G
G
T
L
F
L
D
Pae-AlgB
T
A
E
L
M
E
s
E
L
F
G
H
S
R
G
A
F
T
G
A
. T
E
s
T
L
G
R
V
S
Q
A
D
G
G
T
L
F
L
D
Vha-LuxO
P
K
D
L
I
E
s
E
L
F
G
H
V
K
G
A
F
T
G
A
. A
N
D
R
Q
G
A
A
E
L
A
D
G
G
T
L
F
L
D
Eco-PspF
N
E
N
L
L
D
s
E
L
F
G
H
E
A
G
A
_F
T
G
A
Q
K
R
H
P
G
R
F
E
R
A
D
G
G
T
L
F
L
D
298
109
307
235
240
235
309
318
310
480
239
246
232
107
Rca-NtrC
L G A D G
P S S L L
A RRCGRLVV f[d]222
-J
FIG. 2-12. Alignment of the unique-gap region in the central domains of TyrR
proteins with selected homologs. Amino acid residues conserved in all of the 15
sequences that include both subfamily a (top cluster) and subfamily /3 (lower
cluster) are in shaded boxes. Amino acid residues conserved within the gap region
are shown in open boxes. Rea, Rhodobacter capsulatus; see the legend of Fig. 8 for
other abbreviations used.


93
Kaufman, S. 1987. Phenylalanine 4-monooxygenase from rat
liver. Meth. in Enzymol. 142:3-27.
Kaufman, S., Citron, B. A., Davis, M. and Milstien, S.
1993. The isolation and characterization of clones of 4a-
hydroxytetrahydrobiopterin dehydratase. Adv. Exp. Med.
Biol. 338:97-102.
Kaufman, S and Fisher, D. B. 1970. Purification and some
physical properties of phenylalanine hydroxylase from rat
liver. J. Biol. Chem. 245:4745-4750.
Kwok, T., Yang, J. Pittard, A.J., Wilson, T.J., and Davidson,
B.E. 1995. Analysis of an Escherichia coli mutant TyrR
protein with impaired capacity for tyrosine-mediated
repression, but still able to activate at a70 promoters.
Mol. Microbiol. 17:471-481.
Laemmli, U.K. 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature
(London) 227:680-685.
Lazarus, R. A., S. J. Benkovic and S. Kaufman. 1983.
Phenylalanine hydroxylase stimulation protein is a 4a-
carbinolamine dehydratase. J. Biol. Chem. 258:10960-
10962 .
Letendre, C. H. G. Dickens, and G. Guroff. 1974. The
tryptophan hydroxylase of Chromobacterium violaceum. J.
Biol. Chem. 249:7186-7191.
Letendre. C. H. G. Dickens, and G. Guroff. 1975.
Phenylalanine hydroxylase from Pseudomonas sp. (ATCC
11299a) : purification, molecular weight, and influence of
tyrosine metabolites on activation and hydroxylation. J.
Biol. Chem. 254:1829-1833.
Mendel, D. B. and Crabtree, G. R. 1991. HNF-1, a member of a
novel class of dimerizing homeodomain proteins. J. Biol.
Chem. 266:677-680.
Miller, J. H. 1972. Experiments in molecular genetics. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, N. Y.
Morrett, E., and Segovia, L. (1993) The bacterial enhances-
binding protein family: mechanism of action and
phylogenetic relationship of their functional domains. J.
Bacteriol. 175:6067-6074.
Nakata, H. Yamauchi, T. and Fujisawa, H. 1979. Phenylalanine
hydroxylase from Chromobacterium violaceum: purification
and characterization. J. Biol. Chem. 266:18454-18459.


23
analyzed by using the updated version of sequence analysis
software package offered by the Genetics Computer Group (GCG)
of University of Wisconsin (Devereux et al., 1984) .
Nucleotide Sequence Accession Number
The nucleotide sequence reported in this work has been
assigned Genbank accession number U62581.
Results
Evidence for a Flanking Regulatory Region
The original clone (pJZ9) isolated by Zhao et al. (1994)
produced markedly less phenylalanine hydroxylase activity than
did subclone pJZ9-3a (Fig. 2-2A). A possible explanation was
the presence of a negatively-acting regulatory gene in either
the upstream or downstream flanking region. Plasmid pJS60, in
correlation with its absence of upstream DNA but presence of
downstream DNA, expressed a very high level of activity. Thus,
the upstream region appeared to be responsible for decreased
expression of phhA in E. coli.
Transcriptional fusions were constructed using lacZ as a
reporter gene, as diagrammed in Fig. 2-2B. The results
indicate that the negative effect conferred by upstream DNA
occurs at the transcriptional level.


5
absence of 4a-carbinolamine dehydratase, the dehydration of
the 4a-carbinolamine becomes rate-limiting for the
hydroxylation of phenylalanine. The consequent accumulation
of 4a-carbinolamine results in a small percentage of
rearrangement to the 7-tetrahydrobiopterin isomer (Curtius et
al. 1990) The latter 7-isomer was shown to be a potent
inhibitor of the phenylalanine hydroxylase (Davis et al.,
1992) Under conditions where 4a-carbinolamine and the 7-
isomer are generated, the addition of 4a-carbinolamine
dehydratase markedly inhibits the rate of formation of the 7-
isomer by diverting a greater fraction of the 4a-carbinolamine
to the quinonoid dihydropterin (Davis et al. 1991) Thus, the
dehydratase not only directly catalyzes the dehydration of the
carbinolamine, but also indirectly prevents isomerization to
the inhibitory 7-isomer (Kaufman et al., 1993).
4a-Carbinolamine dehydratase from rat liver has been
cloned and sequenced (Citron et al. 1992). It then became
apparent that this dehydratase is identical to DCoH, a protein
that facilitates the dimerization of hepatic nuclear factor 1
alpha (HNF-la), a homeodomain transcription factor. DCoH was
found to display a restricted tissue distribution and did not
bind directly to DNA. The formation of a stable tetrameric
DCoH-HNF-la complex does not change the DNA-binding
characteristics of HNF-la, but does enhance the
transcriptional activity of HNF-la (Mendel et al. 1991). X-
ray crystallography has revealed DCoH to form a tetramer


16
were made as appropriate. Agar was added at 20 g/liter for
preparation of solid medium. Restriction enzymes, T4 DNA
ligase, DNA-modifying enzymes (New England Biolab or Promaga)
and Taq DNA polymerase (Perkin-Elmer) were used as recommended
by the suppliers. Other biochemicals were purchased from Sigma
Chemical Co. Inorganic chemicals (analytical grade) were from
Fisher Scientific.
Phenylalanine Hydroxylase Assay
Cultures of E. coli JP2255 carrying the various plasmids
specified were grown in 500 ml of LB broth supplemented with
ampicillin (100 xg/ml) at 37C and harvested at late
exponential phase of growth. The cell pellets were
resuspended in 10 ml of 10 mM potassium phosphate buffer (pH
7.4) containing 1 mM dithiothreitol and were disrupted by
sonication for 30 s at 4C using a Lab-Line Ultratip Labsonic
System (Lab-Line Instruments, Inc., Melrose Park, IL) The
resulting extracts were centrifuged at 150,000 x g for 1 hr at
4 C. The supernatant (crude extract) was desalted using
Sephadex G-25 and used for enzyme assay. PhhA was assayed by
following tyrosine formation (Nakata et al., 1979).
Recombinant DNA Techniques
Molecular cloning and DNA manipulation, including plasmid
purification, restriction enzyme digestion, ligation, and
transformation were conducted by standard methods (Sambrook et


44
amino-terminal domain of H. influenzae TyrR appears to be
absent. It is not known whether sequencing errors might
account for this, or whether the equivalent of the amino-
terminal domain might exist separately as a different protein.
A multiple alignment of the central-domain modules of
subfamilies a and (3 was shown in Fig. 2-11. In addition to the
many residues that are absolutely conserved throughout the
family, some residues which may prove to be uniquely conserved
within subfamily a are apparent, e.g., APLL corresponding to
residues 29-32 of Hin-TyrR.
Both Eco-TyrR and Rca-NtrC exhibit deletions in the
"unique-gap region" of the central domain (Fig. 2-12) in
correlation with their regulation of a70 promoters, rather than
a54 promoters. This observation led to the suggestion (Morrett
and Segovia, 1993) that this region of the central domain
might be critical for functional interfacing with a54. Since
this DNA segment of Pae-PhhR is intact with absolute retention
of highly conserved residues, the foregoing hypothesis is
consistent with the successful interaction of PhhR with a a54
promoter. Hin TyrR, on the other hand, is likely to be
deficient in interaction with a54 (like E. coli TyrR) owing
to a 6-residue deletion in this region.
Intervening Region of Divergent Transcription
Since the DNA-binding region of the carboxy terminus of
PhhR is identical at all important residues with E. coli TyrR,


Inactivation of phhR in P. aeruginosa abolished ability to
utilize either L-phenylalanine and L-tyrosine as a sole source
of carbon for growth.
PhhB is a bifunctional protein. It was shown to have 4a-
carbinolamine dehydratase activity as well as regulatory
activity. The expression of phhA was activated by the presence
of phhB in both E. coli and P. aeruginosa. Transcriptional
and translational fusion analysis showed that the regulatory
effect of PhhB on the expression of phhA is at the post-
transcriptional level.
An insertionally inactivated phhB mutant failed to grow
on L-phenylalanine or L-tyrosine as a sole carbon source.
Expression of PhhA in the absence of PhhB causes strong growth
inhibition in E. coli. The inhibitory effect is probably
caused by 7-tetrahydrobiopterin, which is known to be formed
in the absence of PhhB. Since 7-tetrahydrobiopterin is a
potent inhibitor of phenylalanine hydroxylase, this could
account for the inability of phhA in the absence of phhB to
complement E. coli tyrosine auxotrophy. The general inhibition
of growth may be due to inhibition of some unidentified
essential pterin-dependent enzymes.
vi 11


40
PAO-1 PA 103
WT phM. phhK WT rpoN *'
FIG. 2-9. Western blot analysis of phhA expression in mutant
derivatives of P. aeruginosa strains PAO-1 and PA103. The
proteins in crude extracts prepared from cultures grown in LB
medium were separated by SDS-PAGE, and equal amounts of
protein (50/xg) were applied to each lane.


12
4-Hydroxyphenylpyruvate
[PhhC]^
Tyrosine
[PhhA]
4a-Carbinolamine -5^5!I
+
-T etrahy drobi opterin
h2o
Dihydropteridine
[DHPR
Phenylalanine
NAD+ NADH+H+
FIG. 2-1. Initial reactions of phenylalanine catabolism in
mammals. The three structural genes of the phh operon encode
enzymes catalyzing three of the four steps shown. The
abbreviations: PhhA, phenylalanine hydroxylase; PhhB, 4a-
carbinolamine dehydratase; PhhC, aromatic aminotransferase;
DHPR, dihydropteridine reductase. 4a-Carbinolamine is an
alternative designation for 4a-hydroxytetrahydrobiopterin.


39
*Pae-PhhR
*Eco-TyrR
Hin-TyrR_
*Pae-PilR
*Eco-HydG
*Eco-YfhA
Avi-VnfA
Avi-AnfA
Avi-NifA
*Eco-FhlA
*Eco-NtrC
*Pae-AlgB
Vha-LuxO
*Eco-PspF_


68
to replace PhhB in the bacterial system. The results also
ruled out any possible cis effect of phhB on the expression of
phhA.
Expression of phhA in the Presence or Absence of phhB
The trans-complementation study confirmed that
complementation of E. coli tyrosine auxotrophy by phhA
requires the presence of phhB. To understand whether the
requirement of phhB for the complementation was due to
increased expression of phhA in the presence of PhhB, the
expression of phhA in the presence or absence of phhB in E.
coli JP2255 was studied through Western analysis (Fig. 3-3).
A substantial level of PhhA was still detected when only phhA
was present. The presence of either PhhB or DCoH in trans
indeed increased the expression of phhA, but only by about 2-
fold. Although this result indicated that PhhB may regulate
the expression of phhA, much more than the relatively modest
2-3 fold reduction in PhhA level observed in the phhB~ mutant
was expected. This was because little or no PhhA had been
detected in the absence of phhB (pJZ9-5) in E. coli, whereas
a very high level of PhhA was produced in the presence of phhB
(pJZ9-3a) (Zhao et al., 1994). The inconsistency between the
results of this study and that obtained by Zhao et al. (1994)
was found to be due to the incorrectly reported orientation of
the phhA insert in pJZ9-5. Sequencing the DNA insert in pJZ9-5
revealed that the orientation of the phhA gene was opposite to


ACKNOWLEDGMENTS
I wish to express my deep and sincere gratitude to Dr.
Roy A. Jensen, chairman of my supervisory committee, whose
invaluable guidance, constant encouragement, endless ideas,
critical input, and financial support made the fulfillment of
this study possible.
I would also like to thank Dr. Dean W. Gabriel, Dr.
Lonnie 0. Ingram, Dr. James F. Preston, and Dr. Keelnatham T.
Shanmugam for their help, encouragement, advice, and critical
review of the dissertation.
My special thanks are also extended to Dr. Carol Bonner,
Dr. Tianhui Xia, and Wei Gu for their great help in all
aspects of my study, particularly helping me get started
during my first year in the lab.
I am also very thankful to Dr. Randy Fischer, Dr. Prem
Subramaniam, and Gary Xie for their help during this study.
I am indebted to my family, especially to my parents, to
whom this dissertation is dedicated. Without their love,
support, and encouragement, this study could not have been
accomplished. I am also indebted to my brother and sister-in-
law for helping me in taking care of my parents.
Finally, but not least, I wish to express my sincere
appreciation to my wife, Tao Sun, for her love, support,
IV


59
modifying enzymes (New England Biolab or Promega), Taq DNA
polymerase (Perkin-Elmer), and Vent DNA polymerase (New
England Biolab) were used as recommended by the suppliers.
Phenylalanine Hydroxylase Assay
E. coli JP2255 (pJZ9-3a) was grown at 37C in 500 ml of LB
broth supplemented with ampicillin (100 /xg/ml) and harvested
at the late-exponential phase of growth. Cell pellets were
resuspended into 8 ml of 10 mM potassium phosphate buffer (pH
7.4) containing 1 mM dithiothreitol (DTT), and the cells were
disrupted by sonication. The resulting extract was
centrifuged at 150,000 x g for 1 hr at 4C. The supernatant
was desalted using Sephadex G-25 and used as crude extract for
enzyme assay. Phenylalanine hydroxylase (PhhA) was assayed by
following tyrosine formation (Nakata et al., 1979).
Phenylalanine Hydroxylase Stimulation Assay
4a-Carbinolamine dehydratase activity in E. coli(pJZ9-4)
was assayed indirectly using the phenylalanine hydroxylase
stimulation assay (Citron et al. 1992). Reaction mixtures
containing 30 mM potassium phosphate (pH 8.3), catalase (1
mg/ml), 100 /xM NADH, 1 mM phenylalanine, 20 /xg
dihydropteridine reductase, 14.4 /xg rat liver phenylalanine
hydroxylase, and 2.9 /xM 6,7-dimethyltetrahydropterin were
incubated at 25C. Approximately 1 min after the reaction was
started, either buffer (control) or 15 /xg of the E. coli


43
homologous central domain, but the amino-terminal and carboxy-
terminal domains may vary considerably within the family.
Thus, this exemplifies a complex multi-domain protein family
in which family membership is defined by a common ancestral
central domain. Future subdivisions within what is termed
subfamily ¡3 in Fig. 2-8 could likely be defined on the
criterion of homology for the remaining two domains. For
example, Eco-NtrC and Eco-FhlA belong to different mechanistic
subgroups: the two-component regulatory system and direct
response-to-small-molecules, respectively (reviewed by
Shingler, 1996).
Figure 2-8 highlights the emerging homology relationships
of selected paralog and ortholog proteins, with respect to the
central domain. E. coli possesses at least six paralogs, some
of which diverged in a common ancestor that existed prior to
speciation events which generated orthologs. Thus, the
divergence of Eco-NtrC and Pae-PilR was a more recent event
than was the divergence of Eco-NtrC and Eco-PspF. In contrast
to the ancient duplication events which generated all of the
E. coli paralogs (or the P. aeruginosa paralogs) are the
relatively recent duplication events generating the three
paralogs which regulate three distinctly separate nitrogenase
systems in Azotobacter (Joerger et al., 1989).
P. aeruginosa PhhR and E. coli TyrR exhibit homology in
all three domains: 36% identity, amino-terminal; 52% identity,
central; and 47% identity, carboxy-terminal). Curiously, the


57
Table 3-
-1. (continued)
pETllb
T7lac promoter, lacI+ Apr Novagen
pET23
Tllac promoter, lad+ Apr Novagen
pGEM-3Z
T7 promoter, Apr Promega
pGST-DCoH In-frame protein fusion of
glutathione S-transferase and DCoH Citron et al.
pJSlO
phhAB, 2.5-kb Hindi fragment This study-
cloned into pGEM-3Z behind
the T7 promoter
pJSll
phhAB', 1.44-kb Hindi-EcoRV This study
fragment cloned into pACYC177
pJS12
phhAB, 2.5-kb Hindi fragment This study
cloned into pACYC177
pJS51
Hindi-BauMI fragment containing This study
truncated phhA' cloned into
pACYC177
pJS51Z
phhA'-lacZ transcriptional fusion This study
in pACYC177
pJS63
phh'ABC, BamHI-Hindi! fragment This study
cloned into pGEM-3Z behind
the T7 promoter
p JS72
phhA, PCR-generated fragment This study
containing the native ribosome
-binding site and PhhA-coding
region cloned into pET23 behind
T7lac promoter
pJS95
PhhA overexpression vector; This study
PhhA-coding region fused with T7
translational initiation signal
at Ndel site of pETllb
p JS96
PhhA overexpression vector; This study
phhA fused with T7 translational
initiation signal cloned into pUC19
behind lac promoter to constitutively
overexpress PhhA


50
Therefore, this region may possess intrinsic DNA-bending
capabilities.
Function of the phh Qperon
The primary function of the phh operon is clearly not to
accommodate tyrosine biosynthesis since the feedback-inhibited
cyclohexadienyl dehydrogenase which is widely distributed in
gram-negative bacteria exists for this purpose. However, the
phh operon probably provides a fortuitous backup capability
for tyrosine biosynthesis. "Reluctant auxotrophy" for tyrosine
(Patel et al., 1978) can be explained as follows. Mutational
deficiency of cyclohexadienyl dehydrogenase would lead to
accumulation of prephenate, a potent product inhibitor of
chorismate mutase. The subsequent backup of chorismate,
enhanced by lack of early-pathway control in the absence of L-
tyrosine, results in passage of chorismate to the periplasm
where chorismate mutase-F (Gu and Jensen, unpublished data)
and cyclohexadienyl dehydratase (Zhao et al., 1993) generate
L-phenylalanine. Subsequent induction of phenylalanine
hydroxylase completes the alternative circuit to L-tyrosine.
The established function of phenylalanine hydroxylase in
mammals is for catabolism of L-phenylalanine as a carbon
source. We have found that phenylalanine hydroxylase is indeed
essential for use of L-phenylalanine as a sole carbon source
in P. aeruginosa. Thus, inactivation of phhA resulted in
inability to use L-phenylalanine as a sole source of carbon


73
quite protracted and requires 90 minutes or more reach a
maximum. A basal level of PhhB was expressed under non
inducing conditions as seen in lane 2 on the Western blot. It
is unknown whether this is due to expression from a promoter
upstream of phhA or from an internal promoter within the
coding sequence of phhA. Although the latter interpretation
comes to mind because no PhhA band is visualized at zero time,
this could be due to differing sensitivities of the antibody-
probes .
Effect of phhB Knockout in P. aerucrinosa
The phhB gene in P. aeruginosa was inactivated by
chromosomal insertion of the suicide plasmid pUFR/'phhB'/Hgr
through a single homologous crossover event (Fig. 3-5A). The
insertional inactivation of the phhB gene in the resulting
mutant was confirmed by Southern blot analysis (Fig. 3-5B).
The expression of PhhA in a phhB~ background was examined
by Western-blot methodology under fully induced conditions
fostered by growth on LB media (Fig. 3-6) The PhhA level
observed in the phhB~ mutant was reduced about 2-3 fold
compared to the wildtype parent. No PhhA was detected in
phhA~ mutant (negative control) The PhhB level was also
checked at the same time. A low basal level of PhhB was
detected in phhA~ mutant when compared with that in the
wildtype. No PhhB was detected in phhB~ mutant (negative
control).


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
l-xft-M -N jir
Roy A. Jipsen, lichair
Professor of Microbiolgy and
Cell Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree^of Doctor of Philosophy.
Dean W. Gabriel
Professor of Plant Pathology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Lonnie 0. Ingram
Professor of Microbiolgy and
Cell Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degrees of Doctor of .philosophy.
imes F. Preston
'Professor of Microbiolgy and
Cell Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Keelnatham T. Shanmugam
Professor of Microbiolgy and
Cell Science


63
exponential-phase cells were collected by centrifugation, and
the cell pellets were suspended in gel-loading buffer and
heated at 100C for 10 min. Samples of 5-10 /x 1 were loaded
onto two SDS-polyacrylamide gels. After separation of the
proteins by electrophoresis, one gel was stained with
Coomassie blue and the other gel was used for blotting. When
crude extracts were used, equivalent amounts of protein were
loaded into each lane. Western blots were performed according
to Towbin et al. (1979). The proteins were
electrophoretically transferred onto nitrocellulose membranes
and reacted with the polyclonal antiserum at a dilution of
1:1000. Membranes were then incubated with secondary alkaline
phosphatase-labelled anti-rabbit antibodies at a dilution
(1:30,000) and developed by adding NBT and BICP as chromogenic
substrates (Bibco BRL) for alkaline phosphatase.
Gene Inactivation
Both phhB and phhC were inactivated following the method
described by Song and Jensen (1996) To generate the
truncated 'phhB' fragment (308bp), the upper primer 5'-
ACCCAAGCCCATTGCGAAGCCTGCCG-3', and the lower primer 5'-
GTGCGCGCCGCCATGATGAAATCGTT-3' were used. To generate the
truncated 'phhC' fragment (652bp), the upper primer 5'-
GTCGAGCAGGAAACCACCAAGA-3', and the lower primer 5'-
GTTGGCTACGCAGGTCGGTGAG-3' were used. Interruption of the phhB


FIG. 3-5. Inactivation of phhB in P. aeruginosa. (A) Schematic
representation for insertional inactivation of the chromosomal
phhB gene by the integration of the suicide plasmid
pUFR/'phhB'/Hgr through a single homologous crossover. The
resulting Hgr mutant does not contain a complete copy of phhB
gene, but instead has two truncated copies of phhB gene. (B)
Southern-blot analysis of chromosomal DNA from P. aeruginosa
PAO -1 wildtype (lane 1) and mutant JS103 (lane 2) Chromosomal
DNA was completely digested with EcoRI and probed at high
stringency with the truncated 'phhB' fragment of
pUFR/' phhB /Hgr.


70
the lac promoter, rather than in the same orientation as
reported by Zhao, et al. (1994) .
PhhB Regulates Expression at the Post-transcriotional Level
To understand the regulatory role of PhhB in the
expression of phhA, I constructed phhA'-lacZ transcriptional
fusions in E. coli in both multi-copy form (pJS51Z) and a
single-copy form [X[phhA'-lacZ)] In both cases, the presence
of PhhB (pJZ9-4) or DCoH (pGST-DCoH) on a second plasmid did
not result in a higher level of (S-galactosidase as compared to
the control (pUC18) (Table 3-3), indicating that neither PhhB
nor DCoH functions at the transcriptional level. I then
constructed the translational fusion phhA'-'lacZ (pJS105Z)
(Table 3-3), and PhhB or DCoH was again provided in trans on
a second plasmid. (S-Galactosidase activity increased about
two-fold, indicating that PhhB regulates the expression of
phhA at the translational or post-transcriptional level. This
level of activation by PhhB is consistent with the results
obtained from Western-blot analysis in E. coli, where
similarly modest levels of activation in phhA expression were
observed.
Induction of The vhh Ooeron by Phenylalanine in P. aeruginosa
Expression of both phhA and phhB were coordinately
induced by phenylalanine in P. aeruginosa when grown on
minimal fructose medium (Fig. 3-4). The induction process is


85
Although phhA is in fact expressed in the absence of phhB,
phhB does appear to exercise a positive regulatory role in the
expression of phhA with a relatively modest effect of perhaps
2-3 fold. The transcriptional fusion approach seems to have
eliminated the possibility of this regulation being at the
transcriptional level, with the reservation that the
experiments were performed in an E. coli background. I have
shown that PhhR, the positive regulator of the phh operon in
P. aeruginosa, does not interact properly with the E. coli a54
machinery to activate phhA expression. But it is still
possible that PhhB interacts with PhhR as a co-activator
entity in the native P. aeruginosa organism. This would amount
to a parallel with the mammalian system where DCoH is a co
activator for HNFlo' (an upstream enhancer element that can be
considered comparable to PhhR).
However, I have presented results to show that PhhB does
enhance expression of PhhA at the post -transcriptonal level in
E. coli. The 2-3 fold enhancement effects obtained correspond
with the similar 2-3 fold magnitude of effect seen in P.
aeruginosa, when comparing Western blots of PhhA in PhhB+ and
PhhB- backgrounds. If PhhB regulates at a post-transcriptional
level, several possibilities envisioned include: (i) the
sencondary structure of phhA mRNA might mask the ribosomal
binding site. A stem-loop structure has been located in this
area. Binding of PhhB in this region might disrupt the
secondary mRNA structure and enhance translational initiation.


13
mammalian metabolism, the literature encompassing the widely
studied catabolism of aromatic compounds in pseudomonad
bacteria (indeed, in prokaryotes) does not include the
phenylalanine hydroxylase step. Furthermore, L-phenylalanine
(substrate of PhhA) is an extremely poor source of carbon for
growth of P. aeruginosa, whereas L-tyrosine (product of PhhA)
is an excellent carbon source.
Zhao et al. (1994) had previously noted that subclones
lacking the flanking regions around the phh operon possessed
20-fold greater activity for phenylalanine hydroxylase. This
suggested the presence of a regulatory gene. Since an
understanding of the regulation governing the phh operon
should provide important physiological clues about function,
I have analyzed the flanking regions and now report the
characteristics of a regulatory gene, denoted phhR.
Materials and Methods
Materials
The bacterial strains and plasmids used in this study are
listed in Table 1. The LB and M9 formulations (Sambrook et
al. 1989) were used as growth media for E. coli and P.
aeruginosa. Pseudomonas isolation agar (Difco) was used for
isolating Pseudomonas "knockout" mutants. Additions of
ampicillin (100 /xg/ml) chloramphenicol (40 /xg/ml) kanamycin
(50 /xg/ml), tetracycline (25 /xg/ml), mercuric chloride (15
/xg/ml) L-phenylalanine (50 /xg/ml) and thiamine (17 /xg/ml)


19
plasmid Z1918) fused at the BarnU I site within phhA to form
phhA'-lacZ transcriptional fusions. Plasmid pJS61Z has the
same upstream sequence as plasmid pJZ9, and plasmid pJS62Z has
the same upstream sequence as pJZ9-3a. Hence, the phhA'-lacZ
fusions in pJS61Z and pJS62Z should represent the phhA
transcriptional levels in pJZ9 and pJZ9-3a, respectively.
To study regulation of the phhR promoter, the HincII-
BanMI fragment (phhR') was cloned into the pRS1274 lacZY
transcriptional fusion vector at the BarriRI Smal site to create
pJS102(phhR'-lacZ).
3-Galactosidase Assay
/?-Galactosidase activity was assayed under conditions of
proportionality as described by Miller (1972), and specific
activities are expressed in Miller units. The data are the
results of at least two independent assays.
Gene Inactivation
P. aeruginosa is well known for its relatively high
resistance to most antibiotics, which complicates attempts to
use most of available antibiotic-resistance genes as selective
markers for gene replacement. Mercury resistance (Hgr) was
used as a selective marker since P. aeruginosa has been shown
to be sensitive to mercury (Essar et al., 1990; Gambill and
Summers, 1985). Insertional inactivation technique described
by Sophien et al. (1992) utilizes a mobilizable suicide vector


49
BajnH I
CCTAGGCGAGCACCCCGTCCGGCTCGACAACGTACGGCAGCTCCATAAGGACTGTCCGCGCCGGAAGCTAGTGGAAGTCAACGGCCCACTAGTCCCATAAGGTCTGGACCACGAGCCAAA 120
PIREHPLGLQEIGDLYEQCARGEIVKLQRTILTNWVQHET
GGCCTATCACCTACTTTGGCAACAGCAGCCCGACCGCCCGGTGCATGACGCAGCAAAAGTATGCCTOAOGTACTCGCCGAACACCGGCGCCCCAGAAACAACAACAGCAACGGAATAGCT
BPYHI FGNDDPQRAVYQTTKM RBS- U
phhA 4 1 -12
240
PhhR Box 2
TACCGCGTCGGGACGGTCCGTGCAGCCCCGGATAGGGACCGGGCAACGGGGAGGAATCGGCGTTTT CGTAAAGTTTTCCTTACG AATTGGCCTGGGTCGCCTGTTCATTGGGTCAGGCAT
ATGGCGCAGCCCTGCCAGGCACGTCGGGGCCTATCCCTGGCCCGTTGCCCCTCCTTAGCCGCAAAA GCATTTCAAAAGGAATGC TTAACCGGACCCAGCGGACAAGTAACCCAGTCCGTA
TGTTGCTGGTGAGTCTAAC '
ACAACGACCACTCAGATTG .
1 GAAATTTCCGGCCGGGAGTTAAAAAACCGGCCGCGAGCCCATCAGAACGACXACACCGGGCCACGCCATGCGTATCAAAGTGC 4 80
j CTTTAAAGGCCGGCCCTO^TTCTTGOCCGkS33GTt?35tW3TCPKK3tCTTQ3GGGOQ6nQGGG!rO(3
Terminator ?
CQNRVGILRDILNLLVDYGINVNRGEVGGDQGNAIYLLCP
ACTGCCAGAACCGTGTAGGCATCCTCCGCGACATCCTCAACCTGCTGGTCGACTACGGCATCAACGTCAATCGCGGCGAAGTCGGCGGCGACCAGGGCAACGCCATCTACCTGCTCTGTC
Hindi
FIG. 2-13. The intervening sequence between the divergently
transcribed phhA and phhR genes. The number at the end of each
line indicates the nucleotide position. The ribosome binding
sites and putative promoter sites (-12/-24 promoter for phhA,
and -10/-35 promoter for phhR) are indicated. The
translational start sites are indicated by arrows. Two PhhR
boxes are identified. A stem-loop structure is shaded.
Restriction endonuclease recognition sites are marked.


78
(A)
pJS72 [P0
RBS
PhhA
C A T|G GAGlTCCGTATG AAAACG...
Native translational signal MKT
RBS
PhhA
pJS95 lP0||GAAGGAGAlTATACATATG AAAACG
T7 translational signal M K T
1 2 3 4 5
FIG. 3-7. Expression of PhhA using T7 expression system in E.
coli. (A) Construction of PhhA expression plasmids, pJS72
(with the native translational signals) and pJS95 (with 010
translational signals). (B) SDS-PAGE analysis of expressed
PhhA protein. Proteins in whole cell lysate were separated by
SDS-PAGE and stained with Coomassie blue. Lane 1, molecular
weight markers; lanes 2&3, BL21(DE3) harboring pJS72, before
and after 1 mM IPTG induction for 3 h at 30C, respectively;
lanes 4&5, BL21(DE3) harboring pJS95, before and after 1 mM
IPTG induction for 3 h at 30C, respectively.


31
Overproduction of PhhR
PhhR protein was overexpressed in E. coli BL21(DE3) as
detailed under Materials and methods by use of the T7
overexpression system; the construct is illustrated in Fig.2-
6. The initial use of overexpression vectors containing phhR
on the BairiRl-Sphl fragment (see Fig. 2-4) of pJZ9 failed. This
is probably due to autogenous regulation of phhR, judging from
the precedent set by tyrR in E. coli (Argaet et al., 1994).
Accordingly, overexpression was achieved through excision of
DNA upstream of phhR. PCR methodology was used to generate an
intact phhR gene which was fused with the T7 translational
start codon at a Nde I restriction site to create
overexpression plasmid, pJS88. E. coli BL21(DE3) that had been
transformed with pJS88 was induced with 1 mM IPTG for 3 hours
to express PhhR. Whole-cell lysates obtained before and after
IPTG induction were analyzed by SDS-PAGE, as shown in Fig. 2-
6B. Overproduction of a 56-kDa protein was observed, and N-
terminal amino acid sequencing confirmed its synonymy with
PhhR.
Initial attempts to express phhR in E. coli under
physiological conditions indicated that expression of phhR is
highly toxic. The Bglll-EcoRl fragment from pJS88 was cloned
into the BamHI-EcoRI site of pUC19 behind a lac promoter. When
transformed into E. coli DH5q;, transformants achieved only
pinpoint colony size and eventual survivor cells inevitably
had lost the plasmid. Success was finally achieved by use of


33
pACYCl84, a low copy-number plasmid, to create pJS91 which
carried the Bglll-BamHI fragment of pJS88 ligated into the
BamHI site of pACYC184 (Fig. 2-6A). Analysis of 11 plasmids
isolated showed that the orientation of phhR in each case was
opposite to that of the Tcr gene. Presumably, the higher level
of expression expected when driven by the Tcr promoter still
confers an intolerable level of toxicity.
Functional Replacement of E. coli tvrR with phhR
A simple test was used to see whether phhR could
substitute for tyrR as a repressor of the aroF-tyrA operon.
Mutants deficient in TyrR exhibit resistance to m-
fluorotyrosine (Fig. 2-7, middle) whereas tyrR* strains
exhibit sensitivity to growth inhibitory effects of the analog
(Fig. 2-7, left) pJS91 {phhR*) was used to transform an E.
coli tyrR-deficient background (strain SP1313). The ability of
PhhR to replace TyrR is qualitatively apparent (Fig. 2-7,
right) by inspection of the halo of growth inhibition on a
bacterial lawn surrounding a disc containing m-fluorotyrosine
in SP1313 (tyrR' phhR*).
We also examined the ability of PhhR to replace TyrR as
an activator of mtr, encoding a component of a tryptophan-
specific transport system. The phhR* plasmid pJS91 was
transformed into two E. coli X lysogens (Heatwole and
Somerville, 1991) which carried mtr'-lacZ transcriptional
fusions integrated in the chromosome as single-copy fusions.


86
(ii) PhhB might bind phhA mRNA in the 3' region and protect
against nuclease-catalyzed mRNA degradation, (iii) PhhB may
complex with PhhA. This complex may protect PhhA from
proteolysis. I have in fact obtained preliminary evidence for
a PhhA-PhhB complex.
Rationale for Positive Regulatory Role of PhhB
When PhhA is highly expressed in the absence of PhhB, the
E. coli host cells become subject to drastic growth
inhibition. The reasonable explanation for this effect of
PhhB is that removal of the inhibitory effect generated by
PhhA reaction is direct result of 4a-carbinolamine
dehydratase activity. It is known that in the absence of 4a-
carbinolamine dehydratase activity, a 7-isomer of
tetrahydrobiopterin is generated during the reaction catalyzed
by phenylalanine hydroxylase (Davis et al. 1992) and this 7-
isomer is a potent inhibitor of phenylalanine hydrxoylase in
the mammalian sytem. Since potent growth inhibition persists
in wildtype E. coli backgrounds where overexpressed
phenylalanine hydroxylase has no purpose, inhibition cannot be
attributed to inhibition of phenylalanine hydroxylase by the
7-isomer. Two explanations accounting for general growth
inhibition come to mind, (i) The reduced pterin cofactor of E.
coli may be depleted because PhhB is not present for
recycling. Consequently, some pterin-dependent enzymes that
are essential for growth may become limiting. (ii)


FIG. 2-5. Pairwise alignment (GAP program of GCG) of amino acid sequences corresponding
to E. coli TyrR and P. aeruginosa PhhR. The similarity is 66.3% and identity is 45.7%.
The three functional domains are indicated at the right, the central domain also being
shaded. Domain boundaries are based upon those formulated by Morrett and Segovia (1993) .
Alernative boundaries based upon domain segments surviving partial hydrolysis by trypsin
(Cui and Somerville, 1993) are residues 1-190, 191-467, and 468-513 for the three
domains of Eco-TyrR, respectively. In the N-terminal domain, the region between amino
acids 2 and 19 that have a critical role in activation of the expression of E. coli tyrP
and mtr (Pittard, 1996) is double-underlined, and mutations at residues marked in
boldface type abolish TyrR-mediated activation without affecting repression; a second
region between amino acids 92 and 103 which may play a subsidiary role in activation is
also double-underlined; mutations at the residues in boldface type resulted in loss of
function (Pittard, 1996). In the central domain, two ATP-binding sites and a leucine-
zipper motif are underlined. Mutations altering ATP-binding site A and mutations at the
highly conserved residues E-274, G-285, and E-302 abolish TyrR-mediated repression. In
the C-terminal domain, a helix-turn-helix DNA-binding motif is identified with the helix
regions underlined and critical residues in bold print.


8
hydroxylases or 4a-carbinolamine dehydratase have ever been
detected (Vasudevan et al. 1988). Unlike other
dihydropteridine reductases that have been studied, the E.
coli DHPR possesses an FAD prosthetic group, and has
dihydrofolate reductase and pterin-independent oxidoreductase
activities (Vasudevan et al. 1992) .
Regulation of Phenylalanine Hydroxylase
Phenylalanine hydroxylase in mammals is tightly regulated
at different levels. At the protein level, it is
allosterically regulated by phenylalanine (Kaufman, 1987).
The activity of phenylalanine hydroxylase increases at least
20-fold after incubation with phenylalanine (Tourian, 1971).
It is also activated through phosphorylation by a cAMP-
dependent kinase both in vivo and in vitro (Abita et al,
1976) At the DNA level, expression of the phenylalanine
hydroxylase gene in liver and kidney tissues of mice is
enhanced at birth and is induced by glucocorticoids and cAMP
in liver (Faust et al. 1996) Regulatory elements including
a tissue-specific and hormone-inducible enhancer in the
upstream region have been characterized. The enhancer region
contains separate protein-binding sites for the glucocorticoid
receptor and the hepatocyte-enriched transcription factor,
hepatocyte nuclear factor 1 (HNF1) (Faust et al., 1996) HNF1
is a transcriptional activator of many hepatic genes including
albumin, a-antitrypsin, and a- or /3-fibrinogen. Mice lacking


64
or phhC gene in a Hgr isolate was confirmed by Southern
hybridization.
Southern Hybridization
Genomic DNA was extracted from the P. aeruginosa phhB'
strain by the method described by Silhavy et al. (1984).
Southern hybridization was performed as described by Sambrook
et al. (1989). The DNA was completely digested with EcoRI,
separated by electrophoresis in 1% agarose gel, and
transferred to a nylon membrane (Bio-Rad). The DNA was fixed
by baking the membrane under vacuum at 80C for 2 hr and
hybridized at 42C overnight with the truncated 'phhB' (the
same as used for gene inactivation) probes that had been
labeled with biotin-14-dATP using a BioNick labelling system
(GIBCO/BRL). The membrane was washed in 2X SSC (300 mM NaCl,
30 mM sodium citrate, pH 7.0) plus 0.1% SDS (twice for 3 min
each time at room temperature), in 0.2X SSC plus 0.1% SDS
(twice for 3 min each time at room temperature), and in 0.16X
SSC plus 0.1% SDS (twice for 15 min each time at 50C). The
probes were detected with the BluGene nonradioactive nucleic
acid detection system (GIBCO/BRL).
RESULTS
PhhB Has 4a-Carbinolamine Dehydratase Activity
PhhB is a homologue of an established 4a-carbinolamine
dehydratase. To confirm that PhhB catalyzes the 4a-


PhhA
FIG. 3-3. Western blot analysis of PhhA expression in E. coli
JP2255. Proteins in the whole cell lysates of JP2255 carrying
various plasmids were separated by SDS-PAGE and reacted with
rabbit anti-PhhA polyclonal antibodies. Plasmids containing
the gene(s) shown above are as follows: phhA, pJSll; phhB,
pJZ9-4; DCoH, pGST-DCoH. pUC18 was used as the control
plasmid.


30
conservation throughout the entire length of PhhR and TyrR,
and 45.7% of the deduced residues were identical. The N-
terminal domain mediates regulatory modulations, and in TyrR
it binds all three aromatic amino acids. A central domain,
highly conserved throughout the entire family of a54 enhancer
binding proteins, exhibits two established motifs that reflect
the binding of ATP (Pittard, 1996). Site A corresponds to the
ATP-binding pocket motif and site B corresponds to segment 3
of adenylate kinase. In this region a perfect leucine-zipper
motif is apparent in P. aeruginosa PhhR, whereas E. coli
displays an imperfect motif. Residues E-274, G-285, and E-302
were found to be important for TyrR-mediated repression of
aroF-tyrA in E. coli (Yang et al., 1993; Kwok et al., 1995),
and these residues are all conserved in P. aeruginosa PhhR.
The C-terminal domain possesses a helix-turn-helix motif which
is responsible for DNA binding. The absolute conservation of
residues shown to be critical in E. coli (Pittard, 1996)
strongly indicates that PhhR and TyrR might target to a
similar DNA sequences.
Similar to E. coli TyrR, the two aspartate residues and
the lysine residue conserved in the amino-terminal domain of
all response regulator proteins (Stock et al., 1989) were not
found.


81
FIG. 3-8. Expression of PhhB using T7 expression system in E.
coli. Proteins in whole cell lysates were separated by SDS-
PAGE and stained with Coomassie blue. Lane 1, molecular weight
markers; Lanes 2&3, BL21(DE3)/pJSlO, before and after 1 mM
IPTG induction for 3 h at 30C, respectively. Lanes 4&5,
BL21(DE3)/pJS63, before and after 1 mM IPTG induction for 3 h
at 30C, respectively.


PhhA
PhhB
FIG. 3-6. Western blot analysis of PhhA and PhhB expression
in P. aeruginosa PAO strains. Proteins in whole cell lysates
were separated by SDS-PAGE and probed with rabbit anti-PhhA or
anti-PhhB polyclonal antibodies.


This dissertation was submitted to the Graduate Faculty
of the College of Agriculture and to the Graduate School and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
May, 1997
, ^College of ^gr
Dean
griculture
Dean, Graduate School


26
-35 -10 IBS phhX
TGTTOCTOGTGAGTCTAACTGTCACATATTCTTGACGGAAATTTCCGGCCGGGAGTTAAAAAACCGGCCGCGAGCCCATCAGJUlCaiLCiACACCGGGCCACGCCATGCGTATCAAAGTGC 120
M R I K V H 6
Hindi
ACTGCCAGAACCGTGTAGGCATCCTCCX3CGACATCCTCAACCTGCTGGTCGACTACGGCATCAACGTCAATCGCGGCGAAGTCGGCGGCGACCAGGGCAACGCCATCTACCTGCTCTGTC 240
CQNRVGILRDILNLLVDYGINVNRGBVGGDQGNAIYLLCP 46
NMINLQLQSLRPKLBAVPGVPGVKR
GLMPSBRRHLBLNA
HindI
CXJCrGCTCGCCGCCCTGGACTTCCCGGTGCTCTCGGTGGACATGGGCGGGCAGATCGTCGCCGCCAACCGTGCCGCCGCGCAGTTGCTCGGCGTGCGCGTCGACGAGGTGCCGGGGATTC 480
LLAALDPPVLSVDMGGQIVAANRAAAQLLGVRVDBVPGIP 126
BHDBSBALAGA
LTLHRADRVGBRIYH
RKQBLRGFDS 206
RVMAAVVRBARRMAPLDAPLLIBGBTGTGKBLLAR 246
GCPRRVGSDBBVYLDVRVICATQ
ATCACCGTCTCAATGTGCTCTCGCT3CACATCCCGCCGCTGCGCGAGTGCCTGGATGGCCTGGCGCCGCTCGCCGAGCATTTCCTCGACCAGGCCAGCCGGCAGATCGGCTGCGGCCTX3C 1320
HRLNVLSLHIPPLRBCLDGLAPLAEHPLDQASRQIGCGLP 406
1440
446
CCGAGCACATCCGCCTGCCGGATTACGGCGCGCCGCAGCCXjCTGGGGGATnTTCCCTGGAAGGAGACTCGACGCATCGTCGGGCGCTTCGAGAAGGGGTGCTGGAGCGCCTGTTCCGCG 1560
BHIRLPDYGAPQPLGDPSLBGDSTHRRALRBGVLBRLPRB 486
AACATCCGAGCACCCGCCAGTTGGGCAAGCGCCTCGGCGTTTCGCATACCACGGCGGCGAACAAGCTGCGCCAGCATC
HPSTRQLGK
H T T A A N
LRQHGVGQSBG
1680
518
GTTGCAGGCTGGGAAAGAAACGGTGGAGGCGTCGGCCCTTGTGAAAGGGCGGCAGCACGATGAGGCGTGGGGAAGACGGCCAGGATGTTCGCCGAGGGGCGGGATTGTCGCCGCCGAGTA 1800
FIG. 2-3. Nucleotide sequence of the phhR region. The numbers
at the right indicate nucleotide and amino acid positions. The
putative promoter region and ribosome-binding site (RBS) are
indicated with bold print. RBS bases that are complementary to
P. aeruginosa 16S rRNA are overlined. The translational start
site is indicated by a bent arrow and the stop codon by an
asterisk. Nucleotides forming the complementary stems of the
putative transcriptional terminator are marked with tandem
arrowheads. Restriction endonuclease recognition sites are
marked above the nucleotide sequence.


patience, and encouragement during these years of study, and
to my son, Peter, and my daughter, Kerry, for filling the
family with great joy and happiness.
v


20
containing a truncated gene fragment (at both 5' and 3' ends)
and Hgr-cassette, and this suicide plasmid was integrated into
the chromosome by a single homologous recombination event. PCR
was used to generate truncated fragments. To generate a 'phhR'
(6 0lbp) fragment, the upper primer 5'-
CCGTGTAGGCATCCTCCGCGACAT-3', and the lower primer 5'-
CTGGAAGATACTGTCGAAGCCACG-3' were used; to generate the 'phhA'
(639bp) fragment, we used the upper primer 5'-
ACGACAACGGTTTCATCCACTATC-3' and the lower primer 5'-
GGACGAAATAGAGCGGTTGCAGGA-3'. The PCR-generated fragments were
cloned into pCRII (a TA cloning vector) and then excised with
EcoRI. The EcoRI fragments were subsequently cloned into the
EcoRI site of pUFR004 (a mobilizable suicide vector) to create
pUFR/'phhA', and Hgr HindiII-cassette from pJSlOl was inserted
into the Hindlll site of pUFR/'phhA' to create
pUFR/'phhA'/Hgr. pUFR/'phhR'/Hgr was created in a similar
fashion. These plasmids were then used to transform E. coli
strain S17-1 (a mobilizing strain). Strain S17-1 harboring
either pUFR/ 'phhA' /Hgr or pUFR/ 'phhR' /Hgr was used as the donor
in biparental mating with P. aeruginosa performed as described
by Simon et al. (1983) Donor and recipient cells were grown
in LB broth to an OD600 of about 1.0 (E. coli S17-1 at 37C and
P. aeruginosa PAO-1 at 42C) mixed (0.5 ml volume of each) in
a 1.5-ml microcentrifuge tube, and pelleted by centrifugation.
The mating mixture was carefully resuspended in 0.2 ml of LB
broth and spread onto a sterile nitrocellulose filter (0.45-^tm


84
PhhA was produced from pJS97 due to the presence of laclq gene
one the plasmid, the cells carrying pJS97 were better able to
maitain the plasmid, thus being able to overproduce PhhA after
IPTG induction (Lane 4, Fig. 3-9B). Neither pJS96 nor pJS97 by
itself was able to complement E. coli tyrosine auxotrophy.
However, they were able to complement the auxotrophy when phhB
was provided in trans on pJZ9-4 (data not shown). This result
indicates that PhhB was able to remove the inhibitory effect
imposed by overproduction of PhhA on the host cells. To assure
that apparent growth inhibition was not due to excessive
conversion of phenylalanine to tyrosine in the phenylalanine
auxotrophy background of strain JP2255, the plasmids were
moved to a prototrophic background. E. coli DH5a carrying
either plasmid was found to develop only pinpoint colonies on
LB + Amp plates without IPTG induction, confirming that
expression of phhA at higher levels created a general
inhibitory effect on the growth of E. coli.
Discussion
Regulatory Role of PhhB?
The mammalian PhhB homolog, DCoH, has both catalytic and
regulatory functions. It was initially thought that P.
aeruginosa phhB exerted an essential positive role in
expression of phhA since constructs lacking phhB did not
express phhA, as monitored by SDS-PAGE (Zhao et al 1994).


56
Table 3-1. Bacterial strains, plasmids, and phages used in
this study
Strain or
plasmid
Relevant genotype
or description
Source or
reference
E. coli
BL21(DE3)
F" ompT hsdSB (rB~mB~) gal dcm;
with DE3, a X prophage carrying
the T7 RNA polymerase gene
Novagen
DH50!
F'AlacU169 4>80dlacZAM15 hsdR17
recAl endAl gyrA96 thi-1 relAl
supE44
GIBCO/BRL
LE392
F~el4~ (McrA~) hsdR514 (rkmk)
supE44 supF58 lacYl or a (lacIZY) 6
galK galT22 metBl trpR55
Sambrook et al
JP2255
aroF363 pheA361 phe0352 tyrA382
thi-1 strR712 lacYl xyl-15
Baldwin &
Davidson
JS1
SP1313§(phhA'-lacZ)
This study
S17-1
[RP4-2(Tc:Mu)(Km:Tn7)Tra(incP)]
pro hsdR recA Tpr Smr
Simon et al.
SP1313
zah-735: TnlOA (argF-lac) U1691A (tyrR)
Heatwole &
Somerville
P. aeruginosa
PAO -1
Prototroph
Holloway
JS101
PAO-1 phhA, Hgr
Song & Jensen
JS102
PAO-1 phhR, Hgr
Song & Jensen
JS103
PAO-1 phhB, Hgr
This study
JS104
PAO-1 phhC, Hgr
Plasmids
pUC18
pUC19
pACYC177
Ampr lac'IPOZ'
Ampr lac'IPOZ'
P15A replicn, Apr Kmr
Yanisch-
Perron et al.
Yanisch-
Perron et al.
Chang & Cohen


87
Alternatively, the 7-isomer may be a potent inhibitor of a
pterin-dependent enzyme needed for growth. Possible targets of
inhibition could be dihydropteridine reductase or
dihydrofolate reductase. The E. coli dihydropteridine
reductase has been reported to possess broad specificity for
pteridine compounds (Vasudevan et al. 1992).
When phhA was expressed at relatively low levels in E.
coli no growth inhibition, or at least no severe growth
inhibition occured. However, this PhhA was evidently not
functional in vivo ( in the absence of PhhB) because
complementation of tyrosine auxotrophy in the presence of
exogenous L-phenylalanine was unsuccessful. On the other hand,
the joint presence of PhhA and PhhB readily allowed functional
complementation. I concluded that PhhA is an essential target
of the 7-isomer. At the low levels of PhhA expression in the
absence of PhhB, the 7-isomer is generated and inhibits PhhA
function--but not enough 7-isomer is produced to cause general
growth inhibiton. At high levels of PhhA expression in the
absence of PhhB, sufficient 7-isomer is produced to inhibit
one or more enzymes essential for the growth.
With the above background, a rationale to explain a basis
for selection of regulation of PhhA by PhhB is apparent. If it
is correct that the 7-isomer generated from the carbinolamine
pterin product of the PhhA reaction has general antimetabolite
properties, then the significance of PhhB goes beyond its


48
it is likely that PhhR binds to the same binding sites for E.
coli, which are referred as "TyrR boxes" (consensus:
TGTAAAN6TTTACA) This conclusion is also supported by the
ability of PhhR to replace TyrR as a repressor of the aroF-
tyrA transcriptional unit. The location of two "PhhR boxes"
which match the consensus for "TyrR boxes" was shown in Fig.
2-13. PhhR Box 1 is a strong box (with more conserved-symmetry
and higher affinity for TyrR) that overlaps the putative -10
region of the phhR promoter. TyrR boxes in E. coli occur in
tandem with variable spacing (Pittard, 1996), and a TyrR
hexameric molecule is thought to bind both a strong box and a
weak box with DNA looping in between. PhhR Box 2 is a weak box
located in the middle of the intervening region. It seems
probable that by analogy with autorepression of tyrR in E.
coli, both phhR boxes participate in the autogenous repression
of phhR by PhhR, probably with tyrosine as a corepressor.
In the opposite direction of transcription, the a54
promoter for phhABC requires an upstream activator site (UAS) .
PhhR Box 1 may be the most likely UAS, although perhaps both
boxes participate in activation of phhA. L-Phenylalanine and
L-tyrosine, potent inducers of phenylalanine hydroxylase,
presumably are the effector molecules. Since a rpoN mutant
retained low basal level of PhhA, another promoter that is
independent of a54 might be present.
No motif for binding of integration host factor (IHF)
(Friedman, 1988) was located in the intervening region.


35
Strain SP1312 (tyrR+) exhibited the expected elevation of
S-galactosidase activity following growth in the presence of
tyrosine, phenylalanine, or both. However, strain SP1313
(tyrR~) carrying pJS91 (phhR+) produced the control level of
S-galactosidase activity, regardless of the presence or
absence of aromatic amino acids (data not shown). Thus, PhhR
appears to be incapable of replacing TyrR as an activator of
E. coli mtr.
Autogenous Regulation of vhhR
The BamHI-Hind I fragment containing the 5' coding
regions of phhA and phhR and the intervening region (see Fig.
2-4) was fused to lacZ to give the reporter-gene construct
pJS102 (phhR' lacZ) This plasmid construct was introduced
into the tyrR-negative background of strain SP1313 in the
presence or absence of pACYC184 possessing a phhR+ insert. The
results (Table 2-2) demonstrated a repressive effect of phhR*
upon PhhR levels as monitored by measurement of /3-
galactosidase activity. Since the copy number of pJS91 (the
source of PhhR molecules) in this experiment is lower than the
number of repressor target sites provided by the high-copy
number pJS102 and since TyrR boxes are present within seven
other transcriptional units of E. coli, auto-regulation is
undoubtedly grossly under-estimated due to titration of
available PhhR molecules in the system.


52
these enzymes, on the criterion of assessment of specific
activities determined in comparison of tyrR* and t yrR~
backgrounds (data not shown).


36
Table 2-2. Autoregulation of P. aeruginosa phhR in E. coli
SP1313(tyrR') containing pJS102{phhR1 -lacZ)
Second
/3-Galactosidase
levelsb in cells
crown in:
plasmid9
M9C
M9 + F M9
+ Y
pACYC184
550
510
589
pACYC184 (phhR*)
362
376
372
a pACYC184 {phhR*)
is denoted pJS91
in Table 1.
b (S-Galactosidase levels are reported in Miller Units.
c M9 minimal medium was supplemented with 1 mM thiamine-HCl
and, where indicated, 1 mM phenylalanine (F), or 1 mM
tyrosine (Y).


CHAPTER 1
LITERATURE REVIEW
Phenylalanine Hydroxylase in Nature
Phenylalanine hydroxylase (phenylalanine hydroxylase 4-
monooxygenase; EC 1.14.16.1) catalyzes the irreversible
conversion of L-phenylalanine to L-tyrosine (Kaufman, 1987).
In mammals this enzyme catalyzes the initial, obligatory, and
rate-limiting step in the complete catabolism of serum
phenylalanine to C02 and H20 (Kaufman, 1986) A deficiency of
this enzyme causes accumulation of serum phenylalanine,
leading to hyperphenylalanemia. Because metabolism of
phenylalanine is restricted to alterations in the alanyl side
chain of phenylalanine, in the absence of phenylalanine
hydroxylase, the formation and excretion in the urine of
compounds such as phenylpyruvate and phenyllactate occurs.
This condition is called phenylketonuria, a genetic disorder
associated with severe mental retardation in untreated
children (Dilella et al. 1986) Many mutations at the
phenylalanine hydroxylase locus have been identified (Guldberg
et al., 1996).
Phenylalanine hydroxylase has been intensively studied in
mammals for many years. It is a member of a family of enzymes
that also includes tryptophan hydroxylases (EC 1.14.16.4) and
1


62
phhA-coding region together with the upstream T7 translational
start signals were excised from pJS95 as a XJbal fragment and
cloned into pUC18 downstream of a lac promoter to create
pJS96. The Xbal fragment was also cloned into pTrc99A
downstream of the inducible trc promoter to create pJS97.
Two similar plasmids, pJSlO and pJS63, were constructed
to express PhhB. The Hindi fragment containing both phhA and
phhB gene was inserted into pGEM-3Z to create pJSlO, and the
BawHl-Hindlll fragment containing both phhB and phhC was
inserted into pGEM-3Z to create pJS63. The phhB gene was
under the control of a T7 promoter in both plasmids.
Preparation of PhhB-specific Polyclonal Antiserum
PhhB was partially purified by anion-exchange and gel-
filtration chromatography following the methods described by
Zhao et al., (1994). The partially purified PhhB was subject
to SDS-PAGE (12%) and the gel was stained with Commassie blue
R-250. The PhhB band was cut from the gel and used for the
production of polyclonal antiserum in rabbits (Cocalico
Biologicals, Inc., Reamstown, PA) Antiserum was purified by
using an Econo-Pac protein A column (Bio Rad) and further
absorbed with a total cell extract from the PhhB-deficient
mutant JS103.
SDS-PAGE and Western Blot Analysis
SDS-PAGE (12% gel) was performed with the Mini-PROTEAN II
cell (Bio-Rad) by the method of Laemmli (1970). Samples of


80
Table 3-4. Phenylalanine hydroxylase activities in different
expression clones3
Expression
clones
Specific Activity
(nanomoles/min/mg)
JP2255/pJZ9-3a
91.7
BL21(DE3)/pJS72
72.6
BL21(DE3)/pJS95
379.2
3 Cells of E. coli JP2255 harboring pJZ9-3a were grown in LB
broth at 37C and harvested at late exponential phase; cells
of BL21(DE3) harboring pJS72 or pJS95 were grown in LB broth
at 37C to 0.D=1 and induced for 3 hr by addition of 1 mM
IPTG. Crude extracts were used as the enzyme sources.


CHAPTER 2
PHHR, A DIVERGENTLY TRANSCRIBED ACTIVATOR OF
THE PHENYLALANINE HYDROXYLASE GENE CLUSTER
OF Pseudomonas aeruginosa
Introduction
A recent report (Zhao et al., 1994) revealed that
Pseudomonas aeruginosa possesses a tetrahydrobiopterin (BH4) -
dependent monooxygenase that is capable of catalyzing the
phenylalanine hydroxylase reaction. It is encoded by the
proximal member (phhA) of a three-gene cluster. The second
gene, phhB, encodes carbinolamine dehydratase, a key enzyme
within the cycle regenerating BH4. phhC encodes an aromatic
aminotransferase and belongs to the Family-I aminotransferases
(Gu and Jensen, 1996) The reactions, as they are known to
function for the mammalian homologs in the catabolism of L-
phenylalanine, are shown in Fig. 2-1.
The physiological function of phenylalanine hydroxylase
in P. aeruginosa has not been obvious. A primary role in L-
tyrosine biosynthesis seems unlikely because of the
established presence for this purpose of a cyclohexadienyl
dehydrogenase that is widely distributed in gram-negative
bacteria and which is highly sensitive to feedback inhibition
by L-tyrosine (Xia and Jensen, 1990). Although function as an
initial step of L-phenylalanine catabolism has precedent in
11


Copyright 1997
by
Jian Song


58
Table 3-1. (continued)
pJS97
PhhA overexpression vector;
phhA fused with T7 translational
signal cloned into pTrc99A behind
trc promoter
This study
pJSlOl
Hgr-cassette, Apr
Song & Jensen
pJS105
Hindi-Ba/nHI PCR fragment
containing phhA' with a frameshift
This study
pJS105Z
phhA'-'lacZ protein fusion cloned
into pACYC177
This study
pJZ9
phhRABC, Apr
Zhao et al.
pJZ9-3a
phhAB, Apr
Zhao et al.
pJZ9-4
phh'ABC', Apr
Zhao et al.
pJZ9- 5
phhAB', Apr
Zhao et al.
pMC1871
lacZ protein fusion vector
Pharmacia
pTrc99A
Trc promoter, lad* Apr
Pharmacia
pUFRO 04
ColEl replicn, Cmr Mob+ mobP,
lacZa*
DeFeyter et
al.
Z1918
Promoterless lacZ, Apr
Schweizer
Phages
XRZ5
X'hla 'lacZ lacY*
Resental et
al.
XJS1
X$ {phhA'-lacZ) lacY* 'hla
This study


22
heated at 100C for 10 min. Samples of 5-10 /xl were loaded
onto two SDS-acrylamide gels. After separation of the proteins
by electrophoresis, one gel was stained with Coomassie blue R-
250 and the other gel was used for blotting. When crude
extracts were used, equivalent amounts of protein were loaded
in each lane. Western blots were performed according to
Towbin et al. (1979) The proteins were eletrophoretically
transferred onto nitrocellulose membranes and reacted with
polyclonal antibodies raised against PhhA in a rabbit.
N-Terminal Amino Acid Sequencing
PhhR protein produced in E. coli BL21(DE3)/pJS88
following induction by 1 mM IPTG for 3 hours was first
separated from the whole lysate by SDS-PAGE. The proteins
were then blotted to a polyvinylidene difluoride membrane
(Bio-Rad) and were stained with Coomassie brilliant blue R-250
(Sigma). The band corresponding to PhhR was excised from the
membrane and used for sequencing by using an Applied
Biosystems model 407A protein sequencer with an on-line 120A
phenylthiohydantoin analyzer in the Protein Core Facility of
the ICBR at the University of Florida.
DNA Sequencing and Data Analysis
Sequencing of phhR region was performed by the DNA
Sequencing Core Laboratory of the University of Florida. The
nucleotide sequence and the deduced amino acid sequence were


24
(A)
M M
M M H H H M
HHH HHH > H HM
3 a
1 1
a a i s*
1 1 r-J
S .-5
1 Lrf-w
I 3 3
II i 1 i
PhhA activity
phhA phhB
phhC
(nmol/min/mg)
1 kb
pJZ9 2.7
pJZ9-3a 94.0
pJS7 3.5
pJS60 69.7
(B)
pJS61Z 1 tacW
pJS62Z 1 lacZ
P-Galactosidase
activity
(Miller Units)
17
1250
FIG. 2-2. Localization of a regulatory region upstream of the
phh operon. (A) On the right phenylalanine hydroxylase (PhhA)
activities are shown in E. coli JP2255 harboring different
plasmids shown on the left. (B) On the right /6-galactosidase
activities are shown in BW545 harboring the phhA'-lacZ
transcriptional fusions diagrammed on the left.


71
Table 3-3. Levels of phhA expression by phhB and DCoH in
transa.
phhB or DCoH
in trans
(S-Galactosidase Activityb
pJS51Z
X {phhA' -lacZ)
PJS105Z
pJZ9 -4 (PhhB+)
180
13.2
13.1
pGST-DCoH (DCoH+)
191
14.1
10.4
pUC18 (control)
182
15.7
7.3
a Regulation of phhA expression was studied using lacZ as the
reporter gene and /3-galactosidase activities in
transcriptional fusions pJS51Z (phhA'-lacZ) (multicopy) and
X{phhA'-lacZ) (single copy), and translational fusion
pJS105Z {phhA'lacZ) (multicopy) was assayed in absence
(pUC18) or presence of PhhB (pJZ9-4) and DCoH (pGST-DCoH).
b /?-Galactosidase activities are reported in Miller units.


14
TABLE 2-1
. Bacterial strains and plasmids
Strain or
Relevant genotype or
Source or
plasmid
description
reference
E. coli
BL21(DE3)
F- ompT hsdSB (rBmB) gal dcm;
with DE3, a X prophage carrying
the T7 RNA polymerase gene
Novagen
BW545
A (lacU) 169 rpsL
Rosentel et al
DH5a
F~AlacU169 4>80dlacZAM15 hsdR17
recAl endAl gyrA96 thi-1 relAl
supE44
GIBCO/BRL
LE392
F_el4~ (McrA~) hsdR514 (rk~mk+)
supE44 supF58 lacYl or a (lacIZY) 6
galK galT22 metBl trpR55
Sambrook et al
JP2255
aroF363 pheA361 phe0352 tyrA382
thi-1 strR712 lacYl xyl-15
Baldwin &
Davidson
S17-1
[RP4-2(Tc:Mu)(Km:Tn7)Tra(incP)]
pro hsdR recA Tpr Smr
Simon et al.
SP1312
zah-735:Tnl0 A(argF-lac) U169
Heatwole &
Somerville
SP1312
(XSLW20)
SP1312 0 {mtr -lacZ+)
Heatwole &
Somerville
SP1313
SP1312, A(tyrR)
Heatwole &
Somerville
SP1313 SP1313 0 (mtr -lacZ*)
(XSLW20)
P. aeruginosa
Heatwole &
Somerville,
PA103
Prototroph
Totten et al.
PA103NG
rpoN
Totten et al.
PAO -1
Prototroph
Holloway
JS101
PAO-1 phhA, Hgr
This study
JS102
PAO-1 phhR, Hgr
This study


7
Dihvdropteridine Reductase
Dihydropteridine reductase (DHPR; EC 1.6.99.7) is one of
the two essential enzymes involved in recycling the pterin
cofactor for aromatic amino acid hydroxylases. It catalyzes
the reduction of quinonoid dihydropterin to
tetrahydrobiopterin, using NADH as a cofactor. DHPR is an
ubiquitous enzyme in animals, being found in all tissues that
contain the aromatic amino acid hydroxylases (Armarego et al. ,
1984). Close correlation between levels of 4a-carbinolamine
dehydratase and dihydropterine reductase in liver during human
fetal development strongly suggests a physiologically
significant role for both enzymes in tetrahydrobiopterin
regeneration. Genetic defects in DHPR cause malignant
phenylketonuria. A concomitant deficiency of
neurotransmitters such as 3,4-dihydroxyphenylalanine (DOPA)
and 5-hydroxytryptophan reflects the essential coupling of
DHPR to tyrosine hydroxylase and tryptophan hydroxylase as
well (Gudinchet et al., 1992) .
DHPR is also found in bacteria. DHPR has been purified
from Pseudomonas acidovorans (Williams et al. 1976) and E.
coli (Vasudevan et al., 1988). In P. acidovorans, both DHPR
and phenylalanine hydroxylase activities were found to be
higher in cells adapted to a medium containing L-phenylalanine
or L-tyrosine as the sole carbon source than in those grown in
L-asparagine (Williams et al. 1976). Interestingly, DHPR has
also been found in E. coli even though no aromatic amino acid


2
tyrosine hydroxylases (EC 1.14.16.2). All three enzymes
utilize a tetrahydrobiopterin cofactor and molecular oxygen to
hydroxylate their respective aromatic amino acid substrates
(Kaufman and Fisher, 1974). Phenylalanine hydroxylase has
been purified from rat liver where it is an oligomeric protein
(predominantly homotetramers) composed of 52-kDa subunits
(Davis et al. 1996) It has non-heme iron as the active-site
metal. The rat liver hydroxylase was also expressed in E.
coli and purified to homogeneity (Kappock et al., 1995) The
homotetrameric recombinant rat hepatic phenylalanine
hydroxylase is highly active and is identical to the native
enzyme in many properties.
Although mammalian phenylalanine hydroxylase has been
intensively studied, few studies on bacterial phenylalanine
hydroxylase have been done. Phenylalanine hydroxylase has
generally been considered to be of rare occurrence in
prokaryotes, where scattered reports of its existence have
been limited to one phylogenetic division of gram-negative
bacteria. They include Pseudomonas acidovorans (previously
known as Pseudomonas sp. ATCC 11299a) (Guroff & Ito, 1963) .
P. facilis (Decicco & Umbreit, 1964), Alcaligenes eutrophus
(Friedrich & Schlegel, 1972), and Chromobacterium violaceum
(Letendre et al., 1974). Of the three pterin-dependent and
metal-containing hydroxylases, only phenylalanine hydroxylase
from Pseudomonas acidovorans (Letendre et al, 1975) and C.
violaceum (Nakata et al., 1979; Pember et al., 1986) has been


Dedicated to my father and my mother,
whose love, care, and encouragement make it possible
for me to complete this dissertation


25
Identification of phhR
A large open reading frame (Fig. 2-3) located upstream of
the phh structural genes appeared likely to be functional on
the criterion of GCG codon preference analysis. The gene,
denoted phhR, produces a deduced protein having 518 residues,
an anhydrous molecular weight of 56,855, and an isoelectric
point of 7.17. It contains a single tryptophan residue.
Regions corresponding to a possible a70 promoter region
and a factor-independent transcription terminator are marked
in Fig.2-3. A strong ribosome-binding site was not apparent.
Bases that are complementary to P. aeruginosa 16S rRNA at the
3' terminus are marked. Perhaps the "A-richness" of the
initiator region enhances ribosome binding (Ivey-Hoyle and
Steege, 1992).
A physical map is given in Fig. 2-4 of the 5874-bp DNA
segment containing the structural genes of the phh operon, the
divergently transcribed regulatory gene phhR, and a gene
(pbpG) downstream of the phh operon which encodes a
penicillin-binding protein (Song and Jensen, unpublished
data).
Homology of PhhR with E. coli TvrR
The closest homolog of PhhR was found to be E. coli TyrR.
The pairwise GAP alignment (GCG) is shown in Fig. 2-5. TyrR
belongs to a family of modular proteins which usually have
three functional domains. The alignment showed high level of