Molecular characterization of a bifunctional enzyme (dehydroquinate dehydratase/shikimate dehydrogenase) and the post-pr...


Material Information

Molecular characterization of a bifunctional enzyme (dehydroquinate dehydratase/shikimate dehydrogenase) and the post-prephenate pathway enzymes needed for phenylalanine and tyrosine biosynthesis in Nicotiana SPP
Physical Description:
xiii, 157 leaves : ill., photos ; 29 cm.
Bonner, Carol Ann
Publication Date:


Subjects / Keywords:
Nicotiana -- Chemotaxonomy   ( lcsh )
Botanical chemistry   ( lcsh )
Microbiology and Cell Science thesis Ph.D
Dissertations, Academic -- Microbiology and Cell Science -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1994.
Includes bibliographical references (leaves 149-156).
Statement of Responsibility:
by Carol Ann Bonner.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001981577
oclc - 31920213
notis - AKF8508
System ID:

Full Text








Many thanks to Roy A. Jensen for the inspiration and

guidance throughout my studies for this dissertation. I

thank, also, my committee members, Dr. Richard P. Boyce from

the Department of Biochemistry and Molecular Biology; and

from the Department of Microbiology and Cell Science, Dr.

Lonnie O. Ingram, Dr. Robert R. Schmidt and Dr. Keelnatham

T. Shanmugam for their support and encouragement in

completion of this dissertation. Special thanks to all the

members of our lab who have been a great source of help to

me during these studies. Firstly, to Randy Fischer and

Premila Rao, my most grateful thanks for their help and

encouragement in so many ways from the very beginning of my

studies. Thanks to Astride Rodrigues, Jackie Miller,

Michelle Miller for all the work in the tissue culture lab.

Most special thanks to Tianhui Xia, Genshi Zhao, Prem

Subramaniam, Wei Gu and Jain Song for all the discussion and

work we have done together. I would also like to thank the

other members of the microbiology and cell science

department who have helped me with their knowledge and

advice, especially Philip Miller, for all the information

and insight that he gave me with respect to the molecular

biology portion of my dissertation.




ACKNOWLEDGEMENTS........................................ ii

TABLE OF CONTENTS....................................... iii

LIST OF FIGURES......................................... v

LIST OF TABLES........................................ .. viii

ABBREVIATIONS ........................................... ix

ABSTRACT ................................................. xi



Aromatic Amino Acid Biosynthesis in Plants.. 1
Rational for Focus upon DQT and SDH.......... 8
Shikimate biosynthesis and quinate
catabolism...................... ........... 9
Multiple Species of Shikimate Dehydrogenase. 12
The Post-Prephenate Pathway in Plants....... 16

II MATERIALS AND METHODS.......................... 19

Organisms and Extract Preparation............. 19
Enzyme Assays................................. 23
Separation of Pathway Enzymes by
Chromatography.............................. 27
Purification of Pathway Proteins............. 29
Polyacrylamide Gel Electrophoresis........... 33
Kinetic Values............................... 34
Inhibition studies........................... 35
Antibodies .............. .................... 35
cDNA cloning and sequencing.................. 37
Biochemicals................................. 43

ENZYMES IN PLANTS............................. 45

Results..................................... 45
Discussion.................................. 59

AMINO ACID PATHWAY........................... 62

Results .................................... 62
Discussion.................................. 53

V PROPERTIES OF THE S-PROTEINS................... 76

Results .................................... 76
Discussion.................................. .104

PROTEINS... ................................. 108

Results..................................... 108
Discussion.................................. .116

S-PROTEIN, ADH AND PAT....................... 118

Results..................................... 118
Discussion.................................. .132

SUMMARY.... ........... .................................. 148

REFERENCES.... .......................................... 149

BIOGRAPHICAL SKETCH ..................................... 157


Figure Pane

1-1 Aromatic amino acid biosynthetic pathway.......... 3

1-2 Multiple compartmentation of pathways in plant
cells ............................................. 7

1-3 Quinate catabolism and aromatic amino acid
biosynthesis...................................... 11

3-1 DEAE-cellulose chromatography of aromatic pathway
enzymes from N. silvestris....................... 48

3-2 DEAE-Cellulose chromatography of aromatic pathway
enzymes from C. sorokiniana...................... 51

3-3 Temperature optimum for activity of prephenate
aminotransferase in Chlorella sorokiniana........ 52

3-4 Growth curve of N. silvestris in suspension
cultures......................................... 56

3-5 Arogenate dehydratase activities followed
throughout a growth curve of N. silvestris....... 58

4-1 Elution profiles of SP-I and SP-II from Celite
545 column chromatography........................ 64

4-2 Elution profiles of SP-I and SP-II from DEAE-
cellulose chromatography......................... 67

4-3 Elution profiles of S-proteins from an NADP'
specific affinity column.......................... 70

4-4 Celite 545 column chromatography overlap of
arogenate dehydrogenase and SP-II................ 72

4-5 Hydroxylapatite column chromatography separation
of arogenate dehydrogenase and SP-II............. 73

5-1 Elution profile of SDH activities from a Celite
545 column run with a shallow gradient............ 78

5-2 Molecular weight determinations by SDS PAGE and
by silver stain gels of the two S-proteins....... 81

5-3 Gel filtration chromatography of S-proteins...... 83

5-4 Optimal PAGE activity staining for SDH in crude
extract .......................................... 86

5-5 PAGE activity staining of purified S1- and S2-
proteins......................................... 88

5-6 S-proteins pH optima of optima for activity...... 90

5-7 Temperature optima for S-proteins................ 93

5-8 Thermal inactivation of S-proteins............... 95

5-9 Substrate protection against thermal
inactivation of S-proteins....................... 97

5-10 Saturation curves and double reciprocal plots
for SP-I ......................................... 100

5-11 Saturation curves and double reciprocal plots
for SP-II ........................................ 102

6-1 Precipitation of S-proteins with SP-I specific
antibody on Ouchterlony plates .................. 110

6-2 Effect of SP-I specific antibody on SP-I and
SP-II activities ................................ 113

6-3 Comparison of Western blot and activity stained
gels............................................ 115

7-1 Agarose gels of cDNA encoding aromatic pathway
proteins......................................... 121

7-2 Preliminary alignment of translated sequences of
clone SP3 with established DQT sequences......... 124

7-3 Sequencing strategy for cDNA encoding the
bifunctional protein ............................. 126

7-4 Agarose gels of SP3 cDNA fragments during
stages of subcloning.............................. 128

7-5 Nucleotide and deduced amino acid sequence of
cDNA coding for the bifunctional S-protein....... 131

7-6 Multiple amino acid alignment of the AroD
domain of AroD*E with its homologues.............. 135

7-7 Dendrogram of dehydroquinase homologues........... 137

7-8 Multiple amino acid alignment of the shikimate
dehydrogenase domain of AroD*E and its
homologues.... ................................... 141

7-9 Dendrogram for shikimate dehydrogenase
homologues...................................... 143



Table Pare

1-1 Molecular mass and Km values for DQT and SDH....... 15

3-1 Specific activities from N. silvestris and C.
sorokiniana aromatic pathway enzymes.............. 46

3-2 Prephenate aminotransferase amino acid donor
specificity in C. sorokiniana .................... 53

3-3 Prephenate aminotransferase activity after
thermal treatment in C. sorokiniana.............. 53

4-1 Purification of SP-I from N. silvestris
suspension cultured cells........................ 71

4-2 Purification of SP-II from N. silvestris
suspension cultured cells........................ 71

5-1 Kinetic parameters for SP-I and SP-II............. 103

5-2 Effect of PCMB on enzyme activity................ 104

6-1 S-protein antibody effects on N. silvestris and
E. coli extracts................................. 116

7-1 SDH/DQT activities in transformed aroD and aroE
mutant strains of E. coli ........................ 122

7-2 Pairwise comparisons of S-protein with its
homologues....................................... 144

7-3 NAD(P)H binding motif of enzymes................. 145

7-4 Transit peptides of aromatic pathway proteins
and other plant proteins......................... 143








arogenate dehydrogenase
arogenate dehydratase
anthranilate synthase
aneuploid N. silvestris suspension cell line
beta mercaptoethanol
bovine serum albumin
chorismate mutase
carboxymethyl cellulose-52 cation exchange column
chloroplast chorismate mutase
cytosolic chorismate mutase
3-deoxy-D-arabino-heptulosonate 7-phosphate
DAHP synthase, enzyme 1 of aromatic pathway
(DEAE) diethylaminoethyl cellulose-52 anion
exchange column
dehydroquinate dehydratase or dehydroquinase
cytosolic DAHP synthase of plants
chloroplast DAHP synthase of plants
cells of about two generations of exponential
cells in continuous exponential growth
propanesulfonic acid];HPPS)
erythrose 4-phosphate
hydroxylapetite column
high performance liquid chromatography
Interdisciplinary Center of Biotechnology Research
potassium phosphate buffer
Luria-Bertaini medium
LB plus ampicillin
lag, exponential and stationary phases of growth
zolium-bromide; (thiazolyl blue)
nicotinamide adenine dinucleotide
nucleotide triphosphates

OAA oxaloacetate
OPA orthopthalaldyhyde
PAGE polyacrylamide gel electrophoresis
PAT prephenate aminotransferase
PCA protocatechuic acid
PCMB p-chloromercuribenzoate
PDH prephenate dehydrogenase
PDT prephenate dehydratase
PEGo000 polyethylene glycol
PEP phosphoenolpyruvate
PLP pyridoxal 5'-phosphate
PMS N-methyldibenzopyrazine methylsulfate (phenazine
PMSF phenylmethylsulfonyl fluoride
PPA prephenate
PPY phenylpyruvate
PCA protocatechuic acid
QA quinic acid
QDH quinate dehydrogenase
QDT quinate dehydratase
RE restriction enzymes
SA specific activity
S-protein shikimate-bifunctional DQT/SDH protein
SDH shikimate dehydrogenase
SDS sodium dodecyl sulfate
SP-I (S1-protein) bifunctional protein (DQT/SDH), major
peak from Celite 545 chromatography separation
SP-II (S2-protein) bifunctional protein (DQT/SDH), minor
peak from Celite 545 chromatography separation
SHK shikimate

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



Carol Ann Bonner

April 1994

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

Two bifunctional proteins containing catalytic domains

for dehydroquinate dehydratase (dehydroquinase) and

shikimate dehydrogenase were separated from extracts of

Nicotiana silvestris suspension cells by use of a decreasing

ammonium sulfate gradient on a Celite 545 column. The two

apparent isoenzymes, denoted as SP-I and SP-II, were

purified 1000- and 800-fold, respectively. Higher molecular

mass values were obtained for SP-II than for SP-I on the

criteria of sodium dodecyl sulfate (SDS) polyacrylamide gel

electrophoresis (PAGE) and gel filtration chromatography.

Each protein exhibited multiple bands on SDS-PAGE and on

native-PAGE monitored with a specific shikimate


dehydrogenase activity stain. Temperature and pH optima for

catalysis were similar for each protein, as were the

temperatures of inactivation. However, Km values for

shikimate were 0.80 and 0.36 mM for SP-I and SP-II,

respectively. Antibody raised against SP-I cross-reacted

with SP-II, but not with Escherichia coli proteins

possessing the corresponding activities. Antibody screening

of a cDNA library of Nicotiana tabacum was used to isolate

clones expressing the S-protein. Functional complementation

of aroD and aroE E. coli auxotrophs transformed with

plasmids carrying cloned cDNA were successful. Activities

for each enzyme were 15-fold greater in the transformed

mutant strains than in the wild type strain of E. coli.

Analysis of the amino acid sequence deduced from the cloned

cDNA sequence revealed homology with the appropriate

functional domains of the pentafunctional Arol protein of

Saccharomyces cerevisiae and with the monofunctional AroD

and AroE proteins of E. coli. The post-prephenate pathway

enzymes were also studied in Nicotiana silvestris suspension

cells. Levels of arogenate dehydratase activity changed

throughout a growth cycle in suspension cells. Using

specific antibodies, several putative cDNA clones encoding

prephenate aminotransferase or arogenate dehydrogenase were

isolated for further study. The base of comparative

enzymology in higher plants for the aforementioned enzymes

was extended to the unicellular alga, Chlorella sorokiniana.


Prephenate aminotransferase of Chlorella exhibited the

striking high-temperature activity optimum and specificity

for prephenate that is typical of higher plants.




Aromatic Amino Acid Biosynthesis in Plants

In higher plants, the three aromatic amino acids (L-

phenylalanine, L-tyrosine and L-tryptophan) are not only

required for protein synthesis and primary metabolite

synthesis (such as the hormone, indole acetic acid and the

structural component of woody plants, lignin), but also must

be available as starting substrates for a vast array of

secondary metabolites including alkaloids, coumarins,

isoflavones, and tannins (2, 78, 79). Intermediates of the

pathway are also precursors for synthesis of other essential

metabolites in plant cells. The vitamin-like derivatives,

folic acid and ubiquinone, are synthesized from chorismate,

the first branch point intermediate in the aromatic pathway.

Protocatechuic acid is derived from the common pathway

intermediate, dehydroshikimate (Fig. 1). It has been

estimated that up to 60% of plant carbon flows through the

aromatic amino acid pathway (40). Humans as well as other

higher forms of life depend heavily on plants to provide the

three essential aromatic amino acids, and knowledge of

aromatic biosynthesis in higher plants is fundamentally

Fig. 1-1. Aromatic amino acid biosynthetic pathway.
The common portion of the biosynthetic pathway leading to
the branch point intermediate, chorismate, begins with
DAHP synthase [1] catalizing the condensation of
erythrose 4-phosphate and phosphoenolpyruvate and
consists of seven steps. These steps consist of
dehydroquinate synthase [2], dehydroquinate dehydratase
[3], shikimate dehydrogenase [4], shikimate kinase [5],
EPSP synthase [6], and chorismate synthase [7].
Chorismate, at the branchpoint of the pathway is utilized
in one direction to yield L-tryptophan (steps 8-12) and
in another direction by chorismate mutase [13] to produce
prephenate, another branch point intermediate. Organisms
in nature use different routings from prephenate. Higher
plant chloroplasts proceed through prephenate
aminotransferase [14] and arogenate dehydratase [15] to
form PHE, or through prephenate aminotransferase and
arogenate dehydrogenase [16] to form TYR. In other
organisms a second route proceeds through prephenate
dehydratase [19], then through aromatic (or PHE)
aminotransferase [20] to synthesize PHE, or through
prephenate dehydrogenase [17] and through aromatic (or
TYR) aminotransferase [18] to synthesize TYR.

C No c 4C COO "



& N"
1;1 f:C.coo @. Coom N OP.%cSi
N" ON On


S g--t 0

/ /
191 (si


\ -\

1171 1161


0 OH



& r... ^oo"

important. Figure 1 shows a composite schematic of the

biosynthetic steps of the common portion of the pathway and

of the three divergent amino acid branches.

It has been established that a complete aromatic

biosynthetic pathway is present in the chloroplasts of plant

cells (2, 41). The common portion of the pathway begins

with the condensation of erythrose 4-phosphate and

phosphoenol pyruvate to form 3-deoxy-D-arabino-heptulosonate

7-phosphate by DAHP synthase and continues with six more

enzymatic steps to form CHA, the major branchpoint

intermediate. From CHA, the pathway branches to form L-

tryptophan via anthranilate synthase (ANS, enzyme 8) and to

L-phenylalanine or L-tyrosine via chorismate mutase (CM,

enzyme 13). From prephenate, the product of the CM

reaction, different pathway variations have been shown in

nature to complete the synthesis of PHE and TYR (16). In

plant chloroplasts, the post-prephenate pathway begins with

the transamination of PPA to yield L-arogenate by prephenate

aminotransferase (PAT, enzyme 14). Finally, AGN is either

converted to PHE by arogenate dehydratase (ADT, enzyme 15)

or to TYR by arogenate dehydrogenase (ADH, enzyme 16). This

pathway is illustrated with dashed arrows in Fig. 1.

The basic hypothesis underlying my studies is that a

second pathway for aromatic amino acid biosynthesis exists

in the cytosol compartment of higher plants. The supportive

evidence is that DAHP synthase (the early-pathway enzyme)

(32), CM (the mid-pathway enzyme) (23) and ANS (the initial

TRP-branch enzyme) (15) are present as pairs of separately

compartmented isoenzymes (Fig. 1-2). It is also known that

starting substrates for aromatic biosynthesis (E4P and PEP)

are synthesized by duplicate pathways of carbohydrate

metabolism located in both the plastid and cytosolic

compartments (Fig. 1-2). Since most secondary metabolites

are synthesized in the cytosol of plant cells (39) (Fig. 1-

2), it seems likely that an aromatic pathway should be

available in the cytosol to produce aromatic amino acids not

only for protein synthesis but as starting substrates for

secondary metabolism as well.

Cytosolic isoenzymes of DAHP synthase and chorismate

mutase (Ds-Co and CM-II, respectively) are markedly

different from Ds-Mn and CM-I (the corresponding plastid-

localized isoenzymes) in both catalytic and regulatory

properties (32, 23). Such differences and indeed a lack of

homology would be consistent with the endosymbiotic

hypothesis of organelle evolution (55). If cytosolic

species of other pathway steps are present in plant cells,

it would not be surprising to find them to have equally

divergent properties that reflect their xenologous origin at

the time of endosymbiosis. Thus, the alternative post-

prephenate steps shown in Fig. 1 (solid arrows) might be

used in the cytosol. From this perspective it seems

unlikely that dehydroquinase and shikimate dehydrogenase

*H C-r- -H ,C 3-) 4
0 4 ,C 4 (rd

r-1 r a 3 f 3-
Sa sr o

S--,- > r "
o0 0 0 0'
r-l 0 C
(0 r0 ( 4
02 3.02
h o a, O

( U 0 o EU
, .O .r 4
"0 (o o

*-1.5 u U o; M
04 4 41 f :
nJJ0 0J

-o ( 0 0 r
--0 0 r- 0E

0 0
4-1 Hd~4 C=

04 r= 4.4 0

0 0 -

o.~ 00 0
-0 rH 0 q
S 0 rmd
- 5 oa ) "

0cd 0 0,
O- 4 4
a C (d 0 -4

W r= C: rq
0404- 0r^

(d u 91 4J U J3 M

0~K SP

IL V-% IJ o

IL I i

0 IL


would co-exist as a bifunctional protein, as they do in

plastids. Alternative possibilities would be that they are

monofunctional species, or that they are domains of a large

pentafunctional protein like that present in fungi (21).

That the enzymes catalyzing the third and fourth steps

in the common pathway of plastids are bifunctional in higher

plants has been documented (14, 61). In E. coli, DQT and

SDH are each found as monofunctional proteins (18,19).

Yeast and fungi produce the arom protein (21), a

pentafunctional protein bearing functional domains

corresponding to steps 2 through 6 of aromatic amino acid

biosynthesis (Fig. 1-1). It is interesting that when the

arom pentafunctional protein of Neurospora crassa is

subjected to limited proteolysis, the DQT/SDH domains were

retained as an intact and functional fragment (77).

Rationale for Focus upon DOT and SDH

The rationale for choice of the DQT/SDH protein (or S-

protein) for study was as follows. (i) SDH is easily

assayed, has very high activity and the substrates are

readily available. (ii) The S-protein is in the middle of

the common portion of the pathway and provides a link from

the early-pathway isoenzyme pair of DAHP synthase to the

second known mid-pathway pair of isoenzymes of chorismate

mutase. (iii) Two separable shikimate dehydrogenases had

been demonstrated by Rothe et al. (67) in pea seedlings, and

my preliminary experiments had shown this result to be

repeatable in N. silvestris. (iv) Distinctly different

properties of the isoenzyme pairs for DAHP synthase,

chorismate mutase, and anthranilate synthase have been most

useful to demonstrate the separate spatial locations. Since

DQT and SDH were known to coexist as a plastid-localized

bifunctional protein, I thought it likely that the

corresponding cytosolic enzymes would be monofunctional. If

so, this would be a type of differential property that might

allow me to distinguish the two during cell fractionation


Shikimate biosynthesis and quinate catabolism

In monocots, such as Zea mays, there are two

dehydroquinases, one is bifunctional with SDH (specific for

NADPI) and the other is either bifunctional or stably

completed with a NAD' specific quinate dehydrogenase (36).

The pathway steps to quinic acid in plants have not yet been

rigorously established and may originate from E4P and PEP

via DHQ, or from unknown pathways. Quinate has been shown

to accumulate in plant vacuoles and to account for at least

10%, if not more, of the dry weight of certain plant tissues

(2, V. I. Ossipov, personal communications). The quinate

catabolic system from fungi (34,35) overlaps the

biosynthetic shikimate pathway with several analogous

dehydrogenase and dehydratase reactions (Fig. 1-3). An

u a) 4) 5 02t)' ro 0 a) :
-ri 4J 4- c 4 a0 0 -Hq 4J 0
.0 (d mU 0 Ea -H (d 10r :j
(d 0 :4-44 i1 24 4J 0 M
04 4) ..
O m 4 83
-1 0
a) U a( ) -- 0 i

() U
3: (L U 4-4


a,) o
ON Q rHbl
4j c+ a) ma 0
rj d Q (d j Ia)

r~-I 4J t 4 r= Q 4 E
o r. "Ia) ) aE-14-
0-HH M4 H EO

ud UwrUa,02 a )
t4 0 M 4.) -C: 0
o-i i0 rl r A
04 M (d 04
-rl 0 P -ri 4 r=

004-) 0 rd 0oH

a)HO4 r
$44 M2 0 0
(d 4-) H 4-)k~

~ >02
4J r-i W a)
-H 4- rd' 0 0 -H 02
-H M -r 'd 02 0

4 .02 >1
. 2 0 -r -H 40
0 4j 4-) E r(j

r. a) r 4-E>
>4 4 0 d 5

41, U4 U J

-H H1- 1 0l -) a) 04I4-
4- Ua ~IJ ) 0 4-
r d d CJ2 -
04 ro o)S
.rq 41 >, ao 4 ) Q mb
::, 0 'o -ri 0 ril M -r-I Mr
CY> a 04 Eo (d$
j ) r= (d Q) 4

Q a) -rq 0 0 r. a

0 -rl I 0 a)
.H io rd o d 4)
04rd 0 (d

-r- 4 4 0 ) a) p P-r

5 M 0 :4 4 4 W
fd U rO 0 CO Q 4J> 4










o \O




M e


.. -


inducible catabolic dehydroquinase takes DHQ that comes

directly from quinic acid (by a specific NAD' quinate

dehydrogenase) to dehydroshikimate which is converted to PCA

and on to carbon building blocks generated by the TCA cycle.

The dehydroquinase functional domain of the pentafunctional

arom protein of yeast and fungi channels DHQ formed from

DAHP to EPSP, thus bypassing PCA formation and avoiding

futile cycling (17, 21, 28). Both a NADPH specific QDH (V.

I. Ossipov, personal communication) and a quinate

dehydratase (V.I. Ossipov, personal communication) have

recently been shown. The NAD* specific QDH from E. nidulans

also will utilize SHK (34). These two systems offer some

interesting possibilities for overlapping metabolism in

plants. For example, it is conceivable that a QDT might

convert quinate to SHK for synthesis of aromatic amino acids

(Fig. 1-3) to be utilized as precursors for the synthesis of

lignin and other compounds.

Multiple Species of Shikimate Dehydrogenase

Shikimate dehydrogenase has been a popular enzyme

choice for a variety of isoenzyme studies and multiple

activities have been described on many occasions.

The bifunctional DQT/SDH protein (NADP' specific) was

examined in dicots (50). No evidence of a separate DQT or

SDH has been determined, thus far.


Cofactor specificity. Rothe et al. (67) separated two

SDH (NADP') activities from pea seedlings on a Celite 545

column with an ammonium sulfate gradient. Whether DQT

activity coeluted with one or both of the SDH activities was

not examined. Shikimate dehydrogenase was shown to be

located in the stroma of spinach chloroplasts and was

partially purified (30). Mousdale et al. (61) purified a

bifunctional DQT/SDH (NADP' specific) from pea seedling

chloroplasts and also showed by chromatofocusing that there

were two chloroplast isoenzymes and one unstable isoenzyme,

probably located in the cytosol. Two forms of nondissoc-

iable DQT/SDH (NADP') activities and one of QDH (NAD*) were

found in Phaseolus mungo seedlings by Minamikawa (58).

Koshiba (49) also found two forms of DQT/SDH (NADP')

activities from Phaseolus mungo seedlings. In 1988, Ogawa

and Tateoka (64) reported one SDH (NADP*) specific activity

and two SDH (NAD) specific activities (DQT apparently was

not assayed in these fractions) from Phaseolus mungo. This

is the first description of NAD* specific SDH activities and

is surprising since the same methodology for separation was

used by the three groups and each checked for specificity of

cofactors. A recent study of QDH in Phaseolus mungo by Kang

and Scheibe (44) found that the enzyme could use NADP' as

well as NAD*. When SHK was added to the reaction mix

containing quinic acid and NAD', some inhibition was found.

Apparently SHK was not tried as a substrate for QDH. If QDH

also functions as a SDH, competition for two products

(dehydroquinate and dehydroshikimate) would occur and this

could account for the inhibition seen with SHK added to the

reaction mix. This might also explain the earlier report of

the NAD* specific SDH activity by Ogawa and Tateoka (64), if

the enzyme was actually a nonspecific QDH. The qa gene

cluster of Neurospora crassa encodes a functional domain for

a QDH (NADH specific) that will also utilize SHK/NADH (34).

Properties. The range of molecular mass for DQT/SDH

(or SDH, where DQT was not studied) is from 44,000 to

73,000. It is not clear as to whether the lower Mr of

44,000 might represent a monofunctional SDH. SDH activity

is usually assayed in the reverse of the physiological

direction with the substrates, SHK and NADP', rather than

the forward direction with DHS and NADPH:

Dehydroshikimate < > Shikimate

Values of Km have been determined and vary from 0.060 to 1.3

mM for SHK, from 0.010 to 0.10 for NADP', and from 0.130 to

0.180 mM DHQ. Table 1-1 gives estimates of Mr and Km values

for DQT and SDH from several plant species, as well as Mr

from E. coli and from S. cerevisiae.

Inhibition studies with HgCl2, p-chloromercuri benzoate

and oxidized glutathione have suggested that SDH

Table 1-1. Molecular

Source Substrate

Sweet potato SHK
Mung bean 1 DHQ
Mung bean 2 DHQ

mass and

Km (mM)



Km values of DQT and SDH.

Mra Reference

-- 48

57,000 (GE) 49

57,000 (GE)

Tomato SHK 0.038 73,000 (GF) 52
Pea SHK 0.690 59,000 61
NADP 0.013
Corn 1 -- 55,000 (GE) 13
Corn 2 -- 49,000 (GE)
Spinach -- 67,000 (GF) 30
59,000 (GE)
Cucumber SHK 0.060 44,000 (GF) 53
NADP 0.010
Radish SHK 0.860 60,000 (GF) 71
NADP 0.090
Spinach SHK 0.480 44,000 (GF) 71
NADP 0.032 49,000 (GE)
E. coli -- 56,000 (AA) 28b
S. cerevisiae -- 68,000 (AA) 77

aMolecular mass estimated by gel filtration (GF), by PAGE
(GE) or calculated from the deduced amino acid sequence
(AA) .
bMolecular mass of monofunctional DQT and SDH were combined.

sulfhydryl groups act as functional groups in the catalysis

of the enzyme and that partial reversal of inhibition is

accomplished by thiols (48, 53). Aromatic compounds,

metals, and salts have been studied with respect to effect

on DQT and SDH activities. Protocatechuic acid was found to

be a competitive inhibitor of SDH activity in tomato (52,)

and Cl- inhibited DQT activity competitively in pea (61).

Ca2 stimulated SDH activity in Brassica rapa (71).


Molecular genetics. Plant genes encoding DQT or SDH

have not been sequenced, but the monocistronic aroD and aroE

genes have been cloned and sequenced in E. coli (1, 27).

Amino acid sequences are also available for discrete domains

encoding these activities in S. cerevisiae (28) and

Aspergillus nidulans (now known as Emericella nidulans) (17)

as part of the pentafunctional arom gene (aroM). The E. coli

amino acid sequences, aroD (DQT) and aroE (SDH) exhibit only

have about 25 and 21% identity, respectively, with the

corresponding functional domains of the yeast arom protein.

In higher plants, aromatic-pathway genes thus far

cloned and sequenced are limited to DAHP synthase (29),

shikimate kinase (70), EPSP synthase (73), and chorismate

synthase (69) of the common portion of the pathway.

Anthranilate synthase (63) and tryptophan synthase 0 subunit

(3) of the tryptophan branch of the pathway have been cloned

and sequenced.

The Post-Prephenate Pathway in Plants

The dual-pathway hypothesis of aromatic amino acid

biosynthesis specifies that an intact post-prephenate

pathway exists in the cytosol in addition to that already

demonstrated in the plastid compartment (41). If a separate

cytosolic pathway functions: (i) separate genes encoding

isoenzyme counterparts of the plastid-localized enzymes may

exist, (ii) separate genes encoding alternative enzymes

steps may exist, or (iii) the unprocessed preprotein

precursor of the plastid enzymes may function in the


Prephenate aminotransferase. PAT of higher plants was

first characterized in partially purified extracts,

separated from aromatic aminotransferases, and shown to be

unusually specific for its substrate, prephenate. Activity

was optimal at 700C for PAT, a temperature which inactivated

other interfering aminotransferases (5). PAT was purified

about 1000-fold (7), and localization studies showed the

enzyme to be located in the chloroplast (Bonner and Jensen,

unpublished data, 1985; 41, 76). A simple, inexpensive

spectrophotometric assay was developed for PAT by following

the increase in oxaloacetate when the L-aspartate/a-

ketoglutarate couple was used (6). In the same study, the

ASP/aKG couple was used with native PAGE activity stained

gels to relate the position of PAT with respect to other

aminotransferases and to show the disappearance of heat-

inactivated aminotransferases.

Arogenate dehydrogenase. Partially purified ADH was

studied in N. silvestris with respect to allosteric

regulation by TYR by Gaines et al. (31). ADH was partially

purified and regulatory properties were determined in

Sorghum bicolor (22). The enzyme was purified about 1000-

fold (8). PAT, ADH and SDH activities were followed

throughout a growth cycle in N. silvestris suspension cells

(12) and after wounding of potato (59).

Arogenate dehydratase. ADT activity was first

determined in N. silvestris cells by Jung et al. (43). The

enzyme was localized in spinach chloroplasts and allosteric

regulation (activation by TYR and inhibition by PHE) was

studied (43). Siehl and Conn (75) partially purified ADT

from Sorghum bicolor and studied some kinetic and regulatory


Rationale for focus upon the post-prephenate enzymes.

The rationale in choosing to study some aspects of the post-

prephenate enzymes for this dissertation is my long-term

objective to follow up earlier work with molecular-genetic

approaches. Studies of the individual enzymes may shed some

light as to the multiplicity of compartmentation. The

merits are as follows. A) Prephenate aminotransferase (i)

has been characterized and a well-developed HPLC assay

exists (5), (ii) has important biotechnological application

due to its unique aminotransferase properties and the value

of its product, L-arogenate, and (iii) can be purified by an

established procedure (7). B) Arogenate dehydrogenase (i)

has an established purification methodology (8) for

providing purified protein for antibody production that may

be used to probe cDNA libraries, and (ii) is easily assayed

fluorometrically. C) Arogenate dehydratase has an

established HPLC assay (43), but is by far the most

difficult of the three enzymes to assay and purify.



Organisms and extract preparations


Nicotiana silvestris Speg. et Comes suspension cultured

cells originally isolated from haploid leaf material have

been maintained in our laboratory for about twelve years and

are well described physiologically and biochemically (10,

11, 12). Cell populations are subcultured in stationary

phase (7 day), diluted fivefold into fresh Murashige and

Skoog medium (62), and grown under specified conditions in

controlled growth chambers (10). For these experiments

(unless otherwise stated) seven-day old cells were harvested

by filtering through Miracloth (Calbiochem, LaJolla, CA),

rinsed with nanopure water, ground in liquid nitrogen in a

Waring blender and stored at -700C. until use.

C. sorokiniana cells cultured as described by Meridith

and Schmidt (57) were kindly provided by Dr. Robert Schmidt.

Bacterial strains and plasmids

E. coli strains used for transformation include DH5a,

AB1360 aroD (DQT-) and SK494 aroE (SDH-). Plasmids used for

subcloning were pUC18, pGEM5Zf(+) or pBluescript SK+.

Preparation of N. silvestris extract

Crude extracts for assay. Frozen, ground N. silvestris

cells were suspended (1:1) in 100 mM KPO4 buffer at pH 7.3

containing 20% glycerol, 1.0 mM phenylmethylsulfonyl

fluoride and 1.0 mM dithiothreitol, centrifuged at 35,000 x

g at 40C for 20 min. Extracts were treated to 80% of

saturation with ammonium sulfate, centrifuged as above,

followed by resuspension of pellets in EPPS-KOH buffer (pH

8.6) with all of the above components and dialyzed with four

changes of buffer until ammonia could not be detected by

HPLC. Extracts not treated with ammonium sulfate were

desalted using PD10 columns in Epps-KOH buffer as above or

dialyzed overnight with 4 changes of the Epps-KOH buffer.

Extracts for purification of ADH and PAT. Five-day N.

silvestris suspension cells were used in the purification of

ADH and PAT. Crude extracts were prepared as described

above and also included 0.1 mM pyridoxal 5' phosphate and

1.0 mM EDTA in order to stabilize PAT (5).

Extracts for S-protein purification. For purification

of the S-proteins, frozen, ground seven-day N. silvestris

suspension cells (300 g) were suspended into 300 ml of 100

mM KPO4 buffer, pH 7.8, containing 20% glycerol, 1.0 mM DTT,

1.0 mM benzamidine, 0.2 mM PMSF and 0.05 mM leupeptin. A

sample was immediately desalted using a PD10 column,

equilibrated with 50 mM EPPS-KOH buffer at pH 8.6,

containing the same components as above and labeled as crude


Preparation of cultured cells for assay of ADH levels

The isogenic suspension cell-culture line of N.

silvestris was grown in continuous light with gentle shaking

at 150 rev/min. The growth properties of a N. silvestris

suspension cell line (ANS-1) in lag, exponential, and

stationary phases of growth have been characterized

throughout a conventional growth cycle (referred to as E

cells) and for cells which have been maintained in

exponential growth for more than 10 generations, EE cells

(10). For routine subculture, 80 ml of suspension-culture

medium were inoculated with 20 ml of 7-day stationary-phase

cells in 500-ml Erlenmeyer flasks. EE cells (2 g wet

weight) were transferred to fresh medium on day 7 when cells

reached about 1 X 106 cells ml- Daily samples were

harvested using Miracloth for filtration, washed with

distilled water, ground and frozen in liquid nitrogen, and

stored at -700C until extract preparation. Daily sampling

was also done to determine a cell count by mixing 1 ml of a

50 ml culture with 12% chromium trioxide (helps to separate

cell clumps and visualize cells more easily). Samples were

heated to 700C for 2 to 5 min before counting in a Fuchs-

Rosenthal counting chamber at 20x magnification. The

remaining 49 ml of culture were filtered and cells were

dried on preweighed filters made from Miracloth in an oven

at 500C and were weighed daily until constant dry weight

values were obtained. From these two methods, growth curves

were determined.

Preparation of C. sorokiniana extracts

Frozen cell pellets of C. sorokiniana were suspended

into 50 mM PIPES buffer, pH 7.3, containing 1.0 mM DTT, 1.0

mM PMSF, 35 mM KC1, and were broken by 2 passes through a

French Pressure Cell at 20,000 psi. The slurry was

centrifuged at 18,000 x g, at 40C for 15 min, and the

supernatant was next centrifuged at 150,000 x g for 1 h at

40C. Extracts were either desalted and used for column

chromatography, or dialyzed with 50 mM EPPS-KOH buffer with

all protective components as listed above for use in enzyme

assays. (Pipes buffer plus KCl was used in place of EPPS

buffer for column chromatography in order to separate

isoenzymes of chorismate mutase for another study.)

Glycerol and PLP were added to column fractions to protect

several of the enzymes to be assayed (PAT and ADH).

Preparation of bacterial extracts

The mutant E. coli strains (aroD and aroE) carrying the

cDNA clones encoding functional proteins (DQT or SDH,

respectively) were grown in 200 ml of LB/Amp media, washed

and resuspended in KPO4 buffer containing 20% glycerol and 1

mM DTT at pH 7.6, and centrifuged at 150,000 x g, at 40C for

1 h. The supernatants were desalted by passage through

Sephadex G-25 columns before fluorometric assay.

Enzyme assays


Prephenate aminotransferase. Prephenate

aminotransferase was usually incubated with concentrations

of 20 mM L-glutamate and 0.8 mM prephenate (unless otherwise

stated), incubated at 370C for 10 min (or as otherwise

stated) and then assayed by HPLC as an orthopthalaldyhyde

derivative in a 60% methanol/40% of 20 mM KPO4 buffer

system. Peaks of PHE were quantified before and after

acidification of the product of the reaction, AGN, in 1N HC1

for 10 min at 370C.

Aminotransferase couples. Other aminotransferase

assays included an amino acid donor (ASP, TYR, LEU or ALA)

and keto acid (PPY) at concentrations as stated and were

assayed similarly by HPLC, before and after acidification

when applicable. Controls included samples without amino

acid donor or keto acid.

Arogenate dehydratase

Arogenate dehydratase was assayed according to Jung et

al. (43) by measurement of OPA derivatives following HPLC

where the peak of AGN formed were monitored before and after

acidification to PHE. Concentration of 0.5 mM AGN, 0.25 mM

TYR (activator), 50 mM EPPS-KOH buffer at pH 8.6 and enzyme

were included in the reaction sample of 100 Al that were

incubated at 320C for 30 min. Controls included samples

without AGN or enzyme.


Dehydroquinase activity was assayed spectro-

photometrically at 234 nm at room temperature (240C) by

following the rate of increase of product, dehydroshikimate,

when dehydroquinate was provided as substrate. Unless

otherwise stated, a reaction sample included 0.5 mM DHQ,

enzyme and 50 mM EPPS-KOH buffer at pH 8.6. An extinction

coefficient of 12,000 was used to calculate activity (37).

DQT was also assayed fluorometrically via a coupled assay

with SDH. DHS formed from DHQ by the dehydratase reaction

was reduced with NADPH (cofactor for SDH) and the rate of

NADPH disappearance was followed. Reaction samples

contained 0.5 mM DHQ, 0.2 mM NADPH in 50 mM EPPS-KOH at pH

8.6 for column fractions or 50 mM Bis-tris propane buffer,

pH 7.5, for other studies as stated. Controls were carried

out for each enzyme assayed by adding only one substrate for

about three to five min before adding the second substrate.


Dehydrogenase reactions were usually assayed

fluorometrically at room temperature (240C) with cofactor

NAD(P)* for 1 to 2 min before addition of substrate, and the

rate of formation of NAD(P)H was monitored (340 nm emission

and 460 nm excitation). Unless otherwise stated, a total

vol of 400 il contained substrates and 50 mM EPPS-KOH buffer

at pH 8.6. The rate is expressed as fluorometric units per

min (FU min-1). Activities were calculated as nmol min-' by

relating experimental values to a standard curve obtained

with authentic NADPH. Any background rate seen when

cofactor was present alone in reaction sample (only in some

crude extracts) was subtracted from the rate obtained from

reaction sample containing both substrates to give an

appropriate correction. A spectrophotometric assay was also

used occasionally by following the rate of formation of

NAD(P)H at A340o.

Arogenate dehydrogenase. Arogenate dehydrogenase

activity was determined when 0.5 mM AGN was added to the

reaction assay after 0.5 mM NADP' had been added or as

otherwise stated. When appropriate, NAD* was added as

cofactor to the reaction mix.

Prephenate dehydrogenase. Prephenate dehydrogenase

activity was assayed with 0.5 mM NADP* (or NAD*) and 1.0 mM


Shikimate dehydrogenase. Shikimate dehydrogenase

activity was usually assayed in the reverse-of-physiological

direction with 1.0 mM NADP' and 4.0 mM SHK, unless otherwise

stated. When appropriate, NAD* was added to the reaction

sample. In the forward direction, SDH was assayed with

concentrations of 1.0 mM NADPH and 1.0 mM DHS or at

concentrations otherwise stated.

Quinate dehydrogenase. Quinate dehydrogenase was

assayed with 1.0 mM quinate and 1.0 mM NAD or NADP+ in 100

mM glycine buffer at pH 9.5.

Prephenate dehydratase

Prephenate dehydratase was assayed by incubation of PPA

and enzyme in 50 mM EPPS buffer (total vol of 100 Il) at

370C for 10 to 20 min before addition of 100 il IN HC1

followed by further incubation at 370C for 10 min. Controls

included PPA or enzyme with buffer only. Activity was

calculated from spectrophotometric readings at A20 nm and an

extinction coefficient of 17,500.

Protein estimations

Protein concentrations were estimated by the method of

Bradford (14) with BSA as a standard. A micro-assay was

used, as described, to estimate highly purified protein


Definition of activity and specific activity

Activity for all enzyme reactions in this study are

defined as nmol min-1 for the stated amount of protein added

into a standard assay volume. Specific activity is defined

as nmol min-' mg protein-.

Separation of pathway enzymes by chromatography

Column chromatography of N. silvestris extracts

A DEAE-cellulose column for separation of enzymes of

aromatic biosynthesis was prepared similarly as described

under purification. A 100 mg amount of protein was applied

to a bed volume of 102 ml of a DEAE column equilibrated in

50 mM KPO4 at pH 7.5. After washing the column with about 3

column volumes of buffer, a 600 ml gradient from 0 to 0.4 M

KC1 (KP04 buffer) was applied.

Column chromatography of C. sorokiniana extracts

A DEAE-cellulose column (20 ml) was prepared,

equilibrated in Pipes buffer (above), and about 29 mg of

protein from Chlorella were applied. The column was washed

with about 40 ml of the same buffer before a 35 mM to 500 mM

KC1 gradient was added. Finally, 1.0 M KCl/Pipes buffer was

added to remove residual protein from the column.

Appropriate fractions with DQT and SDH activities in the

gradient were pooled and dialyzed against 5 mM KPO4 buffer,

20% glycerol, and 1 mM DTT at pH 7.2. An HA column (10 ml)

was prepared with the 5 mM KPO4 buffer, pH 7.2. A 15-ml

sample from above was applied to the HA column and

activities for DQT and SDH eluted in the wash fractions.

Celite 545 (shallow-gradient application of Fig. 5-1)

A 200-g amount of ground, frozen 7-day suspension cells

was suspended in 200 ml of KPO4 buffer, 7.2 pH, containing

protectants and centrifuged as described above. A 3-ml

amount was removed for dialysis in Epps-KOH buffer and used

as a crude extract. Ammonium sulfate was added to give 75%

of saturation. From the slurry, 70 ml were removed and

prepared for use as a concentrated crude extract. To the

remaining slurry, 40 g of Celite 545 was added, stirred,

poured into a column (90 ml vol) and the eluate collected

into a cylinder. The column was washed with 65% (NH4),SO4 of

saturation in Epps-KOH buffer (about 180 ml) and 8-ml

volumes were collected. Buffer with (NH4)2SO4 at 55% of

saturation (120 ml) was added, followed by a shallow

gradient of 600 ml from 55% to 45% of (NH4)2SO4 saturation in

Epps-KOH buffer. A step gradient of 180 ml of buffer at 45%

of (NH4)2SO4 saturation was applied before the second

gradient of 45% to 0 of (NH4)2SO4 saturation (400 ml) was

added. A final wash of 120 ml of Epps-KOH buffer completed

the elution schedule.

Gel filtration

A Sephadex G-200 gel filtration column was used to

determine the native molecular weight of the S-proteins.

The column was equilibrated with 50 mM Epps buffer, pH 8.6

and DTT. Three ml of concentrated crude extract, and

concentrated SP-I or SP-II combined with 0.1 M sucrose were

applied individually to the same column (column vol of 471

ml, void vol of 119.4 ml). Protein standards of 5 mg each,

alcohol dehydrogenase (150,000 D), BSA (66,000 D) and

carbonic anhydrase (29,000 D) plus sucrose were applied to

the column.

Purification of pathway proteins

Purification of the S-proteins

Crude extract of 540 ml was brought to 60% of ammonium

sulfate saturation at 40C while stirring and was used for

the purification of the S-proteins.

Celite 545 chromatography. After stirring the 60% of

ammonium sulfate saturation slurry for 30 min, 70 g of

Celite 545 was added, with continued stirring for about 20

min. The slurry containing 605 mg protein was applied to a

column and the eluate collected into a large cylinder. The

column volume of 174 ml was washed with 60% of(NH4)2SO4

saturation-Epps-KOH buffer containing 20% glycerol and 1.0

mM DTT at pH 8.6 and was collected in a separate cylinder.

No activity was found in these eluates, although the protein

level was very high. A 500-ml gradient from 60 to 40% of

(NH4)2SO, saturation in the same buffer was applied and 8-ml

fractions were collected. Next a 40% of (NH4)2S04

saturation-Epps-KOH buffer of about 130 ml was applied. A

second 400 ml gradient of 40 to 0% of (NH4)2SO4 saturation-

Epps-KOH buffer was applied, followed by a wash of buffer

without (NH4)2SO4. Two peaks of SDH/DQT activities were

located, pooled individually, and dialyzed against the Epps

buffer as above. Samples of each were saved for assay, and

the remainder of each pool was used in the second

purification step (643 ml, 206 mg of SP-I protein and 348

ml, 97 mg of SP-II protein).

CM-cellulose chromatography. CM cation exchanger

columns were equilibrated with 50 mM Epps-KOH, 20% glycerol

and 1 mM DTT at pH 8.6. SP-I was applied to a bed volume of

123 ml and SP-II was applied to a bed volume of 80 ml. Both

SP-I and SP-II eluted in wash fractions and a small portion

was saved for assay. The remaining portions of the

preparations were used in the third step of purification.

DEAE-cellulose-52 chromatography. Two DEAE-52 anion

exchanger columns were prepared and equilibrated in the same

Epps-KOH buffer as the preceding step. Proteins were

applied, SP-I to bed volume of 107 ml and SP-II to a bed

volume of 72 ml, and washed with the same buffer. A 400-ml

(SP-I) or 300-ml (SP-II) gradient from 0 to 0.6 mM KC1 was

applied to the columns. Fractions of about 6.5 ml were

collected. Single peaks of activity, each eluting at about

0.3 mM KCL were individually pooled and dialyzed against 5

mM KPO4 buffer containing 20% glycerol and 1 mM DTT at pH

7.3. Samples were saved for assay, and the remainder of

each protein was used in the fourth step of purification.

Hvdroxvlapatite column chromatography. SP-I and SP-II

were applied to HA columns equilibrated in the KPO4 buffer

of the previous step (100 ml and 50 ml bed volumes,

respectively). Each protein eluted in the wash and was

applied to the fifth step of purification after samples were

removed for assay.

DEAE KPO4 chromatography. DEAE columns were

equilibrated in 50 mM KPO4 buffer at pH 7.3, before applying

the proteins from the previous step. Gradients up to 0.3 M

KC1 were applied to each column. Single peaks of each

protein were eluted, pooled and dialyzed into 50 mM Epps-KOH

containing 20% glycerol and 1 mM DTT at pH 8.6. Samples

were saved, and the remaining proteins were subjected to the

sixth step of purification.

2'5'ADP Sepharose-4B affinity chromatography. An

NADP'-specific 2'5'ADP Sepharose-4B affinity column was

prepared and equilibrated with the Epps-KOH buffer. The

SP-I sample was applied to the column and washed with

starting buffer before a 300-ml gradient from 0 to 0.18 M

NaC1 and 0 to 12 @8,25 AM NADP' was applied. The protein

eluted at about 4 iM NADP+ in the gradient. The SP-I

Activity peak was pooled and dialyzed into the Epps-KOH

buffer. A sample of SP-I was saved, and the remaining

fraction was used in the seventh step of purification. The

SP-II was applied to the affinity column similarly.

Purification of SP-II was completed with this step at a

final volume of 15 ml and total protein of 6 Ag.

Final steps of SP-I purification. SP-I was applied

once again to a DEAE column equilibrated in Epps-KOH buffer

as previously described. A 200-ml gradient from 0 to 0.6 M

KC1 was applied. The SP-I activity peak fractions were

pooled and dialyzed against Epps buffer. An aliquot was

saved, and 83 ml containing 0.162 mg protein was applied to

the affinity column for the final step of purification. The

completed purification yielded 0.07 mg SP-I protein in 54


Purification of arogenate dehydrocenase

ADH was purified by two methods: A) by a series of

steps of an established procedure which began with a 0 to

45% ammonium sulfate fractionation and ended with the NADP'

affinity column (8), or B) by the elution of ADH from a

Celite 545 column used to separate S-protein activities,

followed by HA column chromatography, DEAE column

chromatography, and the NADP+ affinity column chromatography

described above.

Purification of prephenate aminotransferase

PAT was purified by following an established procedure

(7), which began with a 45 to 75% of ammonium sulfate

saturation precipitation step, followed by a 700C treatment

to inactivate and then precipitate (by centrifugation)

contaminating aminotransferases, and ended with a specific

aminotransferase pyridoxamine phosphate (PMP) affinity



SDS gel electrophoresis

SDS gel electrophoresis was prepared by following the

procedure of Laemmli (51), with a 15% running gel and a 4%

stacking gel. A running buffer of Tris/glycine/SDS at pH

8.8 was used and the gel was run at constant voltage of 63 V

for about 8 h. The gel was stained for about 30 min in

Coomassie Brilliant Blue R250 dye, destined and viewed.

Silver staining

Silver staining of an SDS gel as described by Pharmacia

(Piscataway, NJ) was followed. Washes of methanol/acetic

acid and ethanol/acetic acid were included before incubation

of the gel in oxidizer, followed by addition of the silver

stain, then developer, and finally an acetic acid/water wash

to stop the reaction.

Native gel electrophoresis

A native PAGE was run according to the method of Shaw

and Prasad (74). Gel concentrations were 7% for the

stacking gel and 15% for the running gel. The running gel

was pre-electrophoresed to remove the ammonium persulfate,

and fluorescent light was used to solidify the stacking gel

which contained riboflavin. The gel was run at constant

voltage of 60 V for about 8 h.

Shikimate dehydrogenase activity stained gel

Optimal conditions for SDH activity stain on native

PAGE included 20 mg NADP*, 50 mg shikimate, 10 mg 3-(4,5-


(thiazolyl blue), and 2 mg N-methyldibenzopyrazine methyl

sulfate (phenazine metasulfate) in 50 ml of 100 mM glycine

buffer (without glycerol) at pH 9. Gels were incubated at

370C for 10 to 20 min or at room temperature for 30 min to

an hour. These conditions were used unless otherwise stated

and were modified from the method presented by Davis (24).

Kinetic values

Shikimate dehydrogenase

Enzyme activities were assayed fluorometrically as

described for SP-I and SP-II. In the forward physiological

direction, either DHS or NADPH was held constant at

saturating levels while the remaining substrate

concentration was varied to obtain saturation curves. In

the reverse of physiological direction, SHK and NADP* were

assayed similarly. Double reciprocal plots of the substrate

saturation data were used to determine the Km for each



DQT activities were assayed spectrophotometrically as

described for SP-I and SP-II at varying concentrations of

DHQ to obtain saturation data, and then analyzed

appropriately as above to determine the Km for each protein.

Inhibition studies of S-proteins

Enzyme was incubated for 10 min with p-chloromercuri-

benzoate or with possible protectants {DTT, cysteine or (-

mercaptoethanol (PME)}. SDH was assayed at saturating

conditions and DQT was assayed at 0.1 mM DHQ. For

prevention experiments, DTT, cysteine, or PME was added

about 10 min before addition of substrates and PCMB, and

then rates were determined. For reversal experiments, PCMB

was added to enzyme and buffer for 10 min before assay with

substrates. After a rate (or no rate) was established, DTT

was added to the reaction mix. Concentrations are as stated

in Results.


Preparation of specific antibodies

Antibody preparations were produced by Kel-Farms

(Alachua, FL) with repeated injections of purified SP-I over

the course of several months until antibody titers, tested

by Ouchterlony double diffusion method against SDH, ADH or

PAT, were high enough to bleed the rabbits.

Purification of specific antibodies

Antibodies were purified following a BIO-RAD (Rockville

Center, NY) procedure, by passing each over Econo-Pac 10DG

columns for desalting and then passing through DEAE Affi-Gel

Blue Econo-Pac 10DG columns for serum IgG purification. E.

coli strain DH5a lysate was prepared and incubated with the

purified IgG. This step removed non-specific precipitant

bands observed in the double-diffusion experiments. Pre-

immune rabbit serum were purified similarly.

Antigen:antibody assay

Antigens (SP-I, ADH or PAT) from crude extracts or

extracts from various purification steps were combined 1:1

(v:v) (or as stated) with antibodies. Samples were

incubated at 370C for 10 min and returned to an ice bath,

then centrifuged (5 to 10 min) at full speed in a microfuge

to precipitate the completed antigen:antibody. The

supernatants were assayed for enzyme activities. Controls

were always performed with preimmune rabbit serum.

Ouchterlony assay

Ouchterlony plates were prepared with 1.5% agarose in

sodium barbital and sodium azide and wells were prepared at

precisely measured distances (66). Antibodies or antigens

were diluted with 0.9% NaCl whenever necessary and then

added to the wells. Plates were incubated at 370C


Western blotting

Western blotting procedures from Sambrook et al. (68)

were followed, and bands from either SDS or native PAGE were

transferred to nitrocellulose filter paper. Membranes were

incubated in blocking solution alone and in blocking

solution containing diluted SP-I antibody for one h, washed,

incubated further with blocking buffer containing alkaline

phosphatase-linked goat-anti-rabbit antibody for 1 h, and

then incubated in STP buffer containing MgC21 and

substrates, nitroblue tetrazolium and the p-toluidine salt

of 5-bromo-4-chloro-3-indolyl phosphate. After 1 h,

membranes were viewed and photographed.

Antibody precipitation of DQT and SDH from C. sorokiniana

Antibody made to SP-I was incubated 1:1, 2:1, or 4:1

with Chlorella extract (controls included preimmune

antiserum and extract diluted with buffer). After 10 min at

370C, and centrifugation for 5 min at full speed in a

microfuge, preparations were assayed under standard

conditions for SDH and DQT activities.

cDNA cloning and sequencing

Tobacco leaf cDNA library

A cDNA library prepared from Nicotiana tabacum var.

SR1 tissue culture cells, inserted into pBluescript SK-

plasmids and packaged into Lambda ZAP II vectors, was

purchased from Stratagene, La Jolla, CA. E. coli XL1-Blue

(a host strain used for immunological screening of libraries

constructed in Lambda ZAP expression vectors) and E. coli

SOLR (a non-suppressing host strain for use after excision

of the pBluescript plasmid from the Lambda Zap vector) were

included. Helper phage, ExAssist, efficient for in vivo

excision of the plasmid from the Lambda Zap vector without

replication of the phage genome was also included.

Preparation of cDNA libraries for screening

The cDNA library was first titered, then mixed with the

freshly prepared host strain, XL1-Blue, plated with top agar

on about 80 LB plates at about 20,000 PFU (about 1.6 X 106

PFU/ desired clone) and incubated at 370C. Isopropylthio-3-

D-galactoside treated nitrocellulose filters were applied

about 6 to 8 h later and incubation was continued overnight.

Filters were removed and the immunological screening

procedure of Sambrook et al. (68) was followed.

Screening the cDNA library of Lambda ZAP II

Nitrocellulose filters removed from the plates were

washed, soaked in blocking buffer containing dry milk,

rocked gently in primary S-protein, ADH or PAT specific

antibody/blocking buffer and washed several times. The

filters were next exposed to secondary antibody, goat anti-

rabbit antiserum, washed as before and incubated with

substrates, nitroblue tetrazolium, 5-bromo-4-chloro-3-

indolyl phosphate and with 5 mM MgCl2. The filters were

kept in the dark and gently rocked for one to two hours.

Plaques corresponding to the small, deep blue spots on the

filters were cored from plates, dispensed into SM buffer

plus chloroform and stored at 40C for titering and plaque


Excision of Pbluescript from the Lambda ZAP II vector

Purified phage stocks were incubated with E. coli

strain XL1-Blue and ExAssist helper phage to excise the

pBluescript plasmid from the Lambda ZAP II vector for 2 to

2.5 h at 370C in 2XYT media (see Stratagene instruction

manual for details), heated to 700C and centrifuged. The

supernatant, containing the plasmid, was incubated with host

E. coli strain SOLR, spread on LB plates containing 50 mg

ml-1 ampicillin and incubated overnight at 370C. Colonies

from these plates were streaked on fresh LB-AMP plates for

immediate use and were also stored in DMSO for future use.

DNA purification

Plasmid DNA was purified by two methods. Method 1: The

pBluescript plasmids containing putative cDNA clones coding

for the S-protein, ADH or PAT were purified for sequencing

by the method of the DNA Sequencing Core Facility, ICBR,

University of FL. A colony from the E. coli SOLR strain

from the LB-AMP plate above was incubated overnight in 10 ml

of LB-AMP medium in a 50 ml tube for each clone sample. The

cultures were centrifuged and the pellets were resuspended

in GTE (68) buffer. Treatment with NaOH/SDS was followed by

neutralization with potassium acetate at pH 4.0, and

centrifugation was carried out to remove cellular debris.

The supernatants were treated with DNase-free RNase and

extracted twice with chloroform. DNA was precipitated with

isopropanol and washed with 70% ethanol. To precipitate the

plasmid DNA, the pellets were first dissolved in sterile

water, then 4M NaC1 and 13% PEGo000 was added, followed by

incubation on ice before centrifugation at 40C. The pellets

were rinsed in ethanol, dried under vacuum and resuspended

in sterile water. DNA concentrations were determined

spectrophotometrically at 260 nm and by agarose gel

electrophoresis with a known DNA standard. Method 2: The 10

ml cultures, as prepared above after centrifugation, were

treated with prepared solutions provided by Promega

(Madison, WI) in the Magic Miniprep DNA purification kit.

The DNA was passed through a mini-column and dissolved in

GTE (Promega instructions for preparation) buffer. This

rapid method is preferred for sequencing on a Li-Cor

sequencer available in the Department of Microbiology and

Cell Science.

Competent cells of E. coli strains

E. coli strains DH5a, AB1360 (aroD) and SK494 (aroE)

were grown overnight in 50-ml tubes and inoculated into 1-L

flasks containing 200 ml LB medium. The cells were

incubated at 370C in a shaker at 300 rpm for several hours

until a spectrophotometric reading at A600,, between 0.4 and

0.5 was reached. Cells were harvested by centrifugation in

sterile tubes after incubation on ice, treated with 100 mM

CaCl2 as described by Sambrook (68), resuspended into 100 mM

CaCl2 with 15% glycerol, aliquoted into microcentrifuge

tubes, dropped into liquid nitrogen and stored at -700C


until use. Controls were prepared by streaking auxotrophic

mutant strains on M9 medium plates +/- aromatic amino acids.

Transformation of plasmid DNA

Either pUC18, pGEM5zf(+), or pBluescript SK+ plasmids

were incubated with competent cells of DH5a on ice, heated

to 420C for 90 sec, cooled in an ice bath before addition of

SOC medium (68). The cells were then incubated at 370C for

1 h before spreading on LB-AMP plates. A colony from these

transformed cells was streaked onto a fresh LB-AMP plate.

Controls included spreading of competent cells without

plasmid or plasmid without insert onto LB-AMP plates.

Functional complementation

E. coli mutant strains, aroD (DQT-) and aroE (SDH-),

were prepared for transformation with the pBluescript

plasmids carrying cDNA clones and were plated on LB-AMP

plates. After growth, colonies were streaked onto M9

plates. Controls included nontransformed competent cells or

plasmid spread on LB-AMP plates, and competent mutant

strains spread on M9 plates +/- aromatic amino acids.

Initial sequencing of cloned cDNA

Purified pBluescript plasmid with cloned inserts were

submitted to the DNA Sequencing Core Facility, ICBR, UF, for

initial 5' and 3' end sequencing (from the T3 and T7

promoter sites).

Subcloninq cDNA for complete sequence analysis

Subcloning of cDNA inserted into pBluescript was

performed by use of appropriate restriction enzymes to

obtain fragments for sequencing both strands of the entire

cDNA. Plasmids used for fragment insertion {pUC18,

pGEM5zf(+), or pBluescript SK+} were incubated with the same

restriction enzyme(s), and were treated with calf intestinal

alkaline phosphatase (CIAP) (when only one RE was used) to

prevent rejoining of ends during ligation. Cloned cDNA

fragment samples and plasmid samples were loaded onto 1% DNA

agarose gels at 70V for about 1 h. Fragments and linear

plasmid DNA were excised and treated with the Gene Clean

system (Bio 101, LaJolla, CA), prior to ligation and


Li-Cor sequencing of subcloned cDNA fragments

Plasmid DNA was denatured and reacted with forward or

reverse fluorescent primers, enzymes, 7-Deaza dGTP sequence

extending mix and NTPS according to the directions supplied

by Li-Cor (Lincoln, NE). Three il of each of four prepared

sample mixtures containing one of the four dideoxy

nucleotides was loaded onto wells of a urea polyacrylamide

gel set into the Li-Cor model 4000L sequencer. Parameters

were set in a computerized data collection file and the

sequencer was run overnight. The sequences were analyzed in

a data analysis computerized system.

N-terminal sequencing

About 1 pmol of purified SP-I protein was prepared for

N-terminal sequencing on SDS PAGE, followed by transfer of

the protein onto 3MM paper by electrophoresis. Mini-blot

system protocol and amino acid analysis by acid hydrolysis

was provided by the Protein Chemistry Core Facility,

Interdisciplinary Center of Biotechnology Research,

University of Florida.

Computer analysis

The GCG computer software package (26) was used to

analyze nucleotide and deduced amino acid sequences of

cloned cDNAs.


Arogenate (90% pure) was isolated from a multiple

auxotroph of Neurospora crassa ATCC 36373 (42) or isolated

from a tyrosine auxotroph of Salmonella typhimurium (4).

Prephenate (85% pure) was prepared as the barium salt from

culture supernatants of the tyrosine auxotroph of S.

typhimurium and was converted to the potassium salt with

excess K2SO, prior to use. Dehydroquinate (90% pure) was

chemically synthesized by following the method of Haslam et

al. (37) using a platinum-catalyzed dehydrogenation of

quinic acid (about 90% pure). 2'5' ADP Sepharose-4B and AH-

Sepharose 4B were purchased from Pharmacia, Piscataway, NJ.

Bradford reagent was purchased from Bio-Rad, Rockville


Center, NY. DTT and hydroxylapatite were purchased from

Research Organics, Inc. Cleveland, OH. Shikimate, NADP,

NADPH, benzamidine, leupeptin, and PMSF were purchased from

Sigma Chemical Co., St. Louis, MO. All other chemicals were

purchased through Fisher Scientific Co., Orlando, FL.



The activities of dehydroquinase and shikimate

dehydrogenase (of the common portion of aromatic

biosynthesis), and prephenate aminotransferase, arogenate

dehydratase and arogenate dehydrogenase (of the post-

prephenate portion of the pathway) were examined under

selected physiological conditions in the higher plant N.

silvestris and in the green alga Chlorella sorokiniana.

Prephenate aminotransferase from Chlorella exhibited several

of the striking characteristics of higher plant enzyme.

Arogenate dehydratase activities were followed in a

growth culture cycle of N. silvestris suspension cells

similarly as was followed for PAT and ADH (12).


Specific activities from crude extracts

Specific activities have been compared for several

aromatic amino acid pathway enzymes of N. silvestris

cultured cells and from the green alga C. sorokiniana (Table

3-1). A constant ratio of about 5 for SDH and DQT activity

was found in N. silvestris, compared to a ratio of 1 for

crude extracts of C. sorokiniana. The activities of DQT and

SDH from Chlorella extract were unaffected when incubated

with SP-I N. silvestris specific antibody. Activity was not

found when QDH was assayed in N. silvestris and C.

sorokiniana crude extracts.

Table 3-1. Specific activities from N. silvestris and C.
sorokiniana aromatic pathway enzymes.

Specific Activity (nmol min' mg1 protein)

Enzyme N. silvestris C. sorokiniana

QDT 12.00 6.20
SDH 60.00 6.00
PAT 16.00 3.75
ADT 0.6 NF

NF, not found; Crude extracts were used for assay.

Separation of activities

Column chromatography was used to separate the enzymes

of interest for this study from N. silvestris and C.

sorokiniana. DQT and SDH of the common pathway were not

separable in N. silvestris or in C. sorokiniana after two

steps of column chromatography (DE-52 and HA). Figure 3-1

shows a DEAE-cellulose chromatographic profile of

N. silvestris suspension cell extract with overlapping





0H 0)


4- a)
U0 0

44 rd


um -

tJ4 1)~
ro U

41i Q

0 4~-J
Wi o:

U) 'i-
0) rou

w o co 0N co V
CM C q- d 0

o a d d5 o

0ldH (s oLx) V3u





enzyme activities of PAT, ADH, ADT and SDH (DQT activity was

coincident with SDH activity; not shown). Although ADT and

ADH activities were not found in crude extracts of C.

sorokiniana, they were located in DEAE-cellulose fractions

at very low activities as shown in Fig. 3-2A. ADH was

specific for cofactor NADP'. Prephenate dehydratase and

prephenate dehydrogenase activities were not found in these

column fractions. The ADH and ADT activities overlapped

those of PAT, DQT, and SDH in the KC1 gradient fractions.

The coeluting DQT and SDH activity peak fractions, as shown

in Fig. 3-2B, were pooled, dialyzed and applied to an HA

column where they once again coeluted, this time in the wash

fraction, data not shown.

PAT in C. sorokiniana

Several of the unique properties earlier shown for

prephenate aminotransferase in N. silvestris suspension

cells (5) were studied in C. sorokiniana extracts. A high

temperature optimum of about 700C for PAT was also found in

the alga extract as shown in Fig. 3-3. In Table 3-2, it is

shown that PAT from Chlorella has a preference for GLU as

the amino acid donor in combination with the keto acid

substrate, PPA. As found in higher plants, ASP was utilized

at about 50% of the activity with GLU. Slight activity was

seen with TYR as the amino acid donor in C. sorokiniana but

not in N. silvestris.

Fig. 3-2. DEAE-cellulose chromatography of aromatic
pathway enzymes from C. sorokiniana. Alga extract was
applied to a DE-52 column as described in Materials and
Methods. A) PAT (0), ADH (E), and ADT (A) eluted in the
KC1 gradient (-) and protein was monitored (...). B) SDH
(0) and DQT (0) eluted in the gradient.


4.5 -

S4.0 -

3.5 -

is.5 1.0 1.3

t 2.0 : 0.8 1.8


0.5 0. ***-* : 02 0.4

0 -0



s0 B

70 -

so -

u 50 :". 1.0 1.8

40 0.8 1.

30 0.6 E 1.2

20 0.4 0.8

10 / **..- 0.2 0.4

0 0
0 20 40 80 80

t I


0 20



3-3. Temperature optimum for activity of
aminotransferase in C. sorokiniana. The
range used was from 200C to 800C for 5 min.


Table 3-2. Prephenate aminotransferase amino acid donor
specificity in C. sorokiniana.

L-Amino Acid Donor (10 mM) Specific Activity

GLU 16.0
ASP 8.0
TYR 2.0

aA concentration of 1 mM PPA was used in combination with
the amino acids. S. A. is nmol min- mg protein-

When C sorokiniana extract was incubated at 650C for 10

min and then assayed at 370C, activity with PPY as keto acid

substrate was lost, but activity with PPA remained constant.

At 700C for 20 min, however, about 60% of the PAT activity was

lost (Table 3-3).

Table 3-3. Prephenate aminotransferase activity after
thermal treatment in C. sorokinianaa.

Aminotransferase No 650C 700C 700C
Couple treatment 10 min 10 min 20 min

PPA/GLU 0.61 0.60 0.61 0.25
PPY/GLU 0.29 0 0 0

aReaction mixtures were incubated at 370C following thermal
treatment. Assays contained 1 mM PPA or 10 mM PPY, 10 mM GLU
and 0.05 mg protein.

ADT activities in cultured N. silvestris cells.

When N. silvestris stationary phase cells were diluted

into fresh medium, a substantial elevation of soluble

protein occurred which increased until day 2, leveled off

between days 2 and 3, and began a steady decline between

days 3 and 7 as shown in Fig. 3-4B. This rise ceased before

mid-exponential growth was reached. That soluble protein

synthesis was terminated before reaching a maximum was

indicated by the finding that EE cells achieved a two-fold

higher content of soluble protein. EE cells exhibited a

consistently higher soluble protein content of about 360 mg

g dry wt.-'.

During the transition of cell populations in lag phase

to exponential-phase growth, new synthesis of macromolecular

constituents must have occurred. Figure 3-5A shows that the

rise in ADT activity expressed as units of activity per g

dry wt., paralleled the rise in soluble protein between

subculture and day 2 (Fig. 3-4B). The rise in ADT activity

levels continued until day 3, whereas overall soluble

protein content plateaued between days 2 and 3. Thus,

specific activity or activity expressed on a cell basis

peaked at day 3. Since the specific activity of E cells and

EE cells are similar and since the ratio of soluble protein

to dry weight is greater in EE cells than in E cells, the

activity of ADT expressed as units per g dry wt was greater

in EE cells than in E cells. The decline in specific

Fig. 3-4. Growth curve of N. silvestris in
suspension cultures. A) Cell growth monitored as cell
count or as dry weight. The physiological stages of
growth: lag (L), exponential (E) and stationary (S) are
indicated in each panel. B) Soluble protein content was
determined in samples harvested daily and was related to
the dry wt. and cell number. Comparable protein values
obtained with EE cells (average of ten determinations)
are indicated with stippled bars, the widths of which
indicate the range of variation.






z 150.


.;J 50-

0 1 2 3 4 5 6


Fig. 3-5. Arogenate dehydratase activities followed
throughout a growth curve of N. silvestris. Activity was
followed in samples harvested daily throughout the growth
curve shown in Fig. 2-4. A) Activities expressed as
units/g dry wt., B) as units/cell, C) as specific
activity. The stippled bars indicate the corresponding
activities found in EE cells (average of 10
determinations), the width of the bars indicating the
range of variation. D) total cumulative units of
activity per culture. See Fig. 2-4 for meaning of L, E
and S.

80- ---"--~-1-.-----A
0 60-
a 40



S fc 8 -
< ~ ,
UJ 4
I-- 1 I I I
0 0.24-
>- z
I 3 0.20 -;;--
w i*-
0 0
Q 0.16-

< a 0.12
z E
a L 0E08L s
o -
m 0.04-


0 1 2 3 4 5 6 7



activity (Fig. 3-5C) and units of activity per cell (Fig. 3-

5B) between days 3 and 5 (mid-to late-exponential growth

reflected a rate of enzyme synthesis that was slower than

the growth rate since total units of activity (Fig. 3-5D)

continued to increase during that time. From days 5 to 7

the decline in specific activity reflected a net loss of

enzyme activity.


C. sorokiniana aromatic-pathway enzyme activities

appear to be somewhat lower but consistent with those found

in higher plants. The S-protein from N. silvestris must be

antigenically different from that of C. sorokiniana, since

the S-protein specific antibody did not precipitate the

protein from Chlorella. Specific antibodies made to PAT and

ADH are now available (see Chapter V) to test cross

reactions with these proteins from Chlorella.

Although the coincident DQT and SDH activity peak from

the DEAE column which eluted protein from C. sorokiniana

extract was only purified through two steps, it is

conceivable that further purification would yield a

bifunctional S-protein like that found in higher plants.

Activities from column fractions indicated that DQT activity

was about half that seen for SDH, whereas crude extracts

indicated that activities were about equal. Possible

explanations would be that (i) more than one protein is


present in Chlorella for DQT and/or SDH, (ii) some component

in crude extract inhibits the SDH activity or (iii) DHS, the

product of the reaction is being utilized by another

protein. The second possibility may also apply to ADT and

ADH since activity was not found in crude extracts but was

located in column fractions.

Most interesting, is the similarity of PAT in Chlorella

to that of higher plants, each having the high optimal

temperature for activity. At 650C, the aminotransferase(s)

using the PPY/GLU couple was inactivated. The high

specificity of Chlorella for the substrate couple of PPA and

GLU was also consistent with higher plant PAT (5, 6). From

these data, it is suggested that C. sorokiniana has an

aromatic biosynthetic pathway similar to that established

for higher plant chloroplasts. Supportive evidence would

include the isolation of Chlorella chloroplasts to determine

the location of the aromatic pathway enzymes. The isolation

and sequencing of cDNA clones coding for S-protein, PAT and

ADH (see Chapter VI) may be useful as probes to clone the

appropriate genes in Chlorella.

Arogenate dehydratase activities paralleled a similar

experiment with respect to the chloroplast isoenzyme of DAHP

synthase (Ds-Mn), where growth of cells and enzyme

activities where followed (32). Both enzymes rise in

exponential growth and decline in stationary phase of

growth. ADT exhibited an opposite pattern of activity to

three other aromatic pathway enzymes, SDH, PAT, and ADH.

The three enzymes are known to exist in the chloroplast of

higher plants, but the extent to which a fraction of each of

the total activities measured might be derived from the

cytosol is unknown. All three enzymes exhibited elevated

levels of activity in stationary phase and minimal levels of

activity in exponential-phase growth (11). Aromatic-pathway

enzymes may be sorting out into two different patterns of

expression for reasons yet to be elucidated. Alternatively,

since enrichment of Ds-Co (the cytosolic isoenzyme of DAHP

synthase), occurred in stationary phase, rising levels of

SDH, PAT and ADH in stationary phase may reflect the

expression of cytosolic isoenzymes. If so, the rise in

activity of cytosolic isoenzymes exceeds the decrease of the

plastidial isoenzyme counterparts. These data suggest that

while ADT is present in the chloroplast pathway, it may not

function in the cytosol.If so, perhaps prephenate

dehydratase of the alternative route (activity, not as yet

determined in plants, see Fig. 1-1), functions in the

cytosol to synthesize PHE that may be used extensively for

secondary metabolism.



This chapter includes the separation and purification

of the common pathway bifunctional DQT/SDH proteins from

Nicotiana silvestris suspension cells. Six to eight steps

of column chromatography, including a specific NADP'

affinity column have been used. Prephenate aminotransferase

and arogenate dehydrogenase were also purified by a series

of chromatographic steps.


DOT/SDH purification

In crude extracts, activities for SDH and DQT (specific

activity of 61 and 13, respectively) were detected by

fluorometric or spectrophotometric assays. The enzyme(s)

are not stable at 40C after several days of storage, unless

glycerol and DTT are present. Under these conditions

activities remained stable to freeze/thaw, at either -20 or

-700C. Two peaks of activity for the bifunctional S-protein

were eluted from a Celite 545 column with an ammonium

sulfate gradient (Fig. 4-1). Elution of the major peak

began at about 50% of ammonium sulfate saturation and was

E rl U) 41
: a) cl -1
r- r4d > 4-)
0 d -H (d
(d 4j)

-'- Lt -*p'

cuw~C cV
1 44 C0
rj) a) r .
4-J4..Q 0

0 -H 14
cp 1 Od
(LU) 0-
H 4J 0(d 0

cJ) r.,
(d ;j 0 a ~
04 t jf a

a) 0

'--4~u r

144144 () 044.O
W a)~a)

0 r4 Q >4 1
0H rA 0

4.j) oD r
rx~ 0

0 Q Q

4. 44H 0 CO
0 -ri 04

0 0 D 4~
0 4)-ri4)
-rl 4.) 0 ra 4-)
E-4 (d 0 T

44 r
4 44 0a
m 5

o m
444 (d0)"
U 0 M 04~~

S c "3 V Nic l Ou d

Co C4 Co o o o0 0

31Vd-1nS VinlNOVYViV lN30U3d

8 C 0 0 0 0

0 0 0 0 0 0 0
o (0 A C 0 00

L- u!u na.

0 0 o
c Coc







almost completely removed from the column by the end of a

45% of ammonium sulfate saturation step. The major activity

peak fractions were completely separated from fractions

containing the minor peak of activity and eluted at about

25% of ammonium sulfate saturation. Fractions with DQT

activity were located coincidently with the two peak of SDH

activity. There were no column fractions containing only

one of these two activities. Fractions from each activity

peak were pooled and labeled SP-I for the major peak of

activity and SP-II for the minor peak of activity. Proteins

(SP-I and SP-II) were further purified similarly by

chromatography steps. It was calculated from this

chromatographic step that the major peak, SP-I, accounts for

90% or greater of the total activity in the crude extract.

The fold purifications calculated for the SP-I and SP-II

were based on this assumption. The proteins eluted in the

wash fractions of the CM columns in the second

chromatographic step of purification. Figure 4-2 shows

profiles of SP-I and SP-II elutions from the third step of

purification on DEAE-cellulose columns in Epps buffer at pH

8.6. Each protein peak eluted at about 0.3M KC1 in the

gradient and had coincident DQT and SDH activities. The

fourth chromatographic step was an HA column which eluted

each protein in the wash fractions. This step was followed

by another DEAE column, equilibrated at a lower pH of 7.2.

Each bifunctional protein eluted in the gradient of an NADP*

Fig. 4-2. Elution profiles of SP-I and SP-II from
DEAE- cellulose chromatography. The third step of
purification of the two S-proteins was a DEAE-cellulose
chromatographic column equilibrated in EPPS buffer (see
Materials and Methods for details). SDH (0) activity
remained coincident with DQT (0) activity for both SP-II
(panel A) and SP-I (panel B). The proteins began to elute
at about 0.25MKC1 of the gradient (solid line). Protein
profiles were followed at A280 n (...)




0 10 50 60 70 80 90
0 10 50 60 70 80 90










10 50 60 70 80 90

0.4 -

0.2 -

0.4 <
0.2 M
0 CL


0.4 -

0.2 -

100 110


0.4 Z
0 CL

specific 2'5'-Sepharose-4B affinity column at 1 AM NADP' and

18 mM NaCi (Fig. 4-3). Activities for DQT and SDH remained

coincident for each protein. The fold purification at this

step was 195 for SP-I and 918 for SP-II. Only SP-I was

purified further by repeating the DEAE-cellulose

chromatography step (Epps-KOH buffer), followed by a second

application on the affinity column. The final fold

purification was 1077 for the SP-I. Purification yields,

purities, specific activities and ratios of SDH/DQT

activities are shown in Table 4-1 and in Table 4-2.

Throughout the purification of SP-I and SP-II, the

ratio of shikimate dehydrogenase to dehydroquinase activity

was calculated to be within the range of 4.7 to 6.2 for SP-I

(at an average of about 5.1) and within a range of 4.4 to

5.4 for SP-II (at an average of about 4.8). Each protein was

concentrated on a PM-10 membrane to about 1 or 2 ml before

storage at -700 C.

Purification of ADH and PAT

When ADH was purified along with the SP-II, it was realized

that they eluted with overlapping activity peaks on the

Celite 545 column using the decreasing ammonium sulfate

gradient at about 25 to 20% of saturation as shown in Fig.

4-4. A critical step of separation occurred when SDH did

not bind to HA and ADH bound to HA (Fig. 4-5). Without the

inclusion of this step, the two enzymes co-purified

throughout the entire purification procedure, both eluting

Fig. 4-3. Elution profiles of S-proteins from an
NADP+ specific affinity column. A 2'5'-Sepharose-4B
column with a combined NADP+ (0-10 uM) and NaC1 (0-0.16M)
gradient was used as final steps of purification of the
S-proteins, each eluting at about 0.018M NaCl and luM
NADP. Both SDH (0) and DQT (0) activities coeluted. A)
step six of purification of SP-II protein. B) step eight
of purification of SP-I protein.

A STEP 6 + 0. 0.12
' SP-I z
40 4 z 0.06

0 10 20 30 40 50 60

360 B STEP 8

320 -
280 -

240 -

.I 200 -

160 -
120 0.18
120 -

8- 0.12
80 -
40 4 -0.06

0 10 20 30 40 50 60 70 80

Table 4-1. Purification of S1 protein from N.
silvestris suspension cultured cells.

Vol Protein Purity Yield Purity Yield Ratio
Step ml mg SA fold % SA fold % S/D

Crude 540 605 56 1 100 12 1 100 4.7
Celite 643 206 126 2 70 25 2 65 5.0
CM-52 720 49 412 7 55 94 8 59 4.4
DEAE-1 279 35 561 10 53 113 9 50 5.0
HA 618 8.7 1,515 27 36 257 22 28 5.9
DEAE-2 106 1.95 4,609 82 24 941 78 23 4.9
Aff-1 88 0.42 14,500 259 17 2,342 195 13 6.2
DEAE-3 83 0.16 33,878 605 15 7,365 614 15 4.6
Aff-2 54 0.07 65,877 1176 9 12,923 1077 12 5.1

S/D is shikimate dehydrogenase/dehydroquinase.

Table 4-2. Purification of S2 protein from N.
silvestris suspension cultured cells.

Vol Protein Purity Yield Purity Yield Ratio
Step ml mg SA fold % SA fold % S/D

Crude 540 605 5 1 100 1 1 100 4.6
Celite 348 97 26 5 68 5 5 64 5.0
CM-52 345 69 31 6 59 8 8 73 4.4
DEAE-1 77 31 66 13 55 14 14 53 5.1
HA 104 5 295 59 43 75 75 53 5.4
DEAE-2 31 1 757 151 34 187 187 34 4.1
Aff-1 15 0.006 4,590 918 7 880 880 7 5.2

S/D is shikimate dehydrogenase/dehydroquinate.

100 100

80 80

o S2-Protein
60 E 60

S40 -. 4 -.... 2.0
40 40 .0

20 20 1.0

0 260 280 300 320 340 360

Fig. 4-4. Celite 545 column chromatography overlap of
arogenate dehydrogenase and SP-II. Both SP-II (0) and ADH
(0) elute in the 45 to 0% ammonium sulfate gradient (solid
line) with overlapping activity peaks. Protein was
monitored at A2,, (...) .




40 -


0 20 40 60 80 100

Fig. 4-5. Hydroxylapatite column chromatography
separation of arogenate dehydrogenase and SP-II. The SDH
activity (M) was found in the wash fractions and ADH (0)
eluted at about 120 mM KPO4.



300 0




in similar fractions from the NADP* affinity column.

Therefore, when ADH was purified by the earlier mentioned

procedure an HA column step was included to completely

remove SDH activity. Purification of PAT followed the

previously demonstrated procedure with a purification of

about 250-fold.


The SP-I and SP-II proteins, well separated from each

other on the Celite 545 column, were each purified further

to about 1100- and 900-fold, respectively, by a series of

chromatographic steps. This complete separation was always

repeatable under similar conditions of the Celite 545 column

step (at least 12 times). SDH and DQT activities coeluted

at an activity ratio of about 5 throughout the entire

purification regimen for both bifunctional proteins. These

proteins may be isoenzymes from different compartments

chloroplastt and cytosol) or they may be isoenzymes residing

in different locations within the chloroplast. This second

possibility was reported for two DAHP synthase isoenzymes

(DS-1 and DS-2) presumably located in chloroplasts of

Arabidopsis (2). Further studies which may elucidate

differential properties of the two proteins and localization

studies should provide information with respect to possible

isoenzyme species and their location within the plant cell.


The purified SP-I and SP-II proteins were studied further to

characterize both the DQT and the SDH activities in (Chapter

V). The purified SP-I was used to produce specific antibody

against the protein (see Chapter VI) and for N-terminal

sequencing (see Chapter VII). Purification of PAT and ADH

was completed and then used to produce specific antibodies

to each protein for use as probes in cDNA cloning.



The main focus in my studies of the properties of the

two bifunctional proteins was to detect possible

differential properties which might suggest that these

proteins are isoenzymes. Molecular mass, pH and temperature

optima, thermal inactivation, and Km values for both

functional domains of the bifunctional proteins, SP-I and

SP-II, have been determined. Effects on activity of each

protein by various compounds, including PCMB, were also



Activity peaks on a Celite column

A Celite 545 column resolved about 6 peaks of SDH

activity when a more shallow gradient of ammonium sulfate

was applied than that used during the separation of the S-

proteins, Chapter IV (Fig. 5-1). Dehydroquinase activity

was coincident with SDH activities. Fractions of the six

peaks containing the enzyme activities were pooled

separately, avoiding overlapping fractions as much as

possible, and were labeled as shown in Fig. 5-1. The six

X:- L) 040
41 WU 0)

-4 J
-r- L O


~~40 4J
Lna) 4J (


rA r)-I 0)

-4l a) 14~-4
r-4 0>


L44 44C

4-1 r=o~ V

0-4 H)-


u-t4 j 0
::s4 4 4 )4
Ul 0

o a)~a,

CU (U )m

0 x r=0
LA 1 0) $.q :
V M4J M a)

44 04 r-4 r=
0 r

02 r-40 (
.rA 0E

00 4-I
a) id tLn
riri as

M~ E rI -

i C) (d

u k o





31 n -o LL
ic I


.UwLu nd (*) HOS cv

pooled samples containing DQT/SDH activities were

concentrated and examined for possible differential

properties, substrate requirements for saturation with SHK

and NADP', alternative substrate specificities (i.e., NAD+

and QA), temperature optimum, thermal inactivation, SDS PAGE

and activity-stained native PAGE. No significant

differences were found between the overlapping peak samples

of I to V (data is not shown). All six protein samples were

specific for NADP* and SHK when assayed for SDH activity.

Pool sample IV (or perhaps I through V) corresponds to SP-I,

and Peak VI corresponded to SP-II. SP-I and SP-II exhibited

similar properties except for M,, Km and activity stained


Molecular Mass

A difference was seen in the molecular mass of SP-I and

SP-II as shown in Fig. 5-2. SP-I migrated slightly ahead of

SP-II on SDS gels (A) and on a silver stained SDS gels (B).

Both gels showed bands (or set of bands) at about 59 kD and

40 kD for purified SP-I, and bands at about 61 kD for

purified SP-II. A band at 42 kD can be detected when higher

concentration of SP-II was used in these experiments. The

silver stained gel offered higher resolution and indicated

that the affinity purified SP-I also had a set of bands at

about 29 kD. Native Mr was determined on a Sephadex G-200

column,( Fig. 5-3, panels A and B), to be about 59 kD and 62

kD for SP-I and SP-II, respectively. In crude extract

Fig. 5-2. Molecular weight determinations by SDS
PAGE and by silver stain of the two S-proteins. A) The
SDS gel shows 2 bands for SP-I (584 ng) at molecular
weights of about 60,000 and 40,000 in lane 3. Lane 5 has
a single detectable band at a molecular weight of about
62,000 for the SP-II (155 ng). Lanes 1 and 7 are
molecular weight markers with MW as indicated. B) The
silver staining technique was also used to determine
molecular weights of the S-proteins. Lanes 1, 9, and 16
show molecular weight markers. Lanes 2 through 7
represent SP-I and lanes 11 through 14 represent SP-II at
various stages of purification as follows:

Lane Protein purification step Molecular weight(s)
2 S1 25ul Affinity pure 60,000 40,000
3 S1 10ul Affinity pure 60,000 40,000
4 S1 25ul DEAE-cellulose 60,000 40,000 29,000
5 S1 10ul DEAE-cellulose 60,000 40,000 29,000
6 S1 10ul Affinity pure (AB) 60,000 40,000 29,000
7 S1 25ul Affinity pure (AB) 60,000 40,000 29,000
11 S2 25ul Affinity pure (1) 62,000
12 S2 10ul Affinity pure (1) 62,000
13 S2 25ul Affinity pure (2) 62,000
14 S2 10ul Affinity pure (2) 62,000

(AB) is purified enzyme from several runs that were
combined, passed through the affinity column a third time
and concentrated. For SP-II the (1) represents the
purified protein after the first pass through the
affinity column and (2) represents combined column runs
of SP-II passed a second time through the affinity column
and concentrated.

1 3 5 7

-- 94,000

S 30,000

4 14.400

1 2 3 4 5 6 7 8 910111213141516





Fig. 5-3. Gel filtration chromatography of S-
proteins. A) Concentrated crude extract () pool IV
(SP-I) (0) and pool VI (SP-II) (A) (from the shallow
gradient Celite 545 column) were passed through a
Sephadex G-200 column. The dotted profile represents the
three marker enzymes (alcohol dehydrogenase, BSA and
carbonic anhydrase). B) A molecular weight curve based
on the column profile from A suggests the molecular
weights for the SP-I to be about 59,000 and about 62,000
for SP-II. Crude extract S-protein is suggested to be
about 63,000.


I *I I

0.15- Sephadex G-200 30

o 0.10- 20
0 / .20

0.05. 10

0 200 220 240 260 280 300 320 340 360 380 400 420
3 -

1i SP-II
-- 9
8 -
5 e/ --- SP-I
s S-Proteins



2 -

1 --- I ------------ 1 ----
100 200 300 400

(containing all fractions of the S-protein), the

bifunctional protein was determined to have only one native

molecular weight of about 63 kD.

Activity stained gels

In order to optimize conditions for a native PAGE

activity stain gel, various combinations of substrate and

dyes were studied with concentrated crude extract, data not

shown. All activity stained gels were assayed at optimal

conditions as described in Methods (Chapter II). Activity

stained bands required both substrate, SHK and NADP for

visibility. Crude extract revealed 3 bands on native gels

after application of the activity stain (Fig. 5-4). In Fig.

5-5, purified SP-I and SP-II had 2 activity stained bands

each. SP-II migrated at a faster rate than did SP-I.

pH and temperature optima of S-proteins

Both functional domains of the two bifunctional

proteins were studied with respect to pH optima (Fig. 5-6).

SDH activity was followed in a pH range of 6.5 to 10 in 200

mM Bis Trispropane buffer and showed optimal activity at pH

9.0 for both proteins. DQT, assayed by the coupled assay

method had very low activities when the same buffer was used

over the same pH range for both SP-I and SP-II. A pH range

from 6.1 to 9.5 in potassium phosphate, Epps or glycine

buffers, each at 100 mM resulted in higher activities and

showed optimal activity at pH 7.25 for both proteins. When

the three buffers above were used for the SDH assay there

Fig. 5-4. Optimal PAGE activity staining for SDH
in crude extract. An activity stain of a native
polyacrylimide gel was done at optimal conditions when
concentrated crude extract of 360 ug was applied to the
gel. Three bands are shown after incubation in
substrates at concentrations for optimal activity.

.r~~. 4.





!U:Ar T!;: -- :..

Fig. 5-5. PAGE activity staining of purified SP-I
and SP-II. SP-I was added to lanes 2 and 3, and SP-II
was added to lanes 1 and 4. Activity staining was done
under optimal conditions. A) After incubation for about
15 min, SP-I showed two bands of activity in both lanes.
SP-II, at lower concentrations, showed two bands in lane
4, each migrated faster than the SP-I bands. B) The gel
was incubated further for about 15 min. The SP-II in lane
1 was then visible. C) After incubation for 1 h, 2 bands
were more pronounced for all four lanes.