Louisiana State University
LSU Digital Commons
LSU Historical Dissertations and Theses
Graduate School
1998
Structural Characterization of Beta Carbonic
Anhydrases From Higher Plants.
Michael H. Bracey
Louisiana State University and Agricultural & Mechanical College
Follow this and additional works at: http://digitalcommons.lsu.edu/gradschool_disstheses
Recommended Citation
Bracey, Michael H., "Structural Characterization of Beta Carbonic Anhydrases From Higher Plants." (1998). LSU Historical
Dissertations and Theses. 6655.
http://digitalcommons.lsu.edu/gradschool_disstheses/6655
This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in
LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact
[email protected].
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type o f computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted.
Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted.
Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back o f the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zeeb Road, Arm Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
STRUCTURAL CHARACTERIZATION OF BETA
CARBONIC ANHYDRASES FROM HIGHER PLANTS
A Dissertation
Submitted to the Graduate Faculty o f the
Louisiana State University and
Agricultural and Mechanical College
in partial fulfillment of the
requirements for the degree of
Doctor of Philosophy
in
The Department o f Biological Sciences
by
Michael H. Bracey
B.A., Louisiana State University, 1991
May, 1998
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 9836855
UMI Microform 9836855
Copyright 1998, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zecb Road
Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
Without the contributions o f numerous people, none of this work would have
been possible.
Foremost, I offer my deepest gratitude and great respect to Dr. Sue
Bartlett who has not only proved a superb mentor but also wonderful friend. She has
provided me an environment with a rare combination o f high expectations, vast
intellectual resources, and a thoroughly enjoyable atmosphere which I doubt is surpassed
anywhere.
I have been told that one’s post-doctoral years are the best time o f a
scientist’s academic life. I will be astounded if any period in my scientific life proves
more enjoyable than my graduate training.
I also gratefully acknowledge the members o f my graduate committee: Simon
Chang, Patrick DiMario, Jim Moroney, and Mary Musgrave. Each one has thoughtfully
assisted me in reaching this stage, and I sincerely appreciate their many insights and
discussions.
Additionally, I would like to collectively thank the entire Biochemistry
faculty. Each one o f them has been kind enough to stop what they were doing on one
occasion or another to look for a book, answer a question, or clarify a subject on which I
was unclear. Their friendly open-door attitudes have made my education all the richer.
Past and present lab mates were always ready to lend a hand, an opinion, or a
joke.
These include Steve Pomarico, Carmen Dessauer, Hee Jin Kim, Pilar Tovar, Beth
Floyd, Wei Liu, Katrina Ramonell, and Xiaochun Xi.
I consider each o f them great
friends and look forward to being heckled at meetings. Big thanks also go out to all my
“non-lab” friends who patiently endured my prolonged absences and late arrivals. They
were also gracious enough to kidnap me from time to time.
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
On a more practical note, I acknowledge Joni Haigler and Robyn Migues for
helping with the zinc analysis described in Chapter 2, Jason Christiansen and Steve
Cramer for the EXAFS work, Ray Zielinski for sharing his barley cDNA library, Corey
Jones for constructing the E310A CA mutant, and Wei Liu for providing reagents while
in Stu Linn’s lab at Berkeley.
Finally, my ultimate gratitude goes to my family, especially my mom. Without
her unconditional love, support, and encouragement, I would be nothing resembling
what I am today.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS................................................................................................. ii
LIST OF TABLES.................................................................................................................... v
LIST OF FIGURES.................................................................................................................. vi
ABSTRACT............................................................................................................................. vii
CHAPTER
1
LITERATURE REVIEW..................................................................................1
2
THE ZINC CENTER IN SPINACH
CARBONIC ANHYDRASE.........................................................................17
3
SEQUENCE OF A cDNA ENCODING
CARBONIC ANHYDRASE FROM BARLEY.........................................40
4
A DETERMINANT OF QUATERNARY STRUCTURE
IN SPINACH CARBONIC ANHYDRASE.............................................. 45
5
CONCLUSIONS.............................................................................................61
APPENDICES
A
COPYRIGHT RELEASE FOR CHAPTER 2 .............................................64
B
COPYRIGHT RELEASE FOR CHAPTER 3.............................................65
VITA.......................................................................................................................................... 66
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 2.1 Activity o f and zinc binding to mutants
of spinach carbonic anhydrase....................................................................... 23
Table 2.2 Results o f EXAFS fitting analysis........................................................................29
Table 3.1 Characteristics o f the CA cDNA from barley.................................................. 42
Table 4.1 Interactions o f various CA constructs assayed in the two-hybrid system
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
LIST OF FIGURES
Figure 1.1 Structure o f human CA II....................................................................................... 5
Figure 1.2 Structure o f the Methanosarcina y carbonic anhydrase................................... 12
Figure 2.1 Homology among the plant and bacterial carbonic anhydrases............................. 22
Figure 2.2 Proteolytic susceptibility of selected mutant CAs...................................................25
Figure 2.3 Zn K-edge XAS spectrum....................................................................................... 27
Figure 2.4
Results ofEXAFS curve fitting analysis...............................................................28
Figure 3.1
Nucleotide sequence o f barley carbonic anhydrase
cDNA, NCBI accession number L36959..................................................... 41
Figure 4.1 Summary o f carbonic anhydrase constructs used
to make the two-hybrid fusions.................................................................... 49
Figure 4.2 Carboxy terminal sequence alignment o f
representative plant carbonic anhydrases..................................................... 51
Figure 4.3 Gel filtration sizing of spinach carbonic anhydrase
and the E310A mutant.................................................................................... 52
Figure 4.4 Tentative model of CA’s quaternary structure.................................................58
VI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
It is the goal o f this dissertation research to reveal some aspects of the physical
nature o f spinach carbonic anhydrase as a representative PCA using the techniques of
sequence comparison, molecular biology, and biophysics. Though both a and P carbonic
anhydrases are zinc dependent metalloenzymes, it is clear that the two isoforms do not
adopt the same mechanism for coordinating the active site metal. While aCA binds zinc
through three histidine ligands, PCA cannot due to a lack o f evolutionary conserved
histidines.
Instead, the P family has adopted a ligand scheme incorporating a single
histidine and two cysteines.
This has been determined by systematically mutating
possible zinc ligands in the spinach enzyme and then assaying the resulting variants for
stoichiometric metal binding. Additionally, this conclusion is corroborated by inspection
o f the wild type enzyme’s extended X-ray spectrum. This analysis indicates the metal is
surrounded by two sulfur atoms and two nitrogen or oxygen species.
Secondly, it has been long established that not only do the P isoforms differ from
their a cousins in their multimeric assembly, but subtypes exist within the P family in
which monocot forms assemble into lower molecular weight oligomers while dicot forms
assemble into higher order structures. In an attempt to gain insight into the differences
between monocot and dicot CAs, the CA cDNA from barley, a moncot, was sequenced.
Analysis o f the open reading frame revealed that the barley enzyme lacked ten amino
acids at the carboxyl terminus which are conserved in the dicot isozymes.
It is here
demonstrated that this extension contributes to the difference in multimeric organization
between monocots and dicots. When this extension is deleted from the spinach enzyme,
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the resulting mutant displays an apparent deficit in its ability to form higher order
multimers. Furthermore, this carboxyl extension will interact with the CA holoenzyme in
the yeast two-hybrid system showing that the observed characteristics o f the deletion
mutant do not arise from secondary disruptions, but rather the carboxyl terminus does
participate in intermolecular interactions.
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1
LITERATURE REVIEW
The carbonic anhydrases (CA) represent a diverse and widespread family of
enzymes found in all phylogenetic branches of life so far examined, except for the fungi
(Lindskog, 1997). Presently, this family is divided into three subgroups based on the
primary structure of the proteins, and these subgroups are designated as the a, P, and y
families (Hev/ett-Emmett & Tashian, 1996). Though these isoforms do not share any
appreciable sequence homology with one another, they do share some striking common
features. All CAs are zinc dependent metalloenzymes which catalyze the hydration of
carbon dioxide to form bicarbonate and a proton. In this reaction, a zinc bound water
molecule dissociates to generate a zinc bound hydroxyl and a free proton. This proton is
shuttled to the solvent and the remaining OH* attacks CO 2 to form bicarbonate. Once
this product leaves the active site and a new water molecule binds to the zinc, the
reaction starts over again. Thus, the role of zinc in the reaction scheme is to generate
reactive OH". (Band et al., 1990). Because the reaction converts gaseous CO2 to an ion
and generates H", CAs play a role in many systems to regulate pH and trap inorganic
carbon in the aqueous phase (Lindskog, 1997).
This enzyme family also displays
extraordinarily fast reaction kinetics, with k^t values o f 106 sec'1 for isozymes such as
human CA II (Behravan et al., 1990).
The a Carbonic Anhydrases
The aCAs are by far the most studied and best characterized o f the CA families.
These enzymes were formerly known as the mammalian or eukaryotic isozymes, as they
l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
are found in various tissues in all mammals examined.
Among vertebrates, seven
isozymes have been identified, and they are designated as CA I through CA VII. They
differ from one another in both their reaction kinetics and tissue distributions.
Some
members o f this family are membrane bound, but most aC A s are monomeric, soluble
proteins. Among these isozymes, the CAs are crudely divided into either the “high” or
“low” activity forms based on their kinetic profiles. Additionally, aCA s are seemingly
unique among the three families in that they display esterase activity in addition to their
CO 2 hydration catalysis (Holmes, 1977).
CA II is the most thoroughly studied of the aCAs.
It is widely distributed in
mammalian tissues and can be found in the cytoplasm o f epithelial tissues, red blood
cells, brain, eye, pancreas, kidney, liver, stomach, bone, and uterus (Tashian, 1989).
Among its ascribed physiological roles, CA II is believed to mediate the exchange o f
CO 2 in both the lungs and in capillary beds, buffer the plasma, acidify bodily fluids (such
as gastric juices and urine), and aid the reabsorption o f bicarbonate in the kidney. CA II
is classified as a high activity isozyme and has been compared to the low activity CA I in
attempts to elucidate the structural features which distinguish the two kinetic forms
(Behravan et al., 1990; Behravan et al., 1991).
CA I and CA III are both classified as low activity isozymes, and they are found
primarily in cytosols o f red cells and skeletal muscle, respectively. They display turnover
numbers between one fifth and one hundredth that of CA II (Behravan et al., 1991; Tu et
al., 1990).
These enzymes, like CA II, were thought to play important roles in the
exchange o f CO 2 from the blood to either tissues or the alveoli airspaces (Tashian,
1989). Contradictory to this notion is the observation that a human deficiency in CA I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
does not lead to any observable clinical pathology (Kendall & Tashian, 1977). It would
therefore seem that CA I serves some sort of backup role for CA II despite the fact that
it is the major non-hemoglobin protein in erythrocytes (Lindskog, 1997).
The remaining CA isozymes are less well characterized.
CA VII is a soluble
cytoplasmic enzyme. It has been found in salivary glands and was originally reported to
be membrane bound (Tashian, 1989).
It is presently believed to play a role in the
secretion o f bicarbonate into the saliva. A recombinant form of this enzyme has been
overexpressed, and it displays less than half of the activity o f CA II (Lakkis et al ., 1996).
CA VI is also synthesized in salivary glands but with a signal sequence which directs its
export from the cell. The enzyme itself is present in oral secretions where it is believed
to play a role in the maintenance o f pH (Lindskog, 1997).
CA IV is also secreted but remains membrane bound in the lung and kidney via a
phosphatidylinositol anchor (Zhu & Sly, 1990). It is reported to contain at least one
disulfide linkage necessary for enzymatic activity, and is devoid o f CMinked sugars. The
bovine form, however, purportedly bears five to six AMinked sugars in contrast to its
unadorned human counterpart (Zhu & Sly, 1990). These authors also cite possible roles
for CA IV including facilitating CO 2 efflux from alveoli and aiding bicarbonate
reabsorption in proximal tubules.
Lastly, CA V is a soluble isozyme found in the matrix of the mitochondria. This
localization is conferred by an amino terminal extension which directs its posttranslational transport into the organelle. CA V is presumed to provide bicarbonate as a
substrate for gluconeogenesis (Dodgson & Forster, 1986).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
In addition to the vertebrate isozymes, aC A s have been described in unicellular
species as well. Most notable among these is the photosynthetic alga Chlamydomonas
reinhardtii, which contains at least three a isozymes (Lindskog, 1997). Two o f these,
designated CAH1 and CAH2, are located in the periplasmic space and are believed to
facilitate the concentration of CO2 for photosynthetic carbon fixation (Fukuzawa et al.,
1990; Rawat & Moroney, 1991). In fact, CAH1 gene expression is rapidly upregulated
when cultures are shifted to low CO 2 tensions, and this response has been shown to be
part o f a complex C 0 2 concentrating mechanism designed to minimize photorespiration
(Fukuzawa et al.,
1990).
Curiously, the Chlamydomonas enzymes are post-
translationally processed into a large and a small subunit which remain associated via
disulfides in the tetrameric holoenzyme (Kamo et al., 1990).
Based on a variety of X-ray crystal structures and the extraordinary amino acid
identity among the vertebrate CAs, it is believed that all a isozymes possess nearly
identical tertiary structures.
Inspection of representative structures, such as that for
human CA II, reveal many features o f this enzyme family (Figure l.l) . The protein is
roughly spherical and rich in beta strands. Most o f these strands form a single twisting
sheet which dominates the enzyme’s overall structure. It is this sheet that forms the wall
of a funnel on the surface of the enzyme which leads down to the active site
approximately 15 A below the mean surface (Lindskog, 1997).
At the bottom o f this funnel sits the active site zinc atom coordinated to three
histidine residues. It is this environment which accounts for CA’s extraordinary kinetics
compared to zinc metal in an exposed aqueous environment (B and et al., 1990;
Hakansson et al., 1992). One side o f the cavity is hydrophilic and therefore attracts both
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
Figure 1.1 Structure o f human CA II. This structure was rendered with RasMol by
Roger Sayle using the coordinates 2cba deposited in the Protein Data Bank by
Hakansson et al. (1992).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
water and bicarbonate to facilitate their passage to and from the active site.
Additionally, on this wall sit the residues thought to play important roles in shuttling a
proton from the active site to the outside environment during the generation o f active
site hydroxide from a zinc bound water (Band et al., 1990). Conversely, the opposite
wall o f the funnel is rich in hydrophobic residues. This side is believed binding o f carbon
dioxide and positioning the molecule for attack during hydration (Hakansson et al.,
1992).
Along the funnel leading to the ocCA active site are a number o f residues which
are invariant, but still some are found to correlate with specific isozymes. In an effort to
understand these differences, various papers have appeared in which “wall mounted”
residues have been selectively targeted for mutagenesis in an attempt to re-engineer one
CA isozyme into another. These efforts have revealed the importance o f such residues
as threonine 200 in CA II and how it relates to this isozyme’s remarkable kinetics, while
at the same time demonstrating that apparent distinctions such as a conspicuous
asparagine to valine substitution at position 62 have no readily apparent consequences
(Behravan etal., 1990; Behravan et al., 1991).
The P Carbonic Anhydrases
The P family o f carbonic anhydrases is less well characterized than the a
isozymes. From a structural point of view, this is largely due to the lack o f an X-ray
crystal structure for this class. Formerly, this subtype was identified as the prokaryotic
or bacterial family, as representatives are found in both unicellular prokaryotes and the
chloroplasts o f higher plants. Based on sequence analysis and oligomeric assembly, the P
forms are subdivided into three subtypes: the dicot, monocot, and bacterial isozymes. All
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
three o f these share extensive sequence homology, but subtle differences can be
discerned (Hewett-Emmett & Tashian, 1996).
The bacterial class o f (3CAs is represented in E. coli and Synechococcus. The
cynT gene product in E. coli displays CO 2 hydration activity and is induced by growth in
the presence of cyanate. As a product o f a cyanate-inducible operon, cynT is thought to
to mediate theensure a ready supply of bicarbonate to be used as a substrate by cyanase
in the detoxification o f cyanate to ammonia and carbon dioxide (Guilloton et al., 1992).
The hydration of CO 2 by cynT would thus stimulate the forward reaction o f cyanase by
both supplying reactants and removing products simultaneously. Unlike the aCA s, cynT
exists as an oligomer, and the multimeric state of the enzyme is reportedly influenced by
the availability o f bicarbonate (Guilloton et al.,
1992).
During gel filtration
chromatography, the enzyme was found to shift to a lower molecular weight when the
column was developed in the presence o f bicarbonate.
The icfA gene product in Synechococcus PCC7942 is also a PCA. This gene was
identified in a screen for mutants which require elevated CO 2 concentrations to maintain
carbon fixation (Fukuzawa et al., 1992). Apparently, icfA, like the periplasmic CAHl of
Chlamydomonas, ensures a ready supply o f CO 2 for photosynthesis by this autotroph.
The remaining PCAs are divided between the monocots and the dicots. Among
the latter group, cDNAs encoding isozymes have been cloned from pea, spinach,
tobacco, aspen, Arabidopsis, and several species of Flaveria (Hewett-Emmett &
Tashian, 1996, Larsson et al., 1997). Additionally, evidence for CA protein has been
provided from parsley (Tobin, 1970). These isozymes generally display reaction kinetics
comparable to their a counterparts. For example, the spinach enzyme has a kc*t o f 2 x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
s
105 sec'1, equivalent to human CA I (Rowlett et al., 1994). However, 3CAs do display
altered susceptibility to aC A inhibitors such as the sulfonamides and heavy metals. For
example, spinach CA exhibits a K; for acetazolamide that is more than 103 times weaker
than aC A II (Pocker & Ng, 1974). This difference in inhibition may reflect a difference
in the active site architectures of members o f these two families.
The dicot isozymes are distinguished by their high native molecular weight. The
exact oligomeric makeup o f these holoenzymes has long been a point of contention.
Though they are clearly made up of identical subunits, different techniques have yielded
various native molecular weights, confounding attempts to identify how many subunits
make up the enzyme (Graham et al., 1984). Recently, studies on the pea enzyme using
bifunctional cross-linking agents strengthen the case for an octameric complex
(Bjorkbacka et al., 1997). It may be the case that PCA assumes a non-ideal shape, as
would be the case for the hexameric yCA (discussed below), and thus migrates
aberrantly through gel filtration columns.
Based on studies in Arabidopsis, dicot 3CAs may exist in two distinct
compartments, namely the chloroplast and the cytosol (Fett & Coleman, 1994). Two
distinct cDNAs have been cloned, and one o f them appears to lack a transit peptide.
Additionally, immunological evidence corroborates the existence o f an extrachloroplastic
PCA (Fett & Coleman, 1994). The implications o f this duplication are presently unclear.
Monocot isozymes have been characterized by conceptual translation o f cDNAs
cloned from species such as barley, maize, and rice (Hewett-Emmett & Tashian, 1996).
They are less well characterized than their dicot counterparts, but seem to demonstrate
comparable reaction kinetics (Atkins et al., 1972).
This class is largely distinguished
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
from the dicot isozymes by a comparably lower native molecular weight, and these
enzymes are thought to be dimers (Atkins et al., 1972; Graham et al., 1984).
Additionally, the Tradescantia enzyme displays discemibly different IC 50 values for a
number o f inhibitors when compared to pea. For example, the pea enzyme exhibits an
ICso for nitrate o f 38 pM, while for the Tradescantia CA this value is 175 jxM (Atkins
etal., 1972).
In maize, which engages C4 metabolism, two cDNAs have been cloned by
Burnell and deposited in the nucleotide databases (accession numbers U08403 and
U08401).
As in Arabidopsis, it seems that one o f these appears to encode a
chloroplastic enzyme since it contains an amino terminus similar to transit peptides while
the other gives rise to a cytoplasmic form. It can be speculated that the cytoplasmic
form resides in the mesophyll to provide bicarbonate for PEP carboxylase, but this
hypothesis has not been directly tested.
In higher plants, a role for the chloroplast CA is presently unknown. It has been
traditionally presumed that CA would serve in a capacity similar to the isozymes of
Synechococcns and Chlamydomonas.
That is, chloroplast CA could serve to
concentrate
for the enzyme
the
availability
carboxylase/oxygenase
of CO2
(Rubisco)
and
thus
reduce
ribulose-l,5-bisphosphate
the
metabolic
costs
of
photorespiration. This hypothesis has been partially bome out by the observation that
carbonic anhydrase levels seem to be co-regulated with Rubisco (Hudson et al., 1992).
However, demonstrating this role conclusively has proved difficult.
Coleman used an antisense transgenic approach to reduce the levels of CA in
tobacco by 99% (Majeau et al., 1994). However, he reported no decrease in net CO 2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
fixation by these lines, leading one to conclude that CA does not facilitate carbon
fixation.
Similar results were also reported by Price et al. (1994).
In contrast, Kim
(1997) used the same technique to reduce the levels o f CA in Arabidopsis and achieved
very different results. He found that with a 90% reduction in chloroplast CA activity,
plants displayed no morphological deficits when maintained on tissue culture media
supplemented with sucrose.
However, if sucrose was withdrawn from the media, the
antisense plants either showed a marked decline or died. This lethality was furthermore
shown to be rescuable in the presence o f elevated levels o f carbon dioxide.
These
results, in stark contrast to those of the Price group and the Coleman group, betray a
necessary role for CA in photosynthetic carbon fixation.
The y Carbonic Anhydrases
The yCAs are the most recently recognized members o f the CA family.
Presently, a yCA has only been definitively observed in a member o f the Archaea, though
homologous sequences and expressed sequence tags have been found in such organisms
as Arabidopsis and Synechococcns (Hewett-Emmett & Tashian, 1996).
The enzyme
was isolated from Methanosarcina thermophila and shown to catalyze the hydration of
CO2 (Alber & Ferry, 1994).
This isozyme, like the Chlamydomonas aC A CAH1, is
secreted from the methanogen, but it is expressed in response to nutritional limitations
rather than low CO2 tensions. If the microbe is grown in media containing methanol, it
reduces the methyl group to methane to derive metabolic energy. However, if cultures
are shifted to acetate, Methanosarcina oxidizes the carbonyl group o f acetate to carbon
dioxide to derive energy to reduce the remaining methyl group to methane. It is this
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
resulting CO 2 that is likely acted upon by the yCA, possibly as part of a
bicarbonate/acetate antiport system (Alber & Ferry, 1994).
The Methanosarcina yCA has been crystallized to reveal a rather unusual
structure (Figure 1.2; Kisker et al., 1996). The molecule, like aCA, is dominated by P
strands. However, unlike aCA, the strands o f yCA do not form conventional sheets, but
rather form a three sided, triangular left-handed helix (Figure 1.2). The helix proceeds
through seven to eight turns to form the long axis o f the molecule which has the overall
shape o f a tube. Flanking the P helix is a single a helix which extends approximately the
same length.
To form the holoenzyme, these beautiful subunits assemble into a homotrimer
with parallel orientation.
Based on crystallographic observations, these trimers may
dimerize in a head to head fashion to form a hexamer. In this arrangement, two trimers
become bridged by a P barrel formed by the six amino terminal extensions o f each
subunit. However, it is possible that this further oligomerization is an artifact, as sizing
o f yCA using the analytical ultracentrifuge indicates that it exists in solution as a trimer
(Kisker et al., 1996). When viewed along the long axis of the molecule, the trimer itself
forms a shape resembling a Star o f David with the longitudinal a helix o f each monomer
forming the vertices o f one triangle while the P helices form a second inverted one
(Figure 1.2). The active site zinc atom is coordinated by three histidine ligands which are
virtually superimposable upon the histidines of the active site o f representative aCAs
(Lindskog, 1997). However, unlike the aCAs, the zinc ligands o f the yCA are located at
the subunit interfaces of the trimer. Among the three histidine ligands at each active site,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
A.
B.
Figure 1.2 Structure o f the Methanosarcina y carbonic anhydrase. A. The yCA
monomer viewed perpendicular to the long axis o f the 0 helix. B. The yCA trimer
viewed down the long axis of the 0 helix showing the three fold symmetry o f the
holoenzyme. This structure was rendered with RasMol by Roger Sayle using the
coordinates lthj deposited in the Protein Data Bank by Kisker et al. (1996).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
one donor is supplied by one monomer while the remaining two are supplied by an
adjacent monomer (Kisker et al., 1996). It is thus obvious that multimeric assembly is
required before the yCAs can carry out catalysis.
Overview
In the absence o f a crystal structure for a PCA, structural information for this
class is severely lacking.
It is thus difficult to perform rational mutagenic studies to
elucidate such features as the catalytic mechanism and the nature o f the oligomeric
assembly. Nonetheless, comparisons o f primary structures o f members o f the P family
can offer insights into the structure o f these enzymes. The present work takes advantage
of this approach in an attempt to gain useful information concerning the coordination of
the active site zinc and the formation o f quaternary structure.
References
Alber, B. E. & Ferry, J. G. (1994) A carbonic anhydrase from the archaeon Methanosarcina
thermophila. Proc. Nat. Acad Sci. USA 91: 6909.
Atkins C. A., Patterson B. D. & Graham D. (1972) Plant carbonic anhydrases II:
Preparation and some properties o f monocotyledon and dicotyledon enzyme
types. Plant Physiol. 50: 218-223.
Banci, L., Bertini, I., Luchinat, C. & Moratal, J. M. (1990) The mechanism o f action o f
carbonic anhydrase. Enzymatic and Model Carboxylation and Reduction
Reactions for Carbon Dioxide Utilization. Aresta, M. & Schloss, J. V., eds.
Kluwer Academic Publishers.
Behravan, G., Jonsson, B. -H. & Lindskog, S. (1990) Fine tuning o f the catalytic
properties o f carbonic anhydrase. Studies o f a Thr200 —►His variant of human
isoenzyme II. Ear. J. Biochem. 190: 351.
Behravan, G., Jonasson, P., Jonsson, B. -H. & Lindskog, S. (1991) Structural and
functional differences between carbonic anhydrase isoenzymes I and II as studied
by site-directed mutagenesis. Eur. J. Biochem. 198: 589.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Bjorkbacka, H., Johansson, I. -M., Skarfstad, E., & Foreman, C. (1997) The sulfhydryl
groups o f cys 269 and cys 272 are critical for the oligomeric state of chloroplast
carbonic anhydrase from Pisitm sativum. Biochemistry 36: 4287.
Dodgson, S. J. & Forster, R. E. (1986) Inhibition o f CA V decreases glucose synthesis
from pyruvate. Arch. Biochem. Biophys. 251: 198.
Fett, J. P. & Coleman, J. R. (1994) Characterization and expression o f two cDNAs
encoding carbonic anhydrase in Arabidopsis thaliana. Plant Physiol. 105: 707.
Fukuzawa, H., Fujiwara, S., Yamamoto, Y., Dionisio-Sese, M. L. & Miyachi, S. (1990)
cDNA cloning, sequence, and expression o f carbonic anhydrase in
Chlamydomonas reinhardtir. regulation by environmental CO 2 concentration.
Proc. Nat. Acad. Sci. USA 87: 4383.
Fukuzawa, H., Suzuki, E., Komukai, Y., & Miyachi, S. (1992) A gene homologous to
chloroplastic carbonic anhydrase {icfA) is essential to photosynthetic carbon dioxide
fixation by Synechococcus PCC7942. Proc. Natl. Acad Sci. U. S. A. 89: 4437.
Guilloton, M. B., Korte, J. J., Lamblin, Fuchs J. A. & Anderson, P. M. (1992) Carbonic
anhydrase in Escherichia coli. J. Biol. Chem. 267: 3731.
Graham, D., Reed, M. L., Patterson, B. D., Hockley, D. G., & Dwyer, M. R (1984)
Chemical properties, distribution, and physiology o f plant and animal carbonic
anhydrases. Ann. N. Y. Acad Sci. 429: 222.
Hakansson, K., Carlsson, M., & Svensson, L. A (1992) Structure of native and apo carbonic
anhydrase II and structure of some of its anion-ligand complexes. J. Mol. Biol. 227:
1192.
Hewett-Emmett, D. & Tashian, R. E. (1996) Functional diversity, conservation, and
convergence in the evolution o f the a-, 3-, and y-carbonic anhydrase gene
families. Molec. Phylog. Evol. 5: 50.
Holmes, R. S. (1977) Purification, molecular properties and ontogeny o f carbonic
anhydrase isozymes. Eur. J. Biochem. 78: 511.
Hudson, G. S., Evans, J. R., von Caemmerer, S., Arvidsson, Y. B. C. & Andrews, T. J.
(1992) Reduction o f ribulose-l,5-bisphosphate carboxylase/oxygenase content by
antisense RNA reduces photosynthesis in transgenic tobacco. Plant Physiol. 98:
294.
Kamo, T, Shimogawara, K., Fukuzawa, H., Muto, S. & Miyachi, S. (1990) Subunit
constitution of carbonic anhydrase from Chlamydomonas reinhardtii. Eur. J.
Biochem. 192: 557.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
Kendall, A. G. & Tashian, R. E. (1977) Erythrocyte carbonic anhydrase I: inherited
deficiency in humans. Science 197: 471.
Kim, H. J. (1997) Identification of the physiological role o f carbonic anhydrase using the
antisense technique and investigation of its transcriptional regulation in
Arabidopsis thaliana (L.) Heynh. Ph.D. dissertation, Louisiana State University,
Baton Rouge, Louisiana.
Kisker, C., Schindelin, H., Alber, B. E., Ferry, J. G. & Rees, D. C. (1996) A left-handed
3-helix revealed by the crystal structure of a carbonic anhydrase from the archeon
Methanosarcina thermophila. EMBO J. 15:2323.
Lakkis, M. M., Bergenhem, N. C. H. & Tashian, R. E. (1996) Expression o f mouse
carbonic anhydrase VII in E. coli and demonstration o f its CO 2 hydrase activity.
Biochem. Biophys. Res. Commnn. 226: 268.
Larsson, S., Bjorckbacka, H., Forsman, C., Samuelsson, G. & Olsson, O. (1997)
Molecular cloning and biochemical characterization o f carbonic anhydrase from
Populus tremulax tremuloides. Plant Mol. Biol. 34: 583.
Lindskog, S. (1997) Structure and mechanism o f carbonic anhydrase. Pharmacol. Ther.
74: 1.
Majeau, N., Amoldo, M. A. & Coleman, J. R. (1994) Modification o f carbonic
anhydrase activity by antisense and over-expression constructs in transgenic
tobacco. Plant Mol. Biol. 25: 377.
Pocker, Y. & Ng, J. S. Y. (1974) Plant carbonic anhydrase.
reversible inhibition. Biochemistry 13: 5116.
Hydrase activity and its
Price, G. D., Caemmerer, S. V., Evans, J. R., Yu, J., Lloyd, J., Oja, V., Kell, P.,
Harrison, K., Gallagher & A., Badger, M. R. (1994) Specific reduction of
chloroplast carbonic anhydrase activity by antisense RNA in transgenic tobacco
plants has a minor effect on photosynthetic CO2 assimilation. Planta 193: 331.
Rawat, M. & Moroney, J. V. (1991) Partial characterization of a new isoenzyme of
carbonic anhydrase isolated from Chlamydomonas reinhardtii. J. Biol. Chem.
266: 9719.
Rowlett, R. S., Chance, M. R., Wirt, M. D., Sidelinger, D. E., Royal, J. R., Woodroofe,
M., Wang, Y. F., Saha, R. P. & Lam, M. G. (1994) Kinetic and structural
characterization o f spinach carbonic anhydrase. Biochemistry 33: 13967.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
Tashian, R. (1989) The carbonic anhydrases: widening perspectives on their evolution,
expression and function. BioEssciys 10: 186.
Tobin, A. J. (1970) Carbonic anhydrase from parsley leaves. J. Biol. Chem. 245: 2656.
Tu, C., Paranawithana, S. R., Jewell, D. A., Tanhauser, S. M., LoGrasso, P. V., Wynns,
G. C., Laipis, P. J. & Silverman, D. N. (1990) Buffer enhancement of proton
transfer in catalysis by human carbonic anhydrase III. Biochemistry 29: 6400.
Zhu, L. Z. & Sly, W. S. (1990) Carbonic anhydrase IV from human lung. J. Biol. Chem.
265: 8795.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2
THE ZINC CENTER IN SPINACH CARBONIC ANHYDRASE'
Introduction
Carbonic anhydrase (CA; carbonate dehydratase, EC 4.2.1.1) is a ubiquitous zinc
metalloenzyme which catalyzes the reversible hydration of carbon dioxide. In mammals, CA
plays important roles in facilitating carbon dioxide exchange in capillary beds and alveoli,
maintaining the buffering capacity of blood, and reabsorbing bicarbonate across renal tubules
(Tashian, 1989). Crystal structures for the human isozymes I and II and the bovine isozymes
II and HI have been solved, and they show that these enzymes coordinate an active site zinc
through three conserved histidine residues (for example, see Hakansson etal., 1992).
In higher plants which carry out C3 photosynthesis, the majority o f CA activity can be
localized to the chloroplast stroma where the enzyme's role is unclear, although it may serve
to
concentrate
carbon
dioxide
at
the
active
site
of
ribulose-l,5-bisphosphate
carboxylase/oxygenase (Graham et al., 1984). CA from higher plants is quite different from
the major mammalian isozymes in both primary sequence and multimeric assembly. CA from
C3 dicotyledonous plants is a hexamer with one zinc atom per monomer (Graham et al'.,
1984; Kisel & Graf, 1972) while the major mammalian isozymes are monomeric (Tashian,
1989). Furthermore, sequence analysis reveals no homology between the plant and animal
CA’s, and this suggests that the animal and plant isozymes do not share a common
evolutionary origin (Fukuzawa et al., 1992) and therefore may not share common physical
properties.
Reprinted by permission o f Biochemistry.
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
Until a crystal structure becomes available for a plant-type CA, biochemical
investigations may provide hints o f structural characteristics o f this enzyme. To this end, we
have employed techniques of molecular biology and biophysics to investigate the nature of
the zinc binding site in spinach carbonic anhydrase. We present here results of experiments
utilizing site-directed mutagenesis, elemental analysis, and EXAFS spectroscopy.
Experimental Procedures
Cloning and mutagenesis.
All DNA manipulations including plasmid isolation,
restriction digestions, ligations, and transformations were performed using standard methods
(Sambrook et al., 1989). Site specific mutations were introduced into CA using the Altered
Sites Mutagenesis System (Promega). Mutagenesis changing histidine to glutamine, cysteines
to alanines, glutamates to glutamines, and aspartate to asparagine resulted in the mutants
H210Q, C150A, C213A, E194Q, E266Q, and D152N.
Mutations were confirmed by
sequencing according to a modified dideoxy chain termination method (Fawcett & Bartlett,
1990).
Mutant constructs were cloned into a derivative of the expression vector pKK233-2
(Pharmacia LKB Biotechnology, Inc.) and expressed in E. co/i DH5a for activity
measurements. To facilitate purification, these CA's were further subcloned into a derivative
o f the expression vector RIT2T (Pharmacia LKB Biotechnology, Inc.) to yield constructs,
termed pPAxCA, which upon translation would result in carbonic anhydrase fused to the
carboxyl terminus o f Staphylococcus protein A
Protein isolation and assays. Protein was isolated from the pKK expression system
for activity measurements as follows. E. colt D H 5a harboring the appropriate CA construct
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
or the vector alone was grown at 37° in LB media to an OD<soo o f 1.8. Cells were pelleted
and resuspended in Buffer A (50 mM Tris-HCL, pH 8, 100 mM NaCl, and 1 mM EDTA).
After lysing by sonication, the resulting homogenate was clarified and brought to 56%
saturation with solid ammonium sulfate. Protein was allowed to precipitate for at least one
hour on ice and then pelleted at 32,000x g for 15 minutes.
The resulting pellets were
resuspended and dialyzed overnight at 4° against Buffer A.
Each dialyzed sample was quantitated with the Coomassie protein assay (Pierce)
using BSA as a standard. The presence of recombinant CA in each of the samples was
verified by Western blotting.
Activity measurements of the crude CA preparations were determined according to
the spectrophotometric method of Khalifah (1971). Briefly, 0.5 ml o f an appropriate buflferindicator pair was placed in a cuvette, followed by the addition o f CA extract. The reaction
was initiated by the addition of CO2 saturated water, and the change in absorbance of the
indicator was recorded as a function o f time. We assayed in 25 mM EPPS pH 8 with 0.113
mM phenol red monitored at 555 nm and 25 mM imidazole pH 8 with 0.072 mM pnitrophenol monitored at 400 nm. These buffer-indicator pairs were chosen based on their
similar pK* values.
For each buffer system, an acid calibration curve was used to calculate the buffer
factor O which was then used to convert dA/dt to mmol CO2 sec'1 mg'1. Initial rates of CO2
hydration were determined by extrapolating back to time zero from a linear sampling time.
The hydration rate measured using protein derived from cells harboring the empty pKK233-2
vector was considered spontaneous and subtracted as uncatalyzed background.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
Ellman's assays were performed essentially as described (Riddles etal ., 1983).
Inductively coupled plasma atomic emission spectroscopy (ICP-AESV For the metaJ
analysis, fusion proteins were expressed in E. coli and purified by affinity chromatography
using non-specific IgG linked to Reacti-Gel 6x (Pierce). Purified protein A/CA fusions were
freed of loosely bound zinc using Chelex 100 (Bio Rad). As negative controls, buffer alone
or an amount of BSA comparable to the amount of CA fusion assayed was treated similarly.
Each sample was analyzed for zinc concentration using an ARL 34000 inductively coupled
plasma atomic emission spectrometer and electronics and software from Labco, Inc. For final
determination of zinc binding, readings for the zinc-free BSA standard were treated as
negligible background due to density and viscosity differences within the plasma torch as
compared to the buffer control and were subtracted from readings for the CA fusions.
EXAFS. Wild-type protein A/CA fusion was purified as above from 50 g cell paste.
Protein was concentrated using Minicon-B15 clinical sample concentrators (Amicon) to a
final concentration of ~2 mM and supplemented with 30% ethylene glycol. An 80 pi sample
was loaded into a lucite cell with a 1 mm pathlength and sealed with 0.001" Kapton tape. Xray absorption spectra were recorded at the National Synchrotron Light Source, Brookhaven
National Laboratory, on beamline X10-C. The beamline was run in focused mode with an
Si(l 11) double crystal monochromator configuration. Higher order harmonics were rejected
using a mirror position feedback system (Sansone et a l 1991). Frozen samples were loaded
into an Oxford Instruments liquid helium flow cryostat maintained at -10K. XAFS data were
collected in fluorescence mode using a Canberra Industries 13-element Ge solid-state array
detector (Cramer et al., 1988) while incident beam intensity was monitored with a nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
filled ion chamber. Photon energy was calibrated by simultaneously collecting a transmission
spectrum o f a zinc metal foil and setting the first inflection point energy to be 9659.0 eV.
EXAFS oscillations were extracted from the raw data by routine methods (Cramer et
a l, 1978) and were then quantitatively analyzed using a Levenberg-Marquadt nonlinear leastsquares calculated curve-fitting procedure to minimize differences between the data and
observed EXAFS. Simulations were derived from the curved-wave functional form:
=
sin[2*R, + <#,(*,*,)]
(McKale et al., 1986). For this analysis, theoretical values for both phase and amplitude were
used, and the value g was fixed at 0.9 for all fits.
During fitting, the total Zn coordination
was set to four or five and small changes in the threshold energy (DEo) were fixed at -4.2 eV
(Hubbard et al., 1991), while the interatomic distance (R) and the mean square deviation o f R
(s2) were allowed to vary.
Results
Sequences for the E. coli CA homologue (Sung & Fuchs, 1988), the Synechococcus
CA (Fukuzawa et al., 1992), and CA’s from Arabidopsis (Raines et al., 1992), pea (Roeske
& Ogren, 1990), spinach (Fawcett et al., 1990), tobacco (Majeau & Coleman, 1992), and
barley were analyzed for conservation of the amino acids whose side chains are known to
serve as zinc ligands at enzyme active sites (Vallee & Auld, 1990). Homology alignments
showed that one histidine, two cysteines, two glutamates, and one aspartate are conserved
among the plant-type CA’s (Figure 2.1). Mutagenesis changing these six residues resulted in
mutant forms which displayed altered catalytic activity (Table 2.1).
Western blotting of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
S p in a c h
P ea
T o b acco
A. c h a lia n a
E. c o ll
S y r.ech o .
S p m a c r.
P ea
T o b acco
o h a : lar.a
col 1
S y r.e c h c.
A.
£.
ACSDSRYCPSnVLCFQPGEAFMVRNIANMVPVF DKDKYAGVGAAIEYAVLHLKVENIWI
ACSESR7C?SHYl.IjFQPGEAFVYRHVANLVP?YDQAKYAGTGAAJEYAVI.HLKVSNIW I
AC3DSRVC?SriYI.NFQ?GEAFWRf;iANMVFAY'DKTRYSGVGAAI£YA7T.HLKVENI VVI
ACSDSRYCPSHVLDFQPGDAFVYRNIANMVPPFDKVKYGGVGAAIEYAVLHLKVENIWI
SCSDS.RLVPEL77QREPGDLFVIRNAGNIVPSYGPEP-GGVSASVEYAVAAI.RVSDIVIC
7C S D S R I □PNLITQSGMGELFVIRNA.GNLI PPFG A A N -G G EG A SIE Y A IA A LN IEH W V C
GHSACGG'KGLMSFPCAGPTTTDFIEDW'/KICI.PAKHKVLAilHGNATFAEQCrHCEKEAV
GHSACGGIKGLLSFPFDGTYSTEFIEEWV KIGLPAKAKVKAQHGDAPFAXLCTKCEKEAV
GHSACGGIEGLMSLPADGSESTAFIEDWYKIGLPAKAKVQGEHVDKCFADQCTACEKEAV
GHSACGGIKGLM SFPLDGMM STDFlEDW'/KICLrAKSK'/ISELGDSAFEDQCGRCEREAV
GKSNCGAMTAIAS-C-QCMDHMPAYSHWLRYADSA-RWNEARPHSDLPSKAAAMVRENV
GHSHCGAMKGLLKLM-QTQEDMPLVYDWLQHAQATRRL'/LDNYSGYETDDLVEILVAEfJV
Figure 2.1 Homology among the plant and bacterial carbonic anhydrases. Alignment is
shown of the portion of the plant-type CA’s spanning the six conserved potential zinc ligands
from spinach, pea, tobacco, Arabidopsis, E. coli, and Synechococcus. The alignment begins
with amino acid 149 in the spinach sequence as deposited in GenBank. Asterisks indicate
potential ligands to zinc. Bold asterisks indicate potential zinc ligands that, when mutated,
resulted in CA that bound zinc poorly. Alignment generated by PILEUP o f the UWGCG
programs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
Table 2.1 Activity of and zinc binding to mutants of spinach carbonic anhydrase.
MUTANT
ACTIVITY
mol Zn /
mol CA
monomer
EPPS
imidazole
WT
100
100
1.09
C150A
ND
9
0.03
D152N
1
9
0.81
E194Q
22
5
0.80
H210Q
10
6
ND
C213A
6
5
0.15
E266Q
81
100
1.25
Activities are expressed as percentages of the wild-type enzyme in each buffer system.
The blank-corrected activity of the wild-type was 6.8 mmol C 0 2 sec'1 mg'1 E. coli protein
in EPPS buffer and 76 mmol C 02 sec'1 mg'1 E. coli protein in imidazole. Reported
activities represent an average of four to five independent assays from two preparations.
The assay was limited by an uncatalyzed rate of hydration of 7.9 mmol C 02 sec'1 mg'1 E.
coli protein in EPPS buffer and 32 mmol C 0 2 sec'1 mg'1 E. coli protein in imidazole.
ND, Not detectable.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
lysates from E. coli expressing each of the mutant proteins revealed that all were expressed in
soluble form. (Data not shown.)
The six spinach CA mutants and the wild-type enzyme were also analyzed for their
ability to bind zinc. In order to facilitate purification of the proteins, the mutant forms as well
as wild-type CA’s were expressed in K coli as C-terminal fusions to protein A The wildtype CA fusion protein was catalytically active, indicating that the protein A moiety does not
significantly interfere with proper folding or function of the CA portion o f the fusion.
After
purification, the fusion proteins were depleted of adventitious zinc and analyzed by
inductively coupled plasma atomic emission spectroscopy. Wild-type CA bound zinc in a
ratio of one atom per subunit. However, the mutants C150A, H210Q, and C213A, had
greatly diminished capacities to bind zinc (Table 2.1).
Two of the latter mutant CA’s were more susceptible to proteolysis than the wildtype protein when expressed in E. coli, and during affinity chromatography various
breakdown products co-purify. These breakdown products are evident in Figure 2.2 and are
shown with the wild-type enzyme and two stable mutants for comparison. These products
likely represent various C-terminal truncations that are still able to interact with the IgG
column through the protein A domain at the amino terminus. Regardless, even if intact fusion
represented only 35% o f the total protein assayed in the case of C l 50A and only 10% in the
case o f H210Q, ratios o f zinc to CA monomer would still be less than 0.3 based on backcalculations from the ICP-AES data.
The instability o f the cysteine mutant C150A is not due to the disruption of a disulfide
bond involving this residue. Ellman's assays of the wild-type fusion protein revealed the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
M 1 2 3 4 5
97.4
66.2 - >
42.7 -►
Figure 2.2 Proteolytic susceptibility of selected mutant CAs. SDS-PAGE of the purified CA
fusions assayed for zinc content showing the instability of two mutant proteins that bind zinc
poorly. Lane 1, WT; lane 2, H199Q; lane 3, H210Q; lane 4, C150A; lane 5, C213A.
Molecular weight markers are indicated. The arrowhead at right indicates the protein A/CA
fusion bands.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
presence o f two reduced cysteines per monomer in the native state and all six cysteines per
monomer in the presence o f 6M guanidinium chloride. This indicates that no disulfides exist
in the bacterially expressed spinach CA, though four o f the six cysteines present per monomer
are inaccessible to solvent in the native state.
To further investigate the nature o f the ligands in the wild-type CA fusion protein, we
examined its Zn X-ray absorption spectrum. The fluorescence-detected Zn X-edge spectrum
is shown in Figure 2.3. The CA Zn EXAFS is relatively strong and without a clear beat
pattern (Figure 2.4A). The Fourier transform is dominated by a single peak centered at ~2.3
A with a few minor peaks above the noise between 3-4 A (Figure 2.4B). The pattern is
similar to the transforms observed in EXAFS studies of plastocyanin and other blue Cu
proteins (Scott et al., 1982).
The dominant feature could be simulated by a Zn-S interaction at ~2.3 A, but
additional Zn-N/O interactions near 2 A were necessary to get a good fit. The small features
present beyond the central 2.3 A peak are most probably due to multiple scattering
interactions from an imidazole group and some contributions from the carbons of the
cysteines. Unfortunately, with little knowledge of the symmetry and geometry o f the site,
attempts to fit these features with multiple scattering have not been successful. The quality of
fit for all possible combinations of sulfur and nitrogen/oxygen interactions was judged by
comparing the fit index between the resulting fit and the raw extracted EXAFS, and these
results are summarized in Table 2.2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
200
150
eo
a
50
10000
9500
10500
Entrgy («V)
p " o .e ta ^ C A f e to n edS e “
SPeC,rUm
abS° rp tio n * « * * m ° f t t e
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
2
6
4
2
a.
-4
-6
•8
2
A.
4
6
B
M*1)'
0
10
12
0
1
2
3
4
5
6
R(A)
B.
Figure 2.4 Results of EXAFS curve fitting analysis. A Experimental k3 EXAFS (solid line)
and final fit results for 2 Zn-S — 2 Zn-N/O coordination to Zn (dashed line). B. Fourier
transform of EXAFS (solid line) with final fit results (dashed line).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
Table 2.2 Results of EXAFS fitting analysis.
Model
N
R(A )
a 2 (A2 x 10°)
Fit Index (F)
Tetracoordinate
Zn-S
4
2.31
6.29
314
Zn-S
Zn-N/O
3
1
2.32
2.04
4.08
1.35
212
Zn-S
Zn-N/O
2
2
2.32
2.06
2.03
3.02
171
Zn-S
Zn-N/O
1
3
2.32
2.08
1.00
5.09
255
Zn-N/O
4
2.11
4.90
777
Pentacoordinate
Zn-S
Zn-N/O
2
2.32
2.37
3
2.06
5.63
143
F = [ £(x<.-xJ2k6 ]/n where the difference is between each data point of the experimental
and simulated (xJ EXAFS, and n is the number o f data points in the fitting range.
All fits performed on range k = 2-13 A '1. During fitting, the total Zn coordination was
set to four or five and small changes in the threshold energy (AEJ were fixed at -4.2 eV
(Hubbard et al., 1991), while the interatomic distance (R) and the mean square deviation
of R (cr) were allowed to vary.
(X j
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
Discussion
The plant-type carbonic anhydrases have long been overshadowed by the animal
isozymes, but the former are emerging as rather intriguing enzymes in their own right.
Though it is unlikely that the animal and plant-type CA’s are evolutionarily related
(Fukuzawa et al., 1992), these two groups share many common features. Both are zinc
metalloenzymes which catalyze the hydration of carbon dioxide with remarkable efficiency
and are susceptible to a number of common inhibitors. Among these, the sulfonamides inhibit
the animal isoforms by binding to the catalytic zinc (Vidgren et al., 1990). By analogy, since
sulfonamides also inhibit plant-type CA’s (Pocker & Ng, 1974) it is presumed that the zinc
bound to the plant enzyme is also catalytic in nature. However, in contrast to the animal
enzyme, plant CA’s do not possess three conserved histidines to account for zinc
coordination consistent with the animal model.
Therefore, the two isoforms may have
evolved to coordinate the active site metal by two very different strategies.
To investigate how spinach CA might coordinate zinc, we analyzed available
sequences o f plant-type CA’s for the conservation of all amino acids whose side chains are
known to serve as ligands to active site zincs, namely histidine, aspartate, glutamate, and
cysteine (Vallee & Auld, 1990). Six such conservations were identified and each residue was
targeted for mutagenesis (Figure 2.1). Among these, the aspartate, histidine, and cysteine
mutants each exhibited less than 10% of the activity of the wild-type enzyme when assayed in
EPPS buffer (Table 2.1).
Mutations in the carbonic anhydrase from pea at locations equivalent to those
described here result in similar enzyme performance (Provart et al., 1993).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pea mutants
31
corresponding to the spinach H210Q, C150A, and C213A (H220N, C160S, and C223S,
respectively) had no measurable activity, reinforcing our results with the spinach enzyme.
Additionally, Coleman’s group found the pea E204A to exhibit no activity and E276A to be
rather compromised. However, we find the equivalent, more conservative spinach mutant
E266Q to display near wild-type activity, while E194Q is 22% as active as the wild-type
enzyme. We attribute these differences of results to Coleman’s more severe charged to
aliphatic mutations which may serve to disrupt the pea enzyme more than the conservative
spinach mutation.
Since Coleman’s group did not construct a mutation comparable to
D152N, we are unable to compare our results regarding this variant.
Nonetheless, a
duplication o f enzymatic trends among various mutants from two species provides further
evidence that H210, C 150, and C213 are critical residues in spinach C A
In previous studies on other zinc proteins, site-directed mutagenesis o f potential zinc
ligands followed by the loss of function of the metalloprotein in question has been provided as
evidence implicating residues in metal coordination.
This has been demonstrated for
Ieukotriene A hydrolase (Medina et al., 1991), neutral endopeptidase (LeMoual et al.,
1993), and a potential zinc finger in the glucocorticoid receptor (Seveme et al., 1988). We
have shown here the same correlation in spinach C A of six conserved potential zinc ligands,
four can be mutated to cause a 90% loss of function (Table 2.1). As the zinc o f CA is
presumably catalytic in nature, the loss of the enzyme's ability to coordinate this metal would
necessarily result in a concomitant loss of activity. Thus, based on our activity assays, the
conserved histidine, cysteines, and aspartate are all candidates for zinc ligands in spinach
carbonic anhydrase.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
Additionally, we have definitively ascertained which of these mutations affect the
enzyme's ability to bind zinc rather than affect catalysis through some secondary disruption.
Such a disruption could conceivably occur by various means, for example by abolishing a
hydrogen bond network within the active site as suggested for the pea mutant E276A
(Provart et al., 1993). Through elemental analysis we have established that the histidine and
cysteine mutants uniquely exhibit severely diminished capacities to bind zinc (Table 2.1). This
correlation between a loss of activity and loss of zinc binding has also been shown for the zinc
ligand mutant H94D in human CA II (Kiefer et al., 1993) and mutants o f leukotriene A*
hydrolase (Medina et al., 1991). We interpret this as additional proof indicating that these
three residues are involved in the coordination of the zinc ion, while the aspartate may play an
essential role unrelated to metal coordination.
While the ICP-AES data do not implicate the glutamate and aspartate mutants as
potential zinc ligands, these variants do demonstrate significant catalytic disruptions.
To
investigate if these residues might be involved in a proton transfer step within the active site
as suggested for various amino acids in the animal isozymes, we also assayed in an imidazole
buffer system. As a relatively small buffering molecule, imidazole is thought to complement
the catalytic deficiency of proton shuttle mutants by entering the active site and essentially
replacing the missing functional group.
This property has been exploited in the animal
systems to identify proton shuttling amino acids such as R67 in human CA EH (Tu et al.,
1990). We find that imidazole does indeed restore the activity of E266Q to wild-type levels,
implicating this residue as a possible proton shuttle in the active site. This conclusion has also
been reached for the equivalent pea CA mutation (Provart et al., 1993). The remaining
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
mutants do not exhibit a stimulatory imidazole effect. It is interesting that the activity of
E194Q actually decreases relative to the wild-type enzyme in the imidazole system (Table
2.1). The reasons for this decline are not immediately obvious.
An increased susceptibility to endogenous E. coli proteases during the purification of
C150A and H210Q is demonstrated in Figure 2.2, and this may represent an instability o f the
mutant proteins.
Similar instability has been documented in other systems where zinc
metalloenzymes have been depleted o f zinc and assayed for resistance to trypsin proteolysis.
In the case of the DNA binding domain of the GAL4 transcription factor, the zinc-free
apoenzyme was reduced to peptides upon trypsinization while the zinc-bound form was
cleaved to a 13 kDa core particle (Pan & Coleman, 1989). Similar results were also obtained
with the Gene 32 protein from bacteriophage T4 (Giedroc et al'., 1987).
Furthermore,
differential scanning calorimetry investigations o f the folding processes o f the Gene 32 protein
as well as alkaline phosphatase from E. coli showed metal dependent changes in the
denaturation profiles of these zinc metalloproteins (Keating et al., 1988; Chlebowski &
Mabrey, 1977).
These results imply that the binding of the metal imparts significant
stabilization to these proteins at the level of tertiary structure, or, as suggested for aspartate
carbamoyltransferase, at the level o f quaternary structure (Monaco et al., 1978).
These characteristics may also hold true for spinach carbonic anhydrase. Possibly, the
formation of the active site is necessary for the proper folding of CA monomers; the inability
to bind the metal could disturb elements of the protein's secondary or tertiary structure,
exposing protease sensitive domains. Alternatively, the active sites o f spinach CA may be
located at subunit interfaces and zinc could stabilize the quaternary structure o f the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
holoenzyme.
Though the zinc o f CA is presumably catalytic in nature, this established
correlation o f protein stability and metal binding is a possible explanation for the susceptibility
to proteolysis in our CA mutant proteins.
To further investigate the zinc binding site in wild-type spinach carbonic anhydrase,
we analyzed this enzyme's extended X-ray absorption fine structure.
EXAFS has the
advantage o f providing a direct examination of a metal's environment in metalloproteins.
Analysis of the spectrum o f spinach CA yields evidence that the nearest neighbors o f the zinc
are sulfur and nitrogen/oxygen species. Of the five possible combinations of two distinct ZnS and Zn-N/O interactions in a four ligand system, 2 Zn-S - 2 Zn-N/O gives the best fit, in
agreement with the results o f the mutagenesis. This model represents a 20% improvement in
fit index over the next lowest value (Table 2.2). It is clear that a model with four N/O ligands
coordinated to Zn, as found in mammalian CA, is not consistent with the EXAFS since the
corresponding fit is of much worse quality than the others performed on this system.
Attempts to fit a longer N/O shell at ~2.3 A were also unsuccessful, yielding fit indices - 1 0 4
or causing the second Zn-N/O distance to contract back to ~2.1 A , giving the same result as
the 4 N/O model.
Furthermore, the presence of only two conserved cysteines per
polypeptide in the plant-type CA’s is consistent with our 2 Zn-S - 2 Zn-N/O EXAFS model
and renders the 3 Zn-S - 1 Zn-N/O and 4 Zn-S - 0 Zn-N/O models improbable.
Since four separate CA mutants yielded enzymes with activities of 10% or less, we
also examined the EXAFS data for the possibility of a pentacoordinate zinc as implicated for
adenosine deaminase (Wilson et al., 1991; Bhaumik et al., 1993). The fit index resulting
from the addition o f a third N/O ligand at -2.0 A for a 2 Zn-S - 3 Zn-N/O system does yield a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
16% improvement over the 2 Zn-S -
2
Zn-N/O model. However, several lines o f evidence
discredit this five ligand model. First, the fit indices for the best tetracoordinate model and
the pentacoordiante model lie within the usual error range placed on values o f coordination
derived using the EXAFS technique. Therefore, it is objectively impossible to favor one
model or the other based purely on EXAFS results. Second, when a tetracoordinate zinc is
converted to a pentacoordinate one in a synthetic system, the bond length o f the leaving
group is longer than the four remaining ligand bond lengths (Auf der Heyde & Nassimbeni,
1984). We can find no such assymetry in the CA pentacoordinate model; all N/O bond
lengths remain the same.
Third, only three of our potential ligand mutants exhibit a
diminished capacity for zinc binding. If we define a ligand as a residue that contributes to the
binding of a metal, then the EXAFS data may indicate that D 152 is a near neighbor of the
metal, while the ICP-AES data indicate that this residue is not a ligand per se. Overall, we
find the tetracoordinate system to be a more credible model.
Monozinc enzymes may be classified into families based on the spacing of zinc
ligands and the conservation of amino acids adjacent to them (Vallee & Auld, 1990). A
typical catalytic zinc-binding site in this nomenclature is made up of two closely spaced
ligands Li and L2 , which comprise a zinc binding nucleus separated by an amino acid spacer
X\ plus a third, distant L 3 separated by a spacer Y. In the case o f spinach CA, H210 and
C213 may represent Li and L2, respectively, while C l50 may represent L3, with X=2 and
F=59. A preliminary search o f the SwissProt database for homology with the amino acid
motif encompassing the Lr L 2 nucleus revealed no significant homology with any known zinc
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
metalloenzymes. Thus, the plant-type carbonic anhydrases may represent a novel family of
zinc metalloproteins.
In summary, site-directed mutagenesis of potential zinc ligands in spinach carbonic
anhydrase revealed that mutant proteins in which cysteine 150 was converted to alanine,
aspartate 152 was converted to asparagine, histidine 210 was converted to glutamine, or
cysteine 213 was converted to alanine exhibited severely diminished catalytic activity.
Furthermore, the cysteine and histidine mutants had greatly diminished capacities to bind zinc.
In addition, EXAFS data are consistent with a proposed metal coordination by two sulfiirs,
while the bond lengths suggest tetracoordination, presumably by additional nitrogen or
oxygen donor ligands.
The residues corresponding to cysteine 150, histidine 210, and
cysteine 213 are conserved in all plant-type CA’s sequenced to date, and so our results
suggest that the side chains of these amino acids are the protein ligands to the active site Zn,
while a reactive water molecule may be inferred to complete the coordination sphere. With
this ligand scheme, spinach CA joins alcohol dehydrogenase and cytidine deaminase as the
third zinc metalloenzyme in which a catalytic zinc is coordinated by sulfur ligands (Vallee &
Auld, 1990; Betts et al., 1994). Though a definitive resolution of the role of the conserved
aspartate will await the elucidation of a CA crystal structure or a detailed enzymological
study ofD152 mutants, the plant-type CA’s seem to differ from the mammalian isozymes not
only in primary sequence and quaternary structure, but also in the nature o f the side chains
responsible for binding the catalytic zinc.
References
Auf der Heyde, T. P. E. & Nassimbeni, L. R. (1984) Reaction pathways from structural data:
Dynamic stereochemistry of Zn(II) compounds. Acta Cryst. B40: 582.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
Betts, L., Xiang, S., Short, S. A_, Wolfenden, R , & Carter, C. W. Jr. (1994) Cytidine
deaminase. The 2.3 A crystal structure o f an enzyme : transition-state analog
complex. J. Mol. Biol. 235: 635.
Bhaumik, D., Medin, J., Gathy, K., & Coleman, M. S. (1993) Mutational analysis of active
site residues o f human adenosine deaminase. J. Biol. Chem. 268: 5464.
Chlebowski, J. F., & Mabrey, S. (1977) Differential scanning calorimetry of apo-,
apophosphoryl, and metalloalkaline phosphatases. J. Biol. Chem. 252: 7042.
Cramer, S. P., Tench, O., Yocum, M. & George, G. N. (1988) A 13-element Ge detector for
fluorescence EXAFS. Nucl. Instrum. Methods A 266: 586.
Fawcett, T. W., & Bartlett, S. G. (1990) An effective method for eliminating “artifact
banding” when sequencing double stranded DNA templates. BioTechniques 9:
46.
Fawcett, T.W., Browse, J. A , Volokita, M., & Bartlett, S. G. (1990) Spinach carbonic
anhydrase primary structure deduced from the sequence o f a cDNA clone. J.
Biol. Chem. 265: 5414.
Fukuzawa, H., Suzuki, E., Komukai, Y., & Miyachi, S. (1992) A gene homologous to
chloroplastic carbonic anhydrase (icfA) is essential to photosynthetic carbon dioxide
fixation by Synechococcus PCC7942. Proc. Natl. Acad. Sci. U. S. A. 89: 4437.
Giedroc. D. P., Keating, K. M., Williams, K. R , & Coleman, J. E. (1987) The function of
zinc in gene 32 protein from T4. Biochemistry 26: 5251.
Graham, D., Reed, M. L., Patterson, B. D., Hockley, D. G., & Dwyer, M. R (1984)
Chemical properties, distribution, and physiology of plant and animal carbonic
anhydrases. Ann. N.Y. Acad Sci. 429: 222.
Hakansson, K , Carlsson, M., & Svensson, L. A (1992) Structure o f native and apo carbonic
anhydrase II and structure o f some of its anion-ligand complexes. J. Mol. Biol. 227:
1192.
Hubbard, S. R , Bishop, W. R , Kirschmeier, P., George, S. J., Cramer, S. P., &
Hendrickson, W. A. (1991) Identification and characterization o f zinc binding sites in
protein kinase C. Science 254: 1776.
Keating, K. M., Ghosaini, L. R , Giedroc, D. P., Williams, K. R , Coleman, J. E., &
Sturtevant, J. M. (1988) Thermal denaturation of T4 gene 32 protein: Effects o f zinc
removal and substitution. Biochemistry 27: 5240.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
Khalifah, R. G. (1971) The carbon dioxide hydration activity o f carbonic anhydrase. J. Biol.
Chem. 246: 2561.
Kiefer, L. L., Ippolito, J. A., Fierke, C. A_, & Christianson, D. W. (1993) Redesigning the
zinc binding site of human carbonic anhydrase II: Structure of a His2 Asp-Zn2* metal
coordination polyhedron. J. Am. Chem. Soc. 115: 12581.
Kisel, W., & Graf, G. (1972) Purification and characterization of carbonic anhydrase from
Pisiim sativum. Phytochemistry 11: 113 .
Le Moual, H., Roques, B. P., Crine, P., & Boileau, G. (1993) Substitution o f potential metalcoordinating amino acid residues in the zinc-binding site of endopeptidase-24.11.
FEBS Lett. 324: 196.
McKale, A. G., Knapp, G. S., & Chan, S.- K. (1986) Practical method for full curved-wave
theory analysis o f experimental extended x-ray-absorption fine structure. Phys. Rev.
B 33: 841.
Majeau, N. & Coleman, J. R (1992) Nucleotide sequence of a complementary DNA
encoding tobacco chloroplastic carbonic anhydrase. Plant Physiol. 100: 1077.
Medina, J. F., Wetterholm, A., Radmark, O., Shapiro, R., Haeggstrom, J. Z., Vallee, B. L., &
Samuelsson, B. (1991) Leukotriene A» hydrolase: Determination o f the three zincbinding ligands by site directed mutagenesis and zinc analysis. Proc. Nat. Acad. Sci.
USA 88: 7620.
Monaco, H. L., Crawford, J. L., & Lipscomb, W. N. (1978) Three-dimensional structures of
aspartate carbomoyltransferase from Escherichia coli and of its complex with
cytidine triphosphate. Proc. Nat. Acad Sci. USA 75: 5276.
Pan, T., & Coleman, J. E. (1989) Structure and function of the Zn(II) binding site within the
DNA-binding domain of the GAL4 transcription factor. Proc. Nat. Acad Sci. USA
86: 3145.
Pocker, Y. & Ng, J S. Y. (1974) Plant carbonic anhydrase.
reversible inhibition. Biochemistry 13: 5116.
Hydrase activity and its
Provart, N. J., Majeau, N., & Coleman, J. R (1993) Characterization of pea chloroplastic
carbonic anhydrase. Expression in Escherichia coli and site-directed mutagenesis.
Plant Molec. Biol. 22: 937.
Raines, C.A., Horsnell, P. R., Holder, C., & Lloyd, J. C. (1992) Arabidopsis thaliana
carbonic anhydrase: cDNA sequence and effect o f CO2 on mRNA levels. Plant
Molec. Biol. 20: 143.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
Riddles, P. W., Blakely, R. L., & Zemer, B. (1983) Reassessment of Ellman’s reagent.
Methods in Erizymology 91:49.
Roeske, C. A., & Ogren, W. L. (1990) Nucleotide sequence of pea cDNA encoding
chloroplast carbonic anhydrase. Nucleic Acids Res. 18:3413.
Sambrook, J., Fritsch, E. F., & Maniatis, T. (1989) Molecular Cloning: A Laboratory
Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.
Sansone, M., Via, G., George, G., Meitzner, G., Hewitt, R., & Marsch, J. (1991) in X-ray
Absorption Fine Structure (Hasnain, S. S., Ed.) pp. 656-658, Ellis Horwood, Ltd.,
W. Sussex, England.
Scott, R. A , Hahn, J. E., Doniach, S., Freeman, H. C., & Hodgson, K. 0. (1982) Polarized
x-ray absorption spectra o f oriented plastocyanin single crystals. Investigation of
methionine-copper coordination. J. Am. Chem. Soc. 104: 5364.
Seveme, Y., Wieland, S., Schaflher, W., & Rusconi, S. (1988) Metal binding ‘finger5
structures in the glucocorticoid receptor defined by site-directed mutagenesis.
EMBOJ. 7: 2503.
Sung, Y. C. & Fuchs, J. A. (1988) Characterization of the cyn operon in Escherichia coli
K12. J. Biol. Chem. 263: 14769.
Tashian, R. E. (1989) The carbonic anhydrases: Widening perspectives on their evolution,
expression and function. BioEssays 10: 186.
Tu, C., Paranawithana, S., Jewell, D. A., Tanhauser, S. M., LoGrasso, P. V., Wynns, G. C.,
Laipis, P. J., & Silverman, D. N. (1990) Buffer enhancement of proton transfer in
catalysis by human carbonic anhydrase EQ. Biochemistry 29: 6400.
Vallee, B. L., & Auld, D. S. (1990) Zinc coordination, function, and structure of zinc
enzymes and other proteins. Biochemistry 29: 5647.
Vidgren, J., Liljas, A., & Walker, N. P. C. (1990) Refined structure of the acetazolamide
complex o f human carbonic anhydrase II at 1.9 A. Int. J. Biol. Macromol. 12: 342.
Wilson D. K., Rudolph, F. B. & Quiocho, F. A (1991) Atomic structure of adenosine
deaminase complexed with a transition state analog: Understanding catalysis and
immunodeficiency mutations. Science 252: 1278.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C H A PTER3
SEQUENCE OF A cDNA ENCODING
CARBONIC ANHYDRASE FROM BARLEY’
Introduction
Carbonic anhvdrase (CA) catalyzes the reversible hydration o f CO 2 to generate
bicarbonate and a proton.
Though its physiological function has not been clearly
defined, it had been speculated that CA serves to concentrate CO 2 at the active site o f
ribulose-l,5-bisphosphate carboxylase/oxygenase (Graham et al., 1984).
However,
recent work with transgenic plants expressing CA in the antisense orientation has shown
that the enzyme does not play an essential role in carbon fixation, though the regulation
o f CA levels seems to parallel that of carboxylase (Majeau and Coleman, 1994; Majeau
et al., 1994; Price et al., 1994). Currently, several sequences are available for CAs from
C3 dicots (for a recent listing, see Bracey et al., 1994 and references therein); we present
here what we believe to be the first available sequence of a CA from a C 3 monocot.
Results and Discussion
The cDNA that we have isolated is 1530 base pairs long (Figure 3.1).
Its
essential features are summarized in Table 3.1. Based on a comparison with the dicot
CAs, we predict that the barley clone contains 246 base pairs of 5’ non-translated
sequence and 311 base pairs o f 3’ non-translated sequence. Interestingly, we isolated
two phage clones harboring identical CA inserts with different polyadenylation sites, one
beginning at base 1516 and one beginning at base 1505.
A potential polyadenylation
sequence, AATATAA, occurs at base 1472.
Reprinted by permission o f Plant Physiology.
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
G A A 7TCCT7T
--
TTTCGCTGCT
ACTACTAACA
TTGCATAAAT
CATACTGCTG
■j C A C G C A o T G A T C A A G T A G T A C A G T C G T G C A C C A G C A C C C C A T G A A A T T G
101
rT A C G T T T G T
CACTGTTTGT
TGGTAGAGTA
GACT GTAGAC
ACGCAAATAA
* C, 1
GGTACGTGCA
ACGGGCAGGA
AGCAAGACGT
GGGTACGAGA
CGCTGCAACG
:o i
CCGTGAGACC
AACACGCGGT
GCCAGCTCCT
CCTGCGCACA
GACCCGATGT
^c :
CGTTGCAGAT
TGGGCGGACA
GAGAGGGCCC
GGTCCCCGGT
CTTTGTCTTT
301
GCACACAAGC
GGCAACTGCT
CCATGGACGG
TGTAGTACCA
TCGACAATGC
351
AAATT GCAGC
ACCTGCAGCA
TGAAAATCAA
TAGCACTTGT
ACATTGACGG
•id
OGGTGCCGAT
TGCCGCACTG
CCTGGGCCAC
GTACTACCTC
A CA C T A CT CG
■5 = 1
ACCGCGGCGG
CTAACTGGTG
CTA CGCA A CC
GTCGCGCCCC
GTGCCCGCTC
= ■ ::
CTCCACCATC
GCCGCCAGCC
TCGGCACCCC
CGCGCCCTCC
TCCTCCGCCT
=. c •
CCTTCCGCCC
CAAGCTCATC
AGGACCACCC
CCGTCCAGGC
CGCGCCCGTC
sc:
G CACCTGCAT
TGATGGACGC
CGCCGTGGAG
CGCCT CAAGA
CCGGGTTCGA
~1
GAAGTTCAAG
A C C G A .G G T C T
ACGACAAGAA
GCCCGATTTC
TTCGAGCCGC
'01
TCAAGGCCGG
CCAGGCGCCC
A AGTACATGG
TGTTCGCGTG
CGCCGACTCG
'5 1
GGTGTGTGCC
CGTCGGTCAC
CCTGGGCCTT
GAGCCCGGTG
AGGCCTTCAC
6 01
CATCCG CAAC
ATCGCCAACA
TGGTCCCGGC
CTACTGCAAG
AACAAGTACG
6 51
O C G G G G T T lsG
ATCGGCCATC
GAATACGCCG
TCTGCGCGCT
CAAGGTTGAG
■5Cl
j
. G A T C jT G G
TGATTGGCCA
CAGCCGCTGC
G GTGGAATCA
AGGCTCTGCT
951
CTCGCTCAAG
GAT GGCGCAG
ACGACTCCTT
CCACTTCGTT
GAGGACTGGG
C 01
TCAGGATCGG
GTTCCCGGCC
AAGAAGAAGG
TGCAGACTGA
GTGCGCCTCC
A T G A C C A .C -T G
CACCGTCCTG
GAGAAGGAGG
CCGTCAACGT
6
051
101
GTCCCTCCAG
A A C C T 'C T T G A
CCTACCCGTT
CGTCAAGGAG
GGTGTGACCA
151
AGGGAACCCT
CAAGCTCGTC
GGCGGCCACT
ACGACTTCGT
CTCCGGCAAG
0 01
TT C 3A A A C A 7
j
GGAGCAGTA A A TC TTC CC C
A C C G G T T .A A C
TCCGACATAT
•
ACAAACGTAC
ATATATCAAG
ATATCGTCCG
ATCGATGTGA
AT GCAATGCC
301
.*\T G G G A G T G C
GTACCCGTTA
TTGTCCAGTA
C T G G A .T G C C G
GATGGCCCGA
351
: g tg a a tttg
CCATAAGCAA
TAGAACCTTT
TTTTCTTCAC
CATTTTCTGA
4 01
C G A G G A A .T T G
TACTGCTATG
TGATGCATAA
TTTGATCGTC
TTGTGATCGA
4z
AAGACAT CAT
ATATAAGTTT
A ATATAATAT
TTTCATGAAC
CGTTTACCTT
TTTATTACCC
TTTCA A A AA A
AAAAAAAAA
c.
«
501
Figure 3.1 Nucleotide sequence of barley carbonic anhydrase cDNA, NCBI accession
number L36959. The predicted translation start and stop sites are shown in bold at
positions 247 and 1219, respectively
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
Table 3.1 Characteristics o f the CA cDNA from barley.
Species:
Hordeum vitlgare L.
Gene:
Chromosomal location unknown
Subcellular location:
Chloroplast stroma
Predicted DNA fragments:
5’ flanking region, 246 bp; coding region, 972 bp; 3’ flanking region, 311 bp
Techniques:
XgtlO cDNA library (Ling and Zielinski, 1989) screening using spinach CA cDNA
(Fawcett et al., 1990), cloning into pBluescript vector (Stratagene), restriction
fragment subcloning, and modified Sequenase v2.0 (USB) sequencing o f both
strands (Fawcett and Bartlett, 1990)
Predicted features o f the protein:
The deduced protein sequence consists o f 324 amino acids with a calculated
molecular weight o f 35, 074. The amino terminus is rich in hydroxylated and
positively charged amino acids which likely confer chloroplast localization.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
In the past, comparisons of CAs of both monocot and dicot origin suggested that
these two families would be quite different. The two classes have different quaternary
structure, respond differently to inhibitors, exhibit different stability, and display different
reaction kinetics (Atkins et al., 1972). Despite these observations, the deduced protein
sequence o f barley CA bears remarkable resemblance to the dicot isoforms. When barley
CA was compared to spinach CA using the GCG comparison program BESTFIT, the
predicted mature proteins exhibited 60% identity and 74% similarity if conservative
substitutions are taken into account.
One immediately obvious difference between the inferred barley protein sequence
and the dicot CAs is an apparent truncation at the monocot carboxyl terminus, which
lacks a conserved ten amino acid “tail” present in all dicots so far examined. We are
presently examining this region to investigate its possible role in higher oligomerization
states among the dicot isoforms.
References
Atkins, C. A., Patterson, B. D. & Graham D. (1972) Plant carbonic anhydrases II:
Preparation and some properties of monocotyledon and dicotyledon enzyme
types. Plant Physiol. 50: 218-223.
Bracey, M. H., Christiansen, J., Tovar, P., Cramer, S. P. & Bartlett, S. G. (1994)
Spinach carbonic anhydrase: Investigation o f the zinc-binding ligands by sitedirected mutagenesis, elemental analysis, and EXAFS. Biochemistry 33: 1312413131.
Fawcett, T. W. & Bartlett, S. G. (1990) An effective method for eliminating “artifact
banding” when sequencing double stranded DNA templates. Biotechniques 9:
46-48.
Fawcett, T. W., Browse, J. A., Volokita, M. & Bartlett, S. G. (1990) Spinach carbonic
anhydrase primary structure deduced from the sequence o f a cDNA clone. J.
Biol. Chem. 265: 5414-5417.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
Graham, D., Reed, M. L., Patterson, B. D., Hockley, D. G. & Dwyer, M. R. (1984)
Chemical properties, distribution and physiology o f plant and animal carbonic
anhydrases. Ann. N.Y. Acad. Sci. 429: 222-237.
Ling, V. & Zielinski, R. E. (1989) Cloning o f cDNA sequences encoding the calciumbinding protein, calmodulin, from barley (Hordeum vulgare L.) Plant Physiol.
90: 714-719.
Majeau, N., Amoldo, M. A. & Coleman, J. R. (1994) Modification o f carbonic
anhydrase activity by antisense and over-expression constructs in transgenic
tobacco. Plant Mol. Biol. 25: 377-385.
Majeau, N. & Coleman, J. R. (1994) Correlation o f carbonic anhydrase and ribulose-1,5bisphosphate carboxylase/oxygenase expression in pea. Plant Physiol. 104:
1393-1399.
Price, G. D., von Caemmerer, S., Evans, J. R., Yu, J. W., Lloyd, J., Oja, V., Kell, P.,
Harrison, K., Gallagher, A. & Badger, M. R. (1994) Specific reduction of
chloroplast carbonic anhydrase activity by antisense RNA in transgenic tobacco
plants has a minor effect on photsynthetic CO 2 assimilation. Planta 193: 331340.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4
A DETERMINANT OF QUATERNARY
STRUCTURE IN SPINACH CARBONIC ANHYDRASE
Introduction
Unlike the majority of a carbonic anhydrases, the |3 forms exist as oligomers
(Graham et al., 1984).
The oligomeric states o f a number o f pCAs have been
investigated over the years, and differing results have been reported concerning their
exact quaternary structure.
For example, dicot CAs have been reported as both
hexamers and as octamers when examined by different groups using different techniques
(Graham et al., 1984).
It is currently unclear whether this discrepancy reflects a
misinterpretation o f data, an in vitro artifact, or a bona fide plasticity o f the
oligomerization process. One possible explanation is that CA assumes a non-ideal three
dimensional shape which results in aberrant migration through gel filtration columns.
Accordingly, results obtained by equilibrium sedimentation or cross-linking may be the
most reliable estimates of CA’s exact quaternary structure, and data from the latter
technique suggest that the native enzyme is an octamer (Bjorkbacka et al., 1997).
Thus,
apparent hexamers that elute from gel filtration columns are likely, in fact, octamers.
Nonetheless, gel filtration is still useful for determining the relative populations of
different oligomeric states.
One observation concerning the oligomerization o f PCAs that is agreed upon,
however, is the fact that isozymes derived from dicots differ markedly from those
derived from monocots. For example, the PCAs from spinach, pea, and parsley have all
been reported as hexamers or octamers while the PCAs from barley, wheat, and
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
Tradescantia have been reported as trimers or dimers (Graham et al., 1984; Tobin,
1970; Atkins et al., 1972). This distinction between these two closely related families of
the PCAs led us to examine their primary structures for possible motifs which might
explain this difference in multimeric assembly.
One obvious difference between the
monocot and dicot isozymes is the apparent truncation in the monocot forms o f a ten
amino acid carboxy terminal extension. We first observed this difference upon
characterizing the barley cDNA, which at the time was the only available monocot
sequence (Chapter 3). As additional monocot sequence information became available for
CAs from rice (accession number U08404) and maize (U08403, U08401), it became
clear that this distinction was not a feature unique to barley. In the studies reported here,
we examined the effects o f deletion of this '‘tail” on the oligomeric state of the spinach
enzyme.
Further, we showed that the carboxyl terminus can mediate protein-protein
interactions within the PCA oligomer and propose a speculative model that unifies our
observations and those of Bjorbacka et al. (1997).
Experim ental Procedures
Cloning and mutagenesis.
All DNA manipulations, cloning, and expression of
proteins were performed using standard methods (Sambrook et al., 1989; Bracey et al.,
1994). Site directed mutations were introduced into spinach carbonic anhydrase using
the Altered Sites Mutagenesis System (Promega). The mutant C213A alters an inferred
zinc ligand and has been described previously (Bracey et al., 1994; Chapter 2). The
mutant E310A was generated by engineering an ochre stop codon in the place of
glutamate at amino acid 310, resulting in a carboxy terminal truncation lacking ten amino
acids. Qualitatively, this truncation did not alter catalytic activity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
To analyze protein-protein interactions among the various CA constructs, we
subcloned appropriate fragments into the two-hybrid vectors pAS2-l and pACT2
(Clontech). Constructs expressing the full length mature spinach CA from an engineered
N col site at amino acid 78 were cloned into the two-hybrid vectors to yield the
constructs pAS-CA and pACT-CA.
To accomplish this, pPAxCA (Chapter 2) was
digested with Ncol and Nsil, and the resulting insert was cloned into the NcoI-PstI sites
o f pAS2-l to yield pAS-CA.
Additionally, pKKCA78 (Fawcett et al. 1990) was
digested with Ncol and EcoRI and the resulting insert was ligated into the corresponding
sites in pACT2 to generate pACT-CA.
The two mutants, E310A and C213A, were
cloned into the two-hybrid vectors by starting with constructs in the Pharmacia vector
pKK-233. These plasmids were digested with Ncol and Nsil and the resulting inserts
were either cloned directly into the NcoI-PstI sites in pAS2-1 or into the NcoI-PstI sites
o f LITMUS29 (New England Biolabs).
The LITMUS constructs were then digested
with Ncol and EcoRi and the resulting restriction fragment ligated into the
corresponding sites of pACT2.
This cloning strategy generated the constructs pAS-
E310A, pACT- E310A, pAS-C213A, and pACT-C2l3A.
To express the carboxy
terminal domain alone as GAL4 two-hybrid fusions, an Ncol site was introduced at
amino acid 309 using the Altered Sites Mutagenesis System and the vector pSelect
(Promega). This mutant was then digested with Ncol and EcoRi, and the resulting insert
was subcloned into both pAS2-1 and pACT2 digested with the same restriction enzymes.
This strategy resulted in the expression o f a fragment encoding the ten amino acid
carboxy terminus o f CA in the two-hybrid vectors to yield the fusion constructs pAS-tail
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
and pACT-tail. A summary o f these constructs is depicted in Figure 4.1 where the 5’
end contains the Ncol cloning sites.
Two-hybrid experiments.
To assay interactions of the various CA proteins,
mutants, and domains, cDNAs subcloned into the yeast two-hybrid vectors pAS2-l and
pACT2 were transformed in different combinations into the yeast strain Y190.
Transformants were plated on complete synthetic media lacking both tryptophan and
leucine supplemented with glucose and grown at 30 degrees until colonies were visible.
Colonies were then picked and inoculated into liquid media of the same formulation and
grown out for two days. To assay for 3 galactosidase activity, one milliliter of culture
was pelleted in a microfiige tube, rinsed with buffer A (100 mM sodium phosphate, pH
7; 10 mM KC1; I mM MgSOa), and then resuspended in buffer A with 40 mM Pmercaptoethanol. The resuspended pellet was subjected to five rounds o f freeze thaw by
sequentially placing the tube first in liquid nitrogen and then in a 37 °C water bath. The
assay was initiated by addition o f 5-bromo-4-chloro-3-indolyl-P-D-galactoside (X-gal) to
a final concentration of 1 mg/ml. After a period o f approximately sixteen hours at 30 °C,
development of the blue reaction product was visually verified and compared to both
positive and negative controls.
Gel filtration chromatography. Relative molecular weights were determined for
both spinach CA and the E310A mutant by gel filtration chromatography using Sephacryl
S200 (Sigma). The two CAs were each expressed in E. coli from a derivative o f the
expression vector pKK-233 (Pharmacia), and total clarified lysates were applied to a
calibrated column. Molecular weight standards used to calibrate the column included
sweet potato P-amylase (200 kDa), bovine serum albumin (66 kDa), bovine carbonic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
CA
C213A
E310A
tail
Figure 4.1 Summary o f carbonic anhydrase constructs used to make the two-hybrid
fusions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
anhydrase II (29 kDa), and equine cytochrome c (12.4 kDa).
Peak fractions were
collected which corresponded to the molecular weights o f hexamers, dimers, and
monomers, and these samples were applied to SDS-PAGE gels for Western analysis
using an anti-CA polyclonal antibody (Fawcett et al., 1990).
Results and Discussion
An investigation of the primary structure o f available plant PCAs revealed a
distinct difference between the monocot and the dicot forms (Figure 4.2). The latter
group retains a ten amino acid “tail” at the carboxyl terminus o f the protein which is
missing in the monocot isozymes. We have examined this domain for its potential to
mediate the higher order oligomerization characteristic o f the dicot PCAs which
contrasts with the oligomerization states reported for the monocot isozymes (Graham et
al., 1984).
Both the wild type spinach CA and a corresponding carboxyl truncation were
expressed in E. coli and analyzed for their multimeric assembly by gel filtration. The
wild type enzyme migrated through the column with an apparent relative molecular mass
consistent with hexamers, though this population is presumably octameric for reasons
discussed above (Figure 4.3). We also detected a faint band in samples collected from
fractions corresponding to dimers, though we cannot rule out the possibility that this is a
trail of the hexamer peak. We do note that this bimodal profile was previously observed
for spinach CA and explained to be the result o f an assembly equilibrium in which both
dimeric and hexameric states are present. It was further shown that this dissociation is
not irreversible since the dimers could be concentrated to reform a hexameric species
(Pocker and Miksch, 1978).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
fbidentis
fbrownii
arabidopsis
tobacco
spinach
pea
rice
barley
maize
SLGNLLTYPF VRDGLRNKTL
SLGNLLTYPF VRDGLRNNTL
SLANRCTYPF V R E G W K G T L
SLGNLLTYPF VREGLVKKTL
SLGNLLTYPF VRDGLVKKTL
SLGNLLTYPF VREGLVNKTL
SLENLKTYPF VKEGIANGTL
SLQNLLTYPF VKEGVTNGTL
SLENLKTYPF VKEGLANGTL
ALKGGHYDFV
ALKGGHYDFV
ALKGGYYAFV
ALKGGHYDFV
ALQGGYYDFV
ALKGGYYDFV
KLVGGHYDFV
KLVGGHYDFV
KLIGAHYDFV
NGTFELWALD
NGTFELWALD
NGSFELWELQ
NGGFELWGLE
NGSFELWGLE
KGSFELWGLE
SGNLDLWEP
SGKFETWEQ
SGEFLTWKK
FGLSSPTSV
FGLSSPTSV
FGISPVHSI
FGLSPSLSV
YGLSPSQSV
FGLSSTFSV
Figure 4.2 Carboxy terminal sequence alignment of representative plant carbonic
anhydrases. Asterisks indicate a hydrophobic heptad repeat. The boldface E indicates
the glutamate in spinach CA which was mutated to yield either the truncated E310A
mutant or the Ncol site used to subclone the carboxyl terminus. Sequences are shown
for two species of Flciveria (U08402, U08402), Arabidopsis thaliana (L I8901), tobacco
(M94135), spinach (J05403), pea (X52558), rice (U08404), barley (L36959), and maize
(U08401). Alignment was generated using PELEUP of the Wisconsin GCG sequence
analysis suite. Dicots are underlined.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
CA (WT)
Figure 4.3 Gel filtration sizing o f spinach carbonic anhydrase and the E310A mutant.
Peak fractions corresponding to hexamers, dimers, and monomers were collected,
pooled, and separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis.
The gel was then blotted to nitrocellulose and probed with a rabbit polyclonal antibody
directed against spinach CA. Immunoreactive bands were visualized using a horse radish
peroxidase-coupled secondary antibody and luminol. T, total lysate; H, apparent
hexamers; D, apparent dimers; M, monomers; WT, wild type.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
Conversely, the E310A mutant eluted from the column at a relative molecular
weight corresponding to dimers with a faint signal also present in the hexamer fraction
(Figure 4.3). The band observed in samples collected from the higher molecular weight
fraction appears to be a genuine tail belonging to the dimeric peak based on careful
inspection o f fractions intermediate between the two molecular weights.
Thus,
elimination o f the carboxyl terminus has rendered CA incapable o f forming the higher
molecular weight oligomeric species. We therefore propose that the carboxyl terminus
contributes to a stabilizing interface which mediates either the dimer to tetramer or the
tetramer to octamer transition. Presently, we cannot say whether the E310A mutant is a
dimer or a tetramer, so a definitive verification o f its stoichiometry will await a more
quantitative investigation o f this assembly. Regardless, we have clearly shown that the
tail contributes to an oligomerization step.
To demonstrate that it is, in fact, the tail itself which mediates this interaction and
not that its deletion causes a secondary effect responsible for the disruption, we pursued
studies of CA’s assembly using the two-hybrid system. With this method, we were able
to express the wild type enzyme, two different mutants, and the carboxyl terminus in
various combinations. The results o f this approach are summarized in Table 4.1. We
find that this system demonstrates the assembly o f the wild type spinach CA as well as
the point mutant C213A. We cannot say, however, whether the observed assembly is at
the level o f the dimer, tetramer, or octamer, but only that these two-hybrid fusion
proteins will at least dimerize in vivo.
Curiously, we cannot detect the interaction
between E310A subunits, which we know to be at least dimeric based on the gel
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
T able 4.1 Interactions of various CA constructs assayed in the two-hybrid system.
binding
domain
CA
E310A
C 213A
TAIL
activation domain
CA
+
E310A
C 213A
-
+
+
+
-
-
-
TAIL
-
+
*
-
P-galactosidase activity is expressed as either detectable (+) or not detectable (-). *This
interaction was of significantly weaker strength than the other positive signals.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
filtration data.
The reasons for this are potentially explained by our model for CA
assembly (discussed below), and it may be that the two-hybrid data reflects the assembly
o f tetramers into octamers.
When the carboxyl terminus was co-expressed with the wild type CA, no
interaction was observed. However, we could demonstrate an interaction between the
carboxyl terminus and C213A (Table 4.1). The reasons for this difference are not clear,
but the interaction o f C213A with the tail does demonstrate that the carboxyl terminus is
capable of mediating intermolecular assembly.
Two possibilities exist to explain how the carboxyl terminus might mediate CA’s
oligomerization. The tail may contain specific amino acids which are involved in discreet
interactions at subunit interfaces. Indeed, this very role has been shown for the carboxyl
terminus in the enzyme nucleoside diphosphate (NDP) kinase (Mesnildrey et al., 1998).
Like CA, NDP kinase is found in two different oligomeric states in different phylogenetic
groups.
In prokaryotes, the enzyme is a tetramer while in eukaryotic organisms the
enzyme forms a hexamer, and this difference correlates with the presence or absence o f a
five amino acid carboxyl extension which is found only in the eukaryotic proteins. In the
crystal structure o f this enzyme, the carboxy terminal glutamate hydrogen bonds to
aspartate 115 in the neighboring subunit at the trimer interface within the hexameric
enzyme. Dicot PCAs may have adopted an analogous mechanism for mediating higher
order assembly via the carboxyl extension.
Alternately, the tail may form a structural feature which interacts with domains
on neighboring subunits. If the tail forms a P strand that is included in an intersubunit
sheet or barrel, it would display several amino acids capable of hydrogen bonding,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
including E 3 10, S314, and S318. On the other hand, if the tail forms an a helix, not only
could these interactions possibly stabilize its structure, but a potential hydrophobic
heptad repeat could form a coiled coil (Figure 4.2).
O f these two possibilities, our two-hybrid data suggest that the tail forms some
structural element which interacts with other CA subunits since one would not expect to
see a two-hybrid interaction as a result of a single amino acid hydrogen bond or salt
bridge. However, this model is discredited by the observation that the Self Optimized
Prediction Method, which is a collection o f secondary structure prediction algorithms,
suggests that the carboxyl terminus of spinach CA is largely unstructured (Geouijon &
Deleage, 1994).
A definitive demonstration o f either o f these possibilities will likely
await the elucidation of a crystal structure for this enzyme.
Concerning the quaternary structure o f the enzyme, we have devised a highly
speculative model for CA’s oligomerization which is both testable and consistent with
available data. Aliev et al. (1986) proposed that CA forms a double donut shape with
422 symmetry, and we can build a model o f CA based on this symmetry that accounts
for available data on the subject. Bjorkbacka et al. (1997) were able to use cross linking
agents to isolate monomers, dimers, and tetramers of a tetrameric mutant o f pea CA, but they
were unable to isolate trimers. If we try to reconcile this observation with Aliev’s symmetry,
the two are consistent with a tetramer that has four fold rotational symmetry and which is
built from monomers which each have two different reactive sites for the cross linker used. If
the monomers each possessed two equally reactive sites, the cross linker would have equal
opportunity to bridge any given interface and thus should produce trimers in a statistically
predictable amount. This is not the case. What was observed was likely the result of two
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
separate cross linking events which occur at different rates so that the fast event produces
dimers and the slower event covalently bridges these two dimers to produce a tetramer.
If the tetramer is then regarded as a donut with four fold rotational symmetry, then
the octamer may form a double donut with equivalent interfaces in contact with one another.
This double donut type structure is consistent with the observations o f Aliev et al. (1986)
based on their electron microscopy imaging of the chick pea enzyme. We can then rationalize
our two-hybrid data in the following way. Monomers assemble into dimers with no two fold
symmetry, but these dimers assemble into tetramers that display four fold symmetry.
Alternatively, there may be no intervening dimer step. Further, all four monomers assemble
in such a way that the tetramer has an “N” face which displays all four amino termini and a
“C” face that displays all four carboxyl termini. To form the octamer, then, two tetramers
assemble with their C faces in contact. This explains why the E310A mutant cannot form
octamers; the truncation disturbs the tetramer-tetramer interface.
This model also provides an explanation why our positive two-hybrid transformants
may uniquely reflect the assembly of octamers from tetramers. All the two-hybrid fusions
place the GAL4 domains at the amino terminus. In this arrangement, both the DNA binding
domain and the activation domain are sandwiched between CA and the DNA at the reporter’s
promoter. If the overall structure of CA is sufficiently large, it could sterically hinder the
activation domain from interacting with the transcription apparatus (Figure 4.4). Therefore,
our tetramer assembly would recruit the activator domain to the promoter, but it would be
unable to activate transcription. If, however, an octamer formed via a C - C face interaction,
the opposite side of the octamer would display the activator domain as well, and this activator
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.4 Tentative model of CA’s quaternary structure. This model does not necessarily
imply that the carboxyl termini must interact with one another.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
domain could then interact with the transcription machinery to produce p-galactosidase
mRNA.
Obviously, this model is nothing more than speculation, but it is consistent with
available data and it does provide a framework in which to test its predictions. First, the
model predicts that E310A is a tetramer. This can now be verified using cross linking studies
in the same way that other tetrameric CA mutants have been identified (Bjorkbacka et al.,
1997). Secondly, fluorescence spectroscopy can be used to verify that the tetramer has a
carboxyl face and an amino face. Fluorescent probes can be attached to these regions and
intervening distances can be measured (Liu, 1996). Third, the model predicts that C213A is
an unstable octamer. It must be octameric in order to activate transcription in the two-hybrid
system, but it must be unstable since the tail can interact with it (Table 4.1). This instability
could be verified by equilibrium centrifugation experiments and comparisons could be drawn
between the association constants of this mutant and the wild type enzyme.
Overall, we have shown that an element o f primary structure at the carboxyl
terminus o f dicot CAs stabilizes the oligomeric state o f the spinach enzyme. Deletion of
this extension eliminates the protein’s ability to form higher order oligomers.
Additionally, we have shown that this domain will interact with the C213A enzyme in the
yeast two-hybrid system. The carboxyl tail of the dicot PCAs may therefore be regarded
as an important determinant of quaternary structure.
References
Aliev, D. A., Guliev, N. M., Mamedov, T. G. & Tsuprun, V. L. (1986) Physicochemical
properties and quaternary structure of chick pea leaf carboanhydrase. Biokhimiya
51: 1524.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Atkins, C. A., Patterson, B. D., & Graham, D. (1972) Plant carbonic anhydrases II.
Preparation and some properties o f monocotyledon and dicotyledon enzyme types.
Plant Physiol. 50: 218.
Bjorkbacka, H., Johansson, I. -M., Skarfstad, E., & Foreman, C. (1997) The sulfhydryl
groups of cys 269 and cys 272 are critical for the oligomeric state o f chloroplast
carbonic anhydrase from Pisnm sativum. Biochemistry 36: 4287.
Bracey, M. H., Christiansen, J., Tovar, P., Cramer, S. P., & Bartlett, S. G. (1994) Spinach
carbonic anhydrase: investigation of the zinc binding ligands by site-directed
mutagenesis, elemental analysis, and EXAFS. Biochemistry 33: 13126.
Fawcett, T.W., Browse, J. A., Volokita, M., & Bartlett, S. G. (1990) Spinach carbonic
anhydrase primary structure deduced from the sequence o f a cDNA clone. J.
Biol. Chem. 265: 5414.
Geourjon, C. & Deleage, G. (1994) SOPM: a self optimized prediction method for
protein secondary structure prediction. Prot. Eng. 7: 157.
Graham, D., Reed, M. L., Patterson, B. D., Hockley, D. G., & Dwyer, M. R. (1984)
Chemical properties, distribution, and physiology o f plant and animal carbonic
anhydrases. Ann. N.Y. Acad. Sci. 429: 222.
Liu, W. (1996) Fluorescence studies o f EcoRi restriction endonuclease structure and
dynamics.
Ph.D. dissertation, Louisiana State University, Baton Rouge,
Louisiana.
Mesnildrey, S., Agou, F., Karlsson, A., Bonne, D. M., & Veron, M. (1998) Coupling
between catalysis and oligomeric structure in nucleoside diphosphate kinase J. Biol.
Chem. 273: 4436.
Pocker, Y. & Miksch, R. R. (1978) Plant carbonic anhydrase. Properties and bicarbonate
dehydration kinetics. Biochemistry 17: 1119.
Sambrook, J., Fritsch, E. F., & Maniatis, T. (1989) Molecular Cloning: A Laboratory
Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.
Tobin, A. J. (1970) Carbonic anhydrase from parsley leaves. J. Biol. Chem. 245: 2656.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5
CONCLUSIONS
This dissertation has illuminated, for the first time, a structural feature of the P
carbonic anhydrases which distinguishes this class from the prototypical a isozymes. A
combination of molecular biology and biophysics has been used to show that spinach CA
coordinates the active site zinc with two cysteines and one histidine. This is in contrast to
the a and y isozymes both o f which bind their respective active site zinc ions with three
histidine ligands. Prior to this work, it was generally assumed that the PCA active site
would largely mimic the aC A structure, thereby explaining the identical catalytic
efficiencies of the two CA classes. The distinction illustrated here may provide a basis to
explain the different responses these two classes have to some inhibitors. Furthermore,
the active site model proposed here has subsequently been corroborated by an independent
lab also using extended X-ray absorption fine structure analysis (Rowlett et al., 1994).
This novel ligand scheme now provides additional evidence that the P class is an
evolutionarily distinct carbonic anhydrase family.
The observation that the metal is
bonded to PCA by ligands of very different chemistry than the ligands o f the a class may
supply fertile ground for exploring how catalytic zinc coordination influences the kinetic
characteristics of carbon dioxide hydration in general.
In fact, the likelihood for the
sulfurs o f the cysteine ligands to raise the pK» of the catalytic water could prove relevant
to the potential regulation o f the enzyme’s activity by the increase in stromal pH during
photosynthesis.
Enzymologists will certainly be curious to see if the zinc-sulfur
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
interactions o f PCA impact on the theoretical mechanism o f hydroxyl mediated CO2 attack
by the enzyme.
It has also been shown that the ten amino acids at the carboxyl terminus o f spinach
CA have a profound impact on the enzyme’s quaternary structure. Deletion of this element
results in an enzyme compromised in its ability to form higher order multimeric complexes.
This disruption does not appear to be the result o f a secondary disturbance o f the
enzyme’s tertiary structure as the carboxyl terminus has been shown to independently
interact with the full length enzyme. This finding provides a second glimpse at what may
be a defining difference between monocots and dicots with respect to carbonic anhydrase,
and this difference may impact our understanding of how these two groups of angiosperms
differ with respect to carbon fixation in general.
The nature o f the interaction between the carboxyl terminus and other CA subunits
remains to be determined. However, the model proposed here for multimer architecture
generates several testable hypotheses. A detailed investigation o f the E310A mutant by
analytical ultracentrifugation could yield helpful quantitation of the contribution o f the
carboxy terminal domain to multimer assembly. Further, it may be possible to chemically
synthesize the carboxyl terminus for structural analysis using circular dichroism or nuclear
magnetic resonance spectroscopy. These approaches might yield a model for how the tail
mediates CA assembly.
However, such investigations could prove fruitless if the tail
requires the presence o f other domains in the CA enzyme in order to assume its native
conformation.
The locations o f the amino and carboxyl termini with respect to one another can
also be determined. The model proposed here predicts that amino termini are grouped
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
together on one face o f each tetramer, while the carboxyl termini are grouped on the
alternate face, and it is the latter face which forms the tetramer-tetramer interface. If this
is so, covalent modification o f these termini with appropriate fluorophores can be
exploited to measure distances between these regions by fluorescence spectroscopy. The
current model would be verified if the labelled amino termini are shown to generate
excimers, for example. The same would be true, in turn, for the carboxyl termini.
Overall, it seems clear that the most direct route to understanding the structural
features o f the beta carbonic anhydrases will be to solve the complete structure of a
representative member.
Hopefully, an X-ray crystal structure of a PCA will allow
investigators to extend our understanding o f this enigmatic enzyme.
Reference
Rowlett, R. S., Chance, M. R., Wirt, M. D., Sidelinger, D. E., Royal, J. R., Woodroofe,
M., Wang, Y. F., Saha, R. P. & Lam, M. G. (1994) Kinetic and structural
characterization o f spinach carbonic anhydrase. Biochemistry 33: 13967.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX A
COPYRIGHT RELEASE F O R CHAPTER 2
American Chemical Society
1165 SIXTEENTH STREET. N.W.
WASHINGTON. D.C. 20038
Phone: [202)672-4387 or -4368
Fax: (202) 872-6060 E-malt copyrlghtOaesbig
PUnuCATlONS DIVISION
COPYRIGHT OFFICE
FAX EOHBERs (504) 38B-25S7
DATEt
February 10, 1998
KDCFJLKDOH
TO:
M ichael H. Braoey, L ouisiana S ta te U n iv e rsity
D ept. 02 B io lo g io al S cie n ce s, 508 l i f e Scienoa B uilding
B aton Rouge LA 70803-1715
EAOHi
C. A rlaan Courtney CA s s is ta n t C opyright A d m in istrato r
RE:
Tour l a t t a r d atad January 30, 1998
Than): you f o r your ra o a n t l a t t a r , ra g a rd in g your raquaat fo r perm ission to in clu d a
your p a p a r(a ) o r p o rtio n a o f your p a p a r(a ) , p a r your a tta c h e d l e t t e r , in your
th e s is . P le a se note th e fo llo w in g :
•
I f your p a p e r h a t a lre a d y baan p u b lish e d by ACS, wa would be happy to g ra n t you
th i a p e rm issio n r o y a lty f r e e provided t h a t you p r in t th a re q u ire d ACS co p yright
c r e d i t l i n e on th a f i r a t page of your a r t i c l e : 'R e p rin te d (o r 'R ep rin ted in
p a r t ') w ith p erm ission fro * (rULL REFERENCE CITATION.] C opyright [TEAR] A aeriean
Chemical S o c ie ty .* In c lu d e a p p ro p ria te in form ation t h a t 'a b rack eted .
Mot*: I f you p la n to a u h a lt your th e a ia to OKI
d i a t r i b u t o r , p le a a a inform them t h a t parm iseion
p u b lia h a d ACS a r t i o l a aa p a r t o f y o u r th a a ia ie
ACS co p y rig h t n o tic e (aee above) muat appear on
a rtic le .
•
or to a n o th e r d ie a e r ta tio n
to In c lu d a your alr« a d y
g ra n te d f o r a l l form ate; the
the f i r a t page of th e ACS
I t your p ap er ha» n o t a lre a d y been p u b lia h a d by ACS, you may in clu d e i t in your
t ) e s l a i n p r i n t and m ic ro film fo ra a ta o n ly , provided th a t you p r in t the follow ing
ACS c o p y rig h t
c r e d it l i n e on the f i r s t page of your a r t i c l e : 'R ep rin ted (or
’R ep rin ted in
p a r t '] w ith perm ission from [JOURNAL NAME], in p re ss (or ’subm itted
le x p u b l i c a t i o n ') . Unpublished work c o p y rig h t [CURRENT YEAR] American Chemioal
St c i e t y . "
Mot*: I f you p la n to submit your th a a ia to 0MI or to a n o th er d is s e r ta tio n
d i s t r i b u t o r , you may NOT in c lu d e th a unpublished ACS p ap er in your th e s is i f
d isse m in a tio n o f th e th e s is w ill be i n e le c tro n ic form at, u n til ACS has
p u b lish e d your p a p e r.
A lte r n a tiv e ly , you may includa th a unpublished ACS
p a p e r in your t h a a ia , b u t only i f th a th a a ia w ill be dissem inated in p r i n t or
m ic ro film form ats o n ly . A fte r p u b lic a tio n of the paper by ACS, you may
r e le a s e th% e n t i r e th e s ie fo r e le c tr o n ic dissem in atio n , provided th a t ACS'a
c o p y rig h t c r e d it l i n e appears on th a f i r a t page of th e ACS paper. I f a paper
i s being co n sid ered f o r p u b lic a tio n by ACS but has n o t y e t been pu b lish ed ,
in c lu s io n of th e unpublished p ap e r in a th e a ia th a t w i l l be dissem inated
e le c tr o n io a lly m ight je o p a rd iz e p u b lic a tio n of the p a p er by ACS.
•
fi.rm issio n i s not g ra n te d to p o st any p u b lish e d or unpublished ACS a r t i c l e s on
ar.y w e b site .
•
Ch.h er:
Than:-: you f o r w ritin g .
Q uestions?
P lea s* o a l l me a t 202/872-4368.
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX B
COPYRIGHT RELEASE FOR CHAPTER 3
PLANT
PHYSIOLOGY
Published by the American Society o f Plant Physiologists
COPYRIGHT ASSIGNMENT
Zilux-la-ChUi. Maartoi J. Qulspatb
October 20, 1994
The American S o c i e t y of P l a n t P h y s i o l o g i s t s (ASPP) i s p l e a s e d
t o r e c e i v e and c o n s i d e r your paper f o r p u b l i c a t i o n i n PLANT
PHYSIOLOGY. However, your paper cannot be p u b l i s he d u n t i l t h e
c o p y r i g h t t r a n s f e r f o r * i s s i g ne d and r e t u r n e d . Pl e a se i n d i c a t e
your a c c e p t a n c e o f t h e s e t e r e s o f p u b l i c a t i o n by s i g ni n g t h i s for*
and r e t u r n i n g i t p r o n p t l y t o :
Deborah I . Weiner, Managing E d i t o r , PLANT PHYSIOLOGY,
15501 Monona D r i v e, R o c k v i l l e , MD 20055 USA
Author(s):
Michael H. Br acey, Sue G. B a r t l e t t
T i t l e ( R e f e r r e d t o a s the ’ A r t i c l e * ) :
Sequence of a cDNA Encoding Carbonic Anhydrase Fro*
Ba rle y
Manuscript Nu»ber: P94-1159
Date Received: October 20, 1994
In c o n s i d e r a t i o n o f the p u b l i c a t i o n o f the A r t i c l e , the Authors
gr an t t o ASPP and i t s s u c c e s s o r s a l l r i g h t s i n the A r t i c l e of
whatsoever k i n d , i n c l u d i n g t h o s e now o r h e r e a f t e r p r o t e c t e d by
th e Copyright Laws o f the United S t a t e s and a l l f or ei gn c o u n t r i e s ,
as wel l as any r e n ew a l, e x t e n s i o n , or r e v e r s i o n of c o p y r i g h t , now
or h e r e i n a f t e r p r o v i d e d , i n any c o u n t r y .
The Aut hors w ar r a n t t h a t the A r t i c l e i s an o r i g i n a l work, t h a t i t
has not been p u b l i s h e d el s e whe re i n whole or i n p a r t , t h a t i t
c o n t a i n s n e i t h e r B a t t e r i n f r i n g i n g on t h e cop yr i gh t of any o t h e r
per son, nor B a t t e r t h a t i s i n any way u n l aw f ul . The Authors w i l l
hold ha rml es s and i n d e n n i f y ASPP and i t s eapl oyees a g a i n s t any
c l a i * o r p r o c ee d i n g i n s t i t u t e d as a r e s u l t of a breach of t h e s e
warranties.
ASPP, i n t u r n , grant's t o t h e Authors th e r o y a l t y - f r e e r i g h t of
p u b l i c a t i o n i n any book o f which the A u t h o r ( s) i s the Author or
E d i t o r , s u b j e c t t o t h e e x p r e ss c o n d i t i o n t h a t l awful n o t i c e of
c l a i a o f c o p y r i g h t be g ive n.
The c o r r e s p o n d i n g Author, with the c oncur r ence e i t h e r o r a l l y or
i n w r i t i n g of each o f t he Co-Authors, s houl d s i gn t h i s agreement
on Yfi^_behalf‘-«(C'alLi,g A t h e Authors.
S i g n a t u r e of Cor respondi ng Author.
Date.
Exemption f o r Authors Employed by t h e United S t a t e s Government.
I a t t e s t t h a t t h e A r t i c l e was w r i t t e n as p a r t of the o f f i c i a l d u t i e s
of t he Authors a s employees o f the U.S. government and t h a t t r a n s f e r
of c o p y r i g h t cannot be made.
S i g n a t u r e of Cor respondi ng Author.
Date.
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
AmocUIc EdiiorK
Gloria M. C om al
Malcolm C D r w
Thomas J. CuilfoM*
Sharon 1L Luci|
William L Ogrw
Charles F. Yon an
VITA
Michael Holden Bracey was bom on 16 February, 1969, in Shreveport, Louisiana.
He grew up and attended school there, graduating from Caddo Parish Magnet High in
May, 1987.
He moved to Baton Rouge in August, 1987, to attend Louisiana State
University with a Chancellor’s Alumni Scholarship. Michael received his bachelor o f arts
degree in May, 1991, in French. Following a year at work as a Research Associate for Dr.
Sue G. Bartlett in the Department o f Biochemistry, he joined the graduate program at
L.S.U. in the fall o f 1992.
Since then, Michael has worked on carbonic anhydrase,
thylakoid protein import, and osmotically induced signal transduction while in Dr.
Bartlett’s lab. After graduation in May, 1998, Michael plans to join the research group of
Dr. Ben Cravatt at the Scripps Research Institute in La Jolla, California, where he will
pursue post-doctoral training in neurochemistry and oleamide signaling.
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DOCTORAL EXAMINATION AND DISSERTATION REPORT
Candidate:
Michael H. Bracey
Major Pield:
Biochemistry
Title of Dissertation:
Structural Characterization of Beta Carbonic
Anhydrases from Higher Plants
Approved:
Major Professor and Chairman
Dean of the Graduate School
EXAMINING COMMITTEE:
/
W
,
Date of Examination:
April 8,
1998________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
sr
IMAGE EVALUATION
TEST TARGET ( Q A - 3 )
✓
r
V
1.0
H i | 2. 8
| 25
ta l «
|2 . 2
1 AO
B=_=-.l=
2.0
l.l
1.8
1.25
14
1.6
15 0 m m
IIV W G E.Inc
1653 East Main Street
Rochester, NY 14609 USA
Phone: 716/482-0300
Fax: 716/288-5989
0 1993. Applied Image. Inc.. All Rights Reserved
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.