| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 37, 35033-35038, September 12, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the
Department of Biological and Biomedical
Sciences, University of Durham, South Road, Durham, DH1 3LE, United Kingdom
and ¶Pharmazeutische Biochemie, Pharmazeutisches
Institut, Morgenstelle 8, D-72076 Tübingen, Germany
Received for publication, March 25, 2003 , and in revised form, June 23, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
cAMP is synthesized in mammals by a seemingly ubiquitous family of class III plasma membrane-spanning ACs (transmembrane adenylyl cyclase, tmAC) that mediates cellular responses to extracellular signals. Additionally, a cytosolic form of AC (soluble adenylyl cyclase, sAC) has been identified in mammals that was demonstrated to be molecularly and biochemically distinct from the tmACs (5). Although most abundantly expressed in testis, sAC is expressed ubiquitously (6, 7) and is directly activated by bicarbonate ion in a pH-independent manner (8).
The 
mammalian sAC is more closely related to other prokaryotic class III ACs than
to other mammalian tmACs (5,
9). Consistent with this
phylogenetic relationship, it was demonstrated that a single cyanobacterial
class III AC, cyaC of Spirulina platensis (Arthrospira), was
also stimulated by 
(8). If
stimulation were a general feature
of at least a subset of class III ACs, they would represent the first family
of 
-responsive
signaling molecules. 
is
fundamental to prokaryotic biology; accumulated cytoplasmic
is the primary source of inorganic
carbon transported to the cyanobacterial carboxysome for photosynthesis
(10) and is also hypothesized
to have been the predominant carbon source utilized by oxygenic phototrophs in
the generation of Earth's oxygen atmosphere
(11).
To define the extent to which class III ACs may be stimulated by
we have utilized the
cyaB1 AC gene of the nitrogen-fixing freshwater cyanobacterium
Anabaena sp. PCC7120 as a model system. Cyanobacteria are dependent
upon the accumulation of intracellular
for growth, but the mechanism by
which they detect is unknown and a
major stumbling block in the study of this environmentally important class of
organisms. The genome of Anabaena sp. PCC7120 encodes six AC genes
(12,
13), and cyaB1 codes
for a protein that has an N-terminal autoregulatory GAF (found in
cGMP-phosphodiesterases, adenylyl cyclases, and
FhlA (formate hydrogen lyase transcriptional activator)) domain
that binds cAMP and up-regulates catalytic activity
(14). Biochemical analysis of
the catalytic center of cyaB1 revealed that
stimulates the catalytic activity
of AC by an increase in reaction velocity. In addition we have defined a
residue (Lys-646) essential for
action within the catalytic
center. We have examined the catalytic centers of a number of other
prokaryotic class III ACs and demonstrated that an active site lysine
coordinates in the catalytic cleft
of the subset of ACs that contain an aspartate to threonine active site
polymorphism. On the basis of this hypothesis, we propose that a large number
of prokaryotic class III AC catalytic domains are
.
signaling through cAMP synthesis
is established as a mechanism by which a variety of eukaryotic and prokaryotic
organisms can respond to environmental carbon. This knowledge is of
fundamental importance in understanding the global impact of bicarbonate on
organismal biology.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Nucleotides 1349–1930 of the S. aurantiaca B17R20 cyaB gene (GenBankTM accession number AJ223795 [GenBank] ; gift from Dr. O. Sismeiro, Institut Pasteur) were amplified by PCR, and cloning was performed using standard molecular biology techniques. A discrepancy from the published sequence was noted that gave an amino acid change (P163R). A BamHI and a HindIII site were added at the 5'- and 3'-end, respectively. The cyaB fragment was cloned between the BamHI and HindIII sites of pQE30. The resulting open reading frame codes for amino acids 160–353 of the cyaB adenylyl cyclase with an MRGSH6GS metal-affinity tag at the N-terminal end. Primer sequences are available on request.
Expression and Purification of Bacterially Expressed Proteins—Anabaena cyaB1 wild type and mutant proteins and Mycobacterium Rv12641–397 protein were expressed and purified as previously described (14, 15). Full details of the Mycobacterium Rv1319c protein will be reported elsewhere.
The Stigmatella pQE30-cyaB construct was transformed into
E. coli BL21(DE3)[pREP4]. A culture was grown in Luria Bertani broth
medium containing 100 mg/liter ampicillin and 25 mg/liter kanamycin at 30
°C to an A600 of 0.5. 60 µM
isopropyl-
-D-thiogalactopyranoside was added and the culture
kept at room temperature for 3 h. Cells were harvested by centrifugation at
4,000 x g and washed once with 10 mM Tris-HCl, pH
7.5. The cell pellet was resuspended in 20 ml of buffer A (50 mM
Tris-HCl, pH 8.0, 2.5 mM 1-thioglycerol, 50 mM NaCl) and
disrupted by two treatments in a French Press at 1000 psi. Particulate
material was removed at 31,000 x g for 30 min. The supernatant
was supplemented with 250 mM NaCl, 15 mM imidazole, and
200 µl of Ni2+-nitrilotriacetic acid slurry (Qiagen)
for 30 min. The resin was washed with 3 ml each of buffer B (Tris-HCl, pH 8.0,
2.5 mM 1-thioglycerol, 2 mM MgCl2, 400
mM NaCl, 5 mM imidazole), buffer C (buffer B with 15
mM imidazole), and buffer D (buffer C with 10 mM NaCl).
The enzyme was eluted with 0.4 ml of buffer E (buffer B with 10 mM
NaCl and 150 mM imidazole). The preparation was stabilized with 20%
glycerol and stored at 4 °C.
AC Assay—The AC activity of cyaB1 wild type protein, cyaB1
mutant proteins, and other prokaryotic AC recombinant proteins was assessed in
a final volume of 100 µl
(16). Reactions typically
contained 22% glycerol, 50 mM MOPS-Na as buffer, 2 mM
MnCl2 as divalent metal ion cofactor, and 75 µM
[
-32P]ATP (25 kBq) and 2 mM
[2,8-3H]cAMP (150 Bq) to determine yield during production
isolation (cAMP was not added to assays for cyaB1 holoenzyme). Details of pH,
temperature, and enzyme concentration are provided in the figure legends.
Differences in buffer or cofactor usage are also indicated in the text.
Protein concentration was adjusted to keep substrate conversion at <10%.
Kinetic constants were determined over a concentration range of substrate of
1–100 µM. The data represent the means of several
independent experiments, and error bars represent the S.E.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
sAC and
Spirulina (Arthrospira) cyaC, whereas the remainder
expressed an Asp residue essential for substrate definition in the
corresponding position. Given the conservation of the active site Thr
polymorphism between cyaB1, sAC, and Spirulina cyaC, we investigated
whether cyaB1 was also stimulated by
. We expressed the catalytic domain
of cyaB1 (cyaB1595–859) to include a region of the C terminus
(amino acids 795–828) that had some similarity to a tetratricopeptide
repeat and is essential for production of functional soluble protein in E.
coli (14).
|
The activity of cyaB1595–859 was measured in the presence
or absence of various salts (Fig.
1B). Specific activity was unchanged in the presence of
NaCl and KCl, whereas NaHCO3 and KHCO3 both gave an
2-fold increase of cyaB1595–859-specific activity,
demonstrating that activation of
cyaB1595–859 was independent of the associated cation. We
measured the specific activity of cyaB1595–859 over a range
of concentrations with
Cl– as a control for nonspecific ionic effects
(Fig. 2A). A maximal
2-fold stimulation was seen in the presence of
with an EC50 of 9.6
mM. The GAF-B domain of cyaB1 binds cAMP and activates the AC
catalytic domain (14). cyaB1
therefore acts as a self-activating switch. We asked whether the behavior of
this switch is affected by and
expressed recombinant protein corresponding to the cyaB1 holoenzyme
(cyaB11–859) that contains the GAF domains and examined its
specific activity in the presence or absence of
.
cyaB11–859-specific activity showed a non-linear time
dependence as previously reported
(14); the rate of cAMP
formation was significantly accelerated in the presence of 10 mM
KHCO3, indicating that
activated the GAF-B-mediated positive feedback mechanism of cyaB1
(Fig. 2B). The rate of
cAMP formation was also stimulated in the presence of 10 mM
NaHCO3 but inhibited in the presence of higher concentrations of
NaHCO3, indicating that Na+ may block GAF-B binding of
cAMP or intramolecular
signaling.2
|
cyaB1595–859-specific activity showed a non-linear protein
dependence (Fig. 3), indicating
that homodimerization was necessary for formation of the active site. This has
been independently confirmed by titration of complementary mutant
cyaB1595–859 proteins that are inactive as homodimers but
restored catalytic activity as heterodimers
(14). To determine whether
up-regulated
cyaB1595–859-specific activity by increasing homodimer
formation, we examined the ratio of the
and Cl–-specific
activities as a function of protein concentration. Interestingly, this ratio
remained constant over the range of protein concentrations tested, indicating
that did not affect homodimer
formation. The protein concentration independence of
up-regulation of specific activity
allowed us to make comparisons between experiments in which different
concentrations of protein were assayed (see
Fig. 4).
|
|
We examined the kinetic properties of cyaB1595–859 to
determine whether altered the
behavior of the active site. The activation energy (Ea)
was derived from the linear arm of an Arrhenius plot (tested range 4–47
°C) and was similar in the presence of either Cl– (91.6
± 4.9 kJ/mol) or (97.7
± 3.7 kJ/mol), indicating that
did not fulfill the criteria for a
true catalyst in lowering 
H. The Km value for ATP
at pH 8.5 and 45 °C was 3-fold greater with
(33.8 ± 2.8
µM) than with Cl– (11.8 ± 0.8
µM), indicating a higher requisite substrate concentration to
achieve a given reaction velocity. The corresponding Vmax
values were 2.5-fold greater with
(238.0 ± 36.3 nmol/mg/min) than with Cl– (93.5
± 8.2 nmol/mg/min). A consequence of this is that enzyme efficiency
(kcat/Km) was similar for both ions
but substrate turnover rate (kcat) was 2.5-fold
greater for (7.0
min–1) than for Cl– (2.6
min–1). A Hill coefficient of 1.1 indicated that
neither ion stimulated significant cooperativity of the two catalytic
sites.
The kinetic data implied that
interacts with the catalytic center to alter substrate-binding kinetics. The
catalytic center is in close agreement with a canonical class III catalytic
cleft (17,
18) except for the replacement
of an aspartate (Asp-1018 in AC IIC2; Ref.
17) with a threonine (Thr-721
in cyaB1). Asp-1018 is involved in substrate definition in AC by forming a
hydrogen bond with N6 of the adenine ring of ATP
(17). Thr-721 functionally
replaced this aspartate and may act as a hydrogen acceptor from the purine
ring (14). When assayed at pH
7.5 to eliminate problems with divalent metal ion depletion,
cyaB1595–859-specific activity was stimulated 3-fold
relative to the Cl– activity over the tested range
(0–60 mM )
(Fig. 4A). We
investigated the involvement of the canonical active site residues of a class
III AC in stimulation using point
mutations. Although the basal-specific activities of
cyaB1595–859R732A (transition state stabilization),
cyaB1595–859N728A (transition state stabilization), and
cyaB1595–859D719A (a residue examined for possible functional
homology to Asp-1018 of AC IIC2) were significantly reduced
compared with wild type enzyme, their fold stimulation by
was equivalent (Supplemental Data
Fig. 1). A key difference between Thr-721 of cyaB1 and Asp-1018 of AC
IIC2 is the loss of the aspartate carboxyl group. We reasoned that
possibly mimics the carboxyl group
within the active site but, interestingly,
up-regulation of
cyaB1595–859T721A-specific activity was equivalent to wild
type despite a >99% reduction in basal activity
(Fig. 4B). We noted
that Lys-938 of AC IIC2 (substrate definition and equivalent to
Lys-646 of cyaB1; Ref. 17) was
proposed to act not only as a hydrogen acceptor for the N1
of the ATP purine ring but also as a hydrogen donor to the carboxyl group of
the adjacent Asp-1018 residue
(19). Thus, Lys-646 may form a
stabilizing hydrogen bond with at
a position equivalent to the carboxyl group of AC IIC2. Although
basal activity was reduced by 95%,
activation was completely
abolished in cyaB1595–859K646A in support of this hypothesis
(Fig. 4C). If
mimics a carboxyl group within the
active site, reintroduction of this carboxyl group should ablate
responsiveness. A
cyaB1595–859T721D mutant protein was refractory to
stimulation and had an enhanced
basal-specific activity relative to cyaB1595–859T721A
(Fig. 4D), lending
positive support to this hypothesis. This represents the first description of
a site for action within a
signaling molecule.
Although the amino acid equivalent to Lys-646 of cyaB1 and Lys-938 of AC
IIC2 is conserved in all the ACs examined
(Fig. 1A), we reasoned
that an adjacent threonine or aspartate within the catalytic cleft of a class
III enzyme (i.e. at the position corresponding to Thr-721) could be a
marker for AC responsiveness or
non-responsiveness, respectively. To test this hypothesis we generated
recombinant proteins corresponding to diverse prokaryotic class III ACs with
either a threonine or aspartate at the position equivalent to cyaB1 Thr-721
(Fig. 1A) and examined
them for their response to .
S. aurantiaca B17R20 is a myxobacterium from which two ACs have
been identified (20). We
expressed amino acids 160–353 of cyaB as a recombinant protein
(cyaB160–353) that contained a threonine residue (Thr-293) at
the position corresponding to cyaB1 Thr-721
(Fig. 1A).
cyaB160–353-specific activity was up-regulated by
2-fold relative to the
Cl–-dependent activity (EC50 8.6 mM)
(Fig. 5A), consistent
with the hypothesis that the threonine at amino acid 293 is a marker for
responsiveness. This stimulation
was maintained in the presence of alternative anions to Cl–,
indicating that cyaB160–353 was most likely stimulated by
rather than inhibited by
Cl–.2
|
M. tuberculosis H37Rv is a Gram-negative bacterium and important
human pathogen for which the genome has revealed a number of putative class
III ACs (15,
21,
22). We expressed two ACs that
contain either a threonine (amino acids 356–535 of Rv1319c) or an
aspartate (Rv1264 holoenzyme) at the position corresponding to Thr-721 of
cyaB1 (Fig. 1A).
Consistent with our hypothesis that the threonine residue is a marker for AC
responsiveness,
Rv1319c356–535-specific activity was up-regulated 3-fold
in the presence of over the
concentration range tested (Fig.
5B), whereas Rv12641–397-specific
activity did not respond to over
an identical concentration range (Fig.
5C). The data of Fig.
5 support the hypothesis posited in
Fig. 4 and indicate that
class III AC
domains are widespread in biology and represent the sole candidate mechanism
for detection in an organism.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
, extending the number of
identified class III ACs that are stimulated by
; stimulation was
cation-independent and anion-dependent. The measured EC50 of 9.6
mM is well within the range of calculated intracellular
concentrations for cyanobacteria
(23). Although the inorganic
carbon pool for Anabaena sp. PCC7120 has not been measured, the
related heterocyst-forming species Anabaena variabilis M3 can
accumulate up to 50 mM internal inorganic carbon depending upon the
growth conditions (24). cAMP
production through cyaB1 is therefore likely to be responsive to variations in
intracellular . Intracellular cAMP
has previously been correlated with the rate of
uptake in the cyanobacterium
Anabaena flos-aquae
(25), indicating that the
protein chemistry we describe is functional in vivo.
was able to functionally activate
not only the catalytic domains but also the entire holoenzyme with its
associated GAF and PAS domains. The GAF-B-mediated positive feedback loop
created by cyaB1 may therefore be accelerated by the availability of a fixable
carbon source in Anabaena sp. PCC7120.
did not affect cyaB1 homodimer
formation or lower the activation energy for transition state formation but
did significantly alter substrate binding kinetics by increasing the
Km for ATP and Vmax. The
cyanobacterium Synechococcus PCC6301 (Anacystis nidulans)
has an intracellular ATP concentration of 1 mM (value
calculated from data in Ref.
26). Because the
Km (ATP) for both cyaB1595–859 and
holoenzyme is of the order of <50 µM, it is likely that the
effect of on
Km is biologically irrelevant and that cyaB1 is activated
by in the intracellular
environment by an increase in reaction velocity. Point mutations revealed that
loss of Thr-721 did not affect cyaB1595–859
responsiveness. We demonstrated,
however, that loss of Lys-646 (equivalent to Lys-938 of AC IIC2)
ablated stimulation of specific
activity. In class III ACs that contain an aspartate residue corresponding to
the position of Thr-721, the adjacent lysine in the catalytic center has been
proposed to form a hydrogen bond with the aspartate carboxyl group
(19). We hypothesize that in
cyaB1 can functionally replace
this carboxyl group and is coordinated within the catalytic cleft by Lys-646.
A T721D point mutation was refractory to
in support of this hypothesis. The
enhanced basal activity of T721D relative to T721A may represent an enzyme
mimicking activation. If
does functionally replace the
carboxyl group of an aspartate, it is surprising that
increases Km
(ATP) given that a logical extension of our hypothesis would be that
forms a hydrogen bond with
N6 of the adenine ring and increases affinity for
substrate. It is possible that
binding results in subtle changes in the structure of the substrate-binding
pocket that lowers affinity but optimizes orientation for catalysis. Because
there is no effect on Ea in the presence of
, it is unlikely that this effect
is on the acquisition of the transition state. The increase in
kcat demonstrates that there is an increase in catalytic
activity on formation of the enzyme-substrate complex and this may therefore
occur after formation of the transition state. The exact mechanism of
activation of AC is an interesting
question that requires further investigation. It is significant to note that a
recent study using recombinant human sAC demonstrated that
increased Vmax
with no effect on Km (ATP)
(27). The
Km (ATP) for human sAC is 0.8 mM compared with
11.8 ± 0.8 µM for cyaB1. Differences in the active site
structure between the two enzymes that give this discrepancy in
Km (ATP) may also account for the different effects on the
kinetics of substrate binding in the presence of
.
Independent support for the proposed site of action of
came from studies with recombinant
class III AC domains from other prokaryotic species that contained either a
Thr or an Asp residue corresponding to the position of cyaB1 Thr-721. To date,
all ACs that are responsive to
contain a threonine residue (Anabaena cyaB1, Stigmatella
cyaB, Mycobacterium Rv1319c (this study), mammalian sAC, and
Spirulina (Arthrospira) cyaC, Ref.
8), and those that are
unresponsive contain an aspartate residue (mammalian tmACs, Ref.
8, Mycobacterium
Rv1264 (this study), and
Rv1625c.3 In addition,
mammalian soluble and receptor-type guanylyl cyclases (GC) have also been
demonstrated to be
-non-responsive.4
Presumably the change in the binding pocket of GC relative to AC that allows a
glutamate residue essential for substrate specificity to interact with
N1 and N2 of the guanine ring
(19) would not permit
at the active site.
is ubiquitous in the
intracellular and extracellular aqueous environment.
has a huge impact on the biology
of multiple eukaryotic and prokaryotic systems, but the mechanism by which
organisms detect and respond to fluctuating
is unknown. The expression of
class III AC
domains among diverse prokaryotes and eukaryotes represents the sole mechanism
by which organisms may respond to environmental carbon.
|
FOOTNOTES |
|---|
The on-line version of this article (available at
http://www.jbc.org)
contains a supplemental figure.
A Leverhulme Trust Research Fellow. To whom correspondence should be
addressed. E-mail:
m.j.cann{at}durham.ac.uk.
1 The abbreviations used are: AC, adenylyl cyclase; sAC, soluble adenylyl
cyclase; tmAC, transmembrane adenylyl cyclase; GAF,
cGMP-phosphodiesterases, adenylyl cyclases, and
FhlA (formate hydrogen lyase transcriptional activator); MOPS,
4-morpholinepropanesulfonic acid; PAS, periodic clock
protein, aryl hydrocarbon receptor, and
single-minded protein.
2 M. J. Cann and T. Kanacher, unpublished observations.
3 M. J. Cann, unpublished data.
4 M. J. Cann and D. L. Garbers, unpublished observations.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. A. Kalamidas, M. P. Kuehnel, P. Peyron, V. Rybin, S. Rauch, O. B. Kotoulas, M. Houslay, B. A. Hemmings, M. G. Gutierrez, E. Anes, et al. cAMP synthesis and degradation by phagosomes regulate actin assembly and fusion events: consequences for mycobacteria. J. Cell Sci., September 1, 2006; 119(Pt 17): 3686 - 3694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Bai, L. A. McCue, and K. A. McDonough Characterization of Mycobacterium tuberculosis Rv3676 (CRPMt), a Cyclic AMP Receptor Protein-Like DNA Binding Protein J. Bacteriol., November 15, 2005; 187(22): 7795 - 7804. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Masuda and T.-a. Ono Adenylyl Cyclase Activity of Cya1 from the Cyanobacterium Synechocystis sp. Strain PCC 6803 Is Inhibited by Bicarbonate J. Bacteriol., July 15, 2005; 187(14): 5032 - 5035. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Geng, Z. Wang, J. Zhang, B. Y. Reed, C. Y. C. Pak, and O. W. Moe Cloning and characterization of the human soluble adenylyl cyclase Am J Physiol Cell Physiol, June 1, 2005; 288(6): C1305 - C1316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Kimura, M. Ohtani, and K. Takegawa An Adenylyl Cyclase, CyaB, Acts as an Osmosensor in Myxococcus xanthus J. Bacteriol., May 15, 2005; 187(10): 3593 - 3598. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Gazdik and K. A. McDonough Identification of Cyclic AMP-Regulated Genes in Mycobacterium tuberculosis Complex Bacteria under Low-Oxygen Conditions J. Bacteriol., April 15, 2005; 187(8): 2681 - 2692. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
