| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Max-Delbrück Center for Molecular Medicine, 13092 Berlin;
Max Planck Institute for Physiological and Clinical Research, Bad Nauheim;
Franz-Volhard Clinic at the Max-Delbrück Center for Molecular Medicine, Humboldt University of Berlin; and
§ Institute of Physiology, Humboldt-University of Berlin, Germany
1Correspondence: Ingo Morano, Ph.D., Max Delbrück Center for Molecular Medicine, Robert-Rössle-Str. 10, 13092 Berlin. E-mail: imorano{at}mdc-berlin.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: cAMP kinase • protein phosphorylation • cardiomyocytes
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1C, and the
accessory ß2 and 2
subunits (reviewed in ref. 1
). Enhanced calcium channel
activity on ß-adrenergic receptor stimulation is a classical example
of an ion channel modulation via protein kinase A (PKA)
(1)
-dependent protein phosphorylation leading to increased
contractile force in the myocardium (1
2
3
4
5)
. This increase
is known to be mediated by a cascade reaction involving
G-protein-coupled adenylyl cyclase and activation of PKA
(3
4
5)
. The final step in the signal transduction process,
however, i.e., the site(s) targeted by PKA and the mechanism by which
phosphorylation increased the open probability of calcium channels,
remain incompletely understood. Clearly, both
1C and ß2 subunits
(but not 2) contain potential PKA
phosphorylation sites and are in vitro substrates of PKA
(1)
. We demonstrated previously that in intact dog
myocardium the ß2 subunit is in vivo
phosphorylated after exposure to isoprenaline (6
, 7)
,
suggesting that phosphorylation of the ß2
subunit plays a crucial role in the modulation of calcium channel
activity. This hypothesis was supported by patch-clamp measurements on
expressed calcium channels (8
, 9)
. Results from other
studies, however, have suggested that phosphorylation of the
1C subunit represents the final common pathway
for enhanced calcium channel activity on ß-adrenergic stimulation
(10
11
12
13
14)
and a single serine residue (Ser-1928) was
proposed as the site whose phosphorylation mediates PKA effects
(13
, 14)
. Other studies were unable to reproduce a
stimulatory effect of PKA on calcium channel activity, although similar
cloned calcium channel subunits were expressed (15
16
17)
.
From these findings the hypothesis was drawn that either the subunit
combinations were not identical to those in cardiac myocytes
(17)
or that PKA-dependent regulation requires additional
unidentified components (16
, 17)
.
In this study, we tested the "missing link" hypothesis in native
cardiac preparations and searched for PKA target proteins that are
associated to established channel constituents. Coprecipitation
experiments identified a large (~700 kDa) phosphoprotein tightly
bound to the ß2 subunit of mammalian cardiac
calcium channels. It was surprising to find that sequence analysis of
phosphorylated peptides revealed a close homology with AHNAK
(18)
, a human gene encoding an unusually large 700-kDa
phosphoprotein (19)
.
A three-domain structure was predicted for AHNAK: unique sequence
domains at the amino- and carboxyl-terminal ends of the protein flank a
large internal domain composed of highly conserved repeated elements
(18
, 19)
. The carboxyl-terminal region contains putative
nuclear localization signals and the protein located indeed primarily
in the nucleus (19)
. However, later studies revealed that
AHNAK/desmoyokin is also distributed in the plasma membrane
(20)
and in the cytoplasm (20
, 21)
and
translocates from cytoplasm to the plasma membrane of keratinocytes on
PKC treatment (21)
. Although there is a large body of
structural information about the giant protein encoded by AHNAK, no
definite function has yet been described.
Herein we characterize the cardiac 700-kDa phosphoprotein (pp700) possessing similarities to AHNAK. Coupling of pp700 to the ß2 subunit of calcium channels and the regional in vivo phosphorylation of pp700 on isoproterenol stimulation suggest that this protein plays a role in PKA-mediated signal transduction pathway(s) of cardiac myocytes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
- and ß-MHC in a tissue-specific manner. The mAb to RyR2
(clone: C3–33, mouse IgG) was from Alexis Corp. (Läufelfingen,
Switzerland). Cell culture medium SM20-I, fetal calf serum, penicillin,
streptomycin, gentamicin, trypsin, and glutamine were provided by
Biochrom KG (Berlin, Germany). Enhanced chemiluminescence reaction kit
(ECL) and [
-32P]ATP were from Amersham. The
PKA was prepared as described previously (6)
.
Cardiac tissue
Human left ventricular myocardium was pooled from four
multiorgan donors whose hearts could not be transplanted for technical
reasons. The use of explanted human myocardial tissue for in
vitro investigations was approved by the local ethical committee.
Rats and mice were from a local source. The strains were Wistar rats
(Schö: Wist) and Black 6 mice (PG 129, B6). Pig hearts from a
slaughterhouse were used for comparison to premedicated landrace pigs
(see below).
Sequence-directed antibodies
New Zealand White rabbits were immunized with synthetic peptides
coupled to carriers as described previously (6)
. Antigenic
epitopes comprised the amino acid residues 799–813
(EEEEKERKKLARTASPEKK) in the rabbit cardiac 1C
cDNA (22)
, the amino acid residues 544–567
(YEEEMTDNRNRGRN) of the ß1b (23)
and the carboxyl-terminal amino acid residues (EWNRDVYIRQ) of the
ß2 (24)
and (QRNRPWPKDSY) of the
ß3 (24)
. The anti-ß subunit
"generic" antibody (ßcom) was generated to
a highly conserved sequence (DSYTSRPSDSDVSLE) corresponding to amino
acid residues 21–35 in the rabbit cardiac ß2a
(24)
. The antibodies were affinity purified on peptide
antigen columns as described (25)
and have been
characterized in several models: 1C (26
, 27)
, ß1b (28)
,
ß2 (28
29
30)
, and
ß3 (29
, 30)
. To raise antibodies
against human AHNAK two epitopes were chosen following the strategy of
Stivelman and Bishop (19)
: KIS (KISMPDVDLHLKGPK) and FEN
(KMPKVKMPKFSMPG). Both KIS and FEN antibodies were generated and
affinity-purified as described for the calcium channel antibodies.
Surgery
All experiments were performed in accordance with the standards
established by the German law for the protection of animals. Male
landrace pigs (40–45 kg body weight) were premedicated with azaperone
(Stresnil, Janssen) 2 mg/kg, intramuscular (i.m.). Anesthesia was
induced and maintained by intravenous (i.v.) infusion of pentobarbital
sodium. Pigs were ventilated with an air-oxygen mixture using a Stephan
Respirator ABV (Stephan, Gachenbach, Germany). Ventilation was
controlled by intermittent measurements of arterial blood gases. Blood
pressure was measured with the aid of a Statham P23Db pressure
transducer connected to a saline-filled catheter in the ascending
aorta. The chest was opened by mid-sternal thoracotomy and the heart
was suspended in a pericardial cradle. After surgery, 2 h were
allowed for stabilization.
In vivo phosphorylation by intramyocardial infusion
of isoproterenol
Isoproterenol was administered by intramyocardial infusions
using a peristaltic pump and a 26-gauge hypodermic needle, which was
inserted 6 mm into the left ventricular myocardium. Isoproterenol (1
µM, freshly prepared) or vehicle (150 mM NaCl) was infused for
60 s at an infusion rate of 20 µl/min (31
, 32)
. For
sample retrieval, the needles were removed and the infusion site
biopsied with the aid of a drill-biopsy device (33)
.
Myocardial samples (30 mg wet weight) were retrieved either immediately
after stoppage of the infusion (i.e., at the end of the 60-s infusion
period) or after a 5-min delay (to study reversibility of the effects
of isoproterenol infusion). Biopsies were immediately frozen by
immersion in liquid nitrogen, freeze-dried, and stored in evacuated
glass vials at -20°C until use.
Tissue preparation
Frieze-dried samples from the experimental groups were pooled
and 40-mg samples were homogenized with 0.5 ml buffer F, consisting of
5 mM histidine/HCl, pH 7.4, 50 mM
Na4P2O7,
25 mM NaF, 10 mM EDTA, 0.2 mM dithiothreitol (DTT), 0.1 mM
phenylmethysulfonyl fluoride (PMSF), 1 µM pepstatin A, 17 µg/ml
calpain I, and 7 µg/ml calpain II using a Polytron PT 3000 (position
20) with two bursts of 15 s each. Separation into cytoplasmic and
membrane fraction was achieved by centrifugation of the homogenate at
100,000 x g for 2 h using an Optima TL centrifuge
equipped with rotor TLA 100.3 (Beckman Instruments). Membrane pellets
were resuspended in 1.5 ml of buffer F with a glass-TeflonTM
homogenizer and spun down.
In vitro phosphorylation
Cardiac homogenate proteins (0.5 mg) were solubilized with 1%
digitonin in the presence of protease inhibitors as described
previously (11)
. The mixture was supplemented with 17
µg/ml calpain I, and 7 µg/ml calpain II. Solubilized proteins were
diluted (1:3) with 20 mM Tris/HCl buffer, pH 7.4, containing the
inhibitors and bound to heparin agarose beads (30 µl) equilibrated
with buffer G (0.1% digitonin, 20 mM Tris/HCl buffer, pH 7.4, 80 mM
NaCl, and the inhibitor mixture). The beads were washed twice with
buffer G and the phosphorylation was performed by incubation of the
beads in a final volume of 250 µl with 0.6 µM catalytic subunit of
PKA and 10 µM [-32P]ATP (2–3 nCi/pmol)
for 3 min at 30°C in a medium containing 40 mM HEPES/Tris-buffer, pH
7.4, 10 mM MgCl2, 0.1% digitonin, 1 mM EGTA,
0.1% bovine serum albumin. Cytosolic proteins from biopsies were
phosphorylated under the same conditions on partial enrichment of
calcium channels by binding on heparin agarose. Membrane proteins from
cardiac biopsies were phosphorylated in parallel without binding to
heparin agarose. Phosphorylated membrane proteins were solubilized with
radioimmunoassay (RIA) buffer consisting of 1% Triton X-100, 25 mM
Tris/HCl, pH 7.4, 0.15 M NaCl, and 0.5 mg/ml bovine serum albumin (free
of IgG) and immunoprecipitated.
Immunoprecipitation
Phosphorylated proteins were eluted from heparin agarose beads
with 0.5 M NaCl in RIA buffer and immunoprecipitated with the different
sequence-directed antibodies. Antibody beads were produced by
incubation of 10–20 µg affinity-purified antibody IgG with
preswollen protein A-Sepharose (5 mg) in 0.5 ml RIA buffer for 2 h
at room temperature on a rotating wheel. The beads were washed twice
with RIA buffer, mixed with the phosphorylated proteins, and the
incubation was continued overnight at 4°C. The supernatants were
removed and the beads were subsequently washed twice with RIA buffer
and twice with TBS. Immunoprecipitated proteins were extracted by
sodium dodecyl sulfate (SDS) sample buffer, heated for 2 min at 95°C,
and subjected to SDS-polyacrylamide gel electrophoresis (PAGE) and
autoradiography using Kodak X-omat films.
Peptide sequence analysis
Homogenate proteins (60 mg) from untreated pig hearts were
phosphorylated by PKA and immunoprecipitated with the
ßcom antibody by scaling up the protocol
described above. Precipitated proteins were resolved on a 6.5%
polyacrylamide gel and stained with Coomassie blue. The pp700
corresponding to dried gel bands of four lanes were excised from the
gel with a scalpel, cut in small 1-mm gel cubes, washed in ammonium
bicarbonate/acetonitrile (1:1), and dried under vacuum. After reducing
and alkylating by iodoacetamide, proteolytic digestion was performed at
37°C overnight with 0.2 µg of trypsin (Boehringer, Mannheim) in 300
µl 0.2 M ammonium bicarbonate buffer containing 0.02% Tween-20
(34)
. The resultant peptides were recovered from the gel
pieces by incubating with freshly prepared 2% trifluoroacetic acid
(TFA) at 60°C for 1 h. Eluted peptides in the supernatant were
bound to reversed phase C4 material fixed in an Eppendorf tip and from
there recovered into a small volume of 60% acetonitrile/0.1% TFA.
After approximately fourfold concentration of the eluent in a Speed
Vac, peptides were separated by narrow-bore reverse-phase
high-performance liquid chromatography (HPLC
SMARTTM System from Pharmacia, Uppsala, Sweden)
applying a gradient with increasing concentration of acetonitrile
(buffer C, 0.1% TFA in acetonitrile) from 5 to 45% for 65 min in
buffer A (0.1% TFA in water) and a flow rate of 100 µl/min at room
temperature. The radioactivity of the fractions was measured with a
scintillation counter (LS 6000LL, Beckman Instruments). Peptide
identification in the radioactive fractions was performed by automated
sequence analysis by an ABI Precise Protein Microsequencer (Applied
Biosystems, Foster City, CA). The amino acid sequences of the selected
peptide fractions were compared for similarities to the sequences of
the database (URL address:
http://vega.igh.cnrs.fr/bin/fasta-guess.cgi).
SDS-PAGE and immunoblot analysis
Protein samples were resolved on appropriate polyacrylamide gels
and transferred to nitrocellulose filters according to standard
procedures. The transfers were incubated with affinity-purified primary
antibodies at a concentration of 0.3 to 1 µg IgG/ml for 90 min and
the secondary peroxidase-conjugated anti-rabbit antibody.
Immunoreactive protein bands were visualized with an ECL reaction kit.
Immunocytochemistry
Rat cardiac tissue was prepared as described previously with
slight modifications (35)
. Briefly, adult Wistar rats were
killed by decapitation and hearts rapidly removed. Small pieces of left
ventricular tissue were fixed with 4% formaldehyde (freshly prepared
from paraformaldehyde) in buffer D (0.1 M phosphate buffer, pH 7.4,
containing 5% sucrose) for 30 min at room temperature. After washing
in buffer D, tissue pieces were infiltrated with 2.3 M sucrose and
rapidly frozen in liquid nitrogen (36)
. Semithin sections
of 1-µm thickness were prepared with an Ultra-Cut S ultramicrotome
equipped with a Cryo-Cut FC4S cryoattachment (Leica) and collected on
silane-coated glass slides. To suppress unspecific labeling, the
sections were incubated for 30 min with buffer E consisting of 20 mM
Tris-HCl, pH 7.4, 130 mM NaCl, 0.05% Tween-20, 0.02%
NaN3, and 1% BSA. Immunolabeling was performed
with anti-AHNAK antibodies raised against the KIS epitope diluted in
buffer E to a protein concentration of 30 µg/ml. The primary antibody
was visualized with a Cy3-labeled donkey anti-rabbit-IgG antibody
(Dianova) diluted to 0.2 µg/ml with buffer E. Incubations were
performed overnight at room temperature and 2 h at 37°C,
respectively. Washing steps were carried out with buffer E containing
additional 0.5 M NaCl. Sections were evaluated with an Axioplan
fluorescence microscope (Zeiss) equipped with appropriate filter
systems. Micrographs were taken with an automatic camera (Zeiss) on
Fujichrome Provia 400 film.
Cell culture
Primary heart cell cultures were prepared from ventricular
tissue of 1- to 3-day-old rats by tryptic disintegration as described
(37)
. The SM20-I culture medium was supplemented with 2.76
mM hydrocortisone, 10% fetal calf serum, 2 mM glutamine, 0.002 mM
fluorodeoxyuridine, and 0.02 mg/ml gentamicin. The cells were cultured
on polylysine-coated coverslips for 2–4 days, washed with phosphate
buffer, pH 7.4, and fixed with acetone/methanol (50 v/v) for 5 min at
-20°C. Immunolabeling was performed for 1 h at 37°C with KIS
antibodies at a concentration of 10 µg/ml diluted in buffer E.
Nonspecific labeling was suppressed as described above for the
cryosections. In double labeling experiments, KIS antibodies (10
µg/ml) were applied in combination with monoclonal antibodies against
cardiac specific proteins: anti-MHC mAb 2C9 (diluted 1:100) or
anti-RyR2 mAb (10 µg/ml). Secondary antibodies were Cy3-conjugated
goat anti-mouse IgG (H+L) and DTAF-conjugated goat anti-rabbit IgG
(H+L) from Dianova diluted to 10 µg/ml in buffer E. Cardiomyocytes
were evaluated with a Nikon-Diaphot (Tokyo, Japan) microscope. A
Bio-Rad MRC 600 confocal imaging system (Bio-Rad Laboratories,
Freiburg, Germany) with an argon/krypton laser was used.
Stimulation of PKA and PKC in cultured cardiomyocytes
To stimulate the PKA of cardiomyocytes with isoproterenol, the
culture medium was removed and the cells were washed three times with
HEPES buffer, pH 7.4. Freshly prepared isoproterenol was then added in
this buffer to a final concentration of 10 µM. Stimulation was
stopped after 10 min by addition of cold acetone/methanol. Stimulation
of PKC was performed by addition of phorbol-12,13-dibutyrate for 2 h in SM20-I culture medium at a final concentration of 10 µM.
RNA preparation, reverse transcriptase-polymerase chain reaction
(RT-PCR), and sequencing
Total RNA was prepared from frozen cardiac tissue by the
guanidinium thiocyanate procedure (38)
. One-microgram
aliquots of RNA were converted to first-strand cDNA with 10 pmol of
random hexamers and superscript reverse transcriptase (1 unit, Life
Technologies, Inc.-BRL). Oligodeoxynucleotide primers were designed to
match sequences of human AHNAK (Acc.No. M90902). The following primers
were combined: f1 (CTCGAAGCTCCAGGTCACCATG corresponding to nt 385–406)
and r1 (GTCTCTATGTCCACTCTGGAG complementary to nt 1275–1295); f2
(CTCCAGAGTGGACATAGAGAC corresponding to nt 1275–1295) and r2
(TGCTTTGAACCTGGCACA complementary to nt 1936–1954). The primers r1,
f2, and r2, but not f1, also matched sequences of desmoyokin (Acc.
No. X65157). PCR reactions contained 5 µl of RT reaction as template,
10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, 20 µM
each dATP, dCTP, dGTP, dTTP, 1 µmol each primer, and 1.5 units
Taq-DNA polymerase (Life Technologies, Inc.-BRL). Reactions
were performed with a thermocycler for 32 cycles under the following
conditions: 94°C (1 min); 60°C (1 min); 72°C (90 s); and final
extension at 72°C (10 min). The amplified cDNA fragments were
analyzed by 1% agarose gel electrophoresis and ethidium bromide
staining. Ten to twenty nanograms of gel-resolved PCR fragments were
commercially sequenced on both strands by InViTek (Berlin) using the
ABIPRISM dye terminator cycle sequencing method.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-32P]ATP, and immunoprecipitated
with affinity-purified ßcom antibodies. A
typical experiment shown in Figure 1
illustrates that a phosphoprotein of unusually large size was recovered
in the immunoprecipitates in addition to the calcium channel ß
subunit. Phosphorylation and identity of the 85-kDa
ß2 subunit was documented by autoradiography
and by immunorecognition with an independent
ß2-specific antibody (Fig. 1)
. In contrast, the
large PKA substrate did not cross-react either with
anti-ß2 (Fig. 1)
or with
anti-1C antibodies (data not shown). Based on
the comparison with high-molecular-mass proteins (560-kDa RyR2, 800-kDa
nebulin, data not shown), the ß2-associated
phosphoprotein has an apparent molecular mass of 700 kDa. It was,
therefore, designated as phosphoprotein 700 (pp700). Among mammalian
heart preparations, pp700 was identified in different species on
precipitation with anti-ß2 subunit antibodies,
i.e., in human, rat, and pig (Fig. 1)
, as well as in dog and mouse
(data not shown).
|
The cardiac PKA substrate, pp700, is homologous to AHNAK
To identify pp700 of pig hearts, it was partially sequenced after
in vitro phosphorylation by PKA, immunoprecipitation with
ßcom, and subsequent trypsin digestion. The
resulting peptide map is shown in Figure 2
. Fractions containing radioactive peptides (T26, T36, T41, T45, T47)
were chosen for sequencing. Although the fractions consisted of
clustered peptides, they yielded enough structural information for an
effective protein database search. All peptides sequenced matched
predicted amino acid sequences of AHNAK (Table 1
), a human gene encoding a 700-kDa phosphoprotein (18
, 19)
.
|
|
To clarify the identity of pp700, sequence-directed antibodies were
raised against the human AHNAK-encoded protein. The epitopes were
located within the repetitive sequence motifs and were designated as
KIS and FEN (19)
. Both anti-AHNAK antibodies
immunoprecipitated pp700 of pig heart, albeit KIS precipitated pp700
more efficiently than FEN (Fig. 3A
). Moreover, pp700 of pig hearts was recognized by
anti-AHNAK antibodies on immunoblots after its coprecipitation with the
calcium channel antibodies anti-ßcom and
anti-ß2 (Fig. 3B
). Antibodies
directed against the 1C subunit (Fig. 3B
) as well as against the ß1b- and
ß3-subunits (data not shown) were unable for
pp700 coprecipitation. These findings suggest that pp700 interacts
tightly with the ß subunit of cardiac calcium channels, but not with
the channel-forming 1C subunit.
|
AHNAK gene expression in the myocardium
RT-PCR was performed to estimate AHNAK gene expression in human
and rat myocardium. The primer combination f1/r1 flanking the predicted
AHNAK head portion gave no amplification product (data not shown),
whereas cDNA products of expected size were obtained using downstream
primers f2/r2. Alignment of the partial cDNA from normal human heart
showed 100% identity to the genomic AHNAK sequence. The rat cardiac
cDNA was highly homologous to AHNAK and mouse desmoyokin predicting 81
and 87% amino acid sequence identity, respectively (Fig. 4
). The AHNAK cDNAs amplified from human and rat heart started 1276 and
63 nucleotides downstream from predicted translation initiation sites
of AHNAK and desmoyokin, respectively. They contained two potential
phosphorylation sites for PKA, RLGS21P and
RIS32M. The latter consensus PKA motif is
conserved in human and rat heart.
|
Localization of pp700/AHNAK in rat cardiomyocytes
As a first step toward the understanding of a possible functional
role of pp700/AHNAK, we determined AHNAK location in adult rat hearts
by immunofluorescence microscopy. Incubation of semithin cryosections
of left ventricular tissue with the AHNAK antibody, KIS, resulted in
intense labeling of cardiomyocytes in the region of the plasma membrane
as seen in longitudinal (Fig. 5A
, arrowheads) and cross sections of cardiomyocytes (Fig. 5B
). The labeling pattern includes the intercalated discs as
visible in longitudinal sections (Fig. 5A
, large arrows). In
addition to cardiomyocytes, endothelial cells of capillaries were also
labeled by the KIS antibody (Fig. 5
, small arrows). In both cell types
nuclei remained unlabeled. Hence, in ventricular tissue of adult rat
hearts pp700/AHNAK is detected in cardiomyocytes and endothelial cells,
with preferred sarcolemmal location of the protein in cardiomyocytes.
|
The topology of pp700/AHNAK was further studied by immunofluorescence
microscopy of isolated cultured neonatal rat cardiomyocytes. To assess
the origin of cardiac cells, double labeling experiments were performed
using the AHNAK antibody, KIS, in combination with antibodies against
the cardiac-specific proteins, cardiac MHC and RyR2 (Fig. 6
). Confocal images demonstrate that AHNAK localized primarily to the
plasma membrane of cultured neonatal rat cardiomyocytes (Fig. 6)
. A
weaker diffuse cytoplasmic staining was also visible. The nuclei
remained unlabeled. Preferential plasma membrane labeling by KIS
antibodies remained unaltered in cultured cardiomyocytes after
stimulation of endogenous PKA and PKC using isoproterenol (10 min) and
phorbol-12,13-dibutyrate (2 h), respectively (data not shown).
|
Membrane-localized pp700/AHNAK is the preferential substrate for
endogenous PKA
To determine whether pp700 undergoes in vivo
phosphorylation on ß-adrenergic receptor stimulation, we performed
two-step "back-phosphorylation" experiments. In the first step,
intact pig cardiac muscle was exposed to intramyocardial infusion of
isoproterenol to stimulate in vivo phosphorylation with
endogenous (unlabeled) ATP pools. Cardiac samples were then homogenized
and separated into membrane and cytosolic fractions. In the second
step, the fractions were incubated with
[-32P]ATP and PKA to assay the in
vitro phosphorylation. A diagram of this approach is given in
Figure 7
. The rationale was that sites that are in vivo
phosphorylated should be blocked from subsequent in vitro
phosphorylation with radioactive phosphate.
|
Indeed, membrane preparations derived from isoproterenol-treated
biopsies showed blunted 32P incorporation into
pp700 in vitro compared with controls (Fig. 7
, membranes,
lanes 1, 2). Strong suppression of pp700 in vitro
phosphorylation was observed in immunoprecipitates of both AHNAK and
ßcom antibodies. Five minutes after cessation
of isoproterenol infusion, in vitro
32P incorporation into pp700 increased slightly,
demonstrating that pp700 undergoes reversible in vivo
phosphorylation (Fig. 7
, membranes, lanes 3).
Additional membrane-associated PKA substrates were recovered in the
immunoprecipitates. A 170-kDa protein was obtained by both AHNAK- and
ßcom-directed antibodies; 120- and 60-kDa
proteins were recovered by the ßcom antibody.
It is important that only the 60-kDa PKA substrate (pp60) showed
in vivo phosphorylation, whereas the 170- and 120-kDa
phosphoproteins did not (Fig. 7
, membranes). The pp60 closely resembles
the PKA substrate of dog hearts, which has been previously described by
our group (6
, 7)
. In contrast, the membrane-associated
85-kDa ß2 subunit was hardly in
vitro phosphorylated among the experimental groups (Fig. 7
,
membranes, ßcom Ab), indicating that this
ß2 subunit exhibits a high in vivo
phosphorylation state even under basal conditions. This observation is
in accordance with results from heterologous expression models
(9
, 15)
. Hence, these findings suggest that
ß2 subunit isoforms exist that are
differentially targeted by ß-adrenergic signaling.
Two major in vitro targets of PKA were observed among the
cytoplasmic fractions. The pp700 was recovered by AHNAK antibodies and
the ß2 subunit was obtained by
ßcom antibodies. Notably, only a small fraction
of pp700 was coprecipitated by ßcom antibodies
(Fig. 7
, cytoplasm, ßcom Ab). Moreover,
cytoplasmic pp700 and membrane-associated pp700 differed in the
phosphorylation pattern. Strikingly, in vitro
[32P]-phosphate incorporation into pp700 was
very similar among the experimental groups (Fig. 7
, cytoplasm, AHNAK
Ab), indicating that their PKA phosphorylation sites were not targeted
in vivo. The ß2 subunit also showed
no remarkable differences in [32P]phosphate
incorporation (Fig. 7
, cytoplasm, ßcom Ab).
These results demonstrate that pp700 underwent fractional in
vivo phosphorylation on activation of the endogenous PKA pathway.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Specific
coprecipitation of pp700 and ß2 subunit is
observed with several cardiac preparations spanning the phylogenetic
distance from rodent to human heart. The pp700 and
ß2 subunit exhibit tight interaction because
the complex remains associated in RIA buffer containing 1% Triton
X-100, i.e., conditions that dissociate the channel subunits
1C and ß2. Under
conditions known to preserve
1C/ß2 interaction,
both ß2 and ßcom
antibodies precipitate efficiently the DHP-labeled
1C subunit (29
, 30)
. Thus,
coprecipitation analyses suggest a rank order of binding affinity of
the ß2 subunit that is
pp700/ß2 >
1C/ß2. The inability
of anti-1C antibodies to precipitate pp700
does not necessarily mean that pp700 and 1C
subunit do not interact. One scenario might be that high-affinity
interaction between pp700 and ß2 subunit plays
a structural role to locate pp700 in a position that allows
low-affinity interaction to functional relevant sites of
1C. A similar concept was recently introduced
for the interaction between calcium channel 1C
and ß subunits where the stable 1C/ß
association via conserved binding domains is required for membrane
targeting of 1C, i.e., the structural
component, whereas other, as yet unidentified low-affinity interaction
sites are thought to modulate calcium channel gating properties
(39)
.
AHNAK expression in the mammalian myocardium
The pp700 is similar, if not identical, to the 700-kDa
phosphoprotein encoded by the human gene AHNAK (Swiss-Prot: Q09666).
This finding was surprising because AHNAK has been described as
neuroblast differentiation-associated protein. It was cloned as one of
the genes whose expression is typically repressed in neuroblastoma cell
lines in comparison to other, more differentiated cells of
neuroectodermal origin (18)
. The encoded protein has been
characterized in these cell lines (19)
and in
keratinocytes (20
, 21)
, but cardiac AHNAK expression has
not been addressed so far. In this study, we provide independent lines
of evidence for AHNAK gene expression in cardiomyocytes. A partial cDNA
amplified by RT-PCR was 100% identical to the generic AHNAK sequence.
The cDNA amplified from rat heart exhibits a higher homology to
desmoyokin (20)
than to AHNAK, which likely reflects
species-specific alterations of the AHNAK/desmoyokin gene. Because we
could not amplify human cardiac cDNA with primers spanning the 5' AHNAK
region, the sequence encoding the amino-terminal AHNAK portion may be
spliced in cardiac transcripts. Rather, our results confirm the
translation initiation site of desmoyokin.
Partial amino acid sequence analysis of pp700 and immunodetection of pp 700 by antibodies raised against synthetic AHNAK peptides confirm AHNAK expression in mammalian hearts as well as the similarity between pp700 and AHNAK protein.
Subcellular topology of pp700/AHNAK
In cardiomyocytes, the pp700/AHNAK protein localizes
preferentially to the cell boundary. A similar cell surface location of
AHNAK/desmoyokin was observed in frozen sections of normal human skin
(20)
. This is in contrast to AHNAK/desmoyokin localization
in the nucleus of human cell lines like HeLa, epithelial, melanoma, and
neuroblastoma cells (19)
, and its localization in the
cytoplasm of murine, human, and canine epithelial cell lines
(21)
. After homogenization of cardiac tissue AHNAK
antibodies precipitated pp700 preferentially from the cytoplasmic
fraction. Because purified cardiac sarcolemma do not react with AHNAK
antibodies (Haase et al., unpublished results), we conclude that
pp700/AHNAK is loosely attached to the plasma membrane and is easily
lost during membrane fractionation procedures. This is consistent with
the predicted molecular structure of AHNAK, which has no transmembrane
helices as assessed by an updated version of the hidden Markov model
(40)
. The plasma membrane localization of pp700/AHNAK
remains unaltered in isolated cardiomyocytes after short-term
stimulation of PKA and PKC. This is in contrast to features of
AHNAK/desmoyokin in keratinocytes, where PKC stimulation induces AHNAK
translocation from cytoplasm to the plasma membrane (21)
.
The pp700/AHNAK protein is an in vivo target of PKA
AHNAK has eight potential phosphorylation sites for PKA
(R-X1/2-S/T-X)
(41)
; this prediction is consistent with in
vitro phosphorylation of pp700. The substitution of arginine for
lysine is known as the major deviation in substrate recognition by PKA
(41)
. Obviously, the PKA accepts this substitution to
phosphorylate serine residues within the (K)FS motif of AHNAK variants
C and the (K/R)IS motif of AHNAK variants E in vitro; both
motifs are found among the sequenced radioactive pp700 fractions. It is
interesting to speculate that the PKA may be more discriminating in
cardiomyocytes than in vitro. Back-phosphorylation
experiments support this notion, demonstrating that the majority of
in vitro PKA phosphorylation sites is not targeted in
vivo on isoproterenol stimulation. Taking into consideration that
RIS may be favored in vivo over KIS of recurrent motifs, the
RIS323M site of AHNAK is a leading
candidate to play a physiological role in ß-adrenergic signal
transduction.
An important finding of this study is that cardiac pp700/AHNAK
undergoes fractional in vivo phosphorylation on stimulation
of the ß-adrenergic receptor by isoproterenol. Membrane-associated
pp700 was the preferential PKA target in vivo. These results
are consistent with the hypothesis that scaffold proteins like AKAP79
anchor PKA to physiologically relevant sites within cardiac myocytes
(14)
. It is important that, in vivo
phosphorylation occurs in the pp700 fraction that is tightly bound to
the ß subunit of cardiac calcium channels.
Taken together, we report on the expression of a novel cardiac PKA substrate, pp700, which is homologous to AHNAK and couples to the ß2 subunit of cardiac calcium channels. This interaction probably accounts for pp700/AHNAK phosphorylation by endogenous PKA on stimulation of the ß-adrenergic receptor.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 subunit due to phosphorylation by cAMP dependent protein kinase. Proc. Natl. Acad. Sci. USA 90,10135-101391 subunit. J. Biochem. 120,170-1761 subunit of the cardiac L-type Ca channel by cAMP-dependent kinase. Biochemistry 35,10392-10402[Medline]
-CaMKII isoforms. J. Cell. Biochem. 68,259-268[Medline]
1B and ß subunits during rat brain ontogeny. J. Biol. Chem. 273,14495-14502B-crystallin in rat and human heart. Circulation 96,3466-34761C subunit are independent functions of the ß subunit. J Physiol. (London) 517,353-368This article has been cited by other articles:
|
|
 |
|
  Y. Huang, A. de Morree, A. van Remoortere, K. Bushby, R. R. Frants, J. T. Dunnen, and S. M. van der Maarel Calpain 3 is a modulator of the dysferlin protein complex in skeletal muscle Hum. Mol. Genet., June 15, 2008; 17(12): 1855 - 1866. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. H. Lee, H. J. Lim, S. Yoon, J. K. Seong, D. S. Bae, S. G. Rhee, and Y. S. Bae Ahnak Protein Activates Protein Kinase C (PKC) through Dissociation of the PKC-Protein Phosphatase 2A Complex J. Biol. Chem., March 7, 2008; 283(10): 6312 - 6320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Huang, S. H. Laval, A. van Remoortere, J. Baudier, C. Benaud, L. V. B. Anderson, V. Straub, A. Deelder, R. R. Frants, J. T. den Dunnen, et al. AHNAK, a novel component of the dysferlin protein complex, redistributes to the cytoplasm with dysferlin during skeletal muscle regeneration FASEB J, March 1, 2007; 21(3): 732 - 742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Haase Ahnak, a new player in {beta}-adrenergic regulation of the cardiac L-type Ca2+ channel Cardiovasc Res, January 1, 2007; 73(1): 19 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Haase, J. Alvarez, D. Petzhold, A. Doller, J. Behlke, J. Erdmann, R. Hetzer, V. Regitz-Zagrosek, G. Vassort, and I. Morano Ahnak is critical for cardiac Ca(v)1.2 calcium channel function and its {beta}-adrenergic regulation FASEB J, December 1, 2005; 19(14): 1969 - 1977. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A.G. van der Heyden, T. J.M. Wijnhoven, and T. Opthof Molecular aspects of adrenergic modulation of cardiac L-type Ca2+ channels Cardiovasc Res, January 1, 2005; 65(1): 28 - 39. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Alvarez, J. Hamplova, A. Hohaus, I. Morano, H. Haase, and G. Vassort Calcium Current in Rat Cardiomyocytes Is Modulated by the Carboxyl-terminal Ahnak Domain J. Biol. Chem., March 26, 2004; 279(13): 12456 - 12461. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Komuro, Y. Masuda, K. Kobayashi, R. Babbitt, M. Gunel, R. A. Flavell, and V. T. Marchesi The AHNAKs are a class of giant propeller-like proteins that associate with calcium channel proteins of cardiomyocytes and other cells PNAS, March 23, 2004; 101(12): 4053 - 4058. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Benaud, B. J. Gentil, N. Assard, M. Court, J. Garin, C. Delphin, and J. Baudier AHNAK interaction with the annexin 2/S100A10 complex regulates cell membrane cytoarchitecture J. Cell Biol., January 5, 2004; 164(1): 133 - 144. [Abstract] [Full Text] |
