Cardiovascular Research 61 (2004) 87 – 93
www.elsevier.com/locate/cardiores
Decreased protein and phosphorylation level of the protein phosphatase
inhibitor-1 in failing human hearts
Ali El-Armouche a,1, Torsten Pamminger b,1, Diana Ditz b, Oliver Zolk b, Thomas Eschenhagen a,*
a
Institute of Experimental and Clinical Pharmacology, University Hospital Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany
b
Friedrich-Alexander-University, Erlangen-Nuremberg, Germany
Received 27 August 2003; received in revised form 6 November 2003; accepted 7 November 2003
Time for primary review 26 days
Abstract
Objective: The protein phosphatase inhibitor-1 (I-1) is a highly specific and potent inhibitor of type 1 phosphatases (PP1) that is active
only in its protein kinase A (PKA)-phosphorylated form. I-1 ablation decreases, I-1 overexpression sensitizes h-adrenergic signaling in the
heart. It is controversial whether I-1 expression is altered in human heart failure (HF), likely because its detection in heart is difficult due to its
low abundance. Methods and results: I-1 was >500-fold enriched from left ventricular myocardium (LVM) from patients with terminal HF
(n = 16) and non-failing controls (NF, n = 5) and quantified with an affinity-purified I-1 and a I-1 phosphospecific antiserum. In non-failing I1 protein levels amounted to 126 fmol/mg protein. In failing hearts, I-1 protein levels were reduced by 58% and I-1 phosphorylation by 77%
( P < 0.001 vs. NF). I-1 phosphorylation correlated well with serine-16 phosphorylation of phospholamban (PLB) in the same hearts
( P < 0.001). In contrast, PLB, troponin I (TnI) and PP1 protein and TnI phosphorylation levels did not differ between HF and NF.
Conclusions: The results suggest that the reduction in I-1 protein and phosphorylation in failing human hearts leads to increased phosphatase
activity which in turn may result in reduced phosphorylation of cardiac proteins such as PLB.
D 2003 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
Keywords: Heart failure; Protein phosphatases; Phospholamban; Troponin I; Adrenergic signal transduction
1. Introduction
The protein phosphatase inhibitor-1 (I-1) is an interesting
potential modulator of the phosphorylation/dephosphorylation balance. It acts as a potent and highly specific inhibitor
of protein phosphatase type 1 (PP1) only when phosphorylated by the cAMP-dependent protein kinase A (PKA). As
such it could provide amplification of PKA-mediated signals [1].
Despite its early discovery and its fairly defined role
in h-adrenergic inhibition of glycogen synthesis [1] the
role of I-1 in the heart remained obscure for long.
Recent studies now show that targeted ablation of I-1
* Corresponding author. Tel.: +49-40-42803-2180; fax: +49-40-428034876.
E-mail address: [email protected] (T. Eschenhagen).
1
Both authors contributed equally to this work.
increased PP1-activity, reduced isoprenaline-stimulated
phospholamban (PLB) phosphorylation and impaired hadrenergic contractile responses in mouse heart [2].
Conversely, adenoviral overexpression of I-1 or a constitutively active I-1 mutant sensitized cardiac myocytes
to h-adrenergic stimulation [2,3]. Overexpression of I-1
was associated with increased isoprenaline-stimulated
PLB phosphorylation [3]. These data substantiated an
amplifier role of I-1 in h-adrenergic signaling of cardiac
myocytes.
Uncertainty remains as to the expression level of I-1 in
human heart failure (HF). One study concluded that I-1
protein levels were unchanged, but I-1 phosphorylation
was reduced in samples from patients with HF [2]. Our
own data showed reduced I-1 transcript levels and failed to
detect a specific I-1 protein signal in standard homogenates
[3]. The present study reevaluated this question by utilizing an extraction protocol and an affinity-purified I-1
antiserum.
0008-6363/$ - see front matter D 2003 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.cardiores.2003.11.005
88
A. El-Armouche et al. / Cardiovascular Research 61 (2004) 87–93
reviewed and approved by the Ethical Committee of the
University Hospital Hamburg (Az. 532/116/9.7.1991).
2. Methods
2.1. Recombinant I-1 protein
2.3. Enrichment of I-1
Recombinant I-1 protein was generated by cloning the
complete cDNA of rat heart I-1 into the pGEX-ET expression vector (Stratagene) as described previously [3]. In
short, the plasmid was transformed into competent Escherichia coli BL21 (Novagen) to express I-1/glutathione-Stransferase (I-1 – GST). I-1 – GST was purified from clarified
bacterial lysate by affinity purification with glutathione
agarose beads. Loaded beads were washed three times,
and recombinant I-1 was released by thrombin-cleavage.
2.2. Human myocardial tissue
Failing hearts were obtained from patients undergoing
heart transplantation due to terminal HF (n = 9 dilated
cardiomyopathy, DCM, and n = 7 ischemic cardiomyopathy,
ICM). LV ejection fraction was 16 – 25%, cardiac index 1.7–
2.7 l/min m2. Most patients received ACE inhibitors,
diuretics and cardiac glycosides, seven received calcium
channel blockers, 13 nitrates and three antiarrhythmic drugs
in addition. No patient received h-blockers. Five non-failing
donor hearts (NF) that could not be transplanted for technical reasons were used for comparison. Donor patient
histories or echocardiography revealed no signs of heart
disease (for detail see Table 1). The study conforms with the
principles outlined in the Declaration of Helsinki and was
I-1 was enriched from mouse skeletal muscle (wild-type
[WT] and I-1 knockout [I-1( / )]), rat and rabbit skeletal
muscle (500 mg each) and human left ventricular myocardium(LVM, 1 g) by an optimized extraction procedure [3–
6]. In short, frozen tissue was pulverized with a mortar in
liquid nitrogen, homogenized with a teflon/glass potter in
ice-cold 1.5% trichloroacetic acid (TCA [wt./vol.], 4 mmol/
l EDTA) and centrifuged at 20,000 g for 30 min. The
supernatant was adjusted to 19% TCA, incubated at 4 jC for
12 h, and centrifuged as above. The pellet was resuspended
in 500 mmol/l Tris (pH 8.0), boiled for 10 min and
centrifuged as above. The supernatant was dialyzed for 16
h against water at 4 jC, centrifuged as above and stored at
80 jC for further use. Protein was determined according
to Bradford.
2.4. Western blot
SDS-PAGE and blotting were carried out as described
[3]. Membranes were blocked with 5% (w/v) dried milk in
100 mmol/l Tris, pH 7.5, 0.1% (v/v) Tween 20 and 150
mmol/l NaCl (TBST) for 1 h prior to overnight incubation at
4 jC with the primary antibodies. Primary antibodies were a
custom-made (Eurogentec, Brussels) rabbit polyclonal af-
Table 1
Patient data
Patient #
Age
Gender
Diagnosis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
44
52
50
42
19
47
65
56
44
59
32
n.d.
45
48
66
64
46
54
64
57
65
M
M
M
F
M
M
M
M
M
M
M
n.d.
M
M
M
M
M
M
M
M
M
NF (SAB)
NF (ICB)
NF (SAB)
NF (CIC)
NF (SAB)
DCM
DCM
DCM
DCM
DCM
DCM
DCM
DCM
DCM
ICM
ICM
ICM
ICM
ICM
ICM
ICM
NYHA class
IV
IV
III – IV
IV
III
IV
n.d
IV
III – IV
III – IV
IV
III – IV
III – IV
III – IV
III – IV
IV
LVEF (%)
16
17
25
n.d
n.d.
40
n.d.
n.d.
n.d.
16
20 – 30
n.d.
n.d.
22
n.d.
20
CI (l/min m2)
Drugs
2.1
n.d.
1.7
2.7
1.7
3.5
n.d.
1.3
1.4
1.9
n.d.
n.d.
n.d.
1.8
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
DGNCA
DNR
DGNCA
DGNCAR
DGNCA
DGNA
n.d.
NC
DGNC
DNCAR
DNA
n.d.
DGNR
DGAO
DGN
DGNR
LVEF, left ventricular ejection fraction; CI, cardiac index. Diagnosis: ICB, intracerebral bleeding; SAB, subarachnoidal bleeding; CIC, cerebral ischemia;
DCM, idiopathic dilated cardiomyopathy; ICM, ischemic cardiomyopathy; NF, non-failing donor. Drugs: A, angiotensin converting enzyme inhibitors or
angiotensin II receptor antagonists; C, calcium channel blockers; D, diuretics; G, cardiac glycosides; N, nitrates; R, antiarrhythmics (except h-AR blockers); O,
dopamine/dobutamine; n.d., unknown.
A. El-Armouche et al. / Cardiovascular Research 61 (2004) 87–93
finity-purified antibody against recombinant full-length rat
I-1 that cross-reacts with mouse, rabbit and human I-1 (FB
70, 1/1000), and antibodies against rabbit skeletal muscle I1 (1/200, kindly provided by P. Greengard, G 184), phospho-DARPP-32 (1/2000, Cell Signaling Technology, Beverly, MA), calsequestrin (CSQ, 1/2500; Dianova, Hamburg,
Germany), PP1 (1/500, Upstate, Lake Placid, NY), cardiac
troponin I (TnI, 1/30,000, Chemicon, Temecula, CA),
phospho-TnI (1/30,000, HyTest, Turku, Finland), total
PLB and Ser16-phosphorylated PLB (both 1/5000; PhosphoProtein Research, Bardsey, UK). Immunoblots were
developed with anti-rabbit or anti-mouse IgG-horseradish
peroxidase, subjected to Enhanced Chemiluminescence
PLUS detection reagents (Amersham) and exposed to film
for appropriate times. Densitometric signals on X-ray films
were evaluated with GelDoc (Biorad).
89
2.5. Statistical analysis
Data are mean F S.E.M. Statistical analysis was performed using ANOVA or Students t-test, correlation analysis with Spearman – Rank correlation. P < 0.05 was
considered significant.
3. Results
3.1. Detection and quantitation of I-1 protein in
immunoblots
To identify a specific ‘‘I-1 band’’, immunoblots of
standard homogenates for both WT and I-1( / ) heart
(30 Ag) were probed with G 184 I-1 antibody (against rabbit
Fig. 1. (A) Immunoblots of standard heart homogenates (30 Ag) for WT and I-1( / ) were probed with G 184 I-1 antibody (against rabbit skeletal muscle I-1)
and G 184 incubated with 37.5 Ag recombinant I-1 (for preadsorption, top) as well as with the FB 70 I-1 antibody (against rat heart I-1) and its preimmune
serum (bottom). (B) Immunoblot of TCA-extracts for WT and I-1( / ) skeletal muscle (10 Ag, f 25 mg wet weight) were probed with FB 70 I-1 antibody.
Note, the immunoreactive signal at f 29 kDa in WT skeletal muscle is absent in the I-1( / ) tissue. (C) Immunoblot of TCA-extracts enriched in I-1 isolated
from rat skeletal muscle (5.25 Ag), rabbit skeletal muscle (3.3 Ag) and human LVM (6.0 Ag) were probed with the FB 70 I-1 antibody or for control with
preimmune serum. Note, a protein band of f 29 kDa (rat) and f 26 kDa (rabbit, human) were detected by the I-1 antibody, but not by the preimmune serum.
(D) Immunoblot to quantitate absolute I-1 in TCA-extracts from non-failing human LVM (range of 3.5 – 14 Ag) using the FB 70 I-1 antibody and increasing
amounts for recombinant I-1 for calibration (range of 6.25 – 25 ng).
90
A. El-Armouche et al. / Cardiovascular Research 61 (2004) 87–93
skeletal muscle I-1) and FB 70 I-1 antibody (against rat
heart I-1) and for negative control with G 184 incubated
with 37.5 Ag recombinant I-1 protein (for preadsorption) or
FB 70 preimmune serum. As shown in Fig. 1A, there was
no difference in the pattern of the immunoreacting bands.
This confirmed our previous report that I-1 protein is not
detectable in standard homogenates [3] and other published
data showing that TCA-extracts of at least one whole
guinea-pig heart were necessary for detection of I-1 by
Western blots [6,7]. In contrast, immunoblots of TCAextracts from WT and I-1( / ) skeletal muscle (10 Ag,
f 25 mg wet weight) probed with FB 70 I-1 antibody
showed a strong immunoreactive band at f 29 kDa in
skeletal muscle from WT but not from I-1( / ) mice (Fig.
1B). To further confirm the validity of this extraction
method and the specificity of detecting I-1, immunoblots
of TCA-extracts from rat skeletal muscle, rabbit skeletal
muscle and human LVM were probed with the FB 70 I-1
antibody or preimmune serum. As shown in Fig. 1C a band
of f 29 kDa (rat) or f 26 kDa (rabbit, human) was
detected by the I-1 antibody, but not by its preimmune
serum. The species-dependent difference in size of I-1 and
the apparent molecular mass are in accord published data
[6,8,9].
Fig. 2. (A) PP1 protein levels in homogenates from hearts with dilated (DCM) or ICM or donor hearts (NF) normalized to CSQ. (B) I-1 and phospho-I-1 in
TCA-extracts from the same hearts. Coomassie-stained gels run in parallel demonstrates equal protein loading and similar band pattern. *P < 0.05 vs. NF.
#
Samples and order are identical to (A).
A. El-Armouche et al. / Cardiovascular Research 61 (2004) 87–93
To get an estimate of the absolute level of I-1 in human
heart tissue, a dilution of recombinant rat I-1 (6.25 –25 ng)
was probed with the FB 70 antiserum in parallel with a
dilution of TCA-extract from a non-failing heart (3.5 –14
Ag; Fig. 1D). The intensity of the immunoreactive band at
f 29 kDa of 12.5 ng recombinant I-1 was almost identical
to the band at f 26 kDa of 10.5 Ag TCA-extract protein
(according to 34.2 mg wet weight tissue and 5.3 mg protein
in the initial homogenate). Under the assumption that the
antiserum detects human I-1 with the same affinity as rat I-1
this experiment finds I-1 levels in non-failing human hearts
at 126 fmol/mg protein or 126 nM. This concentration is
well above the IC50 of PP1 inhibition (1 nM).
91
3.2. PP1, I-1 protein and I-1 phosphorylation levels in
failing human hearts
Protein levels of protein PP1 were analyzed in total
homogenates and normalized to CSQ (20 Ag per lane). No
differences were observed (Fig. 2A). I-1 protein and phosphorylation levels were determined in TCA-extracts. 1 g
LV-tissue yielded 120 F 4 mg protein in the initial homogenate and 234 F 13 Ag protein in the TCA-extract (n = 21),
with no differences between the groups. Immunoblots of
these extracts (7 Ag per lanei30 mg wet weight) demonstrated that the signal intensity was significantly smaller in
failing than in non-failing hearts, independent of the etiol-
Fig. 3. (A) Representative blots of total TnI, phospho-TnI, total PLB and phospho-serin-16-PLB in standard homogenates of the same hearts as in Fig. 1. The
blot for CSQ, PP1 (Fig. 1A), TnI and phospho-TnI is identical, the blot for PLB and phospho-PLB (and CSQ) was run in parallel. (B) Correlation of phospho-I1 with phospho-TnI and phospho-PLB, respectively.
92
A. El-Armouche et al. / Cardiovascular Research 61 (2004) 87–93
ogy with a mean reduction of 58% (Fig. 2B, left panel).
Immunoblots of the same homogenates were probed with an
I-1 phosphospecific antibody [10] and demonstrated a mean
reduction of phospho-I-1 in failing hearts by 77% (Fig. 2B,
right panel). Normalization to internal control proteins was
impossible after TCA extraction, but Ponceau S staining of
blots and Coomassie-stained gels run in parallel confirmed
equal protein loading (Fig. 2B, top). The lack of systematic
differences in the band pattern argues against confounders
due to TCA extraction.
3.3. Protein and phosphorylation levels of PLB and TnI in
failing human hearts
PLB and TnI protein and phosphorylation levels were
analyzed in standard homogenates (20 Ag per lane) from the
same samples as above and normalized to CSQ (Fig. 3A).
PLB protein, TnI protein and TnI phosphorylation were
similar in the groups (NF vs. HF: PLB/CSQ 5.58 F 0.60 vs.
5.73 F 0.25; TnI/CSQ 0.63 F 0.02 vs. 0.64 F 0.03; Phospho-TnI/TnI 1.14 F 0.05 vs. 1.13 F 0.04). However, phospho-PLB/PLB ratio was significantly reduced in HF (NF
1.11 F 0.32, HF 0.51 F 0.10, P < 0.05). Phospho-PLB positively correlated with phospho-I-1 (Spearman r = 0.70,
P < 0.001, Fig. 3B).
4. Discussion
This study demonstrates that I-1 is markedly decreased
and dephosphorylated, and therefore, inactive in failing
human hearts. The reduction in I-1 correlates well with
dephosphorylation of PLB. The study took advantage of the
fact that I-1 is heat stable and, in contrast to most other
proteins, not precipitated by low concentrations of TCA [4].
TCA-extracts were prepared under conditions where I-1
phosphorylation state is preserved [4,5]. Immunoblots of
these extracts demonstrated a prominent band at f 26 kDa
that fulfilled the necessary criteria of specificity (see
above). We believe that the band, we and others have seen
before in standard homogenates [2,3] represents a nonspecific cross-reacting protein. This is supported by the
following arguments. (i) The band migrates at slightly
higher molecular weight (f 29 vs. 26 kDa), (ii) is seen
with preimmune serum, (iii) is not blocked by preincubation of the antiserum with recombinant I-1, and (iv) is seen
in cardiac homogenates from I-1 knockout mice (Fig. 1).
One could argue that a protein that is seen in immunoblots
only after 500-fold enrichment is unlikely to play an
important role. However, phosphorylated I-1 is highly
potent (IC50 1 nM; [11]) suggesting that it does not need
to be expressed at high concentrations. Our present analysis
now shows I-1 levels in non-failing human heart to amount
to 126 nM (Fig. 1D). This is clearly above the published
IC50 and not far from estimates made for rat heart (f 500
nM; [4,12]).
Compared to other alterations in protein expression in
human HF, the magnitude of the reduction in I-1 protein
levels by 58% is large. Other well-studied proteins have
been found to be downregulated by only 25 –50% (e.g. h1adrenergic receptors, SR Ca2 + ATPase, PLB, potassium
channels, myosin light chains). We have found previously
that I-1 transcript levels in failing human hearts were
reduced by 50% [3]. In addition, isoprenaline-infusion in
rats led to a 30% downregulation of I-1 mRNA levels [13].
These data suggest that gene expression of I-1, similar to hadrenoceptors, is reduced by chronically elevated catecholamine levels in HF. I-1 phosphorylation was even more
drastically diminished ( 77%). This observation most
likely reflects desensitization of h-adrenergic signaling with
decreased cAMP levels and PKA activation in human HF
(for review see [14]). Another contributor to diminished I-1
phosphorylation could be increased calcineurin activity in
human HF [15], a phosphatase that dephosphorylates I-1.
Evaluation of further downstream elements of the hadrenergic signaling cascade revealed interesting results. In
the identical hearts in which I-1 was downregulated and
dephosphorylated and PLB was dephosphorylated, PP1
protein and TnI protein/phosphorylation levels were unaffected. Unchanged PP1 is in accordance with unchanged
global activity [16], but unchanged TnI phosphorylation
was somewhat counterintuitive because TnI is a substrate
for PKA just like PLB. Moreover, its phosphorylation level
can be experimentally increased by PP inhibition [17,18]
suggesting a role for PP1, and therefore, also I-1. Indeed,
studies in failing human hearts found basal TnI phosphorylation to be reduced [19,20], but others found it unchanged
in human HF [21] and in animal models [22]. Basal
phosphorylation is notoriously prone to experimental conditions, which may explain the discordant results. Nevertheless, the experiments have been performed in identical
samples, showing that phosphorylation of TnI is indeed less
affected by HF than that of PLB. The observation that
isoprenaline-stimulated TnI phosphorylation with a 20-fold
higher sensitivity than that of PLB [23] could offer an
explanation in the sense that basal phosphorylation of TnI
is high, despite h-adrenergic desensitization. In addition,
intracellular compartmentalization of PP1 and I-1 may play
a role. Marks and colleagues [24] have proposed differential
anchoring of PP1 to different compartments to explain
differences between ryanodine receptor and PLB phosphorylation in HF and it is tempting to speculate that I-1
participates in this process. The marked reduction in I-1
identified here will cause less inhibition of PP1. This does
not appear to translate into a globally increased PP1-activity
[16], but goes along with a 2.5-fold increased in SRassociated PP1-activity [16], suggesting that I-1 preferentially localize to the SR. The good correlation between PLB
and I-1 phosphorylation supports the notion of a causal
relationship and argues for preferential affection of the free
SR (PLB) compared with the junctional SR (ryanodine
receptors) or the myofilaments (TnI). Thus, the reduction
A. El-Armouche et al. / Cardiovascular Research 61 (2004) 87–93
in I-1 could be one of the factors that explain differential
affection of TnI and PLB phosphorylation in HF.
PLB is a major regulator of cardiac contractility [25] and
accelerated PP1-mediated dephosphorylation of PLB will
impair Ca2 + resequestration into the SR, reduce systolic
Ca2 + release, and elevate diastolic Ca2 +. The latter would
activate calcineurin, which dephosphorylates I-1 and thereby accelerates a vicious cycle. Thus, the reduction in I-1 is a
newly recognized factor that, besides changes in the adrenergic signaling cascade and sarcoplasmic reticulum function, can contribute to reduced systolic and diastolic
function and to the blunted inotropic response to h-adrenergic stimulation in human HF.
Acknowledgements
We thank Jutta Starbatty for providing expert technical
assistance. We also like to thank Dr. Paul Greengard and
Peter Ingrassia for I-1 knockout mice. This work was
supported by the Deutsche Forschungsgemeinschaft (Es 88/
8-2, GRK 750) and the Johannes und Frieda MarohnStiftung (Arm/00).
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
References
[1] Oliver CJ, Shenolikar S. Physiologic importance of protein phosphatase inhibitors. Front Biosci 1998;3:961 – 72.
[2] Carr AN, Schmidt AG, Suzuki Y, et al. Type 1 phosphatase, a negative regulator of cardiac function. Mol Cell Biol 2002;22:4124 – 35.
[3] El-Armouche A, Rau T, Zolk O, et al. Evidence for protein phosphatase inhibitor-1 playing an amplifier role in h-adrenergic signaling.
FASEB J 2003;17:437 – 9.
[4] Foulkes JG, Cohen P. The hormonal control of glycogen metabolism.
Phosphorylation of protein phosphatase inhibitor-1 in vivo in response to adrenaline. Eur J Biochem 1979;97:251 – 6.
[5] Neumann J, Gupta RC, Schmitz W, Scholz H, Nairn AC, Watanabe
AM. Evidence for isoproterenol-induced phosphorylation of phosphatase inhibitor-1 in the intact heart. Circ Res 1991;69:1450 – 7.
[6] Gupta RC, Neumann J, Watanabe AM, Lesch M, Sabbah HN. Evidence for presence and hormonal regulation of protein phoshatase
inhibitor-1 in ventricular cardiomyocytes. Am J Physiol 1996;270:
1159 – 64.
[7] Ahmad Z, Green FJ, Subuhi HS, Watanabe AM. Autonomic regulation of type 1 protein phosphatase in cardiac muscle. J Biol Chem
1989;264:3859 – 63.
[8] MacDougall LK, Campbell DG, Hubbard MJ, Cohen P. Partial structure and hormonal regulation of rabbit liver inhibitor-1; distribution of
[19]
[20]
[21]
[22]
[23]
[24]
[25]
93
inhibitor-1 and inhibitor-2 in rabbit and rat tissues. Biochim Biophys
Acta 1989;1010:218 – 26.
Lowenstein PR, Shering AF, MacDougall LK, Cohen P. Immunolocalisation of protein phosphatase inhibitor-1 in the cerebral cortex of
the rat, cat and ferret. Brain Res 1995;676:80 – 92.
Snyder GL, Girault JA, Chen JY, et al. Phosphorylation of DARPP-32
and protein phosphatase inhibitor-1 in rat choroid plexus: regulation
by factors other than dopamine. J Neurosci 1992;12:3071 – 83.
Endo S, Zhou X, Connor J, Wang B, Shenolikar S. Multiple structural
elements define the specificity of recombinant human inhibitor-1 as a
protein phosphatase-1 inhibitor. Biochemistry 1996;35:5220 – 8.
Iyer RB, Koritz SB, Kirchberger MA. A regulation of the level of
phosphorylated phospholamban by inhibitor-1 in rat heart preparations in vitro. Mol Cell Endocrinol 1988;55:1 – 6.
El-Armouche A, Jäckel E, Pamminger T, Eschenhagen T. Evidence
for protein phosphatase inhibitor-1 playing a role in h-adrenergic
desensitazion observed in heart failure. Naunyn-Schmiedeberg’s Arch
Pharmacol 2002;365:R88.
El-Armouche A, Zolk O, Rau T, Eschenhagen T. Inhibitory G-proteins and their role in desensitization of the adenelyl cyclase pathway
in heart failure. Cardiovasc Res 2003;60:478 – 87.
Molkentin JD, Lu JR, Antos CL, et al. A calcineurin-dependent transcriptional pathway for cardiac hypertrophy. Cell 1998;93:215 – 28.
Neumann J, Eschenhagen T, Jones LR, et al. Increased expression of
cardiac phosphatases in patients with end-stage heart failure. J Mol
Cell Cardiol 1997;29:265 – 72.
Neumann J, Herzig S, Boknik P, et al. On the cardiac contractile,
biochemical and electrophysiological effects of cantharidin, a phosphatase inhibitor. J Pharmacol Exp Ther 1995;274:530 – 9.
Boknik P, Khorchidi S, Bodor GS, et al. Role of protein phosphatases
in regulation of cardiac inotropy and relaxation. Am J Physiol Heart
Circ Physiol 2001;280:786 – 94.
Bodor GS, Oakeley AE, Allen PD, Crimmins DL, Ladenson JH,
Anderson PA. Troponin I phosphorylation in the normal and failing
adult human heart. Circulation 1997;96:1495 – 500.
Zakhary DR, Moravec CS, Stewart RW, Bond M. Protein kinase A
(PKA)-dependent troponin-I phosphorylation and PKA regulatory
subunits are decreased in human dilated cardiomyopathy. Circulation
1999;99:505 – 10.
Bartel S, Stein B, Eschenhagen T, et al. Protein phosphorylation in
isolated trabeculae from nonfailing and failing human hearts. Mol
Cell Biochem 1996;157:171 – 9.
Boknik P, Fockenbrock M, Herzig S, et al. Protein phosphatase
activity is increased in a rat model of long-term beta-adrenergic
stimulation. Naunyn-Schmiedeberg’s Arch Pharmacol 2000;362:
222 – 31.
Karczewski P, Bartel S, Krause EG. Differential sensitivity to isoprenaline of troponin I and phospholamban phosphorylation in isolated
rat hearts. Biochem J 1990;266:115 – 22.
Marx SO, Reiken S, Hisamatsu Y, et al. Phosphorylation-dependent
regulation of ryanodine receptors: a novel role for leucine/isoleucine
zippers. J Cell Biol 2001;153:699 – 708.
Koss KL, Kranias EG. Phospholamban: a prominent regulator of
myocardial contractility. Circ Res 1996;79:1059 – 63.