![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 294, Issue 2, 620-626, August 2000
Institut für Pharmakologie und Toxikologie, Universität Münster, Münster, Germany
 |
Abstract |
|---|
 Top
 Abstract
 Introduction
Materials and Methods
Results
Discussion
References
|
|---|
In this study we characterized the effects of the protein phosphatase
(PP) type 1 and type 2A inhibitor cantharidin (Cant) and its structural
analogs cantharidic acid and endothall on PP activity, force of
contraction, and myosin light chain phosphorylation in rat aorta. All
compounds inhibited PP activity in homogenates of rat aorta with a rank
order of potency of Cant = cantharidic acid > endothall.
However, only Cant increased force of contraction and myosin light
chain phosphorylation in intact isolated rat aortic rings. Based on
these findings, we investigated the effects of Cant on
-adrenoceptor-mediated vasoconstriction. Cant (1 and 3 µM)
enhanced norepinephrine-induced contraction in endothelium-intact rat
aorta. In contrast, Cant did not affect norepinephrine-induced contraction in endothelium-denuded rat aorta. We suggest that inhibition of PP1 and/or PP2A activities by Cant enhances vascular contractility in endothelium-intact rat aorta by increasing the phosphorylation state of endothelial regulatory proteins.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Phosphorylation
is an important post-translational modification of proteins. For
instance, the phosphorylation of myosin light chain
(MLC20) initiates smooth muscle contraction.
Myosin light chain kinase increases and myosin light chain
phosphatase(s) decrease(s) phosphorylation of
MLC20. The main myosin light chain protein phosphatases (PPs) are PP1 and PP2A, which are comprised of catalytic and one or more regulatory subunits. Different genes encode the catalytic subunits PP1, 1
, 1
, 2A, and 2A (for review,
see Shenolikar and Nairn, 1991
; Wera and Hemmings, 1995; Herzig and Neumann, 2000).
One can distinguish between PP1 and PP2A by their sensitivity toward PP
inhibitors. PP2A is more sensitive to okadaic acid (OA) and cantharidin
(Cant) than PP1. Cant is a less potent inhibitor of PPs than OA but
more economical. Other PP inhibitors such as calyculin A are equipotent
for PP1 and PP2A. However, additional structural derivatives of Cant
inhibit purified PPs. These compounds include cantharidic acid (CA) and
endothall (ETA; Erdödi et al., 1995; Laidley et al., 1996). They
might be useful for correlation of PP inhibition and physiological
function. CA was even claimed to be the active derivative of
Cant (Eldridge and Casida, 1995).
We used these tools to study the hypothetical interaction between PP
inhibition and contractile effects of catecholamines. For instance,
Cant attenuated the relaxant effect of -adrenoceptor stimulation in
bovine coronary arteries, probably by increasing phosphorylation of
MLC20 via inhibition of PP (Knapp et al., 1997). Therefore, we hypothesized that PP inhibition may enhance or attenuate the vasoconstrictory effect of norepinephrine (NE). However, NE did not
induce any vasoconstriction but, in contrast, led to vasorelaxation in
bovine coronary arteries (Knapp et al., 1997). Because NE induces vasoconstriction of rat aorta (Alosachie and Godfraind, 1988), herein
we used isolated rat aorta instead of bovine coronary artery. We first
studied whether PP1 and PP2A are enzymatically present in rat aorta and
investigated the effects of Cant, CA, and ETA on PP activity. For
comparison, we used the well characterized compound OA. Thereafter, we
tried to identify the catalytic subunits of PPs with antibodies.
The next question was whether the inhibitors could actually increase the contraction in aortic preparations and finally whether increased phosphorylation of MLC20 accompanies these contractions. Based on the results of these experiments, we studied the functional interaction of PP inhibition and NE-induced contraction of rat aorta.
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Rat Isolated Aortic Ring Preparation
Male Wistar rats (300-350 g) were sacrificed by cervical
dislocation. The thoracic aorta (aorta thoracica descendens) was dissected and transferred into Krebs-Henseleit solution (KHS) of the
following composition: 118 mM NaCl, 25 mM NaHCO3,
2.5 mM CaCl2, 4.7 mM KCl, 1.2 mM
KH2PO4, 1.2 mM
MgSO4, 11.1 mM glucose, and 0.026 mM
ethylenedinitrilotetraacetic acid continuously gassed with 95%
O2 and 5% CO2. Rat aortae
were cleaned from connective tissue and cut into four rings of an
approximate length of 4 mm. In indicated experiments, the endothelium
was removed by gentle application of 100 µl of Triton X-100 (1%)
followed by several rinses in KHS. Rings were individually mounted in
organ chambers containing KHS and were allowed to equilibrate in KHS
for 60 min under a resting tension of approximately 15 mN (Seasholtz et
al., 1997). After equilibration rings were contracted twice with KCl (75 mM; 25 min) without correction of NaCl in the KHS. After washout, the presence of an undamaged endothelium was checked by relaxation to
carbachol (CB; 1 µM; 10 min) after precontraction with NE (0.1 µM;
10 min) as described in Gray and Marshall (1992). Rings showing less
than 70% relaxation of the maximal NE effect were discarded as having
partially damaged endothelium. Supposedly endothelium-free rings
showing relaxation by CB (1 µM; 10 min) after precontraction with NE
(0.1 µM; 10 min) were excluded from further experiments.
Analysis of MLC20 Phosphorylation and Immunological Identification
The effects of Cant on MLC20
phosphorylation were determined in isolated rat aortic rings used in
the contraction experiments. Rings were freeze-clamped at the end of
contraction experiments (see Results) and stored at 
80°C
for approximately 1 week until biochemical studies. Analysis of
MLC20 phosphorylation by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) was performed as described
(Calovini et al., 1995; Knapp et al., 1999a; Waurick et al., 1999).
Isoelectric focusing (first dimension) was performed in glass
capillaries (12.5 cm in length; 1 mm i.d.) with a pH gradient from 4.5 to 5.4 (pharmalytes; Pharmacia, Uppsala, Sweden). Gels were run for
4.5 h at 600 V constant for the first dimension. The second
dimension was an SDS-electrophoresis with slab gels 10.5 × 9.5 cm, 1 mm in thickness. Gels were stained with Coomassie blue.
MLC20 spots were scanned, quantitated by using
ImageQuant software (Molecular Dynamics, Krefeld, Germany), and each
MLC20 spot was expressed as percentage of whole
MLC20 spots (=100%) detectable on one individual
gel (Waurick et al., 1999).
The identification of the different MLC20
isoforms was based on their different isoelectric points. Because
phosphorylation of MLC20 introduces negative
charges, analysis by 2D-PAGE leads to different
MLC20 isoforms having almost the same molecular
mass but different isoelectric points where the most basic form
(A) represents an unphosphorylated form (Haase and Morano, 1996). In
addition, because the number and assumed phosphorylation state (un-,
mono-, or biphosphorylated) of the other forms (B-F) varies in a
species- and tissue-specific manner (Haase and Morano, 1996) and has
not yet been identified in rat aorta, only relative changes of form A
are reported in this study.
Separated proteins were transferred to nitrocellulose membranes and
were incubated with monoclonal anti-myosin (light chains 20 kDa).
Proteins binding the antibody were visualized with alkaline phosphatase-conjugated goat anti-mouse IgM and color reagents (Knapp et
al., 1999a; Waurick et al., 1999). All MLC20
spots seen on Coomassie gels reacted with this
anti-MLC20 antibody, whereas the essential light
chains were not recognized by this antibody (data not shown).
Immunological Identification of PP1 and PP2A Catalytic Subunits
Rat aortic tissue (aorta thoracica descendens) was homogenized
and separated proteins were transferred to nitrocellulose membranes as
described recently (Knapp et al., 1998, 1999a,b). Antibodies against
the catalytic subunits of PP1 and PP2A at 2-µg/ml dilution in
Buffer A (13 mM Tris, 154 mM NaCl, pH 7.4, containing 5% nonfat dry milk powder) were incubated with the blot overnight. To demonstrate the specificity of bands, antibodies (2 µg/ml) were incubated with
corresponding immunizing peptides for 8 h in Buffer A (PP1, 6.6 µg/ml; PP2A, 16.6 µg/ml) and then incubated with the blot overnight. After several rinses in Tris-buffered saline/Tween 20, the nitrocellulose was incubated with
125I-goat anti-rabbit IgG (ICN Biomedicals,
Eschwege, Germany), diluted 1:1000 in Tris-buffered saline for 4 h
at room temperature. Radioactive bands were visualized in a
PhosphorImager (Molecular Dynamics, Krefeld, Germany).
Phosphatase Activity
Preparation of Homogenates.
Preparation of homogenates from
rat aortae was performed according to the method described previously
(Knapp et al., 1998). Aliquots of homogenates were used for
determination of phosphatase activity.
Phosphatase Assay.
Phosphatase activity was determined as
described previously (Neumann et al., 1993; Knapp et al., 1998) with
[32P]phosphorylase a as substrate.
The reaction was started by adding aliquots of homogenates or aliquots
of peak fractions. Reaction was stopped by addition of 50%
trichloroacetic acid. Precipitated protein was sedimented by
centrifugation and the supernatant was counted in a liquid
scintillation counter.
Protein Determination
Protein was measured according to the method of Bradford (1976).
Chemicals
Benzamidine, NE, CB, leupeptin, phenylmethylsulfonyl fluoride, and Cant were from Sigma (Deisenhofen, Germany). CA and ETA were obtained from Calbiochem (Bad Soden, Germany) and Alexis (Grünberg, Germany), respectively. All chemicals used for high-resolution 2D-PAGE were from Pharmacia Biotech (Freiburg, Germany). Anti-human PP1 (rabbit polyclonal IgG, lot no. 12641), anti-human PP2A (rabbit polyclonal IgG, lot no. 13949), and corresponding immunizing peptides (human PP1 peptide, lot no. 12802; and PP2A peptide, lot no. 13298) were from BIOMOL (Hamburg, Germany). The antibody directed against the regulatory light chains of myosin (mouse monoclonal anti-myosin light chains 20 kDa; clone MY-21) was obtained from Sigma. All other chemicals used were of analytical or best grade commercially available.
Statistics
Results are expressed as mean ± S.E. Significance was estimated by Student's t test for paired and unpaired observations as appropriate. A P value less than .05 was considered significant
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
PP Activity in Homogenates of Rat Aorta.
We measured
phosphatase activity in the absence of Ca2+ with
phosphorylase a as substrate (Fig.
1, control) where only PP1 and PP2A, but
not PP2B and PP2C, are active (Cohen, 1989). These data suggest the
presence of PP1 and/or PP2A.
|
; Neumann et al., 1995; Herzig and Neumann, 2000) is
structurally unrelated to Cant. OA attenuated phosphatase activity with
an IC50 value of approximately 3 nM
(n = 3) and completely blocked PP activity at 1 µM, a
concentration that is specific for the inhibition of PP1 and PP2A.
Cant and CA reduced phosphatase activity in homogenates from rat aorta
with IC50 values of approximately 300 nM. ETA was
less potent with an IC50 value of 2 µM but was
equieffective. Thus, Cant and CA are approximately 100-fold less potent
but as effective as OA (Fig. 1).
Immunological Identification of PP1 and PP2A Catalytic
Subunits.
Next, we wanted to identify the catalytic subunits of
PP1 and PP2A immunologically in rat aorta. Therefore, extracts from rat
aorta were subjected to gel electrophoresis and transferred to
nitrocellulose membranes. Several bands were visualized by the
anti-phosphatase antibodies. The corresponding immunizing peptides only
attenuated the prominent band at the expected molecular mass for PP1 or
PP2A at approximately 37 kDa (Fig. 2).
These data immunologically identify both PP1 and PP2A in the rat aorta.
|
Effects of Cant on Force of Contraction in Isolated Endothelium-Intact Rat Aortic Rings. Isolated rings were equilibrated in bathing solution for 60 min and contracted with KCl (75 mM) twice as described in Materials and Methods. After washout, the presence of an undamaged endothelium was checked. Thereafter, one single concentration of Cant or solvent was added to each arterial ring.
Cant increased force of contraction in a concentration- and time-dependent manner with an EC50 value of 6.5 µM (Fig. 3). Cant (1 µM, n = 7 and 3 µM, n = 20) did not affect force of contraction within 180 min compared with control rings (n = 10). At higher concentrations (5, 7, and 10 µM) Cant led to a slowly developing and sustained increase in force of contraction. The most rapid increase in force of contraction could be observed at 30 and 100 µM Cant. The force generated by Cant (10 µM, 72.3 ± 3.0 mN; 30 µM, 68.7 ± 2.9 mN; and 100 µM, 70.2 ± 2.7 mN) was similar to the force generated by stimulation with 75 mM KCl (70.5 ± 1.0 mN). Thus, Cant is equieffective as KCl in contracting rat aortic rings. Additional experiments in endothelium-denuded preparations revealed that the effects of Cant did not depend on the presence of the endothelium (data not shown).
|
MLC20 Pattern.
Under control conditions 2D-PAGE
resolved four MLC20 forms (A-D with increasing
acidity) in rat aortic rings (Fig. 4,
left), whereas 2D-PAGE separated six MLC20 forms
(A-F with increasing acidity) in rings frozen after treatment with 30 µM Cant for 180 min (Fig. 4, right).
|
). If we summarize form A and form B as unphosphorylated isoforms, the amount of the phosphorylated forms would be approximately 20%. This would be in agreement with data
from others where the amount of total phosphorylated
MLC20 in rat aorta is approximately 10 to 15%
(Jin et al., 1996; Takayama et al., 1996). Those studies used
one-dimensional urea gel electrophoresis for the measurement of
MLC20 phosphorylation. This experimental difference might explain the somewhat higher amount of unphosphorylated MLC20 (if summarizing form A and form B) in this
study compared with the results of others.
In isolated rat aortic rings treated with 7, 10, and 30 µM Cant,
2D-PAGE revealed up to six MLC20 forms. In these
experiments, the amount of the unphosphorylated form A decreased to
17.8 ± 2.0% (7 µM, n = 5), 12.9 ± 0.0%
(10 µM), and 9.2 ± 1.5% (30 µM, n = 4).
Thus, Cant concentration dependently increased force of contraction and
decreased the unphosphorylated MLC20 form A. Fittingly, CA and ETA failed to affect MLC20
phosphorylation in isolated contracting rat aorta (data not shown).
Effects of Cant on NE (0.1 µM)-Induced Contraction in
Endothelium-Intact Preparations.
First, cumulative
concentration-response curves for NE (0.1 nM-100 µM) in
endothelium-intact rat aorta were performed. In these experiments, 0.1 µM NE induced approximately half-maximal contraction of isolated
aortic preparations similarly as described [Alosachie and Godfraind,
1988 (pD2 = 7.51); Xu et al., 1998
(pD2 = 7.09)]. Thus, we used this
concentration (0.1 µM NE) to investigate the effects of Cant on
NE-induced contraction in rat aorta.
|
Effects of Cant on NE (0.1 µM)-Induced Contraction in
Endothelium-Denuded Preparations.
To investigate whether the
effects of Cant on NE-induced contraction depend on the presence of the
endothelium, endothelium-denuded isolated rings were equilibrated in
bathing solution for 60 min and contracted with KCl (75 mM) twice as
described above. After washout, we confirmed the successful removal of
the endothelium functionally. In control rings and rings that were
subsequently treated with 3 µM Cant, respectively, CB (1 µM) did
not relax rings after the first NE-induced contraction
(n = 7 each; Fig. 6).
|
Effects of Cant on NE (0.01-3 µM)-Induced Contraction in
Endothelium-Intact Preparations.
Treatment of endothelium-intact
rings with Cant concentration dependently increased force of
contraction induced by a single concentration (0.1 µM) of NE.
However, if Cant can alter NE-induced contraction, a
concentration-response curve for NE also would be predicted to be
shifted leftward. Therefore, we performed concentration-response curves
for NE in the absence and presence of Cant (1 and 3 µM). In these
experiments, NE (0.01, 0.03, 0.1, 0.3, 1, and 3 µM) was cumulatively
added for 10 min for each concentration. As expected, Cant produced a
nonparallel shift to the left of the concentration-response curve to NE
with an increase of the maximal response, indicating a functional
noncompetitive antagonism (Fig. 7).
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
The main new finding of this report is evidence for a possible link between PP inhibition and NE-induced vasoconstriction. This finding suggests that the endothelium mediates the interaction. We first discuss the characterization of PPs in the rat aorta, then their apparent interaction with NE-mediated vasoconstriction in this vessel, and finally, possible underlying mechanisms.
Identification of PP1 and PP2A in Rat Aorta.
We have
identified PP1 and PP2A by several criteria. Their enzymatic activity
was measurable in the absence of Ca2+ with
phosphorylase a as substrate, which is typical for these PPs
(Cohen, 1989). The IC50 values for Cant and OA
(Cohen et al., 1990; Honkanen, 1993) were comparable to that noted in
homogenates of other tissues and species (Eldridge and Casida, 1995,
Cant, IC50 = 250 nM [mouse femur muscle]; Gong
et al., 1992, OA, IC50 values = 10 nM
[guinea pig ileum] and 8 nM [rabbit femoral artery]). Finally, we
could identify the catalytic subunits for PP1 and PP2A immunologically.
, rabbit skeletal muscle). However, although CA and ETA were able to inhibit phosphatase activity in vitro, only Cant exerted functional effects in
isolated rat aortic rings. The ineffectiveness of CA and ETA is
probably due to impermeability through the cell membrane in smooth
muscle. It could be argued that in contrast they might be membrane
permeant but that their toxic effects limit their contractile
functions. However, 75 mM KCl contracted rat aorta in the continuous
presence of CA and ETA (data not shown). Hence, toxic effects are
unlikely to play a major role. Others claimed that both Cant and ETA
were not permeable across the plasmalemma (Erdödi et al., 1995,
1996) because they did not induce marked morphological changes in 3T3
fibroblasts. Although this impermeability may be the problem for ETA,
this does not hold true for Cant in smooth muscle and cardiac
preparations (Neumann et al., 1995; Linck et al., 1996; Knapp et al.,
1999a,b; this article).
Biochemical experiments demonstrated the effectiveness of Cant in
contracting tissue. It is important to note that these powerful experiments measured force and MLC20
phosphorylation in the very same tissue, which greatly facilitates
their interpretation. Whereas CA and ETA did not affect force of
contraction and the phosphorylation state of
MLC20, Cant increased force and
MLC20 phosphorylation in the same preparations
and this was therefore in all likelihood mediated by PP inhibition.
Based on these data we could use Cant as a tool to study the
interaction of PPs and NE-induced contraction in smooth muscle.
Effects of Cant on NE-Induced Contraction.
Stimulation of
-adrenoceptors in the smooth muscle may involve activation of MLC
kinase but also G-protein-coupled inhibition of
MLC20 phosphatase (Kitazawa et al., 1991; Shirazi
et al., 1994; Somlyo and Somlyo, 1994).
, coronary
arteries; Urabe et al., 1991, rat isolated mesenteric and femoral
arteries). Especially, in isolated rat aorta, the removal of
endothelium from rat aortic rings significantly increases maximal
responses and decreases EC50 values for
-adrenoceptor agonists such as noradrenaline and phenylephrine
(Alosachie and Godfraind, 1988). Thus, the reduced
noradrenaline-induced contraction in the presence of endothelium
depicted in Fig. 5 is in accordance with the observations of many other investigators.
Apparently, the NE-induced contraction in endothelium-intact rings is
mediated by stimulation of 1-adrenoceptors,
whereas the subsequent relaxation is mediated by stimulation of
endothelial 2-adrenoceptors, resulting in
higher [Ca2+]i and
increased activation of endothelial nitric-oxide synthase (NOS). Thus,
nitric oxide (NO) can attenuate
1-adrenoceptor-mediated vasoconstriction
(Bockman et al., 1996). Hence, it is tempting to speculate that
endothelial NOS (eNOS) activity is reduced by Cant, most probably by
phosphorylation (Sase and Michel, 1997).
Indeed, phosphorylation can reduce NOS activity. Phosphorylation of
eNOS by AMP-activated protein kinase or protein kinase C inhibited NOS
activity (Bredt et al., 1992; Chen et al., 1999). Other explanations
also exist. Sutliff et al. (1999) demonstrated the expression of
phospholamban (PLB) in the endothelium of mouse aorta. Unphosphorylated
PLB inhibits Ca2+ uptake (Simmerman and Jones,
1998; Bokník et al., 1999) into the endoplasmic reticulum,
resulting in higher
[Ca2+]i and increased
activation of eNOS (which is activated by Ca2+).
Phosphorylation of PLB increases Ca2+ uptake into
the endoplasmic reticulum. This lowers
[Ca2+]i and should reduce
NO production and thereby would attenuate any endothelium-dependent
relaxation. Conceivably, Cant can increase the phosphorylation state of
PLB in endothelial cells of rat aorta as described in cardiomyocytes
(Neumann et al., 1995). Moreover, PP1 and PP2A are present in
endothelial cells, are inhibited by in vitro Cant, and Cant enters
endothelial cells because it increases the phosphorylation state of
MLC20 in these cells (Knapp et al., 1999b).
It is possible or even likely that additional regulatory proteins
besides eNOS and PLB are phosphorylated in endothelial cells in the
presence of Cant and their phosphorylation may enhance the contraction
by NE in rat aorta. Further work is required to elucidate the
underlying mechanism(s). In summary, the data presented herein support
a possible link between PP inhibition and NE-induced contraction that
is mediated by the endothelium.
| |
Acknowledgments |
|---|
The skillful technical assistance of Insa Post and Barbara Prystaj is gratefully acknowledged.
| |
Footnotes |
|---|
Accepted for publication May 1, 2000.
Received for publication December 21, 1999.
1 This study was supported by the Deutsche Forschungsgemeinschaft and the Interdisziplinäres Zentrum für Klinische Forschung, Münster.
Send reprint requests to: Dr. med. Jörg Knapp, Institut für Pharmakologie und Toxikologie, Westfälische Wilhelms-Universität Münster, Domagkstraße 12, D-48129 Münster, FRG. E-mail: jknapp{at}uni-muenster.de
| |
Abbreviations |
|---|
MLC20, regulatory light chains of
myosin (20 kDa);
PP, serine/threonine protein phosphatase;
OA, okadaic
acid;
Cant, cantharidin;
CA, cantharidic acid;
ETA, endothall;
NE, norepinephrine;
KHS, Krebs-Henseleit solution;
CB, carbachol;
2D-PAGE, two-dimensional polyacrylamide gel electrophoresis;
DMSO, dimethyl
sulfoxide;
NOS, nitric-oxide synthase;
NO, nitric oxide;
eNOS, endothelial NOS;
PLB, phospholamban;
pD2, log
EC50.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
-adrenoceptors in the rat aorta.
Br J Pharmacol
95:
619-629[Medline].-adrenoceptor vasorelaxation in rat thoracic aorta.
Br J Pharmacol
107:
684-690[Medline].-adrenoceptor-mediated vasorelaxation.
Br J Pharmacol
120:
421-428[Medline].1-adrenoceptor subtypes in aortas of 12-month-old spontaneously hypertensive rats.
Eur J Pharmacol
344:
31-36[Medline].
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
) and OA (
) on phosphatase activity (n = 3) in homogenates of rat aorta. Phosphatase activity is expressed as
a percentage of solvent control (Ctr). The concentration of solvent
(DMSO) was constant under all experimental conditions. Abscissa,
concentrations of Cant or OA. Ordinate, phosphatase activity as a
percentage of Ctr. 
denotes the first significant differences versus
Ctr.
;
n = 9), or the solvent DMSO as control (Ctr; 
