Volume 170, number 2
FEBS 1464 May 1984
The calmodulin fraction responsible for contraction in an
intestinal smooth muscle
J.C. Riiegg, G. Pfitzer, M. Zimmer* and F. Hofmann*
II. Physiologisches Institut and *Pharmakologisches Institut, Universitiit Heidelberg, Im Neuenheimer Feld 366,
6900 Heidelberg, FRG
Received 5 April 1984
Freeze-dried fibers of smooth muscle from Taenia coli were used to determine the concentration of
calmodulin responsible for contraction. About 10% of the total intracellular calmodulin (12.6 pmol/kg
wet wt) is directly involved in initiation of smooth muscle contraction.
Calmodulin Myosin light chain kinase Smooth muscle Contraction control Taenia coli
1. INTRODUCTION
The phosphorylation of myosin light chain-2 by
myosin light chain kinase (MLCK) is thought to
trigger smooth muscle contraction. MLCK is pre-
sent in smooth muscle at a concentration of
1 amol/kg wet wt [l] or 2.7 pM and is activated by
calmodulin (CaM) in the presence of Ca”. In vitro
experiments indicated that the [Ca2’] required to
activate either MLCK [2] or contraction of skinned
smooth muscle fibers [3] is inversely correlated
with the [CaM]. In vitro activation of smooth mus-
cle MLCK is further controlled by CAMP-
dependent phosphorylation of MLCK, the
modification of which increases the [CaM] re-
quired to activate the enzyme half maximally from
3 to 50 nM [4]. The [CaM] of bovine uterus
smooth muscle is about 37 PM [5], a concentration
at which the activation of MLCK and the initiation
of contraction should occur at [Ca”] below
0.1 ,uM. In contrast, contraction of living smooth
muscle fibers has not been observed at such low
Ca2+-concentrations [6]. This discrepancy can be
resolved if only part of the total CaM participates
in the activation of MLCK. Using smooth muscle
fibers from T. coli - which were skinned by
freeze-drying [7] - we found that the fibers con-
tain 12.6rmoVkg wet wt or 34pM CaM, but
Published by Elsevier Science Publishers B. V.
about 3-4rM CaM is freely available for the ac-
tivation of MLCK and the initiation of contrac-
tion. This value indicates that functionally no ex-
cess of CaM over MLCK exists in smooth muscle.
2. METHODS
2.1. Preparation of fibers and measurement of
contraction
Smooth muscle fibers from guinea pig T. coli
were shock frozen and freeze-dried at - 20°C as in
[7]. Freeze-dried fibers, approximately 0.5 cm in
length and 0.1-0.2 mm in diameter, were attached
at one end to an AME force transducer and at the
other end to an adjustable micrometer drive.
Fibers were then rehydrated for 1.5 min in an
ATP-free salt solution containing 5 mM EGTA
(rigor solution), followed by an incubation for
10 min (0) or 45 min (u) in relaxing solution
(20 mM imidazole, 4 mM EGTA, 10 mM MgClz,
7.5 mM ATP, 1 mM NaN3, 2 mM DTE, 10 mM
creatinephosphate, 380 units/ml creatine phospho-
kinase, Boehringer, pH 6.7, T = 20”(Z), thereafter
they were exposed to the [CaM] indicated. Con-
traction was initiated by raising the free Ca2+ to
1.6 pM [3]. Tension development (fig.1, table 1)
was measured after 3 min and expressed as a
percentage of the maximum contraction obtained
00145793/84/.$3.00 0 1984 Federation of European Biochemical Societies
383
brought to you by COREView metadata, citation and similar papers at core.ac.uk
provided by Elsevier - Publisher Connector
Volume 170, number 2
FEBS LETTERS May 1984
in the presence of 30,~M Ca’+.
2.2. Determination of calmodulin
Freeze-dried fibers (10 mg dry wt) were in-
cubated at 20°C in 250~1 ATP-free salt solution
containing 4 mM EGTA (rigor solution). This
solution was replaced after 1.0 min by relaxing
solution. Fibers were next incubated for 10 min or
60 min. Thereafter, fibers were homogenized in
350 ~1 extraction solution containing 1% (w/v)
Lubrol WX [8]. Samples were further processed as
in [B] and CaM concentration was determined by
activation of a bovine heart phosphodiesterase us-
ing pig brain CaM as standard. The molecular
mass of CaM was taken as 16.7 kDa.
3. RESULTS
3.1. Fibers skinned by freeze-drying
In initial experiments the basic properties of the
freeze-dried fibers were determined. Rehydrated
fibers relaxed completely in the presence of Mg-
ATP and a free Ca2+
of less than 0.1 PM. They
responded like other types of skinned fibers rapid-
ly and reversibly to micromolar Ca2+ (fig. 1, inset).
Contraction was not elicited by the addition of caf-
feine suggesting that intracellular calcium stores
were non-functional. Contraction was initiated at
threshold concentrations of free Ca2+ of about
1 ,uM. Maximal tension development of freshly
hydrated fibers amounted to OS-l.0 kg/cm2 at a
free Ca2+ of 30pM.
0 I@ o-__.. d,~‘- -~~-~~ ----v
025 05 10
25 50 10
Calmodultn ipMl
Fig.1. Effect of added calmodulin on the contractile
response in freeze-dried smooth muscle fibers. Fibers
were extracted for 0.5 min (0), 10 min (0) and 45 min
(m). The contractile response was then measured in the
absence of added calmodulin (CaM) (0) or in the
presence of the indicated concentration of CaM (0, n ).
The tension developed by the fibers in the absence of
CaM (0) was significantly 0, < 0.05) higher than that
obtained after 10 min extraction in the absence of CaM.
Each point represents x f SE for 7-10 different fibers
and each fiber was used only for one [GM]. Statistical
analysis was performed by the use of a Student’s t-test.
Inset:
reversibility and time course of isometric
contraction cycles elicited by 30pM Ca2+ (at arrow 1).
Isometric contractile responses of freeze-dried
fibers initiated by 1.6 FM Ca2+ decreased with time
and could no longer be elicited when fibers were
preincubated for 45 min in the relaxing solution
Table 1
Contraction of fibers is abolished by preincubation
Ca2+
GM)
10
Extraction time (min)
45
45 + CaM
(070 of maximal contraction)
1.6
21.5 f 6.0 (4) 0.2 f
2 (5) 20.3 f 7.7 (5)
3.5
56.7 + 5.7 (5) 1.9 f
4 (5) 53.7 f 14.9 (5)
Freeze-dried fibers were treated as described in section 2. They were incubated
for 10 or 45 min in relaxing solution. A third set of fibers was first incubated
for 45 min in relaxing solution and then transferred for another 10 min to a
fresh relaxing solution containing 0.5 ,uM CaM. After the extraction period
indicated, all fibers were immersed in a solution containing the indicated free
[Ca”]. Each fiber was used once and values are given as X f SE with the
number of individual fibers in parentheses
384
Volume 170, number 2
FEBS LETTERS
May 1984
(table 1) before challenging with Ca’+. Tension
development in the presence of a maximal Ca2+ of
30pM was not affected. Original contraction
responses were restored after incubation of the
fibers for a few minutes in the presence of 0.5 PM
CaM. This suggested that during the preincubation
period a part of CaM necessary for contraction
was removed from the fibers by diffusion into the
bath medium. The effect of added CaM on con-
traction was again reversed by incubation of these
fibers in a CaM free relaxing solution for 10 min.
This suggests that the freeze-dried fibers were free-
ly permeable to proteins of a molecular mass of
20 kDa but retained larger proteins such as
MLCK.
3.2. Biochemical assays
In the next series of experiments we therefore
determined the total concentration of CaM and
that part of CaM which exchanged rapidly with the
fiber bath medium in the presence of a low free
Ca2+.
Total CaM was 12.6pmol CaM/kg wet wt
which corresponds to an intracellular concentra-
tion of CaM of 34 PM if one assumes that CaM is
only distributed in the intracellular water compart-
ment (table 2). This value is similar to that
reported previously for uterine smooth muscle [5]
and probably describes the total CaM of T. coli
correctly. About 50% of the total CaM was ex-
tracted from the fibers during a 60 min incubation
period indicating that one half of the total CaM is
tightly bound to non-diffusible structures in the
presence of low Ca
2+. Incubation of the fibers for
0.5 min in rigor solution extracted 1% of the total
CaM. Fibers incubated for an other lo-min period
in relaxing solution lost 8% of their CaM which
corresponds to an intracellular CaM of 2.7 ,uM. In
a different set of experiments it was determined
that lo-13% (n = 8) of the total CaM was lost dur-
ing a l-2 min incubation period. These findings
indicated - together with the result that the effect
of added CaM was reversed within 10 min - that
only the freely exchangeable part of CaM was ex-
tracted during the first 10 min. This, therefore,
suggests an intracellular free CaM of about 3 PM
or a tenth of the total CaM.
3.3. Bioassay using skinned fibers
In freeze-dried skinned fibers extracted for
either 10 or 45 min in relaxing solution to remove
endogenous calmodulin, we have established the
relationship between the concentration of added
calmodulin and the relative force evoked at 1.6 pM
Ca2+.
Fig. 1 shows that a calmodulin concentration
of 0.3-l pM is required to develop the same ten-
sion as the force (27.6 + 3.9%; n = 10) evoked by
Table 2
The concentration of intracellular calmodulin
Extraction
Calmodulin extracted
Calmodulin concentration (uM)
Medium
Time
pg/mg dry wt
070
pmol/kg
Extraction
Intracellular
(min)
wet wt
medium
water
Total
_
1.24 + 0.08 (6) 100
12.6”
b
34.0c
Rigor
0.5
0.015 + 0.003 (7)
:::
0.15
0.036
0.4
Relaxation
10
0.10 * 0.02 (3)
1.0
0.23
2.7
Relaxation
60
0.57 f 0.15 (3)
46.0
5.8
1.3
15.6
* The [CaM]/kg wet wt was calculated from the CaM/dry wt by correcting for the lost water (83%, n = 3)
b The [CaM] in the extraction medium was calculated from the CaM/dry wt by correcting for the used dry wt of
fibers (10 mg) and the extraction volume (250 ~1)
’ Intracellular [CaM] was calculated from the [CaM]/kg wet wt using an intracellular water space of 37% of wet
wt [9, lo]. This last value assumes that the total amount of CaM is only distributed in the intracellular water space.
This assumption is certainly not correct since about 50% of the total [CaM] was only extracted in the presence
of detergent, suggesting that this part was largely bound to membranes (see also [8])
All values are corrected for buffer blanks. Original values are given as X f SE with the number of fibers in
parentheses
385
Volume 170, number 2
FEBS LETTERS
May 1984
Ca2+
in a very briefly rehydrated fiber in the
absence of exogenous calmodulin. This means that
in freeze-dried fibers rehydrated for only 1.5 min
in relaxing solution, the endogenous ‘free’
calmodulin available for eliciting contraction had a
concentration of about 0.3-l FM. If appropriate
corrections are made for diffusional losses during
the brief rehydration (table 1) an upper value of
about 3-4pM may be estimated for intracellular
concentration of the free calmodulin responsible
for contraction in living smooth muscle. At this
calmodulin level a Ca2+ concentration of 1.5 PM
would evoke a nearly maximal (i.e., 80%)
contraction.
4. CONCLUSIONS
The results shown in fig.1 and table 2 strongly
indicate that both the free intracellular CaM and
that part of the intracellular CaM directly involved
in contraction control are identical and are about
4 PM. This CaM is very similar to that of MLCK
if the value given in [l] is corrected for the in-
tracellular water space [9,10]. This suggests that
functionally both proteins are present in approx-
imately equimolar concentrations. In the presence
of calcium some part of the free CaM may be
bound to ‘caldesmon’ [ 111. This protein has been
identified in gizzard muscle and it has been sug-
gested that complexation of caldesmon by Ca2+
and CaM is necessary for the initiation of contrac-
tion. It is not known if intestinal smooth muscle
contains caldesmon and what function, if any, is
regulated by this protein.
Nine tenths of the total amount of smooth mus-
cle CaM is not directly involved in the control of
tension development. The function of this large
amount of CaM is unknown. It certainly
represents a compartment which binds Ca2+
without eliciting contraction. From table 2 it can
be calculated that at least 50 pmol Ca2+/kg wet wt
need to be released during contraction to saturate
the 4 Ca2+ binding sites of CaM and thereby to ac-
tivate the MLCK completely. This value is within
the range of that amount of Ca2+ which is released
from intracellular stores during contraction of
smooth muscle [12]. It is, therefore, conceivable
that the large excess of CaM represents a sink of
Ca2+ which prevents tension development by small
amounts of released Ca2+. This consideration
together with the relatively low concentration of
free CaM available for the initiation of contraction
suggests a high calcium requirement for smooth
muscle contraction.
ACKNOWLEDGEMENTS
We thank MS C. Zeugner for expert technical
assistance, MS I. Berger for typing the manuscript
and the Deutsche Forschungsgemeinschaft for sup-
porting this work.
REFERENCES
111
PI
131
[41
151
161
[71
PI
[91
WI
1111
WI
Adelstein, R.S. and Klee, C.B. (1981) J. Biol.
Chem. 256, 7501-7509.
Blumenthal, D.K. and Stull, J.T. (1980)
Biochemistry 19, X08-5614.
Sparrow, M.P., Mrwa, U., Hofmann, F. and
Rtiegg, J.C. (1981) FEBS Lett. 125, 141-145.
Conti, M.A. and Adelstein, R.S. (1981) J. Biol.
Chem. 256, 3178-3181.
Grand, R.J.A., Perry, S.V. and Weeks, R.A.
(1979) Biochem. J. 183, 285-295.
Neering, I.R. and Morgan, K.G. (1980) Nature
288, 585-587.
Stienen, G.J.M., Gtith, K. and Rtiegg, J.C. (1983)
Pfliigers Arch. 397, 272-276.
Kakiuchi, S., Yasuda, S., Yamazaki, R., Teshima,
Y., Kauda, K., Kakiuchi, R. and Sobue, K. (1982)
J. Biochem. 92, 1041-1048.
Casteels, R. (1969) J. Physiol. 204, 193-208.
Jones, A.W., Somlyo, A.P. and Somlyo, A.V.
(1973) J. Physiol. 232, 247-273.
Sobue, K., Miramoto, Y., Fugita, M. and
Kakiuchi, S. (1981) Proc. Natl. Acad. Sci. USA 78,
5652-5655.
Bond, M., Somlyo, A.V.,
Shuman, H. and
Somlyo, A.P. (1983) Proc. Int. Union Physiol. Sci.
XV, 261.
386
