474
chapter 
21 
Muscle and Nonmuscle Contractile Systems
summarizes the control mechanisms in smooth muscle that 
are discussed below.
MLCK is a very specific kinase whose only known 
physiological substrate is MLC in myosin II. The only 
phosphate donor is Mg-ATP. Smooth muscle MLCK (ac-
tually found in almost all cells) phosphorylates LC
2 0
at 
Ser19. There is also a skeletal muscle-specific MLCK and 
an embryonic MLCK. The various MLCKs are about 
950-1000 amino acids long. MLCK has functional do-
mains for: calmodulin binding, actin binding, substrate 
(ATP and myosin) binding, a catalytic domain, several 
target sites for various protein kinases, and a putative au-
toinhibitory site. The sequence and structure of skeletal 
and smooth muscle MLCKs are generally similar at these 
functional sites, but are quite divergent otherwise. MLCK 
is inactive in the absence of Ca-CaM, and its activity in-
creases with increasing [Ca2+]j such that the phosphoryla-
tion of LC
2 0
becomes half-maximal at about 0.3 /xM Ca2+. 
Phosphorylation of a serine in the CaM binding domain 
decreases the affinity of MLCK for CaM and causes a de-
crease in the sensitivity of MLCK activity to increases in 
[Ca
2
+]j. This site appears to be phosphorylated 
in vivo
by 
Ca-CaM-dependent protein kinase II (CaMKII), but not 
by protein kinase C (PKC) or cAMP-dependent kinases. 
However, some studies indicate that this site is phospho-
rylated by protein kinase A (PKA), which might play a 
role in relaxation induced by dilators that increase cAMP, 
but the evidence is conflicting. Also, it appears that the 
[Ca2+]j required to activate CaMKII is about three times 
higher than that required to activate MLCK. This implies 
that Ca-CaM would bind to MLCK and shield the tar-
get serine before CaMKII could act on it, raising doubts 
about the significance of CaMKII in control of smooth 
muscle. It is well established only that Ca-CaM activates 
MLCK.
LC
2 0
phosphorylation produces two pronounced ef-
fects: 
it promotes the ability of myosin 
monomers 
to assemble into filaments, and it markedly increases 
(100-fold) the ATPase activity of myosin compared to un- 
phosphorylated filamentous myosin. How unphosphory- 
lated myosin filaments can remain stable is not certain, 
but may be related to binding to myosin of indepen-
dently expressed MLCK fragments containing the myosin 
binding domain of MLCK. How LC2o phosphorylation 
increases the ATPase activity so dramatically is still 
unknown.
LC
2 0
phosphatase dephosphorylates LC
2 0
. This is a 
trimeric type I protein phosphatase (PP-I) comprising a 
catalytic subunit (M.W. 37,000), a myosin-binding sub-
unit (M.W. 130,000), and a 20,000-M.W. subunit that also 
binds myosin. This PP-I may be inhibited by high arachi- 
donic acid, and inhibited or stimulated by phosphorylation
by PKC or cAMP-dependent kinases, respectively. How-
ever, since the phosphatase is bound to myosin, and the 
ratio of phosphatase to myosin is only about 1:50, there is 
some doubt that LC
2 0
phosphorylation is regulated via the 
phosphatase. MLCs may spontaneously dephosphorylate 
at a significant rate.
In some smooth muscle (especially vascular smooth 
muscle), tension elicited by agonists or K+-induced depo-
larization is often tonically maintained after LC2o phos-
phorylation and sometimes [Ca2+]j has fallen to near basal 
levels. These tonic contractions are characterized by very 
low cross-bridge cycling rates and ATP turnover, and are 
referred to as the 
latch phenomenon.
It is currently be-
lieved that dephosphorylation of attached cross-bridges 
converts them to a form in which ADP is stably bound, 
called 
latch bridges,
thereby dramatically slowing the 
events of the cross-bridge cycle subsequent to the force 
production step. It also appears that in the complete ab-
sence of Ca2+-induced LC
2 0
phosphorylation, the latch 
state cannot be sustained.
Protein kinase C (PKC) may play a role in tonic tension. 
PKC refers to a family of related serine/threonine kinases, 
five of which are found in smooth muscle. PKC is activated 
by diacylglycérol, or DAG. DAG (and also IP
3
) is liberated 
from membranes by the action of phospholipase C (PLC) 
on phosphatidylinositol 4,5-bisphosphate, or by the ac-
tion of phospholipase D (PLD) on phosphatidylcholine. 
A number of receptor-mediated events are transduced by 
activation of these lipases. Some agents that elicit tonic 
contraction (e.g., angiotensin II) activate PLD, thus pro-
ducing DAG (but not IP
3
), which activates PKC with no 
effect on [Ca2+]j. There are at least three sites on LC2o 
that can be phosphorylated by PKC, but it is not known 
which one, if any, of these is involved in the induction of 
contraction and latch.
Caldesmon, or CaD, is a thin filament-linked regulator 
in smooth muscle. CaD is a single long peptide (M.W. 
~90,000). It binds to actin, tropomyosin, myosin, and 
Ca-CaM. The affinity of CaD binding to thin filaments 
is much higher in the presence of tropomyosin than in 
its absence, and 
in vivo,
CaD is probably bound only 
to tropomyosin-containing filaments. CaD inhibits actin- 
activated myosin ATPase by competing with myosin for 
a binding site on actin, and substantially reduces ATPase 
activity at a CaD/actin ratio of 1:10. CaD displaced by 
myosin remains attached to actin, however, via a second 
binding site for which myosin does not compete. It has 
been found that CaD can also bind to the S
2
region of the 
myosin head, which might play a role in stabilizing thick 
filaments that have dephosphorylated.
Reversal of CaD-induced inhibition can occur by sev-
eral means. High concentrations of Ca-CaM inhibit CaD
 
    
