336
10 Proteins of Muscle and the Cytoskeleton
tebrate muscles. The myosins of the nematode
Caenorhabditis elegans and the soil amoeba Acanthamoeba castellani have been particularly thoroughly investigated. In the nematodes, 22 genes
are now known which code for muscle proteins;
mutations in these genes lead to defective motility and abnormalities of the muscle ultrastructure. They include four myosin-HC genes, three
myosin-LC genes, four actin genes, the paramyosin gene unc-15, and the gene unc-22, which encodes the myosin-associated, 600-kDa protein twitchin. Of the four HC genes, A and B are expressed in the 95 muscle cells of the body wall, and C
and D only in the pharynx musculature. Myosin
HC-A is found only in the central region of the
thick filaments and apparently plays a role in
assembly. There are present two LCs of 16 and
18 kDa; two genes are found for the regulatory
18-kDa LC, and these code for identical amino
acid sequences [5, 122]. The amoeba possesses
one myosin II with the usual structure, and four
type-I myosins. Myosin IB is made up of an HC
of 125 kDa and one LC of 27 kDa, whereas myosin IC has one 130-kDa HC and two LCs of
14 kDa. Very recently, a gene has been sequenced
for a relatively large HC of 177 kDa and this
apparently belongs to a high molecular weight
form of myosin I (HMWMI) [111, 171].
Only by DNA-sequence analysis has it been
possible to determine the amino acid sequence of
giant polypeptides such as myosin HC. The first
complete HC sequence obtained was that of myosin B from the body wall of Caenorhabditis
elegans, which is encoded by gene unc-54. This is
described here as an example of the rather uniform HCs of myosin type II. It consists of a total
of 1966 amino acids; the N-terminal 850 amino
acids are folded into the head region, in which are
found certain amino acids responsible for ATP
binding, ATPase activity, and actin and LC binding. There follows a hinge or neck region, in
which the mobility of the bridges is located; and,
finally, the rod region.
The amino acid sequence of the rod region
shows considerable periodicity with period
lengths of 7, 28 and 196 amino acids. The 160
repeats of seven amino acids contain non-polar
amino acids at the first and fourth positions. This
leads to the formation of fields of non-polar sidechains, which allow hydrophobic interaction
between the two HCs, thereby stabilizing the
coiled-coil structure of the rod region. This pattern is interrupted only by four inserted amino
acids (skip residues) and ends at the HMMILMM
border. The LMM region of the HC carries many
charged amino acids at its outer surface, such that
their side-chains produce a charge pattern on the
surface of the coiled-coil with periodicities of 28
and 196 amino acids; the stable structure of the
thick filaments thus relies mainly upon electrostatic interaction. The HC hinge region (S-2) has
no periodicity of charged groups and can therefore move away from the surface of the thick filament when the myosin head makes contact with
an actin filament. The polypeptide chain ends
with a non-helical, C-terminal extension of 24
amino acids, similar to those seen on other myosin HCs from Caenorhabditis, Drosophila and
non-muscle cells but not from vertebrate skeletal
muscle [65].
The HCs of all type-II myosins have a structure
similar to that of Caenorhabditis HC [167, 181,
260]. However, they may be further subdivided
on the basis of particular sequence characters into
two families: the HCs of striated skeletal and
heart muscle, and the HCs of smooth muscle and
non-muscle cells. In the chicken, the HCs of
the smooth, intestinal muscles have 81-84 %
sequence similarity to non-muscle-cells HCs but
only 38-41 % similarity to skeletal-muscle HCs
[124]. Within the two families themselves, there
are also instances of HC isoforms with different,
specific, physiological functions and corresponding deviant structures. Thus, for example, in the
rat there are at least seven muscle-specific HCs
(embryonal skeletal muscle, neonatal, extraocular, fast IIA, fast lIB, a-ventricle/atrium, ~ventricle/slow) encoded by at least 13 genes.
Altogether, six different HCs are expressed in the
breast muscle of the chicken during development,
and seven HC isoforms are present in the various
muscles of the fish Salvelinus alpinus [65, 107,
177,260]. All the smooth muscles of the mammals contain the same two HCs; a third type is
found only in the lung arteries [208]. In vertebrates, the variety of HCs is based on a mUlti-gene
family, which in the case of the chicken includes
31 members. Additional variation in the proteincoding sequences arises by alternative splicing. In
Drosophila, all the HCs of striated muscles come
from one gene and those of non-muscle myosin
come from another. Five of the 18 exons in the
muscle-HC gene can each be spliced in five different ways; theoretically, therefore, there are 480
possible HC isoforms in Drosophila [128, 129]. In
spite of the generally similar geometry of the tail
regions of different myosins, there is evidence for
adaptive evolution of their amino acid sequences.
The proportion of a-helix in the tail region, which
at 0-5°C accounts for almost 100 %, decreases
10 Proteins of Muscle and the Cytoskeleton
tebrate muscles. The myosins of the nematode
Caenorhabditis elegans and the soil amoeba Acanthamoeba castellani have been particularly thoroughly investigated. In the nematodes, 22 genes
are now known which code for muscle proteins;
mutations in these genes lead to defective motility and abnormalities of the muscle ultrastructure. They include four myosin-HC genes, three
myosin-LC genes, four actin genes, the paramyosin gene unc-15, and the gene unc-22, which encodes the myosin-associated, 600-kDa protein twitchin. Of the four HC genes, A and B are expressed in the 95 muscle cells of the body wall, and C
and D only in the pharynx musculature. Myosin
HC-A is found only in the central region of the
thick filaments and apparently plays a role in
assembly. There are present two LCs of 16 and
18 kDa; two genes are found for the regulatory
18-kDa LC, and these code for identical amino
acid sequences [5, 122]. The amoeba possesses
one myosin II with the usual structure, and four
type-I myosins. Myosin IB is made up of an HC
of 125 kDa and one LC of 27 kDa, whereas myosin IC has one 130-kDa HC and two LCs of
14 kDa. Very recently, a gene has been sequenced
for a relatively large HC of 177 kDa and this
apparently belongs to a high molecular weight
form of myosin I (HMWMI) [111, 171].
Only by DNA-sequence analysis has it been
possible to determine the amino acid sequence of
giant polypeptides such as myosin HC. The first
complete HC sequence obtained was that of myosin B from the body wall of Caenorhabditis
elegans, which is encoded by gene unc-54. This is
described here as an example of the rather uniform HCs of myosin type II. It consists of a total
of 1966 amino acids; the N-terminal 850 amino
acids are folded into the head region, in which are
found certain amino acids responsible for ATP
binding, ATPase activity, and actin and LC binding. There follows a hinge or neck region, in
which the mobility of the bridges is located; and,
finally, the rod region.
The amino acid sequence of the rod region
shows considerable periodicity with period
lengths of 7, 28 and 196 amino acids. The 160
repeats of seven amino acids contain non-polar
amino acids at the first and fourth positions. This
leads to the formation of fields of non-polar sidechains, which allow hydrophobic interaction
between the two HCs, thereby stabilizing the
coiled-coil structure of the rod region. This pattern is interrupted only by four inserted amino
acids (skip residues) and ends at the HMMILMM
border. The LMM region of the HC carries many
charged amino acids at its outer surface, such that
their side-chains produce a charge pattern on the
surface of the coiled-coil with periodicities of 28
and 196 amino acids; the stable structure of the
thick filaments thus relies mainly upon electrostatic interaction. The HC hinge region (S-2) has
no periodicity of charged groups and can therefore move away from the surface of the thick filament when the myosin head makes contact with
an actin filament. The polypeptide chain ends
with a non-helical, C-terminal extension of 24
amino acids, similar to those seen on other myosin HCs from Caenorhabditis, Drosophila and
non-muscle cells but not from vertebrate skeletal
muscle [65].
The HCs of all type-II myosins have a structure
similar to that of Caenorhabditis HC [167, 181,
260]. However, they may be further subdivided
on the basis of particular sequence characters into
two families: the HCs of striated skeletal and
heart muscle, and the HCs of smooth muscle and
non-muscle cells. In the chicken, the HCs of
the smooth, intestinal muscles have 81-84 %
sequence similarity to non-muscle-cells HCs but
only 38-41 % similarity to skeletal-muscle HCs
[124]. Within the two families themselves, there
are also instances of HC isoforms with different,
specific, physiological functions and corresponding deviant structures. Thus, for example, in the
rat there are at least seven muscle-specific HCs
(embryonal skeletal muscle, neonatal, extraocular, fast IIA, fast lIB, a-ventricle/atrium, ~ventricle/slow) encoded by at least 13 genes.
Altogether, six different HCs are expressed in the
breast muscle of the chicken during development,
and seven HC isoforms are present in the various
muscles of the fish Salvelinus alpinus [65, 107,
177,260]. All the smooth muscles of the mammals contain the same two HCs; a third type is
found only in the lung arteries [208]. In vertebrates, the variety of HCs is based on a mUlti-gene
family, which in the case of the chicken includes
31 members. Additional variation in the proteincoding sequences arises by alternative splicing. In
Drosophila, all the HCs of striated muscles come
from one gene and those of non-muscle myosin
come from another. Five of the 18 exons in the
muscle-HC gene can each be spliced in five different ways; theoretically, therefore, there are 480
possible HC isoforms in Drosophila [128, 129]. In
spite of the generally similar geometry of the tail
regions of different myosins, there is evidence for
adaptive evolution of their amino acid sequences.
The proportion of a-helix in the tail region, which
at 0-5°C accounts for almost 100 %, decreases