(Received for publication, February 3, 1997, and in revised form, March 24, 1997)
From the Departments of Internal Medicine and
Biochemistry, The University of Iowa, Iowa City, Iowa 52242 and the
§ Department of Physiology, School of Medicine, University
of California at Los Angeles, Los Angeles, California 90024
The function of three of tropomyosin's
sequential quasiequivalent regions was studied by deletion from
skeletal muscle -tropomyosin of internal residues 49-167. This
deletion mutant tropomyosin spans four instead of the normal seven
actins, and most of the tropomyosin region believed to interact with
troponin is retained and uninterrupted in the mutant. The mutant
tropomyosin was compared with a full-length control molecule that was
modified to functionally resemble muscle tropomyosin (Monteiro, P. B.,
Lataro, R. C., Ferro, J. A., and Reinach, F. C. (1994) J. Biol. Chem. 269, 10461-10466). The tropomyosin deletion
suppressed the actin-myosin subfragment 1 MgATPase rate and the
in vitro sliding of thin filaments over a heavy
meromyosin-coated surface. This inhibition was not reversed by troponin
plus Ca2+. Comparable tropomyosin affinities for actin,
regardless of the deletion, suggest that the deleted region has little
interaction with actin in the absence of other proteins. Similarly, the
deletion did not weaken binding of the troponin-tropomyosin complex to actin. Furthermore, Ca2+ had a 2-fold effect on
troponin-tropomyosin's affinity for actin, regardless of the deletion.
Notably, the deletion greatly weakened tropomyosin binding to myosin
subfragment 1-decorated actin, with the full-length tropomyosin having
a 100-fold greater affinity. The inhibitory properties resulting from
the deletion are attributed to defective stabilization of the
myosin-induced active state of the thin filament.
Tropomyosin is an extended -helical coiled-coil that spans
seven actin monomers. Periodicity in tropomyosin's amino acid sequence
and also in its modeled three-dimensional structure have led to the
proposal that it contains 7 and possibly 14 quasiequivalent regions
(1-4). Each
length region is approximately 39
residues long and contains either one or, in the original proposal (1),
a pair of putative actin binding motifs. Deletion of approximately one-half (i.e. 21 residues) of a long period repeat or one
complete such region has little effect on troponin-tropomyosin's
ability to bind actin or regulate the myosin
S-11 thin filament ATPase rate. (4).
However, deletion of
of a putative actin binding motif
abolishes binding and therefore regulation (4). In the present study a
mutant tropomyosin designated r
Tm
(49-167) with three of the
seven regions (regions 2, 3, and 4) deleted has been created. 119 amino
acids are absent from this mutant, corresponding to approximately
3 × 39
residues per region as defined by McLachlan and
Stewart (1). r
Tm
(49-167) is long enough to span only four actin
monomers, yet the present report shows it to behave normally in many
respects.
Recent structural data suggest that tropomyosin can be positioned in three different locations on the actin filament. These positions are dependent upon the presence or the absence of troponin, calcium, and myosin. Troponin in the absence of Ca2+ induces an "off" state in which only weak transitory myosin S-1 binding occurs (5, 6). Tropomyosin in this position blocks actin sites necessary for strong myosin S-1 binding as modeled by Rayment et al. and others (6-10). When either Ca2+ is added or troponin is removed, there is a 25 ° rotation of the tropomyosin from the off state, partially uncovering actin residues involved in myosin binding (7-11). An additional 10 ° rotation of tropomyosin about the actin filament occurs when myosin S-1 is bound to the actin (7, 9, 11, 12).
The structural studies above suggest that tropomyosin-actin
interactions are highly diverse. Accordingly, several conditions were
examined to characterize the deletion tropomyosin of the present
report. The deletion was found not to alter tropomyosin-actin binding
under almost all conditions. However, the deletion had a major effect
on this process in the presence of myosin S-1. Furthermore,
rTm
(49-167) was found to inhibit in vitro motility and thin filament-myosin S-1 MgATPase activity. The functional significance of these interrelated observations is discussed.
One consideration
in designing these experiments was that bacterially expressed forms of
striated muscle tropomyosin are unacetylated and therefore will not
bind actin without troponin (13-15). Monteiro et al. (16)
have shown that a dipeptide (Ala-Ser) at the NH2 terminus
can substitute for acetylation and restore actin binding and
polymerization. Therefore, full-length and deletion tropomyosins were
both modified at the 5 end so that the recombinant forms expressed in
Escherichia coli would mimic acetylated bovine cardiac tropomyosin. A control molecule, designated r
Tm, was created using
rat
tropomyosin as a PCR template. Oligo 1 (5
-GCGCTCGAGCCATGGCTAGTATGGACGCCATCAAG-3
) added a tripeptide,
Met-Ala-Ser, to the 5
end as well as a XhoI and
NcoI restriction sites for cloning purposes. The methionine is removed during processing in the cell (16). PCR oligo 2 (5
-GCGTCTAGATCTTTATATGGAAGTC-ATATCC-3
) contained an error and would
have caused a mistake in Asn281, but the proofreading
mechanisms of Vent DNA polymerase (New England Biolab) corrected the
PCR product to match the template as was confirmed by sequencing.
The internal deletion was accomplished by two PCR reactions that were
cloned individually and later ligated together. Oligo 1 from above and
oligo 3 (5-ATGCGCTGCAGCGCGCCACCAGTGACACCAGCTC-3
) were used to create
the actin binding region 1-containing portion of r
Tm
(49-167).
Oligo 3 introduces a BSSH II site without altering the amino acid
sequence. The other PCR product, encoding putative actin binding
regions 5, 6, and 7, was constructed using oligo 2 and the
BSSHII-containing oligo 4 (5
-ATGCGGCGCGCAAGCTGGTCATCATC-3
). The
number of amino acids deleted was a multiple of seven, thus allowing
proper assembly of the coiled-coil (17). The PCR products were cloned
into pSP72 for confirmation by sequencing and then cloned into pET-3D
for expression in E. coli (DE3).
Bovine cardiac troponin and tropomyosin
(18), rabbit skeletal muscle F-actin (19), and rabbit skeletal myosin
S-1 (20) were purified as described previously. Recombinant rat
striated muscle tropomyosins were purified as described (15) with
revisions. A 40% (242 mg/ml)/70% (231.8 mg/ml) ammonium sulfate
fractionation was done. The protein was dialyzed, 6 M urea
was added, and the protein was loaded on a 80-ml DEAE-cellulose column
at 4 °C. The column had been equilibrated with 50 mM
Tris (pH 8.0), 0.01% NaN3, 0.5 mM EDTA, 0.5 mM DTT, and 6 M urea. This same buffer was used to wash the column after protein loading. The protein was then eluted
using a 0-300 mM NaCl gradient of 360 ml. The pooled
fractions were dialyzed against 10 mM Tris (pH 7.5), 1 mM DTT, 0.01% NaN3. Protein concentrations
were determined using extinction coefficients as determined by Gill and
von Hippel (21). The coefficient for r
Tm
(49-167) was 1.024 × 104 M
1 cm
1. The
full-length r
Tm and bovine cardiac tropomyosin were calculated to
have an extinction coefficient of 1.5360 × 104
M
1 cm
1.
The release of phosphate
from [-32P]ATP (DuPont NEN) in the presence of myosin
S-1 was measured (22) at six 2-min intervals. Conditions and protein
concentrations are as described in the legend to Fig. 5.
Tropomyosins used in binding assays were radiolabeled on Cys190 with [3H]iodoacetic acid (23). Radioactivity from before and after centrifugation at 35,000 rpm in a TLA100 rotor for 30 min at 25 °C were compared (24). Binding curves of tropomyosins or troponin-tropomyosin complexes to actin were fit to the McGhee-Von Hippel equation (23-25). This equation evaluates the binding of a long ligand to a linear lattice. The equation deals with the problem of random gaps in the lattice that are too short for a ligand to fit and explicitly considers cooperative interactions between adjacent ligands that counteract this parking problem. K0 is defined as the affinity of a ligand for an isolated site on the lattice. Any cooperativity between ligands results in a y-fold increase in affinity for a singly adjacent site. The overall binding affinity, Kapp, approximately equals the product yK0.
In Vitro Motility AssaysThe movement of rhodamine phalloidin-labeled thin filaments over rabbit fast skeletal muscle-heavy meromyosin-coated coverslips was observed by video epifluorescence microscopy and analyzed as described previously (26, 27). The experiments were performed at an ionic strength of 0.1 M in the presence of 0.5% methyl cellulose. The temperature of the assay was held constant at 25 °C. Protein concentrations and other conditions are described under "Results" or in Table I.
|
Bacterially expressed skeletal muscle tropomyosin lacks
the NH2-terminal acetylation necessary for
tropomyosin-actin binding in the absence of troponin (13, 28). However,
a fusion chicken -tropomyosin with an NH2-terminal
dipeptide binds to actin normally (16). In this study an analogous rat
fusion
-tropomyosin, r
Tm, was created and compared with bovine
cardiac muscle tropomyosin and to r
Tm
(49-167), which contains
the NH2-terminal dipeptide and has 119 internal residues
deleted. For purposes of comparison, concentrations of both full-length
tropomyosins (0.8 µM) and of r
Tm
(49-167) (1.6 µM) were chosen so that 50% of each would bind to 5 µM actin at low ionic strength (Fig. 1).
As reported previously for the analogous chicken fusion tropomyosin,
r
Tm (Fig. 1, squares) functioned very similarly to muscle
tropomyosin (Fig. 1, circles). More significantly,
r
Tm
(49-167) (Fig. 1, triangles) did not act
differently from r
Tm. All three tropomyosins had decreases in
affinity for actin as the salt concentration increased. Beyond 0.15 M both recombinant tropomyosins did not bind detectably to actin, whereas bovine cardiac tropomyosin appeared to maintain weak
actin binding (see also Refs. 23 and 24).
Recombinant Tropomyosin Binds Actin with Normal Affinity and Cooperativity
Fig. 1 suggests that the tropomyosin internal
deletion has relatively little effect on tropomyosin-actin binding. To
assess this more precisely, tropomyosin-actin binding was analyzed as a
function of the tropomyosin concentration, with representative data in
Fig. 2. These results demonstrate that the stoichiometry of binding was as expected. Both full-length tropomyosins span seven
actins, whereas rTm
(49-167) spans four. Fig. 2 shows that r
Tm
plateaued at approximately 1.5 µM of tropomyosin bound.
Due to the difference in length, 1.75 times as much r
Tm
(49-167) would be needed to saturate the same amount of actin. This was the case
with a value of approximately 2.6 µM tropomyosin bound. Under the conditions presented in the figure (60 mM KCl),
the r
Tm had a Kapp for actin of 1.6 × 106 M
1. Remarkably the
Kapp for r
Tm
(49-167), 0.88 × 105 M
1, was only slightly lower
than that of the full-length molecule. Each Kapp
value is the product of two terms, K0 and
y (see "Materials and Methods"). The 1.8-fold change in
Kapp is within the variation observed from
preparation to preparation (23, 24). For the paired preparations shown
in Fig. 2, the cooperativity parameters were almost identical
(r
Tm
(49-167), y = 35; control r
Tm,
y = 42). The deletion tropomyosin had a slightly lower
K0 (affinity of tropomyosin for an isolated site
on F-actin) compared with full-length tropomyosin (2.5 versus 3.7 × 104
M
1), a 1.5-fold difference. Bovine cardiac
tropomyosin (not shown) bound with an intermediate binding affinity to
the recombinant forms. These results parallel findings by Monteiro
et al. (16), who first determined that the fusion dipeptide
resulted in normal tropomyosin function.
If each quasiequivalent region of tropomyosin contributed equal binding
energy, then removing three of seven such regions would result in
the binding energy. This would mean that
K0 for r
Tm
(49-167) would be 4.1 × 102 M
1 ((r
Tm
K0)4/7 = (3.7 × 104 M
1)4/7 = 4.1 × 102 M
1). This would
be a 90-fold effect of the deletion on K0 and
the same 90-fold effect on Kapp (because
Kapp = yK0). This
calculation is meant as a general estimate for the expected effects of
the deletion, not an exact prediction. However, the observed effects (<2-fold) are much smaller than estimated, suggesting that putative actin binding regions 2, 3, and 4 do not interact with actin under these conditions.
In the
presence of EGTA, increasing concentrations of bovine cardiac troponin
plus either rTm
(49-167) or r
Tm caused inhibition of the
actin-myosin S1 MgATPase rate (Fig. 3,
triangles and circles). Upon the addition of
CaCl2, the activity remained inhibited in samples with
r
Tm
(49-167) (Fig. 3, diamonds). This indicates that
the troponin-deletion tropomyosin complex can inhibit normally but that
this is not reversed by the addition of Ca2+. By
comparison, saturating amounts of troponin-r
Tm activated the ATPase
rate approximately 16-fold when CaCl2 was present (Fig. 3,
squares), as expected with normal regulation. This verifies that dipeptide fusion tropomyosin molecules are capable of activation as well as inhibition.
Other results indicated that when tropomyosin-saturated actin was
studied in the absence of troponin, the shortened tropomyosin inhibited
the MgATPase rate to a greater degree than did full-length recombinant
tropomyosin. In MgATPase assays (not shown) with 1 µM
myosin S-1, 4 µM actin, 3.5 µM rTm, and
other conditions as in Fig. 3, the MgATPase rate was 0.16 s
1, whereas it was 0.06 s
1 with
r
Tm
(49-167). By comparison actin and myosin S-1 alone had a
value of 0.6 s
1. The inhibition of control r
Tm was
4-fold, similar to observations with cardiac tropomyosin (Ref. 18; see
also Refs. 29 and 30). Shortened r
Tm
(49-167), however, produced
a 10-fold inhibition of the MgATPase rate.
Rhodamine-phalloidin labeled F-actin (2 µM) was incubated overnight with various concentrations
of rTm
(49-167). The skeletal muscle heavy meromyosin-propelled
movement of these filaments (diluted to low actin concentration with
supplemental r
Tm
(49-167) in the motility buffer) was studied as
in Ref. 27. Regardless of the overnight conditions, increases in the
final r
Tm
(49-167) concentration caused a decrease in speed and
the number of moving filaments (Table I). In contrast,
control r
Tm had no effect on movement (not shown). The prevention of
all filament movement by sufficient concentrations of
r
Tm
(49-167) (last line of Table I) is consistent with the ATPase
results. The deletion tropomyosin blocks productive actin-myosin
interactions. Significantly, the reduction in speed at intermediate
concentrations of r
Tm
(49-167) indicates that it does not
completely block actin-myosin binding. Rather, it must permit weak
actin-myosin interactions that exert a drag on the actin filament. In
all cases, the addition of 100 nM r
Tm
(49-167)
resulted in no movement. This concentration is lower than expected from
Fig. 2, perhaps due to methylcellulose in the motility buffer.
Motility assays using reconstituted actin-tropomyosin-troponin thin
filaments were also done. As expected, both bovine cardiac tropomyosin
and rTm conferred Ca2+-sensitive regulation in the
presence of troponin. The results with r
Tm
(49-167) resembled the
MgATPase results. 100 nM concentrations of troponin plus
r
Tm
(49-167) resulted in no movement, regardless whether
Ca2+ was added (data not shown).
To understand the inhibitory effect of the deletion, thin
filament assembly was studied in the presence of troponin with or without CaCl2. Ionic conditions were chosen (150 mM KCl) so that binding to actin was weak in the absence of
troponin and in a stronger but still measurable range in the presence
of troponin. Troponin caused rTm
(49-167) to bind actin at least
as tightly as the control r
Tm (Fig. 4). (The deletion
mutant retains most of the tropomyosin region believed to interact with
troponin (31-34).) Full-length and
length tropomyosins in
the presence of troponin bound to actin with very similar
Kapp (2.2 × 106
M
1, 2.6 × 106
M
1, respectively). Interestingly, the
addition of Ca2+ caused a decrease in binding affinity of
both complexes to actin, a 2-fold change in each case. This change is
comparable with that seen with cardiac muscle tropomyosin (23).
Although the deletion abolishes regulation, it does not alter this
effect of Ca2+ on thin filament assembly.
Effect of Myosin S-1 on the Binding of Recombinant Tropomyosin and Mutant Tropomyosin to Actin
Myosin S-1 and heavy meromyosin
greatly strengthen the affinity of tropomyosin for actin (24, 35, 36).
Muscle tropomyosin binding to actin-myosin S-1 is so tight that it is
difficult to measure, even in the presence of high ionic strength (24).
Fig. 5A shows that a similar effect was
observed for rTm. This full-length recombinant tropomyosin did not
substantially bind to actin in the absence of myosin S-1 in the
presence of 0.3 M KCl, but the addition of myosin S-1
increased binding greatly to a Kapp of 9.1 × 106 M
1 (Fig. 5A,
squares versus circles). The effect of myosin S-1 on r
Tm-actin binding is at least 100-fold and probably greater. In
contrast, r
Tm
(49-167) did not bind to actin measurably under the
same conditions, regardless of the presence of myosin S-1 (Fig.
5A, diamonds versus triangles). Deletion of the
internal tropomyosin region had a major 100-fold or larger effect on
tropomyosin binding to actin-myosin S1.
To investigate any effect of myosin S-1 on rTm
(49-167) binding
to actin, positive or negative, these experiments were repeated at
lower ionic strength. At 60 mM KCl, myosin S-1 had a 6-fold strengthening effect on the Kapp of
r
Tm
(49-167) binding to actin (Fig. 5B). This effect
of myosin S-1 is attributable to an increase in tropomyosin's affinity
for an isolated site rather than to a change in cooperativity (the
semilog curves are parallel). Binding of r
Tm to actin was too tight
to measure under these conditions.
Tropomyosin has been shown by several investigators to have different affinities for actin depending upon the presence or the absence of myosin, troponin, and Ca2+ (reviewed in Ref. 37). The addition of troponin increases tropomyosin's affinity for actin (23), perhaps due to troponin-actin interactions (37), and causes it to shift position when Ca2+ is added (7-11). Myosin S-1 also increases tropomyosin's affinity for actin (24, 35, 36), and emerging results (9, 11, 12) indicate that it induces tropomyosin to shift to a third location on the actin, different from either the relaxed state or the Ca2+ state. These different quaternary structures for the thin filament may correspond to the blocked, closed, and opened states proposed in Ref. 38. Our results suggest that the deleted region, tropomyosin regions 2, 3, and 4, is required for stabilizing the myosin-induced state but is not required for other conformational states. The idea that the ends rather than the middle of tropomyosin are most important for actin binding in the absence of myosin is supported by several lines of evidence. First, in electron microscopy data obtained under conditions (0.5 mM MgCl2) poorly favoring tropomyosin-actin binding, Mabuchi (39) observes tropomyosin molecules tethered by one end to the actin filament. Second, the use of carboxypeptidase-A-digested tropomyosin or unacetylated tropomyosin severely decreases tropomyosin's affinity for actin (13, 40), and this effect is not due to diminished cooperativity (15, 41). Finally, Hitchcock-DeGregori and An (17) showed that deletion of tropomyosin regions 2 and 3 had only a 2-fold effect on troponin-tropomyosin binding to actin. This prior study differed from the present results, because two rather than three of the quasiequivalent regions were deleted and because the fusion dipeptide at the NH2 terminus was absent. This latter difference compensates for the absence of acetylation in bacterially expressed tropomyosin and permits experiments in the absence of troponin. Such experiments now demonstrate that the deleted region does not contribute significantly to actin binding when troponin is absent.
The deletion of tropomyosin regions 2, 3, and 4 also had no effect on
troponin-mediated changes in thin filament assembly. For both
full-length and deletion tropomyosins, troponin promoted binding to
actin (compare Fig. 4 with the 150 mM KCl data in Fig. 1).
In these assays the Kapp values for full-length
tropomyosin-troponin and length tropomyosin-troponin are
very close in value. This parallels published findings that deletion of
regions 2 and 3 had little effect (17). Fig. 4 also shows both
recombinant tropomyosin-troponin complexes have a decreased affinity
for actin upon the addition of Ca2+. The decrease in
affinity is comparable between full-length and shortened tropomyosins,
suggesting that the deletion of these three regions does not interfere
with Ca2+-induced changes in the thin filament
conformation. This would be very significant because it suggests that
the shortened tropomyosin is capable of undergoing calcium-induced
changes yet inhibits MgATPase activity (Fig. 3) and in vitro
motility (Table I). Other data will be needed to establish this point
definitively, however.
Full-length rTm binds actin at least 100-fold more tightly in the
presence than in the absence of myosin S-1 and permits actin-myosin S-1
MgATPase activity and in vitro filament movement. Shortened
r
Tm
(49-167) exhibits none of these properties. Because myosin
S-1 slightly promotes r
Tm
(49-167)-actin binding (Fig. 5B), it can be concluded that myosin S-1 and
r
Tm
(49-167) can bind simultaneously to F-actin. The reduced thin
filament sliding speed and fraction of smoothly moving filaments at
intermediate concentrations of r
Tm
(49-167) (Table I) suggests
the presence of a drag force, perhaps from myosin S-1 attachment to a
portion of the thin filament where tropomyosin is bound. Nevertheless, the tropomyosin internal deletion results in destabilization of the
actin-myosin S-1-tropomyosin structure relative to what is found with
normal tropomyosin. Furthermore, the thin filament assembly data in the
absence of myosin indicate that this destabilization is specific for
the myosin-induced conformation of the thin filament.
Strongly bound myosin cross-bridges create an activated state of tropomyosin-containing thin filaments, with increased actomyosin ATPase rates (42-45), troponin affinity for Ca2+ (46-51), actin-myosin affinity (46, 52, 53), and actin-tropomyosin affinity (24, 35, 36). Solution studies of contractile proteins (38, 45) and muscle fiber investigations (54-58) have been explainable in terms of a cooperative transition to one active state for the thin filament, which presumably corresponds to this potentiated state (43). Recent structural data suggest that regulated thin filament have at least three conformations but that only one of these conformations has tropomyosin in a position to allow cross-bridge cycling (11, 59). These results and recent solution studies of actin-myosin S1 interactions in the absence (38) and the presence (60) of ATP point toward the conclusion that there is essentially one active state for the thin filament. One way to test this concept would be to specifically destabilize the myosin-induced or potentiated state of the thin filament and then evaluate the effect on regulation under low myosin, high MgATP conditions. Unexpectedly, tropomyosin with regions 2, 3, and 4 deleted has properties that provide such a test of the mechanism of regulation. The tropomyosin internal deletion destabilizes the actin-myosin-tropomyosin complex and does not change the Ca2+-sensitive assembly of the actin-myosin-tropomyosin-troponin complex, yet the deletion is profoundly inhibitory to actin-myosin cycling. Unless the tropomyosin binds very strongly to actin-myosin S-1, it is inhibitory.
Finally, a relevant prior report is the 1990 study by
Hitchcock-DeGregori and Varnell, which showed that an internal deletion of tropomyosin did not prevent Ca2+-sensitive regulation of
the myosin S-1 MgATPase rate (4). This shortened tropomyosin had only
region 2 deleted and was the normal length, so there is no
conflict between this earlier data and the present report. Rather, the
combined results suggest the possibility that a specific tropomyosin
sequence within region 3 and/or 4 is necessary for thin filament
activation. It may be functionally important, for example, that
r
Tm
(49-167) is missing half of the tropomyosin region (residues
150-180) believed to interact with the globular region of troponin
(31-33, 61-64). However, loss of a specific interaction with troponin
would not explain the inhibitory properties of r
Tm
(49-167) in
the absence of troponin. An alternative possibility is that there are a
critical number of internal tropomyosin regions needed to stabilize the
myosin S-1 induced "on" state or otherwise to restore regulation.
In other words, the crucial difference may be that a
length
tropomyosin permits thin filament activation and a
length
tropomyosin does not, with the specific regions relatively unimportant.
Further experiments may be able to distinguish among these different
explanations.