2368 Biophysical Journal Volume 95 September 2008 2368–2390

Approximate Model of Cooperative Activation and Crossbridge Cycling

in Cardiac Muscle Using Ordinary Differential Equations

John Jeremy Rice,* Fei Wang,y Donald M. Bers,y and Pieter P. de Tombez
*IBM T.J. Watson Research Center, Yorktown Heights, New York; yDepartment of Physiology, Loyola University-Chicago, Maywood, Illinois;
and zCenter for Cardiovascular Research, Department of Physiology and Biophysics, University of Illinois-Chicago, Chicago, Illinois

ABSTRACT We develop a point model of the cardiac myofilament (MF) to simulate a wide variety of experimental muscle
characterizations including Force-Ca relations and twitches under isometric, isosarcometric, isotonic, and auxotonic conditions.
Complex MF behaviors are difficult to model because spatial interactions cannot be directly implemented as ordinary differential
equations. We therefore allow phenomenological approximations with careful consideration to the relationships with the
underlying biophysical mechanisms. We describe new formulations that avoid mean-field approximations found in most existing
MF models. To increase the scope and applicability of the model, we include length- and temperature-dependent effects that
play important roles in MF responses. We have also included a representation of passive restoring forces to simulate isolated cell
shortening protocols. Possessing both computational efficiency and the ability to simulate a wide variety of muscle responses,
the MF representation is well suited for coupling to existing cardiac cell models of electrophysiology and Ca-handling
mechanisms. To illustrate this suitability, the MF model is coupled to the Chicago rabbit cardiomyocyte model. The combined
model generates realistic appearing action potentials, intracellular Ca transients, and cell shortening signals. The combined
model also demonstrates that the feedback effects of force on Ca binding to troponin can modify the cytosolic Ca transient.

INTRODUCTION
This article describes an approximate model of activation and system of equations. Moreover, if computational speed is
force generation in cardiac myofilament that recapitulates desired, then the system must be fairly simple and imple-
many experimental characterizations. Specifically, the exper- mented with ordinary differential equations (ODEs) instead
imental characterizations that weighed most heavily in model of partial differential equations or Monte Carlo approaches
development are described below: typically required for explicit consideration of the spatial
aspects. Much of the following work involves making ap-
1. Steady-state force-sarcomere length relations (F-SL rela-
proximations to maintain a system of ODEs, so emphasis is
tions).
placed on the simplifying assumptions and their inherent
2. Steady-state force-calcium relations (F-Ca relations) in-
limitations. Much of the model derives squarely from work
cluding SL effects.
performed over the last half-century; however, new approx-
3. Steady-state sarcomere length-calcium relations (SL-Ca
imations are developed in the Ca-activation and mean
relations) for unloaded cells.
crossbridge strains that differ from previous work. These
4. Steady-state force-velocity relations (F-V relations).
approximations help bridge the spatial scales where local
5. Isometric twitches including Ca activation and SL effects.
interactions are critical to emergent behavior but cannot
6. Ktr including Ca activation and temperature effects.
be directly implemented in mass-action or mean-field ap-
7. Cell shortening twitches as function of activator Ca.
proaches.
8. Effects of SL control on the intracellular Ca transients.
We develop this model in the middle ground where phe-
The last quarter century has seen the development of nomenological approximations are allowed with careful
models to understand many aspects of myofilament re- consideration to the relationships of the underlying mecha-
sponses. As described in a previous review (1), there are still nisms that cannot be explicitly modeled. We have also at-
difficulties in developing predictive models given that the tempted to strike a reasonable balance between mechanistic
underlying muscle biophysics has yet to be fully resolved. detail and model parsimony while including sufficient cel-
Another difficulty lies in trying to compress the spatial as- lular machinery to recapitulate a wide range of experimental
pects of myofilaments at the molecular level into a tractable protocols. For example, length- and temperature-dependent
effects are included, and the passive restoring force is rep-
Submitted August 9, 2007, and accepted for publication December 13, 2007. resented so that experimental protocols in isolated cell
Address reprint requests to J. Jeremy Rice, Tel.: 914-945-3728; E-mail: shortening can be simulated. Ultimately, we hope that this
[email protected]. model will provide the community with an approximate and
Fei Wang’s current address is TAP Pharmaceutical Products Inc., 675 N. predictive representation that retains enough mechanistic
Field Drive, Lake Forest, IL 60045. underpinnings to provide the flexibility and extensibility that
Donald M. Bers’ current address is Department of Pharmacology, UC existing models do not.
Davis Health System, 451 Health Sciences Drive, Davis, CA 95616.
Editor: David A. Eisner.
 2008 by the Biophysical Society
0006-3495/08/09/2368/23 $2.00 doi: 10.1529/biophysj.107.119487
ODE-Based Model of Cardiac Myofilament 2369

TABLE 1 Parameters for model TABLE 2 Default initial conditions

Parameter Value Units Variable Value Units
Sarcomere geometry SL 1.9 mm
SLmax 2.4 mm NNoXB 0.99 Probability
SLmin 1.4 mm PNoXB 0.01 Probability
lengththick 1.65 mm NXB 0.97 Probability
lengthhbare 0.1 mm PXB 0.01 Probability
lengththin 1.2 mm XBPreR 0.01 Probability
XBPostR 0.01 Probability
Temperature dependence
xXBPreR 0 mm
TmpC Range ¼ 15–37 C
xXBPostR x0 mm
Qkon 1.5 Unitless
IntegralForce 0 (Unit normalized force) s
Qkoff 1.3 Unitless
Qkn_p 1.6 Unitless
Qkp_n 1.6 Unitless
(actin and regulatory proteins). To implement length dependence, we define
Qfapp 6.25 Unitless
the single-overlap fraction of the thick filament (referred to as SOVthick) that
Qgapp 2.5 Unitless
reports the fraction of thick filament that is apposed to single-overlap thin
Qhf 6.25 Unitless
filament. The assumption is that the only effective strongly-bound XBs occur
Qhb 6.25 Unitless
in the single overlap region. Hence, the thick-filament, single-overlap frac-
Qgxb 6.25 Unitless
tion is used in calculations for maximally activated force. This assumption
Ca binding to troponin to thin filament regulation comes directly from classic sliding filament theory (2).
kon 50 mM1 s1 The single-overlap function for the thick filament is shown in Fig. 1 B (see
koffL 250 s1 Eqs. 42–46 for mathematical formulation; please refer to Tables 1–3 for the
koffH 25 s1 parameters and default conditions used in this work). The maximal possible
perm50 0.5 Unitless force corresponds to sarcomere lengths (SLs) in the range 2.3–2.4 mm for which
nperm 15 Unitless the whole thick filament is in the single-overlap region so that SOVthick ¼ 1.
kn_p 50 s1 Between 1.65 and 2.3 mm, the SOVthick decreases at a constant rate as the
Kp_n 500 s1 thin filaments cross over in the center region of the sarcomere. In the range
1.4–1.65 mm, the SOVthick decreases at an even faster rate as the thick fila-
Thin filament regulation and crossbridge cycling
ment is assumed to cross the z-line, and crossbridges are assumed not to form
fapp 500 s1
in the region past the z-line. This aspect to the model is speculative as the
gapp 70 s1
actual interactions between the thick filament and z-line are not currently
gslmod 6 Unitless
understood. However, some experimental characterizations in trabeculae
hf 2000 s1
contract down to sarcomere lengths of ;1.5 mm (3), which supports this
hfmdc 5 Unitless
assumption. The maximal Ca-activated force linearly decreases with sarco-
hb 400 s1
mere length from 2.15 to ;1.7 mm where a faster rate of decrease is seen,
gxb 70 s1
similar to the model prediction. Moreover, experimental protocols in isolated
sp 8 Unitless
cells show sarcomere lengths in the range 1.4–1.5 mm under maximal
sn 1 Unitless
shortening (4). Afterwards the cells recover normal function after relaxation,
Mean strain of strongly-bound states suggesting a nondestructive interaction between the thick filaments and the
x0 0.007 mm z-disk for sarcomere lengths below the thick-filament length.
f 2 Unitless A second overlap fraction is defined for interactions along the length of
the thin filament (referred to as SOVthin). The single-overlap function for the
Normalized active and passive force
thin filament is shown in Fig. 1 B. Note that the single-overlap function for
SLrest 1.9 mm
the thin filament varies between 0.17 at 1.4 mm and 0.64 at 2.4 mm. Hence,
PContitin 0.002 (Unit normalized force)
roughly one-third of the thin filament does not participate in actin-myosin
PExptitin 10 Unitless
interactions, even at sarcomere lengths that produce maximal force (2.3–2.4
SLcollagen 2.25 mm
mm) where 100% of the thick filament can participate (SOVthick ¼ 1). The
PConcollagen 0.02 (Unit normalized force)
difference in the single-overlap function for the thin filament and the thick
PExpcollagen 70 Unitless
filament is attributed solely to the geometry of the sarcomere (see Eqs. 45 and
Calculation of complete muscle response 46). The single-overlap fraction for the thin filament is used to calculate the
Mass 0.00005 (rat) (Unit normalized force) s2 mm1 Ca binding to troponin that depends on crossbridge interaction. Specifically,
0.00025 (rabbit) higher affinity binding can occur in the vicinity of crossbridges, and as such,
Viscosity 0.003 (Unit normalized force) s mm1 the thin filament single-overlap function is used to calculate the Ca binding
constant
Fafterload Range ¼ 0.0–1.0 (Unit normalized force) and activation of the thin filament.
KSE Range ¼ 1.0–200.0 (Unit normalized force) mm1
TABLE 3 Default initial conditions—parameters for default
calcium transient (rat, 22.5C)
METHODS
Parameter Value Units
Description of sarcomere geometry t1 0.02 s
t2 0.11 s
The lengths assumed for the thick and thin filaments are shown in Fig. 1 A.
Caamplitude 1.45 mM
The fraction of crossbridges (XBs) that can strongly bind and generate force
Cadiastolic 0.09 mM
depends on the overlap of the thick filament (myosin) and the thin filament

Biophysical Journal 95(5) 2368–2390

2370 Rice et al.

FIGURE 1 Modeling sarcomere length effects. (A) The assumed sarcomere geometry is defined using the filament lengths as shown. Specific examples are
chosen to show the maximal length (2.4 mm), start of the plateau region (2.3 mm), rest length (1.9 mm), the point where thick filaments contact the z-line (1.65 mm),
and the minimal length (1.4 mm). (B) The thick-filament overlap fraction gives the fraction of myosin heads in the single-overlap regions that can form effective
force-generating actin-myosin interactions. Hence this value gives the maximum normalized force given full activation. The thin-filament overlap fraction is
defined in a similar manner but does not reach unity as the whole thin filament never exists in the single overlap zone. (C) Passive force attributed to titin and
other cytoskeletal elements is shown as a function of sarcomere length. The passive force for cells is assumed to reflect across the abscissa at the rest length. For
trabeculae, the passive force has an additional component attributed to collagen so that force increases steeply above 2.2 mm and effectively limits sarcomere
length to 2.3 mm. (D) In addition to active crossbridge forces and the passive forces just described, the model contains additional components including a
viscosity element and a mass element. The series elastic element is optional and is used to simulate experimental protocols with fixed muscle lengths in which
the internal sarcomeres shorten as compliant end connections are stretched.

While active force of muscles is attributed to the action of cycling cross- and increases total muscle force. Below the rest length, the passive force is
bridges, the complete muscle response involves contributions of other entities negative and hence acts as a restoring force to decrease total force. As shown in
including passive force and other visco-elastic elements as shown in Fig. 1 D. Fig. 1 C, the passive force for cells is assumed to be reflected around the resting
We assume a rest length of 1.9 mm that corresponds to the point of no passive length. The justification for this is that titin is thought to contribute to passive
force as shown in Fig. 1 C. Above the rest length, the passive force is positive force, and passive force will be roughly symmetric around the rest-length of

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2371

titin (as assumed elsewhere, see Fig. 6 in (5)). For this reason, this component is -
ððTmpC 37Þ=10Þ
konT ¼ kon 3 Qkon ; (4)
named for titin, although other sources such as cytoskeletal components could
also contribute. For trabeculae, the passive force has an additional component
so that force increases steeply above 2.2 mm and effectively limits the maximal where kon is 50 mM1 s1 and Qkon is 1.5.
length of cells to 2.3 mm. This feature is assumed to correspond to the effects of The corresponding total rate for unbinding rate for the high- and low-
collagen with convolutions that can initially unfurl easily, but once taut, be- affinity cases are defined as
come very stiff. With both components, the passive force curve matches the
-
ððTmpC 37Þ=10Þ
curvature and steepness of experimental characterizations (3,6), although there koffHT ¼ koffH 3 koffmodspecies 3 Qkoff ; (5)
is variability in the zero crossing that corresponds to the rest-length (e.g., 1.9–
-
ððTmpC 37Þ=10
2.0 mm in (6) versus in 2.0–2.1 mm in (3)). koffLT ¼ koffL 3 koffmodspecies 3 Qk off ; (6)
Other visco-elastic elements are also included. The muscle is assumed to
have a Newtonian viscosity element set to the mean value found experi- where koffH is 25 s1; koffL is 250 s1; kxmodspecies is 1.0 for rat and 0.9 for
mentally (0.3% Fmax mm1 s1 from (6)). A small mass term is also included. rabbit; and Qkoff is 1.2. The off-rate koffH is smaller than koffL by a factor of 10
The effect of the mass is to prevent instantaneous changes in muscle short- to account for the higher affinity of troponin associated with strongly-bound
ening velocity for quick release protocols, a feature that improves the sta- crossbridges. The 10-fold increase is similar to experimental estimates of
bility of the integration of the model equations. Tuning this parameter can ;8.6-fold (7) and $10 fold (8). Note that Qkon . Qkoff so that Ca sensitiv-
also improve response times. Specifically, large values can generate under- ity decreases with lower temperature as suggested by experimental results
damped responses that overshoot and ring. On the other extreme, small mass (9–11).
values can produce overdamped responses. We choose a midrange value
between these extremes. Finally, a linear series elastic element can be in-
cluded to simulate the effects of compliant end connections that occur in real
muscle preparations. Hence the muscle can shorten internally at the active Ca-based activation
force element even through the total muscle length is fixed. No fixed value is
We assume that steep Ca sensitivity in activation results from nearest-
assumed for the elastic element, but instead parametric studies are used to
neighbor interactions of troponin and tropomyosin along the thin filament.
illustrate the effect on muscle responses.
Indeed, explicit modeling of this process underscores the plausibility of this
assumption (1,12). For the modeling here, we seek to avoid explicit spatial
representation of nearest-neighbor interactions as these cannot be repre-
Regulatory Ca-binding to troponin
sented as ODEs. Instead, we assume that thin-filament activation is a steeply
The presence of strongly-bound crossbridges is assumed to increase the nonlinear function of [Ca] as a phenomenological representation of the ef-
binding affinity of the nearby regulatory units (RUs). This is embodied by fects of nearest-neighbor interactions. Similar nonlinear functions have been
assuming Ca binding to two populations of troponin regulatory sites that employed in previous modeling efforts to capture the assumed effects end-to-
correspond to the higher affinity with strongly-bound crossbridges and to end interactions of RUs (13–16).
lower affinity sites without strongly-bound crossbridges. Here, ‘‘high’’ and To implement Ca-based activation, we assume that troponin and tropo-
‘‘low’’ refer to the single regulatory binding site and should not be confused myosin act as RUs that exist in one of two states, N or P (Fig. 2). State N
with the two high-affinity, nonregulatory sites on cardiac Troponin C. The represents a nonpermissive state that prevents the formation of strongly-
high and low affinity sites are calculated as the fractional population with Ca bound crossbridges. State P represents a permissive conformation of the
bound (CaTropH and CaTropL, respectively), regulatory proteins that can permit transitions to strongly-bound crossbridge
states. First, we can consider the case in which no crossbridges can form
d (outside of the single-overlap region between the thick and thin filaments)
CaTropH ¼ konT ½Cað1  CaTropH Þ  koffHT CaTropH ; (1) and use the notation NNoXB and PNoXB to indicate the absence of nearby
dt crossbridges. In this region, the following equations hold:
d
CaTropL ¼ konT ½Cað1  CaTropL Þ  koffLT CaTropL ; (2)
dt d
NNoXB ¼ kn pT 3 NNoXB 1 kp nT 3 PNoXB ; (7)
where konT is the complete rate constant for binding, [Ca] is the concentration dt
of Ca, koffHT is the complete rate constant for unbinding from high-affinity d
sites, and koffLT is the complete rate constant for unbinding the low affinity PNoXB ¼ kn pT 3 NNoXB  kp nT 3 PNoXB : (8)
dt
sites.
While the rates in the model represent a diverse set of state transitions, a The transition rates kn_pT and kp_nT are set so that the fraction of permissive
standard definition format is maintained. The format is explained using the RUs is a nonlinear function of the fraction of RUs with Ca bound and not
following example for generic total rate constant kxT, directly to intracellular [Ca] itself. Mathematically, the nonlinearity is
-
ððTmpCa 37Þ=10Þ incorporated using
kxT ¼ kx 3 kxmod 3 kxmodspecies 3 Qkx ; (3)
where kx is the base rate constant under default conditions; kxmod is a modifier TropRegulatory ðxÞ ¼ ð1  SOVFthin ðxÞÞ3 TropCaL
based on other parameters or states (e.g., crossbridge strain); kxmodspecies is
modifier based on species (e.g., rat or rabbit); and Qkx is the Q10 value for 10 1 SOVFthin ðxÞ3 TropCaH ; (9)
changes in the temperature as specified by TmpC. All transition rates can be
represented in the above form, although not all rates have explicit kxmod and where TropRegulatory(x) is the fraction of thin filament RUs that have Ca
kxmodspecies terms. The net effect of the Q10 terms is to decrease the rates bound; x is the sarcomere length; and SOVFthin(x) is the single-overlap
below the default values as defined at 37C. The T in the subscript differentiates function for the thin filament. We assume nearest-neighbor cooperativity so
the total transition rate kxT from the base-rate value under default conditions that the shift of an RU to a permissive state is represented by a nonlinear
denoted by kx. function called permtot defined as
For these specific examples, the total Ca binding is assumed to be dif-
fusion-limited and is the same for high- and low-affinity cases. We assume a
permtot ¼ ð1=ð1 1 ðperm50 =TropRegulatory ðxÞÞ perm ÞÞ ; (10)
n 0:5
relatively low temperature dependence so that

Biophysical Journal 95(5) 2368–2390

2372 Rice et al.

FIGURE 2 Model construction. States NXB and

PXB represent nonpermissive and permissive con-
formations of the regulatory proteins, respectively.
The next transition is to the XBPreR state, short for
prerotated, that is strongly bound with the head
extended. The transition to the post-rotated force-
generating XBPostR state, short for post-rotated,
represents the isomerization to induce strain in
the extensible neck region. For the activation
process, the fraction of troponin with bound Ca
(TCa) is used to set the transition rate between NXB
and PXB using a strong nonlinearity function to
represent cooperativity. The model assumes that
troponin for regulation has affinity set by the thin-
filament overlap (and hence ultimately sarcomere
length) which tracks the fraction of regulatory
proteins with nearby crossbridges that can attach
(see Fig. 1). Higher affinity is assumed to represent
the cooperative effects of attached crossbridges on
Ca binding. Calculation of apparent Ca binding is
similar but uses thin-filament overlap fraction and
also assumes that affinity increases only after
crossbridges strongly bind to populate the XBPreR
and XBPostR states. The regulatory and apparent Ca
binding terms are calculated separately to avoid a global feedback from strongly-bound crossbridges to Ca binding. Such feedback can produce a
nonphysiological Ca sensitivity (see text for details).

where the half-activation constant perm50 ¼ 0.5 and the Hill coefficient tion of the regulatory proteins, and the nearest myosin is assumed to be in a
nperm ¼ 15. detached or weakly-bound state. In this model, the detached and weakly
Then permtot modifies the forward rate for nonpermissive to permissive bound crossbridge states are lumped together. These states are analogous to
transitions as states NNoXB and PNoXB described above for the case of no nearby myosin.
The XBPreR state is strongly bound, but the myosin head has not isomerized to
-
ððTmpC 37Þ=10Þ
kn pT ¼ kn p 3 permtot 3 Qkn p ; (11) rotate and induce strain in the neck region. Hence, this state contributes to
stiffness but does not generate force in the absence of net motion. The
1
where kn_p ¼ 50 s and Qkn_p ¼ 1.6. Working in the opposite direction, the XBPostR state is a strongly-bound, post-isomerization state in which the
permissive to nonpermissive transition rate is modified by the inverse of crossbridge head has rotated to put distortion equal to x0 in the extensible
permtot in the formulations link. Returning to the weakly bound state is unidirectional and is assumed to
  consume one ATP. In contrast, the other transitions are bidirectional and do
1
inversepermtot ¼ min ; 100 ; (12) not involve ATP hydrolysis. The complete set of equations is
permtot
d
ððTmpC-37Þ=10Þ NXB ¼ kn pT 3 NXB 1 kp nT 3 PXB ; (14)
kp nT ¼ kp n 3 inversepermtot 3 Qkp n ; (13) dt
where kp_n ¼ 500 s1 and Qkp_n ¼ 1.6. Note that a maximum value is placed d
PXB ¼ kn pT 3 NXB  ðkp nT 1 fappT Þ 3 PXB
on inversepermtot to insure that kp_nT is not greater than kp_n 3 100 ¼ 50,000 dt (15)
s1. This limit is set to prevent the numerical integrator from requiring very 1 gappT 3 XBPreR 1 gxbT 3 XBPostR ;
small time steps that result when transition rates are very large. Note that the
limit has very minor effects on model behavior as kn_pT  kp_nT, and all RUs d
XBPreR ¼ fappT 3 PXB  ðgappT 1 hfT Þ 3 XBPreR
are effectively nonpermissive when the limit is reached. dt
1 hbT 3 XBPostR ; (16)
d
Crossbridge cycling—computing XBPostR ¼ hfT 3 XBPreR  ðhbT 1 gxbT Þ 3 XBPostR : (17)
state occupancy dt
As in the work of Razumova et al. (19), the force is proportional to the
For the case of RU activation with subsequent crossbridge formation, the fractional occupancy of the strongly-bound states multiplied by the respec-
situation is somewhat more complicated. Ca-induced changes in the regu- tive mean distortion of these states. The mean distortion states XBPreR and
latory proteins are generally assumed to permit actin-myosin interactions. XBPostR are tracked by variables xXBPreR and xXBPostR, respectively. While
However, strongly-bound crossbridges are also found to produce thin-fila- the full mathematical formulation is presented below, a brief description
ment activation, even in the absence of activator Ca (17,18). To best capture suffices for now. Crossbridges become strongly bound with a transition from
such interactions, activation and crossbridge cycling are combined in a PXB to XBPreR with an assumed distortion of 0. The rotation of the myosin
coupled system (Fig. 2) that is adapted from the work of Razumova et al. from XBPreR to XBPostR is assumed to induce an increase in distortion equal to
(19). This set of states represents an ensemble of myosin heads and the as- x0. Hence, in the absence of net motion between the thick and thin filaments,
sociated actin and regulatory proteins. xXBPreR is 0 and xXBPostR is x0.
State NXB represents a nonpermissive state that prevents the formation of While the basic framework derives from Razumova et al. (19), the tran-
strongly-bound crossbridges. State PXB represents a permissive conforma- sition rates have been modified in both general and specific ways. In general,

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2373

the base rates are larger in the current formulation that corresponds to 37C. consuming detachment transition rate, at least for the protocols simulated in
The current model also includes a term xbmodspecies (1.0 for rat or 0.2 for that study. Hence, only the ATP-consuming detachment transition rate is
rabbit) that scales all crossbridge cycling rates to account for species-based assumed to have strain-dependent terms in the Razumova et al. study. For
differences. For specific changes, the crossbridge attachment rate to the first this study, we carry over the strain-dependence from that study in the rate
strongly-bound state XBPreR is now given by modifier gxbmd, defined as

-
ððTmpC 37Þ=10Þ

fappT ¼ fapp 3 xbmodspecies 3 Qfapp ; expðsp ððx0  xXBPostR Þ=x0 Þ Þ if xXBPostR , x0
2
(18)
gxbmd ¼ ;
expðsn ððx0  xXBPostR Þ=x0 Þ Þ if xXBPostR $ x0
2

where fapp ¼ 500 s1 and Qfapp is defined at the end of this section. Note that
(24)
except for the species and temperature dependence, the rate is fixed. The
original formulation in Razumova et al. (19) has a cooperative attachment where constants sp ¼ 8 and sn ¼ 1 set the effects of strain for positive and
term that is not retained. negative shortening velocities, respectively. The effect of sp . sn is to
The reverse rate is similar except for a modifier gappslmod that increases increase the ATPase rate more for shortening than for lengthening protocols.
the detachment rate at shorter sarcomere lengths. The exact definition is Note that the values chosen differ from those in the original study which
would have corresponded to sp ¼ 1 and sn ¼ 8. The total rate is
-
ððTmpC 37Þ=10Þ
gappT ¼ gapp 3 gapp slmod 3 xbmodspecies 3 Qgapp ; (19) -
ððTmpC 37Þ=10Þ
gxbT ¼ gxb 3 gxbmd 3 xbmodspecies 3 Qgxb ; (25)
gapp slmod ¼ 1 1 ð1  SOVFthick ðxÞÞ 3 gslmod; (20)
1
where gxb ¼ 70 s and Qgxb is defined below.
where gapp ¼ 70 s1; Qgapp is defined at the end of this section; x is the The temperature dependence of the crossbridge cycling transition rates
sarcomere length; and the constant gslmod ¼ 6 is used to scale the effects of are uniformly (except for one case) set to a default Q10 value of 6.25.
the thick-filament, single-overlap fraction on the strongly- to weakly-bound Specifically, Qfapp ¼ Qhf ¼ Qhb ¼ Qgxb ¼ 6:25. By setting the Q10 values
transition rate. to be equal, the relative population of states should be roughly constant as
The construction of gappslmod that increases the detachment rate at shorter temperature changes. While this is an obvious simplification, the values
sarcomere lengths is speculative and ad hoc but has some justification. One or produce reasonable temperature-induced changes in maximal shortening
two strongly-bound crossbridges anywhere along the thin filament may velocity, twitch duration, and Ktr. While 6.25 appears large, values as large as
suffice to hold the whole thin filament permissive even in the absence of 6.7 have been reported for reactions in the crossbridge cycle (23). Note that
activator Ca. We represent this effect by decreasing detachment rates for there is one exception in our model in that Qgapp ¼ 2.5. There are two jus-
conditions for which more crossbridges can be recruited (i.e., as SOVFthick tifications. One is that Qfapp and Qgapp best correspond to k4 and k4 in (23)
(x) increases at longer sarcomere length). In terms of model responses, the with Q10 values equal to 6.7 and 2.5, respectively. In addition, the differ-
construction produces isometric twitches for which the final relaxation has ential in Q10 values in the model produces a maximal Ca-activated force that
faster time rates of force decline as sarcomere length decreases, as seen ex- increases with temperature, as seen in experimental studies (9–11).
perimentally (20,21). Note however, that sarcomere length has been shown
not to affect the tension cost (ATPase rate/force) in experimental studies (22),
so a similar SL-dependence is not applied to gxbT ; the ATP-consuming de- Crossbridge cycling—computing force and
tachment transition rate.
The forward transition rate hfT between the strongly-bound states XBPreR
mean strain
to XBPostR is defined as As in the work of Razumova et al. (19), the force is proportional to the
-
ððTmpC 37Þ=10Þ fraction of occupancy of the strongly-bound states (XBPreR and XBPostR)
hfT ¼ hf 3 hf mod 3 xbmodspecies 3 Qh ; (21)
f multiplied by the average distortion of these states (xXBPreR. and xXBPostR).
 2 ! Mathematically, one can write
xXBPreR
hf mod ¼ exp signðxXBPreR Þ 3 hfmdc 3 ;
x0 Factive } ½ xXBPreR XBPreR 1 xXBPostR XBPostR : (26)
(22) The fractional occupancies of the strongly-bound states are computed as
1 described previously. Note that Eq. 26 constitutes a mean-field approxima-
where hf ¼ 2000 s ; Qhf is defined at the end of this section; and the constant
tion, while spatially explicit approaches calculate force as the expected value
hfmdc ¼ 5 sets the extent to which mean strain of the prerotated state affects the
of developed force for all strongly-bound crossbridges. Specifically, in a
isomerization rate. The net effect is to increase the forward rate as xXBPreR
spatially explicit model, we could write for the population of crossbridges
becomes more negative as occurs during muscle shortening. Conversely, a
lengthening muscle will produce a positive xXBPreR to decrease the isomeriza- Z
tion rate. The backward transition rate hbT from XBPostR to XBPreR. is defined by Factive } ÆFXB æ ¼ ðkXB xÞ 3 PDFðxXB ¼ xÞ dx; (27)
-
ððTmpC 37Þ=10Þ
hbT ¼ hb 3 xbmodspecies 3 Qhb ; (23) where the first term in the integral is the force of an attached crossbridge as a
1
where hb ¼ 400 s and Qhb is defined at the end of this section. linear spring constant kXB multiplied by the distortion x. The second term in
In the original work of Razumova et al. (19), the isomerization transition the integral is the probability density function of an attached strongly-bound
rates (corresponding to hfT and hBT) had no strain dependence. We found that crossbridge with distortion x. This representation is derived from the classic
strain dependence on forward transition rate hfT was needed to produce modeling work of Huxley (24) and is used in more current models with
shortening velocities comparable to experimental measures. In principle, explicit spatial representations that require partial differential equations (e.g.,
similar effects could be produced by a strain-dependent decrease in reverse (25,26)).
rate hbT. However, in this model, strain dependence on backward transition However, for our spatially compressed model, we assume that
rate hbT produces instabilities. Hence, no strain dependence is included in
hbT. As discussed later, the full system of equations can show instability and ÆFXB æ  kXB + ÆXi æÆxXi æ; (28)
i
oscillations under some parameter choices.
In the original work of Razumova et al. (19), the isomerization transition where ÆXiæ is the occupancy of state Xi, ÆxXiæ is the mean distortion of state Xi,
rates play a much smaller role in shaping responses as compared to the ATP- and the summation is over all strongly-bound states. A similar mean-field

Biophysical Journal 95(5) 2368–2390

2374 Rice et al.

approximation is made in previous modeling to decrease the computation xXBPostR that tracks xXBPreR with added strain x0. Turning to Eq. 29, the
complexity and produces reasonable results under many conditions (for a backward transition for isomerization is represented by the term hbT 3
more in-depth discussion, see (19)). Understanding the mean-field approx- (xXBPostR  x0  xXBPreR). In computing xXBPreR, another factor is the effect
imation is key for understanding the model construction that follows. of the transition from a weakly-bound to a strongly-bound state (from state
The mean strain of crossbridge states are computed by assuming full PXB to state XBPreR). These new strongly-bound crossbridges are assumed to
activation of the thin filament. Hence, all RUs are assumed to be permissive, attach with 0 mean distortion, so that a high rate of attachment should de-
and Ca-based activation events plays no role in the strain calculation. Note crease xXBPreR. This effect is incorporated by the fappT 3 (xXBPreR) term,
that assuming full thin-filament activation leads to a different formulation for which will force xXBPreR toward 0 with a rate proportional to fappT.
mean distortion of states than that of earlier work from which the model is So far we have used only attachment rates to compute mean strains. In-
based. In the earlier study (19), mean distortion is assumed to depend on both tuitively, any change in mean distortion as a result of crossbridge cycling
the fractional occupancy of states as well as the transitions between states. To should also depend on detachment rates. In the current formulation, we
carry this approach over to the model here, then mean distortion would in- consider detachment rate indirectly by calculating XBDutyFract
PreR and XBDutyFract
PostR ;
clude the state-occupancy terms (PXB, XBPreR, and XBPostR). However, the which are the fractions (or alternatively, duty cycles) of units in states XBPostR
occupancy of these states is strongly influenced by Ca-based activation, and and XBPreR assuming full thin-filament activation. These values are calcu-
as a result, the kinetics of computing crossbridge strain become strongly Ca- lated by assuming that kn_pT  kp_nT so that only states PXB, XBPreR, and
dependent. We avoided this construction to ensure that the mean distortion of XBPostR are populated. Using the King-Altman rule (27), the steady-state
the states would depend only on the relative sliding of the filaments and the population of states can be determined from the transition rates as

fappT hbT 1 fappT gxbT

XBDutyFract ¼ ; (31)
PreR
gxbT hfT 1 fappT hfT 1 gappT hbT 1 gappT gxbT 1 fappT hbT 1 fappT gxbT
fappT hfT
¼ :
DutyFract
XBPostR (32)
gxbT hfT 1 fappT hfT 1 gappT hbT 1 gappT gxbT 1 fappT hbT 1 fappT gxbT

intrinsic cycling rates of crossbridges. The rationale for the construction In Eqs. 29 and 30, the inverses of XBDutyFract
PreR and XBDutyFract
PostR are used as
depends on the assumption that strong nearest-neighbor coupling between scaling factors for second terms on the right-hand sides to represent the
RUs will produce large stretches of thin filament that are permissive. These dependence on the length of time a crossbridge remains in a given state. If
effects are assumed to be local and not affected by bulk fraction of cycling crossbridges are cycling quickly through the crossbridge cycle, then one can
crossbridges that are represented by the state-occupancy terms. The argu- assume that the rates into the strongly-bound states will be high while the
ments are somewhat involved, and are deferred to the Discussion. total occupancy can be low as result of fast turnover. The inclusion of the
With the assumption of full thin-filament activation, the mean distortion inverses of XBDutyFract
PreR and XBDutyFract
PostR in the calculation of mean strain cap-
xXBPreR and xXBPostR are calculated as tures the effect of turnover rate, on how quickly crossbridge can refresh
strain.
d 1 dSL f  With the above definitions in place, Eqs. 29 and 30 can be interpreted as
xXBPreR ¼ 1 fappT 3 ðxXBPreR Þ
dt 2 dt XBDutyFract
PreR
phenomenological formulations to compute mean distortion as the interplay
of the net motion of thick and thin filament and the effect of crossbridge
1 hbT 3 ðxXBPostR  x0  xXBPreR Þ; (29) cycling. Consider two simple cases. If crossbridges are slowly cycling, upon
d 1 dSL f assuming small values for fappT, hfT, and hbT, the dSL=dt term will dominate.
xXBPostR ¼ 1 DutyFract Then the mean distortion is determined primarily by the net motion of thick
dt 2 dt XBPostR and thin filament. In contrast, when there is no motion between the thick and
3½hfT 3ðxXBPreR 1 x0  xXBPostR Þ; (30) thin filaments (dSL=dt ¼ 0), crossbridge cycling dominates. One can easily
see that xXBPreR will tend to 0 as the weak-to-strong transition will generate
where dSL=dt is the velocity of sarcomere length (note that SL in this instance new crossbridges with net distortions of 0. In contrast, xXBPostR will tend to
is a model variable, although ‘‘SL’’ is the general abbreviation for sarcomere xXBPreR 1 x0 and hence x0. Such results are consistent with current theories
length); f is an empirically derived scaling term; and XBDutyFractPreR and for crossbridge dynamics. In our phenomenological approach, f is an em-
XBDutyFract
PostR are the fraction of units in states XBPreR and XBPostR assuming pirically derived scaling term that weighs the relative contribution of the
full thin-filament activation. dSL=dt term with the contribution of the crossbridge turnover terms. With
The motivation for the mean distortion follows from considering the in- f ¼ 2, the model generates reasonable, albeit phenomenological, values for
terplay of two effects: net motion between the thin and thick filaments and the mean distortions over a wide range of velocities and crossbridge cycling
gain or loss of distortion as crossbridges change states. The first effect is rates.
embodied in the first terms on the right hand sides of Eqs. 29 and 30. Namely,
the dSL=dt terms generate a proportional change in mean crossbridge dis-
tortions that track the net sliding of the thick and thin filaments. The ½- Calculation of normalized active force
scaling term accounts for the effects of sarcomere geometry in which the
thick filaments are symmetric, and the full sarcomere shortening velocity is One complication in developing myofilament models is the method to report
double the net rate of change between half-thick filaments and the associated output force. Similar to previous work in this area (15), we report a nor-
thin filaments. malized force with a maximum value of 1 with no assumptions on the exact
The gain or loss of distortion as crossbridges change state during cycling choice of transition rates. With such an approach, competing models can be
is embodied in the second quantities on the right-hand sides of Eqs. 29 and developed and compared without having to constantly renormalize results.
30. Consider first Eq. 30, which is the simpler of the two. Here xXBPostR The approach can be implemented by choosing scaling factors such that state
assumes a value similar to xXBPreR 1 x0 when the forward transition rate is occupancies are normalized to the maximum values possible under optimal
large. Hence, a high forward rate of isomerization will tend to produce conditions. In the model generated here, this situation occurs for high Ca

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2375

activation, isosarcometric, physiological temperature, and maximal single Running the complete muscle model
overlap of thick and thin filaments. These conditions can be simulated by
assuming kn_pT  kp_nT so that the system is fully activated. Isosarcometric If the simulation is assumed to be isosarcometric, then dSL=dt ¼ 0 and SL is
conditions (dSL=dt ¼ 0) and physiological temperature (37C) produce the fixed at its initial value SL0. If the sarcomere is assumed to contract or ex-
largest values for the transition rates and the maximal steady-state occu- pand, then the following ODE is solved to compute SL,
pancies for force-generating states. Assuming SL ¼ 2.3 mm generates that
SOVthick ¼ 1 and SOVthin ¼ 0.64. d IntegralForce 1 ðSL0  SLÞ 3 viscosity
SL ¼ ; (38)
The two scaling factors for state occupancy computed under optimal dt mass
PrerR and XBPosrR ; which are the fraction of strongly-bound
conditions are XBMax Max
where viscosity and mass are defined as shown in Fig. 1 D. IntegralForce is
crossbridges under the optimal conditions above. In this case, Eqs. 31 and 32
defined so that normalized forces are summed and integrated over time in the
simplify to
formulation
fapp hb 1 fapp gxb Z t
XBPrerR ¼ ;
Max
gxb hf 1 fapp hf 1 gapp hb 1 gapp gxb 1 fapp hb 1 fapp gxb IntegralForce ¼ ðFactive ðxÞ 1 Fpassive ðxÞ  Fpreload
0
(33)
 Fafterload ðxÞÞ dt; (39)
f h
XBPosrR ¼ :
Max app f
where Factive (x) is defined in Eq. 35; and Fpassive (x) is shown in Fig. 1 C (and
gxb hf 1 fapp hf 1 gapp hb 1 gapp gxb 1 fapp hb 1 fapp gxb
defined in the Appendix). The term Fpreload is a constant force that
(34) corresponds to an applied force that would induce an initial sarcomere
length that is larger than the resting length. Hence, this term balances the
Note that Eqs. 33 and 34 are very similar to Eqs. 31 and 32 with the important
passive force so that Fpreload ¼ Fpassive(SL0). The afterload term is used in one
change that the default rate values are used in the latter versus the total rate
of two ways. For an isotonic contraction, the afterload term is fixed after the
values in the former (e.g., fapp versus fappT).
release. For a fixed muscle length (isometric) contraction, the afterload is
The full definition of normalized active force is
computed as a series elastic element (see Fig. 1 D) used to simulate compliant
Factive ðxÞ ¼ SOVFthick ðxÞ ends of the muscle. The exact formulation is

xXBPreR 3 XBPreR 1 xXBPostR 3 XBPostR Fafterload ðxÞ ¼ KSE 3 ðx  SL0 Þ; (40)

3 ; (35)
x0 3 XBPostR
Max
where x is the sarcomere length, and KSE is the stiffness in units of
where x is the sarcomere length. The SOVFthick (x) term is a scaling factor for normalized force per mm.
the contribution of sarcomere geometry to the number of recruitable While the model is essentially defined by the equations alone, a few notes
crossbridges. Note that no XBMax on the implementation are in order. The model source code, parameters to
PreR term exists in the denominator on the
right-hand side of Eq. 35. Under isosarcometric conditions, xXBPreR will be recreate figures, and sample output files are provided in Supplementary
0 so there is no contribution by the XBPreR state under the optimal conditions Material, Data S1. An implementation is also available in CellML, an XML
defined above. markup language to store and exchange computer-based mathematical
models (see http://www.cellml.org/models/rice_wang_bers_detombe_2008_
version01). The model comprises a stiff set of nonlinear ODEs that can be
Apparent Ca-binding to troponin problematic for some numerical integrators. The model is implemented in C
For the regulatory Ca binding as described in Eq. 9, the ratio of low- and code using the CVODE integrator (28). In addition, the model has been
high-affinity troponin units is set by thick- and thin-filament overlap as de- implemented in XPP for which multiple integrators can be selected (http://
termined by sarcomere length. Hence, the regulatory Ca binding assumes a www.math.pitt.edu/;bard/xpp/xpp.html). In XPP, the CVODE integration
higher affinity if thin filament is in the single-overlap region and does not method runs for the widest range of protocols, while other methods often
depend on whether the crossbridges are strongly bound. In contrast, the failed for some protocols or parameter choices. When multiple integration
apparent Ca binding that is assumed to be sensed by a cell is calculated by methods execute successfully, the results are consistent. However, there are
assuming that the affinity of troponin increases only if nearby crossbridges cases where the model can produce ringing and low amplitude oscillations
are in strongly-bound states. In other words, the force-dependent Ca binding that are a property of the equations and not of the integrator choice. This
to troponin that affects the intracellular [Ca] transient is computed differently observation should not be too troubling given that the nonlinear equations are
from the assumed regulatory binding of Ca to troponin that switches on and highly interconnected with feedback terms. Moreover, the mean-field ap-
off the attachment of crossbridges (see Eqs. 1, 2, and 9). The apparent Ca proaches for Ca activation and crossbridge cycling are obvious departures
binding is formulated below. from reality, so inherent stability should not be automatically assumed. In-
The fraction of strongly-bound crossbridges is deed, even real muscle can show oscillations that are more prevalent in
conditions producing submaximal crossbridge cycling (29,30). In the mod-
XBPreR 1 XBPostR eling work here, instabilities are exacerbated by manipulations that lower the
FractSBXB ¼ Max : (36) crossbridge cycling rates (e.g., by lower temperatures) as well as certain
XBPreR 1 XBPostR
Max
parameter choices that produce large strain dependences on transition rates.

Then the apparent Ca binding is calculated by assuming that troponin in the

single-overlap region exhibits high affinity in proportion to FractSBXB as Modification to model rabbit myofilaments
TropApparent ðxÞ ¼ ð1  SOVFthin ðxÞÞ 3 TropL 1 SOVFthin ðxÞ The model is adjusted by decreasing transition rates in the crossbridge cycle
by a factor of 5 to simulate the changes in myosin isoforms (rat is predom-
3 ðFractSBXB 3 TropH inately V1 while rabbit is V3). In the absence of direct experimental data, the
1 ð1  FractSBXB Þ 3 TropL Þ: (37) factor of 5 is set empirically in the model to generate twitch response in rabbit
that look similar to experimental measures. The assumed species difference
The motivation of separately calculating regulatory and apparent binding is is reasonable compared to crossbridge cycling rate differences between rat
described in detail in the Discussion. and guinea pig that has been estimated to be a factor-of-6 faster in rat (31).

Biophysical Journal 95(5) 2368–2390

2376 Rice et al.

The only other changes are a slight increase in Ca sensitivity and a factor-of-5 35). While the mechanism of increased Ca sensitivity in real
increase in the mass term in rabbit which help to improve the rate of relax- muscle is under debate, the behavior of the model can be
ation. These modifications are relatively simplistic, and we expect that more
specific changes in other aspects of the model will be needed to better re-
mechanistically explained. The increased Ca sensitivity re-
capitulate all the species differences. However, a minimal set of changes is sults from a different ratio of high- and low-affinity sites as a
made as there is much less experimental data to characterize the myofilament function of thin-filament overlap fraction (see Fig. 1 B, and
responses in rabbit as compared to rat. see the Appendix for the exact formulation). Activation is
derived from the weighted sum of binding to high- and low-
Coupling to cardiac electrophysiology and affinity binding sites as determined by the thin-filament
Ca-handling mechanisms overlap fraction (see Eq. 9).
The data in Fig. 3 A are isosarcometric and hence can be
One goal of this work is to develop a model of the myofilaments that is directly compared to experimental data with feedback sarco-
suitable for coupling to existing models of cardiac models of cardiac elec-
mere length control via laser diffraction techniques. However,
trophysiology and Ca-handling mechanisms that exist in the literature. To
illustrate this purpose, we coupled our rabbit-modified myofilament model to much of the data in the literature is not SL-controlled and can
the Chicago rabbit ventricular myocyte model (32). These models are cou- have considerable internal shortening as a result of compliant
pled by using the cytosolic Ca concentration ([Ca]c) from the Chicago model end connections. Fig. 3 B shows F-pCa relationships that
as the input to the myofilament model. A feedback pathway exists in that the simulate increasing amounts of internal shortening. Each trace
buffering of the low-affinity, regulatory Ca-binding site on troponin is as-
corresponds to an increasingly compliant end connection,
sumed to be controlled by the apparent Ca binding of the myofilament model
as shown in Eq. 37. One complication exists in that Eq. 37 provides the total specified by smaller KSE, which permits greater degrees of
Ca bound to the regulatory site on troponin, whereas the Chicago model internal shortening (see Eq. 40). As compliance increases,
requires calculation of fluxes on to and off of buffers. To match this con- internal shortening causes decreases in maximal plateau level
struction, we differentiate Eq. 37 with respect to time (see Eqs. 58–64). Note and Ca sensitivity. Note that the apparent cooperativity,
that Ca binding to troponin (TropH and TropL), the thin-filament overlap
quantified by the Hill coefficient, also decreases as compliance
(SOVFthin (x)), and the fraction of strongly-bound crossbridges (FractSBXB)
can also change with time, so the chain rule is applied to terms with these increases. Here the dashed traces show true Hill functions with
variables (see Appendix for details). Hill coefficient ¼ 7.6 for KSE ¼ 50 and a Hill coefficient ¼ 4.0
for KSE ¼ 1. A similar change with an increasing end com-
pliance is found in a earlier modeling study (see Fig. 2 in (34)).
RESULTS Such an observation is consistent with experimental charac-
terizations that show greater apparent cooperativity with sar-
F-Ca and SL-Ca responses
comere length control compared to fixed total muscle length
Fig. 3 A shows steady-state F-pCa relationships with the re- conditions (3,33).
sponse of the model over a range of sarcomere lengths as A third characterization related to steady-state F-pCa is the
labeled. Longer sarcomere length increases both Ca sensi- SL-pCa relation in unloaded isolated cells. In this characteri-
tivity (leftward shift) and maximal plateau force. The steep- zation, the cardiac cell sarcomere length indicates the point at
ness as quantified by the Hill coefficient changes little with which restoring force just balances the actively generated force
sarcomere length. The SL-dependence of F-Ca relations in for the given level of activator [Ca]. Fig. 3 C shows the steady-
Fig. 3 A can be compared with experimental data under state SL-pCa response of the model that can compared to
sarcomere length control (33) with the exception that Ca50 experimental data from isolated cells (35). The response is
values are larger in skinned preparations than what is ex- similar in both the maximal degree of shortening and in the
pected for intact fibers (3). Similar trends are observed in both range of activator [Ca] over which the cell shortens from rest
model and experiment: sarcomere length increases maximal length to maximal shortening. Note that the apparent cooper-
plateau force while the Ca sensitivity shows little change in ativity is less in the SL-pCa relations as compared to F-pCa
the Hill coefficient (the dashed traces show true Hill func- relations under similar conditions (e.g., compare Fig. 3 C with
tions with Hill coefficient ¼ 7.6 for comparison). Note, 3, A and B). In the model, the decrease in apparent coopera-
however, that the shorter sarcomere lengths cannot be ex- tivity results from the transition to shorter sarcomere lengths
perimentally measured, and hence these responses are un- that decrease Ca sensitivity and maximal force so that greater
testable model predictions. The model results at these shorter activator [Ca] is required to continue cell shortening. As
lengths continue the trends at the longer sarcomere length as a shown in Fig. 3 A, a shorter sarcomere length requires greater
result of the mechanisms described below. activator [Ca] for any given level of active force.
The observed SL-based changes come from changes in the
thick- and thin-filament overlap fractions as sarcomere length
Force-velocity relations
changes. The maximal plateau force occurs when the thin
filaments are fully activated and hence reflect the fractional Fig. 4 A shows the model responses for isotonic contractions
recruitment of strongly-bound crossbridges as a function of against different fixed loads. The muscle is activated to a
sarcomere length. This fraction is set by the thick-filament maximal level ([Ca] ¼ 0.01 M), and length is held fixed until
overlap fraction (specifically by the SOVFthick (x) term in Eq. a release at 0.65 s. Directly after release, there is an initial fast

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2377

FIGURE 3 Steady-state responses as a function of Ca.

The plots show isometric force as a function of steady-state
activator [Ca]. (A) Active force is shown for the sarcomere
lengths as labeled that simulate isosarcometric conditions.
The relations are similar to Hill functions as determined by
the Ca-based activation assumed in the model. For com-
parison, two true Hill functions with Hill coefficient ¼ 7.6
are shown by the dashed traces. (B) Total muscle force
(active plus passive) are shown for fixed muscle length for
which internal shortening can occur. The degree of short-
ening is controlled by the stiffness of the series elastic
element as labeled for the different traces (units of KSE are
normalized force per mm extension). The degree of short-
ening from the initial length of 2.2 mm is also shown for
each trace. Two true Hill functions are shown by the dashed
traces for comparison. For KSE ¼ 50, the Hill coefficient ¼
7.6, and pCa50 ¼ 6.1. For KSE ¼ 1, the Hill coefficient ¼
4.0, and pCa50 ¼ 6.0. (C) The steady-state sarcomere
length is shown as a function of activator Ca. Here the
muscle is shortening from the rest length of 1.9 mm against
the passive restoring force. A Hill-like function can be fit to
the sarcomere length as shown by the dashed line. Here the
Hill coefficient ¼ 3.0, and pCa50 ¼ 5.9.

transient (an example is shown by the small arrow in Fig.  is normalized force, and a
where VHill is the Hill fit velocity, F
4 A), after which the muscle contracts at a roughly constant and b are empirically derived constants. The corresponding
velocity for some period. The velocity is determined by av- Hill fit parameters are a ¼ 0.19, 0.19, and 0.16 normalized
eraging over the roughly constant region. When velocity is force units and b ¼ 0.89, 1.6, and 3.7 mm/s for 20, 25, and
plotted as a function of afterload in Fig. 4 B, the typical hy- 30C, respectively. The parameters are chosen to minimize
perbolic shape emerges that can be fit by a modified Hill mean-square error for the data summed over the data points
equation (36), shown at each temperature.
The results in Fig. 4 B compare well with real muscle in

ða 3 Vmax Þ  ðb 3 FÞ several respects. Unloaded shortening values are comparable
VHill ¼  1a ; (41)
F to those measured experimentally. For example, in the model

Biophysical Journal 95(5) 2368–2390

2378 Rice et al.

although high variability can be seen depending on experi-

mental conditions and fitting procedures (see (13)).

Twitch responses
Another common experiment characterization uses dynami-
cally changing activator [Ca] to produce twitches. The sim-
plest situation to simulate is an isosarcometric contraction in
which the sarcomere length remains fixed throughout the
twitch. This situation is simulated in Fig. 5 for which either
sarcomere length (Fig. 5 A) or activator Ca (Fig. 5 B) is
varied. In Fig. 5 A, the sarcomere length is varied from 1.8 to
2.3 mm to show SL-dependent effects. The Ca transient is
the same for each run (see Fig. 5, inset; and see the exact
mathematical formulation in the Appendix) and corresponds
to parameters fit to data at 22.5C in Janssen et al. (38).
Longer sarcomere length increases both peak force and twitch
duration. The increase in peak force reflects the increased Ca
sensitivity and maximal developed force as shown in Fig. 3 A.
The time-to-peak force is relatively constant while increased
sarcomere length leads to longer twitches. These features
correspond well to experimental characterizations for rat at
similar temperature (20,38,39).
Fig. 5 B shows isosarcometric twitches for which the ac-
tivator Ca level is varied while holding sarcomere length
fixed at 2.2 mm. The Ca transient is a scaled version of that
shown in Fig. 5 A. The decreasing levels of activator Ca
produce decreases in twitch force that are similar to those
seen for decreases in sarcomere length in Fig. 5 A. Specifi-
cally, the peak force decreases, while the time to peak is
FIGURE 4 Force-velocity relationships. (A) The figure shows a simula- relatively constant. The decrease in force is accompanied by a
tion of an experiment with a quick release to a fixed afterload. The length is decrease in twitch duration, although the relative changes are
fixed and then released against a fixed afterload at 0.65 s. The traces smaller than those for decreases in sarcomere length. This
correspond to different afterloads from 0 to 0.8, in 0.1 increments in difference can be seen in the inset of Fig. 5 B where the force
normalized force. One additional trace at 0.85 represents isosarcometric
traces are self-normalized to have maximum values equal to
conditions corresponding to maximal force. Note that the shortening
velocity is relatively constant after a fast transient response directly after 1. The lowest peak Ca trace (solid circle) has shorter duration
the release (shown by the arrow). Data are shown for rat at 25C. (B) Force- than the largest Ca trace (asterisk). In comparison, the
velocity relations are generated from the protocol shown in panel A. The shortest sarcomere length trace (cross) has the shortest du-
force is the afterload value and velocity is computed from the relatively ration. The additional effect at the shorter sarcomere length is
constant value obtained after a transient directly after the release. When
that crossbridge detachment rates are increased as sarcomere
plotted in this fashion, the datapoints can be well fit by hyperbolic Hill
relations as shown by the labeled traces. See text for details of the fitting length decreases (see Eqs. 19 and 20).
procedure. A second type of twitch can be simulated in which the cell
contracts against its internal restoring force (similar to the case
for Fig. 3 C). Here the sarcomere length is initially at the rest
Vmax values are 3.82, 8.03, and 22.7 mm/s for 20, 25, and length of 1.9 mm, shortens to smaller length, and then returns
30C, respectively. Similar mean values of 6.13, 12.7, and to the rest length as shown in Fig. 6 A. The different traces
23.4 mm/s are found experimentally for the same range of show increasing levels of activator Ca with the same wave-
temperatures, although some variability is seen across dif- form as in the inset of Fig. 5 A, and the range of peak activator
ferent preparations (n ¼ 26–97) (36). The degree of curvature Ca are the same as in Fig. 5 B. The corresponding force traces
of the hyperbolic Hill fit can be quantified by k ¼ b/Vmax. The are shown in Fig. 6 B. Here the total force is plotted as the sum
k-values are 0.23, 0.20, and 0.16 for 20, 25, and 30C, of active and passive forces. The total instantaneous force is
respectively. Experimental values are generally in the range roughly proportional to the shortening velocity, so shortening
of 0.15–0.25 for a wide variety of muscle preparations (37), stops near the point where total force is 0. However, the effect
and similar values are generally reported for cardiac muscle of the mass also contributes (see Fig. 1 D), so total instanta-

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2379

FIGURE 5 Isometric twitch force as a functions of sarcomere length and

Ca activation. These plots show the active isosarcometric force. (A) sarco-
mere length is varied from 1.8 (y) to 2.3 ( ) mm in increments of 0.1 mm. In
each case, the activating Ca transient is the same as shown in the inset. (B)
The sarcomere length is held constant at 2.3 mm while the peak activating Ca FIGURE 6 Cell shortening twitches as a function of Ca activation. (A)
is scaled down. The traces show the responses for peak values of 1.45 ( ), The cell is allowed to shorten from rest length against the passive restoring
1.25, 1.15, 1.05, 0.95, and 0.85 (d) mM. The inset shows the force transients force. The sarcomere length is shown for the same Ca transients as in Fig. 5
renormalized to have peak values of 1 in each case. The times from 50% B with peak values of 1.45 ( ), 1.25, 1.15, 1.05, 0.95, and 0.85 (d) mM.
activation to 50% relaxation are 0.140 s (y), 0.187 s (d), and 0.223 ( ). Note that increased Ca activation decreases the time to peak shortening
These traces show that decreasing sarcomere length or Ca activation while the relengthening phase shows less dependence on Ca activation. The
decreases the twitch duration. The SL-dependent effect is larger because inset shows self-normalized sarcomere length (1 ¼ rest length, 0 ¼
the crossbridge detachment rate gappT is increased at shorter sarcomere minimum length) for peak values of 1.45 ( ) and 0.85 (d) mM. The times
lengths (see text for details). Data correspond to rat at 22.5C. from 50% shortening to 50% relaxation are 0.247 s ( ) and 0.229 (d). (B)
The total muscle force (active plus passive) is plotted for the corresponding
traces in panel A. Data correspond to rat at 22.5C.

neous force is not exactly proportional to the shortening

velocity. The effect of the mass will be greatest when con- shortening case shows a decreased time-to-peak shortening
traction is fast. For example, the ringing near the bottom of the with increasing Ca. In the shortening case, the decreasing
trace (asterisk) in Fig. 6 B illustrates that force may differ from time-to-peak is accompanied by a faster relaxation for higher
0 when dSL/dt ¼ 0 at the minimum of the asterisked trace in Ca activation levels. Hence the total duration of the cell
Fig. 6 A. The net effect of the mass term is small (,0.005 shortening is roughly the same. In the inset of Fig. 6 A, the
units) even for the case of the fastest shortening rate. smallest (solid circle) and the largest traces are self-normal-
Comparing the isosarcometric traces in Fig. 5 B with the ized and show similar durations (e.g., compare time duration
cell shortening twitches in Fig. 6, similarities and differences at 0.5 normalization). In contrast, the isosarcometric twitches
can be observed with respect to changing activator Ca levels. prolong with slower relaxation for higher Ca activation levels
Raising the Ca level increases peak force and produces a as shown the inset of Fig. 5 B.
larger degree of shortening. Note that while the time-to-peak The differences between the isosarcometric and cell
force is relatively constant in the isosarcometric case, the cell shortening twitches can be further illustrated by simulating

Biophysical Journal 95(5) 2368–2390

2380 Rice et al.

force coincides with the greatest lengthening of the series

elastic element. As the compliance increases, this point oc-
curs at a greater delay from the initiation of the twitch. The
twitch duration decreases because relengthening hastens
relaxation as seen experimentally (21). In the model, the re-
lengthening increases the mean distortion of the strongly-
bound crossbridge states, and in turn, decreases the forward
rotation rate of the crossbridges (see Eq. 22) to hasten force
decline.

Ktr
Another common characterization of muscle is Ktr, the rate of
force development after a sudden length change that is
thought to detach most crossbridges and drop the force to
near zero. Simulation of Ktr experiments are carried out in the
model by first applying a constant level of activator [Ca] until
a steady response is obtained (as shown by the fixed force
level before the 2 s window shown in Fig. 8 A). The cross-
bridge forward transition rates are decreased ( fappT and hfT
are 200-fold slower) and reverse rates are increased (gappT
and hbT are 200-fold faster) for 2 ms to simulate the rapid
removal of strongly-bound crossbridges by the quick release
and restretch that is typical in experimental protocols. Intui-
tively, one could attempt a more direct mapping in the sim-
ulation to the mechanical perturbations in the experimental
protocol. However, we want to simulate crossbridge attach-
ment and force redevelopment that underlies the main phe-
nomenon of Ktr. Attempting to simulate the fast crossbridge
detachment events from the mechanical length changes
FIGURE 7 Fixed muscle twitches with internal shortening. The cell is would increase the complexity of the simulation. Moreover,
held at a fixed total length, but a series elastic element allows for internal one can question the value of simulating the fast force drop
shortening. The traces correspond to different stiffness values of the series
elastic element with values of 50 ( ), 10, 5, 3, 2, 1.4, and 1 (d) where units
for which the theoretical underpinnings are less well under-
of KSE are normalized force-per-micrometer extension. (A) Sarcomere stood than the force redevelopment steps.
length is shown to illustrate the degree of internal shortening. (B) The total As shown in Fig. 8 A, the recovery is well fit by a single
muscle force is shown for the same range of KSE values as in panel A. exponential (dashed traces) with Ktr rates that increase with
Greater degrees of internal shortening produce later times-to-peak force and the activation level. Likewise, experimental results show a
also faster relaxation rates as myocyte relengthening increases crossbridge
strain and has an effect on crossbridge cycling (see text for details). Data
recovery that is well fit by a single exponential with Ktr rate
correspond to rat at 22.5C. that increases with Ca-based activation level in cardiac
muscle (40). While initial theories proposed that Ktr should
reflect crossbridge turnover rates only, later interpretations
suggest an interplay of Ca-based activation and turnover
fixed muscle-length twitches in which the degree of internal rates that causes Ktr rate to increase with Ca-activation level
shortening is changed. In Fig. 7, the different traces corre- (41,42). The results from Fig. 8 A are plotted as a function of
spond to increasing end compliances and larger degrees of Ca level as the 20C trace in Fig. 8 B. Also shown are
internal shortening. For the case with the least end compli- corresponding results for 15 and 25C. Similar to experi-
ance (KSE ¼ 50), there is a very small amount of internal mental results, the rates increase with temperature with the
shortening (asterisked trace in Fig. 7 A). The corresponding divergence increasing at the highest activation levels
force transient in Fig. 7 B is very similar to the isosarcometric (10,11).
twitch in Fig. 5 for SL ¼ 2.2 mm. As the end compliance
increases (KSE decreases), the amount of internal shortening
increases as shown in Fig. 7 A. With greater internal short-
Unloaded cell shortening
ening, the total force as measured at the muscle end changes
to show a later peak and increased rate of relaxation. The We coupled the myofilament model to the Chicago model of
increased time-to-peak results because maximal recorded the rabbit ventricular myocyte (32). Note that all of the

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2381

FIGURE 9 Simulation of cardiac cell with electrophysiology and Ca-

handling mechanisms. (A) The myofilament model developed here is cou-
pled to the Chicago model of the rabbit ventricular myocyte (32). Results are
shown for the combined model (A) and the similar experimental data (B).
The labeled responses show the action potentials, bulk myoplasmic Ca
transients, and cell shortening signals. This figure illustrates suitability of the
myofilament model for coupling with existing models of electrophysiology
FIGURE 8 Ktr as function of Ca-level and temperature. (A) The model is and Ca-handling mechanisms, and the ensemble model recapitulates com-
activated by a constant level of activator [Ca] for 2 s until a steady response mon experimental characterization such as cell shortening. Data correspond
is obtained. To simulate quick release and restretch in real muscle Ktr to rabbit at 37C.
protocols, the crossbridge transitions rates are modified for 2 ms to induce
rapid removal of strongly-bound crossbridges in the model (see text from
details). The recovery is well fit by a single exponential with rate Ktr that
90% and 40% of the default values, respectively. Otherwise
increases with the activation level. The shaded traces show the model default values are used for the Chicago model.
responses, while the dashed overlays show the exponential fits. (B) Ktr is
shown as function of Ca level for three temperatures as labeled. The 20C
trace corresponds to the data in panel A. Similar to experimental results Length effects on the Ca transient
(10,11), the rates increase with temperature with the divergence increasing at
the highest activations levels. Data correspond to rat at the temperature The bulk cytosolic Ca transient from the Chicago model is
labeled. used to compute the binding of Ca to the low affinity regu-
latory sites on troponin in the myofilament model. This step is
straightforward, except that the Ca affinity of this site is a
function of both sarcomere length and the fraction of strongly-
myofilament model data shown up to this point has been for bound crossbridges (Eq. 37). Because of this functional de-
rat at lower temperature. Now the myofilament model is pendence, the amount of Ca bound to troponin will change as
adjusted to replicate rabbit (see Methods) at physiological the fraction of strongly-bound crossbridges changes and as the
temperature (37C). Results are shown in Fig. 9 for the sarcomere length increases or decreases.
combined model and the similar experimental data (43). Fig. 10 shows simulation of effects of internal shortening
The responses show the action potentials, Ca transients, and on the Ca transient. The protocol generates a steady output
cell shortening signals for both model (Fig. 9 A) and exper- by stimulating the cell for nine beats at 1 Hz with fixed
imental characterizations (Fig. 9 B). Note that the particular muscle length with internal shortening. Default values of the
experimental data set here shows a small and prolonged Chicago model are used. The protocol and extent of internal
Ca transient that could be better replicated by decreasing shortening is similar to that in an experimental study (44). For
L-type Ca influx and the forward rate of SERCA pump to beat 10, either the cell is allowed to internally shorten as

Biophysical Journal 95(5) 2368–2390

2382 Rice et al.

force produces augmented Ca-binding to troponin that ini-

tially decreases the Ca transient. Later the increased Ca
bound to troponin is released so that the later Ca transient is
slightly above the internal shortening case (compare solid
circle and asterisked traces). In the experimental study, un-
controlled shortening also increases the Ca transient; how-
ever, a similar crossover feature cannot be detected as the
noise level is too large and presumably would obscure such
an effect if present. However, the crossover effect is seen in
other studies (45,46) using long and short sarcomere length
twitches, which produces larger changes in developed force
and more distinct changes in the Ca transient.

DISCUSSION
An ODE-based model is developed here based on traditional
approaches; however, new formulations of some aspects are
developed to overcome limitations associated with traditional
mean-field approximations. The approximate and spatially
compressed model presented here can recapitulate many of
the commonly measured steady-state and dynamic responses
seen in cardiac muscle. As in all modeling studies, the ability
to generate realistic responses does not prove that the un-
derlying biophysical mechanisms are correctly represented.
The veracity of this statement is obvious for this study as we
have clearly described a number of approximations that do
not match the real biophysics. Specifically, we cannot di-
rectly represent nearest-neighbor interactions of RUs. Also,
force is computed by a mean-field approach using the state-
FIGURE 10 Simulation of internal shortening effects on the Ca transient. occupancy multiplied by the mean strain of the strongly-
The combined myofilament model and Chicago model of the rabbit bound states. We accept these approximations as necessary
ventricular myocyte is used to simulate the effects of cell shortening on
to maintain the system as computationally efficient ODEs that
the Ca transient. The protocol generates a steady output by stimulating the
cell for nine beats with fixed muscle length with internal shortening (KSE ¼ are suitable for large-scale tissue simulations (1,47). In the
1 normalized force units per micrometer extension). Then for beat 10, either following discussion, we focus on several of the limitations of
the cell is allowed to internally shorten as before (d) or held at a fixed our approach. Then our modeling approach is compared with
sarcomere length ( ) to simulate length control. The panels show the resulting existing published models.
sarcomere length (A), force (B), and bulk myoplasmic Ca transient (C) for each
case. As seen in experimental studies, the isosarcometric case shows increased
force and a decrease in the Ca transient. In the model, the increased force
produces augmented Ca-binding to troponin that initially decreases the Ca Limitations
transient. Later, the bound Ca is released, and the later Ca transient is slightly
above the internal shortening case (compare the d and traces). Assumption of spatially homogeneity
The model implicitly assumes several types of homogeneity.
First the model assumes that myofilaments are activated by a
uniform Ca concentration. This assumption conflicts with
before (solid circle) or held at a fixed sarcomere length (as- considerable evidence showing that Ca-induced Ca release is
terisk) to simulate length control. Note that up to beat 10, the inherently spatial with specialized mechanisms to produce
runs are equivalent so that all state variables such as sarco- sarcoplasmic reticulum Ca release in response to local influx
plasmic reticulum loading and intracellular ion concentra- via L-type channels. However, at the level of the myofila-
tions will be the same. The panels show the resulting ments, we expect that Ca is more uniformly distributed over
sarcomere length (Fig. 10 A), force (Fig. 10 B), and bulk the much longer time frame of the force generation. Com-
myoplasmic Ca transient (Fig. 10 C). As seen in experimental putational modeling suggests spatial [Ca] gradients in the
studies, the isosarcometric case shows increased force and a half-sarcomere are largest during the upstroke of the Ca
decrease in the Ca transient as compared to the uncontrolled transient, and the spatial gradients are small by 15 ms after
case with internal shortening. In the model, the increased the peak of the transient (48). Hence, except for the rapid

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2383

upstroke, the gradients are generally small, and [Ca] is nearly Comparison to previous modeling work
uniform in space although changing with time. Sarcomere geometry
A second homogeneity assumption is that all crossbridges
are equally likely to bind and contribute equally to force The length dependence of maximal activated force is as-
generation in the single-overlap region. These assumptions are sumed to reflect the overlap of thick and thin filaments
contradicted by evidence that suggest the intrinsic spacing of resulting from sarcomere geometry. This basic premise can
myosin and actin sites are different so that binding probabili- be traced back to the work of Gordon et al. (2) in skeletal
ties can be a function of the location along the thick and thin muscle. However, such an approach to modeling maximal
filaments (49). Moreover, this work also suggests that com- activated force in cardiac muscle requires assumptions
pliance in the filaments can produce a realignment of the about filament lengths that differ from skeletal muscle (1).
binding site and can introduce cases where crossbridges can The traditional assumption has been that skeletal and car-
contribute different amounts of developed force depending on diac muscle have equivalent sarcomere geometries and fil-
the number and location along the z-disk to m-line direction. ament lengths, and some modeling efforts reflect this
Our model cannot capture these types of compliant realign- premise (14,15). Other modeling efforts (13,53) have used
ment effects. fitting parameters to physiological data on maximal acti-
vated force rather than attempting to model the sarcomere
geometry explicitly. An alternative explanation of the peak
Mean-field approximations for crossbridge cycling of force at lengths .2.0–2.2 mm range found for skeletal
muscle is SL-dependent changes in lattice spacing. Some
As described in Methods, the crossbridge state representation recent modeling efforts (14,54) have included the putative
proposed here considers both the probability of strongly- effects of lattice spacing to explain length-dependent ef-
bound states and the mean distortion of the states. Moreover, fects. Similar lattice spacing effects are not included in the
the mean strain affects the transition rates between cross- model here because appropriate length-dependent effects
bridge states. This representation conflicts with most com- could be simulated based on sarcomere overlap changes
mon notions of crossbridge cycling in which strain affects the alone. Moreover, controversy exists as to whether lattice
transitions rates on an individual crossbridge basis only. spacing changes are large enough to produce length-
However, tracking such interactions requires partial differ- dependent changes in Ca sensitivity and maximal force
ential equations (e.g., (24–26)) or Monte Carlo approaches (1,55,56).
(e.g., (49–51)). In fact, common notions suggest individual The first modeling work to assume different cardiac sar-
strain strongly affects the transition rates, so that mean-field comere geometry was that of Landesberg and Sideman (57);
approaches may prove difficult to apply. Specifically, the however, their experimental justification was mostly lacking.
mean-field approach is best suited for conditions in which More recent characterizations (58) have shown that maximal
the population of states and the corresponding strains of the activated force in cardiac muscle does peak at sarcomere
states are weakly correlated. Despite these observations, the length at 2.3–2.4 mm which is larger than the 2.0–2.2 mm
representation proposed here produces reasonable results for range for skeletal muscle. In addition, maximal activated
the experimental protocols studied. Hence, the approxima- force is seen to linearly increase up to 2.15 mm in trabeculae
tion appears useful for the purposes of this research effort. (3), which contradicts the plateau in force expected for
skeletal muscle geometry in the 2.0–2.2 mm range. Similarly,
maximal force and ATPase rates are found to rise linearly
Simplifications of known complexity
through the 2.0–2.2 mm range (22). Such findings are most
The modeling work contains important simplifications that easily explained by an increase in recruitable crossbridges as
are in conflict with known features of cardiac muscle. As a a result of an increased single-overlap region.
reasonable first approximation, the passive force is repre- The sarcomere geometry in Fig. 1 A is based on physio-
sented as a simple, time-invariant elastic element that is in logical data described above, although we have not found
series with a Newtonian viscous element. This formulation the exact anatomical data to confirm these assumptions.
cannot reproduce experimental data showing more com- However, recent evidence suggests complexity in setting
plexity and time variation in passive force (6). Likewise, the thin-filament lengths in skeletal and cardiac muscles. For
representations of a mass element and springlike series elastic example, a siRNA nebulin knockdown study performed in
elements are other linear simplifications that can only ap- cultured cardiomyocytes resulted in the absence of localized
proximate reality (37). Our three-state crossbridge cycle is a tropomodulin and a 30% increase in thin filament lengths
simplification as many more states can be identified in bio- (59). However, nebulin is thought to be expressed in only a
chemical studies (e.g., (26)). Likewise, we consider only two small fraction of cardiomyocytes in mouse (60) so the exact
states for RUs (permissive and nonpermissive). In contrast, mechanism for producing longer lengths is unclear. In skel-
other research has suggested three states for thin-filament etal muscle that expresses a nebulin in a stoichiometric ratio
activation (52). with thin filaments, the role of setting length is also contro-

Biophysical Journal 95(5) 2368–2390

2384 Rice et al.

versial (60,61). For example, wild-type mouse shows aver- binding cause the Ca-based activation steps to have slow
age thin filament lengths above 1.25 mm in three of four kinetics that mirror the slower crossbridge attachment (42).
skeletal muscle types examined (60). This value is larger than The feedback results in unphysiological responses such as
1.05–1.10 mm, which has been the traditional assumption isosarcometric twitches that show a peak force that occurs
for skeletal and cardiac muscle in most modeling work. The later and later as peak force increases (15). In contrast, the
knockout of nebulin produced shorter but fairly uniform time-to-peak force shows little change with peak force levels
thin filament lengths of ;1 mm in all skeletal muscle types under length-control conditions (20,39). Moreover, the pos-
examined, a finding that suggests a nebulin-independent itive feedback of developed force on Ca-binding can produce
mechanism for setting thin filament lengths (60). The nebulin- a system that can show very long times to reach steady state
independent mechanism may correspond to that in cardiac as well as extreme sensitivity to parameters (47). Specifically,
muscle that is generally assumed to lack nebulin, but the story extreme sensitivity to parameters arise as multiple steady-
may not be that simple. The knockout of nebulin produced state solutions can be reached from slightly different initial
additional features such as ultrastructural abnormalities, conditions.
an inability to maintain myofibrillar integrity during muscle The Ca-based activation in the model developed here re-
contraction, and reduced contraction force. Hence, the re- quires a steep nonlinear relation between the binding of Ca to
moval of nebulin in skeletal muscle should not be considered troponin and the shift from nonpermissive to permissive
to correspond directly to typical cardiac muscle sarcomere RUs. This construction follows directly from earlier work
structure. Clearly, more study is required to understand the (15) that sought a phenomenological approach to capture the
mechanisms of setting thin filament lengths in different putative effects of end-to-end RU interactions. While the
muscle types. approach in the current model is essentially the same and
could still be considered phenomenological, more recent
Ca-based activation
spatially explicit models have shown that end-to-end inter-
action of RUs can produce a strong nonlinear effect. Spe-
In the spatially explicit modeling, activation of RUs (troponin/ cifically, a model using the Ising approach (12) shows
tropomyosin) from a nonpermissive to permissive state strongly cooperative activation of the thin-filament simulated
will increase the Ca affinity by approximately a factor of 10 F-Ca relations that are very similar to Hill functions. More-
for that unit. In the ODE approach, individual units are not over, the model F-Ca relations show slight deviations from
tracked, so such an effect on a unit-by-unit basis cannot be true Hill functions such that the low Ca region is somewhat
directly implemented. A traditional mean-field approxima- more cooperative than the high Ca region as seen in many
tion assumes a uniform increase in the Ca affinity for all experimental studies (64–66). The spatially explicit approach
units. As described previously (1), traditional mean-field is carried over to a second study using Monte Carlo ap-
approximations with global feedback of developed force proaches that include more detail such as crossbridges and
on Ca-binding generate unphysiological responses such as the thick- and thin-filament structure of the sarcomere (51).
F-Ca relations that are not cooperative enough at high and This study also shows that end-to-end interaction of RUs
low Ca regions and are too cooperative in the middle region. produces steep F-Ca relations that resemble those measured
This effect on F-Ca relations is generic and appears in a experimentally.
wide variety of models with different constructions and While the steep nonlinearity in the model can produce
parameter sets (e.g., (13–15)). For a specific example, realistic Ca sensitivity as seen in the spatially explicit models,
compare F-Ca results for models with global feedback on it is not obvious how to incorporate force-dependent Ca-
Ca binding (M1 and M2) with the model without (M5) in affinity in an ODE-based model without generating a tradi-
Fig. 5 of Schneider et al. (14). Another important issue tional mean-field approximation. Our modeling efforts here
arises in that strong global feedback of developed force on propose a novel construction that artificially separates the Ca
Ca-binding can generate hysteresis that is not seen in real binding to troponin that is assumed to control thin-filament
muscle responses (47). Small amounts of global feedback of activation (termed ‘‘regulatory Ca binding’’) and Ca binding
developed force on Ca-binding can be used with few dele- that is sensed by the cell (termed ‘‘apparent Ca binding’’).
terious effects (e.g., (15,62)), However, the resulting F-Ca This approximation produces realistic Ca sensitivity with
relations may not be steep enough, and the small change in F-Ca relations that are similar to true Hill functions (Fig. 3 A).
affinity will not produce the force-dependent changes in the Also, the attachment of strongly-bound crossbridges can
intracellular Ca transient. increase the apparent Ca binding to troponin that is thought
Besides the problems with steady-state responses, global to alter the intracellular Ca transient, as simulated in Fig. 10.
feedback of developed force on Ca-binding can produce However, our approach is an approximation because in
problems in dynamic responses. Specifically, Ca activation reality, there is only one pool of troponin that plays both
kinetics are generally assumed to be faster, whereas the regulatory and buffering roles, although the effective roles
crossbridge cycling kinetics are slower (e.g., (31,38)). In- may depend on the spatial and temporal scales that one
cluding global feedback terms of attached crossbridges on Ca considers.

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2385

FIGURE 11 Schematic representation of 50% thin-fila-

ment activation for different levels of nearest-neighbor
cooperativity between RUs. (A) With no nearest-neighbor
coupling between RUs, one predicts a random arrangement
of activated RUs (raised) and attached, force-generating
crossbridges (hatched). (B) With strong nearest-neighbor
coupling between RUs, the 50% activation point can be
represented by a continuous run of half the RUs in the on-
conformation followed by a run of half the RUs in the off-
conformation. (C) For the extreme amount of nearest-neighbor
coupling between RUs, the whole thin-filament switches
from nonpermissive to permissive in unison. In this case,
50% activation can be represented in a cross-sectional view
of the sarcomere lattice where whole thin filaments would be
either in the on- (d) or the off- (s) conformation. Hence,
recruitment of RUs is at the level of whole thin filaments as
opposed to along the length of the thin filament.

To better illustrate the spatial scales, we propose different cooperativity. From the point of view of the myosin heads
schematic examples of the thin filament as shown in Fig. 11. located up- and downstream, the thin filament is permissive
Assuming no nearest-neighbor coupling between the RUs, a and attachment is facilitated (similar to the depictions in Fig.
random arrangement of permissive RUs is produced, as de- 11, B and C). Hence we assume thin-filament regulation is
picted in Fig. 11 A. Many existing models of the myofila- local to a given filament and not dependent on the bulk
ments have an implicit assumption that the strong binding of number of attached crossbridges elsewhere in the sarcomere.
crossbridges is tightly controlled by the adjacent RUs. If one One additional feature is also important. The sarcomere
assumes little or no nearest-neighbor coupling between the single-overlap fraction sets the number of adjacent myosin
RUs, then recruitment of crossbridges is assumed to be along heads and the potential number of crossbridges that can form
the length to the thin filament in locations corresponding to on each thin filament. For this reason, we have formulated
permissive RUs. In contrast, Fig. 11 B shows an example regulatory Ca binding that depends on sarcomere length but
with strong nearest-neighbor coupling between RUs so that is not dependent on the bulk fraction of attached crossbridges
uniformity is promoted between adjacent RUs. Now the 50% (see Eq. 9).
activation point corresponds to continuous run of RUs in Now consider the Ca binding sensed by the cell on the
the on-conformation followed by a run of RUs in the off- macroscopic scale. Again consider the extreme case for
conformation. While this amount of coupling may be stron- which the whole thin filament switches from nonpermissive
ger than common intuition, our spatially explicit results to permissive in unison. Clearly a single activated thin fila-
predict correlation of RUs states at distances of roughly half ment would not be sensed by the cell in terms of Ca buffering.
the thin filament at 50% activation (12). Fig. 11 C shows an Hence, the first activated thin filament will have a negligible
example of extreme coupling between RUs in which the effect on the Ca transient. However, after a substantial
whole thin filament switches from nonpermissive to per- fraction of thin filaments are activated, one would predict an
missive in unison as suggested by Brandt et al. (67). In this effect. For example, if half of thin filaments are activated,
case, 50% activation can be represented in a cross-sectional then we expect roughly one-half of troponin to have high
view where whole thin filaments would be either in the on- or affinity in the extreme case. In our model, we attempt to
the off-conformation. capture this effect by using the fraction of strongly-bound
We propose that the picture in Fig. 11 C may be closer to crossbridges as a proxy for the recruitment of thin filaments
reality than that of Fig. 11 A. With such a view, the concepts in the cross-sectional plane. There is an additional effect in
of spatial averages can change. A spatially compressed model that not all RUs are in close proximity to a crossbridge be-
computes a single scalar to represent the fraction of permis- cause of the sarcomere geometry. Because of this effect,
sive RUs, and hence, the model cannot directly distinguish sarcomere length also plays a role in determining the number
recruitment in the direction along the filament from recruit- of permissive RUs. Hence, our formulation of apparent Ca
ment in the cross-sectional plane. However, we have chosen binding that is sensed by the cell has contributions from both
the model construction to favor different views of recruitment sarcomere length and the fraction of strongly-bound cross-
for different purposes. For example, early in the activation bridges (see Eq. 37).
process, the thin filament shifts to permissive conformation Besides the spatial aspects just described, the difference in
before crossbridges attach. Once the first crossbridge binds, timescale can effect how the model variables are interpreted.
its associated RU and the neighboring up- and downstream For example, a typical twitch occurs over a longer timescale
RUs will be permissive, assuming high nearest-neighbor that reflects crossbridge kinetics more than Ca activation

Biophysical Journal 95(5) 2368–2390

2386 Rice et al.

events (31,38). At this longer timescale, force and thin fila- neighbor coupling. Neighboring myosins will also sense a
ment activation may track together closely as strongly-bound permissive environment and crossbridges will bind and cycle
crossbridges tend to keep the thin filaments in a permissive with rates that are independent of Ca level. Hence on the local
conformation even after Ca has begun to dissociate from scale, the world is permissive and independent of Ca level, and
troponin (68). Hence, on the macroscopic scale over a long the strain of crossbridges should depend only on their own
timeframe, the thin-filament activation and affinity for Ca intrinsic cycling rates and on the net motion of its thick and
may mirror the bulk population of strongly-bound cross- thin filaments. For these reasons, our computation of mean
bridges, whereas on the microscopic scale or short timeframe, strain does not contain a contribution of mean population of
the bulk attachment of crossbridges should not be an im- the strongly-bound states.
portant variable for local activation events. In all modeling efforts, some level of abstraction is chosen
as a compromise between parsimony and complexity. We
chose a three-state crossbridge cycle and propose that it has
Three-state crossbridge scheme
several advantages over a more parsimonious two-state
The model is constructed around a three-state model that is model for the crossbridge cycle as first proposed by Huxley
adapted from the general approach of Campbell and co- (24). In this formulation, crossbridges are assumed to be ei-
workers (19,69,70). While many more crossbridge states ther detached (i.e., equivalently weakly bound) or attached
have been identified biochemically, the three-state model (i.e., equivalently strongly bound). We can outline the ad-
retains enough machinery to recapitulate many measured vantages of moving to the more complex three-state cross-
phenomena including force-velocity relations, small step bridge cycle in the following three areas:
responses, and harmonic responses (19). The main dif-
ferences in the work here is that we have added a more 1. ATP utilization. A two-state approach requires that cross-
cooperative Ca-activation scheme; included more strain de- bridge detachment proceed via an ATP-utilizing path.
pendence in the crossbridge transition rates; and reworked However, this step is often characterized as slow, so
the calculation of mean strain of the strongly-bound states. relaxation is inhibited unless an unrealistically high ATP
The Ca-activation approach is described above. In the model utilization is assumed or multiple power strokes are as-
developed here, both the crossbridge forward isomerization sumed per ATP hydrolyzed, a feature at odds with most
rate hfT and the ATP-consuming transition rate gxbT are very current theories (1). The inability to reconcile faster relax-
strain-dependent. In contrast, the work of Razumova et al. ation with slow ATPase rates have led to speculation that a
(19) included strain dependence on the ATP consuming fast reverse power stroke can predominate under some
transition only (analogous to gxbT in this study). For the conditions (31,72). With the three-state model, the system
protocols investigated in that study, strain-dependence on the can relax back from the force-generating state to the relaxed
ATP-consuming transition has the dominant effect, and in- state via fast reversible reactions that do not require energy
cluding strain dependence on other transition rates produces usage (ATP hydrolysis) under many conditions. For
relatively minor effects. Another change in the model here is example, during isometric twitches, this pathway will
faster transition rates for crossbridge cycling that correspond predominate and decrease ATP consumption under condi-
to higher temperatures. The higher cycling rates are also tions where no net work is done, as first described by
important to recapitulate a rapid unloaded shortening veloc- Fenn (73).
ity that can be quite large even at subphysiological temper- 2. Crossbridge strain. Another difficulty associated with a
atures (e.g., 23.4 mm/s at 30C (36)). two-state crossbridge representation is how to represent
The modeling work here includes a new formulation of crossbridge strain. The Landesberg-Sideman models (74)
mean strain of the strongly-bound states. In the original model and derivatives use an approach where the strain is an
of Campbell and co-workers, calculation of mean strain has a instantaneous function of velocity implemented with a
dependence on the mean occupancy of the crossbridge states. viscosity-like term for unitary crossbridge force. The
Such an approach makes intuitive sense given that the model viscosity-like effect is also predicted by the Huxley
couples Ca-based activation to crossbridge cycling steps. For model for which the mean-strain of a crossbridge popu-
example, if one envisions the scenario shown in Fig. 11 A, then lation decreases for a constant shortening velocity (6).
the binding of crossbridges and subsequent cycling will However, computing strain as an instantaneous function
closely mirror the Ca-based activation steps. In contrast, if one of velocity precludes the interplay of both shortening
considers the picture in Fig. 11, B and C, then crossbridge velocity and crossbridge turnover in determining strain.
attachment is only affected by the local environment and not In comparison, Negroni and Lascano (53) employ a
by the amount of the activation in the cross-sectional plane. construction of a single spring with a movable attach-
Indeed, on the microscopic scale, consider the first strongly- ment that represents the ensemble of the attached cross-
bound crossbridge in a sarcomere. As the first crossbridge bridges. Here the moving attachment point can allow a
binds, its associated RU and the neighboring RUs up- and resetting of strain with time, similar to the contribution
downstream will be permissive, assuming strong nearest- of both shortening velocity and crossbridge turnover

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2387

rates proposed here (Eqs. 29 and 30). In the three-state to represent rabbit at physiological temperature. This
crossbridge cycle used here, we hope to provide a modified version of the myofilament model is coupled to
more direct mapping to the biophysics of crossbridge the Chicago model of the rabbit ventricular myocyte, and
attachment and strain induction, although admittedly a the integrated model recapitulates the cellular electrophys-
mean-field approximation is still required that is ulti- iology, Ca handling, and myofilament responses. In the
mately at odds with a true spatially explicit calculation integrated model, changing sarcomere length and developed
of strain. force can alter the intracellular Ca transient as seen in ex-
3. Prolongation of relaxation. A side effect of the three- perimental measures. In conclusion, while containing many
state model is that we found a prolongation of twitches at approximations, the model can replicate a wide range of
higher force levels (see inset of Fig. 5 B). The prolon- experimental data. We hope that this model will provide the
gation occurs as more time is required to transition back community with a relatively simple representation of car-
from the PostR state to the PreR state during the relax- diac myofilaments that retains enough mechanistic under-
ation process. The prolongation is important to replicate pinnings to provide flexibility and extensibility for future
isometric twitches for which twitch duration increases model development.
with twitch force (20). In contrast to previous work with
a two-state crossbridge model, two and three cross-
bridges were assumed to attach cooperatively to produce
similar prolongation effects (71). These model results do APPENDIX: ADDITIONAL EQUATIONS
not confirm one mechanism over others, and indeed,
multiple mechanisms could contribute to force-dependent Sarcomere geometry
prolongation of twitches. This is the end of the single-overlap region nearest the z-line:

sovrze ðxÞ ¼ minðlengththick =2; x=2Þ for SLmin # x # SLmax :

CONCLUSIONS (42)

An ODE-based model is developed here based on tradi- This is the end of the single-overlap region nearest the center line:
tional approaches; however, new formulations of some as-
pects were developed to overcome limitations associated sovrcle ðxÞ ¼ maxðx=2  ðx-lengththin Þ; lengthhbare =2Þ
with mean-field approximations. Specifically, we propose for SLmin # x # SLmax ; (43)
that cooperative activation of the thin filament and the
strain-dependent transitions of the crossbridge cycle are
inherently local phenomena that can only be approximately lengthsovr ðxÞ ¼ sovrze ðxÞ  sovrcle ðxÞ; (44)
described by nonspatial, state-variable models. We have
attempted to strike a reasonable balance between mecha-
nistic detail and model parsimony while including suffi- 2 3 lengthsovr ðxÞ
cient cellular machinery to recapitulate a wide range of the SOVFthick ðxÞ ¼ ; (45)
lengththick  lengthhbare
commonly measured steady-state and dynamic responses in
cardiac muscle. Specifically, the steady-state responses are
F-SL, F-Ca, SL-Ca, and F-V relations. Dynamic responses
lengthsovr ðxÞ
are isometric and cell-shortening twitches and Ktr including SOVFthin ðxÞ ¼ : (46)
Ca-activation effects. The model responses are comparable lengththin
to a wide range of experimental data available in the liter-
ature for rat at or near room temperature. With a small
number of parameter changes, the model can be converted Normalized passive force


PContitin 3 ðexpðPExptitin 3 ðx  SLrest ÞÞ  1Þ if x $ SLrest
Ftitin ðxÞ ¼ ; (47)
PContitin 3 ðexpðPExptitin 3 ðSLrest  xÞÞ  1Þ if x , SLrest

PConcollagen 3 ðexpðPExpcollagen 3 ðx  SLcollagen ÞÞ  1Þ if x $ SLcollagen
Fcollagen ðxÞ ¼ ; (48)
0 if x , SLcollagen

Ftitin ðxÞ if isolated cell
Fpassive ðxÞ ¼ : (49)
Ftitin ðxÞ 1 Fcollagen ðxÞ if trabeculae

Biophysical Journal 95(5) 2368–2390

2388 Rice et al.

Calculation of complete muscle response

(
d IntegralForce 1 ðSL0  SLÞ 3 viscosity
SL ¼ if not isosarcometric ; (50)
dt mass
0 if isosarcometric
Z t1
IntegralForce ¼ ðFactive ðxÞ 1 Fpassive ðxÞ  Fpreload  Fafterload ðxÞÞ dt; (51)
t0


Fpassive ðSL0 Þ if SL0 ¼
6 SLrest
Fpreload ¼ ; (52)
0 if SL0 ¼ SLrest
8
< isotonic contraction ðafter releaseÞ
constant
Fafterload if
Fafterload ðxÞ ¼ KSE 3 ðx  SL0 Þ if fixed length with internal contraction: (53)
:
0 otherwise

Equation for simulated calcium transient

 1=ðtt1 1Þ  1=ð1tt2 Þ

t1 2 t1 1
b¼  ; (54)
t2 t2
(
    Cadiastolic    for t # tstart
½CaðtÞ ¼ Caamplitude Cadiastolic : (55)
b
3 exp tttstart
1
 exp  ttstart
t2
1 Cadiastolic for t . tstart

Calculation of fluxes of Ca for apparent Ca binding d d d

lengthsovr ðxÞ ¼ sovrze ðxÞ  sovrcle ðxÞ; (61)
Apparent Ca binding is multiplied by total buffer concentration [Troponin] ¼ dt dt dt
70 mM in the Chicago model:
½TropApparent Ca ¼ ½Troponin 3 TropApparent ðxÞ: (56)
d
Flux of Ca onto the buffer is calculated using time rate of change of Eq. 37
d 2 3 lengthsovr ðxÞ
calculated with the chain rule: SOVFthick ðxÞ ¼ dt ; (62)
d d dt lengththick  lengthhbare
½TropApparent Ca ¼ ½Troponin3 TropApparent ðxÞ; (57)
dt dt

d d d d
TropApparent ðxÞ ¼  SOVFthin ðxÞ 3 TropL 1 ð1  SOVFthin ðxÞÞ 3 TropL 1 SOVFthin ðxÞ
dt dt dt dt
3 ðFractSBXB 3 TropH 1 ð1  FractSBXB Þ 3 TropL Þ 1 SOVFthin ðxÞ
 
d d d d
3 FractSBXB 3 TropH 1 FractSBXB 3 TropH  FractSBXB 3 TropL 1 ð1  FractSBXB Þ 3 TropL ;
dt dt dt dt
(58)

(
1 dSL d
d
sovrze ðxÞ ¼  for x , lengththick ; (59) d lengthsovr ðxÞ
dt 2 dt
0 otherwise SOVFthin ðxÞ ¼ dt ; (63)
dt lengththin
(
1 dSL
d
sovrcle ðxÞ ¼  for 2 3 lengththin  x . lengthhbare ; d d
dt 2 dt XBPreR 1 XBPostR
0 otherwise d
FractSBXB ¼ dt Max dt Max : (64)
(60) dt XBPreR 1 XBPostR

Biophysical Journal 95(5) 2368–2390

ODE-Based Model of Cardiac Myofilament 2389

SUPPLEMENTARY MATERIAL 16. Katsnelson, L. B., and V. S. Markhasin. 1996. Mathematical modeling
of relations between the kinetics of free intracellular calcium and me-
To view all of the supplemental files associated with this chanical function of myocardium. J. Mol. Cell. Cardiol. 28:475–486.
article, visit www.biophysj.org. 17. Trybus, K. M., and E. W. Taylor. 1980. Kinetic studies of the cooper-
ative binding of subfragment 1 to regulated actin. Proc. Natl. Acad. Sci.
USA. 77:7209–7213.
We thank Henk ter Keurs for valuable discussions on how to represent passive
force. Ryan Littlefield provided valuable insights into the nebulin-based 18. Bremel, R. D., and A. Weber. 1972. Cooperation within actin filament
regulation of thin filament lengths. We also thank Jason Yang and Stuart in vertebrate skeletal muscle. Nat. New Biol. 238:97–101.
Campbell for providing very useful feedback in the article preparation. 19. Razumova, M. V., A. E. Bukatina, and K. B. Campbell. 1999.
Stiffness-distortion sarcomere model for muscle simulation. J. Appl.
These studies were supported, in part, by National Institutes of Health Physiol. 87:1861–1876.
grants No. HL-62426 (project 4) and No. HL-75494 to P.P.d.T. and grant
20. Janssen, P. M., and W. C. Hunter. 1995. Force, not sarcomere length,
No. HL30077 to D.M.B. correlates with prolongation of isosarcometric contraction. Am. J. Physiol.
269:H676–H685.
21. Sys, S. U., and D. L. Brutsaert. 1989. Determinants of force decline during
relaxation in isolated cardiac muscle. Am. J. Physiol. 257:H1490–H1497.
REFERENCES
22. Wannenburg, T., P. M. Janssen, D. Fan, and P. P. de Tombe. 1997. The
1. Rice, J. J., and P. P. de Tombe. 2004. Approaches to modeling cross- Frank-Starling mechanism is not mediated by changes in rate of cross-
bridges and calcium-dependent activation in cardiac muscle. Prog. bridge detachment. Am. J. Physiol. 273:H2428–H2435.
Biophys. Mol. Biol. 85:179–195. 23. Wang, G., and M. Kawai. 2001. Effect of temperature on elementary
2. Gordon, A. M., A. F. Huxley, and F. J. Julian. 1966. The variation in steps of the cross-bridge cycle in rabbit soleus slow-twitch muscle
isometric tension with sarcomere length in vertebrate muscle fibers. fibers. J. Physiol. 531:219–234.
J. Physiol. 184:170–192. 24. Huxley, A. F. 1957. Muscle structure and theories of contraction. Prog.
3. Kentish, J. C., H. E. ter Keurs, L. Ricciardi, J. J. Bucx, and M. I. Noble. Biophys. Chem. 7:255–318.
1986. Comparison between the sarcomere length-force relations of 25. Smith, N. P. 2003. From sarcomere to cell: an efficient algorithm for
intact and skinned trabeculae from rat right ventricle. Influence of linking mathematical models of muscle contraction. Bull. Math. Biol.
calcium concentrations on these relations. Circ. Res. 58:755–768. 65:1141–1162.
4. Bassani, J. W., R. A. Bassani, and D. M. Bers. 1995. Calibration of 26. Pate, E., and R. Cooke. 1986. A model for the interaction of muscle
INDO-1 and resting intracellular [Ca]i in intact rabbit cardiac myo- cross-bridges with ligands which compete with ATP. J. Theor. Biol.
cytes. Biophys. J. 68:1453–1460. 118:215–230.
5. Preetha, N., W. Yiming, M. Helmes, F. Norio, L. Siegfried, and H. 27. King, E. L., and C. Altman. 1956. A schematic method of deriving the
Granzier. 2005. Restoring force development by titin/connectin and rate laws for enzyme-catalyzed reactions. J. Am. Chem. Soc. 60:1375–1378.
assessment of Ig domain unfolding. J. Muscle Res. Cell Motil. 26:307–317. 28. Cohen, S., and A. Hindmarsh. 1996. CVODE, a stiff/nonstiff ODE
6. de Tombe, P. P., and H. E. ter Keurs. 1992. An internal viscous solver in C. Comput. Phys. 10:138–143.
element limits unloaded velocity of sarcomere shortening in rat myo- 29. Poggesi, C., C. Tesi, and R. Stehle. 2005. Sarcomeric determinants of
cardium. J. Physiol. 454:619–642. striated muscle relaxation kinetics. Pflugers Arch. 449:505–517.
7. Davis, J. P., C. Norman, T. Kobayashi, R. J. Solaro, D. R. Swartz, and 30. Fukuda, N., and S. Ishiwata. 1999. Effects of pH on spontaneous
S. B. Tikunova. 2007. Effects of thin and thick filament proteins on
tension oscillation in skinned bovine cardiac muscle. Pflugers Arch.
calcium binding and exchange with cardiac troponin C. Biophys. J.
438:125–132.
92:3195–3206.
31. Palmer, S., and J. C. Kentish. 1998. Roles of Ca21 and crossbridge
8. Guth, K., and J. D. Potter. 1987. Effect of rigor and cycling cross-
kinetics in determining the maximum rates of Ca21 activation and relax-
bridges on the structure of troponin C and on the Ca21 affinity of the
ation in rat and guinea pig skinned trabeculae. Circ. Res. 83:179–186.
Ca21-specific regulatory sites in skinned rabbit psoas fibers. J. Biol.
Chem. 262:13627–13635. 32. Shannon, T. R., F. Wang, J. Puglisi, C. Weber, and D. M. Bers. 2004.
A mathematical treatment of integrated Ca dynamics within the
9. Harrison, S. M., and D. M. Bers. 1989. Influence of temperature on the
ventricular myocyte. Biophys. J. 87:3351–3371.
calcium sensitivity of the myofilaments of skinned ventricular muscle
from the rabbit. J. Gen. Physiol. 93:411–428. 33. Dobesh, D. P., J. P. Konhilas, and P. P. de Tombe. 2002. Cooperative
activation in cardiac muscle: Impact of sarcomere length. Am. J.
10. de Tombe, P. P., and G. J. M. Stienen. 1997. The rate of tension
Physiol. Heart Circ. Physiol. 282:H1055–H1062.
redevelopment in rat cardiac muscle: Influence of temperature and
contractile activation level. Circulation. 96:517–518. 34. Solovyova, O., L. Katsnelson, S. Guriev, L. Nikitina, Y. Protsenko,
S. Routkevitch, and V. Markhasin. 2002. Mechanical inhomogeneity of
11. de Tombe, P. P., and G. J. M. Stienen. 2007. Impact of temperature on
cross-bridge cycling kinetics in rat myocardium. J. Physiol. 584:591– myocardium studied in parallel and serial cardiac muscle duplexes:
600. experiments and models. Chaos Solit. Fract. 13:1685–1711.
12. Rice, J. J., G. Stolovitzky, Y. Tu, and P. P. de Tombe. 2003. Ising 35. Lim, C. C., M. H. Helmes, D. B. Sawyer, M. Jain, and R. Liao.
model of cardiac thin filament activation with nearest-neighbor coop- 2001. High-throughput assessment of calcium sensitivity in skinned
erative interactions. Biophys. J. 84:897–909. cardiac myocytes. Am. J. Physiol. Heart Circ. Physiol. 281:H969–H974.
13. Niederer, S. A., P. J. Hunter, and N. P. Smith. 2006. A quantitative 36. de Tombe, P. P., and H. E. ter Keurs. 1990. Force and velocity of
analysis of cardiac myocyte relaxation: a simulation study. Biophys. J. sarcomere shortening in trabeculae from rat heart. Effects of temper-
90:1697–1722. ature. Circ. Res. 66:1239–1254.
14. Schneider, N. S., T. Shimayoshi, A. Amano, and T. Matsuda. 2006. 37. McMahon, T. A. 1984. Muscles, Reflexes, and Locomotion. Princeton
Mechanism of the Frank-Starling law—a simulation study with a novel University Press, Princeton, NJ.
cardiac muscle contraction model that includes titin and troponin I. 38. Janssen, P. M., L. B. Stull, and E. Marban. 2002. Myofilament properties
J. Mol. Cell. Cardiol. 41:522–536. comprise the rate-limiting step for cardiac relaxation at body temperature
15. Rice, J. J., R. L. Winslow, and W. C. Hunter. 1999. Comparison of in the rat. Am. J. Physiol. Heart Circ. Physiol. 282:H499–H507.
putative cooperative mechanisms in cardiac muscle: length dependence 39. de Tombe, P. P., and W. C. Little. 1994. Inotropic effects of ejection
and dynamic responses. Am. J. Physiol. 276:H1734–H1754. are myocardial properties. Am. J. Physiol. 266:H1202–H1213.

Biophysical Journal 95(5) 2368–2390

2390 Rice et al.

40. Wolff, M. R., K. S. McDonald, and R. L. Moss. 1995. Rate of tension 57. Landesberg, A., and S. Sideman. 1994. Coupling calcium binding to
development in cardiac muscle varies with level of activator calcium. troponin C and cross-bridge cycling in skinned cardiac cells. Am. J.
Circ. Res. 76:154–160. Physiol. 266:H1260–H1271.
41. Hancock, W. O., L. L. Huntsman, and A. M. Gordon. 1997. Models of 58. Weiwad, W. K., W. A. Linke, and M. H. Wussling. 2000. Sarcomere
calcium activation account for differences between skeletal and cardiac length-tension relationship of rat cardiac myocytes at lengths greater
force redevelopment kinetics. J. Muscle Res. Cell Motil. 18:671–681. than optimum. J. Mol. Cell. Cardiol. 32:247–259.
42. Campbell, K. 1997. Rate constant of muscle force redevelopment 59. McElhinny, A. S., C. Schwach, M. Valichnac, S. Mount-Patrick, and
reflects cooperative activation as well as cross-bridge kinetics. Biophys. C. C. Gregorio. 2005. Nebulin regulates the assembly and lengths of
J. 72:254–262. the thin filaments in striated muscle. J. Cell Biol. 170:947–957.
43. Bers, D. M. 2001. Excitation-Contraction Coupling and Cardiac 60. Bang, M. L., X. Li, R. Littlefield, S. Bremner, A. Thor, K. U.
Contractile Force. Springer-Verlag, New York. Knowlton, R. L. Lieber, and J. Chen. 2006. Nebulin-deficient mice
44. Janssen, P. M., and P. P. de Tombe. 1997. Uncontrolled sarcomere exhibit shorter thin filament lengths and reduced contractile function in
shortening increases intracellular Ca21 transient in rat cardiac trabec- skeletal muscle. J. Cell Biol. 173:905–916.
ulae. Am. J. Physiol. 272:H1892–H1897. 61. Fowler, V. M., C. R. KcKeown, and R. S. Fischer. 2006. Nebulin: does
45. Backx, P. H., and H. E. Ter Keurs. 1993. Fluorescent properties of rat it measure up as a ruler? Curr. Biol. 16:R18–R20.
cardiac trabeculae microinjected with fura-2 salt. Am. J. Physiol. 264: 62. Robinson, J. M., Y. Wang, W. G. Kerrick, R. Kawai, and H. C. Cheung.
H1098–H1110. 2002. Activation of striated muscle: nearest-neighbor regulatory-unit
46. Kentish, J. C., and A. Wrzosek. 1998. Changes in force and cytosolic and cross-bridge influence on myofilament kinetics. J. Mol. Biol. 322:
Ca21 concentration after length changes in isolated rat ventricular 1065–1088.
trabeculae. J. Physiol. 506:431–444. 63. Reference deleted in proof.
47. Rice, J. J., Y. Tu, C. Pogessi, and P. P. De Tombe. 2007. Spatially 64. Moss, R. L., G. G. Giulian, and M. L. Greaser. 1985. The effects of
compressed cardiac myofilament models generate hysteresis that is not partial extraction of TnC upon the tension-pCa relationship in rabbit
found in real muscle. Pacific Symp. Biocomp. 2008:378–389. skinned skeletal muscle fibers. J. Gen. Physiol. 86:585–600.
48. Wier, W. G., and D. T. Yue. 1986. Intracellular calcium transients 65. Moss, R. L., A. E. Swinford, and M. L. Greaser. 1983. Alterations in
underlying the short-term force-interval relationship in ferret ventric- the Ca21 sensitivity of tension development by single skeletal muscle
ular myocardium. J. Physiol. 376:507–530. fibers at stretched lengths. Biophys. J. 43:115–119.
49. Daniel, T. L., A. C. Trimble, and P. B. Chase. 1998. Compliant 66. Brandt, P. W., R. N. Cox, and M. Kawai. 1980. Can the binding of
realignment of binding sites in muscle: transient behavior and me- Ca21 to two regulatory sites on troponin C determine the steep pCa/
chanical tuning. Biophys. J. 74:1611–1621. tension relationship of skeletal muscle? Proc. Natl. Acad. Sci. USA. 77:
4717–4720.
50. Chase, P. B., J. M. Macpherson, and T. L. Daniel. 2004. A spatially
explicit nanomechanical model of the half-sarcomere: myofilament 67. Brandt, P. W., M. S. Diamond, J. S. Rutchik, and F. H. Schachat. 1987.
compliance affects Ca21-activation. Ann. Biomed. Eng. 32:1559–1568. Co-operative interactions between troponin-tropomyosin units extend
the length of the thin filament in skeletal muscle. J. Mol. Biol. 195:
51. Hussan, J., P. P. de Tombe, and J. J. Rice. 2006. A spatially detailed
885–896.
myofilament model as a basis for large-scale biological simulations.
IBM J. Res. Develop. 50:583–600. 68. Peterson, J. N., W. C. Hunter, and M. R. Berman. 1991. Estimated time
course of Ca21 bound to troponin C during relaxation in isolated
52. McKillop, D. F., and M. A. Geeves. 1993. Regulation of the interaction
cardiac muscle. Am. J. Physiol. 260:H1013–H1024.
between actin and myosin subfragment 1: evidence for three states of
the thin filament. Biophys. J. 65:693–701. 69. Campbell, K. B., M. V. Razumova, R. D. Kirkpatrick, and B. K.
Slinker. 2001. Myofilament kinetics in isometric twitch dynamics. Ann.
53. Negroni, J. A., and E. C. Lascano. 1996. A cardiac muscle model
Biomed. Eng. 29:384–405.
relating sarcomere dynamics to calcium kinetics. J. Mol. Cell. Cardiol.
28:915–929. 70. Razumova, M. V., A. E. Bukatina, and K. B. Campbell. 2000.
Different myofilament nearest-neighbor interactions have distinctive
54. Protsenko, Y. L., S. M. Routkevitch, V. Y. Gur’ev, L. B. Katsnelson,
effects on contractile behavior. Biophys. J. 78:3120–3137.
O. Solovyova, O. N. Lookin, A. A. Balakin, P. Kohl, and V. S.
Markhasin. 2005. Hybrid duplex: a novel method to study the contractile 71. Reference deleted in proof.
function of heterogeneous myocardium. Am. J .Physiol. Heart Circ. 72. Stehle, R., M. Kruger, and G. Pfitzer. 2002. Force kinetics and
Physiol. 289:H2733–H2746. individual sarcomere dynamics in cardiac myofibrils after rapid Ca21
55. Konhilas, J. P., T. C. Irving, and P. P. de Tombe. 2002. Frank-Starling changes. Biophys. J. 83:2152–2161.
law of the heart and the cellular mechanisms of length-dependent 73. Fenn, W. O. 1924. The relation between the work performed and the
activation. Pflugers Arch. 445:305–310. energy liberated in muscular contraction. J. Physiol. 58:373–395.
56. Konhilas, J. P., T. C. Irving, and P. P. de Tombe. 2002. Myofilament 74. Landesberg, A., and S. Sideman. 1994. Mechanical regulation of
calcium sensitivity in skinned rat cardiac trabeculae: role of interfila- cardiac muscle by coupling calcium kinetics with cross-bridge cycling:
ment spacing. Circ. Res. 90:59–65. a dynamic model. Am. J. Physiol. 267:H779–H795.

Biophysical Journal 95(5) 2368–2390