Long on Importance, Short on Mechanism
The Frank-Starling relationship is an intrinsic property of myocardium by which increased length (or ventricular volume) results in enhanced performance during the subsequent contraction.1–3⇓⇓ This relationship appears to be very important in cardiac function because increased venous return and the corresponding increase in end-diastolic volume result in greater stroke volume during the next beat. The ventricles can thus accommodate increased venous return by means of a more vigorous contraction that ejects the greater volume of blood from the heart.
Although the physiological significance of the Frank-Starling relationship is widely appreciated, its cellular basis is not well understood. One hypothesis is that more cross-bridges interact with actin at longer sarcomere lengths due to length-dependent reductions in lateral spacing between thick and thin filaments,3–5⇓⇓ ie, due to closer proximity to actin more crossbridges bind and thereby increase contractile force (see Figure). Measurements in permeabilized myocardium held at constant length have shown that osmotic compression increases force at each [Ca2+]. Thus, the greater Ca2+ sensitivity of force at long lengths can be achieved at short lengths by reducing fiber diameter. Moreover, osmotic compression actually eliminates the length dependence of Ca2+ sensitivity.4 This evidence suggests that lateral filament spacing is a primary determinant of the Frank-Starling relationship, but the evidence is incomplete because studies to date have only measured muscle diameter and not the actual spacing between thick and thin filaments. Diameter and filament separation are likely to increase or decrease in concert, but these changes need not be proportionate because the shape of the muscle cross section can change with stretch or with osmotic compression. Until now, no one has quantified the lateral spacing of thick and thin filaments in such experiments due to the technical difficulty of the measurements.
In this issue of Circulation Research, de Tombe and colleagues6 report results of studies in which x-ray diffraction was used to quantify the lateral spacing of thick and thin filaments in cardiac muscles in which force and length were also measured. The regular structure of the sarcomere produces an x-ray diffraction pattern that reports the lateral spacing between thick and thin filaments, ie, filament lattice spacing. As sarcomere length is increased, lattice spacing decreases and vice versa. The authors assessed force as functions of length and lattice spacing, and using osmotically active compounds, they were able to independently vary lattice spacing and length. To the authors’ great credit, such experiments are technically demanding and have yielded results that could profoundly influence current thinking.
Length-Dependent Variations in Force Can Occur With No Changes in Filament Lattice Spacing
Initial experiments by Konhilas et al6 reproduced important features of earlier work, ie, increases in muscle length or osmotic compression of skinned preparations were found to increase Ca2+ sensitivity. However, they found no effect of osmotic compression on the length dependence of Ca2+ sensitivity, which contrasts with an earlier report5 in which compression eliminated the length dependence. The authors suggest that the difference may be related to differences in species used, but they have not eliminated the possibility that there are unrecognized systematic differences in methods.
Using x-ray diffraction, Konhilas et al6 found that filament lattice spacing decreased when the muscle was stretched or osmotically compressed with dextran, which was the expected result. However, they found that dextran compressed muscle diameter to a greater extent than interfilament lattice spacing, ie, earlier studies overestimated the degree of lattice compression from measurements of muscle diameter. The most important results were as follows: (1) lattice compression with dextran tended to reduce the stretch-dependent decrease in filament lattice spacing, but had no significant effect on the length dependence of Ca2+ sensitivity of force; and (2) osmotic compression to achieve lattice spacings typical of a longer length produced no change in Ca2+ sensitivity of force. These results are not easily reconciled with models in which filament lattice spacing is the sole determinant of the length dependence of Ca2+ sensitivity and at the very least suggest that other mechanisms are involved.
Given the technical difficulty of these measurements, consideration needs to be given to experimental uncertainties. In this regard, sarcomere length was not actually measured during contraction, at least at higher levels of activation, due to disappearance of the diffraction pattern. The inability to consistently measure active sarcomere length reduces confidence in the quantitative relationships between Ca2+ sensitivity or lattice spacing and length, but does not detract from the main finding of the paper, ie, osmotic compression of lattice spacing at short length to achieve the smaller lattice spacing at a long length does not mimic the greater Ca2+ sensitivity observed at the longer length.
Another important issue, studied previously by this group,7 is that skinning of cardiac muscle results in swelling of diameter by about 20%, ie, filament lattice dimensions are greater in skinned than in living preparations. In view of this, the experiments here were likely done at greater than normal filament lattice spacings, although the degree of swelling is not evident from the results, nor is it possible to know whether swelling influenced the results. A puzzling aspect of the results is that compression reduced the length-dependent variations in filament lattice spacing, and yet, living preparations with much smaller lattice dimensions exhibit substantial length dependence of lattice spacing.7 This gives rise to a concern in this and previous studies, voiced by Konhilas et al,6 that the use of high molecular weight polymers might have nonspecific effects on the filament lattice in addition to osmotic compression. To investigate this possibility, new methods are needed for altering filament lattice spacing without osmotically active polymers.
The Mechanism of the Frank-Starling Relationship
The work by Konhilas et al6 has important implications for ultimate understanding of the mechanism(s) underlying the Frank-Starling relationship. They show that osmotic compression that is sufficient to mimic the change in filament lattice spacing observed when resting sarcomere length is increased did not significantly change the Ca2+ sensitivity of force. The conundrum now is that others have been able to mimic the length-dependent changes in Ca2+ sensitivity only by changing fiber diameter,3–5⇓⇓ ie, other than muscle length, lattice spacing is the sole geometrical perturbation that alters Ca2+ sensitivity. Konhilas et al6 suggest a useful way to consider the role of lattice spacing: “… the interfilament spacing theory must be amended to include a variable impact of myofilament lattice spacing on myofilament Ca2+ sensitivity, depending on the overall extent of compression of the myofilament lattice” (page 63).
In view of the present results, other mechanisms must contribute to the length dependence of Ca2+ sensitivity. One possibility is that the number of interacting crossbridges changes with muscle length due to the change in overlap of thick and thin filaments. However, based on the relatively narrow length range in which myocardium operates and the large changes in submaximal forces seen in myocardium over this range, this mechanism would make only minor contributions to length-dependent contraction.1,2⇓ Another idea is that crossbridge binding to actin increases the affinity of Ca2+ binding to the regulatory protein troponin. Results show that this is indeed the case,8 but it is still unclear whether increased Ca2+ binding is mainly a result of crossbridge binding or if increased Ca2+ binding in turn recruits additional crossbridges to the thin filament and increases force.3
Two other mechanisms might contribute to the length dependence of Ca2+ sensitivity in myocardium, with or without length-dependent changes in filament lattice spacing.
Cooperation in Crossbridge Binding
The increase in Ca2+ sensitivity at stretched lengths may involve positive cooperativity in crossbridge binding to actin, ie, initial crossbridge binding facilitates further binding that in turn increases force at any given [Ca2+]. Consistent with this idea, bathing skinned myocardium with a strong-binding derivative of myosin substantially reduces the length-dependent changes in Ca2+ sensitivity of force.9 Presumably, application of the strong binding derivative more nearly saturates the cooperativity of crossbridge binding, so that the activation of force does not vary as much with muscle length. The mechanism of increased cooperativity at long lengths might be due to increased probability of initial crossbridge binding to actin due to reduced lattice spacing or an effect of stretch on crossbridge disposition (below).
Strain of Elastic Proteins
Decreased strain of the elastic protein titin reduces the length dependence of Ca2+ sensitivity of force in myocardium.10–12⇓⇓ Granzier and colleagues11 interpreted their results with a lattice spacing model, ie, a component of titin stress is directed radially rather than axially, so that increased resting force at long lengths would actually pull the thick and thin filaments closer together and increase the likelihood of crossbridge binding to actin (Figure). This is a potentially important idea and is testable with x-ray methods described by Konhilas et al.6 Another possibility is that titin strain alters myosin packing in the thick filament or the orientation of myosin heads along the thick filament backbone. If such changes contribute to increased Ca2+ sensitivity of force at long lengths, the contribution would be eliminated by a reduction in titin strain, which is what has been observed.11,12⇓ These ideas need to be tested quantitatively to address the underlying mechanisms. First, is the effect of titin strain on Ca2+ sensitivity of force observed in skinned myocardium with expanded filament lattice spacing still present at physiological lattice spacing? This question could be studied by assessing effects of titin strain on contraction at compressed lattice dimensions typical of intact myocardium at long and short working lengths. Second, are there changes in cross-bridge disposition at long lengths that might increase the likelihood of crossbridge binding to actin? Assessment of meridional reflections and off-layer layer lines in the x-ray pattern would help resolve this issue.
Konhilas et al6 have provided a much needed quantitative test of the idea that changes in filament lattice spacing account for the increase in Ca2+ sensitivity of myocardial force at long lengths. Their results imply that length-dependent changes in lattice spacing are not the only factor, and possibly not the principal factor, determining the length dependence of Ca2+ sensitivity thought to underlie the Frank-Starling relationship. At the same time, their findings do not suggest a mechanism to explain length-dependent activation in myocardium. Possibilities include effects of stretch on thick filament structure or cross-bridge disposition via the elastic protein titin or length-dependent changes in cooperative processes that modulate activation, but such ideas need to be explored. In this regard, the x-ray methods used by Konhilas et al6 have extraordinary potential for elucidating effects of stretch on the structure of the myofilaments, but such experiments will be even more difficult than those discussed here. The emerging complexity of mechanisms underlying the Frank-Starling relationship recalls the words of A.V. Hill: “There are more things in heaven and earth, Horatio, … and even in … [heart] muscles” (page 22).13
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
- ↵Lakatta EG. Length modulation of muscle performance: Frank-Starling law of the heart.In: Fozzard HA, ed. The Heart and Cardiovascular System. New York, NY: Raven Press Publishers; 1992: 1325–1351.
- ↵Fuchs F, Smith SH. Calcium, cross-bridges, and the Frank-Starling Relationship. News Physiol Sci. 2001; 16: 5–10.
- ↵McDonald KS, Moss RL. Osmotic compression of single cardiac myocytes eliminates the reduction in Ca2+ sensitivity at short sarcomere length. Circ Res. 1995; 77: 199–205.
- ↵Konhilas JP, Irving TC, de Tombe PP. Myofilament calcium sensitivity in skinned rat cardiac trabeculae: role of interfilament spacing. Circ Res. 2002; 90: 59–65.
- ↵Irving TC, Konhilas JP, Perry D, Fischetti R, de Tombe PP. Myofilament lattice spacing as a function of sarcomere length in isolated rat myocardium. Am J Physiol. 2000; 279: H2568–H2573.
- ↵Fitzsimons DP, Moss RL. Strong binding of myosin modulates length-dependent Ca2+ activation of rat ventricular myocytes. Circ Res. 1998; 83: 602–607.
- ↵Cazorla O, Wu Y, Irving TC, Granzier H. Titin-based modulation of calcium sensitivity of active tension in mouse skinned cardiac myocytes. Circ Res. 2001; 88: 1028–1035.
- ↵Fukuda N, Sasaki D, Ishiwata S, Kurihara S. Length dependence of tension generation in rat skinned cardiac muscle. Circulation. 2001; 104: 1639–1645.
- ↵Hill AV. First and Last Experiments in Muscle Mechanics Cambridge University Press; 1970.