|Home | About | Journals | Submit | Contact Us | Français|
Contractility in the borderzone (BZ) after antero-apical myocardial infarction (MI) is depressed. We tested the hypothesis that BZ contractility is also decreased after postero-lateral MI.
Five sheep underwent postero-lateral MI. Magnetic resonance imaging (MRI) was performed 2 weeks before and 16 weeks after MI, and left ventricular (LV) volume and regional strain were measured. Finite element (FE) models were constructed and the systolic material parameter, Tmax, was calculated in the BZ and remote myocardium by minimizing the difference between experimentally measured and calculated LV strain and volume. Sheep were sacrificed 17 weeks after MI and myocardial muscle fibers were taken from the BZ and remote myocardium. Fibers were chemically demembranated, and isometric developed force, Fmax, was measured at supra-maximal [Ca2+]. Routine light microscopy was also performed.
There was no difference in Tmax between pre-MI and remote myocardium 16 weeks after MI. However, there was a large decrease (63.3%, p=0.005) in Tmax in the BZ when compared with the remote myocardium 16 weeks after MI. In addition, there was a significant reduction of BZ Fmax for all samples (18.9%, p=0.023). Myocyte cross-sectional area increased by 61% (p=0.021) in the BZ, but there was no increase in fibrosis.
Contractility in the borderzone is significantly depressed relative to the remote myocardium after postero-lateral MI. The reduction in contractility is due at least in part to a decrease in contractile protein function.
It has been known since the mid-1980s that systolic performance (systolic shortening and wall thickening) is depressed in the non-ischemic borderzone (BZ) after antero-apical myocardial infarction (MI).  It was initially thought that the reduced BZ function was due to mechanical tethering by the infarct. However, finite element (FE) -based inverse calculations of regional contractility, in which optimization routines attempt to match computed and experimentally measured LV volume and strain, suggest that BZ contractility is reduced by more than 50%.  Moreover, BZ contractility appears to vary linearly across the BZ and the BZ may be as much as 3 cm in width after antero-apical MI. 
Myocardial infarction involving the postero-lateral or inferior LV wall behaves differently than MI of the antero-apical LV wall. Postero-lateral MI is less likely to undergo expansion,  and survival after postero-lateral MI is increased.  In addition, chronic ischemic mitral regurgitation (CIMR) is caused by LV remodeling after postero-lateral MI in which the posterior papillary muscle moves laterally. [6, 7] It is therefore possible that the reduction in BZ contractility after postero-lateral MI is different than after antero-apical MI and that LV remodeling and development of ischemic mitral regurgitation may be affected.
The purpose of this study was twofold. First, we determined contractility (systolic material parameter, Tmax) in the BZ and remote myocardium in sheep after postero-lateral MI using magnetic resonance imaging (MRI) -based inverse finite element methods. Second, maximum isometric developed force was measured in vitro in myocardial samples from the BZ and remote myocardium. We tested the hypothesis that there is a significant depression in contractility in the infarct BZ relative to remote myocardium after postero-lateral MI.
The sheep used in this study were treated in compliance with the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resources, National Research Council, and published by the National Academy Press (revised 1996).
Five male adult Dorset sheep underwent postero-lateral MI as previously described. 
Two weeks before and 16 weeks after MI, cardiac MRI with non-invasive tags was performed as previously described.  Endocardial and epicardial surfaces of the LV and right ventricles (RV) were contoured (iContours, Liang Ge, Cardiac Biomechanics Lab, San Francisco, CA) as well as the stroke volumes of the LV and RV as previously described. 
Calculation of systolic strain has been previously described.  Briefly, tag lines were segmented (FindTags, Laboratory of Cardiac Energetics, National Institutes of Health, Bethesda, MD) in each MRI image slice (Figure 1)  and systolic myocardial strains were calculated at mid-wall and around the circumference in each short-axis slice from tag-line deformation using a four dimensional, B-spline-based motion tracking technique (TTT; Laboratory of Cardiac Energetics, Bethesda, MD). 
Finite element models of the LV were created as previously described.  Briefly, finite element models were created using early diastole as the unloaded reference configuration (Figure 2). The infarct region was determined from the post-mortem digital photograph. The BZ extended 1 cm in width from the infarct edge. Cardiac myofiber angles and boundary conditions were assigned as previously described. 
Nearly incompressible, transversely isotropic, hyperelastic constitutive laws for passive  and active myocardium  were modeled in a user-defined material subroutine an explicit FE solver (LS-DYNA, Livermore Software Technology Corporation, Livermore, CA). The diastolic constitutive laws and the diastolic material parameters used have been previously described. 
Kim and colleagues found Ca2+ transient amplitude (in Fura-2 ratio units) in the BZ and remote myocardium to be 1.08 and 0.69 respectively 8 weeks after antero-apical MI in sheep.  Along those lines, we assigned Ca0 = 1 µmol/L for remote zone and Ca0 = 0.65 µmol/L for the BZ. Other material constants for active contraction were assumed to be: t = 25 ms (end systole), (Ca0)max = 4.35 µmol/L, B = 4.75 µm−1, l0 = 1.58 µm, m = 1.0489 sec µm−1, b = −1.429 sec, and lR was set at 1.85 µm. 
The systolic material parameter, Tmax, was calculated in the BZ and remote myocardium by minimizing the difference between experimentally measured and calculated LV strain and volume. The details of the optimization routine used in this study have been previously described. 
After optimization was complete, the Ct parameter in Sun et al, Equation 6, was factored out by setting Tmax = Tmax Ct.  This was done in order to match the form of the modified Hill equation and allow more direct comparison with skinned fiber developed force, Fmax.
Sheep were sacrificed one week after the 16 week MRI. The heart was excised. Cardiac arrest was rapidly achieved by retrograde ascending aortic infusion of a cold hyperkalemic solution (Plegisol, Hospira Inc, Lake Forest, IL; 1 L plus 10 mL of 8.4% NaHCO3 at 4 °C to achieve pH 7.8 at room temperature). The right ventricle was removed. The LV was opened by incising the septum and the anterior leaflet of the mitral valve. A digital photograph was taken of the LV endocardium. A transmural strip of myocardium (approximately 2 × 6 cm) extending circumferentially from the infarct to the posterior wall was excised from the LV. BZ and remote myocardium regions were defined as 0–1 cm and 3–4 cm from the infarct edge, respectively.
Epicardial muscle fibers from BZ and remote myocardium were pinned on silicone substrate at approximate physiological sarcomere length and chemically skinned in a relaxing solution with 1% Triton® X-100 (Sigma Aldrich, St. Louis, MO) for 24 h at 4 °C.  Following chemical treatment, samples were washed in relaxing solution for 2 hours at 4 °C and then stored in a 1:1 solution of relaxing solution plus glycerol at −20 °C.
Measurement of developed force in skinned myocardial fibers was performed as previously described.  Measurements were performed with a permeabilized fiber test system (Model 1400A, Aurora Scientific, Inc., Ontario, Canada). Early samples (n=3) were measured at a passive tension of 0.98 mN/mm2 and sarcomere length was not measured. In later experiments (n=2) sarcomere length was measured by laser diffraction and measurements were performed at a sarcomere length of 2.1 µm. Samples were initially bathed in relaxing and pre-activating solutions before immersion in an activating solution with supra-maximal [Ca2+]. Fmax was normalized to muscle cross-sectional area (mN/mm2). 
Freshly isolated samples from the BZ and remote myocardium were fixed in phosphate-buffered 4% paraformaldehyde (Fisher Scientific, Inc., Fair Lawn, NJ) at 4 °C for at least 24 h. Paraffin-embedded thick sections were stained with hematoxylin and eosin (H&E) and picrosirius red as previously described.  Images of stained cross-sections were analyzed using ImageJ (NIH, Bethesda, MD). Segmentation of myocytes was performed on cells containing only visible nuclei to ensure the representative cross-sectional area along the cell length. Collagen content in picrosirius red-stained sections was quantified using ImageJ thresholding analysis for collagen-positive areas.
All values are expressed as mean ± standard deviation (SD). Multivariable mixed model regression was performed (SAS system for Windows Version 9.1, SAS Institute, Cary, NC). Individual comparisons of Tmax and Fmax were performed with paired t-tests.
LV pressure and volume before and 16 weeks after postero-lateral MI is shown in Table 1. Although LV pressure at ED and ES are unchanged, LV volumes at both ED and ES are increased (24.4%, p=0.1522 and 50.2%, p=0.0788 respectively). Although LV EF is decreased by 15.6% (p=0.0954), total LV stroke volume is unchanged. However, because there is moderate ischemic mitral regurgitation, forward or RV stroke volume is decreased by 28.4% (p=0.033).
The optimized systolic material parameter, Tmax, for pre- and post-MI sheep by region is seen in Figure 4. There was no difference in Tmax between pre-MI and remote myocardium 16 weeks after MI. However, there was a large decrease (63.3%, n=5, p=0.005) in Tmax in the BZ when compared with the remote myocardium 16 weeks after MI.
Regional Fmax after MI is seen in Table 3. Fmax obtained from samples tested at fixed passive tension showed a reduction in the BZ of 19.9% (n=3, p=0.029) when compared with remote myocardium. Fmax obtained from samples tested at fixed sarcomere length showed the reduction in the BZ to be 17.0% (n=2, p=0.086) when compared with remote myocardium. There was a significant reduction of BZ Fmax for all samples (18.9%, n=5, p=0.023). Tmax and Fmax normalized to remote myocardium are seen in Figure 5.
Analysis of representative cross-sectional myocardial slices reveals significant myocyte hypertrophy in the BZ, with the mean BZ myocyte cross-sectional area (CSA) of 190.8 ± 21.2 µm (n=2) compared with remote myocardium of 118.5 ± 2.9 µm (n=2), corresponding to a 61% increase in BZ myocyte CSA (p=0.021). Picrosirius red staining of histological sections shows no replacement fibrosis that could account for the reduction in myocardial force (Figure 6). Image analysis shows no significant change in collagen content, with collagen as a percent of total cross-sectional area to be 5.1 ± 0.3% in the BZ and 5.0 ± 0.4% in the remote myocardium (n=2, p=ns).
The principal finding of this study is that contractility in the borderzone (BZ) is substantially depressed when compared with the remote myocardium after postero-lateral MI in the sheep. Developed force measured in isolated skinned fibers is also substantially depressed in the infarct BZ, suggesting that contractile protein dysfunction is at least partially responsible for the observed reduction in BZ contractility.
Recently, we developed and validated a computationally efficient method of material parameter optimization.  Those methods hold peak intracellular calcium, calcium sensitivity, and sarcomere length constant and only allow the peak force generation capability of myocytes, Tmax, to vary. Such an arrangement necessarily redistributes the effect of other factors that might affect contractility into Tmax. For instance, Ca2+ transient amplitude and calcium handling protein levels,  as well as intracellular phosphocreatine-to-ATP ratio,  are reduced in the BZ after MI and may affect contractile ability. As a consequence, FE-based calculation of Tmax is not directly comparable to the experimentally measured Fmax. However, the fact that both the finite element predicted and experimentally measured results showed a depression of myocyte force generation capability is very reassuring.
Our finite element models assume that the BZ region is 1 cm in width. However, finite element models  and preliminary experimental data from sheep after antero-apical MI (data not shown) suggest that the BZ may be as wide as 3 cm and that there is a linear Fmax gradient between the infarct and the base of the heart. These findings suggest that the region of contractile force reduction is larger than previously thought. It should be noted that a large BZ with reduced contractile function is an attractive therapeutic target and that a recent finite element study of enhanced BZ function suggests that even partial improvement of BZ contractility is sufficient to improve pump function. 
Our study did not directly compare BZ contractility after poster-lateral MI with BZ contractility after antero-apical MI. However, the 63.3% reduction in BZ Tmax seen in the current study is similar to the 49.4% reduction in BZ Tmax after antero-apical MI in our previous study by Sun et al.  Further work is needed but it appears that post-MI BZ dysfunction is independent of the MI location.
The finding that Fmax in the BZ was decreased by nearly 20% suggests that contractile protein dysfunction is at least in part responsible for the observed reduction in BZ contractility. The skinned fiber preparation is a robust experimental method that has been used to investigate the underlying mechanisms of contractile dysfunction in skeletal and cardiac muscle. [25, 26] Skinned fiber measurements are important since they specifically exclude the effects of myocyte calcium handling machinery, permitting bath ion concentrations to be tightly controlled and allowing calcium to directly activate cross-bridge cycling. Skinned fiber measurements provide experimental evidence of dysfunction at the contractile protein level.
While the tests performed at fixed passive tension and fixed sarcomere length differ in magnitude, the proportion of BZ force to remote myocardium remains consistent at approximately 80%. This observation is further demonstrated in experiments (data not shown) where Fmax was measured for BZ and remote myocardium at decreasing sarcomere lengths (2.1, 2.0, 1.9, 1.8 and 1.7 µm), maintaining a BZ-to-remote myocardium ratio of 84.9 ± 2.4%. These results suggest the proportional reduction of BZ force is similar and independent of sarcomere length.
The mechanism of reduced contractile protein function is unclear but is likely mediated by high stress and strain in the BZ myocardium. We know that strain is altered in the BZ after postero-lateral MI. For instance, Rodriguez and colleagues found that longitudinal strain was decreased  and radial-circumferential shear strain was increased in the functional BZ early after circumflex coronary artery occlusion in the sheep. [27, 28] High systolic stress and positive strain in the BZ activates MMPs 2 and 9 [29, 30] and MMP-2 has been shown to directly damage the intracellular contractile proteins myosin light chain 1 and troponin I. [31, 32] Cleavage of such contractile proteins can affect cross-bridge dynamics and should result in reduced myocardial contractility.
There are several limitations in the present study. As above, the accurate measurement of intracellular calcium and ECa50 values is clearly needed. Second, experiments using chemically skinned fiber experiments performed at a fixed passive tension likely had a sarcomere length less than 2.1 µm. Despite demonstrating the preserved BZ-to-remote myocardium ratio for all samples, future experiments should be conducted at an initial diastolic sarcomere length of 2.1 µm.
Finally, comparison between FE based calculations and experimental measurements would be facilitated by using an active contraction material property law that more closely conforms to underlying biophysical mechanisms. The active contraction law described by Guccione and colleagues, that was used in the current study, has proved invaluable in the analysis of left ventricular mechanics after MI  and after application of devices designed to correct heart failure. [33, 34] However, the largely phenomenologic nature of that material property law, that was made necessary by the limited computational power available in 1995, is an increasing limitation. Recently, Rice and colleagues have developed a computationally efficient, active contraction material property law that directly models myocardial action potentials, intracellular calcium transients, and contractile protein cross bridge formation.  The ready availability of cheap parallel processing makes the development of a multi-scale model, in which a material law of the Rice type is coupled with optimization of systolic material parameters, both possible and highly desirable.
Contractility in the borderzone (BZ) is substantially depressed compared with remote contractility in the remote zone after postero-lateral MI in the sheep. Developed force in isolated skinned myofibers is also depressed in the infarct BZ suggesting that decreased contractile protein function is at least in part responsible for the observed reduction in BZ contractility.
Treatment aimed at the prevention or reduction in loss of BZ contractility may significantly improve ventricular pump function. As above, recent finite element based simulations suggest that even a moderate increase in BZ contractility will have a beneficial effect. 
This study was supported by NIH grants R01-HL-84431 (Dr. Ratcliffe), R01-HL-63348 (Dr. Ratcliffe), R01-HL-77921 (Dr. Guccione), and R01-HL-86400 (Dr. Guccione). This support is gratefully acknowledged.