INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CUTANEOUS BURN TRAUMA causes a decrease in cardiac output that does not necessarily parallel the loss of circulating plasma volume. Also, the volume resuscitation that is adequate to replace or exceed intravascular fluid loss does not always restore left ventricular (LV) stroke volume (1, 5). The precise mechanisms responsible for the cardiocirculatory deficits after burn trauma remain poorly understood and controversial.

In a study by Xia et al. (22), the use of ³¹P and ²³Na magnetic resonance spectroscopy (MRS) in isolated perfused hearts confirmed that burn trauma produced cardiac contractile deficits that were intrinsic to the myocardium and independent of burn-induced alterations in preload, afterload, and neurohumoral function. Cardiac dysfunction after burn trauma was paralleled by intracellular Na⁺ accumulation but was not related to either intracellular acidosis or myocardial energy deficits. However, cellular concentrations of Na⁺ and cytosolic free calcium ([Ca²⁺]_i) are metabolically related, and [Ca²⁺]_i is a primary link in excitation-contraction coupling of the heart (16). These data suggest that postburn changes in Na⁺ reported by our laboratory may be paralleled by altered myocyte handling of calcium, contributing to postburn contraction and relaxation deficits.

Whereas burn-mediated changes in [Ca²⁺]_i levels have been studied (6, 11) in adult rat cardiac myocytes by using fura 2 loading and fluorospectrophotometery, we deemed it important to ascertain whether postburn cardiomyocyte calcium levels were altered by the myocyte isolation procedure. Measurements of [Ca²⁺]_i in the isolating contracting heart would likely provide a better assessment of intracellular-extracellular calcium shifts that occur after burn trauma in vivo and would allow simultaneous assessment of cardiac contractile function. Therefore, the aim of the present study was to examine burn-mediated changes in [Ca²⁺]_i levels in the intact beating heart and to correlate intracellular calcium availability with changes in cardiac contractile performance.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental animals and burn procedure. The experimental protocol was reviewed and approved by the Institutional Animal Care and Research Advisory Committee of the University of Texas Southwestern Medical Center at Dallas.

Isolated coronary perfused hearts. Approximately 24 h after burn or sham burn, rats were anesthetized with methoxyflurane and given 100 units of heparin sodium intravenously. The heart was excised through a midline thoracotomy and placed in ice-cold normal saline. A cannula placed in the aorta was connected to a reservoir located 95 cm above the heart for perfusion of the coronary arteries by the Langendorff method. The perfusion solution was a modified phosphate-free Krebs-Henseleit bicarbonate buffer composed of (in mM) 143 Na⁺, 5 K⁺, 126 Cl, 1.2 Ca²⁺, 1.2 Mg²⁺, 25 HCO, 1.2 SO, and 10 glucose, pH 7.4, at 37°C, and bubbled continuously with 95% O₂-5% CO₂. The pulmonary artery was incised to allow drainage of blood or perfusate from the right heart. The perfusate passed through a water-jacketed reservoir maintained at 37°C before reaching the heart. An intraventricular balloon attached to a catheter (0.8 mm outer diameter; OD) was placed in the left ventricle (via left atrium) for monitoring spontaneous heart rate (HR) and peak systolic LV pressure (LVP); LV developed pressure (LVDP) was calculated from LVP and LV end-diastolic pressure. The balloon cannula was connected to a Gould P23 pressure transducer, and functional parameters were measured by using a physiological monitor (Coulbourn; Lehigh Valley, PA) and recorded on a chart recorder (model WR 31001, Western Graphtec; Irvine, CA) (22). The heart was suspended in the center of a saline-filled MR tube (20 mm OD) as previously described by London and colleagues (4, 13). However, this cardiac perfusion technique was modified to increase sensitivity and homogeneity of the resonance spectra. In this regard, the perfused heart was encapsulated within a thin latex bag, and 1,2-bis(2-amino-5,6-difluorophenoxy)-ethane-N,N,N',N'-tetraacetic acid (TF-BAPTA, Molecular Probes; Eugene, OR) perfusate, which had circulated through the coronary circulation, was continuously aspirated through a drain line placed inside the latex bag, producing close approximation of the collapsed latex membrane around the myocardium. Continuous evacuation of the coronary effluent away from the perfused heart eliminated the contaminating effects of TF-BAPTA-containing perfusate, which would normally accumulate around the heart and would confound interpretation of cardiac spectra. Thus the heart was perfused via the coronary arteries with TF-BAPTA-containing perfusate but suspended within a saline-filled NMR tube (Fig. 1). After 10 min of nonrecirculating perfusion via the coronary arteries, the perfusion was changed to a recirculating mode.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Suspension of the perfused heart in a saline-filled nuclear magnetic resonance (NMR) tube. Enclosing the heart within a latex membrane allowed aspiration of cardiac perfusate after a single passage through the coronary circulation.

Calcium indicator, TF-BAPTA. After 15-20 min of control perfusion, 300 ml of 5 µM acetoxymethyl ester (AM) of TF-BAPTA (Molecular Probes) were loaded for 30 min. This agent was developed as a MRS-sensitive fluorinated intracellular calcium ion indicator characterized by high dissociation constant (K_d) to reduce perturbations due to buffering of transients (4). It exhibits two fluorine nuclear magnetic resonances (NMR); one is insensitive to Ca²⁺ chelation and the second shifts by H10 parts per million (ppm) on Ca²⁺ binding.

Magnetic resonance spectroscopy. The heart and the entire perfusion apparatus were lowered into a wide-bore magnet interfaced to a Varian Inova 300 spectrometer, and the field homogeneity was adjusted on the ²³Na-free induction decay. ³¹P spectra were acquired over a ±6,000 Hz spectral width by using 4-K data points with a 33-µs excitation pulse, 200 acquisitions, and a 1.336-s interpulse delay. ¹⁹F spectra were signal averaged over 10,000 scans, with spectral width = 18,001 Hz, acquisition time = 0.203 s, delay = 0.001 s.

MRS data analysis. The resonance areas in ³¹P MRS were determined using a curve analysis program supplied with the Varian software. Phosphocreatine (PCr), ATP, and P_i were assumed proportional to resonance areas. Intracellular pH (pH_i) was determined from the chemical shift difference between P_i and PCr resonances (22). Intracellular magnesium was determined from the chemical shift difference between -ATP resonances (14). [Ca²⁺]_i was determined from the chemical shift difference between Ca²⁺-insensitive fluorine resonance (6F) and Ca²⁺-sensitive fluorine resonance (5F) in the ¹⁹F MRS spectra, using Eq. 1 proposed by Murphy and London (14)

(1)

where K_d is 65 µM, ₀ is the measured shift difference between 6F and 5F of TF-BAPTA, and pH and magnesium (pMg) are the measured pH_i and pMg, respectively. ₁ is the shift difference in the presence of EGTA at pH 12, where there is no protonation (4.88 ppm). ₂ + ₃ and ₄ are the shifts caused by protonation and correspond to 13.16 and 8.28, respectively; the shift caused by protonation of site x (pK_x) = 5.2, the protonation of site y when site x on a model with two protonation sites is already protonated (pK') = 4.9, pK_x = 5.2, pK' = 4.9, and ₅ is the shift difference with excess Ca²⁺. ₆ + ₇ (10.96) and ₈ (7.58) correspond to the shift due to binding of a single Mg²⁺ on the nearest fluorine resonance. Figure 2 provides a representative ¹⁹F NMR spectra from a series of samples containing 4.8 mM TF-BAPTA but varying in total [Ca²⁺] from 0 to 100 mM. The peak at 0 ppm is due to the fluorine nuclei at the 6- and 6'- positions that are insensitive to Ca²⁺ binding, whereas the other peak corresponds to the fluorine nuclei at the 5- and 5'- position shift after TF-BAPTA Ca²⁺ binding.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Representative 1,2-bis(2-amino-5,6-difluorophenoxy)-ethane-N,N,N',N'-tetraacetic acid (TF-BAPTA) 19F NMR spectra from a control heart as perfusate Ca2+ was increased from 0 to 100 nM. ppm, Parts per million.

Statistics. Values were expressed as means ± SE. Differences between burn and sham burn groups were evaluated by an unpaired Student's t-test. Differences were considered to be statistically significant when P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mechanical performance of perfused hearts. The hearts from the rats with cutaneous burn trauma generated significantly lower LVDP (65 ± 10 mmHg) during equilibration than LVDP measured in hearts from the sham burn group (110 ± 15 mmHg, P < 0.01). In addition, LV end-diastolic pressure was higher in burns (13 ± 0.6 mmHg) than that measured in sham burn (9.3 ± 0.6 mmHg, P < 0.05). HR were similar in sham burn (244 ± 54 beats/min) and in burn groups (240 ± 62 beats/min, P > 0.05). Loading with TF-BAPTA caused an initial but transient increase in LVDP in all hearts; however, LVDP returned to 88-90% of control values within 3-5 min after completing TF-BAPTA (Fig. 3). There was no change in LV end-diastolic pressure or HR with TF-BAPTA loading. Coronary perfusion pressure was set at 95 cmH₂O for all hearts and adjusted to maintain coronary flow rate at 7-8 ml/min; TF-BAPTA had no significant effect on coronary perfusion pressure.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Left ventricular developed pressure (LVDP) was calculated in all hearts, both sham and burn, as the difference between systolic and diastolic pressure. TF-BAPTA produced a modest effect on LVDP. Top, representative tracing from a sham burn rat. Bottom, tracing recorded 24 h after burn trauma. LVP, left ventricular pressure.

[Ca²+]_i levels of perfused beating hearts. The average shift difference between 5 and 6F of TF-BAPTA (₀) from this study was 5.04 ± 0.03 (n = 7) for sham burn group and 5.30 ± 0.05 (n = 7) for burn group (Fig. 4, top). Comparison of [Ca²⁺]_i in hearts from burn or sham burn groups is shown in Fig. 4, bottom. With pH_i of 6.97 and pMg of 2.81, the time-averaged myocardial [Ca²⁺]_i in sham burn group was 807 ± 192 nM, a value similar to time-averaged [Ca²⁺]_i reported by Steenbergen and London (17) in perfused beating hearts (630 nM) by using 5F-BAPTA as a Ca²⁺ indicator. Cutaneous burn trauma significantly increased myocardial [Ca²⁺]_i to 3.891 ± 0.929 µM (P < 0.01), supporting our previous observation of postburn Ca²⁺ accumulation in isolated myocytes (6, 11). There was significant negative correlation between cardiomyocyte Ca²⁺ accumulation and decreased LVDP (correlation coefficient = 0.88).

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Top: representative tracing showing the shift between 5- and 6-fluorines (F5 and F6, respectively) of TF-BAPTA. Bottom: comparison of cytosolic free calcium concentration ([Ca2+]i) in sham burn and burn groups. All values are means ± SE; *P < 0.01.

Myocardial energy metabolism. The PCr-to-ATP (PCr/ATP) and P_i-to-ATP (P_i/ATP) ratios measured in the hearts from burned rats were not significantly different from the sham burn values (P > 0.05). ATP and PCr measured in hearts from sham burns were 17.8 ± 0.6 and 26.6 ± 0.7 µmol/g dry wt, respectively; ATP (17.7 ± 0.5 µmol/g dry wt) and PCr (25.7 ± 0.8 µmol/g dry wt) measured in hearts from burns were not significantly different from values measured in shams. However, PCr/P_i measured in hearts from burn rats decreased significantly compared with that measured in hearts from sham burn group (3.33 ± 0.42 vs. 3.85 ± 0.35, P < 0.05). Myocardial pH_i was similar in sham burn and burn groups 24 h after burn trauma (6.97 ± 0.05 vs. 7.02 ± 0.04; P > 0.05). Loading the perfused rat heart with 300 ml of 5 µM TF-BAPTA for 30 min increased P_i resonance but decreased PCr resonance in both burn and sham burn groups (Fig. 5). Whereas TF-BAPTA loading produced minor alterations in the ³¹P MRS spectra, these changes did not mask burn-related changes in the PCr-to-P_i ratio. In addition, ATP resonances in the ³¹P NMRS spectra were not altered by TF-BAPTA loading, as shown in Fig. 5.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Representative 31P magnetic resonance spectroscopy (MRS) spectra of hearts from burn or sham burn groups before and after TF-BAPTA loading.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major finding of this study is that cutaneous burn trauma alters Ca²⁺ homeostasis in heart; the cytosolic Ca²⁺ accumulation paralleled changes in cardiac mechanical function and likely contributed to the postburn myocardial contractile dysfunction. Moreover, this is the first study, to our knowledge, to confirm burn-mediated [Ca²⁺]_i overload in intact beating hearts.

For TF-BAPTA loading and [Ca²⁺]_i measurements, we used a technique introduced by Murphy and co-workers (14). However, our modification included enclosing the heart in a thin latex bag. This modification was related to the fact that TF-BAPTA may leak from the myocytes, accumulating in the solution that bathed the heart and complicating assessment of the MRS spectra. Previous approaches to this problem have included aspirating circulated perfusate from the MR tube, resulting in the heart suspended in an air-filled MR tube. In our modification, the heart was enclosed within a latex bag and perfusate, which had circulated through the coronary arteries and was continuously aspirated via the drainage tube placed within the latex bag. This measure collapsed the latex bag and approximated the latex membrane against the myocardial wall but did not compress the ventricular tissue or restrict ventricular contraction. In this manner, circulated TF-BAPTA-containing perfusate was removed from the bag, and interference of TF-BAPTA that may have leaked from the myocytes into the perfusate was eliminated. The suspension of the heart within a saline-filled MR tube increased the sensitivity as well as the homogeneity of the MR measurements.

Since 1980, ¹⁹F MRS has been described for the measurement of intracellular free Ca²⁺ in a variety of organs (12, 14, 17). This method allows direct evaluation of [Ca²⁺]_i in perfused organs or intact animals, and therefore functional and metabolic status can be determined simultaneously. A review of previously used indicators confirmed that the indicator with the higher K_d gives a higher [Ca²⁺]_i value: DiMe-5F BAPTA (K_d = 46 nM), 89 ± 13 nM described by Kirschenlohr and colleagues (12), 5F-BAPTA (K_d = 710 nM), 544 ± 74 nM described by Steenbergen and co-workers (13), and TF-BAPTA (K_d = 65 µM ), 807 ± 192 nM described in this study (Fig. 4). The most commonly used ¹⁹F MRS Ca²⁺ indicator, 5F-BAPTA, is sensitive for both systolic and diastolic [Ca²⁺]_i measurement. However, its low K_d has several limitations: 1) substantial Ca²⁺ buffering occurs at useful indicator concentrations; 2) 5F-BAPTA decreases LVDP by 75-78% (Z.-F. Xia and J. Horton, unpublished data); and 3) saturation likely occurs when measuring Ca²⁺ levels above 2 µM (4, 13). TF-BAPTA has K_d of 65 µM, a value 100-fold greater than that of 5F-BAPTA. With TF-BAPTA loading, LVDP was maintained at 90% of baseline values in perfused beating hearts, eliminating the cardiodepression that has been previously associated with 5F-BAPTA. Whereas measurements of basal intracellular Ca²⁺ may be subject to greater errors with TF-BAPTA compared with 5F-BAPTA, TF-BAPTA provides a more accurate and sensitive assessment of changes in [Ca²⁺]_i that occur as a result of pathological conditions. In addition, the absence of TF-BAPTA-mediated contractile depression allows adequate assessment of the relationship between changes in [Ca²⁺]_i and alterations in cardiac contractility under a variety of experimental conditions.

Various techniques and models have described a wide range of [Ca²⁺]_i in several organs. In studies of isolated cardiomyocytes, assessment of [Ca²⁺]_i with fluorescent indicators has confirmed cellular Ca²⁺ levels in the range of 100-200 nM. In the perfused intact heart, myocardial [Ca²⁺]_i levels vary with the cardiac cycle. Gated MRS data show a diastolic [Ca²⁺]_i of 177 nM and a systolic [Ca²⁺]_i of nearly 1 µM (12). At a pacing frequency of 1.0 Hz, end-diastolic [Ca²⁺]_i was 198 ± 30 nM, and reducing the pacing frequency to 0.2 Hz lowered [Ca²⁺]_i to 89 ± 13 nM (10). In our study, there were no significant differences in HR in spontaneously beating hearts from sham and burned animals. Therefore, the increased [Ca²⁺]_i measured in hearts from burned animals could not be attributed to differences in HR.

Previous studies have shown that functional conditions also affect [Ca²⁺]_i. Steenbergen and colleagues (17, 18) showed that time-averaged [Ca²⁺]_i in beating hearts was 544 ± 74 nM, whereas time-averaged values were significantly lower in potassium-arrested (352 ± 88 nM) and magnesium-arrested hearts (143 ± 22 nM). Thus from these previous studies it appears that a lower [Ca²⁺]_i correlates with decreased cardiac contractility. In contrast, cardiac contractile dysfunction after burn trauma in our study correlated with significant elevations in [Ca²⁺]_i, suggesting that burn injury alters several aspects of excitation-contraction coupling in the hearts. It should be noted that the intracellular Ca²⁺ levels reported in this study represent an average level between systolic and diastolic [Ca²⁺]_i. In addition, the Ca²⁺ measurements reflect total heart calcium, which include cardiac myocyte calcium plus Ca²⁺ from coronary endothelial cells as well as from other nonmyocyte cell populations. However, because cardiac myocytes constitute the largest cell number within the myocardium, this cell population contributes most significantly to the measured Ca²⁺ signal. Given the contribution of cardiac myocytes to the total calcium levels measured, this NMR technique affords an opportunity to correlate changes in systolic and diastolic function with overall changes in cellular Ca²⁺ levels as well as with pH_i and high-energy phosphate stores.

It is widely accepted that significant intracellular acidosis occurs in models of myocardial ischemia. In our study, the absence of changes in cardiomyocyte pH_i likely excludes coronary hypoperfusion and myocardial ischemia as a primary mechanism underlying postburn cardiac contractile dysfunction and Ca²⁺ accumulation. Previous work (17) from our laboratory has confirmed that burn trauma disrupts several aspects of sarcolemmal function, increasing calcium flux through calcium slow channels, and diminishing myocardial sarcoplasmic reticulum Ca²⁺ efflux channel activity. The attenuation of Ca²⁺ resequestration by this intracellular storage organelle likely contributes to postburn intracellular Ca²⁺ overload (15).

An alternative mechanism by which burn trauma alters cardiomyocyte Ca²⁺ homeostasis may be increased sodium-calcium exchange. Burn trauma promotes significant sodium accumulation by cardiac myocytes (22), and this sodium influx was similar to that described by other studies (3, 19) after trauma. It is likely that the postburn rise in Na⁺ concentration enhances sodium-calcium exchange, contributing to cardiomyocyte Ca²⁺ accumulation.

Whereas we have confirmed a burn-mediated rise in cardiomyocyte sodium and Ca²⁺ levels, the precise mechanism by which a cutaneous burn disrupts cardiomyocyte cation homeostasis remains unclear. Recent studies suggest that inflammatory cytokines, such as tumor necrosis factor (TNF)- and interleukin-1 (IL-1), may play a significant role. For example, we showed (8, 9, 20) that burn trauma triggers cardiomyocyte secretion of TNF- and IL-1, producing myocardial levels of inflammatory cytokines that exceed those measured in the systemic circulation. Furthermore, we showed that anti-TNF- strategies given after burn trauma ameliorate cardiomyocyte secretion of TNF- (2, 20, 23), improve all indexes of cardiomyocyte function (2), and restore Ca²⁺ homeostasis (J. Horton, unpublished data). Furthermore, addition of TNF- to naïve cardiomyocytes has been shown (6) to increase diastolic and systolic intracellular Ca²⁺ levels as measured with fura 2-AM loading of cardiac myocytes. It is likely that the local synthesis of inflammatory cytokines, such as TNF- and IL-1, contribute to disruption of intracellular Ca²⁺ homeostasis, which, in turn, may contribute to cardiac contraction and relaxation deficits. Alternatively, the local synthesis of TNF- and IL-1 may directly mediate cardiac injury and contractile dysfunction, which, in turn, promote intracellular Ca²⁺ accumulation.

While additional studies are required to determine the time course of cardiac contractile depression, local cytokine synthesis, and cardiac sodium-calcium handling after burn trauma, the present study does provide unequivocal evidence that TF-BAPTA and NMR spectroscopy allow simultaneous assessment of intracellular Ca²⁺ levels and contractile performance in the beating heart. In addition, the ability to examine cardiac contractile deficits and with simultaneous measures of cellular sodium-calcium levels in the ex vivo heart model should allow study of cellular signaling mechanisms that occur in pathological conditions, such as trauma, ischemia-reperfusion, and cardiac disease.