INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

-ADRENERGIC RECEPTORS (-ARs) play an important role in the mediation of adrenergic control of cardiac muscle contraction. Any alterations in the dynamics of these receptors would affect myocardial function as a pump. Studies on the regulation of adrenergic receptors have indicated that -ARs are in a dynamic life cycle involving receptor appearance on, and disappearance from, the cell surface (7, 14, 15). Under physiological conditions, cardiac -ARs are distributed in the sarcolemmal membranes and the light vesicles, a distinct cytosolic compartment that is deficient in G_s protein (14, 17). Under certain pathological conditions, however, the distribution of -ARs in the heart can be altered either by internalization from surface membranes to light vesicles or by externalization from intracellular sites to sarcolemmal membranes (4, 9, 14, 15, 17, 26, 30). In septic shock, we have reported earlier that -ARs in the rat heart undergo an externalization from light vesicles to sarcolemma during the early phase of sepsis, whereas they are internalized from surface membranes to intracellular sites during the late phase of sepsis (28). Because myocardial contractility is regulated by catecholamines through -AR mediation, an externalization (overexpression) of -ARs during the initial phase of sepsis is most likely to contribute to the development of the hypercardiodynamic state, whereas an internalization (underexpression) of -ARs during the late phase of sepsis would lead to the formation of hypocardiodynamic state (28). Because phosphorylation/dephosphorylation of receptors has been considered to represent a common mechanism through which recycling of receptors can be regulated (7, 12, 14, 24, 25), the present study dealing with the role of receptor phosphorylation on the biphasic redistribution of -ARs in the rat heart during different phases of sepsis was undertaken in an attempt to uncover the pathogenesis of sepsis-induced alterations in cardiac function.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animal model. Male Sprague-Dawley rats weighing from 270 to 320 g were used. All animals were fasted overnight with free access to water. They were divided into three groups: control, early sepsis, and late sepsis. Sepsis was induced by cecal ligation and puncture (CLP) as described by Wichterman et al. (31) with minor modification. With the rats under halothane anesthesia, a laparotomy was performed (the size of the incision was ~2.5 cm), and the cecum was ligated with a 3-0 silk ligature and punctured twice with an 18-gauge needle. The cecum was then returned to the peritoneal cavity and the abdomen was closed in two layers. Control rats were sham operated (a laparotomy was performed and the cecum was manipulated, but neither ligated nor punctured). All animals were resuscitated with 4 ml/100 g body wt normal saline at the completion of surgery and also at 7 h postsurgery. Animals were fasted, but had free access to water after operative procedures. Hearts were removed from septic and control animals 9 or 18 h postoperation under chloralose and urethan anesthesia and were then subjected to perfusion as described below. Early and late sepsis refers to those animals killed at 9 and 18 h, respectively, after CLP. The mortality rates were 0% (0/20) for control, 15.4% (4/26) for early sepsis, and 36.4% (12/33) for late sepsis.

Phosphorylation of -ARs. Studies of the phosphorylation of -ARs involved labeling of the tissue ATP pool by perfusion of isolated intact hearts with inorganic [³²P]phosphate, isolation of sarcolemmal membranes and light vesicles by homogenization and centrifugation, and quantification of ³²P-labeled -ARs by immunoprecipitation with anti-₁-AR antibody or antiserum. Hearts removed from control or septic rats were retrogradely perfused (nonrecirculating) by the Langendorff technique under constant pressure (70-80 cmH₂O) at 37°C with Krebs-Henseleit (KH) buffer (in mM: 118.4 NaCl, 4.7 KCl, 1.2 MgCl₂, 2.5 CaCl₂, 10 glucose, 25 NaHCO₃, and 0.1 KH₂PO₄/K₂HPO₄; pH 7.3) for 5 min (first perfusion) followed by a 25-min perfusion (recirculating) with KH buffer in the presence of [³²P]H₃PO₄ (2 mCi/50 ml) to label the tissue ATP pool (second perfusion). At the end of the second perfusion, the hearts were perfused (nonrecirculating) for 5 min with KH buffer in the absence of [³²P]H₃PO₄ to wash out radioactivity (third perfusion). In some experiments, 3 × 10⁸ M of isoproterenol was present during the third perfusion. It should be noted that an isoproternol concentration >3 × 10⁸ M caused perfused myocardium to contract. All of the perfusion solutions were gassed continuously with 95% O₂-5% CO₂ throughout the entire perfusion period. At the end of the third perfusion, heart ventricles were rapidly frozen by freeze-clamping with aluminum clamps precooled in liquid nitrogen and then pulverized. The pulverized myocardium was then used for the following: isolation and purification of sarcolemmal and light vesicle fractions, determination of ATP concentration, and examination of the phosphorylated -ARs by immunoprecipitation. The isolation and purification of cardiac sarcolemma and light vesicles were carried out by a procedure involving a series of repeated homogenization, centrifugation, and sucrose gradient separation as described previously (28). Myocardial ATP was extracted by the method of CoGoli and Dobson (5), and its content was quantified by the method of Adams (1). The phosphorylated -ARs were separated and analyzed by SDS-PAGE (7.5-9.5% acrylamide gradient gel) (28).

Determination of the molecular weight of the phosphorylated -ARs. The molecular weight of the phosphorylated -ARs was determined by photoaffinity labeling of cardiac membranes with [¹²⁵I]iodocyanopindolol ([¹²⁵I]ICYP) in the presence of ATP and the catalytic subunit of the cAMP-dependent protein kinase followed by SDS-PAGE and autoradiography (28). The experiments were conducted as described previously (28) except that 1) sarcolemmal membranes and light vesicles were prepared from hearts perfused in the absence of [³²P]H₃PO₄ and 2) the reaction mixture contained 100 mM KCl, 3 mM MgCl₂, 20 mM NaF, 50 mM histidine, pH 7.4, in the absence or presence of ATP (0.2 mM) plus the catalytic subunit of the cAMP-dependent protein kinase (50 U/ml).

Immunoprecipitation of the phosphorylated -ARs. Immunoprecipitation of the phosphorylated -ARs was performed as described by Bahouth et al. (2) and Jahns et al. (8) with modifications. Cardiac sarcolemmal membranes and light vesicles prepared from hearts perfused in the presence of [³²P]H₃PO₄ were solubilized with 1% 3-[(3-cholamidopropyl)dimethyl-ammonio]-1-propanesulfonate (CHAPS) and 0.25% Triton X-100 in buffer A (300 mM NaCl, 0.15 mM CaCl₂, 0.1 mM EGTA, 25 mM NaF, 0.1 mM phenylmethylsulfonyl fluoride, 1 µg/ml soybean trypsin inhibitor, 1 µg/ml aprotinin, 0.75 µg/ml pepstatin A, 2 µg/ml leupeptin, and 25 mM HEPES-Tris, pH 7.4) for 40 min on ice. The solubilisates were diluted 1:1 with buffer A containing 0 NaCl, followed by centrifugation at 104,000 g for 30 min. The resulting supernatants were incubated with 1:50 dilution of anti-₁-AR antiserum SB-03 prepared against a peptide corresponding to amino acids 396-408 in rat ₁-AR (the antiserum SB-03 was a generous gift from Dr. S. W. Bahouth) or 1:200 dilution of polyclonal anti-₁-AR antibody (Affinity Bioreagents) at 4°C for 14 h in a rotating platform. Rabbit serum and rabbit IgG were used as antibody control for anti-₁-AR antiserum SB-03 and polyclonal anti-₁-AR antibody, respectively. Subsequently, 60 µl of rabbit serum-agarose or rabbit IgG-agarose, preblocked with 2% nonfat dry milk in buffer B (buffer A containing 0.1% CHAPS and 0.025% Triton X-100), was added and incubated for 2 h at 4°C. Finally, the resin was washed four times with buffer B and once more with buffer A. The immunoprecipitated proteins were then analyzed by SDS-PAGE. The phosphorylated ₁-AR signals were scanned and quantified with PhosphoImager and ImageQuant (Molecular Dynamics).

Assays of -ARs. Assay of -ARs was performed using ()-[4,6-propyl-³H]dihydroalprenolol ([³H]DHA) as a radioligand as described previously (28), except that sarcolemmal membranes and light vesicles were isolated from hearts perfused in the same manner but in the absence of [³²P]H₃PO₄ for the studies of the phosphorylation of -ARs.

Measurements of the physiological parameters of perfused rat hearts. Physiological parameters such as heart rate, maximum change in pressure over change in time (±dP/dt_max), and left ventricular developed pressure (LVDP) (left ventricle end-systolic pressure minus left ventricle end-diastolic pressure) of perfused rat hearts were measured with a polygraph. A latex balloon catheter filled with saline was inserted into the left ventricle via the left atrium to measure left ventricular pressure. The volume of the balloon was adjusted to produce a left ventricular end-diastolic pressure of 6 mmHg at baseline condition. The catheter was connected to a pressure transducer (Statham P23 Db), and the heart rate, the ±dP/dt_max, and the LVDP were recorded on a polygraph (Grass Instruments, model 79D) equipped with a direct differentiator.

Other measurements. Membrane marker enzyme activities were determined as previously described (28). The protein content of cardiac membranes was determined by the method of Lowry et al. (13). For measurement of heart cAMP content, 100 mg of frozen tissue sample from nonradioactive perfused hearts were homogenized at 4°C in a buffer containing 50 mM Tris · HCl, pH 7.5, and 4 mM EGTA. The homogenates were heated for 10 min in a boiling water bath. After centrifugation, the cAMP in the supernatant was assayed by the method of Tovey et al. (29).

Statistical analysis. The statistical analysis of the data was performed using one-way ANOVA followed by multiple comparison procedures. Statistical significance was accepted at the 95% confidence limit.

Materials. [³H]DHA (90 Ci/mmol) and [¹²⁵]ICYP (2,000 Ci/mmol) were purchased from Amersham and DuPont-NEN, respectively. ()-Alprenolol, ()-isoproterenol, 5'-guanylyl imidodiphosphate [Gpp(NH)p], the catalytic subunit of the cAMP-dependent protein kinase, soybean trypsin inhibitor, aprotinin, pepstatin A, leupeptin, SDS, and EGTA were products of Sigma Chemical. Polyclonal anti-₁-AR antibody was obtained from Affinity Bioreagents. Anti-₁-AR antiserum SB-03 was a generous gift from S. W. Bahouth, The University of Tennessee, Memphis. Other chemicals and reagents were of analytic grade.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Table 1 shows the effects of perfusion on marker enzyme activities and protein yield of various membrane preparations and on myocardial dry-to-wet weight ratio and ATP content of control animals. Na⁺-K⁺-ATPase serves as marker for sarcolemmal membrane, whereas glucose-6-phosphatase serves as marker for sarcoplasmic reticulum. Na⁺-K⁺-ATPase activities were enriched by 28- and 6-fold for sarcolemmal and light vesicle fractions, respectively, in nonperfused hearts. The pattern of enrichment in the activities of Na⁺-K⁺-ATPase in the sarcolemmal and light vesicle fractions of perfused hearts was essentially identical to that of nonperfused hearts. Glucose-6-phosphate activities were the same in nonperfused and perfused hearts, and, furthermore, the activities were not enriched in sarcolemmal and light vesicle fractions in either nonperfused or the perfused hearts. The yield of protein for each specific membrane preparation was virtually identical in nonperfused and perfused hearts. The dry-to-wet wt ratio of the myocardium was also constant between nonperfused and perfused hearts. Myocardial ATP content remained unaltered after perfusion (Table 1). It should be mentioned that, for septic animals, the pattern of enrichment in the activities of the enzymatic markers in each subcellular fraction, the protein yield, the dry-to-wet wt ratio, and the ATP content of the myocardium were essentially identical between the perfused and nonperfused hearts (results not shown). These data indicate that sarcolemmal and light vesicle fractions were highly purified, and, furthermore, the integrity of membrane preparations and the metabolic status of the myocardium were unaffected by the perfusion procedure for the control as well as for the septic animals.

Figure 1 depicts changes in the electrophoretic mobility of -ARs phosphorylated by the catalytic subunit of the cAMP-dependent protein kinase in cardiac membranes isolated from control animals. In the absence of ATP and the catalytic subunit of the cAMP-dependent protein kinase, one single-binding peptide with a molecular mass of 64,000 Da was labeled with [¹²⁵I]ICYP and visualized in both the light vesicle and sarcolemmal fractions (Fig. 1, lanes 1 and 4). The electrophoretic mobility of the ¹²⁵I-labeled 64,000-Da peptides were shifted to 68,000 Da in both membrane fractions when ATP and the catalytic subunit of the cAMP-dependent protein kinase were present in the reaction mixture (Fig. 1, lanes 2 and 5). The labeling of 68,000-Da peptides in the presence of ATP and the catalytic subunit of the cAMP-dependent protein kinase were completely inhibited by a specific -AR antagonist, alprenolol (Fig. 1, lanes 3 and 6), indicating that the 68,000-Da peptides possess -AR specificity. These data indicate that the electrophoretic mobility of -ARs was shifted from 64,000 to 68,000 Da on phosphorylation of the receptors by cAMP-dependent protein kinase. It should be mentioned that, for septic animals, similar shift in the molecular weight was also observed on the phosphorylation of -ARs (results not shown). The 68,000 Da was then used for the identification of the molecular mass for the phosphorylated -ARs in the experiments perfused with [³²P]H₃PO₄ (Figs. 2-4).

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Representative experiment of the autoradiography from SDS-PAGE of -adrenergic receptors (-AR) labeled with [125I]iodocyanopindolol ([125I]ICYP) in cardiac sarcolemma and light vesicles of control rats. Membranes were reacted with [125I]ICYP in the presence (+) or absence () of alprenolol (10 µM), ATP (0.2 mM), or the catalytic subunit of the cAMP-dependent protein kinase (50 U/ml). Numbers at left represent position of molecular mass standards in kDa.

Figures 2 and 3 show immunoprecipitation of the phosphorylated -ARs by specific antiserum/antibody raised against ₁-AR in the sarcolemmal (Fig. 2) and light vesicle (Fig. 3) fractions from control and septic rat hearts after perfusion with [³²P]H₃PO₄. Figure 4 summarizes quantitative results obtained from immunoprecipitation analysis. The phosphorylated ₁-AR (68,000 Da) from immunoprecipitation (Figs. 2 and 3, lanes 2, 4, and 6) comigrates with ₁-AR from Western blotting (Figs 2 and 3, lane 7). No signals from immunoprecipitation were detected in the bands corresponding to 68,000 Da when specific antiserum/antibody was substituted by antibody control (rabbit IgG) (Figs. 2 and 3, lanes 1, 3, and 5). Analysis of the signals by PhosphoImager and ImageQuant (Fig. 4) indicates that, in sarcolemma membrane, the extent of ₁-AR phosphorylation was decreased by 28.5% (P < 0.01) during early phase, whereas it was increased by 33.8% (P < 0.01) during late phase of sepsis (in arbitrary units: 100, 71.5 ± 1.7, and 133.8 ± 2.3 for control, early sepsis, and late sepsis, respectively; n = 4) (Fig. 4A). In light vesicle fraction, the extent of ₁-AR phosphorylation was decreased by 30.6% (P < 0.01) during early sepsis, but it was increased by 30.3% (P < 0.01) during late sepsis (in arbitrary units: 100, 69.4 ± 2.3, and 130.3 ± 4.1 for control, early sepsis, and late sepsis, respectively; n = 4) (Fig. 4B). These results unequivocally demonstrate that the phosphorylation of -ARs in cardiac sarcolemma and light vesicles was decreased during early sepsis, whereas it was increased during late sepsis.

View larger version (75K): [in this window] [in a new window]   Fig. 2.   Immunoprecipitation of the phosphorylated 1-AR in rat heart sarcolemma. Immunoprecipitation of the phosphorylated 1-AR was conducted as described in METHODS (lanes 1-6). Experiments were conducted in the presence (+) of polyclonal anti-1AR antibody containing 1.76 µg protein or in the presence () of rabbit IgG containing 4.6 µg protein as antibody control. Position of the phosphorylated 1-AR from immunoprecipitation comigrates with 1-AR from Western blotting (lane 7). For Western blotting, lane 7 (control rat) was run simultaneously with lanes 1-6 on SDS-PAGE, excised, and transferred to polyvinylidene fluoride membrane and analyzed by Western blotting using identical antibody as for immunoprecipitation. Numbers at left represent position of mass standards. ES, early sepsis. LS, late sepsis.

View larger version (87K): [in this window] [in a new window]   Fig. 3.   Immunoprecipitation of the phosphorylated 1-AR in rat heart light vesicles. Experiments were conducted as described in Fig. 2, except that light vesicle, instead of sarcolemma, was used. Numbers at left represent position of mass standards.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Changes in the phosphorylation of -ARs in the sarcolemmal (A) and light vesicle (B) fractions of rat hearts during different phases of sepsis. Experimental conditions were identical to those for Figs. 2 and 3. Vertical bars indicate standard errors of the mean. Number of experiments for each specific group indicated in parentheses. ** P < 0.01.

Figure 5 depicts changes in the dynamics of -ARs in rat heart sarcolemmal and light vesicle fractions during different stages of sepsis based on [³H]DHA binding studies. In sarcolemmal membrane fraction (Fig. 5A), the maximal binding capacity (B_max; calculated from Scatchard plots) (Scatchard plots not shown) for [³H]DHA binding was increased by 26% (P < 0.05) during early sepsis but was decreased by 39% (P < 0.01) during late sepsis (135.7 ± 3.7, 171.1 ± 2.8, and 83.0 ± 7.4 fmol/mg for control, early sepsis, and late sepsis, respectively). In light vesicle fraction (Fig. 5B), the B_max for [³H]DHA binding was decreased by 30% (P < 0.01) during early sepsis, but was increased by 31% (P < 0.01) during late sepsis (91.1 ± 2.9, 63.6 ± 1.3, and 119.4 ± 3.5 fmol/mg for control, early sepsis, and late sepsis, respectively). The dissociation constant values were unaffected during early and late sepsis in both membrane fractions (sarcolemma: 3.5 ± 0.2, 3.1 ± 0.2, and 3.6 ± 0.1 nM for control, early sepsis, and late sepsis, respectively; light vesicle: 3.3 ± 0.1, 3.3 ± 0.2, and 3.6 ± 0.1 nM for control, early sepsis, and late sepsis, respectively). It should be mentioned that the binding of [³H]DHA in the sarcolemma was inhibited significantly by isoproterenol perfusion in the control, the early-sepsis, and the late-sepsis experiments (Fig. 5A). In contrast, isoproterenol perfusion stimulated significantly the [³H]DHA binding in light vesicles in the control and early-sepsis experiments (Fig. 5B). Because the B_max for [³H]DHA binding was increased in the sarcolemma but was decreased in the light vesicles during early sepsis, and, conversely, the B_max was decreased in the sarcolemma but was increased in the light vesicles during late sepsis, it is concluded that -ARs in the rat heart were externalized from light vesicles to sarcolemmal membranes (hyperexpression) during early sepsis but were internalized from surface membranes to intracellular sites (hypoexpression) during late sepsis.

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Changes in the density of -AR in the sarcolemmal (A) and light vesicle (B) fractions of rat hearts during different phases of sepsis. -AR were assayed as described in METHODS using [3H]dihydroalprenolol ([3H]DHA) as a radioligand. Hearts were perfused in the presence (+) or absence () of isoproterenol (3 × 108 M) during the third perfusion period. Vertical bars indicate standard errors of the mean. * P < 0.05; ** P < 0.01; NS, not significant.

Table 2 shows the modulation of -ARs by guanine nucleotide Gpp(NH)p in sarcolemmal and light vesicle fractions prepared from control and septic rat hearts. In sarcolemmal membranes prepared from control rats, addition of Gpp(NH)p caused a rightward shift in the isoproterenol competition curve, with an increase in IC₅₀ (concentration of unlabeled ligand required to inhibit 50% of the binding of labeled ligand) value from 7 × 10⁷ to 9 × 10⁶ M. These data indicate that, in sarcolemmal membranes prepared from control rat hearts, agonists recognize two classes of -ARs: a high-affinity class of receptors coupled to guanine nucleotide binding stimulation protein (G_s) and a low-affinity class of receptors apparently uncoupled from G_s. In contrast, the competitive binding studies for isopreterenol in the light vesicles prepared from control rat hearts demonstrate only low-affinity binding of isoproterenol with an IC₅₀ value of 5 × 10⁶ M, and this binding was not affected by Gpp(NH)p. These findings indicate that the light vesicles are functionally and presumably physically uncoupled from G_s, consistent with the notion that light vesicles are the intracellular sites of surface receptors (7, 14). During early sepsis, addition of Gpp(NH)p shifted the agonist competition curve to the right, with an increase in IC₅₀ value from 6 × 10⁷ to 6 × 10⁶ M in the sarcolemmal membranes, but it failed to affect the IC₅₀ value of the light vesicles. It should be noted that the IC₅₀ values reported in the literature vary considerably; some investigators reported comparable (3, 22), whereas others reported higher (6, 10) or lower (18, 23), values. These data support the findings presented in Fig. 5 that -ARs are externalized from light vesicles to sarcolemma in rat heart during the early phase of sepsis. During late sepsis, addition of Gpp(NH)p failed to modulate the IC₅₀ values in sarcolemmal and light vesicle fractions because of receptor uncoupling consequent to hyperphosphorylation, which can be seen in the data presented in Figs. 4 and 5.

View this table: [in this window] [in a new window]   Table 2.   Modulation of -adrenergic receptors by Gpp(NH)p in SL and LV fractions prepared from control and septic rat hearts

Figure 6 illustrates the relationship between changes in the intracellular redistribution of -ARs and alterations in the receptor phosphorylation in rat hearts during different phases of sepsis. In the sarcolemmal fraction, the number of -ARs increases (externalization) during the early phase of sepsis (9 h after CLP) coupled with a simultaneous decrease in the receptor phosphorylation, whereas the number of -ARs decreases (internalization) during late sepsis (18 h after CLP) accompanied with a concurrent increase in the receptor phosphorylation (Fig. 6A). In light vesicle fraction, the decrease in the number of -ARs (externalization) during early sepsis coincided with a decrease in the receptor phosphorylation, whereas the increase in the number of -AR (internalization) during late sepsis correlated with a concomitant increase in the receptor phosphorylation (Fig. 6B). Because phosphorylation/dephosphorylation has been considered to be a common mechanism through which recycling of receptors can be regulated (7, 12, 14, 24, 25), our data presented in Fig. 6 strongly suggest that a decrease in the receptor phosphorylation is a mechanism for externalization of -ARs during early sepsis, whereas an increase in the receptor phosphorylation is a mechanism for internalization of -ARs during late sepsis.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Relationship between changes in the intracellular redistribution of -AR and alterations in the receptor phosphorylation in rat hearts (A, sarcolemma; B, light vesicles) during different phases of sepsis. Figures were constructed using data presented in Figs. 4 and 5 in which hearts were perfused in the absence of isopreterenol during the third perfusion period. CLP, cecal ligation and puncture. Vertical bars indicate SE. * P < 0.05; ** P < 0.01 compared with zero time in each specific group.

Table 3 illustrates the baseline and isoproterenol-stimulated physiological parameters in perfused hearts isolated from control and septic rats. Baseline heart rates remained unaltered for all three experimental groups, but they were stimulated by isoproterenol within each respective group. Left ventricle +dP/dt_max was increased by 19.5% (P < 0.01) in early sepsis but was decreased by 39.7% (P < 0.01) during late sepsis under basal conditions, indicating that myocardial basal contractility was enhanced during early sepsis but diminished during late sepsis. Isoproterenol increased basal left ventricle +dP/dt_max in control, early sepsis, and late sepsis, indicating that myocardial basal contractility was enhanced in all three experimental groups, and, furthermore, the enhancement of basal contractility induced during the early stage of sepsis had not reached maximum. Basal left ventricle dP/dt_max was unaffected during early sepsis but it was decreased by 48.3% (P < 0.01) during late sepsis, indicating that myocardial basal relaxation remained unimpaired during early sepsis but was diminished during late sepsis. Isoproterenol, as expected, stimulated basal relaxation in all three experimental groups. These alterations in the parameters of contraction and relaxation were associated with a moderate, but nonsignificant, increase (13.5%; 0.05 < P < 0.1) in LVDP during early sepsis followed by a significant decrease (48.3%; P < 0.05) in LVDP during late sepsis. The basal LVDP was responsive to isoproterenolol stimulation in all three experimental groups. The basal tissue cAMP contents show a biphasic change during the course of sepsis: an increase in early sepsis (26.5%; P < 0.01) followed by a decrease in late sepsis (30.6%; P < 0.01). The basal tissue cAMP contents were stimulated as expected by isoproterenol. The observed changes in the physiological parameters described above illustrate that during early sepsis myocardium is in the hyperdynamic state, whereas during late sepsis myocardium is in the hypodynamic state.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In this study we used isolated intact hearts perfused with [³²P]H₃PO₄ to label the cellular ATP pool followed by isolation and purification of sarcolemmal and light vesicle membrane fractions. The phosphorylation status of the membrane receptors was then examined by immunoprecipitation with anti-₁-AR antiserum/antibody. This approach in monitoring phosphorylation of -ARs has traditionally been of great use in establishing the functional relevance and the significance of receptor phosphorylation (2, 8, 19). During the course of our study, great care was taken to assure that various physiological and biochemical parameters such as energetic state of the myocardium and the purity and integrity of sarcolemmal and light vesicle preparations were not compromised by the perfusion procedure. This was evident by data presented in Table 1 demonstrating that ATP content and the dry-to-wet wt ratio of the myocardium and the marker enzyme activities of various membrane fractions remained unaffected after completion of perfusion. In examining the phosphorylation status of -ARs by PAGE, the shift in the receptor electrophoretic mobility was verified by the in vitro phosphorylation in the presence of exogenous cAMP-dependent protein kinase (Fig. 1). This shift in the receptor mobility on phosphorylation is consistent with findings reported in the literature (12, 24).

After establishing that the metabolic state of the myocardium and the integrity of subcellular membranes remained unaffected by perfusion procedure and after clarifying the shift in the electrophoretic mobility of -ARs on phosphorylation, we then examined the receptor binding characteristics and the incorporation of [³²P]ATP into -ARs. Binding assays using [³H]DHA as a radioligand demonstrate that, in sarcolemmal membrane fraction, the B_max was increased by 26% during early sepsis but was decreased by 39% during late sepsis. In light vesicle fraction, the B_max for [³H]DHA binding was decreased by 30% during early sepsis but was increased by 31% during late sepsis (Fig. 5). These data indicate that -ARs were externalized from light vesicles to sarcolemmal membranes during early sepsis, whereas they were internalized from surface membranes to intracellular sites during late sepsis. The biphasic redistribution of -ARs between the two different intracellular compartments during the progression of sepsis was further verified by Gpp(NH)p modulation studies (Table 2). Gpp(NH)p caused a rightward shift with an increase in the IC₅₀ value in the sarcolemmal membranes of control and early septic rats, whereas it failed to affect the agonist-binding displacement curves in the sarcolemmal membranes of late septic rats as well as the light vesicle fractions of all experimental groups (control, early, and late sepsis) (Table 2). These data are in agreement with the notion that during early sepsis -ARs were externalized to the surface membranes where agonists recognized two classes of -ARs, a high-affinity class of receptors (IC₅₀ = 6 × 10⁷ M) coupled to G_s and a low-affinity class of receptors (IC₅₀ = 6 × 10⁶ M) apparently uncoupled to G_s protein, whereas during late sepsis the -ARs were internalized to intracellular sites where agonists recognized only the low-affinity class of receptor (IC₅₀ = 1 × 10⁶ M) and were deficient in G_s protein (7, 14). It should be pointed out that the results obtained using perfused hearts in regard to changes in the intracellular distribution of -ARs, as reported in this study, were identical to those reported earlier using nonperfused hearts (28). Further investigation on the immunoprecipitation of the phosphorylated ₁-ARs using hearts perfused with [³²P]H₃PO₄ reveals that the incorporation of [³²P]ATP into -ARs in both the sarcolemmal and light vesicles were decreased by 28.5-30.6% (P < 0.01) during early sepsis but was increased by 30.3-33.8% (P < 0.01) during late sepsis (Figs. 2-4). These results, together with those of binding studies (Fig. 5), unequivocally demonstrate that the externalization of -ARs during early sepsis was coupled with a concomitant decrease in the receptor phosphorylation, whereas the internalization of -ARs during late sepsis was accompanied by a simultaneous increase in the receptor phosphorylation (Fig. 6).

-AR consists of two major subtypes, ₁ and ₂, with ₁ being the predominant subtype (~90%) expressed in mammalian heart (27). In our study, the phosphorylated -ARs were identified by the molecular weight corresponding to ₁-AR (Fig. 1) and further verified by immunoprecipitation with specific antiserum/antibody raised against ₁-AR (Figs. 2 and 3). The receptor binding assay was conducted using cold alprenolol, a ₁-selective antagonist (20), to displace [³H]DHA binding (Fig. 5). These data indicate that the -AR we studied represents ₁-AR.

The pathophysiological significance between changes in the intracellular distribution of -ARs and alterations in the phosphorylation of receptors during different phases of sepsis, as reported in this study, is apparent. Progress in the studies of the regulation of -AR function indicates that increased phosphorylation of -AR leads to its functional uncoupling and physical translocation away from the cell surface into a sequestered compartment, and, once the receptors are sequestered, the phosphorylation is reversed, perhaps enabling the receptors to translocate back to the cell surface (7, 12, 14, 24, 25). Because the externalization of -ARs during early sepsis coincides with the reduced state of phosphorylation, whereas the internalization of -ARs during late sepsis parallels the increased state of phosphorylation (Fig. 6), it is logical to conclude that a decrease in the receptor phosphorylation leads to the externalization of -ARs during early sepsis, whereas an increase in the receptor phosphorylation results in the internalization of -AR during late sepsis. These data offer an explanation on the molecular basis that the covalent modification of receptor proteins by phosphorylation/dephosphorylation is a mechanism responsible for the initial hyperexpression during early sepsis followed by a subsequent hypoexpression of -ARs during the late phase of sepsis.

Recent studies using transgenic mice overexpressing the -AR gene, -AR kinase (-ARK), or -ARK inhibitor provide additional evidence establishing the causal relationship between the expression of -ARs and myocardial function and role of receptor phosphorylation in the control of myocardial function. Milano et al. (16) and Rockman et al. (21) reported that transgenic mice overexpressing -AR gene had a 200-fold increase in -AR density, an increase in basal adenylate cyclase activity, and an enhanced myocardial function as evident by increases in left ventricle +dP/dt_max and dP/dt_max (16, 21). In a separate report, Koch et al. (11) found that transgenic mice overexpressing -ARK-1 demonstrated attenuation of isoproterenol-stimulated left ventricle contractility in vivo, damping of myocardial adenylate cyclase activity, and reduced functional coupling of -ARs. Conversely, mice expressing -ARK inhibitor displayed enhanced cardiac contractility in vivo (11). These reports using transgenic animals illustrate a direct link regarding the cause-and-effect relationship between the increase in -AR density and the enhanced myocardial function and, furthermore, the important role of receptor phosphorylation/dephosphorylation in modulating in vivo myocardial function. Our findings of the hemodynamic changes (Table 3) associated with externalization or internalization of -ARs (Figs. 5 and 6) and the reduced or enhanced state of receptor phosphorylation (Figs. 2-4 and 6) in the early or late phase of sepsis are strikingly similar to those reported in transgenic mice overexpressing -AR gene, -ARK-1, or -ARK inhibitor (11, 16, 21). The similarities in the physiological and biochemical alterations between our studies using experimentally induced sepsis animals and those of transgenic mice overexpressing various genes strongly support the contention that 1) a decrease in the phosphorylation of receptors is a mechanism responsible for the externalization (overexpression) of -ARs, which, in turn, results in an enhanced myocardial contractility during the early hyperdynamic phase of sepsis and 2) an increase in the phosphorylation of receptors is a mechanism leading to internalization (underexpression) of -ARs, which, in turn results in a reduced myocardial contractility during the late hypodynamic phase of sepsis.

Perspectives