Chronic allopurinol administration ameliorates maladaptive alterations in Ca2+ cycling proteins and β-adrenergic hyporesponsiveness in heart failure

Anastasios P. Saliaris, Luciano C. Amado, Khalid M. Minhas, Karl H. Schuleri, Stephanie Lehrke, Marcus St. John, Torin Fitton, Chris Barreiro, Cristine Berry, Meizi Zheng, Kristen Kozielski, Virginia Eneboe, Jeff Brawn, Joshua M. Hare


Xanthine oxidase (XO) activity contributes to both abnormal excitation-contraction (EC) coupling and cardiac remodeling in heart failure (HF). β-Adrenergic hyporesponsiveness and abnormalities in Ca2+ cycling proteins are mechanistically linked features of the HF phenotype. Accordingly, we hypothesized that XO influences β-adrenergic responsiveness and expression of genes whose products participate in deranged EC coupling. We measured inotropic (dP/dtmax), lusitropic (τ), and vascular (elastance; Ea) responses to β-adrenergic (β-AR) stimulation with dobutamine in conscious dogs administered allopurinol (100 mg po daily) or placebo during a 4-wk induction of pacing HF. With HF induction, the decreases in both baseline and dobutamine-stimulated inotropic responses were offset by allopurinol. Additionally, allopurinol converted a vasoconstrictor effect to dobutamine to a vasodilator response and enhanced both lusitropic and preload reducing effects. To assess molecular correlates for this phenotype, we measured myocardial sarcoplasmic reticulum Ca2+-ATPase 2a (SERCA), phospholamban (PLB), phosphorylated PLB (P-PLB), and Na+/Ca2+ transporter (NCX) gene expression and protein. Although SERCA mRNA and protein concentrations did not change with HF, both PLB and NCX were upregulated (P < 0.05). Additionally, P-PLB and protein kinase A activity were greatly reduced. Allopurinol ameliorated all of these molecular alterations and preserved the PLB-to-SERCA ratio. Preventing maladaptive alterations of Ca2+ cycling proteins represents a novel mechanism for XO inhibition-mediated preservation of cardiac function in HF, raising the possibility that anti-oxidant therapies for HF may ameliorate transcriptional changes associated with adverse cardiac remodeling and β-adrenergic hyporesponsiveness.

  • xanthine oxidase
  • calcium signaling
  • cardiac contractility and energetics
  • excitation-contraction coupling
  • oxidative stress

excess reactive oxygen species formation (oxidative stress) contributes to both structural (41) and functional (6, 16) abnormalities in the failing circulation due, at least in part, to upregulated myocardial levels of oxidant-producing enzymes such as xanthine oxidase (XO; see Refs. 5, 10, 27, 39). Allopurinol acutely improves Ca2+ responsiveness of stunned myofilaments (48) and restores cardiac mechanoenergetic coupling, implicating XO-derived oxygen species in both abnormal excitation-contraction (EC) coupling and cardiac energetics (5, 10). XO upregulation also contributes to chronic pathophysiological changes in dysfunctional myocardium resulting from myocardial infarction (MI; see Refs. 11 and 46) or genetic cardiomyopathies (29).

We have previously shown that chronic allopurinol administration completely prevents increases in systemic vasoconstriction and ameliorates reductions in myocardial contractility in dogs with pacing-induced heart failure (HF), maintaining normal ventricular-vascular coupling in the failing circulation (1). Using a rat model, our laboratory has shown that that chronic xanthine oxidase inhibition (XOI) induces reverse remodeling, restoring cardiac structure and function, changing the patterns of Ca2+ cycling protein alteration, and reversing alterations in gene expression (29).

Deficiencies in sarcoplasmic reticulum (SR) Ca2+ reuptake mechanisms are considered integral to HF (7) and are associated both with myocardial remodeling and β-adrenergic hyporesponsiveness. In this regard, a genetic abnormality in phospholamban (PLB; see Ref. 42) resulting in decreased phosphorylation of PLB with subsequent constitutive inhibition of the SR Ca2+-ATPase 2a (SERCA) is recently described as causal in familial dilated cardiomyopathy. Furthermore, gene therapy approaches that restore SERCA (8) improve myocardial contractile function, offering support for the notion that derangements in either SERCA or PLB abundance or activity may play significant roles in HF pathophysiology. Accordingly, we tested the prediction that chronically administered allopurinol restores β-adrenergic responsiveness and contractile performance associated with alterations in the expression of key Ca2+ cycling proteins.


The model of pacing-induced HF has become well established (10, 38, 44, 45, 50). When paced at rates of 210–240 beats/min for periods of 3–4 wk, dogs develop a syndrome of HF associated with a dilated left ventricular (LV) cavity, β-adrenergic hyporesponsiveness, depressed myocardial energetics, and abnormal Ca2+ cycling (44).

Surgical implantation of the chronic hemodynamic monitoring equipment.

Mongrel dogs weighing 25–30 kg were used for the study. Animals were chronically instrumented to measure ventricular pressure and dimensions. The dogs were sedated with thiamyl sodium (17.5 mg/kg), and endotracheal intubation was performed. Surgical anesthesia was maintained with 0.5–2% isoflurane. Antibiotic prophylaxis with 25 mg/kg Cefazolin and 80 mg gentamicin intravenously was given.

A Tygon catheter (Norton Plastics and Synthetic Division, Akron, OH) was secured in the right atrium, aorta, and left atrium. All catheters were inserted under direct visualization through purse string sutures. A solid-state miniature pressure transducer (P6.5; Konisberg Instruments, Pasadena, CA) was placed in the apex of the left ventricle for high-fidelity recordings of LV pressure. Endocardial sonomicrometers (Triton Technologies, San Diego, CA) were inserted in the LV septum and free wall to measure anterior-posterior and short-axis dimension. An inflatable cuff was applied to the inferior vena cava to allow acute reduction in venous return to the heart so as to measure pressure volume loops over a wide range of preloads. For the chronic tachycardia pacing, an epicardial pacing lead was attached to the LV free wall and connected to a programmable stimulator (Spectrax; Medtronics) within a subcutaneous pocket. Another pair of pacing leads were secured in the left atrial appendage for the atrial pacing during hemodynamic measurements. All catheters were externalized to the midscapulae and protected by an external jacket. The incision was closed in layers. The dogs were allowed to recover for 7–10 days following surgery, during which time antibiotics and analgesics were administered. The animal protocol conformed to the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health (National Institutes of Health Publication no. 85-23, Revised 1996). All procedures were approved by the Johns Hopkins University Institutional Animal Care and Use Committee.

Rationale for allopurinol dose.

Traditionally used in the treatment of gout and hyperuricemia, the XO inhibitor allopurinol has demonstrated potential benefit in numerous disease states, including ischemia reperfusion injury, inflammatory disorders, and chronic HF (33). When administered orally, it is rapidly absorbed and oxidized by XO into its active metabolite oxypurinol, a noncompetitive inhibitor of XO (33). Although allopurinol has only a 2- to 3-h half-life, that of oxypurinol may be as long as 30 h because of renal reabsorption. Although generally a safe agent, the main side effects of allopurinol include gastrointestinal distress, skin rash, and hypersensitivity reactions that may result in renal failure (33). The dose of allopurinol used in our experiments was based on that used in humans for mild hyperuricemia (100–300 mg daily), adjusting for the lower weight and body surface area of the dogs. This dose is similar to that used in prior studies of chronic allopurinol administration in HF (1, 15).

HF induction protocol.

After being allowed to recover for 7–10 days after implantation of the hemodynamic monitoring equipment, dogs were randomized to receive allopurinol or identically appearing placebo (100 mg/day, orally). After determination of pre-HF hemodynamics, the animals underwent the 4-wk rapid ventricular pacing protocol. Chronic pacing was begun by activating the implanted pacemaker at a rate of 210 beats/min for 3 wk, which was then increased to 240 beats/min for 1 wk. Pacing was confirmed by daily palpation of pulse and weekly comprehensive hemodynamic evaluation.

Dobutamine infusion protocol.

All data were collected in conscious animals standing quietly in a sling apparatus. At the control state and after HF induction, dogs underwent 5-min infusions of escalating concentrations of dobutamine (2.5, 5, 10, and 15 μg·kg−1·min−1) with hemodynamic measurements performed after each dose had achieved steady state.

Hemodynamic data analysis.

The detailed hemodynamic data analysis used for this study has been described previously (10, 39). Briefly, inotropic responses were assessed using peak +dP/dt and the slope of the dP/dt end-diastolic dimension relationship (+dP/dt-EDD), cardiac preload was indexed by LV end-diastolic pressure (LVEDP) and dimension, afterload as effective arterial elastance (Ea), diastolic relaxation by the time constant of isovolumetric relaxation calculated using the Weiss formula (τln; see Ref. 51) and a logistic regression formula validated in HF (τL; see Ref. 43), and contractility by peak +dP/dt. As previously validated, this model uses LV chamber dimension to approximate chamber volume (the 2 parameters correlate closely both before and after HF; see Ref. 10).

Quantification of mRNA.

To compare the levels of SERCA2a, PLB, and the cell membrane Na+/Ca2+ exchanger (NCX) mRNA expression, we performed quantitative PCR on myocardial tissue obtained from normal dogs and dogs from both the allopurinol and placebo group after completion of the full 4-wk pacing protocol. Total RNA was isolated, cDNA was synthesized, and each sample was run in duplicate on a GeneAmp 7900 Sequence Detection System (PE Applied Biosystems, Foster City, CA) and was analyzed using SDS 2.0 software (Applied Biosystems) as we have described (2, 4). The primer sequences for SERCA, PLB, NCX, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; a housekeeping gene used to standardize the input cDNA and provide a reference for our gene of interest) are as follows: SERCA (U94345): forward (2804–2823) CCAGCTGAGCCACTTCCTAC, reverse (3032–3012) GTGGAGGGACATGGACAGGCA; PLB (M16012): forward (240–259) GCCTCAACAAGCACGTCAAA, reverse (407–388) CTCTTCATGGGATGGCAGAT; NCX (M57523): forward (1819–1906), TTGAGATTGGAGAGCCCC reverse (2100–2082) CTCCTCCTCTTTGCTGGTC; and GAPDH (AB038240): forward (166–186) GGCACAGTCAAGGCTGAGAAC, reverse (266–247) CCAGCATCACCCCATTTGAT.

Quantification of protein abundance.

To compare levels of protein expression, Western blot analysis was performed on total heart proteins as described in detail by our group (23). The expression of the following proteins was determined: SERCA, PLB, phosphorylated PLB (P-PLB; Ser16; Upstate Biotechnology), and NCX (Alpha Diagnostic). A monoclonal anti-GAPDH antibody (1:100,000 dilution; Research Diagnostics) was used separately to normalize any potential differences across samples.

Protein kinase A activity.

To measure protein kinase A (PKA) activity, sections of frozen myocardium (100 mg) were homogenized with a Polytron in 1× kinase buffer (Cell Signaling Technologies) supplemented with phenylmethylsulfonyl fluoride (0.5 mM) and protease inhibitor cocktail (Roche Diagnostics). The homogenate was centrifuged at 14,000 g for 15 min at 4°C, and the supernatant was taken for the concentration measurement with bicinchoninic acid reagent (Pierce) and PKA assay. PKA activity was measured using a nonradioactive fluorescent detection Kit (Promega). The cAMP-dependent protein kinase catalytic subunit was diluted to 2 μg/ml in PKA dilution buffer (350 mM K3PO4, pH 7.5, and 0.1 mM dithiothreitol) and used as positive control. Negative control contained only distilled water. For each sample, 5× PKA reaction buffer, PKA-specific peptide substrate (PepTag A1 peptide), 5× PKA activator solution, peptide protection solution, and water were mixed and incubated at 30°C for 1 min. Samples were added and incubated at room temperature for 20 min, loaded on a 0.8% agarose gel (prepared in 50 mM Tris·HCl, pH 8.0), and run at 120 volts for 40–50 min. The bands were photographed under ultraviolet light and quantified by spectrophotometry with 570 nm absorbance. Finally, the PKA activity was calculated using Beer's Law in accordance with the manufacturer's instructions.

Statistical analysis.

All analysis was performed blind to the randomization. All values are expressed as means ± SE. Differences in baseline values were compared using t-tests or ANOVA with Sheffé's post hoc analysis. Differences in the hemodynamic indexes were compared within the groups using repeated-measures ANOVA and between groups using 2-way ANOVA with an interaction term. A level of P < 0.05 was considered statistically significant.



To examine functional effect of chronic allopurinol, we assessed basal and β-adrenergic inotropic responses (see Table 1 for basal and dobutamine-stimulated parameters). HF induction caused characteristic increases in LVEDP and Ea and reductions in contractility -dP/dtmax and dP/dtmax-EDD (Table 1).

View this table:
Table 1.

Basal and dobutamine-stimulated parameters

Before the induction of HF, dobutamine augmented myocardial contractility, increasing dP/dtmax and dP/dtmax normalized by end-diastolic dimension (dP/dtmax-EDD) while enhancing diastolic relaxation as determined by τL (Table 1 and Fig. 1; see Ref. 34). Furthermore, Ea decreased in response to dobutamine, consistent with a vasodilator effect (Fig. 1B). The net result of these hemodynamic effects was increased cardiac performance as evidenced by augmented stroke work (SW) and cardiac output (CO; Table 1 and Fig. 1C).

Fig. 1.

Impact of chronic administration of allopurinol on myocardial hemodynamics. As depicted, heart failure (HF) is characterized by a reduction in dP/dtmax and increase in elastance (Ea), leading to reduced stroke work in both placebo- and allopurinol-treated animals. A: dP/dtmax is significantly decreased in both HF groups. However, this reduction is attenuated by the chronic administration of allopurinol. B: although Ea increased in both HF and allopurinol groups, allopurinol preserved the afterload-reducing effects of dobutamine that were lost with HF. C: there is a reduction in stroke work in both the placebo- and allopurinol-treated groups with the induction of HF. However, this decrease is attenuated in the allopurinol-treated group. Furthermore, the physiological increase in stroke work with escalating doses of dobutamine is lost in the placebo group but preserved in the allopurinol group. *P < 0.001 vs. pre-HF. †P < 0.005 vs. placebo. ‡P < 0.05, ANOVA with repeated measures. ▪, Pre-HF; •, HF; ▴, allopurinol.

With HF induction, the response to dobutamine infusion was characterized by marked attenuation of inotropic (dP/dtmax and dP/dtmax-EDD) and lusitropic (τln) responsiveness and a conversion of the Ea response from vasodilator to a vasoconstrictor. As a result, in HF, dobutamine did not increase SW or CO (see Table 1 and Fig. 3).

Fig. 2.

Impact of chronic administration of allopurinol on gene transcription of Ca2+ cycling proteins determined by quantitative PCR. A: sarcoendoplasmic reticulum Ca2+-ATPase (SERCA) gene expression remains unchanged with the induction of HF in both placebo- and allopurinol-treated animals. B: phospholamban (PLB) gene expression increases with HF induction, and this upregulation is ameliorated by allopurinol. C: Na+/Ca2+ exchanger (NCX) gene expression rises with the induction of HF in placebo- but not allopurinol-treated animals. D: ratio of PLB to SERCA gene expression rises with the induction of HF in placebo- but not allopurinol-treated animals. E: ratio of NCX to SERCA gene expression rises with the induction of HF in both placebo- and allopurinol-treated animals; n = 5, 5, and 4 animals for pre-HF, HF, and allopurinol, respectively. All results are measured in arbitrary units. *P < 0.05 vs. pre-HF. †P < 0.05 vs. HF.

Fig. 3.

Impact of allopurinol on protein abundance of Ca2+ cycling proteins. A: Western blots depicting changes in Ca2+ cycling proteins with HF induction and the effect of allopurinol. As depicted, HF is characterized by unchanged levels of the SERCA, increased PLB, decreased phosphorylated PLB (P-PLB), and increased cell membrane NCX. Blot 1, levels of SERCA protein remain unchanged with HF both with and without HF; blot 2, PLB protein levels increase in HF, but this increase is prevented by allopurinol; blot 3, P-PLB protein levels decrease in HF, but this decrease is attenuated by allopurinol; blot 4, NCX protein levels rise with HF induction, but this increase is prevented by allopurinol. B: graphic depictions of pooled Western blot data; n = 4–5 for each group. *P < 0.05 vs. HF. †P < 0.05 vs. pre-HF.

Allopurinol treatment affected both basal and β-stimulated responses.

First, the decrease in baseline dP/dtmax and dP/dtmax-EDD (Fig. 1) was ameliorated in treated vs. untreated HF animals (Table 1). Allopurinol did not affect the basal increases in Ea and LVEDP. However, treated animals differed from those administered placebo in that the positive inotropic effect of dobutamine was augmented toward normal with allopurinol treatment (Fig. 1). Moreover, the loss of vasodilator and lusitropic responses to dobutamine were restored with allopurinol therapy. Dobutamine not only stimulated decreases in τ but also led to a marked reduction in LVEDP (Table 1 and Fig. 1). The presence of inotropic, vasodilator, and lusitropic responses to dobutamine in allopurinol-treated animals led to overall improvements in cardiac performance as evidenced by preservation in SW and CO (Fig. 1 and Table 1).

Gene expression.

To determine the molecular mechanisms underlying the effects on β-adrenergic responsiveness, we examined whether gene expression of key Ca2+ cycling proteins was affected by HF without and with allopurinol therapy. As previously described for this model (52), over the 4-wk pacing period, SERCA mRNA levels did not change (Fig. 2). In contrast, PLB mRNA levels increased by approximately threefold (Fig. 1, P < 0.01). In addition, mRNA levels of the other major diastolic Ca2+ removal pump, NCX, increased approximately fivefold (P < 0.001). Thus this model resembles the HF phenotype with regard to increased PLB/SERCA and increased NCX (20, 21).

We next examined the impact of chronic XOI on these transcriptional changes. Although allopurinol did not affect SERCA mRNA expression, the increase in both PLB and NCX with HF was ameliorated, leading to near but not complete normalization of the PLB-to-SERCA ratio and significant reduction of NCX/SERCA (Fig. 2).

Protein abundance.

We next performed immunoblotting to determine whether the changes in gene expression resulted in concomitant alterations in protein abundance. Consistent with RNA levels, SERCA protein abundance was unchanged by HF (Fig. 3). However, PLB abundance increased 1.6-fold (5.2 ± 0.8 vs. 8.3 ± 0.8 arbitrary units before and after HF, respectively; P = 0.035), elevating PLB/SERCA ∼2-fold. In addition, P-PLB (Ser16) levels fell ∼3.6-fold, which, in combination with increased PLB abundance, would be expected to amplify SERCA inhibition. Finally, NCX protein increased (P < 0.0005). Allopurinol treatment largely prevented changes in these Ca2+ cycling proteins. The increase in PLB abundance was eliminated by allopurinol, resulting in normalization of the PLB-to-SERCA protein ratio. Additionally, the marked decrease in P-PLB was restored to normal, and the elevated NCX protein was also reduced toward normal (Fig. 3).

PKA activity.

To link the functional changes in β-adrenergic activation with the Ca2+ reuptake cascade, we next evaluated PKA activity. PKA activity was decreased in HF (13.8 ± 0.1 U/ml) compared with pre-HF (15.9 ± 0.1 U/ml; P < 0.01 vs. pre-HF). Allopurinol treatment restored depressed PKA activity to pre-HF levels (16.7 ± 0.2 U/ml; P < 0.01 vs. HF; Fig. 4).

Fig. 4.

Impact of chronic administration of allopurinol on protein kinase A (PKA) activity. As depicted, PKA activity was decreased in HF but was restored it to pre-HF levels with allopurinol. *P < 0.01 vs. pre-HF. †P < 0.01 vs. HF.


The principal new findings of this study are that allopurinol, when administered chronically in dogs with pacing-induced HF, attenuates the development of β-adrenergic hyporesponsiveness and improves myocardial contractile responses through effects attributable, at least in part, to gene expression and protein abundance of key Ca2+ cycling proteins involved in diastolic Ca2+ removal. These findings offer new pathophysiological insights into the mechanisms by which XO contributes to diminished cardiac reserve in the failing heart and suggest an innovative approach to ameliorating both the functional and molecular changes characteristic of the HF phenotype.

XO is upregulated in HF in both humans and a variety of experimental models, including pacing-induced HF (1). From a functional perspective, acute XO inhibition improves Ca2+ sensitivity (35) and mechanoenergetic uncoupling (1). Furthermore, when given chronically in a variety of animal models, allopurinol and oxypurinol cause reverse remodeling (11, 29, 30) and preserve ventricular-vascular coupling (1). A key HF phenotype not previously addressed is that of β-adrenergic hyporesponsiveness. Here we address this issue by showing that, indeed, XO inhibition with allopurinol reverses, at least partially, diminished β-adrenergic contractility. Alterations in β-adrenergic augmentation of contractility are a hallmark of the HF phenotype and are related to reverse remodeling via the Ca2+ cycling proteins.

Our findings agree with previous studies indicating that XOI restores myocardial function in the failing heart (5, 10, 26, 48). Furthermore, recent studies have extended the effects of XOI to chronic postinfarct remodeling in mice, demonstrating that chronically administered allopurinol reduces myocardial oxidative stress with concomitant reduction of myocardial hypertrophy and interstitial fibrosis as well as improved LV function and decreased mortality (11, 46). Given that XO is a key regulator of oxidative stress (3) and the known effects of intracellular redox milieu on gene transcription (24), it is attractive to speculate that allopurinol may directly effect gene transcription of key Ca2+ cycling proteins. Our study is the first to precisely examine molecular changes that could influence both remodeling and EC coupling in HF. Accordingly, preventing maladaptive alterations of Ca2+ cycling proteins represents a novel mechanism for XOI-mediated preservation of cardiac function in HF.

From the integrated hemodynamic perspective, allopurinol preserved the physiological vasodilator response to dobutamine that is lost with HF induction. This offers an interesting and novel mechanistic correlation to the known effect of allopurinol on endothelial function (9, 12). It could be argued that afterload-reducing effects could contribute to restoring cardiac gene expression to pre-HF levels. However, this is unlikely given that the improvements in Ca2+ cycling proteins and cardiac function occurred in a load-independent manner, since baseline preload (LVEDP) and afterload (Ea) were not significantly different between placebo- and allopurinol-treated HF animals. In addition, there is evidence suggesting that both β-adrenergic receptor blockade and angiotensin-converting enzyme (ACE)-I inhibition may affect the transcription of Ca2+ cycling proteins (37) and that, at least in the case of ACE inhibitors, this occurs in an afterload-independent manner (47). In a study by Takeishi and colleagues (47), guinea pigs underwent aortic banding and were treated with ramipril or placebo. Animals receiving ramipril had increased contractility, improved cardiomyocyte Ca2+ transients, and maintenance of SERCA and PLB expression. Thus XOI shares important features of other successful HF therapeutics with regard to transcriptional regulation of EC coupling.

Our group and others (10, 48) previously evaluated the effects of allopurinol in normal control dogs. These studies indicate that there is minimal to no effect on contractility, as measured by dP/dtmax, preload-recruitable SW, and ventricular Ea, and relaxation, as measured by τ. Although XOI with allopurinol exerts minimal effects on normal myocardium (10, 48), it is well documented that XOI with allopurinol and its active metabolite oxypurinol acutely restores to normal several abnormal HF phenotypes, including depressed myocardial contractility (10), mechanoenergetic uncoupling (39), ventricular vascular coupling (1), the fetal gene program (29), and depressed Ca2+ sensitivity in stunned myofilaments (35). Indeed, using a the pacing canine HF model, Ukai and colleagues (48) demonstrated that “acute” infusion of allopurinol improved both stroke volume and dP/dtmax responses to dobutamine, a finding not observed in normal control dogs. In this context, our findings are somewhat surprising given that XOI improves myofilament contraction without influencing the amplitude of Ca2+ influx in stunned (35), failing (36), and nitric oxide synthase 1 knockout mice (22). Thus the present results support a paradigm of cross talk between the myofilament contractile apparatus and Ca2+ cycling ion channels such that chronically administered XOI prevents or ameliorates key molecular changes in these pumps that contribute to depressed Ca2+ cycling. Indeed, it has been proposed that the alterations in Ca2+ cycling proteins are an adaptive response to reduced myofilament Ca2+ sensitivity, increasing their exposure time to Ca2+ (36). Although Ca2+ transient amplitude is blunted in HF, the changes in Ca2+ cycling proteins result in a slowing of Ca2+ extrusion from the intracellular space, thus increasing the time available for Ca2+ to interact with the contractile proteins, thereby enhancing myofilament activation at the cost of impaired diastolic relaxation (36). Our findings of restored contractility, diastolic function, and Ca2+ cycling machinery to pre-HF levels with chronic XOI suggest that allopurinol exerts concerted effects on myofilaments, mitochondria, and the SR, sites central to HF pathophysiology.

These are the first data implicating XO activity in deranged Ca2+ cycling proteins and extend insights into the mechanism of action of XOI with allopurinol in HF. Defective SR Ca2+ reuptake machinery, a central HF phenotype, results from either decreased SERCA abundance (17, 18, 21, 25, 32) or increased PLB inhibition of SERCA. The latter occurs via decreases in the phosphorylation status of PLB, either intrinsically via genetic alteration of PLB (42), by depressed PKA activity (28), or by increases in the levels of protein phosphatase 1 and 2A, phosphatases involved in P-PLB dephosphorylation (40). In canine pacing-induced HF, although the absolute changes that occur in these proteins remain controversial (7, 13, 32), an increased PLB/SERCA is consistently present, suggesting impaired SR Ca2+ reuptake capabilities. Indeed our current findings recapitulate the description of the PLB mutation in humans with an increase in total PLB abundance and simultaneous reduction in PLB phosphorylation. Furthermore, we support the link between PLB phosphorylation and contractility by showing that depressed PKA activity and lowered P-PLB expression in HF are both restored to normal with allopurinol.

In addition NCX, the major pump involved in diastolic Ca2+ extrusion, is uniformly noted to be elevated in human and experimental HF (19, 32). Thus the overall effect of the altered Ca2+ cycling protein expression characteristic of HF is increased extrusion of Ca2+ in the extracellular space and diminished SR Ca2+ reuptake, resulting in alterations in the kinetics and amplitude of the Ca2+ transient with subsequent derangement in the cardiac action potential (32). Interestingly, our finding that allopurinol restores diastolic relaxation is consistent with the observation in canine HF that NCX inhibition improves diastolic relaxation (20).

Our study investigated chronic allopurinol administration on a model of newly evolving HF and thus is limited in that it does not address the effects of this drug on established HF. We have previously addressed this in the spontaneous hypertension and heart failure rat (29) where we demonstrated that chronic XO inhibition with oxypurinol restored both cardiac structure and function, improving fractional shortening and reducing LV end-diastolic dimensions while attenuating alterations in fetal gene expression/Ca2+ handling pathways. Similar findings are reported for chronic ischemic HF. However, in addition to inhibiting XO, allopurinol and its active metabolite oxypurinol have multiple pleotropic effects that may contribute to our findings. As reviewed by Pacher et al. (33), these compounds may act as antioxidants, free radical and superoxide scavengers, and copper chelators. Furthermore, they have been shown to inhibit lipid peroxidation and heat shock protein expression. Although our experiments focused on the effect of allopurinols on the Ca2+ cycling proteins, future studies investigating these pleotropic effects are required to determine their relative contribution to the prevention of β-hyporesponsiveness.

Despite the abundance of experimental data supporting a pathophysiological role for XO in HF, translation to the clinical has heretofore been challenging (33). A recent study of 50 patients with congestive HF revealed that, despite improvements in cholesterol and brain natriuretic peptide, allopurinol had no beneficial effect on exercise tolerance and actually decreased mean hemoglobin concentration (15). Furthermore, preliminary results from a randomized study of oxypurinol in patients with New York Heart Association class III-IV congestive HF suggested no effect on a composite clinical endpoint assessing morbidity and mortality (14). Nevertheless, our study suggests an interesting and potentially novel clinical aspect to XOI. Dobutamine usage is limited by the increased mortality associated with positive inotropes (31) and the development of tachyphylaxis (49). Here we show that chronic administration of allopurinol preserves the positive inotropic and lusitropic effects of dobutamine that are lost in HF. This opens the possibility of coadministration of allopurinol to patients given dobutamine infusions. More studies, particularly looking at the long-term coadministration of allopurinol and dobutamine, are needed to further evaluate this exciting prospect.

In summary, this paper demonstrates that XOI with allopurinol acts at the level of gene transcription to attenuate the maladaptive changes in Ca2+ cycling proteins associated with myocardial remodeling. This preservation is associated with augmentation toward normal of depressed β-adrenergic inotropic responses in HF. Together the findings offer mechanistic insights into the long-term beneficial effects of allopurinol in the failing circulation.


This work was supported by National Institutes of Health Grants RO1 HL-65455 and RO1 AG-025017 (to J. M. Hare), the Johns Hopkins University School of Medicine Institute for Cell Engineering, and the Donald W. Reynold's Foundation.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract