INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODOLOGY RESULTS DISCUSSION REFERENCES

PARAXANTHINE (1,7-dimethylxanthine) is the most abundant of the dimethylxanthines (Fig. 1), accounting for ~80% of the hepatic degradation of caffeine in humans (15). Four hours after the ingestion of 270 mg caffeine, plasma paraxanthine concentrations peaked at 8-10 µM in resting human subjects (15). Plasma concentrations of the other two caffeine metabolites, theobromine (3,7-dimethylxanthine) and theophylline (1,3-dimethylxanthine), were lower (3 and 1.5 µM, respectively) and peaked ~8 h after caffeine ingestion.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Chemical structure of caffeine (1,3,7-trimethylxanthine) and its primary metabolic breakdown products, the dimethylxanthines: paraxanthine (1,7-dimethylxanthine), theophylline (1,3-dimethylxanthine), and theobromine (3,7-dimethylxanthine).

Despite paraxanthine being the most abundant caffeine metabolite in humans, it has received little attention in the literature for its potential ergogenic (7, 11, 15, 22) and sympathomimetic (2) effects. Hetzler and colleagues (11) provided some of the most interesting research with respect to the potential ergogenic effects of paraxanthine in humans. These authors found a rise in plasma free fatty acids 3 h postingestion of 4 mg/kg caffeine. It was concluded that the rise in plasma paraxanthine (which reached ~5 µM) was strongly correlated to the rise in plasma free fatty acids. These results were later supported by Benowitz et al. (2), who administered 2 or 4 mg/kg caffeine, paraxanthine (similar doses as caffeine), or placebo in a crossover design. The authors demonstrated that both caffeine and paraxanthine, at 4 mg/kg, have similar sympathomimetic actions, including increased plasma epinephrine and free fatty acid concentrations.

Another potential ergogenic benefit of paraxanthine involves the reduction of plasma K⁺ concentration ([K⁺]). Because increases in plasma [K⁺] have been implicated in skeletal muscle fatigue (21), mechanisms that reduce the rate or magnitude of plasma [K⁺] increase may attenuate the onset of skeletal muscle fatigue.

Clinically, Hall et al. (9) investigated the incidence of hypokalemia in persons admitted to the hospital with a dimethylxanthine (theophylline) overdose. On admission, plasma theophylline varied from 35 to 156 µg/ml (0.19-0.86 mM), with values over 30 µg/ml (0.17 mM) considered to be overdose. All 22 patients admitted were hypokalemic, with the decrease in plasma [K⁺] being correlated to the serum theophylline level. The authors suggested that the hypokalemia resulted from an acute shift of K⁺ from the extracellular to the intracellular compartment, via stimulation of the Na-K-ATPase, possibly the result of hyperinsulinemia or elevated plasma catecholamines. These observations were corroborated by Van Soeren and colleagues (22), who demonstrated a correlation between the rise in plasma paraxanthine and the decrease in plasma [K⁺]. Within the context of a larger study, it was demonstrated that, 3 h after the ingestion of 6 mg/kg caffeine, decreases in plasma [K⁺] were highly correlated to the rise in plasma dimethylxanthines but not with caffeine. Interestingly, the subjects in this study had cervical lesions and were thus unable to evoke an epinephrine response, demonstrating that the decrease in plasma [K⁺] was not secondary to an epinephrine-induced increase in Na-K-ATPase activity. This work suggested that paraxanthine (and the other dimethylxanthines) could be increasing Na-K-ATPase activity in resting skeletal muscle. This hypothesis has recently been confirmed by Hawke et al. (10) in the perfused rat hindlimb. It was demonstrated that concentrations of paraxanthine between 0.01 and 0.5 mM result in an increase in ouabain-sensitive unidirectional K⁺ influx (i.e., increased Na-K-ATPase activity).

The mechanism(s) by which paraxanthine stimulated skeletal muscle Na-K-ATPase activity in the aforementioned studies is unknown. It has been suggested that increases in intracellular Na⁺ concentration, adenosine receptor antagonism, and increases in intracellular calcium concentration ([Ca²⁺]_i) may all be involved (17). Changes in [Ca²⁺]_i with paraxanthine administration provide a most intriguing possibility because of the numerous roles of intracellular Ca²⁺ as a second messenger (3).

A first step toward understanding the mechanism(s) by which caffeine metabolites can affect skeletal muscle is to understand the potential signaling cascade(s) that may be involved. Thus the purposes of the present study were to determine whether 1) varying pharmacological concentrations of paraxanthine could increase [Ca²⁺]_i in skeletal muscle fibers; 2) a physiological concentration of paraxanthine could increase [Ca²⁺]_i; and 3) the [Ca²⁺]_i responses were below contracture threshold so as to involve Ca²⁺ signaling pathways that may activate the Na-K-ATPase.

It was hypothesized that 1) 0.5-5 mM paraxanthine results in subcontracture increases in [Ca²⁺]_i in isolated, single skeletal muscle fibers, and 2) paraxanthine, at concentrations found in vivo after caffeine ingestion, increases [Ca²⁺]_i in intact skeletal muscle. If paraxanthine is able to elevate [Ca²⁺]_i, then increased [Ca²⁺]_i may provide a mechanism, directly or indirectly, by which paraxanthine could affect skeletal muscle metabolism and ion regulation. The hypothesis that paraxanthine causes subcontracture increases in [Ca²⁺]_i is supported by the results of this study.

	METHODOLOGY

TOP ABSTRACT INTRODUCTION METHODOLOGY RESULTS DISCUSSION REFERENCES

Two experimental methodologies were utilized in this study. The mice used in this study were housed in a controlled environment with food and water available ad libitum. This study was approved by the animal care committees at the University of Sydney, Australia, and University of Guelph, Ontario. Animal care and procedures were performed in accordance with Canadian Council on Animal Care guidelines.

Single-fiber preparation. The flexor digitorum brevis (FDB) muscle was surgically removed from male mice killed by rapid neck disarticulation. The muscle was placed in a Tyrode solution, cleaned of connective tissue and fascia, and then placed in a Tyrode solution containing collagenase (259 units, type 2; Worthington Biochemical, Lakewood, NJ). The Tyrode solution contained (in mM) 121 NaCl, 5 KCl, 1.8 CaCl₂, 0.5 MgCl₂, 0.4 NaH₂PO₄, 5.5 glucose, 24 NaHCO₃, and fetal calf serum albumin (~0.2%). The solution was gassed with 95% O₂-5% CO₂ to obtain a final pH between 7.2 and 7.4. The muscles in the collagenase solution were gently agitated in a 35°C water bath for 45 min. The muscles were gently mixed to allow them to dissociate into single fibers. After collagenase treatment, the muscles were transferred to a collagenase-, Ca²⁺-, and NaHCO₃-free Tyrode solution containing the membrane-permeable fluorescent calcium indicator indo 1-acetoxymethyl ester (AM) (10 µM; Molecular Probes, Eugene, OR) for 30 min at 22°C. After indo 1-AM treatment, the muscle fibers were removed and placed in the gassed control Tyrode solution (indo 1-AM free) until they were utilized.

Experimental protocol 1. Single fibers were placed in a chamber and superfused with a normal Tyrode solution at 22°C. Fiber viability was tested by using a tetanic contraction (100 Hz, 350-ms duration). In some fibers, viability was also assessed by using a Tyrode solution containing 13 mM KCl, whereby high extracellular [K⁺] results in membrane depolarization and large increases in [Ca²⁺]_i (Fig. 2A). If a fiber did not increase [Ca²⁺]_i during stimulation or KCl exposure or subsequently return to baseline after stimulation or washout, the fiber was omitted from the study.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   A: individual fiber response with exposure to 5.0 mM paraxanthine (PX) followed by washout of PX and preexposure to 10 mM procaine. Next, each fiber was then reexposed to 5.0 mM PX in the presence of 10 mM procaine (procaine + PX). This fiber was initially exposed to 13 mM KCl. Fiber viability can be assessed by using high-extracellular K+ concentration because this will result in membrane depolarization and large increases in intracellular calcium concentration ([Ca2+]i). If a fiber did not increase [Ca2+]i during tetanic stimulation or KCl exposure or subsequently return to baseline after stimulation or washout, then the fiber was omitted from the study. B: change in fluorometric ratio with 5.0 mM PX and with PX in the presence of the sarcoplasmic reticulum Ca2+ release channel blocker procaine (10 mM). The protocol for this series is demonstrated in A. Procaine blocked ~75% of the PX-induced increase in intracellular Ca2+. Delta 400/505 nm, difference in fluorometric ratio of 400 to 505 nm. Values are means ± SE. *Significant increase in the fluorometric ratio above baseline, P  0.05. #Significant difference between control PX and PX in the presence of procaine, P  0.05.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   A: response of single fibers to varying PX concentrations at rest and during tetanic contraction. Open bars, change from baseline in fluorometric ratio in response to varying PX concentrations in resting fibers (denoted by an A in B). Solid bars, change in fluorometric ratio during tetanic contraction (100 Hz, 350-ms duration) in the presence of varying PX concentrations compared with a control tetanic contraction (in the absence of PX; denoted by a B in B). Values are means ± SE. Nos. in parentheses, no. of fibers. *Significant change from baseline/control, P  0.05. #Differences between the response to 0.5 mM PX and other PX concentrations at rest or during tetanic contractions, P < 0.05. B: individual fiber response to 5 mM PX. During control (Tyrode) superfusion, the single fiber underwent a tetanic contraction (100 Hz, 350-ms duration). The fiber was then superfused with Tyrode containing 5 mM PX. During this period, the fiber was again exposed to a tetanic contraction (100 Hz, 350-ms duration). The change in fluorometric ratio at rest in response to PX is denoted by an A. The change in fluorometric ratio with PX compared with control tetanic contraction is denoted by a B. The time to peak PX response is calculated as follows: time 0 is the period of time just before the increase in [Ca2+]i, and peak PX response time is the time at which a plateau is reached in [Ca2+]i. The time to peak [Ca2+]i of this particular fiber was longer than the average response time.

Intact muscle preparation. The FDB muscle was surgically removed from mice killed by cervical disarticulation. The muscle was cleaned of connective tissue and fascia and was incubated in a Ca²⁺- and NaHCO₃-free Tyrode solution containing indo 1-AM (10 µM; Sigma Chemical) for 45 min. After indo 1-AM incubation, the muscle was tested for viability by observing the fluorometric ratio response to 13 mM KCl. After this, muscles were maintained in a gassed control Tyrode solution (indo 1-AM free) at 22°C until they were utilized.

Experimental protocol 2. Muscles were placed in a 1-ml quartz vial containing Tyrode solution and placed in a spectrophotometer (LS50, Perkin Elmer, Norwalk, CT). The muscle was illuminated at a wavelength of 350 nm, and the fluorescence emitted was collected at 405 nm (Ca²⁺-bound form of indo 1) and 445 nm (isobestic point). Muscle position within the cuvette was adjusted to provide the largest 405-nm fluorescence signal. After collection of a stable baseline (1 min), paraxanthine was added to the cuvette to produce a final bath paraxanthine concentration of 0.01 mM. Collection of fluorescence data continued for 20 min.

Statistics. The baseline values presented are an average of the final 5 s before drug administration, whereas the peak drug values are a 2-s average of the peak fluorometric response that occurred in the presence of the drug. The data were analyzed by using a paired t-test to compare the baseline and peak drug fluorometric ratios. A one-way ANOVA was performed to test for differences between doses or drugs (in the case of the dimethylxanthines). In the procaine series, a one-way repeated-measures ANOVA was performed to compare the paraxanthine response with the procaine plus paraxanthine response recorded in the same fiber. Where appropriate, Dunnett's post hoc test was used when a significant F ratio was obtained. Significance was accepted at P  0.05. Data presented are means ±SE, unless otherwise stated.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODOLOGY RESULTS DISCUSSION REFERENCES

Single fiber. Exposure of single fibers to paraxanthine resulted in significant increases in [Ca²⁺]_i at all concentrations utilized, and there was a significant difference between the increases in [Ca²⁺]_i at 0.5 and 5 mM paraxanthine (Fig. 3A). The paraxanthine-induced increases in [Ca²⁺]_i recorded during tetanic contraction were also concentration dependent and were consistently larger than those seen with paraxanthine administration at rest (Fig. 3A). The increases in [Ca²⁺]_i elicited by paraxanthine administration were below the contraction threshold, as no muscle shortening was observed, and were also transient (Fig. 3B). The average time to peak [Ca²⁺]_i with 5 mM paraxanthine was 18 ± 2 s, with an average time to return to baseline of 28 ± 5 s. The time to peak was slower with 2.5 mM paraxanthine (25 ± 4 s); however, the time required to return to baseline was similar to that seen with 5 mM paraxanthine (28 ± 11 s). After the transient increase in [Ca²⁺]_i, there was an undershoot below baseline [Ca²⁺]_i before a return to baseline levels. Because of the small increases in the fluorescence ratio with the lower paraxanthine doses, it was not feasible to determine the time courses of the responses accurately.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Response of single skeletal muscle fibers in response to dimethylxanthines [PX, theophylline (TP), and theobromine (TB)]. Nos. in parentheses, no. of fibers. Values are means ± SE. *Significant increase in fluorometric ratio compared with baseline, P  0.05. There was no significant difference among any of the dimethylxanthines in their ability to increase intracellular Ca2+.

1

Intact muscle. Intact FDB muscles (n = 13) exposed to a physiological paraxanthine concentration (0.01 mM) also demonstrated a significant increase in the fluorescence ratio (405/445 nm) of ~4% (1.66 ± 0.12 to 1.72 ± 0.12; P < 0.001) that was transient and subcontracture in nature.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODOLOGY RESULTS DISCUSSION REFERENCES

This is the first study to demonstrate that the primary caffeine metabolite paraxanthine results in transient, subcontracture increases in [Ca²⁺]_i in resting mammalian skeletal muscle. These findings suggest that, for the first time, in vivo paraxanthine concentrations are capable of increasing [Ca²⁺]_i to levels that may initiate Ca²⁺-mediated signaling pathways within mammalian skeletal muscle. Furthermore, these results provide support for the potential role of intracellular Ca²⁺ in mediating methylxanthine's effects on skeletal muscle. It has also been demonstrated that there were no differences among the dimethylxanthines in their ability to, or rate at which they, increase [Ca²⁺]_i. Addition of 10 mM procaine (SR Ca²⁺ channel blocker) successfully reduced the paraxanthine-induced rise in [Ca²⁺]_i by ~79%, indicating that paraxanthine appears to be increasing [Ca²⁺]_i primarily by SR Ca²⁺ release.

The lowest concentration of paraxanthine utilized in this study (0.01 mM) is similar to the plasma paraxanthine concentration (0.008 mM) measured after ingestion of 270 mg of caffeine in humans (14). Considering that caffeine degradation in vivo results in measurable increases in all the dimethylxanthines, the effects on [Ca²⁺]_i seen with theophylline and theobromine may be additive to the effects of paraxanthine and caffeine. In addition, with prolonged caffeine exposure in humans, the accumulation of circulating dimethylxanthines may contribute to the pharmacological effects of caffeine on various tissues (2).

The calculated rate constant of entry for the dimethylxanthines paraxanthine and theophylline in the present study was similar to those reported by Donoso et al. (4) using isolated ventricular myocytes. These researchers calculated the rate constant of entry for caffeine at ~10-fold greater (2.3 ± 0.12 s¹) than for theophylline (0.20 ± 0.02 s¹) and theobromine (0.23 ± 0.03 s¹).

Previous research has offered conflicting results with respect to the transient nature of the increases in [Ca²⁺]_i with methylxanthine administration (13, 14, 19, 24). Similar to the present study, Konishi et al. (14) exposed single frog skeletal muscles to a low methylxanthine concentration (0.3 mM caffeine) and observed a transient increase in the aequorin light signal with no tension development. Similarly, in isolated rat ventricular myocytes, a transient increase in [Ca²⁺]_i was observed with caffeine concentrations of 1-10 mM (18). Also in agreement with the present study, these authors (18) had difficulty detecting increases in [Ca²⁺]_i in resting single fibers exposed to low methylxanthine concentrations (50-500 µM).

Increases in [Ca²⁺]_i have been implicated in a number of cellular signaling events in a variety of tissues (3). The paraxanthine-induced increases in [Ca²⁺]_i have been speculated to play a role in mediating some of the responses associated with caffeine ingestion, including increasing plasma free fatty acid concentration (11) and modulation of Na-K-ATPase activity (10, 17, 22). Support for Ca²⁺ involvement in a signal transduction pathway comes from research demonstrating that increases in ouabain-sensitive ⁴²K uptake (suggestive of increased Na-K-ATPase activity) are maintained for at least 60 min with 0.1 mM paraxanthine administration (10) and are not abolished with 15 min of washout (26). Because the paraxanthine-induced increase in [Ca²⁺]_i is transient, the possibility of secondary phosphorylation/dephosphorylation events (6, 16, 20) or Na-K-ATPase translocation from intracellular stores is more probable than a direct effect of Ca²⁺ on the Na-K-ATPase. Another possibility is that paraxanthine, like caffeine, resulted in an increase in unidirectional K⁺ efflux through K⁺ channels resultant from a rise in [Ca²⁺]_i (23). The concomitant alterations in K⁺ homeostasis may be large enough to result in a stimulation of Na-K-ATPase activity. Further studies are needed to clarify the relationships among increases in [Ca²⁺]_i, phosphorylation events, and Na-K-ATPase activity.

Methodological considerations. The ability of intact muscles loaded with indo 1-AM to produce a "more sensitive" measure of [Ca²⁺]_i compared with single fibers may be a function of the FDB muscle itself. Because the mouse FDB is a small, thin muscle with a central tendon running through it, it is prevented from "curling" during incubation. The diffusion of oxygen, nutrients, and indo 1-AM into the FDB muscle does not appear to be limiting, because the maintenance of energy substrates (T. J. Hawke and D. Batiste, unpublished observations) and measurable changes in fluorescence are very good. The multiple fibers present in the intact muscle preparation result in an enhanced fluorescent response compared with the single fiber.