INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

IT HAS BEEN WELL DOCUMENTED that chronic exercise training induces several adaptations in mammal heart, such as resting and submaximal bradycardia, an increased stroke volume, or improved myocardial contractility (1, 10, 36). The molecular mechanisms involved in such changes, however, are not clearly understood. In this regard, three major biochemical systems are considered to be responsible for the functional characteristics of the myocardium: 1) the metabolic, 2) the contractile, and 3) the Ca²⁺ regulatory systems. In general, the reported studies indicate that exercise training does not alter the glycolytic pathway or the oxidative capacity of cardiac muscle and that few or no training-induced changes occur in myosin ATPase activity or in myosin isoenzyme patterns (1, 10, 15, 37).

Another mechanism that could account for the improved cardiac function in trained animals is an alteration of the Ca²⁺ regulatory systems involved in the excitation-contraction (EC) coupling and relaxation processes. During a normal cardiac contraction-relaxation cycle, extracellular Ca²⁺ influx across the sarcolemmal (SL) membrane, via dihydropyridine (DHP)- sensitive L-type Ca²⁺ channels, triggers a rapid release of Ca²⁺ from the sarcoplasmic reticulum (SR) through a Ca²⁺ channel also referred to as the ryanodine receptor (RyR). Ca²⁺-induced Ca²⁺ release leads to mechanical contraction of the heart. Subsequent activity of Ca²⁺-ATPase in the SR membranes transports a large fraction of the released Ca²⁺ back into the SR lumen, resulting in a rapid decrease in cytosolic Ca²⁺ concentration. The resting diastolic Ca²⁺ is restored by this SR Ca²⁺ uptake, together with Ca²⁺ extrusion by the Na⁺/Ca²⁺ exchanger and Ca²⁺ pump in the SL membrane, although the Ca²⁺ uptake by the SR Ca²⁺-ATPase appears to play the dominant role in relaxation of the heart in several mammalian species (3).

A significant body of evidence exists to support the idea that exercise training elicits alterations in myocardial Ca²⁺ regulation. In young female rats, chronic treadmill running has been shown to produce changes in the lipid composition of cardiac SL associated with an increase in low-affinity Ca²⁺ binding sites (36), altered SL Na⁺/Ca²⁺ exchanger activity via a reduction in the Michaelis-Menten constant for Ca²⁺ (37), and an increase in DHP binding capacity of heart homogenates and SL fractions (38). Treadmill training of old male rats, in turn, results in an increased calcium transport by cardiac SR Ca²⁺-ATPase (33) and an enhanced expression of the gene SERCA2a, which encodes this enzyme (32). Similarly, an increase in the SR Ca²⁺ uptake (12, 19) and SR Ca²⁺-ATPase mRNA levels (7) has also been observed in hearts of swim-trained rats. In contrast, endurance training has been reported to cause no modification in heart SR Ca²⁺ uptake and Ca²⁺-ATPase activity (10, 17, 18, 31), SL Na⁺/Ca²⁺ exchange activity (10), RyR density or modulation by Ca²⁺ (30), or L-type Ca²⁺ channel number or intrinsic function (13). The intensity, duration, and type of exercise training program used are factors that may contribute to these controversial results.

The increased functional demands made by exercise on the heart could also produce adaptations at a different level. Extracellular ATP and adenosine are considered to play an important role as modulators of the cardiovascular system; in this regard, the stimulation of purine receptors modifies processes such as neurotransmitter release, cardiomyocyte contractility, and vascular tone (22). However, the effects of exercise on the ectoenzymes Mg²⁺-ATPase and 5'-nucleotidase (5'-NT), which primarily control extracellular levels of these purines, are at present largely unknown.

On the other hand, the acute effects of physical activity (i.e., a single bout of endurance exercise) on the cardiac Ca²⁺ regulatory systems have been scarcely analyzed. To our knowledge, only two studies have reported that Ca²⁺ uptake by rat myocardial SR was depressed after exhaustive swimming and running exercise (20, 27).

The purpose of this work was to examine the chronic effects of three different exercise-training protocols on the main components involved in the processes of EC coupling and relaxation in rat heart, as well as on the ectonucleotidases involved in extracellular ATP degradation. A complementary aim of our study was to assess the acute effects of a single exhaustive exercise bout on the aforementioned components in sedentary and trained animal hearts.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal Care

Exercise Training Programs

Treadmill Endurance Test

Single Bout of Exhaustive Exercise

Tissue Processing and Homogenate Preparations

Cardiac homogenates were prepared at 0-4°C. The ventricles were minced with scissors and homogenized with a Polytron PT-10 (Kinematica) three times for 5 s at speed setting 3 with a final 1-s burst at speed setting 7 in 6.5 volumes of buffer A (20 mM MOPS, pH 7.0, containing 0.25 M sucrose and protease inhibitors at the following concentrations: 0.23 mM phenylmethylsulfonyl fluoride, 100 µg/ml benzamidine, and 0.5 µg/ml each of aprotinin, leupeptin, and pepstatin A). The resulting homogenate was diluted with five volumes of buffer A, filtered through two layers of cheesecloth, and stored at 80°C to be analyzed. Skeletal muscle homogenates were prepared following the same procedure, except that tissue was homogenized twice for 30 s at the maximal speed setting with a Polytron PT-10. The protein concentration of the homogenates was determined by the method of Lowry et al. (11).

Serum Analyses

Enzyme Activities in Homogenates

Citrate synthase (CS). CS activity was measured at 37°C in the presence of 0.2% Triton X-100 by using the method of Srere (29). This determination was carried out with an aliquot of homogenates, to which defatted bovine serum albumin (final concentration 5 mg/ml) was added before being frozen.

Mg²⁺-ATPase and SR Ca²⁺-ATPase. Mg²⁺-ATPase and SR Ca²⁺-ATPase activities of rat heart homogenates were measured at 37°C as described by Saborido et al. (23). The standard reaction mixture for ATPase assays contained at least 25 mM MOPS, 1 mM EGTA, 15 mM MgCl₂, 200 mM KCl, 10 mM phospho(enol)pyruvate, 2.4 U/ml pyruvate kinase, 10 U/ml lactate dehydrogenase, and 0.27 mM NADH in a final volume of 1 ml. The reaction was started by the addition of 4 mM ATP, and the rate of ATP hydrolysis was calculated from spectrophotometric recording of NADH oxidation at 340 nm ( = 6.22 × 10³ M¹ · cm¹). For the measurement of Mg²⁺-ATPase activity, the reaction medium at pH 7.5 included additionally 5 mM sodium azide and an adequate amount of cardiac homogenate (1 mg tissue wet wt). Because Mg²⁺-ATPase activity decays exponentially as a function of time, the initial rate was calculated from the early part of the reaction progression curves (0-2 min). The nonenzymatic reaction rate and the ATP-independent NADH consumption were assessed by replacing the enzyme sample by buffer or assessed in the absence of ATP, respectively. Mg²⁺-ATPase activity was calculated by subtracting from the overall reaction rate the nonenzymatic and ATP-independent reaction rates.

Ecto-5'-NT. Ecto-5'-NT activity was measured by a method that involves two parallel enzyme activity determinations. The first allows the estimation of total 5'-NT activity (ecto- plus cytosolic-5'-NT). In the second, the presence of 1 mM ,-methyleneadenosine 5'-diphosphate specifically inhibits ecto-5'-NT (40). The difference in activity corresponds to the ecto-5'-NT activity. The reaction mixture contained 50 mM MOPS, pH 7.3, 1 mM EGTA, 5 mM MgCl₂, 1 mM levamisole, and 1 mM KF (inhibitors of nonspecific phosphatases), 0.2% Triton X-100 (to unmask latent activity), and an adequate amount of cardiac homogenate (6 mg tissue wet wt) in a final volume of 1 ml. The reaction was started by the addition of 2.7 mM AMP and was allowed to proceed at 37°C for 1 h, before it was stopped with 0.5 ml of 10% SDS. The inorganic phosphate present in the incubated reaction medium was estimated by the method of Taussky and Shorr (34).

1

1

Binding Assays

Binding of (+)-[³H]PN200-110. Binding of (+)-[³H]PN200-110 (70 Ci/mmol; DuPont-New England Nuclear) to 1,4-DHP receptor (DHPR) was performed essentially as described previously (24), with some modifications. To optimize binding conditions for rat heart homogenates, several factors were checked: amount of cardiac homogenate (5-20 mg tissue wet wt), temperature (25 and 37°C), and time (0.5-3 h) for incubation; divalent cation dependence (none, CaCl₂, and MgCl₂); effects of detergents (0.2 mg/ml saponin and 0.01% Triton X-100); and concentration of unlabeled nitrendipine (0.01-50 µM). Under the established conditions (see below), PN200-110 binding assays were carried out to determine the values of dissociation constant (K_d) and maximum binding capacity (B_max) by using [³H]PN200-110 concentrations ranging from 0.03 to 5 nM and aliquots (equivalent to 12 mg tissue wet wt) of cardiac homogenate from control rats; Scatchard plots were analyzed by linear regression.

Binding of [³H]ryanodine. Binding of [³H]ryanodine (84 Ci/mmol; DuPont-New England Nuclear) to rat heart RyR was performed as described by Sapp and Howlett (25) with some modifications. To optimize binding conditions for rat heart homogenates, several factors were checked: amount of cardiac homogenate (2-8 mg tissue wet wt), temperature (25 and 37°C), and time (0.5-4 h) for incubation; ionic strength dependence (0-1 M KCl); and calcium concentration (0.01-1 mM CaCl₂). Under the established conditions (see below), ryanodine binding assays were carried out to determine values of binding parameters (K_d and B_max), by using [³H]ryanodine concentrations ranging from 0.8 to 35 nM and cardiac homogenate (6 mg tissue wet wt) from control rats; Scatchard plots were analyzed by linear regression.

Data Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Exercise Training Effectiveness

After 12 wk, the three exercise programs induced several changes indicative of a trained state in male rats, as Table 1 shows. First, animal body weight was reduced in all exercising groups compared with the S group, a common finding for trained male rats (5); the decrease was statistically significant only for the T-1 (15% decrease, P < 0.05) but not for T-2 or T-3 training protocols. Second, the results of the treadmill endurance test showed that running time to fatigue was significantly longer (P < 0.01) in trained than in S animals; surprisingly, the observed increase in endurance time was much more remarkable for the T-1 group (9-fold) than for the T-2 group (4.6-fold). Third, the CS activity measured in the homogenates of soleus muscles of T-1 and T-2 trained rats was significantly increased (P < 0.05) with respect to that of S animals, indicating that the exercise programs elicited the well-characterized enhancement of skeletal muscle oxidative capacity (5).

All measured serum parameters (Table 2) for the S and trained animals were within normal ranges, and no statistically significant differences were observed among groups, either in transaminase activities or in total cholesterol, HDL-C, or lactate concentrations. In addition, serum basal levels of testosterone and corticosterone, the most important hormones associated with a stressful condition induced by exercise in rats, were not affected by either of the exercise training programs employed.

Effects of training on heart characteristics are presented in Table 3. Because the heart (ventricles) wet weight of running and S rats was similar, the increased ventricle-to-body weight ratio observed in the T-1 trained animals (P < 0.05) must be attributed to the lower body weight gain during treadmill training. In heart homogenates, the protein content and the oxidative capacity, estimated by CS activity, did not differ among the experimental groups.

Effects of Exercise Training Programs

View larger version (21K): [in this window] [in a new window]   Fig. 1.   [3H]PN200-110 (A) and [3H]ryanodine (B) binding site concentration in cardiac homogenates of sedentary (S) and trained rats (T-1, moderate intensity; T-2, high intensity; T-3, interval running). Data are presented as means ± SD (n = 7-8). There were no statistically significant differences among the experimental groups.

Ryanodine binding in the presence of Ca²⁺ provides a quantitative measure of the number of SR Ca²⁺ channels that open in response to activating Ca²⁺. Under the established experimental conditions, a simple saturation binding was observed, and a K_d value of 2.5 ± 0.2 nM and a B_max value of 22.6 ± 2.3 pmol/g were obtained from Scatchard plot analysis. No significant differences were detected in the amount of [³H]ryanodine specifically bound to heart homogenates from S and trained rats (Fig. 1), suggesting no change due to exercise training in the number of Ca²⁺-activated RyRs.

None of the running programs employed caused significant changes in the SR Ca²⁺-ATPase activity of cardiac homogenates, as illustrated in Fig. 2. Similarly, Mg²⁺-ATPase and 5'-NT activities, plasma membrane-bound enzymes that hydrolyze extracellular nucleotides, were not affected by either of the training protocols (Fig. 2).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Sarcoplasmic reticulum Ca2+-ATPase (A), Mg2+-ATPase (B), and ecto-5'-nucleotidase (C) activities in cardiac homogenates of sedentary and trained rats. Data are presented as means ± SD (n = 7-8). There were no statistically significant differences among the experimental groups. Abbreviations as in Fig. 1.

Effect of Exhaustive Exercise

In all of the aforementioned groups, no changes were detected in heart characteristics (i.e., ventricular wet wt, heart-to-body wt ratio, homogenate protein, and CS activity), but some differences, which confirmed the animal's adaptation to exercise training, were observed in serum parameters (data not shown). Thus in S rats, the exhaustive exercise session caused increases in serum transaminase activities (~60% at 0 h, ~130% at 24 h), lactate concentration (420% at 0 h), testosterone levels (50% at 0 h, 35% at 24 h), and corticosterone levels (71% at 0 h, 50% at 24 h) compared with resting animals; in T-2 trained rats, the final exercise session produced only slight increases in serum transaminase activities (~39% at 24 h) and lactate concentration (30% at 0 h) but marked increases, which returned to control values after 24 h, in testosterone (68% at 0 h) and corticosterone (122% at 0 h) levels.

When Ca²⁺ regulatory proteins were analyzed in heart homogenates (Table 4), no significant alteration in Ca²⁺ channel levels (DHPR and RyR) or in Ca²⁺-ATPase activity was detected, either after the strenuous exercise session or 24 or 48 h postexercise. Similarly, in cardiac homogenates from S and T-2 trained animals, Mg²⁺-ATPase and ecto-5'-NT activities were not modified by the acute exercise bout (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Our results show that, in rat heart, 12 wk of running exercise do not modify the density of Ca²⁺ channels involved in EC coupling or SR Ca²⁺-ATPase activity, at least at the intensities employed in this study. A novel finding is the absence of changes in the Mg²⁺-ATPase, the enzymatic activity that controls the first step in the degradation of extracellular ATP. The three treadmill-running programs employed can be regarded as vigorous exercise for rats and produced the expected increase in exercise tolerance associated with an enhanced oxidative capacity of skeletal muscle. However, although according to Brooks and White (6) the exercise intensity of the three training protocols results in heart rates >85% of rat heart maximum rate, neither increases in heart weight nor changes in myocardial oxidative capacity were observed. The data concur with those of previous studies in which male rats were subjected to similar or more strenuous exercise programs (18, 26), producing significantly lower body weights in runners than in sedentary animals but no differences in heart weights.

Concerning biochemical adaptations in heart induced by endurance training, in light of the controversial results published, it is unclear at present whether the components involved in EC coupling and relaxation undergo modifications. One factor that may account for this uncertainty, at least in part, is the subcellular fraction employed for the measurements. Exercise could conceivably result in an altered distribution of the different membrane vesicles in the various subcellular fractions obtained with routine isolation procedures. This, in turn, could result in an artificial modification in the levels of membrane-system components. In this regard, Chin and Green (8) have shown that rat skeletal muscle fractionation may affect the measured exercise-induced alterations in SR function. To avoid these methodological problems, we have used, in the present study, heart homogenates instead of subcellular fractions. The use of whole muscle homogenates offers the opportunity to examine a more intact preparation and avoids the potential problems of employing nonrepresentative subpopulations of membrane vesicles. It is worth noting that neither the training programs employed herein nor the exhaustive exercise bout modified the protein yield of crude cardiac homogenate. On the other hand, the presence of whole cellular apparatus in the unfractionated homogenates makes accurate enzyme determinations and binding assays, which need to be carefully verified, more difficult. The validity of the procedures for the quantification in rat heart homogenates of Ca²⁺-ATPase, Mg²⁺-ATPase, and ecto-5'-NT activities was established by comparison of the properties of enzyme activities in both homogenate and membrane preparations (Ref. 23 and J. Delgado, unpublished results). Similarly, methods allowing measurement of the concentration of [³H]PN200-110 and [³H]ryanodine binding sites in whole cardiac homogenates were developed, establishing the experimental conditions to minimize nonspecific binding. It must be pointed out that present values of DHPR and RyR densities for S and trained rats were obtained from experiments performed at a single, nonsaturating ligand concentration (2 and 10 nM, respectively) and that we assumed that K_d values were not affected by exercise training, as it has been previously reported for these receptors (30, 38).

The stoichiometry of RyR/DHPR in heart homogenates from control rats was 3.8 ± 0.4, consistent with results from other investigators (4) and supporting the proposed Ca²⁺-induced Ca²⁺ release mechanism for cardiac EC coupling, because each L-type Ca²⁺ channel would have to control several SR Ca²⁺ channels. Neither of the training programs elicited changes in the number of PN200-110 binding sites, although a trend toward decreased density (P < 0.1) was observed in heart homogenates from T-1 trained animals. In female rats, intense treadmill running for 11 wk has been reported to produce an increase in myocardial homogenate and SL DHP binding capacity, which was proposed to underlie the enhanced heart contractility induced by training (38). On the other hand, we have previously shown that a moderate treadmill-training program like the T-1 used here, capable of inducing a marked increase in DHPR levels in homogenates of skeletal muscle, was ineffective in modifying the number of DHP binding sites in heart homogenates (24). The possibility that this apparent discrepancy was due to the difference in the intensity of training appears unlikely because the more intense T-2 and T-3 programs were also ineffective in increasing DHP binding sites. Furthermore, Mokelke et al. (13) have recently reported that whole cell Ca²⁺ current (I_Ca) vs. voltage and I_Ca density vs. voltage relationships, as well as I_Ca inactivation and recovery characteristics, were unaffected in cardiomyocytes isolated from treadmill-trained rats, indicating the absence of variations in the number or intrinsic function of L-type Ca²⁺ channels. These later results and the data from this work, obtained by means of different methodological approaches, do not support the previous proposal that endurance training induces an adaptive response in the cardiac Ca²⁺ channel that controls extracellular Ca²⁺ influx.

Similarly, RyR levels were unmodified by the training programs, although a small, nonsignificant decrease (P < 0.1) that paralleled that of DHPR levels was observed in heart homogenates of T-1 trained rats. In agreement with these results, no changes in both B_max and K_d of ryanodine binding have been observed in heart homogenates of female rats subjected to an intense running program for 16 wk (30). However, these data do not exclude the possibility that training elicits a modification of the properties of the SR channel, involving an improvement of the release of Ca²⁺ necessary for cardiac contraction.

With respect to the SR Ca²⁺-ATPase, our results show that, in heart homogenates, this enzymatic activity was not significantly affected by the training programs irrespective of their intensity. Ca²⁺-ATPase activity and SR Ca²⁺ uptake have been reported not to be modified in hearts of pigs, rats, and dogs trained by running (10, 18, 31). In previously sedentary old rats, treadmill training is capable of inducing increases in the sequestering activity of the cardiac SR as well as in the level of SERCA2a isoform and its mRNA, likely through multifactorial and complex signals, which may include changes in the thyroid status (32). Because the expression of SERCA2a is downregulated in rats by aging, Ca²⁺ transport results increased to the rate observed in sedentary younger rats. Thus it is quite possible that this exercise stimulus is without effect on SERCA2a gene expression in young or mature animals. In support of this idea, mitochondrial cytochrome oxidase activity and cytochrome oxidase mRNA levels, which undergo an aging-dependent reduction, are augmented with running training in old rats (32) but remain unmodified in young rats (15). On the other hand, cardiac SR has been shown to have a greater capacity to sequester Ca²⁺ in sarcotubular fractions of swim-trained vs. normal rats (12, 19) with some exceptions (17). Furthermore, Buttrick et al. (7) observed in hearts of female rats subjected to an intense swimming program an increase in the SERCA2a mRNA levels that was paralleled by an improvement in cardiac relaxation. All of the above results appear to indicate that the type of stimulus (running vs. swimming) could be a main factor in inducing Ca²⁺-ATPase adaptation.

The activity of heart homogenates, herein designated as Mg²⁺-ATPase, is not attributable to mitochondrial ATPase, which was inhibited by azide, or to myofibrillar ATPase, which showed no measurable activity under the employed incubation conditions. The characteristics of Mg²⁺-ATPase activity in cardiac homogenates (23) correspond mainly to those reported for Ca²⁺/Mg²⁺-ATPase and ATP diphosphohydrolase activities of rat heart SL (9, 39), and, taking into account the Mg²⁺-ATPase assay conditions, it seems likely that the activity of both enzymes was measured in the assay. Ca²⁺/Mg²⁺ ATPase and ATP diphosphohydrolase are ectoenzymes, located at the surface of cardiomyocytes and endothelial and smooth muscle cells of coronary vessels (2) where, together with ecto-5'-NT, they can regulate the concentration of extracellular ATP and adenosine. These purines are considered to be important in the modulation of vascular tone, coronary flow, and rhythmic activity of the heart, and adenosine, in particular, is involved in the prevention of deleterious sequelae of ischemia (22). Because exercise induces an increase in the functional demand on the heart, the adaptation of the cardiac ectonucleotidase activities might be associated with improved heart function in trained animals. Our data indicate that exercise training failed to alter Mg²⁺-ATPase and ecto-5'-NT activities. To our knowledge, this is the first time that the effect of exercise training on Mg²⁺-ATPase activity has been evaluated in whole heart homogenates; previously, only basal (or Ca²⁺-independent) Mg²⁺-ATPase activity, present in isolated SR and probably due to contamination of the fraction by superficial membranes, was shown not to change with treadmill training (10, 18). On the contrary, 5'-NT activity was reported to be elevated in cardiac SL vesicles isolated from swim-trained rats (21); the discrepancy with our results could arise because of the exercise paradigm or the subcellular fraction used.

Although exercise training per se seems to induce no changes in cardiac EC and Ca²⁺ regulatory systems when heart was isolated from animals 48 h after the last exercise bout, it is conceivable that a single session of exhaustive exercise may affect myocardial function (35) and, in turn, affect the components of membrane systems basic for the adjustments in intracellular homeostasis; these effects might be different in trained and untrained animals. The session of strenuous exercise before death had no significant effect on either of the Ca²⁺ channels involved in EC coupling or on either of the membrane enzyme activities determined; similarly, no variations were detected in either of the postexercise recovery times. Direct comparison of our results with other studies is difficult because of the very scarce investigations reported in this field, together with differences in the exercise protocol employed (type, duration, and intensity). To our knowledge, there are no previous investigations on the effect of acute exercise on heart DHPR or RyR levels, but a decrease in Ca²⁺ uptake by cardiac SR has been described after a single, exhaustive exercise bout (prolonged swimming or running) in previously sedentary rats (20, 27).

In conclusion, our results do not support that running training elicits alterations in myocardial Ca²⁺ regulatory systems. Nor do ectonucleotidase activities appear to be modified by treadmill training or acute exercise. The lack of a training effect on these biochemical systems may be attributable to insufficient exercise intensity and/or duration, although other studies employing very different exercise programs also failed to find any significant change in these systems (10, 13, 18, 30, 31). However, treadmill training has been shown to produce increases in mean myocyte capacitance and myocyte length, providing evidences for training-induced cardiomyocyte hypertrophy (13, 14). Moreover, there are data indicating that chronic exercise influences the contractile function of different types of cardiac preparations, including left ventricle papillary muscle (36) and isolated myocytes (14), where a modification of Ca²⁺ influx and efflux pathways has been proposed. A possible explanation for the absence of changes detected in the Ca²⁺ regulatory systems when they are studied in vitro may be that there are fast-responding mechanisms for the regulation of these biochemical systems that function in vivo but are no longer active when the tissue is homogenated. In this regard, Nieto et al. (16) have observed, in rats treadmill trained for 10 wk, a desensitization of the cardiac -adrenergic adenylate cyclase transduction system, which is directly involved in several phosphorylation processes of the cardiomyocyte Ca²⁺ regulatory systems. This is an interesting possibility for more detailed future studies.