| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Cardiology and Angiology and 2Institute for Pharmacology and Toxicology, University Hospital Münster, D-48129 Münster; 3Institute for Pathology, University of Essen, D-45122 Essen, Germany; and 4Department of Pediatrics and Cardiovascular Research Center, University of Virginia, Charlottesville, Virginia 22908-1356
Submitted 2 December 2002 ; accepted in final form 2 February 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
heart rate regulation; autonomous nervous system; heart rate variability; sinus node dysfunction; atrioventricular block; atrial fibrillation
Familial occurrence and an autosomal dominant inheritance in some cases suggest a genetic predisposition to sinus node dysfunction (19, 24). Mutations in pacemaker channel genes (e.g., HCN2) could potentially cause sinus node dysfunction (37), but so far the etiology of the disease remains unresolved.
Under physiological conditions, heart rate is mainly determined by the
depolarization rate of pacemaker cells in the sinus node
(6,
32). Autonomic regulation of
pacemaker activity probably alters the function of muscarinic K+
current (IK,ACh), pacemaker current, the L-type
Ca2+ channel, and sustained inward current
(6,
32). In addition to
-adrenoreceptors and muscarinic receptors, A1 adenosine
receptors (A1AR) are active in the sinus node. Short-term
A1AR stimulation causes bradycardia and AV block in humans
(7). Neurotransmitters and
hormones like adenosine that activate IK,ACh all stimulate
pertussis toxin-sensitive G proteins, resulting in the dissociation of the G
protein complex in G
and GTP-bound G
subunits. Liberated
G activates IK,ACh directly by physical
association (14). As adenosine
is constantly present in the interstitial space
(5), chronically enhanced
function of A1AR could contribute to both sinus nodal and AV nodal
dysfunction in vivo.
To study the physiological and pathophysiological relevance of these potential heart rate-regulatory mechanisms, we used transgenic (TG) mice overexpressing A1AR (13, 23). These mice have an increased tolerance to ischemia. Of note, A1AR overexpression slows heart rate in isolated hearts (13, 23). In this study, we describe that overexpression of A1AR decreases the physiological heart rate response to exercise in freely moving mice and slows AV nodal conduction. Furthermore, spontaneous atrial tachyarrhythmias were observed in isolated hearts from TG mice. Our data demonstrate that A1AR contribute to the physiological regulation of heart rate and AV nodal function in mice in vivo and suggest that an enhanced expression or function of A1AR may represent a potential pathomechanism by which A1AR could contribute to sinus node and AV nodal dysfunction in patients.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
-myosin heavy chain promotor) overexpression of A1AR
(23). The mice used for this
study had a 200- to 300-fold overexpression of A1AR
(23,
28). All measurements were
performed in age- and sex-matched littermate pairs of TG and wild-type (WT)
mice. The study conformed with the National Institutes of Health Guide for
the Care and Use of Laboratory Animals (NIH Publication No. 85-23,
Revised 1996). Surface ECG Measurements
Mice were sedated by intraperitoneal application of a mixture of ketamine (30 mg/kg body wt) and xylazine (10 mg/kg) at 3 mo of age. The animals were placed in a recumbent position, and gel-covered silver wire loops were attached around the four limbs of the animals to record a six-lead surface ECG. The signals were preamplified and displayed on paper (amplitude 20 mA/mm, paper speed 100 mm/s, Siemens Megacart; Erlangen, Germany).
Telemetric In Vivo ECG Recordings
Three-month-old mice were instrumented with a telemetric ECG transmitter (Data Science; Minneapolis, MN) in general anesthesia (50 mg ketamine-20 mg xylazine ip until pain reflexes were suppressed). The transmitter was inserted into the peritoneal sac, and the two ECG electrodes were placed subcutaneously in the region of the right shoulder and left leg to approximate lead II of the Einthoven surface ECG (18).
Telemetric ECG protocols. After a postoperative recovery period of 14 days, the telemetric ECG was first recorded for 1 h at rest during daytime. The mice were then placed in a water-filled tank for 5 min (tank size: 20 x 30 cm width and 15 cm depth, water temperature 34 ± 2°C). During that period, the mice had to swim to keep on the water surface. After this exercise challenge that resulted in submaximal physical exercise (4), the mice were put back in their normal cages. The ECG was continuously recorded 1 h before the swimming exercise, during swimming, and for 1 h after swimming using a custom-written data-acquisition software package based on the LabView programming language. Littermate pairs of mice were simultaneously subjected to the protocol.
In four pairs of mice, pertussis toxin (150 µg/kg body wt) was applied intraperitoneally to block pertussis toxin-sensitive G proteins, which mediate the A1AR signal (28). Thereafter, heart rate was continuously monitored for 72 h or until the death of the animals.
Isolated Heart Measurements
Isolated Langendorff-perfused mouse hearts were studied using previously published techniques (9). In brief, hearts of littermate WT and TG mice were rapidly excised via a median thoracotomy and perfused on a vertical Langendorff apparatus (Hugo Sachs Harvard Apparatus; March-Hugstetten, Germany) using a modified 37°C warm oxygenated Krebs-Henseleit solution containing (in mM) 118 NaCl, 24.88 NaHCO3, 41.2 KH2PO, 5.55 glucose, 2 Na-pyruvate, 0.83 MgSO4, 1.8 CaCl2, and 4.7 KCl. After an initial period of 10 min to allow the preparation to stabilize, the heart was perfused at a constant perfusion pressure of 90–100 mmHg, and the flow rate was measured. Thereafter, perfusion was kept at the previously measured flow rate (4 ± 1 ml/min). An octapolar catheter designed for murine electrophysiological measurements (CI'BER mouse, NuMed) was inserted into the right atrium and right ventricle. Two monophasic action potential catheters were placed onto the epicardium of the right and left atrium, and another one on the left ventricle. Catheters were mounted on a custom-made spring-loaded mechanism to ensure gentle and constant pressure and an orthogonal contact of the catheter tip with the epicardial surface. After placement of the catheters, the preparation was allowed to stabilize for 10 min. Thereafter, the spontaneous rhythm was observed for 5 min. The right atrium was paced at constant heart rates (80- to 140-ms pacing cycle length, corresponding to heart rates of 430–750 beats/min) via the octapolar catheter. At each paced heart rate, pacing was continued for 1.5 min to allow for measurement of steady-state action potential durations. Thereafter, right atrial extrastimulation for determination of effective refractory periods was performed. The pacing protocol was repeated during infusion of orciprenaline (0.74 µM continuous infusion) or adenosine [3 mg bolus (11 µmol) over 5 s, followed by a perfusion of 3 mg/l (11 µM)].
ECG Data Analysis
All analyses were performed blinded to genotype. Surface ECG recordings were manually analyzed for PQ, QRS, and QT intervals and heart rate. Telemetric Holter ECG recordings were analyzed for heart rate and PQ intervals during 5-min periods of normal activity (just before swimming), during the 5-min swimming period, and during the last 5 min of the hour after the end of the swimming exercise. All telemetric recordings were analyzed using a semiautomatic custom-written program for off-line analysis of murine ECG signals based on the LabView programming language. The program automatically determined RR intervals; all computed intervals were verified manually. PQ intervals were analyzed in ECG recordings of matched heart rate. Minimal, mean, and maximal heart rate and heart rate quantiles were calculated for each protocol part. For heart rate variability analysis, the SD of RR intervals was calculated over the entire 2-h telemetric ECG protocol, similar to protocols described by others (11). The entire recording period was manually reviewed for arrhythmias.
The recordings in the isolated hearts were analyzed for atrial heart rate; for the presence of higher-degree AV nodal block; for AV nodal conduction times, measured as the time from the latest atrial activation to the earliest ventricular activation; and for atrial and ventricular action potential duration.
Histology and Protein Biochemistry
The heart was excised after a rapid median thoracotomy. The atria were
immersed in formalin and stained using hematoxylin-eosin (H-E) and Sirius red
stains. For protein biochemistry, right and left atrial tissue samples were
shock frozen and prepared using standard methods
(26). For the purpose of this
study, we analyzed the protein levels of phospholamban, the -subunit of
Gi protein (Gi), and sarco(endo)plasmatic
reticulum Ca2+-ATPase (SERCA; SERCA antibodies were a
kind gift of Dr. L. R. Jones; Indianapolis, IN) in the atrial myocardium.
Doppler Echocardiography
Sedated (32 mg/kg body wt ketamine S-13 mg/kg body wt xylazine ip) pairs of TG and WT mice were studied using a Sonos 5500 echocardiography system equipped with a 15-MHz linear transducer and a 12-MHz Doppler transducer (Philips Medical Systems). In addition to conventional echocardiographic parameters, the left atrium was visualized in the parasternal long axis in the plane of the aortic root, and the atrial diameter was measured during early ventricular systole, i.e., at its maximal dimension. Atrial function was assessed using transmitral Doppler flow measurements.
Statistics
All functional analyses (echocardiography, ECG) were performed blinded to genotype. All variables were compared between genotypes by post hoc Student's t-test and ANOVA analyses using Microsoft Excel 2000 and SPSS (version 10) software packages. The incidence of AV nodal block and arrhythmias were compared using Fisher's exact test. Twosided P values <0.05 were considered significant. All values are reported as means ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Surface ECG parameters in sedated animals were not different between TG and WT mice (Table 1). Of note, neither bradycardia nor AV nodal block nor any sign of atrial or ventricular arrhythmia was noted under these experimental conditions. We use the term "bradycardia" in this paper to describe a relative bradycardia of TG mice compared with WT littermates. In our study, WT mice had heart rates comparable to murine heart rates reported by others. Therefore, the term bradycardia also implies that heart rates were lower than those usually observed in mice under the respective experimental conditions. Next, we recorded the ECG in freely moving animals (Fig. 1). Thereby, we could measure the ECG without the influence of anesthetic agents and furthermore assess the effect of exercise on the ECG. During exercise, we expected to observe the so-called "antiadrenergic effect" of A1AR in vivo.
|
|
Minimal heart rate was not different between WT and TG mice in any of the protocol parts, including exercise. Whereas the mean heart rate was not different during normal activity, it was lower during exercise and 1 h after exercise in TG mice. The maximal heart rate was lower in TG mice during all protocol parts (Fig. 2). Heart rate variability, usually regarded as a measure of the balance of sympathetic and parasympthetic tone in vivo, was reduced by 30% in TG mice (Fig. 3). The difference in the SD of RR intervals was still significant when the value was corrected for mean heart rate (data not shown).
|
|
In contrast to the findings in sedated animals, AV nodal conduction was slower in conscious TG mice (Fig. 4). Of note, AV nodal conduction slowing was present at rest and during exercise.
|
Effects of Pertussis Toxin
Pertussis toxin (150 µg/kg body wt) was given intraperitoneally to four pairs of mice to inhibit the function of Gi proteins. This dose was sufficient to block the negative inotropic effects of adenosine and carbachol in isolated left atrial preparations (27, 28). The time course of this effect is shown in Fig. 5. Forty-eight hours after the application of pertussis toxin, bradycardia was completely reversed in freely moving TG mice, whereas protracted AV nodal conduction persisted, albeit in a diminished fashion, compared with WT animals (Fig. 5).
|
Atrial and Ventricular Morphology and Function
Atrial conduction abnormalities and atrial arrhythmias could be due to increased size or altered morphology (fibrosis, necrosis) of the atria. To study these possibilities, the subsequent experiments were performed. Atrial weight (Table 2) and atrial size were not different between TG mice and controls (Table 3). Atrial function as assessed by mitral valve Doppler echocardiography was also not different between genotypes (Table 3). The histology of H-E-stained sections of atria from WT and TG hearts exhibited no difference (data not shown). In contrast to A3 adenosine receptor-overexpressing mice (unpublished observations), no fibrosis or necrosis was observed in TG atria (data not shown).
|
|
Calcium release from the sarcoplasmatic reticulum is the major determinant of cardiac contraction and an important modulator of pacemaker activity (21). To exclude the possibility that altered sarcoplasmatic Ca2+ handling leads to atrial dysfunction, we measured the expression of important Ca2+-regulatory proteins of the sarcoplasmatic reticulum. Their expression was not different between TG and WT atria (Table 2, right atrial measurements). Protein content was also not different in left atrial samples between genotypes, nor were there differences between the right and left atrial protein content (data not shown).
Electrophysiology in Isolated Hearts
To assess heart rate and AV nodal conduction in the intact heart deprived
of autonomic influences, we studied isolated Langendorff-perfused hearts of TG
(n = 7) and WT (n = 9) mice during spontaneous rhythm and
atrial pacing. Mean spontaneous heart rates were lower in hearts from TG mice
than in hearts from WT controls at baseline
(Fig. 6). Orciprenaline, a
-adrenoceptor agonist, accelerated the mean heart rate in TG and WT
hearts, but heart rate remained lower in TG hearts than in WT hearts (mean
heart rate during orciprenaline infusion: TG 479 ± 38 beats/min,
n = 7, vs. WT 646 ± 27 beats/min, n = 9, P
< 0.05). AV intervals were longer in TG hearts during atrial pacing,
consistent with first-degree AV nodal block
(Fig. 6). Despite the relative
bradycardia, third-degree AV nodal block was present spontaneously (n
= 1) or developed during orciprenaline infusion (n = 4) in five of
six TG hearts but in zero of nine control hearts (P < 0.01). The
addition of adenosine into the perfusate (3 mg over 5 s) caused transient AV
nodal block in four of six TG and three of nine WT hearts (P = not
significant). In three of six TG hearts, but in zero of nine control hearts,
nonsustained episodes of spontaneous atrial tachyarrhythmias were observed at
baseline (n = 2; Fig.
6) or after the administration of adenosine (n = 1,
P < 0.01).
|
To test whether A1AR overexpression alters cardiac repolarization, we also measured atrial and ventricular action potential durations (9). As expected, atrial action potential duration was shorter than ventricular action potential duration, and action potential duration decreased during pacing at high heart rates (Table 4). Neither atrial nor ventricular action potential durations were altered in isolated hearts from TG mice (Table 4). While unaltered action potential durations cannot rule out compensated changes in repolarizing currents, these data suggest that A1AR overexpression does not alter the time course of repolarization in epicardial working myocardium under basal conditions.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Overexpression of A1AR depresses the physiological chronotropic response to exercise and slows AV nodal conduction in vivo and in the isolated heart. These in vivo effects are observed under conditions of normal endogenous adenosine production (absence of ischemia). Furthermore, isolated hearts developed episodes of spontaneous atrial tachyarrhythmias at baseline or after challenge with adenosine. Taken together, this phenotype closely resembles sinus node dysfunction.
Heart Rate Response to Exercise
Maximal heart rate was lower in TG mice during all protocol parts,
suggesting that A1AR activation depresses the response to
-adrenoreceptor activation (the so-called antiadrenergic effect),
potentially by decreasing cAMP levels via Gi proteins or by
activating phosphatases (12,
13), thereby causing a
decreased mean heart rate during and after exercise. In small mammals, similar
to our findings, the mean in vivo heart rate is under control of sympathetic
activation. Furthermore, parasympathetic stimulation is probably not very
active in mouse hearts in vivo, as atropine appears to be without effect on
heart rate in freely moving mice (Fig.
5 in Ref. 35).
Conceivably, the maximum heart rate occurs during periods of sympathetic
stimulation in freely moving animals when, e.g., flight reflexes are
activated. During these periods, A1AR activation decreases the
heart rate via the antiadrenergic effect discussed above. These considerations
can explain the fact that maximal heart rates were lower during normal
activity in TG mice. A depressed heart rate increase in response to
sympathetic stimulation could also explain the reduced heart rate variability
in TG mice without increases in basal sympathetic tone. This is in contrast to
patients with heart failure: in these patients, decreased heart rate
variability is explained by increased basal sympathetic action and reduced
parasympathetic effects. An increased basal sympathetic tone could, however,
also be present in TG mice, as mean in vivo heart rates were not different
during normal activity (see Resting and Intrinsic Heart Rate).
Consistent with a G protein-dependent postreceptor signaling pathway, treatment with pertussis toxin, an irreversible blocker of G proteins (27, 28), completely abolished the difference in heart rate between TG and WT mice.
Resting and Intrinsic Heart Rate
Heart rate in the isolated heart reflects the intrinsic heart rate
(15). The mean intrinsic heart
rate was lower in isolated hearts and isolated right atrial preparations from
TG mice compared with littermate control mice
(13,
23,
27). Minimal heart rates, in
contrast, were not different between WT and TG mice in freely moving animals
(Fig. 2, left), nor
was heart rate different in sedated animals
(Table 1). These findings
indicate an intrinsic heart rate-lowering action of A1AR even in
the absence of catecholaminergic or direct sympathetic nerval stimulation.
This effect might be due to amplification of the negative chronotropic effect
of endogeneous adenosine by A1AR overexpression or might even be
due to tonic activation of steps down-stream of the A1AR due to
tense coupling of the nonoccupied ("empty") receptor to subsequent
G proteins in the signal transduction cascade, as discussed for the
2-adrenoceptor
(25).
AV Nodal Conduction Delay
Overexpression of A1AR caused AV nodal conduction delay in the
isolated heart and in vivo, an effect that is comparable to the acute effects
of A1AR agonists on AV nodal conduction in healthy human subjects
(20). In contrast to the heart
rate changes observed, AV nodal conduction was prolonged in freely moving TG
mice irrespective of exercise status, and pertussis toxin did not fully
reverse AV nodal conduction slowing. The negative dromotropic effect of
adenosine has been explained by A1AR activation of the inward
rectifier K+ current IK,ACh (also called
IK,ADo) and by inhibition of stimulatory -adrenergic
effects on the L-type Ca2+ current. However, adenosine
alone causes minimal to no effect on unstimulated, basal L-type
Ca2+ current in AV nodal cells, thereby rendering this
mechanism unlikely to cause AV nodal conduction delay, at least in isolated
(denervated) hearts. Furthermore, the negative dromotropic effect of adenosine
is not diminished in vivo in mice lacking IK,ACh
(35), so
IK,ACh is probably not involved in this pathway. In fact,
work in guinea pig and rabbit AV nodal cells indicates that adenosine mediates
its negative dromotropic effects at concentrations smaller than 2 µM (which
are probably reached in normoxia) rather via activation of a potassium current
with delayed rectifying properties [guinea pig AV nodal cells
(22)]. Taken together, these
findings may suggest that A1AR couple to different ion channels
with different efficiency, using different G proteins as signal transductors.
Furthermore, enhanced expression of G proteins in the AV node
(8) could help to explain the
weaker effect of pertussis toxin on AV nodal function than on heart rate in TG
mice in vivo.
Atrial Arrhythmias
When challenged with adenosine, hearts from TG mice developed nonsustained atrial arrhythmias. This observation resembles those made in patients with the so-called "tachycardia-bradycardia" variant of sinus node dysfunction. The mechanisms resulting in atrial tachyarrhythmias in TG hearts remain to be elucidated. Some data on the effect of bradycardia and adenosine on spontaneous depolarizations allow us, however, to suggest a hypothetical mechanism by which A1AR overexpression could result in atrial tachyarrhythmias: administration of exogeneous adenosine activates K+ channels via A1AR, thereby contributing to afterdepolarizations and triggered activity (3, 30, 36). Furthermore, bradycardia in itself can facilitate afterdepolarizations that cause tachycardias in a model of the long QT syndrome (10). Similarly, afterdepolarizations may have initiated episodes of atrial tachyarrhythmias in TG hearts in the present study. In addition, adenosine-induced atrial arrhythmias could also be maintained by activation of IK,ACh: inactivation of IK,ACh by functional deletion of one of its two constituent proteins (GIRK4) reduces inducibility of atrial arrhythmias in mice (17). Activation of IK,ACh by A1AR overexpression, in contrast, may decrease the threshold of atrial tachyarrhythmias. Further studies are needed to verify these potential mechanisms of tachyarryhthmia induction in TG hearts. Of note, we found no histological alterations (necrosis or fibrosis) that could explain the electrophysiological observations we report here.
Implications
Our data support the concept that the negative chronotropic effects of
adenosine during periods of enhanced catecholaminergic stimulation and the
negative dromotropic effects of adenosine are mediated by A1AR.
Altered activity of A1AR could contribute to the differences in
intrinsic heart rate recorded in vivo in the presence of atropine and
-adrenoreceptor blockers
(1,
15,
16). Further studies may
determine whether these findings in a mouse model with high-level
overexpression of A1AR can be reproduced in conditions that are
closer to human physiology.
Genetic causes for sinus node dysfunction have only been identified in a small proportion of affected patients. Although our observations have been made in TG mice, and not in humans, the search for genetic causes of bradycardia and AV nodal block may be extended to alterations in the A1AR gene, especially when a decreased chronotropic response to exercise is coupled with AV nodal conduction delay or paroxysmal atrial tachyarrhythmias.
During ischemia, adenosine levels are regionally and markedly increased in the heart (2, 5, 23, 29), resulting in an increased activation of adenosine receptors. Our findings support the concept that activation of adenosine receptors could cause reversible bradycardia and AV nodal block in patients with acute inferior myocardial infarction (33). Insufficient perfusion of the sinus nodal area could also chronically elevate adenosine levels and thereby contribute to chronic sinus nodal and AV nodal dysfunction (33, 34).
|
FOOTNOTES |
|---|
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.
* P. Kirchhof and L. Fabritz contributed equally to this study. 
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
-KPQ (LQT3) mice. Cardiovasc Res
57: 1085-1093,
2003.This article has been cited by other articles:
|
|
 |
|
  M. E. Mangoni and J. Nargeot Genesis and Regulation of the Heart Automaticity Physiol Rev, July 1, 2008; 88(3): 919 - 982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Lignon, Z. Bichler, B. Hivert, F. E. Gannier, P. Cosnay, J. A. del Rio, D. Migliore-Samour, and C. O. Malecot Altered heart rate control in transgenic mice carrying the KCNJ6 gene of the human chromosome 21 Physiol Genomics, April 21, 2008; 33(2): 230 - 239. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Knaus, X. Zong, N. Beetz, R. Jahns, M. J. Lohse, M. Biel, and L. Hein Direct Inhibition of Cardiac Hyperpolarization-Activated Cyclic Nucleotide-Gated Pacemaker Channels by Clonidine Circulation, February 20, 2007; 115(7): 872 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Funakoshi, T. O. Chan, J. C. Good, J. R. Libonati, J. Piuhola, X. Chen, S. M. MacDonnell, L. L. Lee, D. E. Herrmann, J. Zhang, et al. Regulated Overexpression of the A1-Adenosine Receptor in Mice Results in Adverse but Reversible Changes in Cardiac Morphology and Function Circulation, November 21, 2006; 114(21): 2240 - 2250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Marin and K. G. Franchini Reduced oxygen supply explains the negative force-frequency relation and the positive inotropic effect of adenosine in buffer-perfused hearts Am J Physiol Heart Circ Physiol, July 1, 2005; 289(1): H131 - H136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Kirchhefer, H. A. Baba, G. Hanske, L. R. Jones, P. Kirchhof, W. Schmitz, and J. Neumann Age-dependent biochemical and contractile properties in atrium of transgenic mice overexpressing junctin Am J Physiol Heart Circ Physiol, November 1, 2004; 287(5): H2216 - H2225. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
