AJP - Heart Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 290: H163-H171, 2006. First published August 26, 2005; doi:10.1152/ajpheart.01308.2004
0363-6135/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
290/1/H163    most recent
01308.2004v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Betsuyaku, T.
Right arrow Articles by Yamada, K. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Betsuyaku, T.
Right arrow Articles by Yamada, K. A.

Overexpression of cardiac connexin45 increases susceptibility to ventricular tachyarrhythmias in vivo

Tetsuo Betsuyaku,1 Nkiruka S. Nnebe,1 Rune Sundset,1,2 Sushmitha Patibandla,1 Charles M. Krueger,2 and Kathryn A. Yamada1,3

1Cardiovascular Division, Department of Medicine, 2Department of Pathology, and 3Center for Cardiovascular Research, Washington University School of Medicine, St. Louis, Missouri

Submitted 1 January 2005 ; accepted in final form 19 August 2005


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Electrophysiological remodeling involving gap junctions has been demonstrated in failing hearts and may contribute to intercellular uncoupling, delayed conduction, enhanced arrhythmias, and vulnerability to sudden death in patients with heart failure. Recently, we showed that failing human hearts exhibit marked increases in connexin45 (Cx45) expression in addition to previously documented decreases in connexin43 (Cx43) expression. Each of these changes results in reduced gap junction coupling. The objective of the present study was to examine functional consequences of increased Cx45 in cardiac gap junctions. Transgenic mice with cardiac-selective overexpression of the developmentally downregulated cardiac connexin, connexin45 (Cx45OE mice) were subjected to in vivo electrophysiology studies in which an intracardiac catheter was used to induce ventricular arrhythmias in anesthetized mice, and in which ambulatory ECG monitoring was used to detect spontaneous arrhythmias in unanesthetized mice. Hearts were analyzed by TaqMan RT-PCR, immunostaining, immunoblotting, and echocardiography. Lucifer yellow and neurobiotin dye transfer was used to assess coupling in transgenic and control myocyte cultures. Cx45 mRNA was two orders of magnitude greater in Cx45OE mice. Cx45-immunoreactive signal at gap junctions increased twofold and total Cx45 protein by immunoblotting increased 25% in Cx45OE mice compared with nontransgenic littermate controls. Functionally, Cx45OE mice exhibited more inducible ventricular tachycardia than controls but did not exhibit any other functional or structural derangements as assessed by echocardiography. Ventricular myocytes isolated from Cx45OE mice exhibited diminished intercellular transfer of Lucifer yellow dye and increased transfer of neurobiotin, consistent with altered cell-to-cell communication. Thus increased myocardial expression of Cx45 results in remodeling of intercellular coupling and greater susceptibility to ventricular arrhythmias in vivo.

arrhythmia; connexins; mouse; telemetry; ventricular tachycardia

HEART FAILURE is a devastating disease that is growing with the world's aging population. The prevalence in the United States alone is estimated at 5,000,000, with an annual incidence of 550,000 cases per year (4). Arrhythmia is a leading cause of death in heart failure. Very little is known, however, about the electrophysiological triggers or the substrates responsible for malignant arrhythmias in heart failure. Extensive electrophysiological remodeling has been demonstrated in the failing heart (5). Ion channel remodeling underlies action potential prolongation and is likely to contribute to arrhythmias initiated by early afterdepolarizations (5). Gap junction remodeling has also been shown to occur in failing hearts from a variety of etiologies (1, 12, 20, 22, 36, 41). It is unclear whether gap junction remodeling in the failing heart results in contractile dysfunction and/or arrhythmogenesis and, if so, what mechanisms are involved (34).

Gap junctions mediate cell-to-cell communication, which is critical during embryogenesis, growth, and development and for normal function of nearly every major organ system of the body. Normal cardiac conduction propagates through the ventricular myocardium via gap junction channels composed primarily of connexin43 (Cx43). Defects in Cx43 expression result in abnormal conduction (10, 15–17) and predilection for enhanced arrhythmogenesis in response to acute myocardial ischemia (27). Indeed, Cx43 remodeling underlies pathophysiological responses to myocardial stunning, infarction, and cardiomyopathy (20, 21, 32).

Cx43 is reduced in the failing heart (1, 12, 35, 41), and diminished coupling resulting from reduced Cx43 expression may contribute to a reentrant arrhythmogenic substrate. However, recent data from Cx43-inducible knockouts (13, 37) suggest that unless Cx43 is reduced by 70–90%, there is neither significant conduction delay nor arrhythmogenesis. Furthermore, Danik et al. (10) reported that heterogeneous loss of Cx43 in addition to substantial downregulation likely contributes to reentrant arrhythmogenesis. Therefore, the more modest reductions in Cx43 observed in most models of heart disease would not likely result in altered myocardial conduction and/or abnormal rhythm generation. Recently, a second cardiac connexin, Cx45, has been shown to be upregulated in human failing hearts (42). Cx45, the first connexin expressed in the heart, is essential for fetal development (24, 25). Subsequently, Cx45 is downregulated, such that it is expressed in much lower abundance than Cx43 in normal adult working ventricular myocytes (3, 24, 25). Although Cx45 is expressed in the conduction system of adult mammalian hearts (8, 9), its precise role in maintenance of cardiac rhythm remains under investigation (2).

The present study was performed to investigate the role of Cx45 in arrhythmogenesis by generating and characterizing transgenic mice overexpressing cardiac Cx45 (Cx45OE mice). We demonstrate here that Cx45OE mice exhibit enhanced ventricular arrhythmias in vivo without overt signs of cardiac structural abnormalities or cardiac contractile dysfunction. Upregulation of Cx45, particularly superimposed on downregulation of Cx43, is one mechanism whereby reduced coupling may result in increased arrhythmias in the failing heart.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Generation of Cx45OE mice. The coding sequence (nt 452–1642) of the mouse Cx45 gene (GenBank accession no. X63100, kindly provided by Dr. Eric Beyer) was cloned downstream from the {alpha}-myosin heavy chain promoter (GenBank accession no. U71441, kindly provided by Dr. Jeffrey Robbins) in pBluescript II SK(+). After sequencing, the construct was sent to the National Institute of Child Health and Human Development Transgenic Mouse Development Facility at the University of Alabama. Twenty-nine mice (C57BL/6J x SJL/J) were returned to us for screening. Mice were genotyped by PCR analysis of ear punch DNA using the following primers: 5'-TGCTCTCTTACCTTCCTCAC-3' (forward) and 5'-CCGGCTGCTCTGTGTTGC-3' (reverse), which generate a 669-bp product. Integration of the transgene and copy number were confirmed by Southern blot analysis of tail DNA from seven transgene-positive founders, with a transgene-specific human growth hormone polyadenylation sequence used as a probe. From the seven founders, we have maintained four transgene-positive lines and have backbred each line into the C57BL/6J background. Mice were housed in a barrier facility under supervision of the Department of Comparative Medicine at Washington University School of Medicine. All procedures were performed in compliance with the American Physiological Society's "Guiding Principles in the Care and Use of Animals"; all protocols were approved by the Washington University Animal Studies Committee.

Isolation of mRNA and TaqMan RT-PCR. Frozen ventricular tissue was homogenized in TRIzol (GIBCO, Invitrogen, Carlsbad, CA) by three 15-s bursts of a Tissue Tearor (BioSpec Products, Bartlesville, OK) at full speed. The homogenate was passed through a QIA shredder (Qiagen, Valencia, CA) and extracted in phenol-chloroform. Total RNA was subjected to Qiagen DNase digestion and purified with Qiagen RNeasy. RNA quantity and purity were assessed by measuring optical density (OD) at 260 nm and the ratio of OD at 260 nm to OD at 280 nm, respectively, using a Beckman DU640B spectrophotometer. RT-PCR (Applied Biosystems, Foster City, CA) was run on 500 ng of RNA per reaction, incubated at 42°C for 1 h and then 72°C for 10 min, and placed on ice for 1 min. TaqMan PCR (Applied Biosystems) was performed on each sample in triplicate. The forward (GAGCAAGCCCTATGCAATGC) and reverse (TCGTTGTCCTCCTCCGTTTC) primers and TaqMan probe (TCCAGAGCCCGGTGTTGTTTCCA, 5'-labeled with 6-carboxyfluorescein and 3'-labeled with 6-carboxytetramethylrhodamine) were synthesized by the Washington University Protein and Nucleic Acid Chemistry Laboratory. Rodent GAPDH was used as an internal control. The GAPDH primers and TaqMan probe (5'-labeled with VIC and 3'-labeled with 6-carboxytetramethylrhodamine) were obtained from Applied Biosystems. PCR was performed in a 96-well Microamp optical plate and analyzed on a sequencing detection system (Prism 7700, Applied Biosystems) for 40 cycles of 95°C/60°C.

Immunoblotting. Proteins in homogenates of frozen tissue (7 µg protein for Cx43 blots and 30 µg protein for Cx45 blots) were resolved by sodium dodecyl sulfate-polyacrylamide (10%) gel electrophoresis and transferred to nitrocellulose membranes for incubation with rabbit anti-Cx43 (1:5,000 dilution; Zymed Laboratories, South San Francisco, CA) or rabbit anti-Cx45 (1:500 dilution; kindly provided by Dr. Thomas Steinberg) antibody as described previously (18). Goat anti-rabbit (1:10,000 dilution; Jackson ImmunoResearch Laboratories, West Grove, PA) and donkey anti-rabbit (1:5,000 dilution; Amersham Biosciences, Piscataway, NJ) antibodies were used as secondary antibodies for Cx43 and Cx45 immunoblotting, respectively. Bands were quantified by densitometry using Adobe Photoshop. Each band was first normalized to its corresponding actin band on the gel; these "corrected" values were then normalized to control samples loaded on each gel.

Immunohistochemistry and confocal microscopy. Frozen tissue sections (10–15 µm) from Cx45OE [n = 25, 37 wk (SD 19)] and control [n = 31, 42 wk (SD 23)] hearts were incubated with anti-Cx45 antiserum (1:250 dilution) as described previously (18). This anti-Cx45 antibody has been validated extensively and shown to be monospecific (18, 38). Paraffin sections of formalin-fixed tissue were subjected to microwave antigen recovery (26) for anti-Cx43 staining using a commercial polyclonal rabbit anti-Cx43 antibody (1:500 dilution; Zymed Laboratories). Cy3-conjugated goat anti-rabbit antibody (1:400 dilution; Jackson ImmunoResearch Laboratories) was used as a secondary antibody as described previously (33). Controls and analysis were performed as described previously (33).

Intracardiac programmed electrical stimulation. Mice were anesthetized with ketamine (87 mg/kg)-xylazine (13 mg/kg) intraperitoneally and intubated with an 18-gauge needle for mechanical ventilation. ECGs were monitored via subcutaneously placed limb leads. A 2-Fr octapolar mouse electrophysiology catheter (CIB'er, NuMed, Hopkinton, NY) was introduced into the right jugular vein and advanced into the right ventricle through the tricuspid valve. Rectal temperature and blood oxygen saturation (model NPB-295, Nellcor Puritan Bennett, Pleasanton, CA) were monitored throughout each study. Mice with excessive bleeding or low oxygen saturation were not included in the data analysis. Right atrial pacing threshold, sinus node recovery time, sinoatrial conduction time, right atrial effective refractory period, atrioventricular (AV) effective refractory period, Wenckebach cycle length, right ventricular pacing threshold, right ventricular effective refractory period, and ventriculoatrial retrograde conduction were recorded (6). Programmed electrical stimulation was performed by investigators blinded to genotype to induce ventricular arrhythmias using a Bloom stimulator. A pacing train of eight stimuli (S1) was administered at a fixed cycle length. Single (S2), double (S2 and S3), and triple (S2, S3, and S4) premature extrastimuli (5 ms, 4 times threshold) were applied by sequential shortening of the coupling intervals until an arrhythmia was induced or until the tissue exhibited refractoriness. Burst pacing consisted of 16 S1 pacing stimuli; initial cycle length was 95 ms, with 5-ms decrements. The following arrhythmias were recorded: premature ventricular complexes (PVCs), premature ventricular couplets (couplets), nonsustained (3–10 beats) ventricular tachycardia (VT), and sustained (>10 beats) VT. Programmed electrical stimulation was performed at two or three sites (base, midway between base and apex, and apex) in each mouse. A total of 19 Cx45OE and 17 control [27 wk (SD 14)] mice completed the protocol, and data are reported here. Data from all four transgenic lines were pooled for analysis.

Ambulatory electrocardiography. Radio-frequency transmitters (PhysioTel TA10EA-F20 or TA10ETA-F20, Data Sciences International, Transoma Medical, Arden Hills, MN) were implanted in abdominal cavities of 14 Cx45OE and 6 control mice [48 wk (SD 11)] under intraperitoneal ketamine (80 mg/kg)-xylazine (16 mg/kg) anesthesia. Leads were tunneled under the skin, sutured, and glued to muscles in the thorax. Cathodal leads were placed on the upper right portion of the thorax, and anodal leads were placed on the chest wall near the apex. Telemetry was performed in the Mouse Physiology Core of the Center for Cardiovascular Research at Washington University School of Medicine 5 days after transmitter implantation on unanesthetized, unrestrained mice. Mice, individually caged, were placed on separate receivers (RPC-1). Lead II ECG signals were recorded (1 kHz) using Dataquest A.R.T. Gold 2.0 Acquisition software (Transoma) for 240 s every 30 min for 24–96 h. Q-T intervals were measured from the beginning of the QRS complex to the end of the T wave. Corrected Q-T interval (QTc) was calculated as Q-T/(R-R/100)1/2. P-R intervals were measured from the beginning of the P wave to the beginning of the QRS complex. Data from all four transgenic lines were pooled for analysis.

Morphology and echocardiography. Light and electron microscopy was performed on four Cx45OE and two control mice as described previously (28). Two-dimensional guided M-mode transthoracic echocardiograms were performed on 19 mice anesthetized with 2,2,2-tribromoethanol [Avertin, 0.01 ml (3%)/g body wt ip] using an Acuson Sequoia echocardiography system with a 13-MHz transducer in the Mouse Physiology Core of the Center for Cardiovascular Research at Washington University School of Medicine as described previously (7, 19). Images were analyzed offline by a single observer blinded to the genotype of the animals.

Dye transfer studies in neonatal ventricular myocytes. Neonatal ventricular myocytes were disaggregated in 0.1% trypsin (Sigma-Aldrich, St. Louis, MO) and cultured in Dulbecco's modified Eagle's medium (GIBCO) with 10% fetal calf serum as described previously (16). Briefly, hearts were removed and incubated in cold (4°C) trypsin for 18 h, incubated for 30 min at 37°C, gently triturated to disaggregate the tissue, filtered through a 35-µm cell strainer, centrifuged at 120 g, preplated for 2 h to reduce fibroblasts in the cell suspension, and, finally, plated in Dulbecco's modified Eagle's medium containing 20% heat-inactivated fetal calf serum (for the first 24 h only), 1% penicillin-streptomycin, and 0.1 mM 5-bromo-2'-deoxyuridine (Sigma). Genotyping by PCR was completed during the initial 18-h trypsin incubation. Myocytes isolated from three to four Cx45OE or control hearts were pooled and plated in 35-mm culture dishes coated with 0.25% laminin (Sigma). After 12–14 days in culture, individual cells were microinjected with Lucifer yellow dye (5% in 300 mmol/l LiCl; Sigma) using borosilicate glass capillary tubing (World Precision Instruments, Sarasota, FL) pulled on a pipette puller (model 750, David Kopf Instruments, Tujunga, CA) and a Picospritzer II (General Valve, Parker Hannifin, Fairfield, NJ). Cultures were fixed in 2% paraformaldehyde 7–9 min after injection. In a separate series of experiments, cells were microinjected with neurobiotin (4% in 150 mM LiCl; Vector Laboratories, Burlingame, CA). After fixation in paraformaldehyde, cells were solubilized with methanol-acetone (1:1) and then incubated with streptavidin-Cy3 (1:400 dilution; Sigma) for 2 h. Transfer of dye through gap junctions into neighboring cells was quantified by counting the number of fluorescent cells surrounding each injected cell.

Statistics. Values are means (SD). Two-group comparisons were made using Student's t-test for group data. Multiple-group comparisons were made using one-way analysis of variance and Tukey or Dunn's post hoc multiple-comparison tests. Fisher's exact test was used for nonparametric (inducible arrhythmia) data. P < 0.05 was considered statistically significant.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Expression and distribution of Cx45 in transgenic overexpressors. Four different lines, 901, 910, 927, and 929, were maintained from transgene-positive founders, each with similar (~5–9) copy number. Total RNA was extracted from 16 adult mouse hearts. Figure 1A shows an ~250-fold increase in mRNA in hearts from Cx45OE mice. Each of the four lines exhibited robust message levels. DNA extracted from Cx45OE mice exhibited a bright transgene band (Fig. 1B) used for genotyping. Cx45OE mice developed normally, were fertile, and aged comparably to their littermate controls. Their hearts exhibited no fibrosis or other light-microscopic or ultrastructural abnormalities (not shown). There was no difference in intercalated disk ultrastructure between Cx45OE and control hearts. Echocardiographic analyses of Cx45OE hearts in vivo demonstrated cardiac structure and function indistinguishable from transgene-negative control hearts (Fig. 1C, Table 1).



View larger version (74K):
[in this window]
[in a new window]
  		Fig. 1. Characterization of cardiac connexin45 (Cx45)-overexpressing (Cx45OE) mice. A: Cx45 mRNA levels determined by TaqMan from total RNA extracted from Cx45OE and transgene-negative control mice. Number of pairs (control and Cx45OE) of hearts is shown in bars. B: mouse genotyping by PCR amplification of DNA extracted from ear punch samples. Examples from 3 mouse lines (901, 910, and 927) are shown. DNA ladder, including markers at bp 2176, 1766, 1230, 1033, and 653 bp (top to bottom), are visible in lane 1. Cx45 transgene band runs at bp 669. –, Transgene-negative control; +, Cx45OE; C, H2O control. C: M-mode echocardiographic images of a control and a Cx45OE mouse demonstrating normal cardiac structure and systolic function.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Echocardiographic parameters

 
Marked increases in expression and distribution of Cx45 were observed in confocal images from left ventricular myocardium from Cx45OE mice (Fig. 2, A and B). Overall, there was a twofold increase (P < 0.001) in Cx45-immunoreactive signal concentrated at gap junctions, from 0.26 %cell area (SD 0.28) in controls to 0.81 %cell area (SD 0.63) in Cx45OE mice (Fig. 2C). The increase in Cx45 immunostaining was due to significant increases in the number and size of gap junctions containing Cx45 (Fig. 2D). One line of Cx45OE mice, the 927 line, exhibited a larger increase in Cx45 signal at gap junctions (Table 2), despite a similar number of gene insertions. Total Cx45 protein expression assessed by immunoblotting of ventricular homogenates was greater in Cx45OE mice (Fig. 2E). Comparable increases in Cx45 expression were observed in all lines. Summarized data show a significant (P = 0.006) 25% increase in Cx45 protein levels (Fig. 2F). The discrepancy between the twofold increase in Cx45-immunoreactive signal and the 25% increase in Cx45 protein expression by immunoblotting is similar to the discordance between Cx45 immunostaining and protein expression observed in Cx43-null mice (18).



View larger version (56K):
[in this window]
[in a new window]
  		Fig. 2. Cx45 expression in Cx45OE hearts. A: confocal image of midmyocardial tissue from a transgene-negative control mouse stained with anti-Cx45 antibody showing diffuse signal at intercellular junctions. Scale bar, 20 µm. B: confocal image of ventricular myocardium from a Cx45OE heart demonstrating robust overexpression of Cx45 at intercellular junctions. Scale bar, 20 µm. C: summarized data showing an overall 2-fold increase in Cx45-immunoreactive signal at intercellular junctions in Cx45OE compared with control mice. *P < 0.001. D: individual gap junction (object) number per field (P < 0.001) and size (P = 0.001) were significantly greater in Cx45OE than in control hearts. E: immunoblot of ventricular homogenates probed with anti-Cx45 antibody showing increased expression of Cx45 in Cx45OE lanes (+) compared with transgene-negative control lanes (–). F: summary of immunoblot data showing 25% increase in total Cx45 protein levels in Cx45OE hearts. *P = 0.006.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Cx45-immunoreactive signal in transgenic line

 
Cx43 immunostaining was nearly identical in ventricular tissue sections from Cx45OE [1.43 %cell area (SD 0.58), n = 13] and control [1.34 %cell area (SD 0.40), n = 15] hearts (Fig. 3, A and B), indicating that gap junction distribution of the major cardiac connexin was not altered in Cx45OE mice. There were no differences in the size or number of gap junctions containing Cx43, nor was there any discernable difference in Cx43 at nonjunctional membranes. Total expression levels of Cx43 protein by immunoblotting were reduced by 14% (P = 0.02) in ventricular homogenates from Cx45OE hearts (Fig. 3, C and D).



View larger version (58K):
[in this window]
[in a new window]
  		Fig. 3. Cx43 expression in Cx45OE mice. A: confocal image of control tissue stained with anti-Cx43 antibody showing typical gap junction staining. Scale bar, 20 µm. B: confocal image from a Cx45OE heart demonstrating Cx43 immunostaining at intercellular junctions comparable to control heart. Scale bar, 20 µm. C: immunoblot of ventricular homogenates probed with anti-Cx43 antibody showing modestly reduced expression of Cx43 in Cx45OE (+) compared with control (–) hearts. D: summary of immunoblot data showing a 14% decrease in Cx43 protein expression in Cx45OE hearts. *P = 0.02.

 
Enhanced arrhythmogenesis in Cx45OE mice in vivo. Programmed electrical stimulation revealed that Cx45OE mice were more susceptible to ventricular arrhythmias. The most specific, single-extrastimulus protocol induced ventricular arrhythmias in three Cx45OE mice (Fig. 4, A and B). None of the control mice were induced by a single extrastimulus. Four Cx45OE mice were inducible at multiple independent sites. Controls, when they were inducible at all, were inducible at only a single site. Finally, sustained VT was induced in three Cx45OE mice. The runs of sustained VT were 12 beats, 21 beats, and >4 s. None of the control mice exhibited sustained VT, even after double or triple extrastimuli or burst pacing. Together, these data indicate that Cx45OE mice are significantly (P = 0.003) more susceptible to inducible ventricular arrhythmias than the transgene-negative controls. Only by using the most aggressive stimulation protocols (including triple extrastimuli and burst pacing) could we equalize inducibility of PVCs and/or VT in Cx45OE (10 of 19) and control (7 of 17) mice, confirming previous findings that highly aggressive protocols can yield nonspecific inducibility (29). There was no difference in ventricular effective refractory periods between Cx45OE and control mice: 58 ms (SD 14) and 59 ms (SD 13), respectively. There were no differences in sinoatrial conduction time, sinus node recovery time, atrial and AV node refractory periods, or AV Wenckebach cycle time between Cx45OE and control mice.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 4. Ventricular tachycardia (VT) in Cx45OE mice in vivo. A: 6-beat run of nonsustained VT (NSVT) induced by a single extrastimulus (S2) in an anesthetized, closed-chest Cx45OE mouse. B: sustained VT induced by a single extrastimulus (S2) in an anesthetized, closed-chest Cx45OE mouse. Top trace, lead II of ECG; bottom trace, intracardiac electrogram recorded from atrioventricular junction (AVJ) showing atrial (A) and ventricular (V) electrograms. C: 2 runs of spontaneous nonsustained VT (horizontal bars) recorded by telemetry in a Cx45OE mouse.

 
Ambulatory ECGs recorded from conscious mice revealed spontaneous ventricular arrhythmias only in Cx45OE mice. Ventricular arrhythmias were not observed in any transgene-negative control mice. Four Cx45OE mice exhibited spontaneous ventricular arrhythmias. Each mouse exhibited premature ventricular couplets. In addition, two of the four mice exhibited runs of VT (Fig. 4C). The trend toward more spontaneous ventricular arrhythmias in Cx45OE mice did not reach statistical significance (P = 0.267) because of the low incidence of arrhythmias. Interestingly, three of the four mice exhibiting spontaneous arrhythmias were descendants of the 927 line of Cx45OE mice, which exhibited a larger increase in Cx45 signal at gap junctions (Table 2). PVCs were observed in 12 of 14 Cx45OE mice and 6 of 6 control mice, consistent with a baseline rate of PVCs exhibited by mice observed previously (7). Although Cx45OE and control mice exhibited a similar PVC rate [0.10 PVCs/h (SD 0.10) and 0.04 PVCs/h (SD 0.01), respectively], 5 of 14 Cx45OE mice had >10 PVCs, while none of the controls had >3 PVCs. There were no differences in heart rate, P-R interval, QRS duration, or QTc between Cx45OE and control mice (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Ambulatory ECG parameters

 
Overexpression of Cx45 alters myocyte gap junction communication. To test the hypothesis that increasing Cx45 in gap junctions diminishes coupling between ventricular myocytes, gap junction communication was measured in neonatal myocytes disaggregated from 1-day-old hearts of Cx45OE and control mice and grown in culture for 12–14 days. TaqMan results demonstrated that Cx45 mRNA taken at day 1 was significantly (P = 0.038) greater in Cx45OE [2.323 corrected arbitrary units (SD 2.394), n = 7] than in control [0.001 corrected arbitrary units (SD 0.001), n = 6] hearts. These levels of mRNA expression were comparable to those observed in 11-day-old Cx45OE [2.193 corrected arbitrary units (SD 0.810), n = 3] and control [0.004 corrected arbitrary units (SD 0.003), n = 5] hearts (P = 0.036).

Dye transfer studies were performed using Lucifer yellow (molecular weight = 457.2, charge = –2) and neurobiotin (molecular weight = 322.8, charge = +1). After Lucifer yellow dye injection, control myocytes (Fig. 5, A and C) were coupled to 32 (SD 10) cells (64 cells injected from 6 different litters). Myocytes from Cx45OE mice, however, were coupled to significantly (P = 0.011) fewer myocytes [17 cells (SD 8), 97 cells injected from 7 different litters; Fig. 5, B, D, and E], indicating reduced intercellular transfer of Lucifer yellow in myocytes from Cx45OE mice. Neurobiotin dye transfer was greater [32 cells (SD 21)] than Lucifer yellow transfer [17 cells (SD 8)] in Cx45OE cells, presumably because of the cationic sensitivity of Cx45 channels (39, 40). Both cationic dyes, neurobiotin and DAPI (4',6-diamidino-2-phenylindole), have been shown to permeate Cx45 channels more efficiently than anionic Lucifer yellow (14, 30).



View larger version (76K):
[in this window]
[in a new window]
  		Fig. 5. Reduced intercellular coupling in Cx45OE myocytes. A: bright-field view of a control ventricular myocyte monolayer. B: bright-field view of ventricular myocytes isolated from Cx45OE mice. *, Cells microinjected with Lucifer yellow. C and D: fluorescence micrographs of microscopic fields in A and B, respectively, showing spread of dye through gap junctions. E: summary of data depicting fewer neighboring cells to which dye spread in Cx45OE cultures after microinjection. *P = 0.011.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Our data support two important conclusions: 1) transgenic mice with cardiac-selective overexpression of Cx45 exhibit enhanced susceptibility to ventricular arrhythmias in vivo, and 2) overexpression of Cx45 in cardiac myocytes results in altered intercellular coupling. Cx45OE mice exhibited spontaneous runs of nonsustained and sustained VT on ambulatory ECG monitoring as well as sustained VT induced by programmed electrical stimulation in vivo. Recently, we reported increased levels of Cx45 expression in ventricular myocardium from patients with end-stage heart failure (42). We hypothesized that the increase in Cx45 expression in failing hearts results in reduced coupling and enhanced arrhythmogenesis. Our present data support this hypothesis.

Cx45OE mice did not exhibit cardiac contractile dysfunction or evidence of structural abnormalities (on the basis of echocardiography and light and electron microscopy) that might have served as fixed anatomic substrates for the genesis of arrhythmias. Increased vulnerability to ventricular arrhythmias in Cx45OE hearts without structural abnormalities may be a relevant model for studying mechanisms leading to lethal arrhythmias in heart failure patients with diffuse primary heart muscle disease without focal scars, who are at substantial risk of arrhythmic sudden death. Although mice are relatively resistant to ventricular tachyarrhythmias and there are limitations to studying arrhythmias in mice (31), our data indicate that Cx45OE mice are more vulnerable to spontaneous and inducible VT in vivo. Our data provide an important link between human studies demonstrating upregulation of Cx45 in failing hearts (42) and genesis of ventricullar tachyarrhythmias in hearts with upregulation of Cx45.

Cx45OE mice exhibited a significant increase in Cx45 expression in gap junctions. The twofold increase in Cx45-immunoreactive signal in confocal images from Cx45OE hearts bears striking similarity to that observed in human failing myocardium (42). Relative to Cx43, Cx45-immunoreactive signal is distributed over a small intercellular junctional area in normal hearts. However, when total gap junction area is reduced in Cx43-null hearts, the area over which Cx45 signal is distributed is dramatically reduced, even though there is no change in total Cx45 protein expression (18). In the present study, a 25% increase in total Cx45 protein (assessed by immunoblotting) was distributed over a threefold greater area at intercellular junctions (assessed by confocal microscopy) in Cx45OE hearts. This increased distribution was reflected by significant increases in the number and size of gap junctions containing Cx45. It is possible that Cx45 may be more readily recruited to gap junctions in Cx45OE than in control mice. It is also possible that intracellular expression of Cx45 accounts, in part, for the discrepancy between the immunostaining and immunoblotting data. Our algorithms for quantitative analysis of immunoreactive signal at gap junctions have been designed to exclude intracellular and background staining. Nevertheless, the increase in intercellular junctional signal observed in Cx45OE mice was accompanied by functional changes in myocyte coupling and enhanced arrhythmia susceptibility in vivo. Interestingly, the transgenic mouse line with the highest level of Cx45-immunoreactive signal in intercellular junctions was the most susceptible to development of spontaneous arrhythmias. These data suggest that the level of Cx45 in gap junctions and the stoichiometry of Cx45:Cx43 in gap junctions are critical for determining arrhythmogenic phenotypes.

Overexpression of Cx45 reduced cell-to-cell coupling in ventricular myocytes isolated from Cx45OE mice as assessed by Lucifer yellow dye transfer. We and others previously showed in heterologous expression systems that overexpression of Cx45 in cells normally expressing Cx43 significantly reduces intercellular coupling (23, 42). Indeed, addition of Cx45 per se in heterologous cell systems reduces gap junction conductance and intercellular coupling without changing Cx43 abundance, localization, or phosphorylation (23), thus producing a dominant-negative effect on intracellular coupling and electrical conductance. Dye transfer studies, although a reasonable surrogate for direct measurements of intercellular coupling, cannot replace dual voltage-clamp recordings for assessing electrical conductance. Indeed, our neurobiotin data support the observation that there is often little correlation between channel conductance and molecular permeability (14, 30, 40). For these reasons, we cannot conclude that reduced intercellular electrical coupling is responsible for the arrhythmogenic phenotype observed in the Cx45OE mice. Cx43 and Cx45 may form heteromeric channels with lower permeability. Alternatively, Cx45 channels may allow for transfer of signaling molecules that influence Cx43 permeability. Further studies are required to elucidate the mechanisms underlying the functional derangements induced by remodeling of Cx45 and Cx43 in gap junctions of diseased hearts.

One plausible mechanism for induction of lethal tachycardias in diseased hearts involves connexin remodeling, leading to reduced intercellular coupling, delayed conduction, and initiation of a reentrant circuit. Danik et al. (10) and Gutstein and colleagues (17) demonstrated that loss of Cx43 in the absence of structural cardiac abnormalities caused marked slowing of conduction velocity and abrupt onset of spontaneous ventricular tachyarrhythmias. Donoghue et al. (11) observed reduced Cx43 message and protein in angiotensin-converting enzyme-related carboxypeptidase-2-transgenic mice with conduction disturbances, sustained VT, and terminal ventricular fibrillation. However, on the basis of the level of conduction slowing observed in mice with modest (50–80%) reduction in Cx43, it is not clear whether the more modest reduction in Cx43 typically observed in diseased hearts is sufficient to result in enough conduction slowing for reentry to occur. Our data from Cx45OE mice indicate that upregulation of Cx45 imparts significant arrhythmogenic potential without significantly changing expression of Cx43 in gap junctions. On the basis of the modest conduction abnormalities observed in Cx43-deficient hearts (15), it is unlikely that Cx45OE mice would exhibit substantial conduction slowing. Although we cannot rule out the possibility that changes in connexin remodeling in Cx45OE mice may result in altered anisotropic properties of Cx45OE hearts, our data are consistent with a scenario in the failing heart in which upregulation of Cx45 combined with downregulation of Cx43 generates synergistic effects to reduce coupling and promote development of arrhythmias. Additional studies are required to determine whether heterogeneous or discontinuous conduction is observed in the Cx45OE hearts and whether uncoupling can be shown to underlie arrhythmogenesis in human ventricular myocardium with altered cardiac connexin expression.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by an American Heart Association Fellowship to T. Betsuyaku and National Heart, Lung, and Blood Institute Grant HL-66350 to K. A. Yamada.


   ACKNOWLEDGMENTS
 
The authors thank Karen G. Green, Evelyn M. Kanter, Colleen Isele, Erin M. Gribben, and William J. Kraft for expert technical help; Dr. Huilin Li for performing Southern blot analyses; Dr. Richard B. Schuessler for use of equipment and space for programmed electrical stimulation studies; and Dr. Jeffrey E. Saffitz for critical review of the manuscript and many helpful discussions.

Present addresses are as follows: T. Betsuyaku, National Nishi Sapporo Hospital, Sapporo, Japan; N. S. Nnebe, Penn State University, State College, PA; R. Sundset, Dept. of Medical Physiology, University of Tromsoe, Tromsoe, Norway; S. Patibandla, Clinical Cardiology Research, National Institutes of Health, Bethesda, MD; C. M. Krueger, Sigma Chemical, St. Louis, MO.


   FOOTNOTES
 

Address for reprint requests and other correspondence: K. A. Yamada, Cardiovascular Division, Box 8086, Washington Univ. School of Medicine, 660 South Euclid Ave., St. Louis, MO 63110 (e-mail: kyamada{at}wustl.edu)

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.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Akar FG, Spragg DD, Tunin RS, Kass DA, and Tomaselli GF. Mechanisms underlying conduction slowing and arrhythmogenesis in nonischemic dilated cardiomyopathy. Circ Res 95: 717–725, 2004.[Abstract/Free Full Text]
  2. Alcoléa S, Jarry-Guichard T, de Bakker J, Gonzàlez D, Lamers W, Coppen S, Barrio L, Jongsma H, Gros D, and van Rijen H. Replacement of connexin40 by connexin45 in the mouse. Impact on cardiac electrical conduction. Circ Res 94: 100–109, 2004.[Abstract/Free Full Text]
  3. Alcoléa S, Théveniau-Ruissy M, Jarry-Guichard T, Marics I, Tzouanacou E, Chauvin JP, Briand JP, Moorman AFM, Lamers WH, and Gros DB. Downregulation of connexin 45 gene products during mouse heart development. Circ Res 84: 1365–1379, 1999.[Abstract/Free Full Text]
  4. American Heart Association. Heart Disease and Stroke Statistics—2004 Update. Dallas, TX: Am. Heart Assoc., 2003.
  5. Armoundas AA, Wu R, Juang G, Marbán E, and Tomaselli GF. Electrical and structural remodeling of the failing ventricle. Pharmacol Ther 92: 213–230, 2001.[CrossRef][ISI][Medline]
  6. Berul CI, Aronovitz MJ, Wang PJ, and Mendelsohn ME. In vivo cardiac electrophysiology studies in the mouse. Circulation 94: 2641–2648, 1996.[Abstract/Free Full Text]
  7. Betsuyaku T, Kovacs A, Saffitz JE, and Yamada KA. Cardiac structure and function in young and senescent mice heterozygous for a connexin43 null mutation. J Mol Cell Cardiol 34: 175–184, 2002.[CrossRef][ISI][Medline]
  8. Coppen SR, Dupont E, Rothery S, and Severs NJ. Connexin45 expression is preferentially associated with the ventricular conduction system in mouse and rat heart. Circ Res 82: 232–243, 1998.[Abstract/Free Full Text]
  9. Coppen SR, Severs NJ, and Gourdie RG. Connexin 45 ({alpha}6) expression delineates an extended conduction system in the embryonic and mature rodent heart. Dev Genet 24: 82–90, 1999.[CrossRef][ISI][Medline]
  10. Danik SB, Liu F, Zhang J, Suk HJ, Morley GE, Fishman GI, and Gutstein DE. Modulation of cardiac gap junction expression and arrhythmic susceptibility. Circ Res 95: 1035–1041, 2004.[Abstract/Free Full Text]
  11. Donoghue M, Wakimoto H, Maguire CT, Acton S, Hales P, Stagliano N, Fairchild-Huntress V, Xu J, Lorenz JN, Kadambi V, Berul CI, and Breitbart RE. Heart block, ventricular tachycardia, and sudden death in ACE2 transgenic mice with downregulated connexins. J Mol Cell Cardiol 35: 1043–1053, 2003.[CrossRef][ISI][Medline]
  12. Dupont E, Matsushita T, Kaba RA, Vozzi C, Coppen SR, Khan N, Kaprielian R, Yacoub MH, and Severs NJ. Altered connexin expression in human congestive heart failure. J Mol Cell Cardiol 33: 359–371, 2001.[CrossRef][ISI][Medline]
  13. Eckardt D, Theis M, Degen J, Ott T, van Rijen HVM, Kirchhoff S, Kim JS, de Bakker JMT, and Willecke K. Functional role of connexin43 gap junction channels in adult mouse heart assessed by inducible gene deletion. J Mol Cell Cardiol 36: 101–110, 2004.[CrossRef][ISI][Medline]
  14. Elfgang C, Eckert R, Lichtengberg-Fraté H, Butterweck A, Traub O, Klein RA, Hülser DF, and Willecke K. Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells. J Cell Biol 129: 805–817, 1995.[Abstract/Free Full Text]
  15. Eloff BC, Lerner DL, Yamada KA, Schuessler RB, Saffitz JE, and Rosenbaum DS. High-resolution optical mapping reveals conduction slowing in connexin43 deficient mice. Cardiovasc Res 51: 681–690, 2001.[Abstract/Free Full Text]
  16. Guerrero PA, Schuessler RB, Davis LM, Beyer EC, Johnson CM, Yamada KA, and Saffitz JE. Slow ventricular conduction in mice heterozygous for a Cx43 null mutation. J Clin Invest 99: 1991–1998, 1997.[ISI][Medline]
  17. Gutstein DE, Morley GE, Tamaddon H, Vaidya D, Schneider MD, Chen J, Chien KR, Stuhlmann H, and Fishman GI. Conduction slowing and sudden arrhythmic death in mice with cardiac-restricted inactivation of connexin43. Circ Res 88: 333–339, 2001.[Abstract/Free Full Text]
  18. Johnson CM, Kanter EM, Green KG, Laing JG, Betsuyaku T, Beyer EC, Steinberg TH, Saffitz JE, and Yamada KA. Redistribution of connexin45 in gap junctions of connexin43-deficient hearts. Cardiovasc Res 53: 921–935, 2002.[Abstract/Free Full Text]
  19. Kanno S, Lerner DL, Schuessler RB, Betsuyaku T, Yamada KA, Saffitz JE, and Kovacs A. Echocardiographic evaluation of ventricular remodeling in a mouse model of myocardial infarction. J Am Soc Echocardiogr 15: 601–609, 2002.[CrossRef][ISI][Medline]
  20. Kaprielian RR, Gunning M, Dupont E, Sheppard MN, Rothery SM, Underwood R, Pennell DJ, Fox K, Pepper J, Poole-Wilson PA, and Severs NJ. Downregulation of immunodetectable connexin43 and decreased gap junction size in the pathogenesis of chronic hibernation in the human left ventricle. Circulation 97: 651–660, 1998.[Abstract/Free Full Text]
  21. Kitamura H, Ohnishi Y, Yoshida A, Okajima K, Azumi H, Ishida A, Galeano EJ, Kubo S, Hayashi Y, Itoh H, and Yokoyama M. Heterogeneous loss of connexin43 protein in nonischemic dilated cardiomyopathy with ventricular tachycardia. J Cardiovasc Electrophysiol 13: 865–870, 2002.[CrossRef][ISI][Medline]
  22. Kostin S, Dammer S, Hein S, Klovekorn WP, Bauer EPO, and Schaper J. Connexin 43 expression and distribution in compensated and decompensated cardiac hypertrophy in patients with aortic stenosis. Cardiovasc Res 62: 426–436, 2004.[Abstract/Free Full Text]
  23. Koval M, Geist ST, Westphale EM, Kemendy EM, Civitelli R, Beyer EC, and Steinberg TH. Transfected connexin45 alters gap junction permeability in cells expressing endogenous connexin43. J Cell Biol 130: 987–995, 1995.[Abstract/Free Full Text]
  24. Krüger O, Plum A, Kim JS, Winterhager E, Maxeiner S, Hallas G, Kirchhoff S, Traub O, Lamers WH, and Willecke K. Defective vascular development in connexin 45-deficient mice. Development 127: 4179–4193, 2000.[Abstract]
  25. Kumai M, Nishii K, Nakamura K, Takeda N, Suzuki M, and Shibata Y. Loss of connexin45 causes a cushion defect in early cardiogenesis. Development 127: 3501–3512, 2000.[Abstract]
  26. Kwong KF, Schuessler RB, Green KG, Boineau JP, and Saffitz JE. Differential expression of gap junction proteins in the canine sinus node. Circ Res 82: 604–612, 1998.[Abstract/Free Full Text]
  27. Lerner DL, Yamada KA, Schuessler RB, and Saffitz JE. Accelerated onset and increased incidence of ventricular arrhythmias induced by ischemia in Cx43-deficient mice. Circulation 101: 547–552, 2000.[Abstract/Free Full Text]
  28. Luke RA, Beyer EC, Hoyt RH, and Saffitz JE. Quantitative analysis of intercellular connections by immunohistochemistry of the cardiac gap junction protein, connexin43. Circ Res 65: 1450–1457, 1989.[Abstract/Free Full Text]
  29. Maguire CT, Wakimoto H, Patel VV, Hammer PE, Gauvreau K, and Berul CI. Implications of ventricular arrhythmia vulnerability during murine electrophysiology studies. Physiol Genomics 15: 84–91, 2003.[Abstract/Free Full Text]
  30. Martinez AD, Hayrapetyan V, Moreno AP, and Beyer EC. Connexin43 and connexin45 form heteromeric gap junction channels in which individual components determine permeability and regulation. Circ Res 90: 1100–1107, 2002.[Abstract/Free Full Text]
  31. Nerbonne JM. Studying cardiac arrhythmias in the mouse—a reasonable model for probing mechanisms? Trends Cardiovasc Med 14: 83–93, 2004.[CrossRef][ISI][Medline]
  32. Peters NS, Coromilas J, Severs NJ, and Wit AL. Disturbed connexin43 gap junction distribution correlates with the location of reentrant circuits in the epicardial border zone of healing canine infarcts that cause ventricular tachycardia. Circulation 95: 988–996, 1997.[Abstract/Free Full Text]
  33. Saffitz JE, Green KG, Kraft WJ, Schechtman KB, and Yamada KA. Effects of diminished expression of connexin43 on gap junction number and size in ventricular myocardium. Am J Physiol Heart Circ Physiol 278: H1662–H1670, 2000.[Abstract/Free Full Text]
  34. Saffitz JE and Yamada KA. Do alterations in intercellular coupling play a role in cardiac contractile dysfunction? Circulation 97: 630–632, 1998.[Free Full Text]
  35. Severs NJ. Gap junction alterations in the failing heart. Eur Heart J 15 Suppl D: 53–57, 1994.
  36. Severs NJ. Gap junction remodeling in heart failure. J Card Fail 8: S293–S299, 2004.
  37. Van Rijen HVM, Eckardt D, Degen J, Theis M, Ott T, Willecke K, Jongsma HJ, Opthof T, and de Bakker JMT. Slow conduction and enhanced anisotropy increase the propensity for ventricular tachyarrhythmias in adult mice with induced deletion of connexin43. Circulation 109: 1048–1055, 2004.[Abstract/Free Full Text]
  38. Van Veen TAB, van Rijen HVM, and Jongsma HJ. Electrical conductance of mouse connexin45 gap junction channels is modulated by phosphorylation. Cardiovasc Res 46: 496–510, 2000.[Abstract/Free Full Text]
  39. Veenstra RD, Wang HZ, Beyer EC, and Brink PR. Selective dye and ionic permeability of gap junction channels formed by connexin45. Circ Res 75: 483–490, 1994.[Abstract/Free Full Text]
  40. Veenstra RD, Wang HZ, Beblo DA, Chilton MG, Harris AL, Beyer EC, and Brink PR. Selectivity of connexin-specific gap junctions does not correlate with channel conductance. Circ Res 77: 1156–1165, 1995.[Abstract/Free Full Text]
  41. Wang X and Gerdes AM. Chronic pressure overload cardiac hypertrophy and failure in guinea pigs. III. Intercalated disc remodeling. J Mol Cell Cardiol 31: 333–343, 1999.[CrossRef][ISI][Medline]
  42. Yamada KA, Rogers JG, Sundset R, Steinberg TH, and Saffitz JE. Up-regulation of connexin45 in heart failure. J Cardiovasc Electrophysiol 14: 1205–1212, 2003.[CrossRef][ISI][Medline]



This article has been cited by other articles:


Home page
 Cardiovasc ResHome page
N. J. Severs, A. F. Bruce, E. Dupont, and S. Rothery
Remodelling of gap junctions and connexin expression in diseased myocardium
Cardiovasc Res, October 1, 2008; 80(1): 9 - 19.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
290/1/H163    most recent
01308.2004v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Betsuyaku, T.
Right arrow Articles by Yamada, K. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Betsuyaku, T.
Right arrow Articles by Yamada, K. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.