Skip Navigation

Cardiovascular Research 2004 63(4):700-708; doi:10.1016/j.cardiores.2004.05.017
© 2004 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Fransen, P.
Right arrow Articles by De Keulenaer, G. W
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fransen, P.
Right arrow Articles by De Keulenaer, G. W
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2004, European Society of Cardiology

Endocardial endothelium modulates subendocardial pHi of rabbit papillary muscles: role of transendothelial HCO3– transport

Paul Fransen*, Regis R Lamberts1, Jan Hendrickx and Gilles W De Keulenaer

Department of Pharmacology, University of Antwerp, Groenenborgerlaan 171, Antwerp B-2020, Belgium

* Corresponding author. Tel.: +32-3-820-2587; fax: +32-3-265-3276. Email address: paul.fransen{at}ua.ac.be lamberts{at}physiol.med.vu.nl

Received 21 November 2003; revised 8 April 2004; accepted 24 May 2004


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objective: This study investigated whether endocardial endothelial cells contribute to intracellular pH (pHi) regulation of subjacent cardiomyocytes. Methods: Right ventricular rabbit papillary muscles were loaded with the pH-sensitive dye 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester (BCECF-AM) to measure pHi and HCO3– equivalent influx or efflux in muscles with intact endocardial endothelium (+EE) and in muscles with the endocardial endothelium removed (–EE). Results: In steady-state conditions, pHi was consistently higher in +EE than in –EE muscles (7.38±0.03, n=39 vs. 7.27±0.04, n=20, p<0.05). In +EE muscles, removal of HCO3– from the buffer solution or adding 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS), an inhibitor of HCO3– transport, reduced pHi from 7.38 to 7.24±0.06 (n=14) and to 7.16±0.10 (n=10), respectively, whereas in –EE muscles pHi decreased slightly from 7.27 to 7.15±0.05 (n=14) and to 7.13±0.04 (n=7). In addition, HCO3– equivalent efflux during alkali loads by NH4Cl pulses was smaller in +EE muscles than in –EE muscles (0.89±0.50 vs. 1.99±0.12 mmol/l/min, n=5, p<0.05) and was inhibited by DIDS. HCO3– equivalent influx during recovery from acid load imposed upon wash out of NH4Cl, was larger in +EE muscles than in –EE muscles (2.15±0.54 mmol/l/min, n=6, vs. 1.06±0.20 mmol/l/min, n=5). 5-(N-Ethyl-N-isopropyl)amiloride (EIPA), an inhibitor of Na+/H+ exchange, decreased HCO3– equivalent influx by 50% in both muscle groups but influx was still significantly higher in +EE than in –EE muscles (1.18±0.23 vs. 0.57±0.07 mmol/l/min, p<0.05). Finally, endocardial endothelial cells cultured on collagen-coated inserts established a DIDS-sensitive transendothelial HCO3–gradient. Conclusion: These data suggest that the endocardial endothelium maintains transendothelial fluxes of HCO3– from luminal (blood) to basal (muscle) side of the cells, which modulate pHi regulation in subjacent cardiomyocytes.

KEYWORDS Endothelial function; Acidosis; Ion exchangers; Ion transport; Na/H exchanger


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
Cardiac endothelial cells cross-interact with subjacent cardiomyocytes both in the foetal and in the adult heart, mostly through paracrine cardio-active substances, thereby modulating growth, rhythmicity and contractility of cardiomyocytes [1]. Using a perfused intact multicellular preparation, Muller-Borer et al. [2] speculated that the endocardial endothelium also participated in the regulation of intra- and extracellular pH of cardiomyocytes by creating pH gradients from endocardium to deeper muscle layers. Direct evidence for an endocardial endothelial control of pH of its subjacent muscle environment is, however, lacking. Based on previous observations, we speculated that the endocardial endothelium could function as an active physicochemical barrier that creates transendothelial ion fluxes and gradients [3]. Observations in favour of such a ‘blood–heart’ barrier included specific features such as the extensive intercellular overlap with abundant gap junctions at the endocardial endothelial cell borders as well as an asymmetrical localisation of ion channels [4] and of Na+/K+-ATPase [3] in endocardial endothelial cells.

Strict control of intracellular pH (pHi) in cardiomyocytes is essential for the mechanical performance, excitability and electrical conduction of the heart. Control of pHi in cardiomyocytes depends on chemical buffering and on the activity of membrane ionic transporters including acid extruders (Na+/H+ exchanger and Na+/HCO3– cotransporter) and acid loaders (Cl–/HCO3– exchanger and Cl–/OH– exchanger or Cl–/H+ cotransporter) [5,6]. Apart from these cellular pH-regulating mechanisms, however, previous observations have suggested that cell-to-cell interactions may also participate in the regulation of pHi of cardiomyocytes. Indeed, measurements of pHi of cardiomyocytes in various buffer systems have revealed different results when single cardiomyocytes were studied after they had been removed from the organ, as compared to intact multicellular preparations with cell-to-cell interactions present [6–15].

Based on these observations, we investigated in isolated intact cardiac muscles, whether endocardial endothelial cells affect pHi of cardiomyocytes, and whether transendothelial HCO3–/H+ fluxes are involved.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1. Isolated cardiac muscle preparation
Right ventricular papillary muscles were dissected from hearts of New Zealand white rabbits (n=85) and mounted in modified Krebs–Ringer solution (KR) as described previously [3]. The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Muscles were mounted horizontally on top of a NIKON Diaphot inverted microscope in a 5-ml organ bath with a glass coverslip as bottom, stimulated at 36 contractions/min and superfused at constant flow of 3.5 ml/min with KR, pre-gassed with a mixture of 95% O2–5% CO2 (35.0±0.5 °C, 7.4±0.1 pH, 320 mOsm osmolarity). KR contained (mmol/l): 128 NaCl, 4.7 KCl, 2.4 MgSO4·7H2O, 20 NaHCO3, 1.2 KH2PO4, 9 glucose, 1.8 CaCl2 and 0.02 atenolol. To work in nominally HCO3–-free conditions, KR was replaced by HEPES-buffered solution (pH=7.4±0.1 with 1 M NaOH) with the same composition except for HEPES 20 mmol/l instead of NaHCO3.

2.2. Endocardial endothelial damage
Experiments were divided into two groups. In one group, the endocardial endothelium of muscles was selectively destroyed upon dissection of muscles from the heart by immersion of the muscles for 1 s in 0.5% Triton X-100, followed by abundant wash with Triton-free solution (–EE muscles). This method has been validated as for its completeness and selectivity of endocardial endothelial damage with concomitant preservation of underlying myocardium [16]. Subsequently, these muscles were treated in the same way as muscles from the second group in which the endocardial endothelium was left intact (+EE muscles). To verify preservation of an intact endocardial endothelium in +EE muscles, at random immunostaining for platelet endothelial cell adhesion molecule (PECAM) was performed. In agreement with previous observations, stainings consistently showed that more than 85% of the endothelial surface was intact [17].

2.3 Measurement of pHi
Muscles were loaded for 20–30 min with the membrane permeable acetoxymethyl ester form of the pH sensitive indicator 2',7'-bis-(2-carboxyethyl)-5,6-carboyfluorescein (BCECF-AM, Molecular Probes) (2.5 µmol/l in HEPES-buffered solution with 0.02% Pluronic-F127 and 3 mmol/l 2,3-butanedionemonoxime). Excitation wavelengths (485 and 445 nm) were delivered at frequencies of 1, 0.67 or 0.5 Hz with a DeltaRAM Multiwavelength Illuminator (Photon Technology International, PTI, USA). Emission light intensity (dichroic mirror, emission filter PO# OROM9712127, WO#A018983, Omega Optical, USA) was measured with a microscope photometer (D-104, PTI). The single emission at dual excitation (485/445 ratio) was analysed with Felix software (PTI). Before starting experiments, horizontal and vertical aperture of the microscope photometer was set on part of the surface of the muscle (about 50% of the two-dimensional flat plane) and the mean background emission values for excitation at 485 and 445 nm were determined over a period of 120–180 s. These background values were real-time subtracted from the emission values during experiments. At the end of a number of experiments, fluorescence emission ratio was calibrated in vitro by the high K+-nigericin method [18]. There were no significant differences between calibration curves in +EE and –EE muscles (data not shown).

To investigate the amount of BCECF loaded to the muscle, transverse cryosections were made of six papillary muscles, which were treated in the same way as the experimental muscles. The sections were co-stained with propidiumiodide to visualise the nuclei (1.5 µg/ml, H-1300, Vectashield Mounting Medium, Vector). Images were obtained with an Olympus fluorescence microscope (BX40) equipped with a CCD camera (Sensicam, PCO, Germany).

2.4. Experimental protocols
To study HCO3– transport in rabbit papillary muscles, the following experimental protocols were applied: (1) measurement of pHi changes during removal of HCO3– buffer (replacement by Hepes buffer); (2) measurement of pHi changes by incubating muscles for 20–30 min with 100 µM 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) and evaluation of HCO3– equivalent efflux during recovery from alkali loads installed by addition of 10 mM NH4Cl; (3) measurement of pHi changes by incubating muscles for 20–30 min with 5 µM 5-(N-ethyl-N-isopropyl)amiloride (EIPA) and evaluation of HCO3– equivalent influx during recovery from acid loads installed by wash out of 10 mM NH4Cl; (4) evaluation of HCO3– influx/efflux during recovery from alkalinisation and acidification (vehicle controls of (2) and (3)). All measurements of pHi in +EE or –EE muscles were performed in different muscles.

2.5 Determination of HCO3– equivalent efflux and influx during NH4Cl pulses (Fig. 1)
During a 300-s pulse with 10 mmol/l NH4Cl, the initial alkalinisation of pHi gradually decreased due to efflux of HCO3– via activity of Cl–/HCO3– exchange (recovery from alkali load [19,20]). The slope of this decrease determines the rate of HCO3– equivalent efflux (HCO3– equivalent efflux=slope*total buffering power). Apparent intracellular buffering power (βt) was measured from the initial change of pHi after an alkaline load imposed by 10 mmol/l NH4Cl in the different solutions. Addition of NH4Cl caused a substantial increase of pHi, which was significantly larger in HEPES-buffered solutions than in HCO3–-buffered solutions. In HEPES (absence of HCO3– and CO2), pHi increased from 7.24±0.08 to 7.98±0.07 and 7.29±0.05 to 7.84±0.11 in +EE and –EE muscles, respectively. βt equals the apparent intrinsic buffering power (βi) or {Delta}[NH4+]i/{Delta}pHi and was 18.2±2.4 mmol/l/pH in +EE muscles (n=5) and 20.8±2.7 mmol/l/pH in –EE muscles (n=5, p>0.05). In the presence of HCO3– and CO2, pHi increased from 7.40±0.07 to 7.80±0.13 and from 7.30±0.03 to 7.63±0.06 in +EE and –EE muscles, respectively. Here, apparent βt (=βi+βCO2, with βCO2, the buffering power caused by intracellular CO2 and HCO3–) was 56.1±6.9 mmol/l/pH in +EE muscles (n=5) and 59.4±4.5 mmol/l/pH in –EE muscles (n=5, p>0.05) and was significantly larger than in HEPES-buffered solution (p<0.001). In HCO3– solution, addition of 100 µM DIDS (n=5) or 5 µM EIPA (n=5) increased apparent βt to respectively 91.4±26.5 and 122.7±22.9 mmol/l/pH in +EE muscles and 113.0±12.4 and 135.0±20.0 mmol/l/pH in –EE muscles. There were no significant differences between apparent βt in +EE and –EE muscles in any condition (Fig. 1).


Figure 1
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Determination of HCO3– equivalent efflux and influx during and following NH4Cl pulses. At time 0 s, 10 mmol/l NH4Cl was applied, leading to alkalinisation of pHi. Recovery from alkalinisation occurs via activity of Cl–/HCO3– exchange mainly (dotted line). After 300 s, wash out of NH4Cl imposes an acid load, which is thereafter removed by activity of Na+/H+ exchange and Na+/HCO3– cotransport (dotted line).

 
After 300 s, wash out of NH4Cl imposed an acid load by trapping protons into the muscle cells. Subsequently, muscles recovered from this acid load by the activity of Na+/HCO3– cotransport (HCO3– influx) and Na+/H+ exchange (H+ efflux) [21]. The slope of pHi recovery determines the rate of HCO3– equivalent influx (H+ efflux) (HCO3– equivalent influx=slope*βt).

2.6 Cell culture inserts and determination of luminal and basal HCO3– concentration
Endocardial endothelial cells were isolated from the right ventricle of porcine hearts or from both ventricles of rat and rabbit hearts. The cells were cultured as described previously [3,22]. Rabbit and porcine endocardial endothelial cells were cultured in Medium 199 plus 20% foetal bovine serum, whereas rat endocardial endothelial cells were cultured in DMEM plus 10% foetal bovine serum. First passage cells were seeded on collagen-I-treated inserts (0.45 µm pore diameter, Becton Dickinson, USA) and kept in culture for 24, 48 or 72 h. Luminal and basal media were collected and the concentration of HCO3– in 100 µl samples was determined with Vitros 950 (OCD, Ortho Clinical Diagnostics, USA).

2.7. Statistics
All data are expressed as mean±S.E.M. Repeated measure analyses of variance and two-tailed unpaired or paired t-tests were performed on the raw data (after logarithmic transformation for homoscedasticity) in order to analyse the effects of buffer solution, pharmacological substance and the influences of an intact endocardial endothelium (GraphPad PRISM). Differences were considered statistically significant when p<0.05.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1. Cell types involved in the global BCECF signal
The relative contribution of endocardial endothelial cells to the total BCECF fluorescence signal measured in +EE muscles was determined by comparing signal intensity before and after removal (cotton rubbing) of endocardial endothelial cells. In seven muscles, BCECF emission decreased with 16.0±4.4% for excitation at 485 nm and with 17.4±5.3% at 435 nm. Accordingly, about 15% of the BCECF signal in +EE muscles was derived from the endocardial endothelium and 85% from the underlying muscle. Consistently, transverse cryosections through BCECF-loaded papillary muscles (n=6, Fig. 2) showed a 15–20-µm-thick layer of stained subendocardial cardiomyocytes, which largely outnumbered the mass of stained endothelial cells in the endocardium.


Figure 2
View larger version (47K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Transverse cryosections through a rabbit papillary muscle, which was loaded for 30 min with 2.5 µM BCECF-AM. Cell nuclei were stained with propidiumiodide (A), whereas B shows BCECF distribution (5 x objective (left) and 40 x objective (right)). BCECF-stained endocardial endothelium (<1 µm) and subjacent cardiomyocytes (20–25 µm).

 
3.2 pHi in +EE and –EE muscles
In HCO3– conditions, basal pHi of +EE muscles was significantly more alkaline than pHi of –EE muscles (7.38±0.03, n=39 vs. 7.27±0.04, n=20, p<0.05). In the absence of HCO3– (HEPES buffer), basal pHi of +EE muscles was significantly lower (7.24±0.06, n=14, p<0.05), whereas pHi of –EE muscles was only slightly lower (7.15±0.06, n=14, p=0.08). In addition, switching physiological buffer solution to a HEPES buffer ("removal of HCO3–") in +EE muscles acidified pHi with –0.17±0.03 pHi (n=9), but only minimally affected pHi in –EE muscles (–0.05±0.04, n=8, p<0.01, Fig. 3A). Accordingly, in HEPES conditions, the pHi difference between +EE and –EE muscles was blunted (p=0.28).


Figure 3
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 (A) Net acidification ({Delta}pHi) by removal of HCO3– buffer (replaced by HEPES buffer) in +EE (n=9) and –EE muscles (n=8). (B) Net acidification of +EE (n=10) and –EE (n=6) muscles in HCO3– buffer following inhibition of Cl–/HCO3– exchange and Na+/HCO3– cotransport with 100 µM DIDS (added 30 min before measurement). Note the larger acidifying effects by HCO3– removal or by DIDS in +EE muscles. (#p<0.05: +EE vs. –EE).

 
To verify these results, DIDS, an inhibitor of HCO3– transporters, was added to a different group of +EE and –EE muscles in HCO3–-buffered solutions. Similarly as for removal of HCO3–, inhibition of HCO3– transport by 100 µM DIDS induced a robust acidification of +EE muscles from 7.44±0.06 to 7.16±0.10 (n=10, p<0.05), but only slightly reduced pHi in –EE muscles (from 7.25±0.02 to 7.13±0.04, n=7, p<0.005) (Fig. 3B). Accordingly, inhibition of HCO3– transport abolished the pHi difference between +EE and –EE muscles (p=0.91).

3.3 pHi in +EE and –EE muscles during NH4Cl pulses
Above results suggest that pHi in papillary muscles is dependent on endocardial endothelium in conditions where HCO3– fluxes are allowed, and suggest an endothelium-dependent transport of HCO3– from the luminal (bathing solution) to the basal (muscle) side of the endothelium. To validate these results and to further characterize HCO3– fluxes, experiments were performed with the NH4Cl pulse technique, during which HCO3– fluxes were analysed.

Fig. 4 shows representative examples of pHi changes induced by 10 mmol/l NH4Cl. As explained in Section 2, the downward slope of the recovery from an alkali load following the initial pHi increase is attributed to HCO3– equivalent efflux due to activity of Cl–/HCO3– exchange [19,20]. The speed of recovery from an alkali load (steepness of downward slope and thus rate of HCO3– efflux, black broken lines in Fig. 4A and B) was consistently smaller in +EE muscles than in –EE muscles. Consistent with the premise that this slope represents HCO3– flux, the slope of this curve was diminished following inhibition of Cl–/HCO3– exchange with DIDS (grey broken lines in Fig. 4A and B). Fig. 5 summarises the mean data of the experimental groups. HCO3– efflux was 0.89±0.50 mmol/l/min in +EE muscles (n=5), which was significantly smaller than HCO3– efflux of 1.99±0.12 mmol/l/min in –EE muscles (n=5, p<0.05). As further shown in Fig. 5, 100 µM DIDS robustly attenuated HCO3– efflux in +EE and –EE muscles and abolished the difference in efflux between both muscle groups. These results are consistent with the hypothesis that efflux of HCO3– during an alkaline load in +EE muscles is attenuated by the presence of endocardial endothelial cells and suggest a transendothelial influx of HCO3–.


Figure 4
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 pHi of a +EE (A) and a –EE (B) muscle in HCO3– buffer (black) and HCO3– buffer plus 100 µM DIDS (grey, applied 30 min before NH4Cl). At time 0 s, 10 mmol/l NH4Cl was applied for 300 s. The initial alkali load was followed by a gradual decrease of pHi because of HCO3– efflux via Cl–/HCO3– exchange. Broken line depicts the approximate slope of recovery from alkali load. Note the steeper slope of pHi decrease in –EE muscles compared to +EE muscles.

 

Figure 5
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 HCO3– equivalent efflux following an alkali load in +EE (n=5) and –EE (n=5) muscles in HCO3– buffer and HCO3– buffer plus 100 µM DIDS (*p<0.05: HCO3– vs. HCO3–+DIDS, #p<0.05: +EE vs. –EE). Note the larger HCO3– efflux rate in –EE than in +EE muscles and the inhibition of HCO3– efflux by DIDS.

 
Fig. 6 shows representative examples of muscle pHi changes following acid loads imposed by wash out of NH4Cl. As explained in Section 2, the upward slope following wash out of NH4Cl is attributed to activity of Na+/H+ exchange (H+ efflux) and Na+/HCO3– cotransport (HCO3– influx). If endocardial endothelial cells regulate a transendothelial influx of HCO3–, as suggested by the previous observations, it would be expected that +EE muscles recover faster from acid loads than –EE muscles. Consistently, Fig. 6 shows that the slope of pHi recovery was steeper in the +EE than in the –EE muscle. Importantly, EIPA, an inhibitor of NHE that forces pHi recovery to be driven by HCO3– influx only, diminished the slopes of acid load recovery by about half, but preserved the differences between +EE and –EE muscles. Fig. 7 summarises mean data of the experimental groups. In +EE muscles (n=6), pHi recovered from 7.10±0.07 to 7.39±0.06 pHi with HCO3– equivalent influx of 2.15±0.54 mmol/l/min, whereas in –EE muscles (n=5) recovery was from 7.20±0.09 to 7.36±0.07 pHi with lower HCO3– equivalent influx of 1.06±0.20 mmol/l/min (p=0.066). In the presence of EIPA (condition in which Na+/HCO3– cotransport is the only acid extruder), recovery was from 7.12±0.05 to 7.38±0.05 pHi in +EE muscles and from 7.16±0.09 to 7.33±0.08 pHi in –EE muscles. In both muscle groups, EIPA reduced HCO3– equivalent influx by about 50% to, respectively, 1.18±0.23 (p=0.028) and 0.57±0.07 mmol/l/min (p=0.066). In conditions where Na+/HCO3– cotransport is the only acid extruder, HCO3– influx is significantly reduced in the absence of endocardial endothelium (p=0.023). These results are consistent with the hypothesis that HCO3– equivalent influx during an acid load in +EE muscles is favoured by the presence of endocardial endothelial cells.


Figure 6
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 pHi of a +EE (A) and a –EE (B) muscle in HCO3– buffer (black) and HCO3– buffer plus 5 µM EIPA (grey, added 20 min before NH4Cl). At time 300 s, 10 mmol/l NH4Cl was washed out, which led to acid load. Subsequently, recovery from acid load occurred by HCO3– influx via Na+/HCO3– cotransport and/or H+ efflux via Na+/H+ exchange, with the former only being operable in the presence of EIPA. Broken lines depict the approximate slope of recovery from acid load. Note the steeper slope of pHi increase in +EE muscles compared to –EE muscles and the slower pHi increase in the presence of EIPA.

 

Figure 7
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 HCO3– equivalent influx in +EE (n=6) and –EE (n=5) muscles in HCO3– buffer and HCO3– buffer plus 5 µM EIPA. (*p<0.05: HCO3– vs. HCO3–+EIPA, #p<0.05: +EE vs. –EE). Note the larger HCO3– influx rate in +EE than in –EE muscles and the partial inhibition of HCO3– influx by EIPA.

 
Finally, when both acid extruders (Na+/H+ exchange and Na+/HCO3– cotransport) were inhibited by removal of HCO3– and addition of EIPA, recovery was absent in three out of six experiments and reduced by 83±9% at the mean (data not shown), indicating that Na+/H+ exchange and Na+/HCO3– cotransport are the major acid extruders in cardiac muscle.

3.4 Transendothelial HCO3– transport
Above data on cardiac muscles suggest that the endocardial endothelium creates a transendothelial pH gradient through transport of HCO3– ions. To validate this observation, porcine and rat endocardial endothelial cells were cultured for 24, 48 and 72 h and the HCO3– concentration of the medium at luminal and basal side of the cells was measured (Fig. 8A and B). At each time interval, the basal concentration of HCO3– was significantly higher than the luminal concentration resulting in a transendothelial HCO3– gradient. The gradient was preserved during the progressive acidification of the medium over time due to metabolic activity of the cells. Incubating luminal and basal medium with 100 µM DIDS 12 h before measurement of HCO3– concentrations, abolished the small HCO3– gradient, which developed after 48 h culture of rabbit endocardial endothelial cells (Fig. 8C).


Figure 8
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8 Concentrations of HCO3– at the luminal and basal side of cultured rat (A), porcine (B) and rabbit (C) endocardial endothelial cells (n=3), which were grown for indicated times (A and B) or for 48 h (C) on collagen-I coated inserts. HCO3– concentration was significantly different between luminal and basal side of the cells (ANOVA, p=0.0001 for A and p<0.0001 for B) and decreased significantly with time (ANOVA, p<0.0001 for A and p=0.007 for B). In the rabbit, luminal HCO3– concentration was lower than basal HCO3– concentration (p=0.05) in control conditions, whereas following incubation of luminal and basal medium with 100 µM DIDS, luminal and basal HCO3– concentrations were not different (p=0.42). (*p<0.05, **p<0.01: basal vs. luminal HCO3– concentration).

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
In the present study, we observed that pHi of subendocardial myocardium of isolated papillary muscles was consistently and significantly higher in muscles with intact endocardial endothelium. In these muscles, DIDS, an inhibitor of HCO3– transport, or removal of HCO3– from the buffer solution, reduced pHi to the same level as in muscles with damaged endocardial endothelium, but had no or only a slight effect on pHi when endocardial endothelium had been damaged prior to DIDS administration or HCO3– removal. From these observations, we hypothesized the presence of an endocardial trans-endothelial HCO3– flux that directly modulates pHi in cardiomyocytes. This evidence was further strengthened by measurements of a DIDS-dependent trans-endothelial HCO3– gradient over a monolayer of cultured endocardial endothelial cells. Furthermore, using the NH4Cl pulse technique in papillary muscles, we observed that following perturbations in pHi, the endocardial endothelium mediated compensatory responses for pHi homeostasis in the subendocardium.

4.1 Transendothelial HCO3– flux and regulation of pHi of cardiomyocytes
Alkali loads (pHi>7.25) in cardiomyocytes are counteracted by HCO3– equivalent efflux via Cl–/HCO3– exchange mainly and to a lesser extent by Cl–/OH– exchange [6]. One of the major observations of the present study is that HCO3– equivalent efflux rate in a cardiac muscle exposed to an alkaline pulse is twice as high when the endocardial endothelium is damaged (2 vs. 1 mmol/l/min, respectively). This observation suggests trans-endothelial flux of HCO3– towards subjacent muscle. Consistently, damaging endocardial endothelium in a multicellular cardiac muscle preparation results in an efflux rate of HCO3– that approaches that of an isolated single cardiomyocyte (2–9 mmol/l/min [19]). Accordingly, although the lower equilibration rate of ion fluxes in multicellular muscle preparations is usually explained by a slower diffusion rate in the interstitial space [21], our data rather suggest that the presence of endocardial endothelial cells and the associated transendothelial HCO3– flux may participate in this difference. By contrast, enhanced Cl–/HCO3– exchange activity via higher pHi [6] was unlikely to be the cause of higher efflux in –EE muscles as pHi under NH4Cl was lower in –EE than in +EE muscles (7.72±0.07 vs. 7.94±0.13). In addition, NH4+ entry through potassium transporting pathways [21] was also unlikely to contribute as the decrease in pHi during the alkali load was completely inhibited by blocking Cl–/HCO3– exchange.

Acid loads (pHi<7.25) in cardiac muscle cells are removed by HCO3– influx via Na+/HCO3– cotransport and efflux of H+ via Na+/H+ exchange. An important observation of this study was that cardiac muscles recovered from acid loads (imposed by wash out of NH4Cl) twice as fast in the presence than in the absence of an intact endocardial endothelium. The difference in recovery rate between +EE and –EE muscles was significant even when recovery occurred only through influx of HCO3–. Accordingly, these data again support the concept that endocardial endothelium contributes to HCO3– influx in cardiac muscle and that endocardial endothelium creates a trans-endothelial flux of HCO3– anions from the luminal to the basal side. Of note, it is unlikely that pHi-induced differences in Na+/HCO3– cotransport and Na+/H+ exchange activity [6] contributed to differences in HCO3– influx rate between +EE and –EE muscles as there was no significant difference between the minimum recovery pHi for both muscle groups.

4.2. Endocardial endothelium as a physicochemical barrier
The endothelium interacts with the circulating blood and its immediate tissue environment by releasing diffusible substances or expressing membrane receptors. In the present study, we present evidence that in addition to these recognized mechanisms the endocardial endothelium may affect its environment by creating an active trans-endothelial physicochemical gradient for ions. Although this is a novel finding for endocardial endothelium and its relation to the subjacent myocardium, trans-endothelial ion fluxes have previously been described in endothelia of other organs. For example, at the blood–brain barrier, trans-endothelial ion fluxes are essential to preserve the unique interstitial environment of the brain [23,24]. In addition, in corneal endothelium, trans-endothelial ion fluxes underlie the corneal fluid pump, which is essential to counterbalance fluid leak into the cornea through the corneal endothelium [25]. Similarly as described for endocardial endothelium in the present study, corneal trans-endothelial ion fluxes include HCO3– fluxes. In corneal endothelium, these fluxes are directed from stroma (basolateral side) to aqueous humor (luminal side) and are created by combined activity of Na+/HCO3– cotransporter-1 (expressed at the endothelial basolateral side [26,27]) and carbonic anhydrase IV (expressed at the endothelial luminal side [25,28,29]).

Compared with corneal endothelium, transendothelial HCO3– flux in endocardial endothelium is directed in the opposite sense, i.e. from the luminal to the basal side of the cells, suggesting that the asymetrical organisation of ion transport in endocardial endothelium is opposite to the organisation in corneal endothelial cells. In agreement with this hypothesis, we previously described that the {alpha}1 subunit of Na+/K+ ATPase in endocardial endothelium was localised at the luminal side of the cell membrane [3] and thus opposite to the basolateral localisation in corneal endothelium [25]. The mechanisms of trans-endothelial HCO3– flux in endocardial endothelium, however, as well as the precise localisation of related ion carriers and pumps, needs further investigation. Together with the asymmetrical distribution of ion channels and pumps in endocardial endothelial cells [3,4] the transendothelial transport of HCO3– and/or H+ provides further evidence for the concept that the endocardial endothelium functions as an active blood–heart barrier, which controls subendothelial and global ion homeostasis of the cardiomyocytes.

Importantly, however, in the present study, only the presence and absence of endocardial endothelial cells was investigated, and we cannot exclude that other cardiac cells (in capillaries and microcapillaries, or even fibroblasts) could have an additive effect on muscle pHi.

4.3. Physiological implications
4.3.1. Multicellular vs. single cell preparations
It is well known that pHi of intact cardiac tissue responds differently than pHi of isolated single cardiomyocytes. For example, whereas removal of HCO3– does not affect pHi in isolated single cardiomyocytes [6,9,14,15], it acidifies intact hearts or undamaged cardiac muscle (the present study and Refs. [8,11,12,30,31]). Surprisingly, in the present study, when the endocardial endothelium of cardiac muscles was intentionally damaged, removal of HCO3– also failed to significantly affect pHi. Hence, pHi of cardiac muscle with damaged endocardial endothelium responds similarly as pHi of isolated single cardiomyocytes. Based on the observations in this study, we now postulate that the difference in pHi behaviour of different experimental preparations is explained by the presence or absence of transendothelial HCO3– fluxes in the preparation. Accordingly, +EE muscles, contrary to isolated single cardiomyocytes and to –EE muscles, but similarly to corneal endothelial cells [25], accumulate HCO3– at steady-state physiological pHi, and display a larger activity of Na+/HCO3– exchange than of Cl–/HCO3– exchange.

4.3.2. Limitations of the study
In the present study, by reporting signals from isolated papillary muscles loaded with the intracellular pH-indicator BCECF, we intended to measure pHi of cardiomyocytes in cardiac tissue, which is, however, a multicellular preparation containing also other types of cells. It was found that BCECF loading was confined to the outer 15–20 µm of the muscle surface, suggesting that the major pHi signal is from the cardiomyocytes beneath the thin endocardial endothelial cell layer. Also, we found that maximally 15% of the total BCECF signal in +EE muscles was due to BCECF loaded in the endocardial endothelium. Therefore, it cannot be excluded that BCECF signals from non-cardiomyocytes in isolated papillary muscles contribute to some extent to the total muscle BCECF signal measured in the present study. The relatively acidic pHi of endocardial endothelial cells (6.97±0.06 pHi, n=11), however, at least in culture (personal observations), indicates that this confounding factor cannot explain our results. Indeed, pHi of +EE muscles was more alkaline than of –EE muscles, suggesting that our results are rather underestimations of the modulatory effects of endocardial endothelium on cardiac muscle pHi. On the other hand, transsarcolemmal HCO3– flux of cardiomyocytes may be affected not only by the endocardial endothelium, but also by paracrine substances or electrophysiological signals [13,14].

4.3.3. pH regulation in cardiac pathologies
Disturbed calcium handling, e.g. in cardiac infarction or during development of cardiac hypertrophy or heart failure, mediates contractile dysfunction and contributes to genesis of ventricular arrhythmias. It may, however, be an adaptive process secondary to other stimuli to cope with altered contractile demands. Increased expression of Na+/HCO3– cotransport, Na+/H+ exchange and Cl–/HCO3– exchange has been described in hypertrophied myocardium of spontaneously hypertensive rats and following myocardial infarction [20,32–34]. Specific inhibition of Na+/H+ exchange, knocking out the Na+/H+ exchanger gene, but also administration of anti- Na+/HCO3– cotransporter antibody has been shown to be beneficial in ischemia/reperfusion injury [33–36]. The present observation that the endocardial endothelium mediates transendothelial flux of HCO3– to the cardiomyocytes suggests a potential role for a (dys)functional endocardial endothelium in the pathogenesis of cardiac disease.


   Acknowledgements
 
This work is supported by Concerted Research Project UA 1998: "Endothelium and cellular infiltrate in tissue remodelling". The authors wish to acknowledge Prof. Dr. Dirk L. Brutsaert for his constructive comments; Dr. Jan Van den Bossche and Dr. Annick Wauters for bicarbonate concentration measurements; Mss. Pascale Van Tongelen for cell culture.


   Notes
 
1 Present address: Laboratory for Physiology, VU University Medical Center, Van der Boechorststraat 7, Amsterdam 1081 BT, The Netherlands. Back

Time for primary review 35 days


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Brutsaert D.L. Cardiac endothelial–myocardial signaling: its role in cardiac growth, contractile performance, and rhythmicity. Physiol. Rev. (2003) 83:59–115.[Abstract/Free Full Text]
  2. Muller-Borer B.J., Yang H., Marzouk S.A.M., Lemasters J.J., Cascio W.E. pHi and pHo at different depths in perfused myocardium measured by confocal fluorescence microscopy. Am. J. Physiol. (1998) 275:H1937–H1947.[Web of Science][Medline]
  3. Fransen P., Hendrickx J., Brutsaert D.L., Sys S.U. Distribution and role of Na+/K+ ATPase in endocardial endothelium. Cardiovasc. Res. (2001) 52:487–499.[Abstract/Free Full Text]
  4. Manabe K., Ito H., Matsuda H., Noma A., Shibata Y. Classification of ion channels in the luminal and abluminal membranes of guinea-pig endocardial endothelial cells. J. Physiol. (1995) 484:41–52.[Abstract/Free Full Text]
  5. Sun B., Leem C.H., Vaughan-Jones R.D. Novel chloride-dependent acid loader in the guinea-pig ventricular myocyte: part of a dual acid-loading mechanism. J. Physiol. (1996) 495:65–82.[Abstract/Free Full Text]
  6. Leem C.H., Lagadic-Gossmann D., Vaughan-Jones R.D. Characterization of intracellular pH regulation in the guinea-pig ventricular myocyte. J. Physiol. (1999) 517:159–180.[Abstract/Free Full Text]
  7. Ellis D., Thomas R.C. Direct measurement of the intracellular pH of mammalian cardiac muscle. J. Physiol. (1976) 262:755–771.[Abstract/Free Full Text]
  8. Fülöp L., Szigeti G., Magyar J., et al. Differences in electrophysiological and contractile properties of mammalian cardiac tissues bathed in bicarbonate- and HEPES-buffered solutions. Acta Physiol. Scand. (2003) 178:11–18.[CrossRef][Web of Science][Medline]
  9. Puceat M., Clement O., Vassort G. Extracellular MgATP activates the Cl–/HCO3– exchanger in single rat cardiac cells. J. Physiol. (1991) 444:241–256.[Abstract/Free Full Text]
  10. Dart C., Vaughan-Jones R.D. Na+-HCO3– symport in the sheep cardiac Purkinje fibre. J. Physiol. (1992) 451:365–385.[Abstract/Free Full Text]
  11. Grace A.A., Kirschenlohr H.L., Metcalfe J.C., et al. Regulation of intracellular pH in the perfused heart by external HCO3– and Na+–H+ exchange. Am. J. Physiol. (1993) 265:H289–H298.[Web of Science][Medline]
  12. Kusuoka H., Marban E., Cingolani H.E. Control of steady-state intracellular pH in intact perfused ferret hearts. J. Mol. Cell. Cardiol. (1994) 26:821–829.[CrossRef][Web of Science][Medline]
  13. Cingolani H.E., Alvarez B.V., Ennis I.L., Camilion de Hurtado M.C. Stretch-induced alkalinization of feline papillary muscle: an autocrine-paracrine system. Circ. Res. (1998) 83:775–780.[Abstract/Free Full Text]
  14. Aiello E.A., Petroff M.G., Mattiazzi A.R., Cingolani H.E. Evidence for an electrogenic Na+-HCO3– symport in rat cardiac myocytes. J. Physiol. (1998) 512:137–148.[Abstract/Free Full Text]
  15. Skolnick R.L., Litwin S.E., Barry W.H., Spitzer K.W. Effect of ANG II on pHi, [Ca2+]i, and contraction in rabbit ventricular myocytes from infarcted hearts. Am. J. Physiol. (1998) 275:H1788–H1797.[Web of Science][Medline]
  16. Brutsaert D.L., De Keulenaer G.W., Fransen P., et al. The cardiac endothelium: functional morphology, development, and physiology. Prog. Cardiovasc. Dis. (1996) 39:239–262.[CrossRef][Web of Science][Medline]
  17. De Keulenaer G.W., Andries L.J., Sys S.U., Brutsaert D.L. Endothelin-mediated positive inotropic effect induced by reactive oxygen species in isolated cardiac muscle. Circ. Res. (1995) 76:878–884.[Abstract/Free Full Text]
  18. Thomas J.A., Buchsbaum R.N., Zimniak A., Racker E. Intracellular pH measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ. Biochemistry (1979) 18:2210–2218.[CrossRef][Web of Science][Medline]
  19. Xu P., Spitzer K.W. Na-independent Cl–-HCO3– exchange mediates recovery of pHi from alkalosis in guinea pig ventricular myocytes. Am. J. Physiol. (1994) 267:H85–H91.[Web of Science][Medline]
  20. Chiappe de Cingolani G.E., Morgan P., Mundina-Weilenmann C., et al. Hyperactivity and altered mRNA isoform expression of the Cl–/HCO3– anion-exchanger in the hypertrophied myocardium. Cardiovasc. Res. (2001) 51:71–79.[Abstract/Free Full Text]
  21. Frohlich O., Wallert M.A. Methods of measuring intracellular pH in the heart. Cardiovasc. Res. (1995) 29:194–202.[Free Full Text]
  22. Fransen P.F., Demolder M.J., Brutsaert D.L. Whole cell membrane currents in cultured pig endocardial endothelial cells. Am. J. Physiol. (1995) 268:H2036–H2047.[Web of Science][Medline]
  23. Ennis S.R., Ren X., Betz A.L. Mechanisms of sodium transport at the blood–brain barrier studied with in situ perfusion of rat brain. J. Neurochem. (1996) 66:756–763.[Web of Science][Medline]
  24. Hsu P., Haffner J., Albuquerque M.L., Leffler C.W. pHi in piglet cerebral microvascular endothelial cells: recovery from an acid load. Proc. Soc. Exp. Biol. Med. (1996) 212:256–262.[CrossRef][Medline]
  25. Bonanno J.A. Identity and regulation of ion transport mechanisms in the corneal endothelium. Prog. Retin. Eye Res. (2003) 22:69–94.[CrossRef][Web of Science][Medline]
  26. Bonanno J.A., Giasson C. Intracellular pH regulation in fresh and cultured bovine corneal endothelium: II. Na+:HCO3– cotransport and Cl–/HCO3– exchange. Invest. Ophthalmol. Vis. Sci. (1992) 33:3068–3079.[Abstract/Free Full Text]
  27. Lane J., Wigham C.G., Hodson S.A. A chloride-activated Na+/HCO3–coupled transport activity in corneal endothelial membranes. Biophys. J. (2000) 78:2493–2498.[Web of Science][Medline]
  28. Kuang K.Y., Xu M., Koniarek J.P., Fischbarg J. Effects of ambient bicarbonate, phosphate and carbonic anhydrase inhibitors on fluid transport across rabbit corneal endothelium. Exp. Eye Res. (1990) 50:487–493.[CrossRef][Web of Science][Medline]
  29. Bonanno J.A., Guan Y., Jelamskii S., Kang X.J. Apical and basolateral CO2–HCO3– permeability in cultured bovine corneal endothelial cells. Am. J. Physiol. (1999) 277:C545–C553.[Web of Science][Medline]
  30. Spitzer K.W., Bridge J.H. Relationship between intracellular pH and tension development in resting ventricular muscle and myocytes. Am. J. Physiol. (1992) 262:C316–C327.[Web of Science][Medline]
  31. Camilion de Hurtado M.C., Perez N.G., Cingolani H.E. An electrogenic sodium-bicarbonate cotransport in the regulation of myocardial intracellular pH. J. Mol. Cell. Cardiol. (1995) 27:231–242.[Web of Science][Medline]
  32. Perez N.G., Alvarez B.V., Camilion de Hurtado M.C., Cingolani H.E. pHi regulation in myocardium of the spontaneously hypertensive rat. Compensated enhanced activity of the Na+–H+ exchanger. Circ. Res. (1995) 77:1192–1200.[Abstract/Free Full Text]
  33. Baartscheer A., Schumacher C.A., van Borren M.M., Belterman C.N., Coronel R., Fiolet J.W. Increased Na+/H+-exchange activity is the cause of increased [Na+]i and underlies disturbed calcium handling in the rabbit pressure and volume overload heart failure model. Cardiovasc. Res. (2003) 57:1015–1024.[Abstract/Free Full Text]
  34. Allen D.J., Xiao X.-H. Role of the cardiac Na+/H+ exchanger during ischemia and reperfusion. Cardiovasc. Res. (2003) 57:776–782.
  35. Wang Y., Meyer J.W., Ashraf M., Shull G.E. Mice with a null mutation in the NHE1 Na+–H+ exchanger are resistant to cardiac ischemia–reperfusion injury. Circ. Res. (2003) 93:776–782.[Abstract/Free Full Text]
  36. Khandoudi N., Albadine J., Robert P., et al. Inhibition of the cardiac electrogenic sodium bicarbonate cotransporter reduces ischemic injury. Cardiovasc. Res. (2001) 52:387–396.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Fransen, P.
Right arrow Articles by De Keulenaer, G. W
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fransen, P.
Right arrow Articles by De Keulenaer, G. W
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?