| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1 Medicine and
2 Cell Biology and Anatomy, Confocal microscopy and the
H+-sensitive fluorophore
carboxyseminaphthorhodafluor-1 (SNARF-1) were used to measure either
intracellular pH (pHi) or
extracellular pH (pHo) in
isolated, arterially perfused rabbit papillary muscles.
Single-excitation, dual-emission fluorescent images of the endocardial
surface and underlying myocardium to a depth of 300 µm were
simultaneously recorded from perfused cylindrical muscles suspended in
a controlled atmosphere oriented oblique to the focal plane.
Contraction was inhibited by the addition of butanedione monoxime. In
separate muscles, pHo was measured during continuous perfusion of SNARF-1 free acid.
pHi measurements were made after
the muscle was loaded with SNARF-1/AM and the extracellular space was
cleared of residual fluorophore. Initial experiments demonstrated the
uniformity of ratiometric measurements as a function of pH, image
depth, and fluorophore concentration, thereby establishing the
potential feasibility of this method for quantitative intramural pH
measurements. In subsequent experiments, the method was validated in
isolated, arterially perfused rabbit papillary muscle during normal
arterial perfusion and as pHi and pHo were altered by applying
CO2 externally, exchanging HEPES and bicarbonate buffers, and changing
pHi with
NH4Cl washout. We conclude that in
situ confocal fluorescent microscopy can measure pHi and
pHo changes at the endocardial
surface and deeper endocardial layers in arterially perfused
ventricular myocardium. This method has the potential to study
pHi regulation in perfused
myocardium at boundaries where diffusion of gases, metabolites, and
peptides are expected to modify processes that regulate
pHi.
carboxyseminaphthorhodafluor-1; ventricular myocardium; carbon
dioxide; acidosis; alkalosis
CYTOSOLIC pH is an important modulator of
cellular metabolism, ionic homeostasis, membrane channel function,
intracellular signaling, and cell-to-cell communication. The regulation
of intracellular pH (pHi) is
tightly controlled, and changes in
pHi are minimized by complex
processes involving sarcolemmal ion exchangers, cotransporters, cellular consumption of nonvolatile acids, and chemical
buffering. In cardiac tissue these processes have been
studied using the distribution of weak acids and bases (27),
31P NMR in whole heart
preparations (10, 13, 14, 22), and liquid ion-exchange microelectrodes
in cultured, isolated cells and tissue (8, 9, 26, 30). Although these
techniques have provided important insights into the regulation of
pHi in cardiac tissue, they lack
spatial and/or temporal resolution to access local changes in
pHi anticipated from diffusion at
the endocardial surface or through diffusion processes occurring
between the ventricular myocyte and adjacent vascular endothelium or
vessels. Recent developments in fluorescent indicators and in situ
confocal microscopy provide an alternative approach. Confocal
fluorescence microscopy with pH-sensitive fluorophores allows
pHi and extracellular pH
(pHo) changes to be measured
with spatial and temporal resolution that surpasses microelectrode or
NMR techniques, respectively.
In this study, carboxyseminaphthorhodafluor-1 (SNARF-1), a
single-excitation, dual-emission fluorescent pH indicator, was used in
conjunction with confocal fluorescence microscopy to measure pHi and
pHo in the superficial layers of
an arterially perfused in situ rabbit papillary muscle suspended in a
humidified atmosphere. Our studies show that ratiometric imaging of the
pH-sensitive fluorophore permits quantitative measurements of
pHi and
pHo from the endocardial surface
to a depth of ~300 µm. The absence of a significant inner-filter
effect and the ability of ratiometric imaging to compensate for
differences in fluorophore concentrations and attenuation of emitted
light in the deeper cell layers made this technique possible.
Although spatial resolution diminished beyond a depth of 100-125
µm and cellular structure could not be resolved below this depth, the
mean pH values calculated from multicellular regions of interest remain valid.
In summary, this new method permits the determination of either
pHi or
pHo in in situ arterially perfused
ventricular myocardium at the endocardial surface and deeper
endocardial layers. This method has the necessary spatial and temporal
resolution to evaluate the effects of mobile, diffusible compounds such
as CO2 or vasoactive peptides on
the regulation of pH at the blood-tissue boundaries. In addition, this
method has the potential to examine the role of pH in microvascular
regulation in the superficial layers of ventricular myocardium, i.e.,
within 50-100 µm of the surface of perfused in situ cardiac tissue.
Perfusion of Papillary Muscle and Ventricular Septum
ABSTRACT
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
INTRODUCTION
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
MATERIALS AND METHODS
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
, 25 HCO3, 0.4 HPO24, and 20 glucose). The isolated
heart was placed on a dissection tray, where the atria and left
ventricular free wall were removed. The left ventricular surface of the
septum was secured to a wax platform, and the septal artery was
cannulated, secured with a purse-string suture, and perfused with a
modified Tyrode (M-Tyrode) perfusate. This perfusate included the
addition of insulin (1 U/l), heparin (400 U/l), albumin (2 g/l), and
dextran [average mol wt 70,000; 40 g/l]. In
each experiment the elapsed time between cross-clamping of the aorta
and perfusion was <6 min.
The muscle and cannula were transported to a custom-made chamber (15)
as shown in Fig. 1. The nonperfused portion
of the right ventricle was removed, exposing two to three right
ventricular papillary muscles. The wax platform with the septal
preparation was lowered into the chamber, and one papillary muscle
(diameter = 1.3 ± 0.5 mm, n = 16)
was suspended in the chamber by securing its tendon to an adjustable
pin. With the muscle attached to the pin, a 40-45° angle was
formed between the muscle and the chamber base. The chamber was covered
to maintain a constant humidified atmosphere around the muscle. A
peristaltic pump (Digi-Staltic, Masterflex, Barrington, IL) controlled
perfusate flow to the chamber. Perfusate temperature was maintained at
36-37°C by passing the perfusate through a heated water bath
in the base of the chamber before it reached the muscle. Intra-arterial
pressure was monitored continuously with a pressure transducer (Millar,
Houston, TX) and recorded on a strip-chart recorder. The intra-arterial
pressure was maintained between 30 and 50 mmHg by adjustment of the
perfusion flow rate (1.3-1.5
ml · min1 · g
tissue1). This is a
normal perfusion pressure for the small arteries (i.e., diameters
~120-160 µm) of the rabbit heart. In rabbit, 40-50% of
the peripheral coronary resistance is located in vessels with diameters
>150 µm (6).
|
Control and Measurement of PO2, PCO2, and pH
During normal perfusion, the PCO2 and PO2 of the atmosphere surrounding the muscle were matched to the PCO2 and PO2 of the perfusate. The volume fractions of O2 and CO2 in the atmosphere of the chamber were frequently sampled ~0.5 mm from the muscle surface with an O2/CO2 gas analyzer (Illinois Instruments, Ingleside, IL). The pH, PO2, PCO2, and [HCO3] of the
perfusate were measured with a blood gas analyzer (Instrumentation
Laboratory, Lexington, MA). PO2 and
PCO2 in the perfusate were adjusted in a membrane gas exchanger. A schematic drawing of the perfusion system is shown in the lower portion of Fig.
2. The perfusate pH was continuously
monitored by a glass pH electrode connected to the perfusion line at
the inlet to the chamber. The relative amount of
CO2 was adjusted to yield a
perfusate pH of 7.40 ± 0.07 during control perfusion
(PO2 = 223 ± 35 mmHg,
PCO2 = 37 ± 3 mmHg). A major
advantage of this method was the ability to precisely control and match
the PCO2 of the chamber and
perfusate.
|
SNARF-1 Perfusion for pHo Measurements
pHo was measured in the perfused rabbit papillary muscle preparation during several interventions as the muscle was continuously perfused with M-Tyrode perfusate buffered with bicarbonate or HEPES. The constituents of the HEPES perfusate were (in mM) 149 Na+, 4.5 K+, 0.49 Mg2+, 1.8 Ca2+, 133 Cl, 0.4 HPO24, 10 HEPES, and 20 glucose and
insulin (1 U/l), heparin (400 U/l), albumin (2 g/l), and
dextran (average mol wt 70,000; 40 g/l). The perfusate contained
SNARF-1 free acid (20 µM, Molecular Probes) and butanedione monoxime
(BDM; 20 mM, Sigma). BDM was added to inhibit electromechanical
coupling and contractions, to prevent motion artifact related to
contraction. To validate pHo
measurements acquired from SNARF-1 fluorescence a novel,
custom-designed, solid-state electroplated iridium oxide pH electrode
(18) was placed on the muscle surface. This
pHo electrode
continuously monitored pHo changes
and recorded the pHo
on a strip-chart recorder. Additionally, in several studies, after
the muscle was perfused with SNARF-1, a fluorophore-free perfusate was
used to eliminate SNARF-1 from the extracellular compartment.
Extracellular washout of SNARF-1 was complete within 15 min, i.e.,
fluorescent emission intensities were equal to background emissions.
SNARF-1/AM Loading for pHi Measurements
For pHi measurements, muscles were loaded with the acetoxymethyl ester of SNARF-1 (SNARF-1/AM). SNARF-1/AM enters the cells and is converted to SNARF-1 free acid by intracellular esterases. SNARF-1/AM was dissolved in dimethyl sulfoxide at a concentration of 1 mg/ml. This stock solution was diluted in M-Tyrode perfusate for a final concentration of 10 µM. The dextran concentration of the M-Tyrode perfusate during SNARF-1 loading was reduced to 10 g/l. The papillary muscles were perfused with SNARF-1/AM perfusate for 30 min at 37°C. To improve intracellular SNARF-1 loading, the papillary muscles were simultaneously superfused by placing a single papillary muscle inside a piece of longitudinally cut plastic tubing filled with SNARF-1/AM perfusate. The superfusate in the tubing was replenished several times during SNARF-1 loading. In addition, a Grass stimulator (Grass Instruments, Quincy, MA) applied excitatory current pulses (0.5 ms at double threshold) to the apical end of the papillary muscle through an insulated platinum wire. While the muscle preparation spontaneously beat at 0.5-1.0 Hz, preliminary observations revealed improved SNARF-1 loading (i.e., higher fluorescence emission intensities) when the muscles were stimulated at 2 Hz. After SNARF-1 loading, the muscle preparation was perfused with a fluorophore-free M-Tyrode perfusate containing FCS (10%, Life Technologies), BDM (20 mM), and a normal dextran concentration (40 g/l). The superfusion tubing was removed from the chamber, and electrical stimulation was discontinued. Results from initial studies indicated that the fluorophore was retained within the cells for a longer period of time when FCS was used during SNARF-1/AM washout. Conversely, when FCS and/or BDM was added to the perfusate during SNARF-1 loading very low-emission intensities were recorded, suggesting poor intracellular loading. All intracellular fluorescence images were collected after an extracellular fluorophore washout period of 15 min.Intracellular SNARF-1 concentration was estimated after each
experimental protocol. In 5 µM increments, SNARF-1 was added to
HEPES-Tyrode (in mM: 149 Na+, 4.5 K+, 0.49 Mg2+, 1.8 Ca2+, 133 Cl, 0.4 HPO24, 10 HEPES, and 20 glucose) with the pH adjusted to five values ranging from 6.4 to 7.6 (e.g., 6.4, 6.7, 7.0, 7.3, and 7.6). The SNARF-1 solutions were placed in a single-well
chamber slide covered with a coverslip. Emission fluorescence
intensities from samples were recorded using identical system settings,
objectives, and image pair acquisition times (i.e., total time to
acquire emission image pairs) per experimental protocol. Intracellular
fluorophore concentrations were estimated by comparing the magnitude of
emission fluorescence intensities acquired in the samples with those
acquired in the muscle. The SNARF-1 solutions with concentrations
similar to the muscle preparation were subsequently used for pH
calibrations (see pH Calibration).
Confocal Fluorescence Microscopy and Image Analysis
The chamber was placed on a modified stage of a Nikon microscope (Nikon, Melville, NY) equipped with a K2SBIO disk-scanning confocal attachment. A schematic drawing of the optical elements is shown in the top portion of Fig. 2. The 45-µm-wide curved slits in the scanning disk allowed uniform 10% light transmission and video rate confocal imaging. Light from a mercury-arc lamp (100 W) passed through a 546-nm excitation filter and was focused on the papillary muscle suspended inside the chamber. Emitted SNARF-1 fluorescence from the excited muscle passed through a 570-nm long-pass dichroic reflector. Emission image pairs were collected serially through a 620-nm long-pass filter (Image620) and 585 ± 10-nm band-pass filter (Image585). Emission filters were located on a rotating filter wheel in the optical path. Filter position and image pair acquisition time were software controlled. A cooled charge-coupled device camera (Princeton Instruments, Trenton, NJ) with a 1,317 × 1,035-pixel array (6.8 × 6.8-µm pixel size) collected a 12-bit image at each emission wavelength (i.e., emission intensities ranged from 0 to 4,096 arbitrary units). A schematic of the image acquisition system is shown on the right side of Fig. 2. Before SNARF-1/AM loading, background autofluorescence images of the muscle in the chamber were acquired and stored at each emission wavelength (Image620,bkgd and Image585,bkgd, respectively). To determine the SNARF-1 fluorescence ratio (ImageRatio), background autofluorescence was subtracted at each emission wavelength as calculated in Eq. 1.
|
(1) |
SNARF-1 emissions were observed through an Olympus SPLAN10 air objective (0.3 NA). When the objective was lowered through the opening in the chamber cover, a rubber shield attached to the objective helped to contain the atmosphere within the chamber. To prevent condensation from the humidified atmosphere on the objective a flexible Kapton heater (Minco Products, Minneapolis, MN) was wrapped around the objective with the temperature maintained at 40°C. Figure 3 illustrates the muscle alignment used to acquire a ratiometric confocal image, simultaneously capturing the endocardial surface and deeper endocardial layers. To assist in muscle alignment, focus, and angle measurement, a drop of perfusate with a 1% concentration of 1.0-µm-diameter yellow-green fluorescent latex microspheres (excitation and emission wavelength 490 and 515 nm, respectively) was placed on the surface of the muscle. Microspheres that adhered to the surface of the muscle were focused, allowing accurate placement of the microscope stage, alignment of the muscle along its longitudinal axis, and focus at the muscle surface. The angle of the muscle was determined by measuring the depth of focus along the longitudinal axis as shown in Fig. 3. In this orientation using an air objective and no coverslip, all depth-of-focus measurements were scaled by a factor of 0.73 to adjust for the mismatch of refractive indexes (i.e., refractive index of air is 1, refractive index of vertebrate muscle is 1.38) (12).
|
Intracellular fluorescence emissions were evaluated by selecting two regions of interest (ROI) from within the ratioed image. Each ROI encompassed multiple cells within the confocal slice, providing an average pH measurement. The ROI were centered along the longitudinal axis of the muscle, with one ROI placed near the edge of the image where endocardial surface measurements were recorded and the second ROI placed near the edge of the image where fluorescence from the deeper endocardial layers was recorded. Centering the muscle to acquire images along its longitudinal axis reduced measurement variability produced by the cylindrical muscle geometry.
pH Calibration
To calibrate SNARF-1 emissions, an in vitro pH calibration was performed after each experimental protocol. SNARF-1 was added to HEPES-Tyrode with the pH adjusted to five values ranging from 6.4 to 7.6 for pHi measurements and from 6.4 to 8.0 for pHo measurements. The SNARF-1 concentration ranged from 15 to 30 µM, based on intracellular dye concentration estimates previously described (see SNARF-1/AM Loading for pHi Measurements). The calibration solutions were placed in a single-well chamber slide covered with a coverslip. Emission fluorescence pairs were collected 100 µm below the surface of the liquid using identical system settings, objectives, and image pair acquisition times per experimental protocol. Background images were acquired with SNARF-1-free calibration solution. Background corrected ratios were created by division of the individual component fluorescence images as calculated from Eq. 1. Titration curves of the ratio vs. pH were calculated and used to convert tissue fluorescence emission ratios to pH.| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Validation of Ratiometric Measurement Technique for Determination of pH at Different Depths
Experiments were designed to determine the effect of path length and fluorophore concentration on the acquired fluorescence emissions and resulting ratiometric measurements. These initial studies established conditions necessary to quantitatively measure pH from SNARF-1 fluorescence in papillary muscle.Measurement of pH in a homogeneous aqueous volume. SNARF-1 fluorescence was measured in homogeneous aqueous solutions, with pH ranging from 6.5 to 7.5, as the fluorophore concentration was decreased from 50 to 25 µM and as the depth of focus was extended from the surface to 200 µm below the surface of the volume. The results, acquired 200 µm below the surface of the liquid, are shown in Fig. 4. In the 50 µM SNARF-1 solution, as the pH of the solution increased from 6.5 to 7.5, emission intensities recorded through the 620-nm long-pass filter increased, whereas the emission intensities recorded through the 585 ± 10-nm band-pass filter decreased. As the concentration of SNARF-1 decreased from 50 to 25 µM, the intensity of the emission fluorescence decreased significantly. However, at both SNARF-1 concentrations, the emission intensity ratios were the same, verifying that the fluorescence emission ratios were proportional to pH and independent of the fluorophore concentration. Furthermore, the difference in fluorescence emissions collected at 10 and 200 µm below the surface, although not shown in Fig. 4, was <2%.
|
Measurement of pH in a homogeneous semisolid volume. To simulate fluorophore measurements in thick tissue, SNARF-1 fluorescence was measured in a semiopaque solid. For this measurement, gelatin was dissolved in HEPES-Tyrode at 100°C. After the gelatin cooled to 25°C with a pH of 7.3, SNARF-1 (20 µM) was added to the gelatin, forming a semiopaque solid in which SNARF-1 was homogeneously distributed. The results, shown in Fig. 5, establish that SNARF-1 fluorescence can be measured at depths up to 300 µm without significant inner-filter effects. Furthermore, only a 2% decrease in fluorescence emission intensities occurred as the focal depth of the image plane moved from the surface to 300 µm below the surface. The observation that ratiometric imaging of SNARF-1 fluorescence compensated for differences in light attenuation caused by path length and fluorophore concentration supported the potential use of this method in tissue.
|
In Vivo Measurement of pHo
The ability to image SNARF-1 fluorescence to quantitate relative changes in pHo was assessed by the simultaneous measurement of pHo with a solid-state electroplated iridium oxide pH electrode (18) placed on the muscle surface. Papillary muscles were continuously perfused with M-Tyrode at a pH of 7.45, containing SNARF-1 (20 µM) and BDM (20 mM). As shown in Fig. 6, during steady-state perfusion with the bicarbonate-buffered M-Tyrode perfusate and a PCO2 of 40 mmHg, pHo at the muscle surface measured ~7.3 as measured by both fluorescence microscopy and the pHo electrode. This was slightly more acidic than the perfusate pH, measured by the glass pH electrode at the chamber inlet. A transient decrease in pHo was initiated by changing the perfusate to a non-bicarbonate-HEPES buffer, with CO2 present. This decreased the pH of the perfusate to 6.7 within 3 min. The pHo at the muscle surface showed no initial change during the switch in perfusate but gradually decreased to a pHo of 6.8 and 6.75 as measured by the extracellular iridium oxide pH-sensitive electrode and SNARF-1 fluorescence, respectively. When CO2 was removed from the atmosphere and perfusate the surface pHo returned to 7.2, slightly lower than the starting pHo, a finding previously reported by Vanheel et al. (26). The difference in pHo measured by the surface electrode and SNARF-1 fluorescence during the switch to HEPES and withdrawal of CO2 is most likely caused by the surface effect of CO2 withdrawal being sensed more rapidly by the surface electrode. However, at steady state both methods measured the same pHo. Figure 6 clearly shows the response of the SNARF-1 fluorescence measurement to changes in pH. The response to the detected change in H+ activity by SNARF-1 fluorescence was rapid and preceded the response of the extracellular iridium oxide pH electrode by ~1 min (i.e., response time of the extracellular iridium oxide pH electrode).
|
To further evaluate the relative changes in pHo, at the surface and intramural layers, measurements of pHo were acquired as perfusate pH was decreased in a stepwise manner. In this series of experiments the papillary muscles were continuously perfused with HEPES perfusate, SNARF-1 (20 µM), and BDM (20 mM) at 100% O2 saturation. Fluorescence images were recorded at 1-min intervals at the muscle surface and ~250 µm below the surface. Representative results are shown in Fig. 7 as the perfusate pH was decreased from 7.37 to 6.54. Initially, the pHo values both at the surface and below the surface of the muscle were more acidic than the perfusate pH. As the perfusate pH decreased, a corresponding decrease in pHo was observed. An interesting and consistent finding was a significant pHo difference between the muscle surface and deeper intramural layer that remained nearly constant as pHo decreased.
|
pHo Changes Induced by CO2
Chamber PCO2 was used to alter carbonic acid concentration and pHo in the superficial layers of the perfused papillary muscle and to assess ratiometric imaging as a means to measure pHo changes at the surface and deeper intramyocardial cell layers. Results are shown in Fig. 8. During continuous perfusion at constant PCO2 and pH (perfusate PCO2 = 33 mmHg, PO2 = 256 mmHg, pH = 7.55) confocal images were acquired. The resultant ratiometric images provided pHo measurements from the muscle surface to a depth of ~200-260 µm. Initially, the PCO2 values surrounding the muscle and perfusate were equal and measured pHo ranged from 7.40 at the surface to 7.58 in the deepest layer, consistent with the intrinsic pHo gradient found in this preparation with the HEPES-buffered perfusate. As the chamber PCO2 was reduced from 33 to 7 mmHg, a superficial PCO2 gradient was established (i.e., low PCO2 on muscle surface compared with deeper layers). Consequently, the surface pHo rose more compared with pHo changes in the deeper layers and the initial pHo difference between the surface and deeper endocardial layers disappeared. When the superficial PCO2 gradient was reversed by increasing the chamber PCO2 from 7 to 340 mmHg, the initial pHo difference was reestablished and pHo decreased to 6.47 at the surface and 6.71 in the deeper layers. Returning the atmosphere PCO2 to 33 mmHg restored the pHo to initial values.
|
Measurement of pHi
After the assessment of SNARF-1 emission fluorescence in aqueous solutions, the validation of the pH calibration technique, and confirmed measurement of pHo during CO2 manipulation, our goal was to measure pHi in the perfused rabbit papillary muscle preparation during normal perfusion. After SNARF-1 loading and SNARF-1/AM washout, fluorescence emission image pairs were collected at 1-min intervals until the fluorescence emission intensities decreased to the background noise level. The muscles (n = 9) were continuously perfused at a pH of 7.48 ± 0.03, and the PO2 (240 ± 8 mmHg) and PCO2 (33 ± 1 mmHg) of the chamber atmosphere and perfusate were matched. All ratiometric images included the muscle surface and deeper endocardial layers (280 ± 20 µm below the surface).In a representative study shown in Fig. 9, the emission intensity ratios corresponding to pHi remained constant over a 60-min period. With continuous perfusion at a pH of 7.43 the average pHi, recorded 100 µm below the surface, was 7.14 ± 0.01. However, the average fluorescence collected through the emission filters, >620 nm and 585 ± 10 nm, steadily decreased over this same time period. Like the SNARF-1 ratiometric measurements in an aqueous solution, this study demonstrated that the emission intensity ratios correct for variability in the fluorophore concentration.
|
Modulation of pHi With NH4Cl Washout
The recovery of pHi from acid loading induced by external NH4Cl addition and removal was used to assess changes in myocardial pHi. In this study the papillary muscles (n = 4) were placed in the chamber and perfused with HEPES perfusate (pH = 7.40) saturated with 100% O2. After 10 min of control perfusion, NH4Cl (10 µM) was perfused to the muscle for 15 min, followed by 20 min of NH4Cl-free perfusate. Figure 10 shows the pHi response measured ~100 µm below the surface of the papillary muscle during this intervention. As expected, NH4Cl perfusion produced a transient alkalosis, with pHi increasing from 7.2 to 7.6 during the first 10 min and then decreasing to 7.4. Conversely, when NH4Cl was removed a transient acidosis was measured, decreasing to pH 6.7 before recovering to control levels.
|
Hypercapnic Acidosis
Analogous to the experiment where chamber PCO2 was used to monitor pHo changes at the surface and deeper endocardial layers, a PCO2 pulse was used to assess changes in pHi. Initially, fluorescence was collected from the muscle during control perfusion (n = 6). Hypercapnic acidosis was created by simultaneously increasing the PCO2 in the perfusate and chamber atmosphere surrounding the muscle while adjusting the N2 to maintain a constant PO2. During control perfusion with PCO2 at 37 ± 3 mmHg, pHi measured 7.17 ± 0.09. When PCO2 was increased to 170 ± 33 mmHg, pHi decreased, reaching a steady state of 6.84 ± 0.05 within 10 min. Consequently, the perfusate pH decreased from 7.40 ± 0.07 to 6.65 ± 0.07. After acidosis, PCO2 in the chamber and perfusate was returned to the initial levels and the muscle was allowed to recover. As expected with hypercapnic acidosis the overall decrease in pHi was 0.33 ± 0.03 pH units, ~44% of the 0.75 ± 0.05 decrease in perfusate pH. This change in pHi was within the physiological range of 6.68-6.95 as measured in cardiac myocytes by similar methods (21).Representative pHi changes during hypercapnic acidosis are shown in Fig. 11. During control perfusion a pHi gradient was observed, with the surface pHi more acidic than the pHi of deeper endocardial layers ~200 µm below the muscle surface. When the PCO2 in the chamber and perfusate were increased, pHi decreased ~0.3 pH units. Consequently, the pHi gradient between the surface and deeper layers was reversed as the change in pHi in the deeper layers was greater than that at the surface. This observation suggests more active proton extrusion or buffering in the surface layers of the muscle. On return to control PCO2 the pHi returned to normal levels, with pHi of the deeper layers increasing more than the surface pHi.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Contemporary methods used to measure
pHi in tissue include
ion-selective microelectrodes and NMR. Although these methods
contribute enormously to our understanding of cellular pH regulation,
inherent technical restrictions of limited spatial resolution with
microelectrodes and decreased temporal resolution with NMR preclude
their practical application to the study of cellular pH regulation in
the vicinity of boundaries between different ionic and/or
metabolic environments or between cells in situ. Ratiometric
fluorescent microscopy with the pH-sensitive fluorophore SNARF-1 has
successfully measured pHi in
isolated cardiac cells (3, 5, 21, 24). Confocal microscopy,
when combined with ratiometric imaging of SNARF-1 in an arterially
perfused papillary muscle, provided a novel means to measure dynamic
intracellular and extracellular pH changes from a multicellular
preparation in compact ventricular myocardium. The principal findings
of this study show that quantitative measures of
pHi and
pHo are possible from both the
surface and deep endocardial tissue layers in perfused ventricular
tissue. The maximal depth to which
pHi or
pHo can be measured is limited by
the working distance of the objective lens and the penetration of the
excitation and emitted light. However, in practical terms the limit for
pH measurements, using fluorescent emissions from SNARF-1, is 
300 µm below the muscle surface.
Initial experiments addressed the uniformity of ratiometric measurements as a function of pH, image depth, and fluorophore concentration, thereby establishing the potential feasibility of this method for quantitative intramural pH measurements. The method was validated in an isolated, arterially perfused rabbit papillary muscle during normal arterial perfusion as pHo was altered by applying CO2 externally and by changing the perfusate buffer. Additionally, relative changes in pHi were evaluated during NH4Cl application and withdrawal and as CO2 in the atmosphere and perfusate were increased.
Confocal Microscopy, SNARF-1 Ratiometric Fluorescence Imaging, and Arterially Perfused Papillary Muscle Preparation
The determination of changes in intramural pHo and pHi in the perfused rabbit papillary muscle presented several technical challenges. Although confocal microscopy collects thin optical sections from thick specimens, acquisition of images from deep within the tissue required arterial perfusion to uniformly deliver the pH-sensitive fluorophore to the tissue. Additionally, confocal fluorescence microscopy allowed relatively fast image pair acquisition (e.g., ratiometric fluorescent images typically acquired in <20 s). Acquiring focused images of the beating papillary muscle was not feasible; therefore, BDM was added to the perfusate to inhibit electromechanical coupling and abolish contractions. Although BDM reversibly abolishes cardiac contraction, the cardiac myocytes continue to propagate transmembrane action potentials (17, 29). A study by Liu et al. (16) concluded that BDM was acceptable in electrophysiological studies despite a small decrease in conduction velocity and membrane conductances. When the muscle was mechanically quiescent, thin optical sections were obtained with the muscle aligned oblique to the focal plane, providing simultaneous fluorescence emissions from the endocardial surface and deeper endocardial layers. This technique provided increased temporal resolution compared with NMR and a spatial resolution suitable for multicellular tissue preparations (i.e., less than whole heart resolution but greater than single-cell resolution).Ratiometric Imaging of SNARF-1 Fluorescence
In comparison to other fluorophores available for the measurement of pH, SNARF-1 has several advantages. In the arterially perfused papillary muscle preparation the acetoxymethyl ester of SNARF-1 distributes homogeneously and is readily taken up by rabbit myocardium. Cytosolic esterases convert the membrane-permeable SNARF-1/AM to SNARF-1, which is a useful indicator for measuring pH changes ranging from 6.3 to 8.6 (11). In addition, the spectral characteristics of SNARF-1 permit ratiometric imaging from deep within the muscle, i.e., SNARF-1 is excited and emits fluorescence at longer wavelengths compared with other commonly used fluorophores like 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein. This property of SNARF-1 fluorescence is advantageous in this application because the long wavelength of the excitation and emission spectrum enhances the excitation and transmission of emitted fluorescence from the subendocardial cell layers. The longer wavelengths are attenuated less than shorter-wavelength light, and the lower excitation energy wavelengths reduce the potential for photobleaching and photodamage. Finally, the use of a single-excitation, dual-emission fluorophore versus a dual-excitation, single-emission fluorophore eliminated effects of chromatic aberrations on the ratio of the emitted fluorescence.A theoretical concern associated with the collection of emitted fluorescence from the deeper tissue layers is the loss of emitted light caused by absorption and scatter, which may severely restrict quantitative analysis of the image, i.e., the inner-filter effect. As shown in Figs. 4 and 9, emission wavelengths were attenuated as the concentration of the fluorophore decreased. Additionally, the measured intensity of the emitted light, from SNARF-1 homogeneously distributed in a semiopaque gel, decreased as the image depth increased (see Fig. 5). However, emission ratios of the fluorophore were independent of the concentration and focal plane depth, correcting for any inner-filter effects. Fluorescence ratios measured as a function of fluorophore concentration and path length resulted in a <2% difference between the ratios. This is in agreement with previous assessments of SNARF-1 as a pHi indicator (2, 23).
Loss of the intensity of the fluorescent emissions over time is
observed in isolated cells (2, 4, 28) and in our arterially perfused
tissue preparations. Not unexpectedly, loss of the intensity of SNARF-1
fluorescence over time was a significant limitation. A primary cause of
the decrease in fluorescence intensity is related to the leakage of
SNARF-1 from the cell. This was determined by the lack of any visible
fluorophore in the muscle preparation (i.e., SNARF-1-loaded muscles
were very pink) and low fluorescence emissions from multiple sites on
the perfused rabbit papillary muscle 60-90 min after SNARF-1
loading. Although photobleaching contributes to some loss in emission
intensity, Bassnett et. al. (2) showed that at a pH < 7.0 the
emission ratios of SNARF-1 appear to be independent of bleaching,
whereas the rate of photobleaching increases with pH 
7.3. The pHi of the perfused rabbit
papillary muscle under control conditions ranged from 7.04 to 7.23. Given the rate of loss of SNARF-1 from the intracellular space,
fluorescence imaging was only practical for ~60-90 min after the
cells were loaded. Studies show that probenecid, an anion-transport
blocker, reduces fluorophore loss when added to a perfusing solution
(1, 7). When probenecid (0.1 mM) was used in this preparation there was
no measurable effect on the rate at which fluorescence intensity emissions decreased; however, the application of this technique requires further study.
With the loss of SNARF-1 from the cells, an obvious concern is the
contribution of extracellular fluorescence to the
pHi measurement. Although we did
not measure the concentration of extracellular SNARF-1 caused by cell
loss, it is unlikely that cellular loss of SNARF-1 and accumulation
into the extracellular space made any significant contribution to the
fluorescence measurement. All muscles were continuously perfused with a
fluorophore-free solution (i.e., a perfusion flow rate of 1.3-1.5
ml · min1 · g
tissue1), which served to
constantly dilute any fluorophore leaking into the extracellular space.
Validation of pHi Calibration Technique
In isolated cardiac myocytes calibration of the intracellular fluorescence emission ratio and its relationship to pH is accomplished by incubating the preparation with a variety of ionophores (monensin, nigericin, etc.) and equilibrating pHi and pHo by perfusing with buffers of known pH (25). Studies show the intracellular calibration method to be superior to the in vitro method (2, 3, 20, 23), because the spectral characteristics of the dye are measured in the same environment during both the experiment and the calibration. Review of the literature suggests that because of the heterogeneity of the intracellular space and intracellular redistribution of SNARF-1 (5, 19, 20, 23), pHi measurements with SNARF-1 may involve significant errors when an in vitro method is used to calibrate pHi. These are important considerations when obtaining subcellular pHi measures. Initially, an in vivo calibration was attempted in the arterially perfused papillary muscle. However, this approach caused an abrupt and irreversible increase in arterial vascular resistance. In addition, visible heterogeneity of perfusion, heterogeneous distribution of the ionophore, and pH calibration solution proved this technique to be unreliable and impractical in this muscle preparation. Consequently, an in vitro calibration was used to establish a ratio scale. The major focus of this study was to evaluate relative changes in pHi from multicellular regions of the tissue rather than subcellular pHi. Hence, the in vitro calibration and resulting ratio scale proved to be an adequate method to monitor relative pH changes. A HEPES-buffered Tyrode solution was used for the pH calibration standards. We choose this simplified approach rather than try to duplicate the cytosolic constituents of the different cell types in the papillary muscle. Previous studies with single cells indicates that individual cells often have unique calibration curves, and these differences may be caused by the cytosolic constituents (3, 19, 23). It is possible that our calibration technique overestimated pHi. Nevertheless, using the in vitro calibration, when the muscle was perfused with M-Tyrode the average resting pHi measured 7.13 ± 0.03, which is consistent with values obtained with intracellular microelectrodes in the same muscle preparation (i.e., pHi of 7.03 ± 0.03; Ref. 30), thus validating this calibration technique. The relative changes in pHi and pHo are reliable, whereas the absolute values may be limited by the technical limitations noted. It should be noted that if absolute pHi is of prime importance an in vivo calibration should be used or the pHi verified with microelectrodes as was shown with the pHo measurements.Endocardial-to-Subendocardial pH Gradient
An initial finding during the development of the method was that pHo and pHi of the superficial endocardial cells were consistently more acidic than the pHo and pHi measured from cells ~200-300 µm below the surface. Although we can speculate why these differences exist, they are not likely to be related to a measurement artifact or an experimental artifact. The pH gradient was present in all experiments during steady state and was measured after the muscle preparations had reached equilibrium (i.e., 1 h after cannulation) and before any interventions. Indeed, the consistent measurement of pHo by two separate methods was shown as the perfusate was switched from a bicarbonate perfusate to a HEPES-buffered perfusate. Second, in a HEPES-buffered perfusate the pHo of the muscle decreased coincident with a decrease in perfusate pH, yet the pHo gradient remained fairly constant. Finally, with a bicarbonate-buffered perfusate the pHo gradient could be altered by externally decreasing and increasing CO2. Subsequently, the narrowing of pHi gradient, and in some cases the reversal of the pHi gradient, during the application of CO2 suggests that the local differences in pH regulation can be detected at the endocardium and deeper endocardial layers. In this case the greater effect of CO2 in the deeper layers indicates that the buffering capacity or the ability of the subendocardial cardiac myocytes to extrude protons is less than that of the superficial cells. Electron microscopy of the top 50 µm of the papillary muscle dissected from the right ventricle shows the surface of the papillary muscle to be covered with a layer of endothelial cells with underlying connective tissue measuring 20 ± 10 µm in depth (data not shown). The observed pHi differences may relate to this structural difference and the different characteristics and properties of the endocardial endothelial cells compared with ventricular myocardium. It is possible that in vivo the endocardial pHi has a different set point compared with the underlying ventricular myocardium. These differences may be related to local effects of endogenous vasoactive peptide mediators or endothelium-derived relaxing factor or to differences in intrinsic buffering capacity. The absence of a potentiation of the gradient between the pHi at the surface and that at deeper layers by externally applied CO2 is consistent with the known high diffusibility of CO2 and intracellular buffering capacity of myocardium.Future Applications
Integration of confocal fluorescent microscopy and the ratiometric measurement of the emission spectra of the H+-sensitive fluorophore SNARF-1 provides a reliable method to measure dynamic changes in pHi and pHo at the surface and deeper endocardial layers of the perfused ventricular myocardium. The high spatial and time resolution of this method makes it feasible to study the influence of diffusion of gases and metabolites on the regulation of ionic homeostasis in ischemic and reperfused myocardium bordering normal myocardium. Specifically, the role of the production, diffusion, and accumulation of CO2 for the regulation of H+ homeostasis in the ischemic border zone may be studied. The high spatial and temporal resolution of confocal microscopy used in conjunction with fluorophores sensitive to ions, metabolites, and other molecules may permit the study of intercellular signaling among endothelial cells, smooth muscle cells, and ventricular myocytes in tissue.| |
ACKNOWLEDGEMENTS |
|---|
The authors acknowledge the technical assistance of Darrell R. Sandiford, Samuel Tesfai, Ammasi Periasamy, Hal Mann, and David Lashley.
| |
FOOTNOTES |
|---|
This work was supported by National Heart, Lung, and Blood Institute Grants RO1 HL-48769, PO1-HL-27430, and T32-HL-07793 and Office of Naval Research Grant N00014-96-0283.
S. A. M. Marzouk was on leave from the Department of Chemistry, Ain Shams University, Cairo, Egypt.
Address for reprint requests: B. J. Muller-Borer, Univ. of North Carolina School of Medicine, Div. of Cardiology, CB#7075, Burnett-Womack Bldg., Chapel Hill, NC 27599-7075.
Received 26 February 1997; accepted in final form 22 July 1998.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
1.
Arkhammer, P. T.,
T. Nilsson,
and
P. O. Berggren.
Glucose-stimulated efflux of indo-1 from pancreatic beta-cells is reduced by probenecid.
FEBS Lett.
73:
182-184,
1990.
2.
Bassnett, S.,
L. Reinsch,
and
D. C. Beebe.
Intracellular pH measurement using single excitation-dual emission fluorescence ratios.
Am. J. Physiol.
258 (Cell Physiol. 27):
C171-C178,
1990
3.
Blank, P. S.,
H. S. Silverman,
O. Y. Chung,
B. A. Hogue,
M. D. Stern,
R. G. Hansford,
E. G. Lakatta,
and
M. C. Capogrossi.
Cytosolic pH measurement in single cardiac myocytes using carboxy-seminaphthorhodafluor-1.
Am. J. Physiol.
263 (Heart Circ. Physiol. 32):
H276-H284,
1992
4.
Buckler, K. J.,
and
R. D. Vaughn-Jones.
Application of a new pH-sensitive fluoroprobe (carboxy-SNARF-1) for intracellular pH measurement in small isolated cells.
Pflügers Arch.
417:
234-239,
1990[Medline].
5.
Chacon, E.,
J. M. Reece,
A. Nieminen,
G. Zahrebelski,
B. Herman,
and
J. J. Lemasters.
Distribution of electrical potential, pH, free Ca2+ and volume inside cultured adult rabbit myocytes during chemical hypoxia: a multiparameter digitized confocal microscopic study.
Biophys. J.
66:
942-952,
1994[Medline].
6.
Chilian, W. M.,
C. L. Eastman,
and
M. L. Marcus.
Microvascular distribution of coronary vascular resistance in beating left ventricle.
Am. J. Physiol.
251 (Heart Circ. Physiol. 20):
H779-H788,
1986
7.
Di Virgilio, F.,
J. A. Steinberg,
and
S. C. Silverstein.
Inhibition of fura-2 sequestration and secretion with organic anion transport blockers.
Cell Calcium
11:
57-62,
1990[Medline].
8.
Ellis, D.,
and
R. C. Thomas.
Direct measurement of the intracellular pH of mammalian cardiac muscle.
J. Physiol. (Lond.)
262:
755-771,
1976
9.
Furusawa, K.,
and
P. M. T. Kerridge.
The hydrogen ion concentration of the muscles of the cat.
J. Physiol. (Lond.)
63:
33-41,
1927.
10.
Garlick, P. B.,
G. K. Radda,
and
P. J. Seeley.
Studies of acidosis in the ischaemic heart by phosphorous nuclear magnetic resonance.
Biochem. J.
184:
547-554,
1979[Medline].
11.
Haugland, R. P.
Molecular Probes Handbook of Fluorescent Probes and Research Chemicals, 1992-1994 (5th ed.). Eugene, OR: Molecular Probes, 1992, p. 129.
12.
Hell, S. W.,
G. Reiner,
C. Cremer,
and
E. H. K. Stelzer.
Aberrations in confocal fluorescence microscopy induced by mismatches in refractive index.
J. Microsc.
169:
391-405,
1993.
13.
Hollis, D. P.,
R. L. Nunnally,
G. J. Taylor,
M. L. Weisfeldt,
and
W. E. Jacobus.
Phosphorus nuclear magnetic resonance studies of heart physiology.
J. Magn. Reson.
29:
319-330,
1978.
14.
Jacobus, W. E., I. H. Pores, S. K. Lucas, M. L. Weisfeldt, and J. T. Flaherty.
Intracellular acidosis and contractility in the normal and
ischemic heart examined by 31P
NMR. J. Mol. Cell. Cardiol. 4, Suppl. 3: 13-20, 1982.
15.
Kléber, A. G.,
and
C. B. Reigger.
Electrical constants of arterially perfused rabbit papillary muscle.
J. Physiol. (Lond.)
385:
307-324,
1987
16.
Liu, Y.,
C. Cabo,
R. Salomonsz,
M. Delmar,
J. Davidenko,
and
J. Jalife.
Effects of diacetyl monoxime on the electrical properties of sheep and guinea pig ventricular muscle.
Cardiovasc. Res.
27:
1991-1997,
1993
17.
Marijic, J.,
N. Buljubasic,
D. F. Stowe,
L. A. Turner,
J. P. Kampine,
and
Z. J. Bosnjak.
Opposing effects of diacetyl monoxime on contractility and calcium transients in isolated myocardium.
Am. J. Physiol.
260 (Heart Circ. Physiol. 29):
H1153-H1159,
1991
18.
Marzouk, S. A. M., S. Ufer, R. Buck, W. E. Cascio, and T. A. Johnson. Electrodeposited iridium oxide pH electrode for
measurement of extracellular myocardial acidosis during ischemia.
Anal. Chem. In press.
19.
Opitz, N.,
E. Merten,
and
H. Acker.
Evidence for redistribution-associated intracellular pK shifts of the pH-sensitive fluoroprobe carboxy-SNARF-1.
Pflügers Arch.
427:
332-342,
1994[Medline].
20.
Owen, C. S.
Comparison of spectrum-shifting intracellular pH probes 5'(and 6')-carboxy-10-dimethylamino-3-hydroxyspiro[7H-benzo[c]xanthene-7,1'(3'H)-isobenzofuran]-3'-one and 2',7'-biscarboxyethyl-5(and 6)-carboxyfluorescein.
Anal. Biochem.
204:
65-71,
1992[Medline].
21.
Ramsey, J.,
C. Austin,
and
S. Wray.
Differential effects of external pH alteration on intracellular pH in rat coronary and cardiac myocytes.
Pflügers Arch.
428:
674-676,
1994[Medline].
22.
Salhany, J. M.,
G. M. Piepers,
S. Wu,
G. L. Todd,
F. G. Clayton,
and
R. S. Eliot.
31P nuclear magnetic resonance measurement of cardiac pH in perfused guinea-pig hearts.
J. Mol. Cell. Cardiol.
11:
601-610,
1979[Medline].
23.
Seksek, O.,
N. Henry-Toulmé,
F. Sureau,
and
J. Bolard.
SNARF-1 as an intracellular pH indicator in laser microspectrofluorometry: a critical assessment.
Anal. Biochem.
193:
49-54,
1991[Medline].
24.
Spitzer, K. W.,
and
J. H. B. Bridge.
Relationship between intracellular pH and tension development in resting ventricular muscle and myocytes.
Am. J. Physiol.
262 (Cell Physiol. 31):
C316-C327,
1992
25.
Thomas, J. A.,
R. N. Buchsbaum,
A. Zimniak,
and
E. Rackler.
Intracellular pH measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ.
Biochemistry
18:
2210-2218,
1979[Medline].
26.
Vanheel, B.,
A. de Hemptinne,
and
I. Leusen.
Intracellular pH and contraction of isolated rabbit and cat papillary muscle: effect of superfusate buffering.
J. Mol. Cell. Cardiol.
17:
23-29,
1985[Medline].
27.
Waddell, W. J.,
and
R. G. Bates.
Intracellular pH.
Physiol. Rev.
49:
285-329,
1969
28.
Wieder, E. D.,
H. Hanf,
and
M. H. Fox.
Measurement of intracellular pH using flow cytometry with carboxy-SNARF-1.
Cytometry
14:
916-921,
1993[Medline].
29.
Wiggins, J. R.,
J. Reiser,
D. F. Fitzpatrick,
and
J. L. Bergey.
Ionotropic action of diacetyl monoxime in cat ventricular muscle.
J. Pharmacol. Exp. Ther.
212:
217-224,
1980
30.
Yan, G.,
and
A. G. Kléber.
Changes in extracellular and intracellular pH in ischemic rabbit papillary muscle.
Circ. Res.
71:
460-470,
1992
This article has been cited by other articles:
|
|
 |
|
  T. Manna, D. A. Thrower, S. Honnappa, M. O. Steinmetz, and L. Wilson Regulation of Microtubule Dynamic Instability in Vitro by Differentially Phosphorylated Stathmin J. Biol. Chem., June 5, 2009; 284(23): 15640 - 15649. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z.-Q. Xiong, P. Saggau, and J. L. Stringer Activity-Dependent Intracellular Acidification Correlates with the Duration of Seizure Activity J. Neurosci., February 15, 2000; 20(4): 1290 - 1296. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
= tan
y/
) and 585 ± 10 nm (
) were acquired 200 µm below surface of sample volumes.
B: corresponding emission intensity
ratios (
). Note that with decreasing fluorophore concentration
recorded emission intensities decreased, whereas ratios remained
constant.
) and extracellular pH
(pHo) in perfused rabbit
papillary muscle measured at endocardial surface from SNARF-1
fluorescence (
). Muscle was continuously perfused with SNARF-1 (20 µM) during
conditions in which CO2 was
matched in atmosphere and perfusate, as perfusate was switched to HEPES
perfusate with CO2 present, and as
CO2 was withdrawn from system with
100% O2 saturation.
) as perfusate pH was decreased from 7.37 to 6.54. Muscle was perfused with SNARF-1 (20 µM) in HEPES perfusate with
100% O2 saturation.
) and 585 ± 10 nm (
) and
estimated pHi (
