| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Laboratory of Cardiac
Energetics, The oxygenation
state of myoglobin and the redox state of cytochrome
c provide information on the
PO2 in the cytosol and mitochondria,
respectively. An optical "window" from ~540 to 585 nm was found
in the pig heart in vivo that permitted the monitoring of myoglobin and
cytochrome c without interference from
Hb oxygenation or blood volume. Scanning reflectance spectroscopy was
performed on the surgically exposed left ventricle of pigs. Difference
spectra between control and a total left anterior descending coronary
artery occlusion revealed maxima and minima in this spectral region
consistent with myoglobin deoxygenation and cytochrome c and
b reduction. Comparison of in vivo
data with in vitro fractions of the heart, including Hb-free tissue
whole heart and homogenates, mitochondria, myoglobin, and pig red blood
cells, reveals minimal contributions of Hb in vivo. This conclusion was
confirmed by expanding the blood volume of the myocardium and
increasing mean Hb O2 saturation
with an intracoronary infusion of adenosine (20 µg · kg
oxygen; mitochondria; oxygen consumption; facilitated diffusion; pig; dog; cytochrome b; cytochrome
c; diffusion; computer simulation; hemoglobin
THE LABILE VISIBLE LIGHT absorption of the heart in
vivo is mostly dependent on the oxygenation status of myoglobin and Hb, the redox state of the cytochromes, and myocardial blood volume. Thus
the reflectance spectroscopy can monitor each of these important parameters in myocardial O2
transport. Although visible optical spectroscopy has been used to study
the perfused heart, papillary muscles, and isolated myocytes (6, 12,
19, 20, 28, 42, 46), less work has been attempted on the heart in vivo (5, 25, 29, 34, 35, 39, 40, 43). Four factors make it difficult to use
reflectance spectroscopy on a beating, blood-perfused heart:
1) motion artifacts,
2) changes in blood volume,
3) spectral overlap between
absorption of myoglobin, Hb, and the cytochromes, and
4) limited depth of light
penetration. Infrared spectroscopy has been used to overcome the
limited path length but requires multicompartmental modeling to
quantify changes in Hb oxygenation and content, making the residual
measures of myoglobin oxygenation and cytochrome redox state difficult
(34, 35).
In preliminary studies the band at ~560 ± 20 nm of the reflected
optical spectrum appeared to be largely insensitive to absorbance changes related to Hb oxygenation and blood volume changes. The hypothesis was generated that this "window" in the optical
spectrum is not significantly influenced by Hb, permitting the direct
detection of cytochromes and myoglobin. To test this hypothesis,
comparisons between in vivo heart optical spectra with and without
blood as well as in vitro porcine blood and blood-free myocardial
homogenates were performed. Studies were also performed on porcine
hearts in vivo during normal perfusion, ischemia, and
adenosine-induced coronary vasodilation to evaluate the effects of
changes in blood volume and Hb oxygenation. Finally, studies were
conducted on the effect of increased workload on myocardial
oxygenation. Computer simulations of
O2 delivery to the myocardium were
also performed using recent functional and morphological data.
A secondary hypothesis was generated that the lack of Hb influence was
due to the short path length caused by the high optical absorption of
tissue chromophores and the low capillary content of red blood cells
(RBCs). This second hypothesis is developed further in the companion
article (15).
Animal preparation.
All animal care research protocols were approved by the Animal Care and
Use Committee at the National Institutes of Health and conform to the
standards of the American Physiological Society. Domestic swine of
either gender, weighing 25-55 kg, were studied in an acute
open-chest preparation. Animals were premedicated intramuscularly with
ketamine, xylazine, and butorphanol (Torbugesic). Anesthesia was
induced with
ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
APPENDIX
1 · min1),
which had no significant effect on the 540- to 585-nm region. These
results also suggested that myoglobin
O2 saturation was not blood flow
limited under these conditions in vivo. Work jump studies with
phenylephrine also failed to change cytochrome
c redox state or myoglobin
oxygenation. Computer simulations using recent physical data are
consistent with the notion that myoglobin O2 saturation is >92% under
basal conditions and does not change significantly with moderate
workloads. These studies show that reflectance spectroscopy can assess
myocardial oxygenation in vivo. Myoglobin
O2 saturation is very high and is
not labile to moderate changes in cardiac workload in the open-chest
pig model. These findings indicate that myoglobin does not contribute
significantly to O2 transport via
facilitated diffusion under these conditions.
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
APPENDIX
MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
APPENDIX
-chloralose (10 g/l) at a dose of 10-15 ml/kg iv.
Anesthesia was maintained with 5-10 ml/kg iv boluses every
1-2 h as needed or, in some animals, by a heated continuous infusion at 5-10
ml · kg1 · h1
to maintain adequate anesthesia. Ventilation was provided by a
hospital-grade servoventilator (model SV900, Siemens). Acid-base status
and blood gases were monitored. The arterial pH was maintained between
7.35 and 7.45 by adjusting the ventilator or giving intravenous bicarbonate. The PO2 was maintained
at >100 mmHg. Body temperature was maintained by keeping the
operating room temperature elevated (30-32°C) with a heated
water blanket and Mylar heat-reflecting blankets.
1 · min1 iv for 7 min) was used
to study the effects of an increase in cardiac work on myoglobin
oxygenation. Adenosine (20 µg · kg1 · min1
for 7 min) and total LAD occlusion were used to estimate the physiological dynamic range of the oxygenation effects in each animal
as well as the effects of Hb oxygenation and blood volume. Control data
were acquired for ~2.5 min before the phenylephrine infusion. Spectra
were then collected continuously through the infusion protocol. In some
animals it was necessary to repeat the phenylephrine dose after
repositioning of the optics because of heart movement that placed the
aperture mask out of the field of illumination (see below). After
proper adjustment of the optics, the aperture remained in the
illumination field throughout the control and experimental periods.
In a subset of animals the spectral effects of blood removal on the
ischemia difference spectrum were evaluated in the intact heart. Animals were also fitted with an inflatable occluder placed proximal to the first distal branch of the LAD, allowing generation of
transient reversible ischemia. Control blood-perfused spectra were collected at an acquisition rate of 100 ms for 3 min, and then
ischemia was initiated for an additional 3 min. No cardiac gating was used in these studies. Reperfusion spectra were also collected after occlusion to ensure restorative myocardial oxygenation. After reperfusion the superior and inferior vena cava were ligated distal to the base and apex, respectively, and a 1-cm-long right atrial
vent was placed between the ligatures. Cardioplegia was initiated by
passing ice-cold (2°C) buffer (153 mM NaCl, 16 mM MgSO4, and 16 mM KCl equilibrated
with 100% O2) at 200 mmHg
through a 14-gauge needle, which was inserted into the ascending aorta after cross clamping. Evacuated fluids were aspirated to minimize cross
contamination of blood on the surface of the heart. The clearance of
the blood was evaluated from observation of the venous effluent from
the heart. Oxygenated myocardial blood-free spectra in situ were
obtained as described for control blood-perfused animals in the
cold-arrested heart. Ischemia spectra were collected by
allowing the tissue to warm and collecting data at 10-min intervals until a stable ischemia spectrum was obtained. In all cases,
felt and dark current spectra were obtained for computation of
corrected optical density (OD).
Reflectance spectroscopy.
Localized reflectance spectroscopy was measured as shown in Fig.
1. A black felt mask (~25
cm2) was spot glued, with care
taken to avoid obvious surface vessels, with
n-butyl cyanoacrylate (Vetbond)
adhesive to the distal anterior LV myocardium in an area predicted to
be served by the LAD on the basis of surface diagonal branches. A round
~1-cm hole was cut out of the middle of the felt mask. This mask
served to localize the measurements to one discrete portion of the
myocardium. Regions free of obvious epicardial fat and blood vessels
were selected during positioning of the mask. Illumination light was
from a high-intensity tungsten source. A small fraction of the output was selected by an integrating sphere and projected onto the heart. The
amount of light was adjusted to optimize the dynamic range of the
detector. The projection and detection optics were in a nonconfocal
arrangement and oriented obliquely relative to the surface of the heart
to minimize specular reflections. Illuminating and detecting from a
larger region around the aperture in the mask ensured that the entire
region of interest was always detected. This greatly minimized
translational motion-related artifacts as well as concerns regarding
heterogeneity of the surface optical properties. Theoretically, motion
should cause intensity shifts because of changes in angle of incident
light, with no changes in the spectral characteristics. All data were
ratiometrically corrected for any spectral characteristics of the black
felt mask and detector with the use of a felt reference spectrum. All
changes in OD (
OD) reported were calculated by subtracting the
logarithm of reflected light under experimental conditions from the
logarithm of the reflected light under control conditions. This is
equivalent to the 
log(experimental condition/control).
|
Spectral fitting. The linear least-squares methodology follows from the previous work of French et al. (13). Spectral fitting was performed using a linear least-squares fitting routine resident in the SigmaPlot (Jandel) software package. Fits were evaluated from the R2 values as well as visual inspection of the frequency dependence of the residuals.
Blood flow measurements. Fluorescent microsphere blood flow was measured in a subset of animals. Approximately 2-2.5 × 106 E-Z Trac microspheres were injected into the left atrium before and during a given experimental perturbation. Different-color spheres were used for each measurement. Reference blood samples (20 ml) were obtained from the femoral vein at a rate of 10 ml/min during each injection. At the end of the study, myocardial samples taken from the LAD zone and the non-LAD zone were quantified through the E-Z Trac Investigator Partner Services (Interactive Medical Technologies, Los Angeles, CA).
Tissue samples. Triton X-100-solubilized pig heart homogenate was prepared as previously described (1). Briefly, animals were anesthetized and surgically prepared as discussed above, except for instrumentation on the heart itself. The animals were heparinized, and the hearts were rapidly removed. The hearts were then immediately perfused in a retrograde fashion via aortic cannulation with ice-cold saline until no evidence of blood was apparent in the effluent (1-2 liters of perfusate). Samples (5-10 g) of the free LV wall were then dissected free of fat and connective tissue and finely minced over ice. This sample was then homogenized in an equal weight of 100 mM phosphate buffer (pH 7.1). Care was taken not to heat the sample to prevent the formation of metmyoglobin. Aliquots of this homogenate were dissolved 1:5 in Triton X-100 to solubilize the tissue. Solid material was removed by gentle centrifugation, and the supernatant was used as a blood-free homogenate of porcine heart. Isolated pig heart mitochondria and Triton X-100 extracts were prepared as described by Mootha et al. (30).
Myoglobin determination. Hb-free tissue extracts of the pig heart were used to determine the concentration of myoglobin. Samples were split into two 1-ml aliquots. One sample (reference sample) was supplemented with 10 mM sodium ascorbate and 200 µM NaCN to reduce cytochromes c and a (1). The remaining sample (experimental sample) was completely deoxygenated and reduced using Na2S2O4. Difference spectra between the reference and experimental samples were collected from 450 to 620 nm on a scanning spectrophotometer (Perkin-Elmer). The reduced cytochrome a in the reference and the experimental sample resulted in only myoglobin contributing to the difference spectrum in the 595- to 620-nm region. Thus the concentration of myoglobin was determined by the absorbance difference between 595 and 620 nm in these difference spectra.
The extinction coefficient for horse myoglobin (Sigma Chemical, St. Louis, MO) at these wavelength pairs in oxygenated-deoxygenated spectra was determined on the same instrument. A myoglobin stock solution was first completely reduced by titration with Na2S2O4. The progress of the titration was followed optically. The reduced myoglobin was oxygenated by exhaustive room air bubbling. Samples were prepared with oxygenated and deoxygenated (retreated with Na2S2O4) myoglobin and run in a differential mode on the spectrophotometer. Protein concentrations from 0.2 to 0.8 mg/ml were measured, and the extinction coefficient for the difference between 620 and 595 nm (OD620 595)
in the oxygenated deoxygenated myoglobin was determined to be 4,715 OD620 595 · cm1 · M1
with a molecular weight of 16,900.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
APPENDIX
|
|---|
In vivo cardiac ischemia: control difference reflectance
spectra.
Control and ischemia reflectance spectra and the corresponding
difference spectrum from an in vivo pig heart are shown in Fig.
2, A and
B. The minimal changes below 435 nm
are presumably due to the high tissue absorbance at these wavelengths.
At longer wavelengths, calculated OD is negative (less absorbance
during ischemia). This was observed in most of the
ischemia studies. This systematic effect appears associated
with the swelling of the LV surface closer to the detector during
ischemia. In contrast to previous work on perfused hearts
during the transition from hypoxia to oxygenation, the physiological
constraints of the in vivo beating heart eliminated the ability to hold
the heart relative to the optics, resulting in the normal isobestic
points being nonzero. Spectral absorbance differences are best observed
from control-ischemia reflectance OD spectra (Fig.
2B). Minima in the OD spectrum
correspond to the maximal decreases in absorbance, which occur at 480, 535, and 582 nm. Maxima in the spectrum were at 520, 550, 563, and 605 nm.
|
OD spectra are not
at the wavelengths expected for changes in Hb content or Hb
O2 saturation. Thus we attempted
to determine the tissue chromophores contributing to this spectrum.
Toward this goal, the in vivo OD data were compared with in vitro
data collected from several fractions of heart, as summarized in Fig.
3. Blood-free pig myocardial homogenates
were used to represent total tissue chromophores, isolated porcine
mitochondria for the mitochondrial chromophores, purified horse
myoglobin for cytosolic myoglobin, and pig RBCs/Hb for blood
constituents. In general, difference spectra were collected with and
without
Na2S2O4
to remove O2 and reduce
cytochromes. Normalized spectra are presented for the blood-free homogenate representing all cellular chormophores, cytochromes, and
myoglobin in the naturally occurring ratio of concentrations. Blood-free pig myocardial homogenates treated with KCN (0.2 mM) and
sodium ascorbate (5 mM) selectively reduced cytochrome
c at 550 nm and cytochrome
a,a3 at 605 nm
(1). Purified myoglobin deoxygenation spectra have maxima at 563 nm and
minima at 581 nm. Cytochrome b has a
deoxygenation absorbance peak at 563 nm when blood-free, myoglobin-free
mitochondrial suspensions are reduced with
Na2S2O4.
|
OD spectrum corresponded
most closely with the
Na2S2O4
OD spectrum of the blood-free myocardial homogenate. The shoulder at
550 nm corresponds to cytochrome c.
The minimum at 581-582 nm corresponds to the myoglobin
deoxygenation not the minimum of pig blood Hb at ~577 nm.
Qualitatively, the relative maximum and relative minimum observed in
vivo correspond within 1 nm of the individual components identified in
the in vitro experiments. These data are consistent with the hypothesis that the in vivo ischemia absorbance changes in the 540- to
585-nm region are dominated by the cellular chromophores, mitochondrial cytochromes, and cytoplasmic myoglobin.
The extent to which Hb contributes to the in vivo
ischemia-related absorbance changes was evaluated
quantitatively by intentionally contaminating blood-free tissue
homogenates with pig blood and monitoring the spectral minimum in the
580-nm region. The in vivo spectra demonstrate a relative minimum in
the ischemia-to-control ratio spectrum at 581.9 ± 0.9 nm
(n = 6). In vitro deoxygenation absorbance spectra of blood-free myocardial homogenates had a minimum
at 581.3 ± 0.7 nm (n = 5) that was
not significantly different from that in vivo
(P > 0.05, t-test). Titration of extract minima with pig Hb is presented in Fig. 4. The
minimum frequency was very sensitive to Hb contamination. Contaminating
the blood-free myocardial homogenate with the equivalent of ~0.3%
RBC by volume shifted the minimum to 579.1 ± 1.0 nm, which was
significantly different from the pure homogenate
(P < 0.05, paired
t-test). Approximately 1%
contamination of myocardial homogenate with RBC shifted the minimum
even closer to that of pure blood (577.0 ± 0.2 nm,
P < 0.01, paired
t-test). Thus ~0.3% RBC volume
contamination would be predicted to start shifting the position of this
minimum by ~1-2 nm, which could be detected in the in vivo
spectra. It is also important to note that blood contamination of
>0.5% RBC volume resulted in complete masking of the cytochrome
c shoulder at 550 nm (not shown),
further supporting the notion that the blood contamination was very
small. These data indicate that absorption changes associated with
ischemia in vivo represent primarily myoglobin and the
cytochromes with very little contribution from Hb. The potential
reasons that Hb is not detected are reviewed in
DISCUSSION.
|
Spectral fitting. To attempt to quantify the contributions of cytochromes and myoglobin to the in vivo ischemia spectra, linear least-squares routines were used to fit the in vivo data. Two analyses were performed. The first involved fitting a combination of in vitro mitochondria and myoglobin (Fig. 5) difference spectrum to the in vivo ischemia difference spectrum. With use of the following equations, a linear least-squares fit of Eq. 3 to the in vivo spectra was performed
|
(1) |
|
(2) |
|
(3) |
OD spectra collected in vitro,
Imito and
IMb are the relative amplitudes of
the mitochondrial and myoglobin contributions, C1 and
C2 are constants,
and Amito and
AMb are the
amplitude-adjusted mitochondria and myoglobin absorbance spectra that,
in combination, provided the best fit (f) to the in vivo
spectrum. The models used and the results of the spectral
fitting routine are presented in Fig. 5. Good fits of the in vivo data
with the mitochondria-myoglobin model were obtained using this approach
with minimal residuals (R2 = 0.97 ± 0.03, n = 6). The
IMb/Imito
ratio for the models used was 29.4 ± 3.0 (n = 5) for in vivo spectra and 29.3 ± 1.0 (n = 4) for the blood-free
Triton X-100 extracts. The similarity of in vivo and in vitro
IMb/Imito
ratios suggests that the relative amounts of myoglobin and mitochondria
chromophores are the same in the blood-free extracts and in vivo. To
confirm this, the in vitro blood-free homogenate spectrum was the only
model system used to fit the in vivo data with the same linear
least-squares strategy. The single blood-free homogenate spectrum was
found to be very similar
(R2 > 0.96, n = 5) to the in vivo spectrum. Fits
using Hb alone or Hb and mitochondria model spectra resulted in
consistently lower R2 values. These
results support the hypothesis that the Hb contamination to the in vivo
optical spectrum was minimal and that spectral contributions from
cytochrome c and myoglobin alone, in
the relative concentrations found in tissue extracts, were adequate to
fit the in vivo data in this spectral region.
|
OD spectrum of the intact heart
with and without blood was evaluated. In Fig.
5B, the ischemic OD spectra of the
same heart perfused in situ with blood and thoroughly perfused with
cardioplegia solution are presented. A linear regression was also
performed in the region from 540 to 615 nm between the two data sets to
quantitatively evaluate the shape differences of the spectra. The
absorption minimum at ~580 nm is maintained along with the cytochrome
shoulder at ~550 nm in both spectra. The linear regression in the
540- to 585-nm window showed minimal residuals with
R2 > 0.99. Beyond the 585-nm point the absorbance differences between the
conditions increased as the tissue absolute absorbance decreased and
effective optical path length through the tissue increased (Fig.
5B). These data confirm that the
OD spectrum between 540 and 585 nm is minimally perturbed by tissue
Hb content. However, above this window, significant differences in the
ischemic difference spectrum are observed between blood-perfused and
blood-free tissue.
Effects of motion on reflectance spectrum.
The felt mask is intended to ensure that only light coming from the
open aperture is collected for analysis. This ensured that the same
region of the myocardium was evaluated during the experimental
perturbations, minimized the inclusion of large surface blood vessels,
and reduced the influence of translational motion on the amplitude of
the reflected signal. By illuminating a region on the felt mask larger
than the open aperture, the open aperture was evenly illuminated and
detected. Amplitude variations still occurred as a result of the heart
moving closer to or farther from the optics and changes in the angle of
incidence of the light with the rotation of the heart. An example of a
motion-related artifact is shown in Fig. 6,
where the signal intensity at five wavelengths is plotted as a function
of time after an R wave trigger. The amplitude variation across the
cardiac cycle is on the same order of magnitude as the
ischemia-specific absorbance changes. However, the relative
spectral characteristics of the light are unchanged in several spectral
bands. All wavelengths between 540 and 585 nm have similar time
courses, whereas 592 nm is equal in intensity to the other wavelengths
at some times but at other times deviates significantly. This is
demonstrated spectrally at six phases of the cardiac cycle ranging from
midsystole to late diastole in Fig. 7.
Despite baseline offsets, the spectra remain flat across the cardiac
cycle from 540 to 585 nm but not at longer wavelengths such as 592 nm.
Thus cardiac motion does not introduce spectral changes in the 540- to
585-nm region, despite large amplitude offsets introduced by
respiratory and cardiac motion.
|
|
Epicardial myoglobin oxygenation.
The oxygenation status of myoglobin and the redox state of the
cytochromes were evaluated in vivo by comparing ischemia and adenosine experiments. Full-scale deoxygenation of myoglobin and reduction of the cytochromes were defined by the absorbance changes after 2.5 min of total LAD occlusion in the beating heart. Full-scale oxygenation of myoglobin was defined by near-maximal coronary vasodilation achieved by intracoronary adenosine infusion (20 µg · kg1 · min1).
An example of the reflectance spectra and the absorbance change spectrum associated with myocardial ischemia is shown in Fig. 2. During ischemia, OD between 581.5 and 559 nm
(OD581.5 559), a wavelength pair-sensitive myoglobin oxygenation, was only
0.046 ± 0.036. This represents a maximal amplitude change
associated with the deoxygenation of myoglobin. In contrast,
near-maximal coronary vasodilation with intracoronary adenosine
(coronary blood flow was 2.9 ± 0.6 times control, Table
1) had a minimal effect on
the reflectance spectra (Fig. 8) within the bandwidth
from 540 to 585 nm. On average,
OD581.5 559
was only +0.001 ± 0.004 during adenosine infusion
(P < 0.05 vs. ischemia, paired t-test,
n = 6). Thus adenosine did not
significantly change absorbance at wavelength pairs maximally sensitive
to the oxygenation status of myoglobin. Because of the lack of spectral
response to adenosine, the linear least-squares analysis was not
performed. These data indicate that myoglobin is fully oxygenated
within the limits of this measurement technique and under these
experimental conditions. Hemodynamic characteristics are summarized in
Table 1. In paired experiments on animals with this instrumentation and
anesthesia (n = 8), coronary venous
PO2 averaged 23.1 ± 2.1 mmHg and
varied by <1.6 mmHg across multiple interventions. In addition,
hematocrit averaged 29.2 ± 3.1 for these young animals and was
stable across time.
|
|
Effect of myocardial workload.
The hemodynamic results for the phenylephrine protocol are presented in
Table 1. Optical spectra for the phenylephrine study are summarized in
Fig. 9. The difference between
559 and 581 nm in the control vs. phenylephrine infusion was not
significant at 0.003 ± 0.002 OD (paired
t-test,
n = 6).
|
Myoglobin concentration.
The myoglobin concentration in the pig heart was determined because of
the discrepancies in the literature. The elimination of blood and
cytochrome contamination is critical in any optical assessment of
myoglobin in heart tissue. In Fig. 10,
difference spectra for
Na2S2O4-treated
vs. raw extract, which will have a myoglobin and a cytochrome
contribution, and
Na2S2O4-treated
vs. cytochrome-reduced (KCN and sodium ascorbate treatment) extract, which should have only myoglobin contributions, are presented for
comparison. The extract was devoid of a significant contribution from
Hb on the basis of the difference spectra minimum at 581 nm (see
above). The contribution of cytochrome to the difference spectrum was
minimized by the prior reduction of the cytochromes. This was best seen
as the removal of the cytochrome c
shoulder at 550 nm and the shift of the absorbance maximum in the
600-nm region from a mixture of cytochrome
a and myoglobin to pure myoglobin at
595 nm. The 595-nm peak was used to avoid contamination from cytochrome
b that is not fully reduced by the
cyanide-ascorbate treatment.
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
APPENDIX
|
|---|
With use of a combination of a physical aperture and appropriate optics, reflection spectroscopy data were collected from the in vivo pig heart. These results were consistent with the notion that the 540- to 585-nm region of the reflected optical spectrum is dominated by cytoplasmic myoglobin and mitochondrial cytochromes. Vascular Hb O2 saturation did not significantly influence the 540- to 585-nm region of the in vivo spectrum. The vasodilation studies suggest that the myoglobin oxygenation is high and that the myoglobin oxygenation and cytochrome c redox states are not affected by alterations in workload two to three times the resting anesthetized states.
The hypothesis that the 540- to 585-nm band is not influenced by vascular Hb is supported by the comparison of in vitro data with in vivo spectra as well as physiological and nonphysiological manipulations in vivo. The difference spectrum between the control and ischemic heart had minima (581 nm for myoglobin) and maxima (550 nm for cytochrome c and 563 nm for myoglobin) similar to Hb-free tissue extract, suggesting that Hb was not significantly contributing to the difference spectrum. The 581-nm minimum is a key finding, since it differentiates between Hb and myoglobin in this spectral region. The 550-nm shoulder for cytochrome c clearly shows that the cytoplasmic chromophores are contributing to the reflectance spectrum. The isobestic points (545 and 558 nm) in the ischemia spectra are also consistent with the blood-free saline-perfused rabbit heart (21). Finally, all these characteristics in the 540- to 585-nm region were maintained in the ischemic difference spectrum of the blood-free pig heart. Titration of tissue extracts with blood indicates that the RBC contamination must be less than ~1% RBC to maintain an absorption minimum at 581 nm.
Beyond the maxima and minima of the spectra, the spectral shape of the in vivo ischemia difference spectrum was fit using a linear model. The model components were 1) in vitro oxidized-reduced spectra from isolated mitochondria representing cytochromes c and b as well as oxy-deoxymyoglobin spectra from pure myoglobin or 2) in vitro control Na2S2O4 spectra from blood-free myocardial homogenates representing mitochondrial and cytoplasmic chromophores in the naturally occurring mole fractions. Good fits of the in vivo data were obtained with either model. It is important to note that mitochondria and myoglobin alone could be used to fit the in vivo spectrum and that the appropriate ratio of these components, found in vivo, was maintained in the fit without including a vascular Hb component. Attempts using Hb and mitochondria models resulted in much poorer fits to the in vivo data.
These in vitro comparisons suggest that vascular Hb does not contribute to the reflected visible spectrum of the in vivo pig heart and that this region of the spectrum is dominated by cytoplasmic chromophores.
To confirm the lack of vascular Hb contribution in the 540- to 585-nm region in vivo, adenosine was used to increase the blood volume of the myocardium and to increase the mean Hb O2 saturation by raising venous PO2. The coronary infusion of adenosine increased blood flow more than threefold (Table 1) and should have roughly doubled tissue blood volume (26, 31) and increased venous Hb saturation >70% (2). Despite these large changes in Hb content and oxygenation, adenosine infusion resulted in no specific absorption changes in the 540- to 585-nm wavelengths. Outside the 540- to 585-nm region, especially at >600 nm, alterations in the reflectance spectrum occurred that are likely the result of blood volume or O2 saturation changes. The effects observed in the >600-nm region are consistent with the lower tissue-blood absorption in this region, leading to more blood sensitivity. Similar results were obtained in the comparison of blood-perfused and blood-free hearts (Fig. 5B). Previous studies have also observed spectral changes of >600 nm with ischemia in the dog heart (39) that may be caused by blood volume changes. The effects of adenosine in vivo, where induced increases in blood volume and oxygenation did not affect the 540to 585-nm region, are consistent with a minimal influence of Hb in this region of the reflectance spectrum in vivo.
The contribution of blood to the in vivo spectral characteristics through 540-585 nm was estimated to be <0.5% of RBC volume per volume of myocardium on the basis of in vitro measures. This conclusion is consistent with the simulations of photon migration in the pig heart (15). This blood content is much lower than the values for whole heart hematocrit (i.e., RBC volume / tissue volume) obtained with invasive techniques, which vary from ~2 to 9% depending on the methodology (11, 17, 32, 36). The relative insensitivity of the ischemia absorbance changes to the total tissue blood Hb may be related to the high light absorption of Hb, myoglobin, and the cytochromes at these wavelengths as well as the dynamic range of absorption changes expected for Hb. The extinction coefficient for myoglobin from 540 to 585 nm is ~2.5 times higher than at 460-510 nm and ~6-8 times higher than at 600-640 nm. Effectively, this means that the path length in the 540- to 585-nm band will be much shorter than that at wavelengths with lower absorbance. Shorter path length translates into lower probability of interaction with highly dispersed RBCs in the tissue. In addition, the RBC Hb absorption of light in this bandwidth is about two orders of magnitude higher than that of tissue. Because of this high RBC absorbance, a large fraction of photons (540-585 nm) that enter a significant blood vessel never escape for detection (15). Thus, as light enters and scatters through the myocardium, the exiting photons are highly weighted to those that only scatter through the cytosol and never encounter an RBC. A theoretical analysis of this process is presented in the companion article with use of Monte Carlo simulations and the physical data collected from the pig heart (15).
With the assumption that only the smallest vessels contribute to the
spectrum, the majority of the blood detected optically will be in the
capillary space. A capillary volume of 8% for the pig (45) and the
lowest estimate of cardiac capillary hematocrit of 12% (38) still
result in an RBC volume of ~1%, which is still above the detection
threshold established in the in vitro titration experiments (Fig. 4).
However, a significant fraction of heart wall blood is on the venous
side of the capillary bed, where the O2 saturation is low. Thus any
further decrease in O2 saturation with ischemia will be small relative to the entire dynamic
range used in the in vitro titration. Thus a reduction in the Hb
contribution by a factor of 
2 could occur, because a large amount of
Hb is not saturated with O2 under
control conditions. In the presence of adenosine the venous Hb
O2 saturation increases to
~60-70%. However, no evidence of Hb contamination was observed
with subsequent ischemia in these hearts, despite the fact that
the dynamic range of Hb O2
saturation was increased. This suggests that the effect of Hb dynamic
range was not dominant in determining the "visibility" of Hb in
these studies. We speculate that the combination of these two effects,
reduced influence of large vessels and reduced dynamic range of the
venous Hb, may explain the lack of Hb contribution to the ischemic
difference spectrum.
The adenosine infusion experiment also provides some insight with regard to O2 delivery to the myocardium. Because adenosine increases the venous Hb O2 saturation, it is assumed that the myoglobin and cytochrome redox state will trend to a more oxygenated and oxidized state, respectively, if PO2 was limiting under control conditions. No changes in myoglobin or cytochrome redox state were observed with adenosine infusion. Furthermore, it has been shown that the cytochromes are highly oxidized and myoglobin fully oxygenated in the Triton X-100 tissue extracts (1). Thus the Na2S2O4 extract vs. control difference spectrum should represent the difference between completely oxidized and reduced cytochrome as well as between oxy- and deoxymyoglobin. As shown in the model fitting, this in vitro extract difference spectrum provided a good model of the in vivo control-ischemia spectrum collected. Because ischemia should induce a near-complete reduction of the cytochromes and deoxygenation of myoglobin in vivo, it is reasonable to reach the conclusion that the agreement between the extract and in vivo difference spectra also implies that myoglobin was nearly fully oxygenated and the cytochromes were highly oxidized. Both of these results, the adenosine effects and model fitting, are consistent with the hypothesis that the myoglobin is nearly fully oxygenated and that the PO2 at the mitochondrion is in excess of that required to oxidize the cytochrome chain under control conditions. These results in the pig are similar to those of Chen et al. (7) using 1H NMR to detect deoxymyoglobin in the exposed dog heart. In these NMR studies, no deoxymyoglobin was detected under control conditions with experimental limitations on the order of ±10%. These NMR data also suggest that myoglobin is nearly maximally saturated with O2 under control conditions.
In contrast, invasive studies suggest that the mean myoglobin O2 saturation is only ~50%. Coburn et al. (9) using 14CO methods, Losse et al. (27) using O2-sensitive electrodes, and Gayeski et al. (16) using microspectrophotometry on rapidly frozen tissue sections found that the mean heart myoglobin saturation was ~50% (PO2 ~4 mmHg on the basis of affinity of myoglobin used), suggesting that O2 could be limiting for oxidative phosphorylation in significant volumes of the tissue (9). The current optical study is not consistent with these previous studies, since no evidence for a significant amount of deoxymyoglobin or O2 limitation to the redox state of cytochrome c was found. The reasons for this discrepancy between the noninvasive data (optics and NMR) and previous invasive techniques are unknown. Potentially, the fact that tissue samples were collected or electrodes were inserted into the tissue may have contributed to lower PO2 values recorded with the invasive procedures.
To evaluate whether a high mean myoglobin oxygenation is physically realistic, an O2 delivery model was developed for the in vivo pig heart. This model was derived from the work of Groebe (18) for skeletal muscle. The details of the model are presented in the APPENDIX along with the results. This three-dimensional simulation reveals that the total myoglobin O2 saturation could be on the order of 92% (see Table 3). Simulation of the adenosine flow changes results in an increase in myoglobin saturation to 95% (see Table 3). This predicted 3% change in saturation is well within the noise of the current optical studies. Similar conclusions were found in the control state of the dog heart (see Table 4), consistent with the 1H NMR data. Thus a physical model of O2 delivery in the myocardium is consistent with the hypothesis that myoglobin O2 saturation is >90% under control conditions.
The effect of work was evaluated to gain some insight into myocardial O2 delivery. Increasing cardiac work with systemic infusions of phenylephrine caused no change in the myoglobin oxygenation or cytochrome c redox state (Fig. 9). Similar results were obtained in three studies with dobutamine or aortic constriction to increase work (not shown). These results suggest that the O2 delivery was adequate to maintain tissue PO2, despite two- to threefold increases in O2 consumption. This is similar to the conclusions of previous invasive studies that found the apparent muscle PO2 to be constant over a wide range of flows and workloads, even though the absolute PO2 was determined to be very low, as discussed above. Simulation of these workload data also reveals that the measured increases in coronary flow were adequate to maintain the total myoglobin saturation at very high levels (see Table 3). This was also true for the dog heart in vivo with use of data for near-maximal workloads (44), where large declines in myoglobin oxygenation were observed only at the highest workloads imposed (see Table 4).
Facilitated diffusion of O2 by myoglobin depends on a gradient of oxymyoglobin in the cytosol (47). The cytosolic oxymyoglobin gradient must be small, since myoglobin is nearly completely oxygenated under control and moderate workload conditions. The combined effects of high myoglobin O2 saturation (present study; 7), low myoglobin diffusion coefficient in muscle (23), and relatively low concentration of myoglobin in heart result in a minimal contribution of myoglobin to O2 transport. The simulations suggest that the myoglobin contribution to O2 transport is <1% over the entire myocardium. Myoglobin-facilitated O2 diffusion approaches 5% in the 10% of the myocardium capillary diffusion bed farthest from entry of the arterial blood. At the workloads reached in this study and those simulated for the dog (see APPENDIX), the contribution of myoglobin to facilitated diffusion seems to be minimal in the heart until near-maximal rates of respiration are achieved. Similar conclusions were reached by Cole et al. (10) for the perfused dog heart after reduction of myoglobin facilitated diffusion with inhibitors. Myoglobin-facilitated O2 diffusion may play a more important role at very high workloads, ischemia, or hypoxic conditions. However, under normal physiological conditions, myoglobin may serve more as a temporal buffer, reducing oscillations in tissue PO2 as a result of phasic flow and metabolic activity. Funk et al. (14) proposed a similar temporal buffer role for another large cellular metabolic buffer in the heart, creatine phosphate. Thus both of these large buffer pools in the heart, oxymyoglobin and creatine phosphate, may be more important for the temporal maintenance of cellular homeostasis and not for cytosolic transport of metabolites.
The lack of change in tissue PO2 with workload also implies that the gross tissue levels of O2 are not contributing to the regulation of O2 consumption or regional blood flow within the dynamic range of these measures. With regard to metabolism, the measures of cytochrome c redox state are interesting since an increase in metabolic activity in isolated mitochondria, induced by ADP or Pi increases, result in a net oxidation of cytochrome c (4). No evidence for modification in cytochrome c redox state was observed in this study with moderate increases in workload. However, the cytochrome c redox change associated with full transitions from rest to maximum velocity in mitochondria is only 8% (4). The moderate transitions in workload [~30% of the maximum respiratory rate (30)] made in this study should cause changes in cytochrome c well below the signal detection levels for this in vivo measurement.
With regard to the regulation of blood flow, the vasodilation associated with the workload challenges was not associated with significant changes in oxymyoglobin or cytochrome c. These results suggest that global decreases in cellular PO2, within the sensitivity limits of these measures, are not required for increases in coronary flow induced by increases in afterload, the main phenylephrine effect. Thus a simple hypoxia feedback loop is not adequate to explain the regulation of blood flow, as has been suggested previously by many investigators (3).
One of the major limitations of this optical approach is the limited
path length that samples only the epicardial region of the myocardium.
In the companion article (15), the mean path length was estimated to be
~1.3 mm with a mean penetration depth of ~180 µm (400-µm mean
maximum depth). The in vivo optical measure was still heavily weighted
by tissue dependence on vascular supply of
O2 and not surface diffusion in
this open-chest preparation, since large decreases in myoglobin
oxygenation were observed with the removal of vascular
O2 supply (ischemia). This
was confirmed by calculating the diffusion penetration of
O2 from the surface, with
myoglobin diffusion facilitation, by using the physical parameters in
the APPENDIX and the one-dimensional
facilitated diffusion equation of Wyman (49). In Fig.
11, PO2
is supported by surface diffusion from the epicardium at different
workloads. The maximum penetration was only 40 µm. These
calculations suggest that a small component of the epicardium is
supported via nonvascular diffusion of
O2 in an open-chest preparation.
It is important to note that the
1H NMR data, which are not as
critically attenuated transmurally as the optical signals, yield
similar data in the intact dog heart [i.e., no detectable
deoxymyoglobin under control conditions (7)].
|
No quantitation of myoglobin O2 saturation or cytochrome redox state is provided. This limitation is due to the inability to reach the extremes in O2 delivery to establish the 0 and 100% points. A regional infusion of adenosine was used in an attempt to reach maximum PO2. However, simulations (see APPENDIX) suggest that even this threefold increase in blood flow was not adequate, with the myoglobin saturation only reaching 95%. Limitations of the O2 delivery model are outlined in the APPENDIX. We know of no way to attain fully oxidized cytochrome c in vivo. Even regional ischemia may not be adequate to generate complete anoxia, since surface diffusion and collateral perfusion could result in some O2 delivery. This technology, like many nondestructive in vivo tools, is limited to directional changes by use of internal controls. However, this qualitative information can still provide useful insights into physiological regulation in vivo.
Finally, the optical spectroscopy technology and 1H NMR sampled a rather large region of the myocardium. Large gradients in myoglobin saturation or cytochrome c redox state could exist in small regions of the myocardium that would not be easily detected with these gross measures, as shown in the simulations. Thus conclusions of this study apply only to an average response over a ~0.8-cm2 area of the epicardium. However, if any highly labile steep O2 gradients do exist (22), they must not make up a significant volume of the myocardium. This restriction limits the distribution of these gradients to highly localized regions of limited total volume in the myocardium if they are going to be missed with use of these approaches.
| |
APPENDIX |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
APPENDIX
|
|---|
Recent quantitative data on the geometry and distribution of
capillaries in the pig heart (24, 45) as well as information on
myoglobin O2 affinity (37),
content (this study), and diffusion in muscle (23) provide new insights
for a calculation of the mean myoglobin saturation in the pig heart in
vivo. Another area of some controversy is the myoglobin concentration
in the pig heart. Recent fractionation studies have suggested a pig
heart myoglobin concentration of only 8 × 105 mol/kg (33). However,
we directly determined the myoglobin content of pig hearts to be 3.6 × 104 mol/kg, which
is more consistent with previous studies in other animal species (for
examples see Ref. 48). The reason for this discrepancy is unknown.
The basic model of Groebe (18) was used for these simulations with the
use of the data referred to above to modify it for heart muscle.
Definitions of variables and input data are presented in Table
2. This model takes into account the RBC
distribution and O2 unloading
along the capillary, O2 diffusion
in the carrier-free region (CFR) between the capillaries and tissue,
and the facilitated diffusion of
O2 by Hb in the RBC and myoglobin
in the muscle cell. The basic geometry for the model is shown in Fig.
12.
|
|
The model was previously described in detail by Groebe (18) for skeletal muscle. We have adapted this model to geometric and physical parameters associated with the pig and dog heart (Table 2). The first step was the calculation of the PO2 at the outside of the capillary RBC [P(z)RBC] as a function of a longitudinal coordinate (z) along the length of the capillary. This is accomplished by using the global values of capillary flow, tissue O2 consumption, and blood O2 content. In this model it is assumed that all the capillaries in a given region of the heart are participating in O2 delivery. The unloading of O2 from the RBCs follows the model of Clark et al. (8).
The diffusion of O2 through the CFR is modeled by using the following equation at each PRBC along the capillary length (z) and the radial coordinate (r) up to the myoglobin-containing tissue (assumed to be 3.5 µm from the center of the capillary)
|
![− [<A><AC>V</AC><AC>˙</AC></A><SC>o</SC><SUB>2</SUB>(<IT>R</IT><SUP><IT>2</IT></SUP><SUB><IT>k</IT></SUB><IT> − R</IT><SUP>2</SUP><SUB>CFR</SUB>)]/(2<IT>D</IT><SUB>CFR</SUB>&agr;<SUB>CFR</SUB><IT>l</IT><SUB>RBC</SUB>) ln (<IT>r</IT>/<IT>R</IT><SUB>RBC</SUB>)](/content/vol277/issue2/fulltext/H683/img005.gif)
|
(A1) |
O2 is
O2 consumption,
RCFR is the
radius of CFR,
DCFR is
O2 diffusion coefficient in CFR,
CFR is
O2 solubility in CFR,
lRBC is length of
RBC in capillary, and
RRBC is radius of
RBC. This simulation resulted in a two-dimensional array
PCFR(z,r),
providing the PO2 as a function of
z and
r in the CFR.
The effective PO2 at the CFR-tissue boundary for each capillary position [P(z)CFR limit] is then used to calculate the radial PO2 through the myocardium [Pm(z,r)]. Within the muscle, myoglobin-facilitated O2 diffusion is included and the muscle PO2 is calculated as a function of z and r as follows
|
![+ (<A><AC>V</AC><AC>˙</AC></A><SC>o</SC><SUB>2</SUB>/<IT>D</IT><SUB>m</SUB>&agr;<SUB>m</SUB>)[0.5(<IT>r</IT><SUP>2</SUP> − <IT>R</IT><SUP>2</SUP><SUB>CFR</SUB>) − <IT>R</IT><SUP>2</SUP><SUB>K</SUB> ln (<IT>r</IT>/<IT>R</IT><SUB>CFR</SUB>)]](/content/vol277/issue2/fulltext/H683/img007.gif)
|
(A2) |
|
(A3) |
![P<SUB>m</SUB>(<IT>z</IT>, <IT>r</IT>) = 0.5D<SUB>F</SUB> + 0.5[D<SUP>2</SUP><SUB>F</SUB> + 4P<SUB>eff</SUB> (<IT>z</IT>, <IT>r</IT>)P<SUB>50</SUB>]<SUP>0.5</SUP>](/content/vol277/issue2/fulltext/H683/img009.gif)
|
(A4) |
m is
O2 solubility of muscle,
DMb is myoglobin diffusion coefficient in muscle,
CMb is total myoglobin concentration in heart, and P50 is
half-saturation PO2.
DF is an intermediate term and has no physical meaning. The
muscle myoglobin saturation [SMb(z,r)]
was determined by numerically solving the following equation for each
position in the muscle
![S<SUB>Mb</SUB>(<IT>z</IT>, <IT>r</IT>) = P<SUB>m</SUB>(<IT>z</IT>, <IT>r</IT>)/[P<SUB>m</SUB>(<IT>z</IT>, <IT>r</IT>) + P<SUB>50</SUB>]](/content/vol277/issue2/fulltext/H683/img010.gif)
|
(A5) |
The relative contributions of O2
(DO2) and
oxymyoglobin (DOMb) diffusion to
the overall O2 transport
(
O2)
were determined using the following relationship
|
(A6) |
|
(A7) |
|
(A8) |
|
(A9) |
O2, DO2, and
DOMb were also estimated in the
regions of the lowest PO2 by
integrating a 10% volume of the muscle with the lowest
PO2.
These equations were numerically solved for the entire capillary
diffusion domain using IDL (RSI). An example of the
Pm and myoglobin
O2 saturation
(SMb) in the pig heart is shown
in Fig. 13. The capillary and CFR regions have been
excluded for clarity.
|
The effects of changes in cardiac perfusion and workload were modeled using the physiological parameters for control, adenosine, and phenylephrine from this study (Table 2) and for the in vivo dog described by von Restorff et al. (44) (see Table 4). No attempt to modify the functional capillary density, or Hill coefficient, with workload was made in these simulations.
Three-dimensional plots for the pig heart simulation are presented in
Fig. 13 for the muscle PO2 and in
Fig. 14 for myoglobin O2
saturation under control conditions. The CFR zone was omitted from
Figs. 13 and 14 for clarity. Because of the high affinity of myoglobin
for O2, despite the large drop in
tissue PO2 down the capillary, the
myoglobin remains highly saturated with
O2. Summaries of the data for the
pig are found in Table 3. In addition to the myoglobin
saturation, the end-capillary PO2 is
shown, which is also in good agreement with experimental values. The
lowest PO2 calculated was always well
above the critical PO2 for oxidative
phosphorylation in this model. The relative contribution of
myoglobin to O2 transport over the entire
muscle volume was calculated to be <3%. This is even true if it is
calculated for the 10% of the tissue with the lowest tissue
PO2.
|
|
Simulations were also performed for the in vivo dog; because this is a
common model for cardiovascular studies,
1H NMR studies have found little
deoxygenated myoglobin under resting conditions, and excellent data on
blood flow and metabolic rate with exercise are available. These data
are summarized in Table 4. Under control and moderate
workloads, myoglobin remained highly oxygenated, as observed in the pig
studies and the previous 1H NMR
studies in dogs. Only at the highest workloads, or cardiac metabolic
rates, did the total myoglobin saturation decrease markedly. At the
highest workloads achieved by von Restorff et al. (44), some tissue
hypoxia was predicted by the model, suggesting that O2 limitation may be limiting
performance. A similar conclusion was reached in the dog with regard to
the contribution of myoglobin to
O2 transport. When the whole
muscle is considered, the myoglobin contributed <1% to the
O2 transport. At the highest
workloads in dogs, myoglobin supported ~20% of the
O2 transport in the 10% of the
tissue with the lowest PO2.
|
Any simulation of complex physiological processes is limited by the inherent simplicity of the model used as well as the accuracy of the physiological, anatomic, and physical data utilized. Of particular concern in the present model are the diffusion coefficient of myoglobin and the fraction of the capillaries available for O2 delivery as a function of workload. Furthermore, it has been shown that countercurrent flow exists in the capillaries of the heart (41). This countercurrent effect would decrease the tissue gradients even further. For these reasons and because of other limitations, this model is not accurate. However, within our current understanding of the physical and physiological parameters concerning O2 delivery to the heart, this simulation provides some useful insights. First, the observation by both methods that myoglobin is nearly fully saturated with O2 in the in vivo pig heart is not physically impossible. The short diffusion distances and high flow contribute to this phenomenon. Second, steep gradients in myoglobin oxygenation will not be detected by these gross measures, which integrate the entire myoglobin pool, and not selectively the regions at risk. Third, these results suggest a limited role for myoglobin in O2 transport in the heart in vivo under control conditions because of its low concentration, small diffusion coefficient, and maintenance of relatively high PO2. Fourth, no evidence for an O2 limitation of oxidative phosphorylation was found, except near maximum rates of work in the dog.
| |
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: R. S. Balaban, National Heart, Lung, and Blood Institute, National Institutes of Health, Bldg. 10, Rm. B1D416, MS 1016, Bethesda, MD 20892 (E-mail: rsb{at}zeus.nhlbi.nih.gov).
Received 26 February 1998; accepted in final form 22 February 1999.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
APPENDIX
|
|---|
1.
Balaban, R. S.,
V. K. Mootha,
and
A. Arai.
Spectroscopic determination of cytochrome c oxidase content in tissues containing myoglobin or hemoglobin.
Anal. Biochem.
237:
274-278,
1996[Medline].
2.
Balaban, R. S.,
J. F. Taylor,
and
R. Turner.
Effect of cardiac flow on gradient recalled echo images of the canine heart.
NMR Biomed.
7:
89-95,
1994[Medline].
3.
Broten, T. P.,
J. L. Romson,
D. A. Fullerton,
D. M. Van Winkle,
and
E. O. Feigl.
Synergistic action of myocardial oxygen and carbon dioxide in controlling coronary blood flow.
Circ. Res.
68:
531-542,
1991
4.
Chance, B.,
and
C. M. Williams.
The respiratory chain and oxidative phosphorylation.
Adv. Enzymol.
17:
65-134,
1956.
5.
Chance, B.,
J. R. Williamson,
D. Famieson,
and
B. Schoener.
Properties and kinetics of reduced pyridine nucleotide fluorescence of the isolated and in vivo rat heart.
Biochem. Z.
341:
357-377,
1965.
6.
Chapman, J. B.
Fluorometric studies of oxidative metabolism in isolated papillary muscle of the rabbit.
J. Gen. Physiol.
59:
135-154,
1972
7.
Chen, W.,
J. Zhang,
M. H. Eljgelshoven,
Y. Zhang,
X. H. Zhu,
C. Wang,
Y. Cho,
H. Merkle,
and
K. Ugurbil.
Determination of deoxymyoglobin changes during graded myocardial ischemia: an in vivo 1H NMR spectroscopy study.
Magn. Reson. Med.
38:
193-197,
1997[Medline].
8.
Clark, A.,
W. J. Federspiel,
P. A. A. Clark,
and
G. R. Cokelet.
Oxygen delivery from red cells.
Biophys. J.
47:
171-181,
1985[Medline].
9.
Coburn, R. F.,
F. Ploegmakers,
P. Gondrie,
and
R. Abboud.
Myocardial myoglobin oxygen tension.
Am. J. Physiol.
224:
870-876,
1973.
10.
Cole, R. P.,
B. A. Wittenberg,
and
P. R. Caldwell.
Myoglobin function in the isolated fluorocarbon-perfused dog heart.
Am. J. Physiol.
234 (Heart Circ. Physiol. 3):
H567-H572,
1978.
11.
Crystal, G. J.,
H. F. Downey,
and
F. A. Bashour.
Small vessel and total coronary blood volume during intracoronary adenosine.
Am. J. Physiol.
241 (Heart Circ. Physiol. 10):
H194-H201,
1981.
12.
Fabel, H.,
and
D. W. Lubbers.
Measurements of reflection spectra of the beating heart, in situ.
Biochem. Z.
341:
351-356,
1965.
13.
French, S.,
P. R. Territo,
and
R. S. Balaban.
Correction for inner filter effects in turbid media: fluorescence assays of mitochondrial NADH.
Am. J. Physiol.
275 (Cell Physiol. 44):
C900-C909,
1998
14.
Funk, C.,
A. Clark,
and
R. J. Connett.
How phosphocreatine buffers cyclic changes in ATP demand in working muscle.
Adv. Exp. Med. Biol.
248:
687-692,
1989[Medline].
15.
Gandjbakhche, A. H.,
R. F. Bonner,
A. E. Arai,
and
R. S. Balaban.
Visible-light photon migration through myocardium in vivo.
Am. J. Physiol.
277 (Heart Circ. Physiol. 46):
H698-H704,
1999
16.
Gayeski, T. E. J.,
and
C. R. Honig.
Intracellular PO2 in individual cardiac myocytes in dogs, cats, rabbits, ferrets, and rats.
Am. J. Physiol.
260 (Heart Circ. Physiol. 29):
H522-H531,
1991
17.
Gonzalez, F.,
and
J. B. Bassingthwaighte.
Heterogeneities in regional volumes of distribution and flows in rabbit heart.
Am. J. Physiol.
258 (Heart Circ. Physiol. 27):
H1012-H1024,
1990
18.
Groebe, K.
An easy-to-use model for O2 supply to red muscle. Validity of assumptions, sensitivity to errors in data.
Biophys. J.
68:
1246-1269,
1995[Medline].
19.
Hassinen, I. E.,
J. K. Hiltunen,
and
T. E. S. Takala.
Reflectance spectrophotometric monitoring of the isolated perfused heart as a method of measuring the oxidation-reduction state of cytochromes and oxygenation of myoglobin.
Cardiovasc. Res.
15:
86-91,
1981[Medline].
20.
Heineman, F. W.,
and
R. S. Balaban.
Effects of afterload and heart rate on NAD(P)H redox state in the isolated rabbit heart.
Am. J. Physiol.
264 (Heart Circ. Physiol. 33):
H433-H440,
1993
21.
Heineman, F. W.,
V. V. Kupriyanov,
R. Marshall,
T. A. Fralix,
and
R. S. Balaban.
Myocardial oxygenation in the isolated working rabbit heart as a function of work.
Am. J. Physiol.
262 (Heart Circ. Physiol. 31):
H255-H267,
1992
22.
Jones, D. P.
Intracellular diffusion gradients of O2 and ATP.
Am. J. Physiol.
250 (Cell Physiol. 19):
C663-C675,
1986
23.
Jurgens, K. D.,
T. Peters,
and
G. Gros.
Diffusivity of myoglobin in intact skeletal muscle cells.
Proc. Natl. Acad. Sci. USA
91:
3829-3833,
1994
24.
Kassab, G. S.,
and
Y.-C. B. Fung.
Topology and dimensions of pig capillary network.
Am. J. Physiol.
267 (Heart Circ. Physiol. 36):
H319-H325,
1994
25.
Kedem, J.,
A. Mayevsky,
J. Sonn,
and
B. A. Acad.
An experimental approach for evaluation of the O2 balance in local myocardial regions in vivo.
Q. J. Exp. Physiol.
66:
501-514,
1981
26.
Liu, Y.-H.,
R. C. Bahn,
and
E. L. Ritman.
Microvascular blood volume-to-flow relationships in porcine heart wall: whole body CT evaluation in vivo.
Am. J. Physiol.
269 (Heart Circ. Physiol. 38):
H1820-H1826,
1995
27.
Losse, B.,
S. Schuchhardt,
and
N. Niederle.
The oxygen pressure histogram in the left ventricular myocardium of the dog.
Pflügers Arch.
356:
121-132,
1975[Medline].
28.
Makino, N.,
H. Kanaide,
R. Yoshimura,
and
M. Nakamura.
Myoglobin oxygenation remains constant during the cardiac cycle.
Am. J. Physiol.
245 (Heart Circ. Physiol. 14):
H237-H243,
1983.
29.
Mills, S. A.,
F. F. Jobsis,
and
A. V. Seaber.
A fluorometric study of oxidative metabolism in the in vivo canine heart during acute ischemia and hypoxia.
Ann. Surg.
186:
193-200,
1977[Medline].
30.
Mootha, V. K.,
A. E. Arai,
and
R. S. Balaban.
Maximum oxidative phosphorylation capacity of the mammalian heart.
Am. J. Physiol.
272 (Heart Circ. Physiol. 41):
H769-H775,
1997
31.
Morgenstern, C.,
U. Holjes,
G. Arnold,
and
W. Lochner.
The influence of coronary pressure and coronary flow on intracoronary blood volume and geometry of the left ventricle.
Pflügers Arch.
340:
101-111,
1973[Medline].
32.
Myers, W. W.,
and
C. R. Honig.
Number and distribution of capillaries as determinants of myocardial oxygen tension.
Am. J. Physiol.
207:
653-660,
1964.
33.
O'Brien, P. J.,
H. Shen,
L. J. McCutcheon,
M. O'Grady,
P. J. Byrne,
H. W. Ferguson,
M. S. Mirsalimi,
R. J. Julian,
J. M. Sargeant,
R. R.M. Tremblay,
and
T. E. Blackwell.
Rapid, simple and sensitive microassay for skeletal and cardiac muscle myoglobin and hemoglobin: use in various animals indicates functional role of myohemoproteins.
Mol. Cell. Biochem.
112:
45-52,
1992[Medline].
34.
Parsons, W. J.,
J. C. Rembert,
R. P. Bauman,
F. G. Duhaylongsod,
J. C. Greenfield, Jr.,
and
C. A. Piantadosi.
Myocardial oxygenation in dogs during partial and complete coronary artery occlusion.
Circ. Res.
73:
458-464,
1993
35.
Parsons, W. J.,
J. C. Rembert,
R. P. Bauman,
J. C. Greenfield, Jr.,
and
C. A. Piantadosi.
Dynamic mechanisms of cardiac oxygenation during brief ischemia and reperfusion.
Am. J. Physiol.
259 (Heart Circ. Physiol. 28):
H1477-H1485,
1990
36.
Rakusan, K.
Vascular capacity and hematocrit in experimental cardiomegaly due to aortic constriction in rats.
Can. J. Physiol. Pharmacol.
49:
819-823,
1971[Medline].
37.
Schenkman, K. A.,
D. R. Marble,
D. H. Burns,
and
E. O. Feigl.
Myoglobin oxygen dissociation by multiwavelength spectroscopy.
J. Appl. Physiol.
82:
86-92,
1997
38.
Silverman, D. A.,
and
K. Rakusan.
Red blood cell spacing in rat coronary capillaries during the cardiac cycle.
Microvasc. Res.
52:
143-156,
1996[Medline].
39.
Snow, T. R.,
L. H. Kleinman,
J. C. Lamanna,
A. S. Wichsler,
and
F. F. Jobsis.
Response of cytochrome a,a3 in the in situ canine heart to transient ischemic episodes.
Basic Res. Cardiol.
76:
289-304,
1981[Medline].
40.
Snow, T. R.,
E. Vanoli,
G. De Ferrari,
M. Stramba-Badiale,
and
D. T. Dickey.
Response of cytochrome a,a3 to carbon monoxide in canine hearts with prior infarcts.
Life Sci.
42:
927-931,
1988[Medline].
41.
Steinhausen, M.,
H. Tillmanns,
and
H. Thederan.
Microcirculation of the epimyocardial layer of the heart.
Pflügers Arch.
378:
9-14,
1978[Medline].
42.
Tamura, M.,
N. Oshino,
B. Chance,
and
I. A. Silver.
Optical measurements of intracellular oxygen concentrations of rat heart in vitro.
Arch. Biochem. Biophys.
191:
18-22,
1978.
43.
Thorniley, M. S.,
A. Lahiri,
B. Glenville,
C. Shurey,
G. Baker,
U. Ravel,
J. Crawley,
and
C. J. Green.
Non-invasive measurement of cardiac oxygenation and haemodynamics during transient episodes of coronary artery occlusion and reperfusion in the pig.
Clin. Sci. (Colch.)
91:
51-58,
1996[Medline].
44.
Von Restorff, W.,
J. Holtz,
and
E. Bassenge.
Exercise induced augmentation of myocardial oxygen extraction in spite of normal dilatory capacity in dogs.
Pflügers Arch.
372:
181-185,
1977[Medline].
45.
White, F. C.,
L. Nimmo,
Y. Nakatani,
and
C. M. Bloor.
Compensatory angiogenesis during progressive right ventricular hypertrophy.
Am. J. Cardiovasc. Pathol.
4:
51-68,
1992[Medline].
46.
Williamson, J. R.,
and
D. Jamieson.
Metabolic effects of epinephrine in the perfused rat heart. Comparison of intracellular redox states, tissue PO2, and force of contraction.
Mol. Pharmacol.
2:
191-205,
1966
47.
Wittenberg, B. A.,
and
J. B. Wittenberg.
Transport of oxygen in muscle.
Annu. Rev. Physiol.
51:
857-878,
1989[Medline].
48.
Wittenberg, J. B.
Myoglobin-facilitated oxygen diffusion: role of myoglobin in oxygen entry into muscle.
Physiol. Rev.
50:
559-632,
1970
49.
Wyman, J.
Facilitated diffusion and the possible role of myoglobin as a transport mechanism.
J. Biol. Chem.
241:
115-121,
1966
This article has been cited by other articles:
|
|
 |
|
  J. Feygin, A. Mansoor, P. Eckman, C. Swingen, and J. Zhang Functional and bioenergetic modulations in the infarct border zone following autologous mesenchymal stem cell transplantation Am J Physiol Heart Circ Physiol, September 1, 2007; 293(3): H1772 - H1780. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Hogan, C. M. Stary, R. S. Balaban, and C. A. Combs NAD(P)H fluorescence imaging of mitochondrial metabolism in contracting Xenopus skeletal muscle fibers: effect of oxygen availability J Appl Physiol, April 1, 2005; 98(4): 1420 - 1426. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Takahashi and K. Asano Mitochondrial respiratory control can compensate for intracellular O2 gradients in cardiomyocytes at low PO2 Am J Physiol Heart Circ Physiol, September 1, 2002; 283(3): H871 - H878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Schenkman Cardiac performance as a function of intracellular oxygen tension in buffer-perfused hearts Am J Physiol Heart Circ Physiol, December 1, 2001; 281(6): H2463 - H2472. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. French, C. Giulivi, and R. S. Balaban Nitric oxide synthase in porcine heart mitochondria: evidence for low physiological activity Am J Physiol Heart Circ Physiol, June 1, 2001; 280(6): H2863 - H2867. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhang, K. Ugurbil, A. H. L. From, and R. J. Bache Myocardial oxygenation and high-energy phosphate levels during graded coronary hypoperfusion Am J Physiol Heart Circ Physiol, January 1, 2001; 280(1): H318 - H326. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. Gandjbakhche, R. F. Bonner, A. E. Arai, and R. S. Balaban Visible-light photon migration through myocardium in vivo Am J Physiol Heart Circ Physiol, August 1, 1999; 277(2): H698 - H704. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Papadopoulos, V. Endeward, B. Revesz-Walker, K. D. Jurgens, and G. Gros Radial and longitudinal diffusion of myoglobin in single living heart and skeletal muscle cells PNAS, May 8, 2001; 98(10): 5904 - 5909. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
