INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

REGULATION OF OXIDATIVE PHOSPHORYLATION is crucial for the survival of cells under physiological and pathophysiological fluctuations of either oxygen delivery or demand. By extrapolating in vitro data, one might think that oxygen availability hardly affects mitochondrial oxidative metabolism in normal hearts because the cytochrome c oxidase Michaelis-Menten constant (K_m) for oxygen is <1 mmHg in isolated mitochondria (8, 31), whereas in vivo coronary venous PO₂ is ~20 mmHg (21, 33). Nevertheless, the oxygen transport from capillary to mitochondria in in vivo cardiac tissue is not simple. Besides coronary capillary PO₂ extracellular (from capillary blood to plasmalemma) and intracellular (from plasmalemma to mitochondrial inner membrane) PO₂ gradients determine PO₂ at mitochondria. Both gradients increase as cellular oxygen demand increases, thus lowering PO₂ at mitochondria. Even in normal hearts, these oxygen gradients are so steep that small decreases in coronary oxygen supply bring intracellular PO₂ to ranges in which mitochondrial metabolism is reset (23, 28, 35). In these regulatory PO₂ ranges, in situ myoglobin (Mb) is partially desaturated (7, 14, 35, also see Ref. 1).

In the extracellular space, oxygen diffusion is relatively slow because this compartment lacks an oxygen carrier (such as Mb in the cytosol). Thus large oxygen concentration gradients are produced despite a very short (<0.5-2 µm) diffusion distance (13). In the intracellular space of cardiomyocytes, oxygen diffusion may be significantly accelerated by Mb-facilitated oxygen diffusion (18, 33). However, a much longer (~10 µm in rat ventricular myocytes) diffusion distance from the plasma membrane to the center of cell may also produce significant radial oxygen concentration gradients. The mitochondria located near the plasma membrane should then be exposed to substantially higher PO₂ than those located near the center of the cell. This allows measurement of adaptive changes in mitochondrial oxidative phosphorylation to PO₂ reductions within single cardiomyocytes.

We demonstrated that in isolated single rat cardiomyocytes treated with carbonyl cyanide m-chlorophenylhydrazone (CCCP) and at low extracellular PO₂ (<15 mmHg), there are radial PO₂ gradients within the cell such that the core oxygen pressure is near zero (27), and, as a result, the NADH fluorescence is significantly increased (26). These experiments certainly demonstrated that intracellular oxygen diffusion may limit mitochondrial oxidative phosphorylation. The question remained, however, as to how cardiomyocytes deal with such relatively slow intracellular oxygen diffusion. Our previous studies did not shed light on the role of physiological regulatory mechanism (the respiratory control) that should interact with diffusional oxygen supply because we disrupted the regulation by an uncoupler of oxidative phosphorylation. Thus we undertook the present study to demonstrate that mitochondrial respiratory control can compensate for such insufficient intracellular oxygen diffusion so that electron transfer is sustained in mitochondria locating in the hypoxic cell core.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

Isolation of single cardiomyocytes. Prior approval for the experiment was obtained from the Animal Research Committee, Yamagata University School of Medicine. Single cardiomyocytes were isolated from adult male Sprague-Dawley rats using type 2 collagenase (Worthington) and suspended in a HEPES buffer solution containing (in mM) 130 NaCl, 6 HEPES, 10 glucose, 5.4 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 2.5 pyruvate, 10 taurine, and 2 glutamate, supplemented with 0.1% bovine serum albumin. All the reagents used in this study were of the highest grade.

Increasing oxygen consumption. To address the effects of radial intracellular gradients of diffusional oxygen delivery on mitochondrial metabolism, it is important to significantly increase mitochondrial respiration because the magnitude of radial oxygen gradients is proportional to intracellular flux of oxygen. This was done by two distinct techniques. In the first group, mitochondrial oxygen consumption (O₂) was elevated with the use of 1 µM CCCP, an uncoupler of oxidative phosphorylation, in Ca²⁺-free HEPES solution at room temperature. In the second group, mitochondrial respiration was stimulated by electrical pacing of cardiomyocytes at 300 beats/min for 10 min in the presence of 2 µM norepinephrine and 1 mM Ca²⁺ at 36°C. For this purpose, a pair of 6 × 0.2-mm -platinum wires was placed on the bottom quartz glass in the measuring cuvette and connected to an electrical stimulator (model SEN-3301, Nihon Kohden).

Measurement of Mb oxygen saturation with subcellular spatial resolution. Intracellular heterogeneities of oxygen concentration were assessed by the spectrophotometric measurement of fractional oxygen saturation of Mb (S_Mb) with a subcellular spatial resolution. Reconstruction of intracellular oxygenation consists of the following steps. First, an aliquot of 6-µl cell suspension was placed on the airtight measuring cuvette on the stage of the inverted microscope (model IX-70, Olympus) and was superfused with humidified mixed gas of a given PO₂ (balance N₂). Transmitted light image of a single cell at 435.2 nm (an absorption peak of deoxy Mb) was captured using a 16-bit back-illuminated charge-coupled device (CCD) camera (model SV512, PixelVision). Second, cell suspension was consecutively exposed to anoxic superfusion gas (>99.999% N₂) for at least 5 min and the second cell image capturing was conducted. Third, the superfusion gas was again replaced with 20% oxygen for >5 min, followed by the third cell image capture. A blank-field image was obtained to calculate the optical density (OD) of respective cell images. Finally, S_Mb was calculated by the following equation (27), with a spatial resolution that one pixel on a computer monitor corresponds to 0.26 µm

(1)

All of the image arithmetic was carried out with IPLab software (Scanalytics).

(2)

View larger version (37K): [in this window] [in a new window]   Fig. 1.   Identifying myoglobin (Mb) light absorption in transmitted light image. A: transmitted light image of a cell at 435.2 nm was binalized, and regions with higher intensity are indicated in green. B: localization of mitochondria was visualized by rhodamine 123 fluorescence in the same cell and indicated in red. C: superimposition of images A and B revealed that regions with lower transmitted light intensity or higher optical density (OD) mainly represent mitochondria (cytochrome light absorption).

Measurement of mitochondrial oxidative metabolism with subcellular spatial resolution. Dependency of mitochondrial oxidative metabolism on diffusional oxygen delivery was assessed by the high spatial resolution imaging of mitochondrial NAD(P)H (26). We assumed that local increases in NADH indicate limitation of electron transfer in the respiratory chain due to insufficient oxygen supply in that particular region. Mitochondrial NADH was assessed from autofluorescence on ultraviolet (UV) excitation (9, 30).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

Oxygen consumption. In a suspension of cardiomyocytes containing ~2 × 10⁴ cells/ml, O₂ was determined by the conventional PO₂ electrode method (27). At room temperature (25°C) in the absence of extracellular Ca²⁺, 1 µM CCCP-stimulated mitochondrial respiration from 46 ± 19 to 334 ± 108 nmol O₂ · min¹ · 10⁶ cells¹ (n = 12). Normal hearts can increase steady-state oxygen utilization 20-fold (34). A 7.3-fold increase in O₂ in the present CCCP-treated quiescent cardiomyocytes may then represent a moderate increase.

1

1

Radial S_Mb gradients. In 1 µM CCCP-treated cardiomyocytes, the transition between full oxygenation and deoxygenation of Mb produced ~3% changes in transmitted light intensity (incident light intensity = 100%, Fig. 2B). When the same cell was superfused with hypoxic gas, radial profiles of transmitted light intensity near plasma membranes were similar to those in hyperoxia, whereas those near the center were close to those in anoxia (Fig. 2B). Thus radial gradients of S_Mb were demonstrated in a single cardiomyocyte (Fig. 2C). In the electrically paced cardiomyocytes, there were again significant parallel shift of OD profiles between hyperoxia and anoxia (Fig. 3A). Although the hypoxic profile was not completely analogous to those of hyperoxia and anoxia with respect to the absolute position of troughs (due to hypercontracture), calculation of S_Mb using corresponding trough values successfully demonstrated significant radial gradients of S_Mb (Fig. 3B).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Representative data demonstrating reconstruction of radial fractional Mb saturation (SMb) profile in a single cardiomyocyte treated with 1 µM carbonyl cyanide m-chlorophenylhydrazone (CCCP). A: a rectangular region of interest (ROI) was defined in the cell image and radial profile of transmitted light intensity was measured along the direction indicated by the arrow. Values were then averaged over adjacent 20 pixels (5.2 µm) along the long axis of a cell. B: radial profiles of OD when the cell was superfused with hyperoxic, hypoxic, and anoxic gases. Finally, SMb along the short axis of the cell was calculated by Eq. 1 (C).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Representative data demonstrating reconstruction of radial SMb profile in an electrically paced single cardiomyocyte. A: radial profile of OD within a rectangular ROI (5.2 µm wide) was measured for anoxic, hypoxic, and hyperoxic superfusion. To avoid contamination of cytochrome light absorption, only trough values of OD435-OD424 were used for SMb calculation (indicated by numbers). Note that absolute position of troughs differs between hyperoxic/anoxic and hypoxic profiles due to hypercontracture of cell. B: nadir SMb was estimated by interpolation using a hyperbolic function.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Radial SMb gradients in single cardiomyocytes with moderately increased respiration. A: 1 µM CCCP; B: pacing. SMb near plasma membranes and the center of the cell is shown, whereas the oxygen concentration of superfusion gas was changed. Peripheries 1 and 2 indicate regions ~4 µm inside from plasma membranes. *P < 0.05.

PO₂ in the hypoxic core. Before we compare mitochondrial metabolism in different intracellular oxygen supply, we should determine the level of hypoxia in the present study. The CCCP experiment was conducted at 25°C, whereas the pacing experiment was done at 36°C. The calculation of cell core PO₂ from S_Mb then needs adjustments for the temperature dependency of Mb oxygen binding. In the CCCP-treated cardiomyocytes, cell core S_Mb during superfusion with 2% and 4% oxygen were 0.32 and 0.75, respectively (Fig. 4). Because PO₂ at half-maximal saturation of Mb oxygen binding is 0.78 mmHg at 25°C (24), corresponding cell core PO₂ were 0.4 and 2.3 mmHg, respectively. In the electrically paced cardiomyocytes, the nadir S_Mb was 0.56 and 0.76 for 2% and 4% oxygen-superfused gases, respectively (Fig. 4), which corresponds to 2.6 and 6.4 mmHg with an assumption that PO₂ at half-maximal saturation is 2.03 mmHg at 36°C (24). These differences in the nadir intracellular PO₂ suggest that O₂ of the electrically paced cardiomyocytes might not have reached the level attained by CCCP. In isolated rat cardiomyocytes, Stumpe and Schrader (25) demonstrated a reversible downregulation of O₂ and contractile activity at a low PO₂ (6 mmHg) without any effect on cellular energy state. Thus an alternative explanation is that in electrically paced cardiomyocytes, mitochondria locating in the oxygen-depleted core might have downregulated their O₂, thus reducing radial PO₂ drop. In either case, we could produce a similar deoxygenation level in the cell core in cardiomyocytes with intact and abolished respiratory control (2% oxygen-superfused electrically paced cells and 4% oxygen-superfused CCCP-treated cells, respectively).

Radial gradients of mitochondrial oxidative metabolism in CCCP-treated cardiomyocytes. In cardiomyocytes, autofluorescence under UV epi-illumination mainly represents mitochondrial NADH (9). Thus the magnitude of NADH oxidation can be continuously assessed in single cardiomyocytes. At least three factors affect the level of NADH in mitochondria: activities of mitochondrial dehydrogenases, cytosolic energy state (phosphorylation potential, [ATP]/[ADP][P_i]), and cytochrome c oxidation ([c²⁺]/[c³⁺]) (31). The last one takes place at the complex IV of the respiratory chain and is quite sensitive to oxygen availability at mitochondria (5, 9). To assess oxygen dependency of mitochondrial metabolism, we compared NADH fluorescence in different regions within one cardiomyocyte where mitochondria were exposed to different PO₂.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Representative data demonstrating reconstruction of the radial NADH profile in a 1 µM CCCP-treated single cardiomyocyte. A: as in the SMb measurement, a rectangular ROI (5.2 µm wide) was defined and autofluorescence intensity was measured along the short axis of the cell. B: NADH fluorescence was normalized to that for the fully reduced state. au, Arbitrary units.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Radial NADH gradients in single cardiomyocytes with moderately increased respiration. NADH fluorescence intensities near plasma membranes and the center of cell are shown, whereas oxygen concentration of superfusion gas was changed. Peripheries 1 and 2 indicate regions ~4 µm inside from plasma membranes. In 1 µM CCCP-treated (A) and electrically paced cardiomyocytes (B), the NADH level at nonlimiting PO2 was 44 ± 5% and 63 ± 6% of full NADH reduction, respectively. *P < 0.05.

Insignificant anoxic core in electrically paced cardiomyocytes. Figure 7 illustrates the reconstruction of radial NADH profile in an electrically paced single cardiomyocyte. To assess the roles of mitochondrial respiratory control under insufficient mitochondrial oxygen supply, we compared NADH heterogeneities in the electrically paced cells and the CCCP-treated cells at similar cell core PO₂. In both 2% oxygen-superfused paced cells and 4% oxygen-superfused CCCP-treated cells, Mb near plasma membranes was almost fully saturated with oxygen (S_Mb = 0.93 ± 0.10 and 0.95 ± 0.05, respectively; see Fig. 4). Consequently, NADH in these regions was almost fully oxidized (11 ± 24% and 12 ± 10% reductions from the aerobic level, respectively; Fig. 6), indicating that mitochondrial oxygen metabolism in these regions was never limited by oxygen supply. In contrast, at the cell core, level of NADH reduction was significantly higher in the CCCP-treated cells (50 ± 14% reduction from the aerobic level; Fig. 6A) than in the electrically paced cells (9 ± 13% reduction from the aerobic level) at similar PO₂ (2.3 and 2.6 mmHg, respectively; Fig. 6B). These results indicate that electrically paced single cardiomyocytes can suppress accumulation of NADH caused by insufficient oxygen supply to mitochondria. Thus mitochondria with intact respiratory control seem to maintain almost normal electron transfer in the hypoxic core.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Representative data demonstrating reconstruction of NADH profile in an electrically paced single cardiomyocyte. A: as in the SMb measurement, a rectangular ROI (5.2 µm wide) was defined and autofluorescence intensity was measured along the short axis of the cell. B: NADH fluorescence was normalized to that for anoxic superfusion. To avoid inner filter effect (absorption of NADH fluorescence by Mb), only fluorescence peaks were used for calculation of radial NADH profile. Note that absolute position of peaks differs between anoxic and hypoxic profiles due to hypercontracture of the cell.

View larger version (12K): [in this window] [in a new window]   Fig. 8.   Magnitude of the anoxic core in single cardiomyocytes with moderately increased oxygen consumption. Elevations of NADH fluorescence at the cell core from regions near the plasma membrane were calculated. Because NADH fluorescence levels at nonlimiting oxygen differed significantly between the CCCP-treated and electrically paced cells (Fig. 6), values were represented as percent reductions from respective aerobic fluorescence levels (during superfusion with 20% oxygen). *P < 0.05.

Regulation of mitochondrial oxidative metabolism in oxygen depletion. In CCCP-treated and electrically paced single cardiomyocytes, mitochondria in relatively uniform biochemical environment but exposed to substantially different PO₂ behaved quite differently. Mitochondrial respiration and electron transfer in electrically paced cells are regulated by the mitochondrial respiratory control mechanism (i.e., dependency on energy state), whereas it is abolished in CCCP-treated cells. We conclude that mitochondria with intact respiratory control can continue electron transfer in the respiratory chain when diffusive oxygen supply is limited. This mechanism may maintain the electrochemical gradient across the mitochondrial membrane, the key parameter for mitochondrial functions. Thus mitochondrial physiological regulatory mechanism can compensate for the relatively slow oxygen diffusion within cardiomyocytes.

(3)