Beta-cell replacement and regeneration: Strategies of cell-based therapy for type 1 diabetes mellitus.

Pancreatic islet transplantation has demonstrated that long-term insulin independence may be achieved in patients suffering from diabetes mellitus type 1. However, because of limited availability of islet tissue, new sources of insulin producing cells that are responsive to glucose are required. Development of pancreatic beta-cell lines from rodent or human origin has progressed slowly in recent years. Current experiments for ex vivo expansion of beta cells and in vitro differentiation of embryonic and adult stem cells into insulin producing beta-cell phenotypes led to promising results. Nevertheless, the cells generated to date lack important characteristics of mature beta cells and generally display reduced insulin secretion and loss of proliferative capacity. Therefore, much better understanding of the mechanisms that regulate expansion and differentiation of stem/progenitor cells is necessary. Here, we review recent advances in the identification of potential cellular sources, and the development of strategies to regenerate or fabricate insulin producing and glucose sensing cells that might enable future cell-based therapies of diabetes mellitus type 1.

Human Krüppel-like factor 11 inhibits human proinsulin promoter activity in pancreatic beta cells.

AIMS/HYPOTHESIS: The Krüppel-like factor 11 (KLF11; TIEG2), a pancreas-enriched Sp1-like transcription factor, is a known negative regulator of pancreatic exocrine cell growth. A recent study indicated KLF11-induced activation of the human proinsulin promoter (hInsP). MATERIALS AND METHODS: We investigated the functional role of KLF11 in pancreatic beta cells. RESULTS: Endogenous KLF11 mRNA expression was found in whole rat pancreas, human pancreatic islets and INS-1E beta cells and was profoundly reduced by high glucose in INS-1E. Cotransfections of INS-1E and beta-TC3 beta cells with a human (h)KLF11 expression plasmid and an hInsP-driven reporter plasmid resulted in a substantial dose-dependent and glucose-independent inhibition of proinsulin promoter activity. 5'-deletion of hInsP demonstrated that hKLF11 acts via DNA sequences upstream of -173 and requires the beta cell-specific transcription machinery, since hKLF11-mediated inhibition of promoter activity was abolished in HEK293 cells. Besides a previously described GC box, we further identified a CACCC box within the hInsP, both putative KLF11-binding motifs. Electrophoretic mobility shift analysis (EMSA) verified binding of in vitro translated hKLF11 to the GC box, but neither hKLF11-induced inhibition nor basal hInsP activity was altered by mutation or 5'-deletion of the GC box. In contrast, CACCC box mutation substantially reduced basal promoter activity and partially diminished hKLF11 inhibition, although binding of in vitro translated hKLF11 to the CACCC box could not be verified by EMSA. CONCLUSIONS/INTERPRETATION: In rodent beta cell lines, we demonstrate hKLF11overexpression-mediated inhibition [corrected] of human proinsulin gene expression and characterise a prominent role for the CACCC box in maintaining basal proinsulin promoter activity.

Tumor necrosis factor alpha induces the expression of the nuclear protein p8 via a novel NF kappaB binding site within the promoter.

p8 is a widely expressed HMG-I/Y-like transcription factor which is involved in regulating cell proliferation and tissue stress. Several studies describe a strong upregulation of p8 expression during inflammatory processes like pancreatitis and LPS-induced sepsis. Here we demonstrate that TNFalpha, which is an important inducer of innate defence against gram-negative bacteria, significantly stimulates p8 protein production in H4IIE rat hepatoma cells within 2 hours. Since a putative NF kappaB motif has been described, we further tested whether TNFalpha stimulates p8 expression via activation of NF kappaB. We characterized the TNFalpha-induced binding of NF kappaB to this motif. We show that the TNFalpha-induced NF kappaB pathway contributes to the induction of p8 during pancreatitis and LPS-induced inflammation.

Human breast adipocytes express interleukin-6 (IL-6) and its receptor system: increased IL-6 production by beta-adrenergic activation and effects of IL-6 on adipocyte function.

Adipocytes produce the inflammatory cytokine interleukin-6 (IL-6); however, it is not known whether these cells express the IL-6 receptor system, how the secretion of this cytokine is regulated, and whether it has a function within adipose tissue. Using cultured human breast adipocytes, we investigated the expression of IL-6 and its receptor system, the effects of IL-6 on main adipocyte functions, and the regulation of IL-6 secretion by catecholamines and glucocorticoids. In the culture system, immunohistochemistry demonstrated expression of IL-6 and its receptor system, consisting of the ligand-binding IL-6 receptor and the signal-transducing protein gp130, in mature adipocytes, but not in undifferentiated adipocyte precursor cells. In freshly isolated adipocytes, RT-PCR detected messenger ribonucleic acids encoding the above proteins. Chronic incubation of adipocytes with 1 nmol/L IL-6 during adipose differentiation reduced glycero-3-phosphate dehydrogenase (GPDH) activity, a marker of adipocyte differentiation, and triglyceride synthesis to 67 +/- 9% of the basal level (mean +/- SEM; P < 0.05) only on day 21. Incubation of differentiated adipocytes with 10 nmol/L IL-6 for 24 h also resulted in a reduction of GPDH activity to 81 +/- 5% (P < 0.05). On the other hand, 24-h exposure to 10 nmol/L IL-6 increased basal glycerol release by 42 +/- 12% (P < 0.01) and isoproterenol-induced glycerol release by 21 +/- 6% (P < 0.05). The same concentration of IL-6, however, did not alter basal or insulin-stimulated glucose transport. IL-6 secretion was acutely and chronically stimulated by 1 micromol/L isoproterenol (peak of 6.2-fold after 3 h; P < 0.001) and only moderately suppressed by 100 nmol/L cortisol (-36 +/- 10%; P < 0.001). In conclusion, human breast adipocytes release substantial amounts of IL-6 and express IL-6 receptor and gp130. The secretion of IL-6 by adipocytes is strongly stimulated by beta-adrenergic activation and is modestly suppressed by glucocorticoids. IL-6 reduces GPDH activity and stimulates lipolysis, suggesting an autocrine/paracrine role of this cytokine in human adipose breast tissue.

Immunohistochemical and ultrastructural localization of leptin and leptin receptor in human white adipose tissue and differentiating human adipose cells in primary culture.

Leptin is mainly produced in white adipose tissue and acts both at distant sites and locally at the tissue from which it originates. The cellular and subcellular localization of leptin and its receptor (Ob-receptor [Ob-R]) and their relationship to various stages of fat cell maturation have not been characterized as yet. Therefore, we analyzed leptin and Ob-R by using reverse transcriptase-polymerase chain reaction, immunohistochemistry, and ultrastructural immunogold labeling in human white adipose tissue and in human adipocyte cell cultures at early and late stages of differentiation. Both leptin and its receptor were present in mature unilocular fat cells. The thin cytoplasmic rim of the adipocytes exhibited the strongest expression of both leptin and Ob-R. At early stages of differentiating human adipocytes, leptin was mainly expressed in multilocular preadipocytes, whereas the Ob-R was found predominantly on fibroblast-like cells. Other cellular components of human white adipose tissue were characterized by anti-CD31 for endothelial cells, anti-CD68 for macrophages, and antibodies specifically labeling B-cells and T-cells. In addition to fat cells, endothelial cells were immunopositive for the full-length leptin receptor. On the ultrastructural level, leptin was mainly found attached to cellular membranes and in small alveolate vesicle-like structures in the cytoplasm of adipocytes. Leptin was also present on the cell membranes of endothelial cells and macrophages. We conclude that the expression of the Ob-R in human white adipose tissue is not restricted to adipocytes but is present in resident endothelial and immune cells. Ultrastructural localization studies revealed an association of leptin with cell membranes and small vesicles. The cellular and subcellular distribution of leptin and its receptor suggests an important autocrine and paracrine role for leptin in human adipose tissue.

Effects of a novel corticotropin-releasing-hormone receptor type I antagonist on human adrenal function.

Corticotropin-releasing hormone (CRH) is the principal regulator of the hypothalamic-pituitary-adrenal (HPA) axis and an activator of the sympathoadrenal (SA) and systemic sympathetic (SS) systems. Mental disorders, including major depression and, more recently, Alzheimer's disease have been associated with dysregulation of the HPA axis and the SA/SS systems. Treatment of rats or monkeys with the novel CRH receptor type 1 (CRH-R1) antagonist antalarmin inhibits the HPA and/or the SA/SS axes. This is the first study to examine the potential direct effect of antalarmin on human adrenal function. Adrenocortical and adrenomedullary cells were characterized by double-immunohistochemistry with anti-17 alpha hydroxylase (cortical cells) and anti-chromogranin A (chromaffin cells). Expression of CRH, ACTH, CRH type I and type II receptor mRNA were analyzed by reverse-transcription (RT) PCR. Human adrenal cortical and/or chromaffin cells in co-culture were incubated with CRH, antalarmin, and both CRH and antalarmin in vitro. Exposure of these cells to corticotropin or vehicle medium served as positive and negative controls, respectively. Cortical and chromaffin tissues were interwoven in the human adrenals, and both in situ and in the co-culture system the endocrine cell types were in close cellular contact. ACTH, CRH, and CRH-R1 and CRH-R2 mRNAs were expressed in the human adrenal as determined by RT-PCR. CRH (10-8 M) led to a moderate increase of cortisol release (145.7 +/- 20.0%) from cortical and chromaffin adrenal cells in co-culture. This effect corresponded to 41.8% of the maximal increase induced by ACTH (10-8 M). The action of CRH was completely inhibited by antalarmin. CRH, ACTH, and both CRH-R1 and CRH-R2 mRNAs are expressed in the adult human adrenal gland. CRH stimulates cortisol production in cortical and chromaffin cell co-cultures. This effect is blocked by antalarmin, a selective CRH-R1 receptor antagonist, suggesting that CRH-R1 receptors are involved in an intraadrenal CRH/ACTH control system in humans.

The role of interleukin-6 in the human adrenal gland.

Interleukin (IL)-6 is a potent activator of the hypothalamic-pituitary-adrenal (HPA) axis on all levels in humans, and appears to play a pathogenic role in conditions related to chronic stress and physiological ageing; with physiological ageing showing a similar hormonal and immunological pattern to chronic stress. IL-6 and its receptor IL-6R are co-expressed at similar sites in the human adrenal gland, which seems to be an important source of IL-6 production. In vitro, in primary cultures of adrenal gland cells, chronic exposure to IL-6 stimulates adrenocortical steroid release in a time- and dose-dependent manner. This explains the high systemic cortisol levels in the absence of adequate plasma concentrations of corticotropin (ACTH) observed in patients after long-term treatment with IL-6. It could therefore be concluded that in situations of prolonged stress, when corticotropin-releasing hormone and ACTH release are suppressed by feedback inhibition due to circulating glucocorticoids, IL-6 maintains the elevated glucocorticoid levels by direct stimulation of adrenocortical steroidogenesis via autocrine/paracrine mechanisms.

Interleukin-6 and the interleukin-6 receptor in the human adrenal gland: expression and effects on steroidogenesis.

Interleukin (IL)-6 is a potent activator of the human hypothalamicpituitary-adrenal axis. After chronic administration of IL-6 in humans, there is a substantial elevation of cortisol, whereas ACTH levels are blunted. Thus, we investigated whether IL-6 and/or the IL-6 receptor (IL-6R) are expressed in the human adrenal gland and whether IL-6 could cause the release of steroid hormones by a direct action on adrenal cells in primary culture. The expression of IL-6 and IL-6R was investigated with RT-PCR and immunohistochemistry, and the effects on human adrenal steroidogenesis were tested with IL-6 in vitro. To avoid effects mediated by macrophages, we depleted adrenal primary cultures from macrophages using specific mouse antihuman CD68 and sheep antimouse IgG conjugated magnetic beads. The results showed that 1): IL-6 and IL-6R are expressed in adrenal cell cultures, including all cell types and those depleted of macrophages; 2) IL-6R is mainly expressed in the zona reticularis and the inner zona fasciculata; positive signals from the zona glomerulosa and the medulla occurred in single cells; and 3) IL-6 regulates adrenal synthesis of mineralocorticoids, glucocorticoids, and androgens in vitro, dependent on time and dose, in the absence of macrophages. After 24 h, aldosterone secretion increased to 172 +/- 28% SEM, cortisol to 177 +/- 27% SEM, and dehydroepiandrosterone to 153 +/- 20% SEM of basal secretion. These findings, in combination with previous investigations, suggest that IL-6 exerts its acute action via the hypothalamus and the pituitary. In the adrenal gland, however, IL-6 seems to be a long-term regulator of stress response, integrating the responses of all cortical zones to stimuli from the immune and endocrine system.

Direct effects of interleukin-6 on human adrenal cells.

Interleukin-6 (IL-6), which is expressed in the human adrenal gland, was found to be a very potent activator of the human HPA axis. So far nothing is known about a local paracrine or autocrine influence of IL-6 within the human adrenal. In this study, the expression of IL-6 and the IL-6 receptor by human adrenal cells in vitro could be demonstrated by immunohistochemistry. Possible effects of IL-6 on steroid release were tested by incubating human adrenal cells in vitro with IL-6 [10(-8) M]. Adrenal steroids were stimulated by IL-6: aldosterone 184 +/- 23, cortisol 198 +/- 19, DHEA 140 +/- 8 and androstenedione 136 +/- 5 (results are means +/- s.e.m. in %). In conclusion, IL-6 can act directly on human adrenal cells and appears to be an important paracrine or autocrine factor.

Transmural bioenergetic responses of normal myocardium to high workstates.

The response of myocardial high-energy and inorganic phosphates (HEP and Pi, respectively) and associated changes in myocardial blood flow, lactate uptake, and O2 consumption (MVo2) rates were examined in an open-chest canine model during progressively increasing workloads achieved by catecholamine infusion. HEP and Pi levels (measured with transmurally localized 31P-nuclear magnetic resonance spectroscopy) were unaffected by moderate increases in the level of energy expenditure but were significantly altered by high workloads, especially in the subepicardium. The MVo2 and HEP data from three different protocols that utilized pharmacological augmentation of blood flow demonstrated that the maximal rate of myocardial energy production during inotropic stimulation was dictated by perfusion limitation. This limitation was more severe in the subepicardial layer at the high workloads despite equivalent or even higher increases in blood flow to this layer, reflecting a preferential enhancement of demand in the outer layer by catecholamines. In contrast, under basal conditions, existence of a marginal perfusion limitation was evident in the inner but not in the outer layer.

Effects of dobutamine on myocardial blood flow, contractile function, and bioenergetic responses distal to coronary stenosis: implications with regard to dobutamine stress testing.

To determine the effects of dobutamine stimulation on myocardium distal to a coronary stenosis, transmural spatially localized phosphorus 31 nuclear magnetic resonance measurements of myocardial high-energy phosphate compounds (adenosine triphosphate and phosphocreatine), inorganic phosphate, and blood flow and systolic wall thickening were made in 8 open-chested dogs. Data were collected under (1) control conditions, (2) after the application of a moderate coronary stenosis, (3) during infusion of dobutamine with continuing stenosis, and (4) after the release of the stenosis with continuing dobutamine. Stenosis was associated with concordant reductions of subendocardial blood flow, wall thickening, and high-energy phosphate, and mild elevation of inorganic phosphate; subepicardial measurements were essentially unchanged. During dobutamine infusion, blood flow increased in all myocardial layers. Wall thickening returned to control values in the subendocardium and increased nonsignificantly in the subepicardium. Additional loss of high-energy phosphate occurred only in the subepicardium. The data suggest that improved contractile function associated with dobutamine infusion resulted from the inotropic effects of dobutamine and was made possible by the improved blood flow it produced. The data indicate that measurements of blood flow and contractile function do not reliably predict the transmural myocardial metabolic responses to inotropic perturbations in the hypoperfused heart. Taken together, the present findings yield insights with regard to the interpretation of diagnostic dobutamine stimulation testing with single photon emission tomography, radionuclide angiography, and echocardiography.

High-energy phosphate responses to tachycardia and inotropic stimulation in left ventricular hypertrophy.

Spatially localized nuclear magnetic resonance (NMR) spectroscopy was used to examine the effect of tachycardia and inotropic stimulation on myocardial ATP, creatine phosphate (CrP), and inorganic phosphate (Pi) in animals with left ventricular hypertrophy (LVH). Studies were performed in eight normal dogs and seven dogs with moderate LVH produced by banding the ascending aorta. 31P-NMR spectra were obtained from five layers across the LV wall, while blood flow (BF) was measured with microspheres during control conditions, pacing at 200 and 240 beats/min, and during dobutamine infusion (Dob). Myocardial ATP and CrP levels were normal in the LVH hearts during control conditions. Pacing did not alter the transmural distribution of perfusion or the levels of CrP, ATP, and Pi in normal hearts. In contrast, in four of seven LVH hearts, pacing decreased the subendocardial/subepicardial (ENDO/EPI) BF ratio and caused depletion of CrP and appearance of Pi characteristic of ischemia in the subendocardium. Dob produced greater increases in the heart rate x LV systolic pressure product (RPP) and greater increases of Pi and decreases of CrP in LVH than in normal hearts; however, at comparable elevations of RPP the alterations of Pi and CrP were similar in both groups. Although Dob decreased the ENDO/EPI in LVH hearts, Dob-induced alterations in CrP and Pi were uniform across the LV wall. Increasing myocardial BF with adenosine or carbochromen did not reverse the alterations in Pi or CrP produced by Dob. We conclude that 1) ENDO perfusion abnormalities during tachycardia in LVH do produce ENDO subendocardial ischemia; 2) when the degree of augmentation of mechanical performance is considered, the metabolic changes induced by Dob were similar in normal and LVH hearts; 3) Dob-induced alterations in Pi and CrP were not related to inadequate perfusion, since increasing coronary BF did not reverse these changes; and 4) alterations of Pi and CrP during Dob infusion were not more prominent in the ENDO, indicating that the decreased ENDO/EPI flow did not cause ENDO ischemia but may reflect relatively lower O2 demands in this region during inotropic stimulation.

Responses of myocardial high energy phosphates and wall thickening to prolonged regional hypoperfusion induced by subtotal coronary stenosis.

The response of the myocardium to prolonged or chronic ischemia may differ from the well documented changes that occur acutely subsequent to the onset of hypoperfusion. Therefore, we have examined in an instrumented canine model and using spatially localized spectroscopy to achieve transmural differentiation, the myocardial HEP and Pi levels as well as wall thickening in situ during prolonged ischemia induced by sustained coronary artery stenosis. The results demonstrate that subtotal coronary artery occlusion causes immediate and transmurally inhomogeneous decreases in the myocardial HEP content and increase in the Pi/CP ratio; however, during prolonged mild hypoperfusion, metabolic changes occur which lead to statistically significant recovery of CP (but not ATP) and disappearance of Pi despite the persistence of reduced blood flow and oxygen supply. Upon release of the occlusion, the previously ischemic muscle recovered blood flow, and some (but not all) of its preischemic contractile function without parallel changes in the HEP levels. It is concluded that normal HEP and Pi levels cannot be equated with either the absence of underperfusion or insensitivity of NMR spectroscopy to ischemia. Rather, it is imperative that both functional and spectroscopic measurements are performed simultaneously to distinguish between ischemic myocardium which is adapted versus unadapted to the hypoperfusion.

Contrast-enhanced first pass myocardial perfusion imaging: correlation between myocardial blood flow in dogs at rest and during hyperemia.

The sensitivity of contrast-enhanced MR first pass perfusion imaging in detection and quantification of hypoperfused myocardium was evaluated using an instrumented, closed-chest dog model where graded regional hypoperfusion was induced by applying predetermined levels of stenosis to the left anterior descending artery (LAD). All measurements were performed at rest and under stress induced by dipyridamole (DIP). Myocardial perfusion was assessed both with MR and radiolabeled microspheres injected immediately before the administration of the MR contrast agent. Ultrafast MR imaging was performed using a Turbo FLASH sequence with a 180 degrees inversion prepulse. A Gd-DTPA bolus was injected into the left atrium and T1-weighted images were acquired with every heart beat. Signal intensity measured from the images in regions of the LAD and left circumflex (LCx) perfusion beds was plotted against time to generate signal intensity versus time curves (SI time curve). Various flow indices were derived according to the indicator dilution theory, and compared with and without volume correction due to vasodilation to the myocardial blood flow (MBF) calculated from radiolabeled microspheres. Correlation of the MR and MBF data demonstrated that different transmural and regional myocardial perfusion levels can be easily visualized in the perfusion images and accurately monitored by the SI time curves. Detection of the impairment of myocardial perfusion improved significantly after administration of DIP. The inverse mean transit time calculated from the SI time curve was found to yield a linear correlation to absolute MBF derived from the microsphere data. These results suggest that with intracardiac injections of exogenous contrast agent, myocardial perfusion can be assessed parametrically with first pass contrast enhanced ultrafast MRI.

Effects of amiodarone with and without polysorbate 80 on myocardial oxygen consumption and coronary blood flow during treadmill exercise in the dog.

Since amiodarone has been reported to possess antianginal activity, this study examined the effects of amiodarone on coronary blood flow and myocardial oxygen consumption during exercise. Studies were performed in 14 chronically instrumented dogs trained to run on a motor-driven treadmill. Left circumflex coronary artery blood flow was measured with an electromagnetic flowmeter while aortic and coronary sinus catheters allowed measurement of myocardial oxygen extraction. During control conditions, graded exercise resulted in progressive increases in heart rate, aortic pressure, and coronary blood flow. Two preparations of amiodarone, 5 mg/kg, one dissolved in sterile water and the other in 10% polysorbate 80, were given intravenously to separate groups of dogs. Amiodarone in sterile water caused no hemodynamic changes at rest. However, the increase in heart rate during exercise was blunted after amiodarone, so that heart rate during the heaviest level of exercise was significantly less than during control exercise. Coronary blood flow and myocardial oxygen consumption were unchanged. Amiodarone with polysorbate 80 also blunted the increase in heart rate during exercise, but in addition caused a significant decrease in aortic pressure both at rest and during exercise. Myocardial oxygen consumption and coronary blood flow were significantly decreased after administration of amiodarone with polysorbate 80 at rest and during all exercise levels. Amiodarone with or without polysorbate 80 did not change myocardial oxygen extraction. These data demonstrate that amiodarone exerts a negative chronotropic effect during exercise. However, the decreased arterial pressure and myocardial oxygen consumption were not due to amiodarone, but was seen only with the combination of amiodarone dissolved in polysorbate 80.

Transmural high energy phosphate distribution and response to alterations in workload in the normal canine myocardium as studied with spatially localized 31P NMR spectroscopy.

Spatially localized phosphorus-31 nuclear magnetic resonance (31P NMR) spectroscopy has been applied to the study of the normal canine myocardium to measure the relative content of high energy phosphates across the left ventricular wall. Transmural NMR data were acquired in five voxels spanning the wall of the left ventricle using the FLAX-ISIS technique. The validity of the FLAX-ISIS approach in acquiring localized spectra for transmural studies and in providing quantitative information from the localized spectra was examined rigorously by studies involving phantoms, intact rats, and the canine myocardium in vivo. The results indicated that (1) this technique yields spatially resolved spectra with partial overlap between adjacent voxels and virtually no overlap between every other voxel; (2) in the canine heart, signals from subepicardium, midwall, and subendocardium can be detected separately without cross contamination; and (3) relative metabolite contents within a voxel and among voxels can be quantitated. Transmural 31P NMR spectra were acquired with cardiac gating on 29 separate animals either at early systole or late diastole, and at three different workloads with the heart rate peak systolic pressure product (RPP) increasing from 6000 mmHg/min to 35,000 mmHg/min. The data revealed that in the normal canine myocardium, the creatine phosphate (CP) content and the CP/ATP ratio was significantly lower in the subendocardium than in the subepicardium. ATP levels were transmurally constant. Both the CP content and the CP/ATP ratio measured for each voxel remained unaltered in relation to either the phase of the cardiac cycle or approximately fourfold increase in workload. Free ADP levels calculated for each voxel showed that ADP was relatively higher in the subendocardium than the subepicardium, and in all transmural layers was higher than its apparent Km for oxidative phosphorylation. In this domain changes in ADP content with workload and MVO2 are not expected and were not observed.

Correlation between transmural high energy phosphate levels and myocardial blood flow in the presence of graded coronary stenosis.

Spatially localized nuclear magnetic resonance spectroscopy was used to investigate with transmural differentiation the response of myocardial high energy phosphate compounds and inorganic orthophosphate (Pi) to graded reductions in coronary blood flow caused by sustained coronary stenosis. In an open-chest model, localized 31P nuclear magnetic resonance spectra from five layers across the left ventricular wall were obtained simultaneously with transmural blood flow measurements during control conditions and during sustained graded reductions in intracoronary pressure. Both the blood flow, and high energy phosphate and Pi contents displayed transmural heterogeneity in response to decreases in intracoronary pressure. The subendocardial creatine phosphate (CP) level remained unchanged as blood flow was reduced to approximately 0.7 ml/min/g wet wt and decreased precipitously beyond this critical flow level. The relation between CP and flow in the midmyocardium and especially in the subepicardium was more complex. Subepicardial CP content did not correlate well with blood flow; however, in cases in which a coronary stenosis resulted in subendocardial hypoperfusion but subepicardial flow was near or above normal, a close correlation was present between subepicardial and subendocardial CP levels. ATP levels in all layers remained unaltered until blood flow was severely reduced. These results demonstrate that 1) the myocardial high energy phosphate and Pi levels at any transmural layer are not generally determined by O2 and blood flow limitation under basal conditions; 2) during subtotal coronary occlusion, increased oxygen extraction is able to meet myocardial needs until a critical level of stenosis is reached; 3) below a critical flow level, subendocardial CP and Pi contents are closely correlated with absolute subendocardial blood flow; and 4) in the presence of a coronary stenosis, subepicardial CP and Pi contents may change even in the absence of perfusion deficit secondary to loss of subendocardial function.