Correction to: Multiple receptor tyrosine kinases converge on microRNA-134 to control KRAS, STAT5B, and glioblastoma.

Following publication of the article, the author named as "B Dey", wished to point out that his full name is "Bijan K. Dey". This was not reflected in the typesetting of the article, and as a consequence the article is not visible on Pub Med when a search is conducted on his full name.

Multiple receptor tyrosine kinases converge on microRNA-134 to control KRAS, STAT5B, and glioblastoma.

Receptor tyrosine kinases (RTKs) are co-deregulated in a majority of glioblastoma (GBM), the most common and most deadly brain tumor. We show that the RTKs MET, EGFR, and PDGFR regulate microRNA-134 (miR-134) in GBM. We find that miR-134 is downregulated in human tumors and cancer stem cells and that its expression inversely correlates with the activation of MET, EGFR, and PDGFR. We demonstrate that miR-134 inhibits cancer cell and stem-cell proliferation, survival, and xenograft growth, as well as cancer stem-cell self-renewal and stemness. We identify KRAS and STAT5B as targets of miR-134, and establish molecular and functional links between RTKs, miR-134, KRAS/STAT5B and malignancy in vitro and in vivo. We show that miR-134 induction is required for the anti-tumor effects of RTK inhibitors. We also uncover the molecular pathways through which RTKs regulate miR-134 expression and demonstrate the involvement of MAPK signaling and the KLF4 transcription factor. We therefore identify miR-134 as a novel RTK-regulated tumor-suppressive hub that mediates RTK and RTK-inhibitor effects on GBM malignancy by controlling KRAS and STAT5B.

Effect of Akt inhibition on scatter factor-regulated gene expression in DU-145 human prostate cancer cells.

The cytokine scatter factor (SF) (hepatocyte growth factor) transduces various biologic actions, including cell motility, invasion, angiogenesis and apoptosis inhibition. The latter is relevant to understanding the role of SF in promoting tumor cell survival in different contexts, for example, detachment from basement membrane, growth in metastatic sites and responses to chemo- and radiotherapy. Previously, we showed that SF protects cells against apoptosis owing to DNA damage, by a mechanism involving phosphoinositol-3-kinase/c-Akt signaling. Here, we used DNA microarray assays to identify c-Akt-regulated genes that might contribute to cell protection. DU-145 human prostate cancer cells were transfected+/-a dominant-negative mutant Akt, treated+/-SF and analysed for gene expression using Affymetrix arrays. These studies identified SF-regulated genes for which induction was c-Akt-dependent vs -independent. Selected microarray findings were confirmed by semiquantitative and quantitative reverse transcription-polymerase chain reaction. We tested the contribution of four SF-inducible/c-Akt-dependent genes (AMPD3, EPHB2, MX1 and WNT4) to protection against adriamycin (a DNA topoisomerase IIalpha inhibitor) using RNA interference. Knockdown of each gene except EPHB2 caused a small but significant reduction in the SF cell protection. The lack of effect of EPHB2 knockdown may be due to the fact that DU-145 cells contain a single-mutant EPHB2 allele. A combination of three small interfering RNAs blocked most of the protection by SF in both DU-145 and T47D cells. These findings identify novel c-Akt-regulated genes, some of which contribute to SF-mediated cytoprotection.

Efficacy of the non-peptide CGRP receptor antagonist BIBN4096BS in blocking CGRP-induced dilations in human and bovine cerebral arteries: potential implications in acute migraine treatment.

Calcitonin gene-related peptide (CGRP) is a potent vasodilator in brain vessels and it has been implicated in the pathogenesis of migraine headache. Blocking post-junctional CGRP receptors, mediators of trigeminal-induced vasodilation, has been suggested as a potential antimigraine strategy. In this study, we tested the ability of a new non-peptide CGRP receptor antagonist, BIBN4096BS, to inhibit the CGRP-induced dilation in human and/or bovine brain vessels and compared it to that of the antagonist alpha-CGRP(8-37). BIBN4096BS and alpha-CGRP(8-37) both blocked the alpha-CGRP-induced dilation in bovine middle artery segments with respective potency (pK(B) values) of 6.3 and 7.8. In human pial vessels, BIBN4096BS was particularly potent. When tested at 10(-14)-10(-9) M concentrations, it induced a rightward shift in the alpha-CGRP concentration-response curve and yielded a biphasic Schild plot suggesting interaction with more than one receptor population, as was also indicated by the significant best fit of the alpha-CGRP-induced dilation in human brain vessels with a two receptor site interaction. Schild plot analysis in the linear portion of the BIBN4096BS inhibition curve revealed interaction with one high affinity site (pA(2) value approximately 14). In bovine vessels, both alpha-CGRP(8-37) and BIBN4096BS concentration-dependently reversed a pre-established CGRP-induced dilation ( approximately 59 and 85%, respectively), BIBN4096BS being approximately tenfold more potent than alpha-CGRP(8-37) (respective pIC(50) values of 7.5 and 6.75). In human middle cerebral and middle meningeal arteries, BIBN4096BS reversed the alpha-CGRP-induced dilation (> or =70%) by interaction with two different receptor populations: it exhibited a high affinity for one population (pIC(50) value approximately 13) and a lower affinity for the other (pIC(50) value approximately 8). The present data demonstrate that BIBN4096BS is a very potent antagonist that could, depending on its bioavailability and in vivo affinity, be of potential benefit in the acute treatment of migraine headache by blocking and/or reversing the CGRP-mediated dilation of intracranial vessels induced by activation of trigeminovascular afferents.

A hammerhead ribozyme suppresses expression of hepatocyte growth factor/scatter factor receptor c-MET and reduces migration and invasiveness of breast cancer cells.

PURPOSE: Hepatocyte growth factor/scatter factor (HGF/SF), via its receptor c-MET, has been implicated to play a pivotal role in breast cancer development and progression. This study examined a transgene-consisting of a combination of U1snRNA, hammerhead ribozyme, and antisense, designed to inhibit c-met expression-and its impact on the migration and in vitro invasion of breast cancer cells. EXPERIMENTAL DESIGN: A hammerhead ribozyme targeting human c-MET was cloned into a modified pZeoU1EcoSpe vector and transfected into breast cancer cells MDA MB 231 and MCF-7 by electroporation. Expression of MET mRNA and protein was determined. Migration and in vitro invasiveness of transfected cells were also analyzed. RESULTS: Breast cancer cells were transfected with the ribozyme-containing plasmids. Stable transfectants manifested an almost complete loss of MET mRNA and protein, as shown by reverse transcription-PCR, Northern blotting, and Western blotting, respectively, whereas the wild-type plasmid had no effects. Met-ribozyme transfected cells exhibited reduced migration and in vitro invasiveness through extracellular matrix (Matrigel), compared with the wild-type cells and cells transfected with empty plasmid. CONCLUSIONS: These data show that targeting c-MET by way of a hammerhead ribozyme encoding antisense to c-MET is an effective approach in reducing the invasiveness of breast cancer cells.

Signaling pathways in the induction of c-met receptor expression by its ligand scatter factor/hepatocyte growth factor in human glioblastoma.

Scatter factor/hepatocyte growth factor (SF/HGF) and its tyrosine kinase receptor c-met are developmentally expressed, neuroprotective, and tumorigenic within the CNS. In the present study SF/HGF is shown to induce the expression of c-met in two human glioblastoma cell lines, U-373 MG and T98G, and the signaling pathways involved in this induction are dissected. SF/HGF activated mitogen-activated protein kinase (MAPK) and inhibition of either Ras or MAPK-kinase completely inhibited SF/HGF-mediated c-met induction. Inhibition of phospholipase-C (PLC) did not affect c-met induction in either cell line. Inhibition of phosphoinositide 3-kinase (PI3-kinase) substantially reduced c-met induction by SF/HGF in T98G cells but had no effect in U-373 MG cells. Protein kinase C (PKC) inhibition reduced c-met induction in T98G cells but not in U-373 MG cells. SF/HGF induced the expression of c-fos and c-jun mRNA and increased the levels of AP-1 transcription factor in both cells lines as determined by AP-1-luciferase reporter expression. Transfection of either cell line with TAM-67, a dominant negative for the jun transactivation domain, completely inhibited AP-1 and c-met induction by SF/HGF. These results support a model of c-met induction by SF/HGF in human glioma cells that uniformly involves Ras, MAPK, and AP-1 and additionally involves PI3-kinase and PKC in some cell lines.

Scatter factor/hepatocyte growth factor protects against cytotoxic death in human glioblastoma via phosphatidylinositol 3-kinase- and AKT-dependent pathways.

We have shown recently that the multifunctional growth factor, scatter factor/hepatocyte growth factor (SF/HGF), and its receptor c-met enhance the malignancy of human glioblastoma through an autocrine stimulatory loop (R. Abounader et al., J. Natl. Cancer Inst., 91: 1548-1556, 1999). This report examines the effects of SF/HGF:c-met signaling on human glioma cell responses to DNA-damaging agents. Pretreating U373 human glioblastoma cells with recombinant SF/HGF partially abrogated their cytotoxic responses to gamma irradiation, cisplatin, camptothecin, Adriamycin, and Taxol in vitro. This cytoprotective effect of SF/HGF occurred at least in part through an inhibition of apoptosis, as evidenced by diminished terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling index and reduced DNA laddering. Anti-c-met U1/ribozyme gene transfer inhibited the ability of SF/HGF to protect against single-strand DNA breakage, DNA fragmentation, and glioblastoma cell death caused by DNA-damaging agents, demonstrating a requirement for c-met receptor function. Phosphorylation of the cell survival-promoting kinase Akt (protein kinase B) resulted from SF/HGF treatment of U373 cells, and both Akt phosphorylation and cell survival induced by SF/HGF were inhibited by phosphatidylinositol 3-kinase inhibitors but not by inhibitors of mitogen-activated protein kinase kinase or protein kinase C. Cytoprotection by SF/HGF in vitro was also inhibited by transient expression of dominant-negative Akt. Transgenic SF/HGF expression by intracranial 9L gliosarcomas reduced tumor cell sensitivity to gamma irradiation, confirming the cytoprotective effect of SF/HGF in vivo. These findings demonstrate that c-met receptor activation by SF/HGF protects certain glioblastoma cells from DNA-damaging agents by activating phosphoinositol 3-kinase-dependent and Akt-dependent antiapoptotic pathways.

Glioma inhibition by HGF/NK2, an antagonist of scatter factor/hepatocyte growth factor.

Strategies that antagonize growth factor signaling are attractive candidates for the biological therapy of brain tumors. HGF/NK2 is a secreted truncated splicing variant and potential antagonist of scatter factor/hepatocyte growth factor (SF/HGF), a multifunctional cytokine involved in the malignant progression of solid tumors including glioblastoma. U87 human malignant glioma cells that express an autocrine SF/HGF stimulatory loop were transfected with the human HGF/NK2 cDNA and clonal cell lines that secrete high levels of HGF/NK2 protein (U87-NK2) were isolated. The effects of HGF/NK2 gene transfer on the U87 malignant phenotype were examined. HGF/NK2 gene transfer had no effect on 2-dimensional anchorage-dependent cell growth. In contrast, U87-NK2 cell lines were approximately 20-fold less clonogenic in soft agar and approximately 4-fold less migratory than control-transfected cell lines. Intracranial tumor xenografts derived from U87-NK2 cells grew much slower than controls. U87-NK2 tumors were approximately 50-fold smaller than controls at 21 days post-implantation and HGF/NK2 gene transfer resulted in a trend toward diminished tumorigenicity. This report shows that the predominant effect of transgenic HGF/NK2 overexpression by glioma cells that are autocrine for SF/HGF stimulation is to inhibit their malignant phenotype.

BIIE0246, a potent and highly selective non-peptide neuropeptide Y Y(2) receptor antagonist.

1. BIIE0246, a newly synthesized non-peptide neuropeptide Y (NPY) Y(2) receptor antagonist, was able to compete with high affinity (8 to 15 nM) for specific [(125)I]PYY(3 - 36) binding sites in HEK293 cells transfected with the rat Y(2) receptor cDNA, and in rat brain and human frontal cortex membrane homogenates. 2. Interestingly, in rat brain homogenates while NPY, C2-NPY and PYY(3 - 36) inhibited all specific [(125)I]PYY(3 - 36) labelling, BIIE0246 failed to compete for all specific binding suggesting that [(125)I]PYY(3 - 36) recognized, in addition to the Y(2) subtype, another population of specific NPY binding sites, most likely the Y(5) receptor. 3. Quantitative receptor autoradiographic data confirmed the presence of [(125)I]PYY(3 - 36)/BIIE0246-sensitive (Y(2)) and-insensitive (Y(5)) binding sites in the rat brain as well as in the marmoset monkey and human hippocampal formation. 4. In the rat vas deferens and dog saphenous vein (two prototypical Y(2) bioassays), BIIE0246 induced parallel shifts to the right of NPY concentration-response curves with pA(2) values of 8.1 and 8.6, respectively. In the rat colon (a Y(2)/Y(4) bioassay), BIIE0246 (1 microM) completely blocked the contraction induced by PYY(3 - 36), but not that of [Leu(31), Pro(34)]NPY (a Y(1), Y(4) and Y(5) agonist) and hPP (a Y(4) and Y(5) agonist). Additionally, BIIE0246 failed to alter the contractile effects of NPY in prototypical Y(1) in vitro bioassays. 5. Taken together, these results demonstrate that BIIE0246 is a highly potent, high affinity antagonist selective for the Y(2) receptor subtype. It should prove most useful to establish further the functional role of the Y(2) receptor in the organism.

Reversion of human glioblastoma malignancy by U1 small nuclear RNA/ribozyme targeting of scatter factor/hepatocyte growth factor and c-met expression.

BACKGROUND: Expression of scatter factor (SF), also known as hepatocyte growth factor (HGF), and its receptor, c-met, is often associated with malignant progression of human tumors, including gliomas. Overexpression of SF/HGF in experimental gliomas enhances tumorigenicity and tumor-associated angiogenesis (i.e., growth of new blood vessels). However, the role of endogenous SF/HGF or c-met expression in the malignant progression of gliomas has not been examined directly. In this study, we tested the hypothesis that human glioblastomas can be SF/HGF-c-met dependent and that a reduction in endogenous SF/HGF or c-met expression can lead to inhibition of tumor growth and tumorigenicity. METHODS: Expression of the SF/HGF and c-met genes was inhibited by transfecting glioblastoma cells with chimeric transgenes consisting of U1 small nuclear RNA, a hammerhead ribozyme, and antisense sequences. The effects of reduced SF/HGF and c-met expression on 1) SF/HGF-dependent induction of immediate early genes (c-fos and c-jun), indicative of signal transduction; 2) anchorage-independent colony formation (clonogenicity), an in vitro correlate of solid tumor malignancy; and 3) intracranial tumor formation in immunodeficient mice were quantified. Statistical tests were two-sided. RESULTS: Introduction of the transgenes into glioblastoma cells reduced expression of the SF/HGF and c-met genes to as little as 2% of control cell levels. Reduction in c-met expression specifically inhibited SF/HGF-dependent signal transduction (P<.01). Inhibition of SF/HGF or c-met expression in glioblastoma cells possessing an SF/HGF-c-met autocrine loop reduced tumor cell clonogenicity (P =.005 for SF/HGF and P=.009 for c-met) and substantially inhibited tumorigenicity (P<.0001) and tumor growth in vivo (P<.0001). CONCLUSIONS: To our knowledge, this is the first successful inhibition of SF/HGF and c-met expression in a tumor model directly demonstrating a role for endogenous SF/HGF and c-met in human glioblastoma. Our results suggest that targeting the SF/HGF-c-met signaling pathway may be an important approach in controlling tumor progression.

Scatter factor/hepatocyte growth factor gene transfer increases rat blood-glioma barrier permeability.

Malignant gliomas are associated with a dysfunctional blood-tumor barrier (BTB) that causes substantial morbidity. Scatter factor/hepatocyte growth factor (SF/HGF) is a multifunctional growth factor that correlates with glioma malignancy and has several biological properties that suggest a role in enhancing blood-glioma barrier permeability. In this study, we examined the effects of glioma cell SF/HGF expression on BTB permeability to horseradish peroxidase (HRP). Fischer 344 rats bearing intrastriatal 9L tumors engineered to secrete SF/HGF (9L-SF) and SF/HGF-negative control tumors (9L-neo) received intracardiac injections of HRP and were rapidly decapitated. Densitometric analysis of brain sections reacted with diaminobenzidine showed significantly greater extravascular HRP surrounding SF/HGF-secreting tumors than 9L-neo tumors of comparable size (p<0.05). HRP enzymatic activity associated with striata containing SF/HGF-expressing tumors was 1. 6-fold greater than that of striata containing control tumors (p<0. 05). Northern analysis showed that expression of vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) did not differ between 9L-neo and 9L-SF tumors. These data demonstrate that SF/HGF expression by intracerebral glial tumors can enhance BTB permeability independent of changes in VEGF/VPF expression.

Expression of neuropeptide Y receptors mRNA and protein in human brain vessels and cerebromicrovascular cells in culture.

Neuropeptide Y (NPY) has been suggested as an important regulator of CBF. However, except for the presence of Y1 receptors in large cerebral arteries, little is known about its possible sites of action on brain vessels. In this study, we sought to identify the NPY receptors present in the human cerebrovascular bed. Specific Y1 receptor binding sites, localized on the smooth muscle of human pial vessels and potently competed by NPY, polypeptide YY (PYY), and the selective Y1 receptor antagonist BIBP 3226, were identified by quantitative radioautography of the Y1 radioligand [125I]-[Leu31, Pro34]-PYY. In contrast, no specific binding of the Y2-([125I]-PYY3-36) and Y4/Y5-(125I-human pancreatic polypeptide [hPP]) radioligands could be detected. By in situ hybridization, expression of Y1 receptor mRNA was restricted to the smooth muscle layer of pial vessels, whereas no specific signals were detected for either Y2, Y4, or Y5 receptors. Similarly, using reverse transcriptase-polymerase chain reaction (RT-PCR), mRNA for Y1 but not Y2, Y4, or Y5 receptors was consistently detected in isolated human pial vessels, intracortical microvessels, and capillaries. In human brain microvascular cells in culture, PCR products for the Y1 receptors were exclusively found in the smooth muscle cells. In cultures of human brain astrocytes, a cell type that associates intimately with brain microvessels, PCR products for Y1, Y2, and Y4 but not Y5 receptors were identified. Finally, NPY significantly inhibited the forskolin-induced cAMP production in smooth muscle but not in endothelial cell cultures. We conclude that smooth muscle Y1 receptors are the primary if not exclusive NPY receptors associated with human brain extraparenchymal and intraparenchymal blood vessels, where they most likely mediate cerebral vasoconstriction.

Associations between neuropeptide Y nerve terminals and intraparenchymal microvessels in rat and human cerebral cortex.

Neuropeptide Y (NPY) can influence local brain perfusion, possibly via direct relationships with the microvascular bed. To evaluate this possibility, the authors quantitatively analyzed by light and electron microscopy the morphological associations between immunostained NPY neuronal elements and intraparenchymal microvessels in the rat and human cerebral cortex. At the light microscopic level in the rat frontoparietal cortex, about 16% of NPY neurons and large proximal processes as well as a subset of nerve terminals not affected by double sympathectomy were associated with penetrating arterioles and local microvessels. In human temporal cortex, a dense network of NPY nerve fibers was observed, many of which approached and/or contacted intracortical vessels. At the ultrastructural level, 14% of NPY axonal varicosities in the rat cerebral cortex were considered perivascular and associated with capillaries (approximately 70%) or microarterioles (approximately 30%). They were particularly enriched in the immediate vicinity (< 0.25 micron) of the microvessels, where the perivascular astrocytic leaflets represented a frequent target. In human cerebral cortex, NPY varicosities were observed in proximity to microvessels corresponding primarily to capillaries. Perivascular NPY varicosities never established synaptic junctions with vascular or astroglial elements. The results show that central NPY nerve terminals associate with microvessels and perivascular astroglial cells in the rat and human cerebral cortex. Thus, NPY released from these nerves could possibly influence (via a parasynaptic mode of action) vascular and/or astrocytic functions depending on the distribution of NPY receptors in these cellular compartments. These results provide morphological support for the effects of NPY on brain perfusion and homeostasis.

Parallel changes of blood flow and heterogeneity of capillary plasma perfusion in rat brains during hypocapnia.

Plasma perfusion patterns were investigated in brain capillaries during decreased cerebral blood flow induced by hyperventilation. Anesthetized rats were decapitated 3-4 s after being given an intravenous bolus injection of Evans blue dye. the measured steep increase of the arterial dye concentration at this moment ensures that different capillary plasma transit times are reflected in different intracapillary dye concentrations. The observed heterogeneity of capillary plasma transit time was expressed as the coefficient of variation (means +/- SD) of the intracapillary dye concentrations. For comparison, cerebral blood flow was determined at comparable PCO2 values in a second experimental group. At arterial PCO2 values between 40 and 25 mmHg, the cerebral blood flow and the coefficient of variation of the intracapillary dye concentration decreased with decreasing PCO2, whereas at PCO2 values <25 mmHg cerebral blood flow and coefficient of variation did not correlate with the arterial PCO2. However, it cannot be excluded that the coefficient of variation of the intracapillary dye concentration increases between 25 and 14 mmHg and decreases between 14 and 10 mmHg. It is concluded that the reduction of cerebral blood flow measured during moderate hypocapnia is paralleled by a decreased heterogeneity of the brain capillary perfusion. During severe hypocapnia this relationship is lost, indicating a potential disturbance of the cerebral microcirculation.

Characterization of neuropeptide Y (NPY) receptors in human cerebral arteries with selective agonists and the new Y1 antagonist BIBP 3226.

1. We have characterized pharmacologically the receptor subtype(s) responsible for the neuropeptide Y (NPY)-induced vasoconstriction in human cerebral arteries. NPY, PYY and several of their derivatives with well defined affinities at the known Y1 and Y2 receptor subtypes were used. Moreover, we tested the ability of the new Y1 receptor antagonist, BIBP 3226, to antagonize the NPY-induced cerebral vasoconstriction. 2. NPY, PYY and their agonists with high affinities at the Y1 receptor subtype ([Leu31-Pro34]-NPY and [Leu31-Pro34]-PYY) elicited strong, long lasting and concentration-dependent contractions of human cerebral arteries. Compounds with Y2 affinity such as PYY3-36 or NPY13-36 either elicited a submaximal contraction at high concentrations or failed to induce any significant vasomotor response. Also, the application of NPY or the specific Y1 agonist, [Leu31-Pro34]-NPY, to human cerebral vessels pretreated with the Y1 agonist, NPY13-36, resulted in contractile responses identical to those obtained when these compounds were tested without prior application of NPY13-36. 3. The order of agonist potency at the human cerebrovascular receptor was: [Leu31-Pro34]-NPY = [Leu31-Pro34]-PYY > or = NPY > PYY > PYY3-36 > > > NPY13-36, which corresponded to that reported previously at the neuronal and vascular Y1 receptors. 4. Increasing concentrations (10(-9)-10(-6) M) of the Y1 receptor antagonist, BIBP 3226, to human cerebral vessels caused a parallel and rightward shift in the NPY dose-response curves without any significant change in the maximal contractile response. The calculated pA2 was 8.52 +/- 0.13, a value compatible with the reported affinity at the rodent and human Y1 receptor. 5. We conclude that Y1 receptors exclusively, mediate the NPY-induced contraction in human cerebral arteries and we show that BIBP 3226 is a potent and competitive antagonist of this YI-mediated vasoconstriction.

Patterns of capillary plasma perfusion in brains in conscious rats during normocapnia and hypercapnia.

The present study aimed to investigate the distribution pattern of plasma flow velocities in brain capillaries. We tested the hypothesis that plasma flow velocities are heterogeneous in the brain capillaries of normocapnic conscious rats and become more homogeneous during increased cerebral blood flow induced by hypercapnia. We developed a method that makes it possible to detect the distribution pattern of plasma flow velocities from the intravascular dye concentrations measured in different capillaries. Evans blue was injected intravenously as a bolus, and 3 to 4 seconds later the rats were decapitated. During this period, a steep increase in arterial dye concentration was verified by frequent arterial blood sampling. Under such conditions, divergent plasma flow velocities in different capillaries yield unequal intravascular dye concentrations. Dye concentrations were measured in several hundred capillaries of brain cryosections using quantitative fluorescence microscopy based on calibration curves obtained from anesthetized rats. The results show a high degree of variation in the intravascular dye concentration during normocapnia. During increasing stages of hypercapnia, the variation was gradually reduced. The coefficient of variation (SD/mean-100) of intracapillary dye concentration decreased from 76% at normocapnia to 22% at extreme hypercapnia (PCO2 of 87 mm Hg), thus showing an inverse correlation with arterial PCO2 (r = .97). The heterogeneity of intravascular dye concentrations observed in the present experiments indicates heterogeneous velocities of plasma perfusion in different brain capillaries during normocapnia and a more homogeneous distribution pattern during hypercapnic hyperemia.