Oxygen-regulated transferrin expression is mediated by hypoxia-inducible factor-1

Transferrin (Tf) is a liver-derived iron transport protein whose plasma concentration increases following exposure to hypoxia. Here, we present a cell culture model capable of expressing Tf mRNA in an oxygen-dependent manner. A 4-kilobase pair Tf promoter/enhancer fragment as well as the 300-base pair liver-specific Tf enhancer alone conveyed hypoxia responsiveness to a heterologous reporter gene construct in hepatoma but not HeLa cells. Within this enhancer, a 32-base pair hypoxia-responsive element was identified, which contained two hypoxia-inducible factor-1 (HIF-1) binding sites (HBSs). Mutation analysis showed that both HBSs function as oxygen-regulated enhancers in Tf-expressing as well as in non-Tf-expressing cell lines. Mutation of both HBSs was necessary to completely abolish hypoxic reporter gene activation. Transient co-expression of the two HIF-1 subunits HIF-1alpha and aryl hydrocarbon receptor nuclear translocator (ARNT)/HIF-1beta resulted in enhanced reporter gene expression even under normoxic conditions. Overexpression of a dominant-negative ARNT/HIF-1beta mutant reduced hypoxic activation. DNA binding studies using nuclear extracts from the mouse hepatoma cell line Hepa1 and the ARNT/HIF-1beta-deficient subline Hepa1C4, as well as antibodies raised against HIF-1alpha and ARNT/HIF-1beta confirmed that HIF-1 binds the Tf HBSs. Mutation analysis and competition experiments suggested that the 5' HBS was more efficient in binding HIF-1 than the 3' HBS. Finally, hypoxic induction of endogenous Tf mRNA was abrogated in Hepa1C4 cells, confirming that HIF-1 confers oxygen regulation of Tf gene expression by binding to the two HBSs present in the Tf enhancer.     

The redox-sensitive human antioxidant responsive element induces gene expression under low oxygen conditions

Transient transfection studies of human HepG2 and mouse Hepa hepatocarcinoma cells with a reporter gene construct regulated by a human antioxidant responsive element (ARE) from the NQO1 gene demonstrated that the element is responsive to low oxygen conditions. The antioxidant N-acetyl L-cysteine (NAC) strongly inhibited basal aerobic reporter gene activity in HepG2 cells without obviously affecting the hypoxic induction, as is consistent with ARE sensitivity to oxidative stress in aerobic cultures. Electrophoretic mobility shift (EMS) assays of nuclear extracts of HepG2 and Hepa cells lysed under aerobic or hypoxic conditions or after exposure to the phenolic compound 3-(2)-tert-butyl-4-hydroxyanisole (BHA), showed specific and constitutive protein binding to the ARE under all of these conditions. Taken together, these findings show that the ARE can mediate gene expression in response to low oxygen conditions. Co-ordinately regulated expression of ARE-dependent genes, such as phase II detoxification enzymes, may be an important phenotype of solid tumors containing significant regions of pathophysiological hypoxia.     

Enhancement of gene expression under hypoxic conditions using fragments of the human vascular endothelial growth factor and the erythropoietin genes

PURPOSE: Selective gene expression in response to tumor hypoxia may provide new avenues, not only for radiotherapy and chemotherapy, but also for gene therapy. In this study, we have assessed the extent of hypoxia responsiveness of various DNA constructs by the luciferase assay to help design vectors suitable for cancer therapy. MATERIALS AND METHODS: Reporter plasmids were constructed with fragments of the human vascular endothelial growth factor (VEGF) and the erythropoietin (Epo) genes encompassing the putative hypoxia-responsive elements (HRE) and the pGL3 promoter vector. Test plasmids and the control pRL-CMV plasmid were cotransfected into tumor cells by the calcium phosphate method. After 6 h hypoxic treatment, the reporter assay was performed. RESULTS: The construct pGL3/VEGF containing the 385 bp fragment of the 5' flanking region in human VEGF gene showed significant increases in luciferase activity in response to hypoxia. The hypoxic/aerobic ratios were about 3-4, and 8-12 for murine and human tumor cells, respectively. Despite the very high degree of conservation among the HREs of mammalian VEGF genes, murine cells showed lower responsiveness than human cells. We next tested the construct pGL3/Epo containing the 150 bp fragment of the 3' flanking region in the Epo gene. Luciferase activity of pGL3/Epo was increased with hypoxia only in human cell lines. The insertion of 5 copies of the 35-bp fragments derived from the VEGF HREs and 32 bp of the E1b minimal promoter resulted in maximal enhancement of hypoxia responsiveness. CONCLUSIONS: The constructs with VEGF or Epo fragments containing HRE may be useful for inducing specific gene expression in hypoxic cells. Especially, the application of multiple copies of the HREs and an E1b minimal promoter appears to have the advantage of great improvement in hypoxia responsiveness.     

Effects of antioxidant vitamins on renal and hepatic erythropoietin production

An important role in O2 sensing has been assigned to microsomal and membrane-bound b-type cytochromes which generate regulatory reactive O2 species (ROS). Recently, ROS have been shown to suppress the in vitro synthesis of erythropoietin (Epo). We investigated the potential of the antioxidant vitamins A, E and C to enhance renal and hepatic Epo production. Renal effects were studied in isolated serum-free perfused rat kidneys. In control experiments without antioxidant vitamins, Epo secretion amounted to 441 +/- 23 mU/g kidney (mean +/- SEM, N = 5) during the three hour period of hypoxic perfusion (arterial pO2 35 mm Hg). Epo secretion significantly increased to 674 +/- 92 mU/g kidney (N = 7) when vitamins A (0.5 microgram/ml), E (0.5 microgram/ml) and C (10 micrograms/ml) in combination were added to the perfusion medium. The effects of the single vitamins were studied in Epo-producing hepatoma cell cultures (lines HepG2 and Hep3B). Vitamin A induced a dose-dependent increase (half-maximal stimulation at 0.2 microgram/ml) in the production of immunoreactive Epo during 24 hours of incubation (such as 680 +/- 51 U Epo/g cell protein in HepG2 cultures with 3 micrograms/ml retinol acetate compared to 261 +/- 15 U/g in untreated controls; N = 4). In contrast, vitamin E (tested from 0.05 to 500 micrograms/ml) and vitamin C (tested from 2 to 200 micrograms/ml) did not increase Epo production in hepatoma cell cultures. Thus, while vitamins E and C may have the potential to protect cells from oxidative damage, vitamin A exerts a specific stimulation of Epo production. Preliminary evidence suggests that this effect of vitamin A involves increased mRNA levels of hypoxia-inducible factor 1 alpha (HIF-1 alpha).     

Mouse hypoxia-inducible factor-1alpha is encoded by two different mRNA isoforms: expression from a tissue-specific and a housekeeping-type promoter

Hypoxic induction of erythropoietin (Epo) and other oxygen-dependent genes is mediated by the hypoxia-inducible factor-1 (HIF-1), a heterodimeric transactivator consisting of an alpha and a beta subunit. We previously found that the mouse gene encoding HIF-1alpha harbors two alternative first exons (I.1 and I.2), giving rise to two different HIF-1alpha mRNA isoforms. Here, we show by RNase protection analysis that the exon I.1-derived mRNA isoform is differentially expressed in mouse tissues, being highest in kidney, tongue, stomach, and testis, but undetectable in liver, whereas the exon I.2 mRNA isoform is ubiquitously expressed. Sequence and methylation analysis showed that, in contrast to exon I.1, exon I.2 resides within a region showing typical features of a CpG island, known to be associated with the 5' end of housekeeping genes. We identified a 232-bp minimal exon I.2 promoter that strongly induced reporter gene expression in mouse L929 fibroblasts and Hepa1 hepatoma cells. In contrast to L929 cells, the exon I.1 promoter was inactive in Hepa1 cells and hypoxic exposure (1% O2) markedly reduced exon I.2 promoter activity in Hepa1 cells. Prolonged exposure of mice to hypoxia (7.5% O2 for up to 72 hours) also caused a decrease in liver HIF-1alpha mRNA, whereas aldolase mRNA levels increased. These findings might be related to the relatively low Epo levels in the adult liver.     

Coronary bypass surgery: what is changing? Analysis of 3834 patients undergoing primary isolated myocardial revascularization

We have previously demonstrated in primary cancer of the uterine cervix that tumor hypoxia, as determined polarographically, is strongly associated with clinical malignant progression of the disease. Having applied a similar methodological approach to investigate loco-regional relapses, we found a pronounced shift to more hypoxic oxygenation profiles in the recurrent tumors than in the primary tumors. Median pO2 values in 53 pelvic recurrences were significantly lower than the median pO2 values of 117 primary tumors of comparable sizes (7.1 +/- 1.1 mmHg vs. 12.1 +/- 1.0 mmHg, p = 0.0013). The differences in tumor oxygenation between primary and recurrent tumors mirrored the differences in the patients' 5-year survival probabilities. In the cohort of patients with pelvic relapses, median tumor pO2 < 4 mmHg indicated a significantly shorter median survival time as compared to median tumor pO2 > or = 4 mmHg. Our results further support our thesis that in cervical cancer, tumor hypoxia and clinical aggressiveness in terms of resistance to therapy and tumor dissemination, are interrelated.     

Microvascular permeability of the isolated rat heart to various solutes in well-oxygenated and hypoxic conditions

The aim of this study was to investigate the effect of a moderate degree of hypoxia on coronary vascular permeability to various lipophobic solutes. Using the multiple indicator dilution method the permeability-surface area (PS) products were determined for 125I-albumin, 125I-insulin and 57Co-cyanocobalamin in perfused rat hearts (flow approximately 10 ml.min-1.g-1) either with well-oxygenated (pO2 approximately 96 kPa) or hypoxic (pO2 approximately 45 kPa) solutions. The PS products for albumin, insulin and cyanocobalamin during the well-oxygenated equilibration period were 0.20 +/- 0.03, 0.29 +/- 0.06 and 2.0 +/- 0.3 ml.min-1.g-1 (mean +/- SE), respectively, relative to 131I-gamma-globulin. The PS products for these solutes 15 min after the induction of hypoxia were 1.3 +/- 0.3, 0.8 +/- 0.1 (p < 0.05) and 2.1 +/- 0.2 (p < 0.05), respectively. In hearts perfused with well-oxygenated solution for 75 min, the PS products for these solutes remained stable throughout the period of the study. Electron-microscopic examination of hypoxic tissues showed the presence of endothelial gaps of approximately 1 micron which were underlined by an intact basal lamina. We conclude that a moderate degree of hypoxia produces a large increase in permeability of albumin and insulin but has no effect on the PS products for cyanocobalamin and that the endothelial gaps are the likely mechanism of the observed increase in permeability.     

IGFBP-2 expression in a human cell line is associated with increased IGFBP-3 proteolysis, decreased IGFBP-1 expression and increased tumorigenicity

Insulin-like growth factors (IGF-I and -II) play an active role in cell proliferation. In biological fluids, they are non-covalently bound to high-affinity binding proteins (IGFBPs), at least 6 species of which have been identified to date, but with poorly defined functions. One of these IGFBPs, IGFBP-2, is secreted by most cell lines and appears to be involved in cell proliferation. A human epidermoid carcinoma cell line, KB 3.1, which produces IGFBP-1 and -3 and small amounts of IGFBP-4, but no IGFBP-2, was stably transfected with an expression vector comprising IGFBP-2 complementary DNA (cDNA), whose expression was placed under the control of the constitutive and ubiquitous cytomegalovirus promoter. After an s.c. injection of these IGFBP-2-expressing KB 3.1 cells into nude mice, tumours developed more quickly than in controls, they were 3 to 4 times larger and grew about 3 times as fast. Concomitant with IGFBP-2 expression in these tumours, were a decrease in IGFBP-1 expression and an increase in IGFBP-3 proteolysis, both of which increase the bioavailability of the IGF-II produced by the cells. The increased IGFBP-3 proteolysis most probably resulted from amplified expression of tissue-type plasminogen activator (t-PA) and depression of its inhibitor (PAI-I) observed in IGFBP-2-expressing xenografts. Our findings suggest that IGFBP-2 plays a role in this model of experimental tumorigenesis via a mechanism that remains unclear, but appears to involve increased protease activity and IGF-II bioavailability.     

A 2-week pretreatment with 13-cis-retinoic acid + interferon-alpha-2a prior to definitive radiation improves tumor tissue oxygenation in cervical cancers

BACKGROUND: We have evaluated the tumor tissue pO2 in cervical cancers in patients treated with 13-cis-retinoic acid and interferon-alpha-2a prior to and during radiotherapy. PATIENTS AND METHODS: From June 1995 through April 1997, 22 patients with squamous cell carcinoma FIGO IIB/III of the cervix who were scheduled for definitive radiotherapy with curative intent received additional treatment with 13-cis-retinoic acid (cRA, isotretinoin) plus interferon-alpha-2a (IFN-alpha-2a) as part of a phase-II protocol. cRA/IFN-alpha-2a started 14 days prior to radiotherapy (1 mg per kilogramme body weight cRA orally daily plus 6 x 10(6) IU IFN-alpha-2a subcutaneously daily). After this induction period, standard radiotherapy was administered (external irradiation with 50.4 Gy in 28 fractions of 1.8 Gy plus HDR-brachytherapy). During radiotherapy, cRA/IFN-alpha-2a treatment was continued with 50% of the daily doses. Tumor tissue pO2-measurements were performed prior to and after the cRA/IFN-induction period as well as at 20 Gy and at the end of radiotherapy with an Eppendorf-pO2-histograph. RESULTS: In 11 out of the 22 patients, pO2-measurements were performed prior to the cRA/IFN-induction therapy. The median pO2 of these untreated tumors was 17.7 +/- 16.3 mm Hg. The relative frequency of hypoxic readings with pO2-values below 5 mm Hg ranged from 0% to 60.6% (mean 24.3 +/- 21.0%). After the 2-week induction period with cRA/IFN, the median pO2 had increased from 17.7 +/- 16.3 mm Hg to 27.6 +/- 19.1 mm Hg (not significant). In all 5 patients with hypoxic tumors prior to cRA/IFN (median pO2 of 10 mm Hg or less), the median pO2 was above 20 mm Hg after the 2-week cRA/IFN-induction. In this subgroup of hypoxic tumors, the median pO2 increased from 6.3 +/- 2.7 mm Hg to 27.0 +/- 5.6 mm Hg (p = 0.004, t-test for paired samples). The frequency of hypoxic readings (pO2-values < 5 mm Hg) decreased from 44.7 +/- 17.1% to 2.0 +/- 2.5% (p = 0.012, t-test for paired samples). There was, however, no obvious volume reduction after 14 weeks of cRA/IFN on clinical examination. A complete clinical remission of the local tumor was observed in 19/22 patients after radiotherapy and additional cRA/IFN-alpha-2a-treatment. In primarily hypoxic tumors (with a median pO2 below 10 mm Hg prior to treatment), 4/5 achieved complete remission. CONCLUSIONS: Pretreatment with cRA/IFN improves oxygenation of primarily hypoxic cervical cancers. The mechanisms of action remain unclear and further investigation of the combination regimen is recommended.     

Hypoxia and vascular endothelial growth factor stimulate angiogenic integrin expression in bovine retinal microvascular endothelial cells

PURPOSE: Integrins alphavbeta3 and alphavbeta5 are cell-to-matrix adhesion molecules that have been reported to mediate vascular cell proliferation and migration. The authors investigated the regulation of expression of these angiogenic integrins by hypoxia and vascular endothelial growth factor (VEGF) in retinal microvascular endothelial cells in culture. METHODS: Cultured bovine retinal capillary endothelial cells were exposed to human recombinant VEGF under normoxic (95% air, 5% CO2) conditions to assess the effects of VEGF. Hypoxia studies were performed under lower oxygen concentration (0.5%-1.5% O2) induced by nitrogen replacement in constant 5% CO2 conditions. Integrin family mRNA and protein expression were assessed by northern blot analysis and immunoprecipitation. RESULTS: VEGF (25 ng/ml) increased integrin alphav, beta3, and 35 mRNA after 24 hours 6.1+/-0.8-fold (P < 0.001), 5.9+/-1.1-fold (P < 0.001), and 1.9+/-0.2-fold (P < 0.01), respectively. Similarly, hypoxia stimulated gene expression of integrin alphav and beta3 after 24 hours by 5.1+/-1.7-fold (P < 0.01) and 3.0+/-0.5-fold (P < 0.01), respectively, and integrin beta5 after 9 hours 1.4+/-0.2-fold (P < 0.05). This hypoxia-induced, integrin alphav mRNA elevation was inhibited significantly by anti-VEGF neutralizing antibody. Also, a conditioned medium from confluent endothelial cells maintained under hypoxic conditions for 24 hours produced a 7.1+/-1.1-fold increase (P < 0.001) in integrin alphav mRNA expression after 24 hours, which was reversed by anti-VEGF neutralizing antibody. Induction of integrin alphav by VEGF and hypoxia was confirmed in the protein level. CONCLUSIONS: These data suggest that hypoxia stimulates expression of vascular integrins alphavbeta3 and alphavbeta5 in retinal microvascular endothelial cells partially through autocrine-paracrine action of VEGF induced by the hypoxic state.     

Lack of induction of substance P gene expression by hypoxia and absence of neurokinin 1-receptor mRNAs in the rat carotid body

Peripheral chemoreceptors are commonly thought to respond to hypoxia by releasing neurotransmitters from the type 1 cells of the carotid body; these molecules then bind to post-synaptic receptors on the carotid sinus nerve. The tachykinin substance P (SP) may act as an important neurotransmitter/neuromodulator in hypoxic chemotransmission in peripheral arterial chemoreceptors. In order to elucidate the role of SP in modulating hypoxic chemotransmission, we have used quantitative in situ hybridization histochemistry, to determine the effect of hypoxia on SP gene induction, and the localization of neurokinin 1 (NK-1) receptor mRNA in the carotid body and petrosal ganglia complex in rats at 21 days post-natal age. For comparison, we also determined: (1) the effect of hypoxia on tyrosine hydroxylase (TH) gene induction and (2) the localization of the mRNA encoding the D2-dopamine receptor. SP mRNA was not detected in the rat carotid body during normoxia and its expression was not induced after a 1 h of exposure to hypoxia (10% O2/90% N2), a stimulus that was sufficient to cause a significant increase (P < 0.01) in TH mRNA levels in the carotid body. Both SP and TH mRNAs were abundantly expressed in multiple cells in the petrosal and the jugular ganglia. However, these mRNAs were not co-localized and SP and TH mRNA levels were not affected by hypoxia in these ganglia. Although D2-dopamine receptor mRNA was abundantly expressed in the rat carotid body, we found no evidence of NK-1 receptor mRNA in the carotid body. In contrast, both NK-1 receptor mRNA and D2-dopamine receptor mRNA were present in petrosal ganglion cells. In the rat, SP does not appear to modulate hypoxic chemotransmission by being made in and released from type 1 cells in the carotid body, and neither does SP modulate the activity of type 1 cells by binding to NK-1 receptors on these cells.