Coupling Between 02 Sensors and Regulators by Chemical Modifications

5.1. Oxygen Sensing and HIF-la Phosphorylation

The first evidence indicating that phosphorylation reactions are involved in the regulation of HIF-1 activity were obtained by electrophoretic mobility shift assay (EMSA) which showed that treatment of nuclear extracts from hypoxic Hep3B cells with phosphatase disrupts the HIF-1/DNA complex (101). Furthermore, application ofthe serine/threonine kinase inhibitor 2-aminopurine, the tyrosine kinase inhibitor genistein, as well as the serine threonine phosphatase inhibitor NaF inhibited HIF-1-DNA-binding activity and HIF-1 stabilization under hypoxic conditions (100). These data suggested that phosphorylation, as well as dephosphorylation, could both be involved in HIF-1 activation. Several kinases were shown to be activated under hypoxia in different cell types and are therefore possible candidates for HIF-1 activation. This is the case for some of the mitogen-activated protein kinase (MAPK) family members and for phosphatidylinositol-3 kinase (PI3-K) and protein kinase B (PKB, also known as AKT) (66, 71, 81, 82, 94).

Several reports demonstrated that the pathway involving PI3-K, which generates phosphatidylinositol-3,4,5-phosphate [PI(3,4,5)P3], which then regulates the activity or subcellular localization of a variety of signaling molecules such as phosphatidiylinositol-dependent kinase (PDK) and PKB, plays an important role in activating the HIF pathway. The co-transfection of a HIF-1 reporter plasmid either with dominant-negative vectors for PI3-K or for PKB impaired the activation of HIF-1 as well as VEGF gene transcription in hypoxic NIH3T3R cells (71). Likewise, overexpression of PKB enhanced HIF-1 a levels and stimulated HIF-1-dependent PAI-1 expression as well as EPO-HRE-regulated reporter gene activity in primary rat hepatocytes and HepG2 cells (56).

However, in contrast to growth factors, hormones and coagulation factors including PDGF, angiotensin II, insulin and thrombin as well as to H202 which activate PKB within minutes, enhanced phosphorylation of PKB has been observed after hours of exposure to hypoxia (8). Interestingly, these agonists were also able to stimulate the HIF pathway in a ROS- and PI3-K/PKB-dependent manner independent from the 02 tension (68, 75, 84, 97). This suggested a cross talk between these agonist-dependent pathways and the 02 signaling cascade leading to the activation of HIF. This assumption was further supported by observations that MAP kinases (including p38 MAP kinase and/or p42/44 MAP kinase) also contribute to the activation of the HIF pathway. Overexpression of the p38 upstream kinases MKK3 and MKK6 resulted in enhanced HIF-la levels and stimulated HIF-1-dependent PAI-1 expression as well as EPO-HRE-regulated reporter gene activity under normoxic and hypoxic conditions (52) whereas inhibition of p38 MAP kinase prevented thrombin-induced HIF activation in a ROS-dependent manner (32). On the other hand, p42/44 MAP kinase has been shown to directly phosphorylate HIF-1 a in vitro under normoxic conditions (81) but overexpression ofp42/44 MAP kinase did not alter HIF-la protein levels and did not stimulate HIF-1-dependent PAI-1 expression or EPO-HRE-regulated reporter gene activity (52). Moreover, in contrast to p38 MAP kinase, p42/44 MAP kinase was not activated by H202 and did not contribute to thrombin-induced activation of the HIF cascade (32) but stimulated angiotensin II-dependent activation of HIF (82). The reasons for these discrepancies are not clear at the moment but may relate to cell type specific differences in kinase expression and sensitivity to hypoxia and ROS.

5.2. Oxygen Sensing and HIF-la Hydroxylation

The amount of the HIF-la protein is rapidly increased when cells are exposed to hypoxia. In contrast, under normoxic conditions, the HIF-la protein is remarkably unstable suggesting that the formation of the active HIF-1 dimer depends mainly on hypoxia-induced stabilization of HIF-la. Two transactivation domains (TAD) were identified within HIF-a which confer the sensitivity towards 02. Both domains are 100% conserved in human, mouse and rat HIF-1 a. The N-terminal TAD (TAD-N) is present within a region (amino acids 401 -603) which was found to be critically involved in the 02-dependent destabilization of the HIF-la protein via proteasomal degradation (45, 84) and was thus named the oxygen-dependent degradation domain (ODD) (45).

In normoxia, HIF a-subunit destabilization is mediatedby 02-dependent hydroxylation of at least two proline residues within the ODD (47, 68, 97). This process allows binding of the von Hippel-Lindau tumor suppressor protein (pVHL) (70, 75, 78, 97). pVHL is found in a multiprotein complex with elongins B/C, Cul2, and Rbx1 forming an E3 ubiquitin ligase complex called VEC. This modular enzyme then initiates degradation by the ubiquitin-proteasome pathway (46, 48, 49).

The C-terminal TAD (TAD-C, ranging from amino acid 776-826), and the TAD-N have been shown to interact with coactivators such as CREB-binding protein (CBP)/p300) (3, 18, 21), the steroid receptor coactivator-1 (SRC-1), transcription intermediary factor-2 (TIF-2) and redox factor-1 (Ref-1) (12), thereby facilitating enhanced transcriptional activity. A new enzyme named factor inhibiting HIF (FIH) (69) prevents the recruitment of the coactivator CBP/p300 by hydroxylating an asparaginyl residue in the TAD-C (63). F1H, like the HIF prolyl hydroxylases, is an 02-, oxoglutarate-, iron-, and ascorbate-dependent enzyme and may thus also be considered as a putative 02 sensor (39, 63, 65, 72).

5.3. ROS as Messengers in the Oxygen Sensing Cascade

Whenever oxygen is not completely reduced to water within the organism or a cell,, oxygen intermediates such as and can be generated. Besides their ability to exert oxidative stress resulting in the damage of membranes, the oxidation of proteins and the mutation of DNA, ROS are also important determinants for the normal growth and metabolism in a variety of cells. The latter has become evident for the last decade and pointed to a rather widespread and exciting role of H202 and ROS as second messengers for various signals in bacteria, plants and mammalian cells (16, 83, 93). Since the production of ROS increases proportionally with the 02 tension they are likely to be involved in the modification of transcription factors modulating gene activity in response to the ambient Po2. The role of ROS as 02 messengers has been supported by the finding that, similar to a typical response to hypoxia, treatment of healthy human volunteers with the antioxidant N-acetyl-cysteine (NAC) enhances the hypoxic ventilatory response (HVR) and blood erythropoietin concentration (40). Thus, NAC or its biochemical derivates, cysteine and glutathione, appear to mimic hypoxia. Furthermore, the proposal that H202 and derived ROS such as OH' can serve as mediators ofthe 02 signal was based on findings in HepG2 cells and carotid body preparations showing the presence of a low output oxidase which may be an NADPH oxidase isoform able to convert oxygen to superoxide and thus act as an oxygen sensor (33). Subsequently, Po2-dependent OH' production has been demonstrated in hepatoma cells and primary hepatocytes implicating that hypoxia is associated with decreased OH" levels. The response to hypoxia can be mimicked under normoxia by the application of the OH' radical scavenger dimethyl thiurea (DMTU) which reduces OH' levels (20, 50) (Fig. 3). In line with these observations were findings by several laboratories demonstrating decreased amounts ofROS under hypoxic conditions in the lung (4).

Figure 3. Reduction of OH' formation by hypoxia in primary rat hepatocyte cultures. Mimicry of venous P02 by DMTU (0.5 mM). Hepatocytes were cultured for 24 hrs under normoxia ( 16% 02). At 24 hrs, the media was changed and cells were further cultured for 2 hrs under 8% 02 or in the presence of DMTU (16% 02). In each experiment the OH' level measured in the control under 16% 02 was set equal to 100%. Values are means±SE of 3 independent culture experiments with 32 measurements per point each. * PO.01 16% 02 vs. 8% 02, ** P<0.01 control vs. DMTU.

5.4. Oxygen Sensing and Modification of HIF-la by ROS

The concept that H202 and other ROS play an important role in oxygen sensing was further supported by findings demonstrating that application of H202 repressed the hypoxia- and HIF-1 -dependent EPO production in HepG2 cells (89) and decreased the upregulation of tyrosine hydroxylase expression in PC 12 cells by hypoxia. Addition of H202 prevented the downregulation of the glucagon-dependent PCK mRNA expression by hypoxia in primary hepatocytes (51). These findings implicate that addition of H202 prevents the hypoxic

Figure 3. Reduction of OH' formation by hypoxia in primary rat hepatocyte cultures. Mimicry of venous P02 by DMTU (0.5 mM). Hepatocytes were cultured for 24 hrs under normoxia ( 16% 02). At 24 hrs, the media was changed and cells were further cultured for 2 hrs under 8% 02 or in the presence of DMTU (16% 02). In each experiment the OH' level measured in the control under 16% 02 was set equal to 100%. Values are means±SE of 3 independent culture experiments with 32 measurements per point each. * PO.01 16% 02 vs. 8% 02, ** P<0.01 control vs. DMTU.

DMTU

response and restores the normoxic response. This assumption was further supported by findings demonstrating that addition of H202 to cells grown under hypoxia resulted in the destabilization of the HIF-1 a protein in Hep3B cells (44) and the HIF-2a protein in HeLa cells (106). Furthermore, treatment with the antioxidants pyrrolidine dithiocarbamate (PDTC) and NAC increased HIF-la levels in alveolar type II epithelial cells (38). The redox processes modifying both HIF-la and HIF-2a appeared to predominantly affect the C-terminal transactivation domain (TAD-C). Within this domain, the cysteine 800 of HIF-la and the cysteine 848 of HIF-2a seem to be critical for transactivation, and the oxidation/reduction state ofthese cysteines is dependent on the presence ofRef-1 (12, 21, 44, 97). The regulation of HIF-1 DNA binding activity by ROS appeared to be evolutionary conserved since Drosophila HIF-D was also shown to be sensitive to redox modifications (7, 76, 77).

Whereas superoxide anion radicals are less likely to act as a second messenger since they are not freely diffusible, their dismutation product H202 is more suitable to function as a second messenger in Due to its freely diffusible non-charged character it can participate in two- and one-electron transfer reactions in the cells. Usually H202 is degraded by glutathione peroxidase in the cytosol and mitochondria or by catalase in peroxisomes or non-enzymatically converted into hydroxyl anions and hydroxyl radicals (OH*) in the presence of Fe2+ in a Fenton reaction:

Thus, a Fenton reaction adjacent to proteins or even transcription factors may affect Fe-S clusters or cysteine residues within regulatory proteins. It has been demonstrated that this H202-degrading reaction takes place in a perinuclear compartment and is involved in 02 modulated gene expression (20, 50) as well as in carotid body nervous discharge (15). Thus, it appears likely that the redox-sensitive HIF-1 a may be a target within an 02 sensing system involving ROS and a Fenton reaction. However, it remains open in which cellular compartment the Fenton reaction takes place and whether transcription factors regulating the 02-dependent gene expression or ion channels triggering nervous activity are located in this compartment.

5.5. Measurement of Intracellular Reactive Oxygen Species

Most of the chemicals commonly used to detect ROS may interact with a variety of radicals. Therefore, it appears to be difficult to use an indicator reacting with OH" in more or less specific way and to allow the detection and subcellular localization of the Fenton reaction. This problem was solved by the use of the non-fluorescent dye dihydrorhodamine 123 (DHR 123) which can be converted by OH' into fluorescent rhodamine 123 (RH 123) (20, 50, 53).

Thereby, the OH' rearrange the 7i-system of DHR 123, yielding fluorescent RH 123 (20). Furthermore, specific experimental conditions are required to measure the Fenton reaction in response to hypoxia such as: i) a non-photo toxic irradiation of the cells to avoid ROS generation during fluorescence excitation; ii) a minimal intracellular DHR 123 deposition to minimize secondary RH 123 dye distribution via diffusion and channel transport (103); iii) the start of kinetic measurements under conditions where ROS levels are expected, like hypoxia, since DHR 123 to RH 123 conversion is irreversible; iv) the maintenance of physiological conditions oftissue culture during the measurements to minimize cell stress; and v) an independent proof of intracellular ROS turnover.

L

0

* *

* m ^

Figure 4. Oxygen-dependent generation of OH". One single HepG2 cell cultured in hypoxia for 60 mins was treated for 5 mins with 30 nM DHR123 before rhodamine 123 fluorescence (in white) was visualized by 2P-CLSM (A). Then the same cell was exposed to normoxia and imaged again (C). White perinuclear spots indicate an increased oxdation of DHR 123 to RH 123 under normoxia due to an enhanced Fenton reaction mediated OH' generation. The same HepG2 cell under hypoxia (B) and normoxia (D) was challenged with light of a wavelength range between 400 and 500 nm for 5 seconds to mediate OH" generation by photoreduction. Normoxia shows a significant higher OH' generation mediated by photoreduction than hypoxia. Dimensions of the X, Y, Z axis are given in fim.

Figure 4. Oxygen-dependent generation of OH". One single HepG2 cell cultured in hypoxia for 60 mins was treated for 5 mins with 30 nM DHR123 before rhodamine 123 fluorescence (in white) was visualized by 2P-CLSM (A). Then the same cell was exposed to normoxia and imaged again (C). White perinuclear spots indicate an increased oxdation of DHR 123 to RH 123 under normoxia due to an enhanced Fenton reaction mediated OH' generation. The same HepG2 cell under hypoxia (B) and normoxia (D) was challenged with light of a wavelength range between 400 and 500 nm for 5 seconds to mediate OH" generation by photoreduction. Normoxia shows a significant higher OH' generation mediated by photoreduction than hypoxia. Dimensions of the X, Y, Z axis are given in fim.

These criteria can be fulfilled by using two photon confocal laser scanning microscopy which is not phototoxic due to infrared light (43). RH 123 fluorescence was registered by a photo multiplier, digitized and visualized. The signal-to-noise ratio was determined and images were deconvolved using the Maximum Likelihood Estimation (MLE) method. The data were reconstructed with the Application Visualization System (Waltham). Calculation ofisosurfaces was performed using a marching cube algorithm (96). In our measurements, a 5-min incubation of cells (with 30 |iM DHR 123) kept under physiological conditions in a microscope tissue culture chamber enabling observation at 37°C under a variable gas atmosphere gave an optimal dye deposit which was fully convertible to fluorescent RH 123 only under normoxia in combination with 5-sec blue light illumination. This drastic ROS increase under normoxia was contrasted by the nearly-missing illumination reaction under hypoxia, which renders short term blue light illumination as an ideal prove for intracellular ROS turnover (Fig. 4). Under these conditions hypoxia was accompanied by a decrease in ROS levels, in agreement with earlier studies (23).

Figure 5. A Fenton reaction in transmission of the 02 signal and regulation of 02 modulated genes. Po2 is sensed via a heme protein producing H202 below the threshold exerting oxidative stress. H202 is then diffused to the close vicinity of the nucleus due to the Fenton reaction, yielding OH'. Under a high Po2, OH" oxidizes SH-groups (e.g., the cystein 800 in HIF-la) and shifts the balance to the oxidized state. Furthermore, 02 which escapes binding by the heme protein may be additionally used by the asparagine (FIH) and proline hydroxylases (PHD'S) which modify HIF-la directly to inhibit cofactor recruitment and to mediate VHL binding to promote proteasomal degradation, respectively. In hypoxia, reduced ROS levels and decreased HIF-la hydroxylation lead to nuclear translocation of HIF-la, its dimerization with HIF-ip (ARNT), recruitment of cofactors and binding to hypoxia response elements (HRE). In addition, crosstalk between different signaling cascades activated by high or low ROS levels may also influence the stability of HIF-ip, thus allowing to fine tune HIF-1-dependent target gene expression.

Figure 5. A Fenton reaction in transmission of the 02 signal and regulation of 02 modulated genes. Po2 is sensed via a heme protein producing H202 below the threshold exerting oxidative stress. H202 is then diffused to the close vicinity of the nucleus due to the Fenton reaction, yielding OH'. Under a high Po2, OH" oxidizes SH-groups (e.g., the cystein 800 in HIF-la) and shifts the balance to the oxidized state. Furthermore, 02 which escapes binding by the heme protein may be additionally used by the asparagine (FIH) and proline hydroxylases (PHD'S) which modify HIF-la directly to inhibit cofactor recruitment and to mediate VHL binding to promote proteasomal degradation, respectively. In hypoxia, reduced ROS levels and decreased HIF-la hydroxylation lead to nuclear translocation of HIF-la, its dimerization with HIF-ip (ARNT), recruitment of cofactors and binding to hypoxia response elements (HRE). In addition, crosstalk between different signaling cascades activated by high or low ROS levels may also influence the stability of HIF-ip, thus allowing to fine tune HIF-1-dependent target gene expression.

However, other reports described enhanced ROS levels in hypoxia which decreased by returning to normoxia although the ROS-induced dye oxidation should be irreversible (13, 14). These findings may be explained by intensity changes of the oxidized dye, but do not appear to be due to changes in ROS.

Our findings suggest that the Fenton reaction primarily occurs in the cytoplasm. Since only 0.15% of the electron flow during mitochondrial respiration gives rise to the formation of ROS, this rather small amount of ROS is unlikely to escape to the cytoplasm due to the effective mitochondrial ROS scavenging systems including manganese superoxide dismutase and glutathione peroxidase (95). Our findings thus underline the importance of cytoplasmic, but not mitochondrial ROS as second messengers in 02 sensing.

Was this article helpful?

0 0
Reducing Blood Pressure Naturally

Reducing Blood Pressure Naturally

Do You Suffer From High Blood Pressure? Do You Feel Like This Silent Killer Might Be Stalking You? Have you been diagnosed or pre-hypertension and hypertension? Then JOIN THE CROWD Nearly 1 in 3 adults in the United States suffer from High Blood Pressure and only 1 in 3 adults are actually aware that they have it.

Get My Free Ebook


Post a comment