Up-regulation of Hypoxia Inducible Factor 1α by Isoflurane in Hep3B Cells

No human or animal subjects were used for this study. All reagents were obtained from Sigma Chemical (St. Louis, MO) unless specified in the text. Hep3B cells (American Type Culture Collection, Manassas, VA) were maintained in minimum medium at 37°C with Earle’s salts supplemented with 10% fetal bovine serum (Hyclone, Logan, UT), essential amino acids, pyruvate, 100 U/ml penicillin, and 0.1 mg/ml streptomycin. Cells were used for experiments when they were 80% confluent. Hypoxia and normoxia protocol was set up as we described previously.13Briefly, on the day of the experiment, the medium was replaced with a thin layer of fresh medium (0.15 ml/cm²) to decrease the diffusion distance of the ambient gas. Hypoxia was produced by placing cells in a humidified airtight chamber (approximately 5 l in volume) with inflow and outflow valves, and hypoxic gas mixture (1% O², 5% CO², and 94% N²) was delivered at 10 l/min for 25 min. The airtight incubator was kept at 37°C for preset time periods, whereas that which received no intervention was kept under normoxia condition (21% O², 5% CO², and 74% N²) for the preset time periods.

Isoflurane exposure was performed as described previously.14,15Briefly, cell dishes were kept in the airtight chamber housed within a water jacketed incubator maintained at 37°C. An in-line calibrated anesthetic agent vaporizer was used to deliver isoflurane to the gas phase of the culture wells. Normoxic gas was administered at a flow rate of 3 l/min, until the appropriate effluent concentration of the anesthetic was achieved. This required approximately 10–20 min. The flow rate was then decreased and maintained at 900 ml/min for preset periods at 37°C. Effluent isoflurane, oxygen, and carbon dioxide concentrations were continuously monitored via  a sampling port connected to an anesthetic agent analyzer (5250, RGM; Datex-Ohmeda, Madison, WI). The pH of the medium before and after exposure was 7.4. The agents applied during the incubation, in addition to various concentrations of isoflurane, included cobalt chloride and desferrioxamine (hypoxia mimetic, 100 μm), cycloheximide (inhibitor of protein translation, 100 μm), and MG132 (a proteasome inhibitor, causes cells to accumulate HIF-1α protein, 50 μm). In experiments involving these compounds, they were added to the culture media just before the start of hypoxia or isoflurane stimulation, and the media was replaced at the end of the stimulation.

Whole cell extracts were prepared as we have described previously.13A total of 100 μg protein was loaded onto a 10% SDS–polyacrylamide gel, and after electrophoresis, it was blotted onto nitrocellulose membranes. The primary rabbit anti-human HIF-1α (Upstate, Charlottesville, VA) polyclonal antibodies were used at 1:1,000 dilution, and anti-HIF-1β monoclonal antibody (Novus Biologicals, Littleton, CO) was used at 1:1,000 dilution. The anti-rabbit immunoglobulin G secondary antibody (KPL, Gaithersburg, MD) was used at 1:2,000 dilution, and the signal was analyzed by enhanced chemiluminescence (Chemicon, Temecula, CA). The intensity of each band was quantified with the software Quantity One-4.2.3 (Bio-Rad, Hercules, CA) and normalized to HIF-1β by density analysis.

Cells were harvested and RNA was isolated as we described previously.13Real-time quantitative polymerase chain reaction (PCR) estimation of messenger RNA (mRNA) levels was performed as described previously.16Real-time quantitative PCR was performed by using SYBR green ??? as fluorescent dye (TaKaRa Biotechnology, Dalian, China). Primers were commercially obtained (Shanghai Sangon Biologic Engineering & Technology and Service Co. Ltd., Shanghai, China) (table 1). Amplification and detection were performed by using an ABI PRISM 7700 detection system (Applied Biosystems, Foster City, CA) as follows: 1 cycle at 95°C for 10 s, 40 cycles at 95°C for 5 s, and 60°C for 30 s. Real-time PCR was performed in triplicate reactions with 20 ng complementary DNA in a final volume of 10 μl containing 1 × SYBR Green Master Mix and 200 nm of both primers. Agarose gel electrophoresis, purification, and DNA sequencing confirmed the identity of the PCR products. Reference gene β-actin, which was not altered by isoflurane or any other treatment applied in this study (data not shown), was used for normalization of the expression data. All reverse-transcription and real-time quantitative PCR was performed in triplicate from three separate culture dishes.

The levels of VEGF protein secreted by the cells in the medium were determined by the VEGF enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, MN). In brief, subconfluent cells were cultured in serum-free medium under normoxia overnight, followed by incubation with indicated stimuli for 16 h. The medium was collected, and VEGF protein concentrations were measured according to the manufacturer’s instructions. The results were normalized to the number of cells per plate.

Transient transfection of a luciferase reporter driven by HRE is one of the most sensitive systems for the measurement of HIF-1 activity. The Dual-luciferase Reporter Assay System exploits the differing biochemical requirements for luminescence of the firefly (Photinus pyralis ) and sea pansy (Renilla reniformis ) luciferase proteins. This allows for the sequential quantitative measurement of both luciferase activities in a single protein extract. The luciferase reporter plasmids (pGL3-luc) harboring HRE from human VEGF gene promoter region (−88/+54) and its HRE mutant were kindly provided by Amit Maity, M.D., Ph.D. (Assistant Professor of Radiation Oncology, Department of Radiation Oncology, University of Pennsylvania School of Medicine, Philadelphia, PA). The Renilla  luciferase expression plasmid pRL-SV40 was obtained from Promega (Madison, WI).

Hep3B cells were plated 5 × 10⁴cells/well on the day before transfection. Each plate was transfected with a mixture of 200 ng reporter gene plasmid and 50 ng control plasmid pRL-SV40 for transfection efficiency, premixed with Lipofectamine 2000 and Opti-MEMI reduced serum medium (Invitrogen, Shanghai, China). After 6 h, cells were recovered overnight and subsequently incubated with or without isoflurane under normoxic conditions with or without desferrioxamine for another 16 h. The cells were harvested, and the luciferase activity of each well was measured in the same dosage of cell lysate with the use of Dual-luciferase Reporter Assay System (Promega).

The small interference RNA (siRNA) oligonucleotide for HIF-1α was designed as described previously.17Sense and antisense RNA strands (5′-UGUGAGUUCGCAUCUUGA UdTdT-3′) and (5′-AUCAAGAUGCGAACUCACAdTdT-3′) with two 5′ deoxy-thymidine overhangs were commercially synthesized (KanChen Biology Co. Ltd., Shanghai, China) and annealed at a final concentration of 20 μm each by heating at 95°C for 1 min and incubating at 37°C for 1 h in annealing buffer (20 mm potassium acetate, 6 mm HEPES-KOH, pH 7.4, and 0.4 mm magnesium acetate). Transfection of siRNA was performed at a concentration of 100 nm using Lipofectamine 2000. As a control for HIF-1α–siRNA, we used either a corresponding random siRNA sequence (scrambled-siRNA: sense 5′-AGUUCAACGACCAGUAGU CdTdT-3′ and antisense 5′-GACUACUGGUCGUUGAdTdT-3′) or in case of luciferase reporter gene assay also luciferase siRNA (sense 5′-CUUACGCUGAGUACUUCGAdTdT-3′ and antisense 5′-UCGAAGUACUCAGCGUAAGdTdT-3′) targeting firefly luciferase.18Control cells were transfected without oligonucleotides under the same conditions. After transfection, the cells were split in aliquots and grown in six-well dishes for 24 h and then exposed to normoxic or indicated condition for preset time periods. Cells were lysed and total RNA or protein was extracted as described above.

One-way analysis of variance was used to compare mean values across multiple treatment groups with a Tukey post hoc  multiple comparison test. In all cases, statistical significance was defined as P < 0.05. All data are expressed throughout the text as mean ± SD.

The biologic activity of HIF-1 is determined by the protein expression and transactivation activity of HIF-1α subunit. Isoflurane stimulated HIF-1α accumulation in a concentration-dependent manner in Hep3B cells; an increase could be seen at 1%, whereas maximal induction was attained at 2% (fig. 1A). In contrast, HIF-1β expression remained unchanged by isoflurane or any other treatment applied in the study, attesting to the fact that HIF-1β is constitutively expressed. So we used it as a loading control as described previously.19Time course studies were then performed on Hep3B cells stimulated with 2% isoflurane. An increase could be seen after a 2-h stimulation, and maximal induction was attained between 4 and 8 h in the presence of isoflurane (fig. 1B). However, this induction is slow as compared with the hypoxic induction of HIF-1α in these same cells. HIF-1α induction in Hep3B cells had already attained a maximum after a 2-h stimulation under hypoxia (fig. 1B). This indicates that different mechanisms might be implicated in the activation of HIF-1α by hypoxia and by isoflurane in these cells. HIF-1α is predominantly regulated at the protein level by oxygen-dependent prolyl hydroxylases. Cobalt chloride and desferrioxamine cause accumulation of HIF-1α protein through inhibition of prolyl hydroxylases and thus mimic hypoxic signaling.20As a positive control, we stimulated Hep3B cells with these reagents and isoflurane for the same period of time. HIF-1α protein level in normoxic cells was increased to a similar level by desferrioxamine (100 μm), cobalt chloride (100 μm), and 2% isoflurane and hypoxia (1% O²) for 6 h (fig. 1C). The combination of isoflurane and desferrioxamine resulted in greater increases in HIF-1α protein levels than those seen with either treatment alone. Desferrioxamine-treated cells were used in place of hypoxic cells because reoxygenation occurred when isoflurane was added. Taken together, these results show that the incubation of Hep3B cells with isoflurane results in a time- and concentration-dependent elevation of HIF-1α protein levels.

To obtain a better understanding of the process involved in HIF-1α accumulation in response to isoflurane, we investigated the effect of isoflurane on the amount of HIF-1α mRNA. Hep3B cells were incubated with hypoxia–desferrioxamine–isoflurane for 6 h (fig. 2A) or stimulated with isoflurane for indicated time periods (fig. 2B). The change in HIF-1α mRNA did not reach statistical significance under all these conditions, suggesting that isoflurane does not regulate HIF-1α mRNA transcription just as hypoxia in Hep3B cells.21To elucidate posttranscriptional mechanisms, we examined whether isoflurane would stabilize HIF-1α. Therefore, Hep3B cells were exposed to desferrioxamine–isoflurane for 6 h and then returned to normoxia. Although isoflurane increased HIF-1α protein levels when incubated with desferrioxamine, it did not affect HIF-1α degradation on reoxygenation (fig. 3A). On the transition from desferrioxamine to reoxygenation, HIF-1α completely disappeared within 15 min in desferrioxamine and desferrioxamine plus isoflurane–treated cells. Inhibition of the proteasome pathway by MG132 led to a high level of HIF-1α, and isoflurane induced HIF-1α accumulation in the presence of MG132. This elevated expression of HIF-1α in the presence of desferrioxamine and MG132 prevented us from detecting any effect of isoflurane on HIF-1α levels (figs. 3B and C). To analyze the isoflurane effect on HIF-1α synthesis, we performed a time course of HIF-1α disappearance in the presence of the protein translation inhibitor, cycloheximide. The addition of cycloheximide prevented the isoflurane-dependent induction of HIF-1α (fig. 3D). In the presence of cycloheximide, the half-life of HIF-1α was greater than 60 min in desferrioxamine-treated cells but less than 15 min in isoflurane-treated cells (fig. 3E). Together, these results suggest that isoflurane increases HIF-1α protein levels through a translation-dependent pathway but not via  the inhibition of its degradation.

The data presented above show that the level of HIF-1α protein in Hep3B cells is increased upon isoflurane challenge, by promoting its translation. We next asked was the effect of isoflurane was on HIF-1 target genes. HO-1, iNOS, and VEGF are considered typically HIF-1–regulated genes.22–24Hypoxia significantly increased HO-1, iNOS, and VEGF mRNA expression, and isoflurane induced mRNA levels of these genes in a concentration-dependent manner (fig. 4A). Identical induction was observed between iNOS, HO-1, and VEGF by isoflurane. Thus, all further experiments were only performed on VEGF. Treatment of the cells with isoflurane for 16 h resulted in VEGF protein accumulation (fig. 4B).

We then tested the ability of isoflurane to activate an HIF-1–dependent luciferase reporter gene construct. Exposure of Hep3B cells to isoflurane or 1% O²for 16 h resulted in an increase of luciferase expression, as found for the endogenous VEGF gene (fig. 4C). The induction of luciferase expression by isoflurane disappeared when the reporter gene was under the control of a promoter lacking functional HRE element. Hence, the increment of HIF-1 target gene expression observed after isoflurane stimulation is attributable, at least in part, to the induction of HIF-1 acting through HRE. To further demonstrate that HIF-1α is required for the induction of VEGF mRNA expression in response to isoflurane, Hep3B cells were mock transfected or transfected with a siRNA^HIF-1αthat targets HIF-1α mRNA for degradation. After transfection, the cells were exposed to desferrioxamine or isoflurane. As expected, Hep3B cells were efficiently cleared from HIF-1α protein, although a faint band was still visible in siRNA^HIF-1αand isoflurane-treated cells (fig. 5A). A subsequently performed luciferase reporter gene assay revealed a drastic loss of desferrioxamine and isoflurane-dependent induction (fig. 5B). VEGF mRNA levels were slightly reduced in cells treated with siRNA^HIF-1αin normoxia and dramatically reduced in isoflurane-induced VEGF mRNA expression (fig. 5C). However, the effect of siRNA^HIF-1αwas modest (approximately 80% the control values), indicating that HIF-1 pathway is not the only mechanism involved in isoflurane-dependent induction of VEGF mRNA. As an additional control for siRNA transfection, we used luciferase-siRNA targeted against firefly luciferase. Luciferase-siRNA inhibited firefly-luciferase reporter gene expression more strongly when compared with the effects of HIF-1α-siRNA (fig. 5D).

Isoflurane is known to affect the expression of HIF-1–responsive genes.6–12However, little is known about the extent to which induction of HIF-1, the master regulator of oxygen homeostasis, is affected by the presence of isoflurane. Herein, for the first time, we provide evidence that isoflurane up-regulates HIF-1α protein expression and activity in a time- and concentration-dependent manner; elevates HIF-1–responsive genes HO-1, iNOS, and VEGF mRNA levels; and induces VEGF through an HIF-1–dependent pathway in Hep3B cells. Moreover, in contrast to hypoxia, which is a major activator of HIF-1α, isoflurane does not regulate HIF-1α through the inhibition of its degradation but via  a translation-dependent mechanism.

In this study, hypoxia treatment significantly increased HIF-1α protein levels in Hep3B cells, which is consistent with previous findings that hypoxia stimulates HIF-1α expression in the cell line.21We also found that isoflurane could increase HIF-1α protein level under normoxic conditions. The possibility that isoflurane might increase HIF-1α gene expression was rejected because HIF-1α mRNA levels were not elevated by isoflurane treatment. The lack of effect of isoflurane on transcriptional activation of HIF-1α was not surprising in view of the fact that hypoxic regulation of HIF-1α is also primarily determined by stabilization of HIF-1α protein.2Further, isoflurane did not affect HIF-1α degradation during reoxygenation. The findings indicated that isoflurane-inducible accumulation of HIF-1α protein might be due to an increase of HIF-1α protein synthesis. MG132 blocks the proteasomal degradation machinery, so the rate of HIF-1α accumulation is at large a function of the rate of HIF-1α synthesis in the presence of MG132. Isoflurane significantly enhanced HIF-1α protein levels in the presence of MG132 under normoxic conditions. The effect of isoflurane was abrogated by cycloheximide, an inhibitor of translation. A comparison of the half-life of HIF-1α in the presence of isoflurane and desferrioxamine shows that isoflurane does not stabilize the HIF-1α protein. Nevertheless, we cannot exclude the possibility that isoflurane can regulate the translation of a protein which inhibits HIF-1α degradation.

We also demonstrate that mRNA for HO-1, iNOS, and VEGF are elevated in the presence of isoflurane. Preinduction of HO-1 and application of carbon monoxide have been shown to prevent reperfusion injury after liver transplantation or after systemic ischemia in several experimental studies.11,25iNOS has been shown to play an important role in cardioprotection or neuroprotection induced by many stimuli including isoflurane.7–9It could be argued that the observed effect of isoflurane on HIF-1–responsive gene expression is independent of HIF-1 and through a response element other than HRE. To investigate this possibility, we transiently transfected Hep3B cells with a reporter plasmid construct expressing the firefly luciferase gene under the control of the wild-type HRE from VEGF gene promoter or an HRE-mutated version of this promoter. HIF-1α was indispensable for isoflurane-induced stimulation of HIF-1–dependent reporter gene and target gene VEGF expression as shown by knocking down HIF-1α by siRNA. Activation of VEGF by HIF-1 contributes to cardioprotection.26Therefore, isoflurane alone increases HIF-1α protein levels under normoxic conditions and might potentiate hypoxia-induced HIF-1α expression. Increased synthesis by isoflurane cooperates with hypoxic stabilization and accumulation of HIF-1α that is due to inhibition of prolyl hydroxylase activity, and ensures enhanced expression of the HIF-1 target genes such as VEGF.

Recent studies have shown that a number of nonhypoxic stimuli could strongly increase HIF-1α in a cell-specific manner.27Interestingly, some of these increases have been shown to be equal to or greater than the hypoxic induction of HIF-1α. In this study, we show that in Hep3B cells, isoflurane is an excellent nonhypoxic stimulator and activator of the HIF-1 transcription factor. However, the signal transduction pathway of the induction by isoflurane remains unclear. Cumulative evidences suggest that isoflurane may exert protection against hypoxic–ischemic injury via  the activation of phosphatidylinositol 3-kinase and intracellular calcium pathways. Isoflurane has been shown to activate phosphatidylinositol 3-kinase pathway,28,29and this signaling cascade has also been shown to play a major role in the induction of HIF-1α protein levels by different nonhypoxic stimuli.30–36Further, studies suggest that activation of phosphatidylinositol 3-kinase pathway preferentially increases HIF-1α protein translation.35,36Therefore, isoflurane may have induced HIF-1αvia  a phosphatidylinositol 3-kinase pathway. Other study demonstrated that calcium signaling stimulates translation of HIF-1α during hypoxia.37Isoflurane neuroprotection in hypoxic hippocampal slice cultures involves increases in intracellular calcium and mitogen-activated protein kinases.38–40Further investigation will be required to define whether either indicated signaling or other pathways are involved in this process.

It has been demonstrated that halothane reversibly blocks hypoxia-induced HIF-1α protein accumulation and transcriptional activity at clinically relevant doses.41Differential physiologic effects of volatile anesthetics are well known. For example, halothane and isoflurane produce strikingly different inotropic and vasodilatory effects.42In the experiment setting of healthy volunteers, isoflurane and halothane differentially affect the hypoxic chemoreflex loop.43Hepatic HO-1 expression was differentially regulated by several volatile anesthetics.10In addition, isoflurane induces protection of human myocardium against anoxia injury but not for halothane,44,45and isoflurane altered several genes involved with neurotransmitter transport, signaling and cellular structure, whereas halothane produced few detectable changes in primary cortical neurons.46Therefore, these previous studies as well our current study supports our hypothesis that the two anesthetics (isoflurane and halothane) genuinely differ in their actions on HIF-1 activation. These studies suggested, despite HIF being regulated primarily by oxygen availability, other signals received by the cell modulate this response. Alternatively, these findings might indicate the regulation of HIF-1α by anesthetics is partly dependent on the stimulus used and the cell type studied.

In our study, isoflurane at concentration of 1–4% up-regulated HIF-1α in Hep3B cells. These isoflurane concentrations are clinically relevant, because 1 minimum alveolar concentration (MAC) is 1.15%,47and isoflurane concentrations higher than 1 MAC are frequently used during anesthesia. In the liver, because of the unidirectional blood flow from the portal vein and hepatic artery to the central vein and because of the oxygen-consuming metabolic processes of the cells along the sinusoids, an oxygen gradient is formed. Therefore, the oxygen tension decreases from 65 mmHg in the periportal area to 35 mmHg in the perivenous area.48HIF-1α is present in mice liver under normoxic conditions and is further increased in response to systemic hypoxia (6% O²).49Our data for HIF-1α were based on the human hepatoma cell line Hep3B cell, and its use to test the regulation of HIF-1α is of great interest. However, the immortalized cell line cannot be considered as equivalent for differentiated hepatocytes. It would be useful to test whether the regulatory events described in Hep3B cells might be applicable to normal human liver. Although it is not appropriate to extrapolate our results directly to in vivo  conditions or clinical practice, our data suggest a potential site for volatile anesthetics to act in the liver. On the other hand, we treated cells with the anesthetics for rather short periods, whereas in the clinical setting, human bodies are usually exposed to anesthetics for longer periods than we used here. An increase in iNOS has been shown after the end of isoflurane anesthesia up to 24 h,7and changes in cerebral protein expression after desflurane anesthesia occur as rapidly as 3 h after the start of anesthesia and last up to 72 h.50Conversely, using organotypic hippocampal slices, it has been shown that isoflurane-induced neuronal degeneration is dependent on the duration of isoflurane exposure.15,51Further studies are needed from the point of view of time.

Our results suggest that HIF-1α should be involved in the activation of known hypoxic genes by isoflurane. However, we do not suggest that the HIF-1 pathway is the only mechanism involved in the activation of these genes. Multiple signaling pathways are activated by isoflurane in cells, and a number of pathways have been shown to play an important role in the activation of certain hypoxia-responsive genes. We believe that along with the other pathways and transcription factors, HIF-1 permits the maximal activation of these genes. Two recent studies by Hoetzel et al. ,10,11based on the regulation of hepatoprotection enzyme HO-1 by isoflurane in liver, revealed exciting novel signal transduction pathway involving protein kinase C and phospholipase A². Unfortunately, HIF-1α levels were not determined in these studies. It is hoped that future work will elucidate the impact of HIF-1 on the protective effect of isoflurane in a variety of physiologic and pathophysiologic conditions.

The authors thank Amit Maity, M.D., Ph.D. (Assistant Professor of Radiation Oncology, Department of Radiation Oncology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania), for providing the reporter plasmid construct and Jing Yuan Fang, Ph.D. (Professor of Gastroenterology, Shanghai Institute of Digestive Disease, Renji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China), for excellent technical assistance.