Human solid tumors contain hypoxic regions that have considerably lower oxygen tension than normal tissues. These impart resistance to radiotherapy and anticancer chemotherapy, as well as predisposing to increased tumor metastases. To develop a potentially therapeutic protein drug highly specific for solid tumors, we constructed fusion proteins selectively stabilized in hypoxic tumor cells. A model fusion protein, oxygen-dependent degradation (ODD)-β-galactosidase (β-Gal), composed of a part of the ODD domain of hypoxia-inducible factor-1α fused to β-Gal, showed increased stability in cultured cells under a hypoxia-mimic condition. When ODD-β-Gal was further fused to the HIV-TAT protein transduction domain (TAT47–57) and i.p. injected to a tumor-bearing mouse, the biologically active fusion protein was specifically stabilized in solid tumors but was hardly detected in the normal tissue. Furthermore, when wild-type (WT) caspase-3 (Casp3WT) or its catalytically inactive mutant was fused to TAT-ODD and i.p. injected to a tumor-bearing mouse, the size of tumors was reduced by the administration of TAT-ODD-Casp3WT but not by TAT-ODD-mutant Casp3. TAT-ODD-Casp3WT did not cause any obvious side effects on tumor-bearing mice, suggesting specific stabilization and activation of the fusion protein in the hypoxic tumor cells. These results suggest that the combination of protein therapy using a cytotoxic TAT-ODD fusion protein with radiotherapy and chemotherapy may provide a new strategy for annihilating solid tumors.

Solid tumors comprise >90% of all of human malignancies. Furthermore, biologically and therapeutically significant hypoxia occurs in many solid tumor masses. The hypoxic tumor cells are more resistant to radiotherapy and chemotherapy than their well-oxygenated counterparts and are capable of stimulating angiogenesis by the release of several growth factors (e.g., vascular endothelial growth factor, angiogenin, and platelet-derived growth factor) for endothelial cells (1, 2, 3). Although the specific delivery of anticancer gene products to hypoxic regions of solid tumors has been studied extensively (4, 5, 6, 7, 8, 9), an effective technique has not been established.

The transcription factor, HIF2-1α, is an important mediator of the hypoxic response of tumor cells and controls the up-regulation of a number of factors vital for solid tumor expansion, including angiogenic factors, such as vascular endothelial growth factor (10). The up-regulation of HIF-1α by decreased cellular O2 concentration involves increased mRNA expression, protein stabilization, nuclear localization, and trans-activation (11). Perhaps the most striking regulation occurs exclusively at the level of protein expression via changes in the half-life of the HIF-1α protein in response to hypoxia both in cultured cells and in vivo(12, 13, 14).

The ODD domain of HIF-1α is located in its central region and consists of ∼200 amino acid residues. The ODD domain controls the degradation of HIF-1α by the ubiquitin-proteasome pathway, and the deletion of this entire region is required for DNA binding and trans-activation in the absence of hypoxic signaling (15). Recently, the product of von Hippel-Lindau tumor suppressor gene was reported to mediate ubiquitination and proteasomal degradation of HIF-1α under normoxic conditions by interaction with the core of the ODD domain of HIF-1α (16, 17, 18).

HIV-1 TAT has been used to deliver functional biomolecules into cells. Although the entire protein can be used for this purpose, it is more efficient to use the highly basic region containing residues 49–57. Through covalent attachment to TAT47–57, several proteins have been delivered into cells, including an inhibitor of human papillomavirus type 16 (19), ovalbumin into the MHC class I pathway (20), the cyclin-dependent kinase inhibitors p27kip1(21) and p16INK4a(22), and a Casp3 protein (23). TAT79–57 has also been successfully used in vivo to deliver β-Gal into all tissues of the mouse, including the brain (24).

Casp3 is a major executioner protease in known apoptotic pathways. Casp3 remains dormant until the initiator caspases activate it by direct proteolysis. Activated Casp3 cleaves the inhibitor of caspase-activated DNase, resulting in cell death (25, 26, 27). Activation of Casp3 under hypoxic conditions has been reported in several cell types, such as ventricular myocytes, endothelial cells, and squamous cell lung carcinomas (28, 29, 30, 31).

Here we demonstrated that part of the ODD domain conferred oxygen-dependent stability to a fusion protein and that the covalent attachment of TAT47–57 and ODD548–603 to bioactive β-Gal allowed quick delivery and specific stabilization in a solid tumor. Furthermore, the replacement of β-Gal with Casp3 reduced tumor masses without any obvious side effects, demonstrating the feasibility and efficacy of protein therapy using a TAT-ODD fusion protein for solid tumors.

Cell Cultures and Reagents.

The human lung adenocarcinoma cell line A549, the human embryo kidney cell line HEK293, the human pancreatic cancer cell line CF/PAC-1, and the mouse fibroblast cell line NIH/3T3 were obtained from American Type Culture Collection. A549, HEK293, and NIH/3T3 were maintained in 5% FCS-DMEM, and CF/PAC-1 was maintained in 10% FCS-Iscove’s modified Dulbecco’s medium (Life Technologies, Inc., Rockville, MD) supplemented with penicillin (100 units/ml) and streptomycin (100 μg/ml). Cbz-LLL (Peptide Institute, Inc., Osaka, Japan) was dissolved in DMSO to 10 mm. Pimo hydrochloride (Hypoxyprobe-1; Natural Pharmacia International, Inc., Research Triangle Park, NC) was dissolved in normal saline (10 mg/ml).

DNA Constructs.

To construct pCH/ODD plasmids encoding a series of ODD-β-Gal fusion proteins, pCH was first constructed by annealing the oligonucleotides containing Kozak and NLS sequences (5′-AGCTTGACATGGCGCCTAAGAAGAAGAGGAAGA-3′ and 5′-GATCTCTTCCTCTTCTTCTTAGGCGCCATGTCA-3′) and inserting the annealed DNA fragment between HindIII and BglII sites of the pCH110 vector (Amersham Pharmacia Biotech, Milwaukee, WI). Then cDNA fragments encoding various lengths of the ODD domain of human HIF-1α were inserted between BglII and KpnI sites of pCH. The cDNA fragments encoding ODD were amplified by PCR using corresponding oligonucleotides as primers from cDNA prepared from total RNA from A549. The ODD fragments in pCH/557-574, pCH/562–569, pCH/557-571, and pCH/560-574 were prepared by annealing the corresponding oligonucleotides.

To construct pBAD/PC, pBAD/3-0, and pBAD/557-574, which encode the β-Gal protein fused to TAT, TAT-ODD/3-0, and TAT-ODD557–574, respectively, pCH/TAT/PC, pCH/TAT/3-0, and pCH/TAT/557-574 were first constructed. To prepare pCH/TAT/3-0, pCH/TAT/557-574, or pCH/TAT/PC, oligonucleotides encoding TAT47–57 were inserted into the BglII site of pCH/3-0 and pCH/557-574 or substituted with the BglI-KpnI ODD/3-0 fragment of pCH/3-0, respectively. Then, pCH/TAT/3-0, pCH/TAT/557-574, and pCH/TAT/PC were digested with HindIII, blunted with klenow-fragment, and further digested with SacI. The blunt-SacI DNA fragments were inserted into blunted BamHI (413) and SacI sites of pBAD/His/LacZ (Invitrogen, Carlsbad, CA), resulting in pBAD/3-0, pBAD/557-574, and pBAD/PC. To construct pGEX/3-0/Casp3WT encoding TAT-ODD/3-0-Casp3, Casp3 cDNA was amplified by PCR using corresponding oligonucleotides as primers from cDNA prepared from total RNA from A549 and substituted for β-Gal cDNA in pCH/TAT/3-0, and a DNA fragment encoding TAT-ODD/3-0-Casp3WT was inserted to SmaI and SalI sites of pGEX-6P-3 (Amersham Pharmacia Biotech, Uppsala, Sweden). Casp3MUT encoding a mutant Casp3(C163M) was constructed as described previously (23) and then substituted for Casp3 in pGEX/3-0/Casp3WT.

DNA Transfection, β-Gal Assay, and X-gal Staining.

For β-Gal assays, cells (1 × 104/well) were seeded onto 24-well plates and cotransfected with 2 μg of pCH plasmids (Fig. 1, A and B) and 0.2 μg of pRL-CMV (Promega, Madison, WI) as described (32). The following day, the cells were trypsinized, and aliquots were plated in duplicate onto 24-well plates (one well without Cbz-LLL and the other with 5 μm Cbz-LLL). Later (24 h), the cells were lysed with 200 μl of 1 × Reporter Lysis Buffer (Promega). The lysates (50 μl) were subjected to the calorimetric enzyme assay in triplicate using a β-Gal Enzyme Assay System (Promega), and obtained A405 nm was normalized by luminescence counts (10 s) of 10-μl lysates assayed with a Luciferase Assay System (Promega). The relative β-Gal activities were calculated by dividing the normalized counts of the lysate from the cells cultured with Cbz-LLL with the ones of the corresponding lysate from the untreated cells. The experiments were repeated at least three times. For X-gal staining, the cells transfected with pCH plasmids or treated with TAT-fusion proteins were washed three times with PBS, fixed with 1% formaldehayde-0.2% glutaraldehyde solution, washed twice with PBS, and then treated with an X-Gal staining solution (5 mm potassium ferrocyanide, 5 mm potassium ferricyanide, 1 mm MgCl2, and 0.04% X-gal) at 37°C. The experiments were repeated twice with similar results.

Hypoxia.

Hypoxia was achieved by incubating cells in a humidified environment at 37°C in CO2 incubators maintained at 94% N2, 5% CO2, and 1% O2.

Preparation of TAT-fusion Protein and in Vitro Assay.

To prepare TAT-, TAT-ODD/3-0-, and TAT-ODD557–574-β-Gal proteins, pBAD/PC, pBAD/3-0, and pBAD/557-574 were transformed into LMG194 bacteria (Invitrogen). TAT-fusion proteins were purified over a Ni2+-NTA column (Qiagen) following the manufacturer’s instruction manual. The β-Gal activities of the final preparations were determined by β-Gal assay (Promega). To prepare TAT-ODD/3-0-Casp3WT and TAT-ODD/3-0-Casp3MUT, pGEX/3-0/Casp3WT and pGEX/3-0/Casp3MUT were transformed into BL21(DE3)pLysS (Novagen, Madison, WI), and the fusion proteins were purified using Glutathione Sepharose 4B and PreScission Protease (Amersham Pharmacia Biotech) following the manufacturer’s instruction manual.

Xenograft Tumor Growth, TAT-fusion Protein Injection, and Immunohistochemical Analysis.

CF/PAC-1, human pancreatic cancer cells (2 × 106) were s.c. injected into the flanks of 6–8-week-old nude mice (CLEA Japan, Inc., Tokyo, Japan). The mice carrying tumors were i.p. injected with 100 μl of TAT-β-Gal, TAT-ODD-β-Gal protein (containing 100 units of β-Gal activity), or TAT-ODD-Casp3 protein 2–3 weeks after the xenograft. To detect the hypoxic regions, the tumor-bearing mice were i.p. injected with Hypoxyprobe-1 (60 mg/kg) 2 h after TAT-fusion protein administration. Later (2 h), the tumors (∼300 mm3) and the livers were surgically excised, fixed in 4% formaldehyde, and embedded in paraffin for hematoxylin and anti-β-Gal or antipimonidazol staining or were dropped into liquid N2 immediately after excision and kept at −80°C until staining for X-gal. Immunostaining of the tissue sections for the hypoxia marker, Pimo, and β-Gal was carried out with a Hypoxyprobe-1 kit (Natural Pharmacia International, Inc., Belmont, MA) and rabbit anti-β-Gal Ab (ICN Pharmaceuticals, Inc., Aurora, OH) by the indirect immunoperoxidase methods as described previously (33). X-gal staining of the tissue sections was done as described previously (34).

Assay for Apoptosis.

NIH/3T3 cells (1 × 104/well) were seeded onto 24-well plates and preincubated for 24 h in the presence of the solvent (DMSO) or 5 μm Cbz-LLL. Then, TAT-ODD-Casp3 proteins (7.5 μg/30 μl) were added to the fresh culture medium, and the cells were further incubated for 24 h under normoxic (20% O2) or hypoxic (1% O2) conditions. The photos were taken under an inverted microscope.

Assay for Casp3 Activity.

TAT-ODD-Casp3 proteins (7.5 μg/30 μl) were added to the culture of HEK293 cells (5 × 105 Cells/0.5 ml) and incubated for 4 h in the presence of the solvent (DMSO) or 10 μm Cbz-LLL. Then the cells were harvested, and Casp3 activity in a 50-μg cell lysate was detected as the amount of chromophore p-nitroanilide after cleavage from the labeled substrate DEVD-p-nitroanilide (CPP32/Casp3 colorimetric protease assay kit; MBL, Nagoya, Japan) and quantified using a microtiter plate reader at 405 nm.

Determination of the Core of the ODD Domain Responsible for ODD of an ODD-fusion Protein.

Because the stability of HIF-1α is regulated through the ODD domain, we assumed that the stability of a protein fused to an appropriate region of ODD domain would be under the same control as HIF-1α. To identify the region responsible for ODD of an ODD-fusion protein, we constructed a series of fusion proteins composed of parts of the ODD domain of HIF-1α and β-Gal as a bioactive reporter. The plasmids encoding the NLS and various lengths of the ODD domain fused to the NH2 terminal of β-Gal (Fig. 1, A and B) were transfected to HEK293 cells. The β-Gal activity of these cells was examined by X-gal staining after culturing in the presence or absence of Cbz-LLL, a peptide aldehyde that inhibits the major peptidase activities of proteasomes and reduces the degradation of proteins (35). Therefore, culture with Cbz-LLL renders the cells in a hypoxia-mimic condition (36). The results of X-gal staining were summarized as the ODD regulation, which refers to the increased β-Gal activity and increased number and thickness of blue color in X-gal-stained cells in Fig. 1,B. In the cells transfected with the plasmids encoding fusion proteins consisting of β-Gal and the HIF-1α residues 401–447 (pCH/0-1), 401–497 (pCH/0-2), 401–547 (pCH/0-3), 448–497 (pCH/1-2), 448–547 (pCH/1-3), or 579–603 (pCH/4-0), the X-Gal staining pattern was same between the Cbz-LLL-treated and untreated cells (Figs. 1,B and 2,A). On the other hand, in the cells transfected with the plasmids encoding the fusion proteins consisting of β-Gal and the HIF-1α residues 401–603 (pCH/0-0), 448–603 (pCH/1-0), 498–603 (pCH/2-0), 548–603 (pCH/3-0), or 548–583 (pCH/3-4), more cells were stained deeper with X-gal in the Cbz-LLL-treated cells than the untreated ones (Figs. 1,B and 2 A).

To determine the core of ODD region responsible for ODD of the ODD-fusion proteins, we further narrowed down the ODD region. The residues 548–583 have 18 amino acid sequences (557–574) conserved among the related HIFs and between species. We constructed a plasmid encoding the fusion protein consisting of β-Gal and the HIF-1α residues 557–574 (ODD557–574) and examined its ODD regulation as above. Although a considerable number of cells was stained with X-gal in the absence of Cbz-LLL, more cells were stained deeper in the Cbz-LLL-treated cells than the untreated ones (Fig. 2,A), suggesting that ODD557–574 still retained the ODD function. However, the stability of fusion proteins containing the HIF-1α residues of 562–569 (pCH/562–569), 557–571 (pCH/557–571), or 560–574 (pCH/560–574) were no longer under the control of oxygen tension (Fig. 1 B), indicating that ODD557–574 contained the minimum ODD residues necessary for the ODD regulation of an ODD-fusion protein.

Furthermore, to examine the importance of the NLS for ODD regulation, we constructed pCH/557–574 Δ NLS and pCH/3-0 Δ NLS, which had the NLS sequence deleted from their vector constructs (Fig. 1,A), and examined the stability of fusion proteins with β-Gal assay after culturing their transfectants in the presence or absence of Cbz-LLL. The ODD regulation was significantly decreased in the NLS-less fusion proteins (Fig. 2 B), confirming the importance of the nuclear localization of proteins for the ubiquitin/proteasome-mediated degradation through the ODD domain of HIF-1α. For additional study, we used the fusion proteins containing NLS and HIF-1α residues 548–603 (ODD/3-0) or ODD557–574.

Construction of TAT-ODD Fusion Protein and Its Oxygen-dependent Instability in Vitro.

To apply the ODD-fusion proteins to in vivo study, we fused TAT47–57 to the NH2 terminal of ODD-β-Gal and created a TAT-ODD-β-Gal triple fusion protein; proteins without TAT transmembrane sequences are not delivered to any tissues (24). We first examined its ODD regulation in vitro. When the control TAT-β-Gal fusion protein was added to the culture medium of A549 cells, 100% of cells were stained by X-Gal, irrespective of the oxygen condition (Fig. 2,C, a and c), indicating that TAT-β-Gal fusion protein was efficiently delivered and stably localized in the cells. The addition of 10 times more β-Gal protein alone to the culture medium resulted in no staining of the cells (data not shown). When the optimal amount of TAT-ODD/3-0-β-Gal was added to the culture medium of A549 cells, the cells were stained significantly milder with X-Gal under normoxic condition than the cells under hypoxic condition (Fig. 2,C, b and d). Similar results were obtained by the experiment using TAT-ODD557–574-β-Gal (data not shown). However, the degradation of the fusion protein under normoxic condition (20% O2) or in the absence of Cbz-LLL was less efficient in the experiment using the pCH/557–574 plasmid (Fig. 2, A and B). Therefore, we used TAT-ODD/3-0-β-Gal for in vivo study.

Selective Stabilization of TAT-ODD-β-Gal in the Hypoxic Regions of the Solid Tumor.

Next, we applied TAT-ODD/3-0-β-Gal to in vivo study to examine if it was delivered to tumors and specifically stabilized in the hypoxic regions of the tumors. Mice bearing tumors resulting from xenografts of a human pancreatic cancer cell line CF/PAC-1 were i.p. injected with a control protein, TAT-β-Gal, and the tumors were subsequently analyzed for the delivery and stability of the control protein with X-Gal staining and anti-β-Gal Ab. Entire tumors were stained with both X-Gal and anti-β-Gal Ab, indicating that bioactive β-Gal was delivered efficiently (Fig. 3,A, a and c). However, when TAT-ODD/3-0-β-Gal was used for the same experiment, only parts of tumors were stained with X-Gal and anti-β-Gal Ab (Fig. 3,A, b and d). To evaluate if these areas in the tumor corresponded to the hypoxic regions, the tumor-bearing mice were injected i.p. with Pimo, a hypoxia marker (37), after the injection of the fusion proteins, and their tumors were stained with anti-Pimo and anti-β-Gal Ab. The regions detected with monoclonal anti-Pimo and rabbit anti-β-Gal Ab were indeed similar (Fig. 3 B). These data suggest that although TAT-ODD/3-0-β-Gal was delivered into the tumors, it was degraded in normoxic regions and lingered in the hypoxic tumor cells.

To examine the stability of TAT-ODD/3-0-β-Gal in normal tissue, we examined the liver of the tumor-bearing mice with X-gal staining and anti-β-Gal Ab. The entire liver from the mice injected with TAT-β-Gal stained with both X-Gal and anti-β-Gal Ab, confirming that bioactive β-Gal was delivered efficiently to the tissues (Fig. 4,A, a, c, and e). On the other hand, β-Gal activity, as well as β-Gal protein, was hardly detected in the liver from the mice injected with TAT-ODD/3-0-β-Gal (Fig. 4 A, b, d, and f). These results suggest that TAT-ODD/3-0-β-Gal was degraded efficiently in the normal tissue.

Antitumor Effect of TAT-ODD-Casp3 Fusion Protein.

Because TAT-ODD/3-0-β-Gal was stabilized specifically in the hypoxic solid tumors, a cytotoxic TAT-ODD/3-0 fusion protein should have an antitumor effect. Casp3WT and Casp3MUT (C163M), encoding a catalytically inactive Casp3 (23), were fused to the COOH terminal of TAT-ODD/3-0, and their Casp3 activities and ODD regulation were first examined in vitro. Significant numbers of cells underwent apoptosis when NIH/3T3 and A549 cells were treated with TAT-ODD/3-0-Casp3WT under hypoxic condition (Fig. 5,A and data not shown). Moreover, significant increase of Casp3 activity was observed in the cells treated with TAT-ODD/3-0-Casp3WT only in the presence of Cbz-LLL (Fig. 5,B). We then injected TAT-ODD/3-0-Casp3WT and TAT-ODD/3-0-Casp3MUT into the tumor-bearing mice. TAT-ODD/3-0-Casp3WT significantly and dose dependently reduced the tumor masses for a couple of days, whereas TAT-ODD/3-0-Casp3MUT did not influence tumor growth (Fig. 5 C). These results suggest that Casp3 was stabilized and activated in the cells under hypoxic conditions, resulting in the hypoxic tumor cell death. The application of fusion protein ≤20 mg/kg did not cause any obvious side effects, such as weight loss, nor was the appearance or behavior of the tumor-bearing mice altered during the observation period.

We have presented a novel protein therapy targeting solid tumors. To design a protein drug, we used the TAT protein transduction domain, which quickly and efficiently delivered a bioactive protein to all tissues and took advantage of the selective stabilization of the ODD-fusion protein under hypoxic conditions. The TAT-β-Gal-fusion protein was delivered successfully in vivo into all tissues of the mouse (Ref. 24; Figs. 3,A, a and c; and 4, a, c, and e), whereas TAT-ODD-β-Gal was detected in only a part of the solid tumors (Fig. 3,A, b and d). To examine if the regions where TAT-ODD-β-Gal was detected were hypoxic, we compared them with the regions detected with Ab against a hypoxia marker (anti-Pimo Ab). The regions were similar, although not identical, and anti-β-Gal Ab was detected in an apparently larger area than the regions detected with anti-Pimo Ab (Fig. 3 B). We speculated that the regions surrounding the hypoxic regions in the tumor would have a relatively low oxygen tension and would therefore have slower degradation of TAT-ODD/3-0-β-Gal.

Because the hypoxic regions with the lowest oxygen tension would be just a few percentage of the total tumor mass, removal of the hypoxic regions alone would not influence the tumor mass significantly. The administration of TAT-ODD/3-0-Casp3WT, however, not only suppressed the tumor growth but also reduced the tumor masses (Fig. 5 C), suggesting that TAT-ODD/3-0-Casp3WT must have stabilized in the hypoxic regions, as well as their outskirts, in a distribution similar to TAT-ODD/3-0-β-Gal.

When applying a cytotoxic TAT-ODD fusion protein as a protein drug, we must minimize the damage to normal tissues. To achieve this, at least three things have to be considered for additional study: (a) dosage; (b) administration; and (c) characteristics of the cytotoxic protein fused to TAT-ODD. Addressing the first point, determination of the optimal dose of the fusion protein is the crucial part of this protein therapy. Because the degradation of ODD-fusion proteins depends on the cellular ubiquitin/proteasome system, their degradation rate is restricted by the capacity of the cellular ubiquitin/proteasome system. In vitro experiments using TAT-ODD-β-Gal fusion protein indicated that the ODD regulation was evident at the optimal concentration of the ODD-fusion proteins, but at higher concentrations, the ODD regulation was hardly observed (data not shown). The fusion of an additional ODD domain to the COOH-terminal end of the protein might increase the efficiency of the ODD degradation. Addressing the second point, although we gave only one injection of the TAT-ODD fusion protein to the tumor-bearing mice, repeated administrations might be effective in reducing the tumor mass. The timing of the administration and the combination with radiotherapy and chemotherapy constitute the next phase of this study. Lastly, to minimize the side effects, the cytotoxic activity should preferably be indirect, and a certain level of accumulation of proteins should be needed to achieve cytotoxicity. We used a dormant form of Casp3, a nuclear executioner caspase, as a cytotoxic protein fused to TAT-ODD. We expected that Casp3 would be active in solid tumor cells under hypoxic conditions, because Casp3 activation has been observed under hypoxic conditions in vivo as well as in vitro (Refs. 28, 29, 30, 31 and Fig. 5,AB). It should be emphasized that although TAT-ODD-Casp3 was able to induce apoptosis in mouse cells under hypoxic condition (Fig. 5,A), 20 mg/kg TAT-ODD/3-0-Casp3WT did not cause any obvious side effects in the tumor-bearing mice, consistent with the data that β-Gal activity and protein were hardly detected in the normal tissue (Fig. 4). Although we successfully demonstrated the antitumor effect of TAT-ODD-Casp3, there might be better cytotoxic protein candidates. Therefore, it is imperative to conduct an additional study.

We determined the minimum ODD residues (557–574) that can regulate the ODD of a fusion protein. This region is also located within the minimal NH2-terminal trans-activation domain (532–585; Refs. 38 and 39), which has been identified recently as the region critical for interaction with von Hippel-Lindau to mediate the ubiqutin-proteasomal degradation of HIF-1α under normoxic conditions (16, 17, 18). The ODD residues required for the ODD regulation of HIF-1α stability (36) were different from the ones required for the regulation of ODD-fusion proteins (Fig. 1 B). Srinivas et al.(40) obtained similar results using a Gal4/hHIF-1α (530–652) fusion protein. At present, we are unable to explain these differences. The ODD regulation of HIF-1α per se might need an additional regulatory factor. Because the ODD regulation of ODD/3-0 was tighter than those of ODD557–574 and ODD548–583, there might be some regulatory element within the region 584–603. The molecular mechanism of the ODD regulation in ODD-fusion proteins is under investigation.

Fig. 1.

Identification of HIF-1α ODD regions responsible for ODD of β-Gal fusion proteins. A, schematic representation of ODD-β-Gal reporter plasmid. B, summary of the effects of different ODD regions on ODD regulation of ODD-β-Gal fusion proteins. The amino acid numbers indicated for each ODD fragment correspond to the amino acid number of HIF-1α. The ODD regulation refers to the increased β-Gal activity and increased number and thickness of blue color in X-gal-stained cells cultured with 5 μm Cbz-LLL as compared with no treatment. +*, the ODD regulations were observed, but relatively higher β-Gal activity was observed in the transfectants cultured without 5 μm Cbz-LLL.

Fig. 1.

Identification of HIF-1α ODD regions responsible for ODD of β-Gal fusion proteins. A, schematic representation of ODD-β-Gal reporter plasmid. B, summary of the effects of different ODD regions on ODD regulation of ODD-β-Gal fusion proteins. The amino acid numbers indicated for each ODD fragment correspond to the amino acid number of HIF-1α. The ODD regulation refers to the increased β-Gal activity and increased number and thickness of blue color in X-gal-stained cells cultured with 5 μm Cbz-LLL as compared with no treatment. +*, the ODD regulations were observed, but relatively higher β-Gal activity was observed in the transfectants cultured without 5 μm Cbz-LLL.

Close modal
Fig. 2.

In vitro study of ODD regulation of β-Gal fusion proteins. HEK293 cells were transfected with the plasmids indicated in the figure and treated with or without 5 μm Cbz-LLL for 24 h. pCH is the control plasmid encoding β-Gal and no ODD sequences. The cells were stained with X-gal (A) or applied for β-Gal assay (B). In pCH/3-0ΔNLS and pCH/557–574ΔNLS plasmids, NLS sequences (see Fig. 1 a) were deleted from pCH/3-0 and pCH/557-574, respectively. Relative β-Gal activity = (normalized β-Gal activity in the lysates from the cells treated with Cbz-LLL)/(normalized β-Gal activity in the lysates from the untreated cells). In C, TAT-fusion protein (1 unit of β-Gal activity/ml) was added to the culture medium of A549 cells. The cells were cultured for 18 h under normoxic (20% O2) or hypoxic (1% O2) conditions and then stained with X-gal for 2 days.

Fig. 2.

In vitro study of ODD regulation of β-Gal fusion proteins. HEK293 cells were transfected with the plasmids indicated in the figure and treated with or without 5 μm Cbz-LLL for 24 h. pCH is the control plasmid encoding β-Gal and no ODD sequences. The cells were stained with X-gal (A) or applied for β-Gal assay (B). In pCH/3-0ΔNLS and pCH/557–574ΔNLS plasmids, NLS sequences (see Fig. 1 a) were deleted from pCH/3-0 and pCH/557-574, respectively. Relative β-Gal activity = (normalized β-Gal activity in the lysates from the cells treated with Cbz-LLL)/(normalized β-Gal activity in the lysates from the untreated cells). In C, TAT-fusion protein (1 unit of β-Gal activity/ml) was added to the culture medium of A549 cells. The cells were cultured for 18 h under normoxic (20% O2) or hypoxic (1% O2) conditions and then stained with X-gal for 2 days.

Close modal
Fig. 3.

Selective stabilization of TAT-ODD-β-Gal in solid tumors. In A, the enzymatic activity and quantification of β-Gal protein in the tumor sections were assessed by X-gal staining (a and b) and anti-β-Gal Ab (c and d) after i.p. injection of TAT-β-Gal (a and c) or TAT-ODD/3-0-β-Gal (b and d) fusion proteins. In B, Pimo was i.p. injected 2 h after i.p. injection of fusion proteins. The samples were analyzed with anti-Pimo Ab (right) or anti-β-Gal Ab (left). The Ab-stained samples (Aa and Ac; B) were also stained with hematoxylin. Bar, 100 μm.

Fig. 3.

Selective stabilization of TAT-ODD-β-Gal in solid tumors. In A, the enzymatic activity and quantification of β-Gal protein in the tumor sections were assessed by X-gal staining (a and b) and anti-β-Gal Ab (c and d) after i.p. injection of TAT-β-Gal (a and c) or TAT-ODD/3-0-β-Gal (b and d) fusion proteins. In B, Pimo was i.p. injected 2 h after i.p. injection of fusion proteins. The samples were analyzed with anti-Pimo Ab (right) or anti-β-Gal Ab (left). The Ab-stained samples (Aa and Ac; B) were also stained with hematoxylin. Bar, 100 μm.

Close modal
Fig. 4.

Rapid degradation of TAT-ODD-β-Gal fusion protein in normal tissue. Analysis of β-Gal protein and enzymatic activity was assessed by anti-β-Gal Ab (a–d) and X-gal staining (e and f), respectively, in the liver. The livers were removed 4 h after i.p. injection of TAT-β-Gal (a, c, and e) or TAT-ODD/3-0-β-Gal (b, d, and f) fusion proteins to tumor-bearing mice and then stained with X-gal for 3 days. c and d, the higher magnification of a and b, respectively. The sections for X-gal staining (e and f) were not stained with hematoxylin. Bar, 100 μm.

Fig. 4.

Rapid degradation of TAT-ODD-β-Gal fusion protein in normal tissue. Analysis of β-Gal protein and enzymatic activity was assessed by anti-β-Gal Ab (a–d) and X-gal staining (e and f), respectively, in the liver. The livers were removed 4 h after i.p. injection of TAT-β-Gal (a, c, and e) or TAT-ODD/3-0-β-Gal (b, d, and f) fusion proteins to tumor-bearing mice and then stained with X-gal for 3 days. c and d, the higher magnification of a and b, respectively. The sections for X-gal staining (e and f) were not stained with hematoxylin. Bar, 100 μm.

Close modal
Fig. 5.

The antitumor effect of TAT-ODD-Casp3. In A, NIH/3T3 cells were treated with WT or Mut TAT-ODD-Casp3 proteins or the buffer and cultured for 24 h under normoxic (20% O2) or hypoxic (1% O2) conditions. In B, WT or Mut TAT-ODD-Casp3 proteins or the buffer were added to the culture of HEK293 cells and incubated for 4 h in the presence of the solvent only (dark bars) or Cbz-LLL (gray bars). The cells were harvested, and Casp3 activity in a 50-μg cell lysate was measured and shown in the figure. C, tumor-bearing mice were i.p. injected with the buffer only (0 mg/kg) or 2 or 20 mg/kg TAT-ODD/3-0-Casp3 fusion proteins (Mut or WT). Average total volume of the tumors was ∼300 mm3. The size of tumors was measured every day after injection, and the relative tumor mass compared with the mass on the injected day was indicated in the figure. Tumors were measured with calipers, and tumor volume was calculated as 0.5 × L × W2. The results are the mean of six independent tumors ± SD.

Fig. 5.

The antitumor effect of TAT-ODD-Casp3. In A, NIH/3T3 cells were treated with WT or Mut TAT-ODD-Casp3 proteins or the buffer and cultured for 24 h under normoxic (20% O2) or hypoxic (1% O2) conditions. In B, WT or Mut TAT-ODD-Casp3 proteins or the buffer were added to the culture of HEK293 cells and incubated for 4 h in the presence of the solvent only (dark bars) or Cbz-LLL (gray bars). The cells were harvested, and Casp3 activity in a 50-μg cell lysate was measured and shown in the figure. C, tumor-bearing mice were i.p. injected with the buffer only (0 mg/kg) or 2 or 20 mg/kg TAT-ODD/3-0-Casp3 fusion proteins (Mut or WT). Average total volume of the tumors was ∼300 mm3. The size of tumors was measured every day after injection, and the relative tumor mass compared with the mass on the injected day was indicated in the figure. Tumors were measured with calipers, and tumor volume was calculated as 0.5 × L × W2. The results are the mean of six independent tumors ± SD.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

2

The abbreviations used are: HIF, hypoxia-inducible factor; ODD, oxygen-dependent degradation; β-Gal, β-galactosidase; NLS, nuclear localization signal; Cbz-LLL, N-carbobenzoxyl-l-leucinyl-l-leucinyl-l-norvalinal; Pimo, Pimonidazole; Ab, antibody; WT, wild type; Casp3, Caspase-3; MUT, mutant.

We thank Drs. Thoru Shibata and Gen Kondoh for discussion; Dr. Nobtake Akiyama, Ataru Kobayashi, and Shigeki Masui for technical advice and valuable reagents; and Naoko Murakami for technical assistance.

1
Höckel M., Vaupel P. Tumor hypoxia: definitions and current clinical, biologic and molecular aspects.
J. Natl. Cancer Inst. (Bethesda)
,
93
:
266
-276,  
2001
.
2
Mueller-Klieser W. Tumor biology and experimental therapeutics.
Crit. Rev. Oncol. Hematol.
,
36
:
123
-139,  
2000
.
3
Teicher B. A. Hypoxia and drug resistance.
Cancer Metastasis Rev.
,
13
:
139
-168,  
1994
.
4
Kung A. L., Wang S., Klco J. M., Kaelin W. G., Jr., Livingston D. M. Suppression of tumor growth through disruption of hypoxia-inducible transcription.
Nat. Med.
,
6
:
1335
-1340,  
2000
.
5
Brown J. M. Exploiting the hypoxic cancer cell: mechanisms and therapeutic strategies.
Mol. Med. Today
,
6
:
157
-162,  
2000
.
6
Lemmon M. J., van Zijl P., Fox M. E., Mauchline M. L., Giaccia A. J., Minton N. P., Brown J. M. Anaerobic bacteria as a gene delivery system that is controlled by the tumor microenvironment.
Gene Ther.
,
4
:
791
-796,  
1997
.
7
Dachs G. U., Coralli C., Hart S. L., Tozer G. M. Gene delivery to hypoxic cells in vitro.
Br. J. Cancer
,
83
:
662
-667,  
2000
.
8
Koshikawa N., Takenaga K., Tagawa M., Sakiyama S. Therapeutic efficacy of the suicide gene driven by the promoter of vascular endothelial growth factor gene against hypoxic tumor cells.
Cancer Res.
,
60
:
2936
-2941,  
2000
.
9
Binley K., Iqball S., Kingsman A., Kingsman S., Naylor S. An adenoviral vector regulated by hypoxia for the treatment of ischaemic desease and cancer.
Gene Ther.
,
6
:
1721
-1727,  
1999
.
10
Ryan H. E., Lo J., Johnson R. S. HIF-1α is required for solid tumor formation and embryonic vascularization.
EMBO J.
,
17
:
3005
-3015,  
1998
.
11
Semenza G. L. Regulation of mammalian O2 homeostasis by hypoxia-inducible factor 1.
Annu. Rev. Cell Dev. Biol.
,
15
:
551
-578,  
1999
.
12
Jiang B. H., Semenza G. L., Bauer C., Marti H. H. Hypoxia-inducible factor 1 levels vary exponentially over a physiologically relevant range of O2 tension.
Am. J. Physiol.
,
271
:
C1172
-C1180,  
1996
.
13
Huang E., Arany Z., Livingston D. M., Bunn H. F. Activation of hypoxia-inducible transcription factor depends primarily upon redox-sensitive stabilization of its α subunit.
J. Biol. Chem.
,
271
:
32253
-32259,  
1996
.
14
Yu A. Y., Frid M. G., Shimoda L. A., Wiener C. M., Stenmark K., Semenza G. L. Temporal, spatial, and oxygen-regulated expression of hypoxia-inducible factor-1 in the lung.
Am. J. Physiol. Lung Cell. Mol. Physiol.
,
275
:
L818
-L826,  
1998
.
15
Huang L. E., Gu J., Schau M., Bunn H. F. Regulation of hypoxia-inducible factor 1α is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway.
Proc. Natl. Acad. Sci. USA
,
95
:
7987
-7992,  
1998
.
16
Tanimoto K., Makino Y., Pereira T., Poellinger L. Mechanism of regulation of the hypoxia-inducible factor-1α by the von Hippel-Lindau tumor suppressor protein.
EMBO J.
,
19
:
4298
-4309,  
2000
.
17
Ohh M., Park C. W., Ivan M., Hoffman M. A., Kim T. Y., Huang L. E., Pavletich N., Chau V., Kaelin W. G. Ubiquitination of hypoxia-inducible factor requires direct binding to the β-domain of the von Hippel-Lindau protein.
Nat. Cell Biol.
,
2
:
423
-427,  
2000
.
18
Maxwell P. H., Wiesener M. S., Chang G. W., Clifford S. C., Vaux E. C., Cockman M. E., Wykoff C. C., Pugh C. W., Maher E. R., Ratcliffe P. J. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis.
Nature (Lond.)
,
399
:
271
-275,  
1999
.
19
Pepinsky R. B., Androphy E. J., Corina K., Brown R., Barsoum J. Specific inhibition of a human papillomavirus E2 trans-activator by intracellular delivery of its repressor. DNA.
Cell Biol.
,
13
:
1011
-1019,  
1994
.
20
Kim D. T., Mitchell D. J., Brockstedt D. G., Fong L., Nolan G. P., Fathman C. G., Engleman E. G., Rothbard J. B. Introduction of soluble proteins into the MHC class I pathway by conjugation to an HIV tat peptide.
J. Immunol.
,
159
:
1666
-1668,  
1997
.
21
Nagahara H., Vocero-Akbani A. M., Snyder E. L., Ho A., Latham D. G., Lissy N. A., Becher-Hapak M., Ezhevsky S. A., Dowdy S. F. Transduction of full-length TAT fusion proteins into mammalian cells: TAT-p27Kip1 induces cell migration.
Nat. Med.
,
4
:
1449
-1452,  
1998
.
22
Gius D. R., Ezhevsky S. A., Becker-Hapak M., Nagahara H., Wei M. C., Dowdy S. F. Transduced p16INK4a peptides inhibit hypophosphorylation of the retinoblastoma protein and cell cycle progression prior to activation of Cdk2 complexes in late G1.
Cancer Res.
,
59
:
2577
-2580,  
1999
.
23
Vocero-Akbani A. M., Heyden N. V., Lissy N. A., Ratner L., Dowdy S. F. Killing HIV-infected cells by transduction with an HIV protease-activated caspase-3 protein.
Nat. Med.
,
5
:
29
-33,  
1999
.
24
Schwarze S. R., Ho A., Vocero-Akbani A., Dowdy S. F. In vivo protein transduction: delivery of a biologically active protein into the mouse.
Science (Wash. DC)
,
285
:
1569
-1572,  
1999
.
25
Woo M., Hakem R., Soengas M. S., Duncan G. S., Shahinian A., Kagi D., Hakem A., McCurrach M., Khoo W., Kaufman S. A, et al Essential contribution of caspase 3/CPP32 to apoptosis and its associated nuclear changes.
Genes Dev.
,
12
:
806
-819,  
1998
.
26
Enari M., Sakahira H., Yokoyama H., Okawa K., Iwamatsu A., Nagata S. A caspase-activated DNase that degrades DNA during apoptosis and its inhibitor ICAD.
Nature (Lond.)
,
391
:
43
-50,  
1998
.
27
Liu X., Li P., Widlak P., Zou H., Luo X., Garrard W. T., Wang X. The 40-kDa subunit of DNA fragmentation factor induces DNA fragmentation and chromatin condensation during apoptosis.
Proc. Natl. Acad. Sci. USA
,
95
:
8461
-8466,  
1998
.
28
de Moissac D., Gurevich R. M., Zheng H., Singal P. K., Kirshenbaum L. A. Caspase activation and mitochondrial cytochrome C release during hypoxia-mediated apoptosis of adult ventricular myocytes.
J. Mol. Cell. Cardiol.
,
32
:
53
-63,  
2000
.
29
Derrick M., He J., Brady E., Tan S. The in vitro fate of rabbit fetal brain cells after acute in vivo hypoxia.
J. Neurosci.
,
21
: (RC138)
1
-5,  
2001
.
30
Yamamoto K., Morishita R., Hayashi S., Matsushita H., Nakagami H., Moriguchi A., Matsumoto K., Nakamura T., Kaneda Y., Ogihara T. Contribution of Bcl-2, but not Bcl-xL and Bax, to antiapoptotic actions of hepatocyte growth factor in hypoxia-conditioned human endothelial cells.
Hypertension
,
37
:
1341
-1348,  
2001
.
31
Volm M., Mattern J., Koomagi R. Inverse correlation between apoptotic (Fas ligand, caspase-3) and angiogenic factors (VEGF, microvessel density) in squamous cell lung carcinomas.
Anticancer Res.
,
19
:
1669
-1671,  
1999
.
32
Chen C., Okayama H. High-efficiency transformation of mammalian cells by plasmid DNA.
Mol. Cell. Biol.
,
7
:
2745
-2752,  
1987
.
33
Raleith J. A., Chou S-C., Tables L., Suchindran S., Varia M. A. Relationship of hypoxia to metallothionein expression in murine tumors.
Int. J. Radiat. Oncol. Biol. Phys.
,
42
:
727
-730,  
1998
.
34
Akiyama N., Matsuo Y., Sai H., Noda M., Kizaka-Kondoh S. Identification of a series of transforming growth factor β-responsive genes by retrovirus-mediated gene trap screening.
Mol. Cell. Biol.
,
20
:
3266
-3273,  
2000
.
35
Rock K. L., Gramm C., Rothstein L., Clark K., Stein R., Dick L., Hwang D., Goldberg A. L. Inhibitors of the proteasome block the degradation of most cell proteins and the generation of peptides presented on MHC class I molecules.
Cell
,
78
:
761
-771,  
1994
.
36
Sutter C. H., Laughner E., Semenza G. L. Hypoxia-inducible factor 1α protein expression is controlled by oxygen-regulated ubiquitination that is disrupted by deletions and missense mutations.
Proc. Natl. Acad. Sci. USA
,
97
:
4748
-4753,  
2000
.
37
Varia M. A., Calkins-Adams D. P., Rinker L. H., Kennedy A. S., Novotny D. B., Fowler W. C., Jr., Raleigh J. A. Pimonidazole: a novel hypoxia marker for complementary study of tumor hypoxia and cell proliferation in cervical carcinoma.
Gynecol. Oncol.
,
71
:
270
-277,  
1998
.
38
Jiang B. H., Zheng J. Z., Leung S. W., Roe R., Semenza G. L. Transactivation and inhibitory domains of hypoxia-inducible factor 1α. Modulation of transcriptional activity by oxygen tension.
J. Biol. Chem.
,
272
:
19253
-19260,  
1997
.
39
Pugh C. W., O’Rourke J. F., Nagao M., Gleadle J. M., Ratcliffe P. J. Activation of hypoxia-inducible factor-1; definition of regulatory domains within the α subunit.
J. Biol. Chem.
,
272
:
11205
-11214,  
1997
.
40
Srinivas V., Zhang L., Zhu X., Caro J. Characterization of an oxygen/redox-dependent degradation domain of Hypoxia-Inducible Factor α (HIF-α) proteins.
Biochem. Biophys. Res. Commun.
,
260
:
557
-561,  
1999
.