Since HCCR-1 has been shown to have tumorigenic features previously 1314 we focused on the biological functions of HCCR-1 in the cell. In order to determine the functional domain of HCCR-1 at the molecular level, Pfam version 18.0 was used to analyze its amino acid sequences and identified with the LETM1 homologous domain at amino acids 75 to 346 of HCCR-1 (Fig 1A). Sequence alignment between the domain from HCCR-1 and from the LETM1 protein using CLUSTALW 15 showed a 57.61% sequence similarity. From the analysis using Pfam, we also identified several LETM1 domain-containing proteins, in various eukaryotic species, from human through fungi as well as plant (Fig 1B); these findings suggest that the domain has widespread application in organisms and is well conserved during evolution, and that the LETM1 domain may have an important biological function in the normal life cycle.

Since some of the LETM1 domain-containing proteins have been identified as mitochondrial proteins 8, we analyzed the subcellular localization of the LETM1 domain-containing proteins using the TargetP version 1.1 16 and Predotar version 1.03 17. As shown at Table 1, almost all proteins were predicted to be mitochondrial proteins with a high score (Table 1). It strongly suggests that LETM1 domain may play an important role in mitochondria.

In order to examine the subcellular localization of HCCR-1, cDNA was subcloned into a pEGFP-N1 vector. COS-7 and MCF-7 cells were cultured on a pre-coated coverslip and transiently transfected with the pEGFP-N1 vector (Fig 2A, GFP) or the vector carrying the human HCCR-1 cDNA (Fig 2A, HCCR-1/GFP). The cells were then further incubated for 24–28 h and their mitochondria were stained with MitoTracker for 15 min at 37°C. The cells were fixed, mounted, and analyzed by confocal microscopy (Fig 2A, top through forth panels). We also used the HEK293 cell line which stably expresses the V5 tagged HCCR-1 (Fig 2A, HEK293/HCCR-1-V5). The cultured cells on the pre-coated coverslip were stained with MitoTracker, fixed with 100% methanol for 1 min at -20°C. Mouse anti-V5 antibody and goat anti-mouse IgG antibody conjugated with horseradish peroxidase (HRP) were applied to the cells for 1 h at room temperature. The stained cells were mounted and analyzed by confocal microscopy (Fig 2A, Bottom panel). The HCCR-1 fluorescence images revealed a staining pattern resembling mitochondria in all tested cell lines (Fig 2A, HCCR-1/GFP and HCCR-1-V5). The visualized mitochondria, as a result of incubation with a mitochondria-specific dye (Fig 2A, MitoTracker), were overlaid with HCCR-1/GFP or HCCR-1-V5 (Fig 2A, merge) while the vector alone (Fig 2A, GFP) did not merge with the mitochondria; these findings suggest that HCCR-1 is localized at the mitochondria.

Sequence analysis of the HCCR-1 protein, using the Hopp-Woods method of calculating hydrophilicity and TMHMM Server Version 2.0 18, revealed the presence of two hydrophobic segments, one extending from 139 to 161 residues which was predicted a transmembrane domain (TM) and the other extending from position 81 to 98 amino acids which may be involved in anchoring to a membranous component. Therefore, HCCR-1 is predicted to be an integral membrane protein.

To examine whether HCCR-1 is a mitochondrial membrane protein, the HEK293/HCCR-1-V5 cell lines were studied for subcelllular fractionation experiments. Nuclear (Fig 2B, lane1), post-mitochondrial supernatant (Fig 2B, lane2), and mitochondrial (Fig 2B, lane3) fractions were separated by differential centrifugation. The isolated mitochondria were solubilized by treatment with alkaline sodium carbonate 19 and ultra-centrifuged to separate the soluble proteins (Fig 2B, lane 5) from the insoluble integral membrane proteins (Fig 2B, lane 4). Each of the fractionated proteins were displayed by SDS-PAGE and analyzed by immunoblotting using antibodies to the V5 epitope (Fig 2B, top panel), voltage dependant anion channel 1 (VDAC-1) which is expressed on the mitochondrial membrane 20 (Fig 2B, middle panel) and superoxide dismutase-2 (SOD-2) on the mitochondrial matrix 21 (Fig 2B, bottom panel). The HCCR-1 protein was found in the mitochondrial fraction (Fig 2B, lane3) which is consistent with the previous result shown in figure 2A. Treatment with alkaline sodium carbonate did not solubilize the mitochondrial integral membrane protein VDAC-1 (Fig 2B, middle panel, lane 4), but did disrupt the mitochondrial membrane as a result of exposing the matrix protein SOD-2 (Fig 2B, bottom panel, lane 5). HCCR-1 protein also remained in the pellet fraction (Fig 2B, top panel, lane 4); these findings suggest that HCCR-1 is exclusively localized in the mitochondrial membrane.

Since HCCR-1 was localized to the mitochondrial membrane, we evaluated the topological structure of HCCR-1. Bioinformatic analyses using the TMHMM Server Version 2.0 18, showed that the NH₂-terminus of HCCR-1 is exposed to the cytoplasm, and also predicted that HCCR-1 is a type II integral membrane protein.

To examine the HCCR-1 topology, the intact mitochondria were isolated from HEK293/HCCR-V5 cell lines and incubated with and without proteinase K or proteinase K plus Triton X-100. Each treatment was subjected to SDS-PAGE and visualized by immunoblotting using antibodies to the V5 epitope (Fig 2C, top panel), TOM40 which expresses on the outer mitochondrial membrane 45 (Fig 2C, second panel), TIM23 on the inner mitochondrial membrane 67 (Fig 2C, third panel), and SOD-2 on the mitochondrial matrix (Fig 2C, bottom panel). TIM23 and SOD-2 proteins were protease resistant in the mitochondria, which indicate that these proteins were not exposed to the surface. By contrast, Tom40, an outer membrane protein protruding to the surface, was sensitive to proteinase K and weakly detected by its polyclonal antibody (Fig 2C, lane 2). Interestingly, the immunoblot analysis, using anti-V5 monoclonal antibody, detected two sized proteins – a full sized HCCR-1-V5 (about 39 kDa) and a little smaller sized protein (about 32 kDa); this indicates that the NH₂-terminus of HCCR-1 was partially degraded by the proteinase K, while the COOH-terminus of HCCR-1, containing V5 epitope, was protected from the protease activity in the intact mitochondria. This suggests that HCCR-1 is located on the outer mitochondrial membrane (Fig 2C). This result also suggests that the topological structure of HCCR-1 is indeed consistent with the bioinformatic prediction as shown in figure 2D. Upon disintegration of the membrane proteins by treatment with Triton X-100, all proteins were digested by proteinase K, showing that none of them was intrinsically protease-resistant (Fig 2C, lane 3).

The targeting signal, on mitochondrial proteins, often exists on their DNA fragments coding for the amino terminal region of the protein 222324 and does not have a conserved sequence, but rather share a common physical property. This feature is the ability to form an amphipathic α-helix, one in which positively charged amino acids are localized to one surface of the helix 252627. This signal sequence is recognized by the receptors on the mitochondrial outer membrane and imported into mitochondria. We have shown that HCCR-1 is also a mitochondrial protein. This protein may be posttranslationally transported to mitochondria and then imported into mitochondria by recognition of the signal sequence.

To determine the mitochondrial targeting signal of HCCR-1, three different sizes of the NH₂-teminal DNA fragments were subcloned into a pEGFP-N1 expression vector (Fig 3A). Each of the constructs was transfected into COS-7 cells and the mitochondria were stained and examined by confocal microscopy (Fig 3B). HCCR D1-1, which contains NH₂-terminal segments at positions 1 to 78 residues, was localized diffusely in the cytoplasm and nuclei (Fig 3B, third panel). This distribution is more like that of GFP alone (Fig 3B, top panel). By contrast, the deletion mutants, HCCR D1-2 and D1-3, which contained at position 1 to 110 and 1 to 173 respectively, exhibited a typical mitochondrial pattern, and were overlaid with their mitochondrial staining (Fig 3B, forth and bottom panels); this suggests that the D1-2 region which partially contains LETM1 domain, contains the mitochondrial targeting sequence and D1-1 region alone cannot reach the mitochondria. We also tested these constructs in NIH3T3 and MCF-7 cells, and obtained the same results as in COS-7 cells (Data not shown).

The 2^nd structural analysis of the D1-2 region, using GOR IV secondary structure prediction method 28, identified an α-helical domain from 92 to 110 which belongs to LETM1 domain of HCCR-1 and this domain represented an amphipathic α-helical feature which often plays a role in the mitochondrial targeting signal 252627. To further characterize the HCCR-1 targeting signal, the positions 1 to 56 were deleted from HCCR D1-1, -2 and -3 constructs (Fig 4A). Each construct was transfected into COS-7 cells and examined by confocal microscopy (Fig 4B). The HCCR D2-1 construct exhibited a diffuse pattern in the cytoplasm and nuclei resembling that of GFP alone in COS-7 cells (Fig 4B, top panel). The HCCR D2-2 construct contains an amphipathic α-helical domain. This construct also showed a diffuse distribution in cytoplasm and nuclei (Fig 4B, middle panel), which is different from the pattern of HCCR D1-2, and suggests that the NH₂-terminal region from 1 to 56 is also required for mitochondrial targeting. However, some of the D2-2 mutant was transported to mitochondria (Fig 4B, middle panel), which suggests that this region between 57 and 110 as well as the region from 1 to 56 is involved in the mitochondrial targeting. When the HCCR D2-3 construct was expressed in COS-7 cells, it was not detected in nuclei but in cytoplasm and other organelles (Fig 4B, bottom panel). The D2-3 mutant contains the transmembrane domain which is moderately hydrophobic 18. This may result in absence in nuclei and reaching into the mitochondria or other organelles 2930.

Therefore, the mitochondrial localization of HCCR-1 requires its D1-2 region, which partially belongs to the LETM1 domain of HCCR-1. By contrast, D1-1 which contains NH₂-terminus of HCCR-1 and D2-2 which belongs the LETM1 domain of HCCR-1, were prevented from localizing to mitochondria. This suggests that the D1-1 region may be co-related with the D2-2 region for mitochondrial targeting.

In order to confirm the previous results obtained by confocal microscopy experiments, COS-7 cells expressing HCCR D1-2 and D2-2 were subjected to subcellular fractionation. Nuclear (Fig 5A and 5B, lane 1), post-mitochondrial supernatant (Fig 5A and 5B, lane2), and mitochondrial (Fig 5A and 5B, lane3) fractions were separated by differential centrifugation, and analyzed by Western blotting using antibodies to the GFP (Fig 5A and 5B, top panel) and the mitochondrial membrane protein VDAC-1 (Fig 5A and 5B, bottom panel). As expected, the deleted mutant D1-2 was exclusively localized to the mitochondrial fraction (Fig 5A), while the D2-2 exhibited all fractions (Fig 5B). These data confirmed that the D1-2 domain is essential for the mitochondrial targeting of HCCR-1.

At the figure 4B and 5B, most of D2-2 and D2-3 mutants are located on mitochondria although some of them are failed to reach to mitochondria. This suggests that HCCR-1 cytosolic form such as HCCR-2 which is deleted at position 1 to 56 amino acids, may exist. This cytosolic form could act as a negative regulator of p53 or could not. However, we propose that this form may not affect the HCCR function because it may also mainly localize at mitochondria.

To verify the role of HCCR-1 on mitochondrial outer membrane and p53 negative regulation role, we used the HEK293 cell line which stably expresses the vector alone (HEK293/Vector) and V5 tagged HCCR-1 (HEK293/HCCR-1), respectively. The cells were stimulated with the 40 J/m² of the short-wave length light (UVC) ray, 100 nM staurosporine or 300 nM actinomycin D for trigging apoptosis. After the stimulation, the cells were stained with propidium iodide and analyzed by FACS. The cell number of apoptotic sub G₀/G₁-phase in the HEK293/HCCR-1 cells after treatment with UVC or staurosporine was less increased in comparing with that in HEK293 and HEK293/Vector cells, in respectively (Fig 6). The cell number of apoptotic phase in the HEK293/HCCR-1 after treatment with actinomycin D was not significantly changed at all tested cell line (Fig 6). Our flow cytometric data indicate that UVC or staurosporine induced-nuclear apoptosis is retarded in HEK293/HCCR-1 cells, which suggest that HCCR-1 induces HEK293 cells more resistance to UVC-ray and staurosporine-triggered apoptosis, but not in actinomycin D.

We have shown that overexpression of the cDNA encoding HCCR-1 led to the detection of the protein in mitochondira compartment in COS-7, MCF-7 and HEK-293 cells (Fig 2A). However, the previous immunofluorescence study showed that HCCR was predominantly localized in the plasma membrane and cytoplasm of hepatocellular carcinoma 31. Interestingly, our resent analysis of HCCR mRNA using reverse transcription-PCR and sequencing has identified several HCCR isoforms from various cancerous cell lines and tissues (data not shown). The expression of these isoforms also depended on the cell type and tissues, while some of the isoforms were predicted to be different subcellular organelles. This could be invoked, for example, to account for the generation of the other cellular compartment pool of HCCR.

HCCR-1 has been isolated and identified as a strong oncoprotein with tumorigenic features 1314. Interestingly, this protein has been localized to the mitochondrial membrane which has been deeply related to the intrinsic apoptotic signal 132. Here, we showed that the oncoprotein HCCR-1 represses mitochondrial dysfunction caused by its uncontrolled expression. This may prevent the apoptotic signal, block the senescence, and induce cancer development. Therefore, the present data now make it possible to investigate mitochondrial function and mitochondria-mediated apoptosis in HCCR-1 overexpressed cells.

Human embryonic kidney (HEK) 293 (ATCC CRL-1573), COS-7 (ATCC CRL-1651) and MCF-7 breast cancer (ATCC HTB-22) cells were obtained from the American Type Culture Collection (Manassas, VA). HEK293/HCCR-1-V5 cell line 13 which stably expressed V5 tagged HCCR-1 in HEK293 cells, was maintained in DMEM (Gibco, Grand Island, NY) containing 200 μg/ml G418, 10% FBS (Hyclone, Logan, Ut) and 1% PenStrep (Gibco, Grand Island, NY). COS-7 cell line was maintained in RPMI (Gibco, Grand Island, NY) supplemented with 10% FBS and 1% PenStrep. All cells were incubated at 37°C, 5% CO₂.

HCCR-1, and its deletion mutants, were amplified using a Peltier Thermal Cycler PTC-200 (MJ Research, MA) with the primers DL1-F (5'-CCG

CTCGAG

CTCGAG

For the DNA transfection, all the recombinant plasmids were prepared using Qiagen columns (Qiagen Inc, Valencia, CA) and DNA transfection was performed according to the manufacturer's instructions using LipofectAMINE 2000 reagent (Invitrogen, Carlsbad, CA). Plasmid DNA (0.5 μg) was transfected to COS-7 cells grown on a 6 well dish and the cells were incubated for 24 -28 h before analysis.

Subcellular fractionation of HEK293/HCCR-1-V5 and COS-7 cells was carried out according to the mitochondria isolation kit (PIERCE, Rockford, IL). Briefly, cells were harvested by centrifuging at 850 g for 2 min and the pellet was suspended with 800 μl of reagent A, and then incubated exactly 2 min on ice. Next 10 μl of Reagent B was added to the suspended solution and incubated for 5 min on ice with vortexing at maximum speed every minute. Then 800 μl of Reagent C was added to the solution and inverted the tube several times to mix. The solution was centrifuged at 700 g for 10 min at 4°C, and the pellet was used for crude nucleic fraction. The supernatant was continuously centrifuged at 12,000 g for 15 min at 4°C and transferred to a new tube for the post-mitochondrial supernatant fraction. The pellet was washed with 500 μl of Regent C, and used for isolation of the mitochondrial fraction.

For the isolation of the mitochondrial membrane fraction, the isolated mitochondria were resuspended in 200 μl of freshly prepared 100 mM sodium carbonate pH 11.5 and incubated on ice for 30 min, and centrifuged at 100,000 g for 30 min at 4°C. The supernatant was used as a mitochondrial matrix fraction. The pellet was resuspended with the same volume (200 μl) of 1X SDS loading buffer and used as a mitochondrial membrane fraction.

Intact mitochondria were isolated from cultured HEK293/HCCR-1-V5 cells and diluted with MSB buffer (150 mM Potassium acetate, 5 mM Magnesium acetate, 50 mM HEPES pH 7.6, 200 mM Sucrose, 1 mM DTT) and treated with or without 50 μg/ml of proteinase K or proteinase K plus 1% Triton X-100 at room temperature for 30 min. Stop proteolysis was performed by adding 15% (w/v) trichloroacetic acid (TCA, Sigma-Aldrich Corporation, MO) and incubating for 15 min on ice. TCA pellets were collected by microcentrifuging for 5 min at 8,000 g, at room temperature. The supernatants were carefully removed and the pellets were suspended in 30 μl of 1X SDS sample buffer by vortex mixing and subjected to Western blot analysis.

Each fraction was quantified with the BCA Protein Assay Reagent kit (PIERCE, Rockford, IL). Samples in 1X SDS sample buffer were subjected to electrophoresis in a SDS-PAGE and transferred to nitrocellulose by standard procedures. Membranes were blocked with 5% nonfat milk and 0.05% Tween 20 in TBS (Tris-buffered saline, pH8.0) for 1 h at room temperature. After washing with 0.05% Tween 20 in TBS, the membranes were incubated with primary antibodies, anti-V5 (Invitrogen, Carlsbad, CA), anti-GFP (Santa Cruz Biotechnology, Santa Cruz, CA), anti-TOM40 (Santa Cruz Biotechnology, Santa Cruz, CA), anti-TIM23 (Santa Cruz Biotechnology, Santa Cruz, CA), anti-SOD-2 (Santa Cruz Biotechnology, Santa Cruz, CA), or anti-VDAC1 (Santa Cruz Biotechnology, Santa Cruz, CA) antibodies. After washing with 0.05% Tween 20 in TBS, immunoblots were incubated with horseradish-peroxidase conjugated rabbit anti-mouse IgG (Santa Cruz Biotechnology, Santa Cruz, CA) or goat anti-rabbit IgG (KPL, Gaithersburg, MD), and proteins were visualized using enhanced chemiluminescence according to the manufacturer's instructions (PIERCE, Rockford, IL).

Coverslips were washed with HCl, three times with distilled water and twice with 100% Ethanol. Dried coverslips were then coated with 5 μg/ml Poly-L-Lysine (Sigma-Aldrich Corporation, MO) for overnight at 37°C. COS-7 and HEK293/HCCR-1-V5 cells were seeded onto precoated coverslips in 6 well plates. Then following one day incubation, the cells were transiently transfected using LipofectAMINE 2000 (Invitrogen, Carlsbad, CA). After 24–28 h further incubation, 100 nM MitoTracker (Molecular Probes, Eugene, OR) was added to the medium and incubated for 15 min at the same culture condition and washed three times with 2 ml PBS before fixation. The cells on the coverslips were then fixed with 4% paraformaldyhyde containing 4% sucrose at 4°C for 10 min. To stain V5 tagged HCCR-1, HEK293/HCCR-1-V5 cells were fixed with 100% methanol at -20°C for 1 min and washed three times with PBS and incubated with mouse anti-V5 antibody (Invitrogen, Carlsbad, CA) at a dilution of 1:300 for 1 h at room temperature. They were washed with PBS, Alexa Fluor 488 goat anti-mouse IgG antibody (Molecular Probes, Eugene, OR) at a dilution of 5 μg/ml was applied for 1 h at room temperature. The cells were washed three times with PBS and mounted using ProLong Gold Antifade Reagents (Molecular Probes, Eugene, OR). Fluorescent images were analyzed and taken using a Bio-Rad MRC-1024MP laser scanning confocal microscope (Bio-Rad, Hercules, CA).

The cells were stimulated with the 40 J/m² of the short-wave length light (UVC) ray and incubated then for 2 days, or with 100 nM staurosporine and 300 nM actinomycin D for overnight, in respectively. After the stimulation, cells were washed two times with ice-cold PBS and stored at -20°C with 5 ml of 70% Ethanol for a day. For the Propidium iodide staining, cells were resuspended and collected by centrifugation at 4°C and washed two times with PBS and then resuspended with 0.5 ml of PBS. 10 μl of RNase (100 μg/ml) were added and the cells were incubated at 37°C for 20 min. After removing RNAs, 1 μl of Propidum iode (50 μg/ml) were added and incubated at 37°C for 10 min for staining. And then analyzed on a BRYTE HS flow cytometer (Bio-rad, Herts, England).