Introduction

RB94 is produced by translation of the wild-type retinoblastoma (RB) gene (RB110) from the second in-frame AUG codon and lacks the N-terminal 112 amino acids present in RB110.¹ We have shown that RB94 is significantly more cytotoxic than wild-type RB110 to all tumor cell types studied to date, including bladder cancer cell lines, and it kills human cancer cells whatever the genetic defects may be in the malignant cells.^{1, 2} However, normal human cells appear to have no or limited susceptibility to RB94 toxicity although RB94 is cytotoxic to genetically altered transformed human cells such as those immortalized with E6 and E7.^{1, 2, 3} Differences between RB94 and RB110 include the fact that RB94 has a longer half-life than RB110, being in an active hypophosphorylated form for longer periods of time^{1, 2} and causes rapid telomere erosion and chromosomal crisis leading to what was initially thought to be caspase-dependent apoptosis.³

As systemic delivery of RB94 will soon be used in a phase 1 gene therapy study⁴ we wished to examine the RB94-induced changes and the time course of their appearance more closely and determine if even earlier RB94-specific changes might be identified. This in turn might lead to a better understanding of the basis for the unique cytotoxicity observed in cancer cells produced by RB94 without showing evidence to date of any tumor resistance. The same RB-negative UC14 and RB-positive UC9 bladder cancer cells were used in this study as were previously utilized as was the adenoviral RB94 construct.³ RB94 expression in this adenovirus is under tetracycline control and can be blocked by doxycycline, which allows RB94-specific changes to be identified. Ad-RB110 also was used as a positive control to show that the results were RB94 specific and also not adenoviral related.

Materials and methods

Cell lines, cell culture and photography

The bladder cancer cell line, UC14, which lacks both the RB-1 and p16 gene and UC9, which is RB positive and p16 negative, were used in this study.⁵ These cells were maintained cultured in minimal essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), penicillin and streptomycin. Normal human urothelial cells (NHUCs) were also utilized which had been immortalized by human telomerase reverse transcriptase (hTERT; TERT-NHUCs).⁶ They otherwise appear to be entirely normal and were grown as previously described.⁶ Cells were cultured on glass coverslips for the immunochemical studies. For phase microscopic studies they were grown in six-well cell culture dishes and an inverted phase-contrast microscope (Olympus CKX31) was used to observe cell changes. Using a conventional digital camera and adaptor, the cell status for each time point and treatment was recorded at 100, 200 and 400 magnifications.

Time points and test groups

The time points focused upon were 24 and 48 h after treatment, although some experiments were maintained for 72 and 96 h and in others earlier time points were recorded. Cells were transduced with 100 multiplicity of infection (MOI) of either Ad-RB94 or Ad-RB110.

AdRB94 plus doxycycline (1 g ml^-1) was also used to prevent RB expression as a viral control, as Ad-RB94 is under control of tetracycline regulation. The sources of the viruses and the transduction procedure have previously been described.³

RB94 and TUNEL staining

The monoclonal anti-RB antibody (QED Biosciences, San Diego, CA) and the Oncor Apoptotag kit (Chemicon International, Temecula, CA) were used for RB94 and TUNEL staining, respectively. RB94 immunochemical staining was carried out as described ² and the TUNEL assay was performed according to the manufacturer's instructions.

Confocal studies

The apoptosis procedure was carried out according to the protocol from the manufacture (Chemicon International, Billerica, MA). Briefly, UC14 cells grown on coverslips were fixed with 1% paraformaldehyde in phosphate-buffered saline (PBS), pH 7.4. After washing three times with PBS, cells were post-fixed in precooled methanol:acetic acid 2:1 for 5 min at 20 °C and again washed two times with PBS. The cells were then incubated with equilibration buffer for 10 s at room temperature, and followed by 1 h incubation with TdT enzyme at 37 °C. The enzymatic reaction was stopped by 10 min incubation with stop buffer at room temperature. The working strength anti-digoxigenin conjugate and monoclonal anti-RB antibody (1:1000) were applied to the cells and incubated at 4 °C overnight. After washing three times with PBS, goat anti-mouse Alexa 594 antibody was added to the coverslips followed by incubation at room temperature for an hour. Finally, the slides were mounted with mounting medium (DAKO, Carpinteria, CA) containing 0.5 g ml^-1 4,6-diamidino-2-phenylindole (DAPI). The staining status was evaluated using an Olympus IS71 with FV500 confocal laser microscope.

Caspase assay

Western blotting was carried out to measure caspase 3 and 9 cleavage. Both antibodies were purchased from Cell Signaling (Danvers, MA). Briefly, cultured cells were lysed in cold lysis buffer (1% Triton X-100, 1 mM EDTA, 150 mM NaCl, 50 mM Tris-HCl, 0.2 mM Na₂VO₄, 10 mg ml^-1 each of leupeptin, phenylmethylsulfonyl fluoride, and apotine; Roche Molecular Biochemical, Indianapolis, IN). A total of 50 g of each protein sample was loaded in each lane. Bound antibody was detected using the enhanced chemiluminescence detection kit (Pierce Biotechnology Inc., Rockford, IL).

Cytochrome c release analysis by western blot

Cells were trypsinized and resuspended in 100 l cytosolic lysis buffer containing 25 mM Tris, 5 mM MgCl₂ and complete mini proteinase inhibitors (Roche, Mannheim, Germany) at pH 7.4. The cells were set on ice for 5 min before spinning at 14 000 rpm for 5 min. The supernatant (cytosolic fraction) was transferred to a new centrifuge tube and stored at -20 °C. To the pellet above, 100 l lysis buffer containing 1% Triton X-100, 150 mM NaCl, 25 mM Tris, and complete mini proteinase inhibitors at pH 7.4 was added. After being lysed on a rotator for 30 min at 4 °C, it was then spun at 14 000 rpm for 5 min. The supernatant (mitochondrial fraction) was transferred to a new centrifuge tube and stored at -20 °C. Western blot was carried out with anti-cytochrome c monoclonal Ab from BD Biosciences Pharmingen (San Diego, CA) at concentration of 1:2000.

MTT assay

The various cells were seeded into 24-well plates at 4 10⁴ cells per well before infection. The cells were infected for 2.5 h as described above, and at different time points the medium was removed, and 200 l of medium were added containing 1 mg ml^-1 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). After 3 h, the reaction was stopped with 200 l of N,N-dimethylformamide lysis buffer, and the resultant solution read at A595 nm with a microreader. All time points were recorded in triplicate. Statistical analysis was carried out using the General Linear Models of the Statistica software (StatSoft Inc., Tulsa, OK).

Analysis of DNA fragmentation by field-inversion gel electrophoresis

The formation of 50 kb DNA fragments was monitored by field-inversion gel electrophoresis (FIGE) as described previously.⁷ Cells were harvested by exposure to trypsin, resuspended in 100 l PBS and mixed with 100 l of a buffer containing 150 mM NaCl, 10 mM Tris, pH 7.5, 10 mM EDTA and 2% low melting point agarose. The mixtures were transferred to the wells of a FIGE plug mold and incubated for 30 min at 4 °C. The plugs were ejected and incubated for 16 h at 37 °C in a lysis buffer containing 50 mM Tris, pH 8.0, 10 mM EDTA and 0.2 mg ml^-1 proteinase K (Promega Inc., Madison, WI). The plugs were washed 3 for 1 h at 4 °C in a buffer containing 10 mM Tris, pH 8.0, 1 mM EDTA and inserted into the wells of a 1% agarose gel. DNA fragments were then resolved by FIGE (FIGE Mapper, Biorad Inc., Hercules, CA) in 0.5 Tris-borate-EDTA buffer at a forward voltage of 180 V and a reverse voltage of 120 V for 16 h using a switch time ramp of 0.1–3.5 s (Program no. 4 provided by the manufacturer). Gels were then stained with ethidium bromide and images were obtained using a gel documentation system (Fotodyne Inc., Hartland, WI).

Studies on apoptosis-inducing factor

The siRNAs targeting apoptosis-inducing factor (AIF) have been described.⁸ The AIF siRNA was kindly provided by Dr Katsumi Kitagawa at St Jude Children's Research Hospital, Memphis, TN. 2 10⁵ UC14 cells were seeded in six-well plate 1 day before the transfection. On the day of transfection 10 l of AIF siRNA (forward sequence: CUUGUUCCAGCGAUGGCAU, reverse sequence: AUGCCAUCGCUGGAACAAG) was added in 50 l Opti-MEMI serum-free medium (Invitrogen, Carlsbad, CA) and the mix incubated for 5 min at room temperature. In another tube 0.75 l Lipofectamin (Invitrogen) was mixed in 50 l serum-free medium for 5 min at room temperature. The contents of each was then added together, mixed and incubated for 15 min. Meanwhile, 500 l of serum-free medium was added to each well after washing once with PBS. The mixture of AIF siRNA and lipofectamin was then added to each well. The cells were incubated for 4 h at 37 °C followed by medium change in MEM+10% FBS. After 72 h incubation, the cells were split and treated with Ad-RB94 as described above for subsequent experiments. The siRNA knockdown of AIF was confirmed by western blotting.

Results

Changes at the phase microscopic level

Following Ad-RB94 treatment various morphological changes were seen that were not found in either Ad-RB94 plus doxycycline-treated or Ad-RB110-transduced UC14 or UC9 cells. These were especially evident by 36–48 h after treatment. They included cellular enlargement, micronucleation (seen in approximately 10% of the cells) and peripheral nuclear chromatin condensation (seen in approximately 30–40% of the cells; Figure 1b). The peripheral nuclear condensation and blebbing was also readily seen by DAPI staining (Figure 1b, insert). In addition a significant number of cells began to detach into the tissue culture medium by 48 h after treatment (Figure 1b). None of the changes were seen in Ad-RB110- or Ad-RB94+doxycycline-treated cells.

Figure 1.

Morphological microscopic changes in RB-negative UC14 cells following Ad-RB94 transfection. (a) Normal morphology in untreated cells. (b) Typical morphological changes seen 48 h after Ad-RB94 treatment of UC14 cells. These included micronucleation (red arrows) seen in approximately 10% of the cells and approximately 30% of the cells with enlarged nuclei showing a peripheral nuclear condensed ring-like appearance (black arrows). Note the round floating cells also seen, which represent RB94-induced cells detaching from the dish (green arrows). Original magnification 400. The insert in the lower right shows 4,6-diamidino-2-phenylindole (DAPI) staining of a cell with the peripheral nuclear condensation noted in phase contrast by black arrows.

Full figure and legend (116K)

RB94 and TUNEL staining

The RB-negative UC14 cells were stained for RB94 following Ad-RB94 treatment to evaluate the transduction frequency or stained to determine the number of TUNEL-positive cells present. The earliest time that the UC14 cells showed significant RB94 staining was at 24 h post-transduction but these cells were TUNEL negative. The RB94-positive cells became TUNEL positive 36 h after treatment. TUNEL staining was seen primarily in cells with peripheral nuclear chromatin condensation but was also present in some micronucleated cells. At this time point as well as 48 h after treatment the percentages of RB-positive and TUNEL-positive cells that remained attached were almost equal being between 40 and 50% of the cells (Figures 2a and b, respectively). Confocal analysis of cells stained for both TUNEL and RB confirmed that TUNEL-positive cells were also RB94 positive (Figure 2c).

Figure 2.

RB94 and TdT-mediated dUTP nick end labeling (TUNEL) staining after Ad-RB94 treatment. (a) RB94 staining (brown nuclear staining) was seen in approximately 40–50% of the initially RB-negative UC14 cells 48 h after Ad-RB94 treatment. (b) TUNEL staining (brown nuclear staining) was observed in approximately the same percentage of cells 48 h afterAd-RB94 treatment as the percentage of RB94-positive cells as shown in (a). Original magnification 400. (c) Confocal examination showing both RB94 and TUNEL staining in the same cells. Plate 1—4,6-diamidino-2-phenylindole (DAPI) staining (blue), plate 2—TUNEL staining (green), plate 3—RB94 staining (red), plate 4—TUNEL and RB94 staining (merged).

Full figure and legend (158K)

In contrast, cells which were RB positive after being treated with Ad-RB110 showed no morphological changes and were TUNEL negative. Moreover, cells treated with Ad-RB94+doxycycline also showed neither RB-positive staining nor were they TUNEL positive, again indicating the doxycycline had completely blocked the RB94 protein expression and that the TUNEL-positive cells originated as a direct result of RB94 treatment. TUNEL-positive cells with similar morphology as seen in UC14 were also apparent 24 h after Ad-RB94 treatment of UC9 cells and were seen primarily in cells with peripheral nuclear chromatin condensation (not shown).

In addition, although at the 24 h time point after Ad-RB94 treatment at least 50% of the UC14 cells showed RB94 nuclear staining, including cells with micronucleation, by 96 h after Ad-RB94 treatment the majority of attached cells were no longer RB94-positive, indicating that most of the RB94-expressing cells were no longer viable and had detached from the dish.

Lack of caspase activation or cytochrome c changes in TUNEL-positive attached cells

No cleavage of caspase 3 and 9 occurred in the attached cells at the 48 h time point (Figures 3a and b), despite the fact that the cells were TUNEL positive (Figure 1e). Examination of cleaved caspase 3 in these TUNEL-positive cells using confocal microscopy and an anti-caspase 3-specific antibody which recognizes cleaved caspase 3 also showed no caspase 3-positive cells (not shown). Cleavage of both caspase 3 and 9, however, did occur in the cells after they had detached and floated in the medium 48 h after Ad-RB94 treatment (Figures 3a and b, respectively).

Figure 3.

Lack of caspase 3 and caspase 9 cleavage or cytochrome c translocation in attached cells 48 h after Ad-RB94 treatment. (a and b) No caspase 3 or caspase 9 cleavage, respectively, was seen in the cells which were still attached 48 h after treatment, although many of these cells were TdT-mediated dUTP nick end labeling (TUNEL) positive (Figure 2b). Cleavage of caspases 3 and 9 was observed in the detached, floating cells, however (c) no cytochrome c translocation from the mitochondria to the cytoplasm was observed after Ad-RB94 treatment unlike that observed after treatment with staurosporine (STS) shown as a positive control producing typical caspase-dependent apoptosis. In addition, cytochrome c is only found in the mitochondria of control cells as expected illustrating the success of the purity of the mitochondrial and cytoplasmic fractions.

Full figure and legend (91K)

In addition, no translocation of cytochrome c from the mitochondria into the cytoplasm, a component of typical apoptosis, was seen prior to or at the same time TUNEL-positive cells were observed (Figure 3c). Both were examined at 27, 29, 36 and 48 h time points. The expected cytochrome c translocation into the cytoplasm was seen after treatment with staurosporine, which produces typical caspase-dependent apoptosis and all the cytochrome c was also seen in the mitochondria of control cells, indicating that the mitochondrial and cytoplasmic fractionation was successful.

Ad-RB94 produces AIF related 50 kb DNA fragmentation which is blocked by AIF inhibition

Although numerous attempts were made to document typical oligonucleosomal DNA fragmentation and the production of DNA laddering by RB94 in the attached TUNEL-positive cells, none could be identified. In contrast, the production of high molecular weight (HMW) 50 kb DNA fragmentation was readily observed in both UC9 and UC14-attached cells following Ad-RB94 treatment (Figure 4a). Caspase-independent apoptosis and peripheral nuclear condensation has been reported in association with the production of 50 kb DNA fragmentation which involved translocation of the AIF from the mitochondria to the nucleus.^{9, 10, 11} Therefore, AIF siRNAs were used to determine the role of AIF status in Ad-RB94-produced 50 kb DNA fragmentation and to document if this fragmentation could be reversed whether or not it also inhibited the cytotoxicity caused by Ad-RB94. In fact AIF protein could readily be blocked in Ad-RB94-treated cells by AIF siRNA (Figure 4a), which in turn completely blocked Ad-RB94-induced 50 kb DNA fragmentation in both UC9 and UC14 cells (Figure 4b). However, RB94-induced apoptosis as measured by TUNEL formation (data not shown) or morphological changes (Figure 4c) was not blocked by AIF siRNA.

Figure 4.

Ad-RB94 producing 50 kb DNA fragmentation involves apoptosis-inducing factor (AIF) but is not a mechanism of RB94-produced cytotoxicity. (a) AIF siRNA blocks AIF protein production in control and Ad-RB94-treated cells (arrow). (b) Ad-RB94 produces 50 kb DNA fragmentation in UC9 and UC14 cells which was inhibited by AIF siRNA as shown by field-inversion gel electrophoresis (FIGE). (c) Morphological changes seen in Ad-RB94-treated cells, including enlarged nuclei with a peripheral nuclear-condensed ring-like appearance (black arrows) and cellular detachment (Figure 1b), were not blocked by AIF siRNA. Original magnification 400. All cells were evaluated 48 h after Ad-RB94 treatment in each experiment.

Full figure and legend (165K)

Ad-RB94 is not cytotoxic to normal urothelial cells

Long-term stable normal human urothelial cells expressing hTERT have recently become available.⁵ These TERT-NHUCs were examined to determine if they showed any growth inhibition or cytotoxicity to Ad-RB94. The NHUCs are easily transfected by adenoviral constructs and at an MOI of 100 over 50% of the cells show expression of the transfected gene such as Ad-gal at 24 or 48 h after treatment. Ad-RB94 produced no inhibition of cell growth as measured by the MTT assay, no morphological changes as seen by phase microscopy and no apoptosis as examined by TUNEL (Figure 5).

Figure 5.

Lack of cytotoxicity produced by Ad-RB94 in telomerase reverse transcriptase-normal human urothelial cells (TERT-NHUCs). (Left) No decrease in cell growth is seen in the TERT-NHUCs after Ad-RB94 treatment. Ad-gal was also used as a control at the same multiplicity of infection (MOI) as Ad-RB94 and in which approximately 50% of the cells show gal staining 48 h after treatment. (a and b, right panels) No morphological changes are seen 48 h after Ad-RB94 treatment compared to untreated control TERT-NHUCs. (c and d) Only rare TUNEL-positive cells were seen in either control or Ad-RB94-treated cells (arrows) indicating that Ad-RB94 did not produce apoptosis as measured by TUNEL in TERT-NHUCs compared to cancer cells (Figure 2b). Original magnification 400.

Full figure and legend (124K)

Discussion

Finding new agents for the treatment of cancer is a major focus of cancer research. RB94 has been found to have characteristics that make it a rather unique and potentially new modality for the treatment of various cancers. First, RB94 has produced cell death in all types of cancer cells, independent of their genetic changes. In addition, to date no cancer cell has been found to be resistant to RB94, as no long-term-positive cells have been obtained following RB94 transfer, unlike the case following RB110 treatment. Moreover, RB94 has not been cytotoxic to normal human cells^{1, 2, 3} including the NHUC cells studied here. Therefore, RB94 should have a very favorable therapeutic index.

As a clinical trial is soon planned using systemic delivery of RB94 for various cancers⁴ it is important to better understand the mechanism(s) by which RB94 causes cell death in cancer cells. We have shown that RB94 produces rapid telomere erosion and chromosomal crisis is human cancer cells but not human normal cells, which could be a key mechanism of RB94-produced tumor cell kill.³ However, such changes are most apparent approximately 48 h or later after RB94 exposure. Therefore, discovering even earlier significant changes produced by RB94 may provide a window to allow a better understanding of the mechanism(s) of RB94-produced cancer cell death.

In this paper we have documented that the initially RB-negative UC14 cells became TUNEL positive by 36 h after Ad-RB94 transfection and in the wild-type RB-positive UC9 cells TUNEL-positive cells were present even at 24 h after Ad-RB94 treatment. Such TUNEL positivity was found primarily in cells with peripheral nuclear chromatin condensation. However, these TUNEL-positive cells, which still remained attached to the culture dishes showed no evidence of caspase 3 or 9 cleavage, including examination of caspase status at the single cell level using confocal analysis. Subsequently, caspase 3 and 9 cleavage occurred in those cells, which detached from the dish, the latter beginning approximately 48 h after treatment. Therefore, caspase cleavage is only seen as a late event in RB94-produced cytotoxicity.

These results are consistent with previous studies where we found that poly-(ADP-ribose)polymerase cleavage occurred in cells but only after they detached from the dish.³ In recent studies we have also found that various tumor cell lines are consistently TUNEL positive (unpublished results) following RB94 treatment and it may be possible to use the presence of increased TUNEL-positive cells in tumors following RB94 treatment as a marker of RB94-produced cell death in the clinical setting. However, we found no cytochrome c translocation from the mitochondria to the cytoplasm following RB94 transfection. Such transfer is usually observed in caspase-dependent apoptosis and these negative findings provided additional support that the early cytotoxic changes following Ad-RB94 treatment are caspase independent.

Of initial interest, HMW 50 kb DNA fragment was produced by RB94-positive cells in both UC9 and UC14 cell lines (Figure 5a). These observations were consistent with stage 1-initiated apoptosis in which caspase-independent HMW 50 kb DNA fragmentation as well as peripheral nuclear chromatin condensation occur. It has been proposed that such DNA fragmentation is produced through an AIF-induced mechanism.^{7, 8, 9} Indeed, we found that blocking AIF using AIF siRNA inhibited HMW 50 kb DNA fragmentation produced by RB94 (Figure 4b). Nevertheless, such inhibition of 50 kb DNA fragmentation did not block RB94-induced cytotoxicity as measured by the number of TUNEL-positive cells or morphological changes observed, including peripheral nuclear chromatin condensation or cellular detachment. Therefore, the formation of HMW 50 kb DNA following RB94 treatment was a change not necessary for the cytotoxicity produced by RB94.

Future studies will continue to concentrate on the molecular mechanism(s) of RB94-produced cytotoxicity in a larger number and types of human cancer cells. We believe these additional studies are need not only because of the potential clinical importance of RB94 as a new modality of treatment but also because of the possibility that some of the changes produced by RB94 may involve a new mechanism of apoptosis. Understanding such molecular events also could help reveal the reason(s) for a lack of RB94 drug resistance to cancer cells observed to date as well as the absence of cytotoxicity to normal human cells. Finally, based on our present studies, finding an increase in the number of TUNEL-positive tumor cells after RB94 treatment may be an important early marker of RB94 efficacy in the phase 1 cancer gene therapy trial, whereas caspase cleavage may be a later marker. Both of these possibilities will be addressed in the clinical trial.

References

1. Xu HJ, Xu K, Zhou Y, Li J, Benedict WF, Hu SX. Enhanced tumor cell growth suppression by an internal AUG codon initiated retinoblastoma protein. Proc Natl Acad Sci USA 1994; 91: 9837–9841. | Article | PubMed | ChemPort |
2. Xu HJ, Zhou Y, Seigne J, Perng GS, Mixon M, Zhang C et al. Enhanced tumor suppressor gene therapy via replication-deficient adenovirus vectors expressing an N-terminal truncated retinoblastoma protein. Cancer Res 1996; 56: 2245–2249. | PubMed | ChemPort |
3. Zhang X, Multani AS, Zhou JH, Shay JW, McConkey D, Dong L et al. Adenoviral-mediated retinoblastoma 94 produces rapid telomere erosion, chromosomal crisis, and caspase-dependent apoptosis in bladder cancer and immortalized human urothelial cells but not in normal urothelial cells. Cancer Res 2003; 63: 760–765. | PubMed | ISI | ChemPort |
4. Pirollo K, Rait A, Zhou Q, Yu W, Hogrefe R, Chang EH et al. Tumor- targeting nanocomplex delivery of tumor suppressor RB94 enhances chemosensitization of bladder carcinoma cells in vitro and in vivo. Clin Cancer Res 2008; 14: 2190–2198. | Article | PubMed | ChemPort |
5. Sabichi A, Keyhani A, Tanaka N, Delacerda J, Lee IL, Zou C et al. Characterization of a panel of cell lines derived from urothelial neoplasmas: genetic alterations, growth in vivo, and the relationship of adenoviral-mediated gene transfer to Coxsackie adenovirus receptor expression. J Urol 2006; 175: 1133–1137. | Article | PubMed | ChemPort |
6. Chapman EJ, Hurst CD, Pitt E, Chambers P, Aveyard JS, Knowles MA. Expression of hTERT immortalises normal human urothelial cells without inactivation of the p16/Rb pathway. Oncogene 2006; 25: 5037–5045. | Article | PubMed | ISI | ChemPort |
7. McConkey DJ. Calcium-dependent, interleukin 1beta-converting enzyme inhibitor-insensitive degradation of lamin B1 and DNA fragmentation in isolated thymocyte nuclei. J Biol Chem 1996; 271: 22398–22406. | PubMed | ChemPort |
8. Niikura Y, Dixit A, Scott R, Perkins G, Kitagawa K. BUB1 mediation of caspase-independent mitotic death determines cell fate. J Cell Biol 2007; 178: 283–296. | Article | PubMed | ChemPort |
9. Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow BE, Brothers GM et al. Molecular characterization of mitochondrial apoptosis-inducing factor. Nature 1999; 397: 441–446. | Article | PubMed | ISI | ChemPort |
10. Daugas E, Susin SA, Zamzami N, Ferri KF, Irinopoulou T, Larochette N et al. Mitochondrio-nuclear translocation of AIF in apoptosis and necrosis. FASEB J 2000; 14: 729–739. | PubMed | ISI | ChemPort |
11. Susin SA, Daugas E, Ravagnan L, Samejima K, Zamzami N, Loeffler M et al. Two distinct pathways leading to nuclear apoptosis. J Exp Med 2000; 192: 571–580. | Article | PubMed | ISI | ChemPort |

Acknowledgements

This study was supported by Grant CA-097127 from the National Cancer Institute to WFB and a Career Development Award to JZ from The GU SPORE in Bladder Cancer CA-091846.