Apoptosis induced in normal human hepatocytes by tumor necrosis ...

© 2000 Nature America Inc. • http://medicine.nature.com

ARTICLES

Apoptosis induced in normal human hepatocytes by tumor necrosis factor-related apoptosis-inducing ligand MINJI JO1, TAE-HYOUNG KIM1, DAI-WU SEOL2, JAMES E. ESPLEN1, KENNETH DORKO1, TIMOTHY R. BILLIAR2 & STEPHEN C. STROM1


Department of Pathology, School of Medicine, 200 Lothrop St. BST S-450, University of Pittsburgh, Pittsburgh, Pennsylvania 15261, USA Department of Surgery, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania 15261, USA Correspondence should be addressed to S.C.S.; email: [email protected]

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) has been reported to induce apoptosis in various tumor cells but not in nontransformed, normal cells1–3. Preclinical studies in mice and nonhuman primates have shown that administration of TRAIL can induce apoptosis in human tumors, but that no cytotoxicity to normal organs or tissues is found3,4. The susceptibility of tumor cells to TRAIL and an apparent lack of activity in normal cells has lead to a proposal to use TRAIL in cancer therapy. Here, we assessed the sensitivity of hepatocytes from rat, mouse, rhesus monkey and human livers to TRAIL-induced apoptosis. TRAIL induced apoptosis in normal human hepatocytes in culture but not in hepatocytes isolated from the other species. Human hepatocytes showed characteristic features of apoptosis, including cytoplasmic shrinkage, the activation of caspases and DNA fragmentation. Apoptosis and cell death in human hepatocytes was massive and rapid, occurring in more than 60% of the cells exposed to TRAIL within 10 hours. These results indicate that there are species differences in sensitivity to TRAIL, and that substantial liver toxicity might result if TRAIL were used in human cancer therapy. In preliminary studies with human hepatocytes exposed to recombinant human tumor necrosis factor-related apoptosis-inducing ligand (rhTRAIL), more than 60% of the hepatocytes were killed after 6 hours of treatment with rhTRAIL at concentrations of 200 and 400 ng/ml. We used 200 ng/ml rhTRAIL for all subsequent studies. In contrast to reports that rhTRAIL does not induce apoptosis in normal cells1,3, normal human hepatocytes showed extensive apoptosis after 4 hours of treatment with rhTRAIL (Fig. 1a and b). Although there were slight differences between different human cases in the timing of the apoptotic response, we examined 20 different human cases, and for each, apoptosis was clearly evident in more than 60% of the cells within 10 hours. Human fetal hepatocytes and adult cells infected with hepatitis C virus were susceptible to rhTRAIL-induced apoptosis (data not shown). As preclinical studies in mice and monkeys showed no systemic cytotoxicity after the injection of rhTRAIL (refs. 3,4), we isolated parenchymal hepatocytes from livers of rats, mice and rhesus monkeys, and exposed them to rhTRAIL. rhTRAIL did not induce apoptosis in parenchymal hepatocytes from any species other than human (Fig. 1c). Apoptosis was not induced in human liver epithelial cells, a replicating cell line (Fig. 1c), indicating a specificity of rhTRAIL for the parenchymal hepatocyte. Consistent with published work, human tumor cells in culture were sensitive to rhTRAIL. Human epidermoid carcinoma cells (A431) and the hepatocellu564

lar carcinoma cell lines AKN-1 and HepG2 were susceptible to rhTRAIL-induced apoptosis. The morphology of cells after the addition of rhTRAIL was characteristic of apoptotic cells, with blebbing and shrinkage of cytoplasm (Fig. 1a). Fragmentation of DNA was evident after as little as 2 hours of treatment with rhTRAIL and increased with longer exposures (Fig. 1d). Caspase activation is a common feature of cell death through apoptosis. To confirm that TRAIL-induced cell death was the result of apoptosis, we measured, by western blot analysis, caspase activation in human hepatocyes exposed to rhTRAIL. Caspases involved in apoptosis induced by tumor necrosis factor and Fas were also activated in rhTRAIL-induced apoptosis. Initiator caspases 8, 9 and 10 were activated in TRAIL-induced apoptosis in normal human hepatocytes. Western blot analysis of cellular proteins with antibodies against caspase 8 that detect the precursor form (55 kDa) and the active, cleaved form (18 kDa) of caspase 8 showed that the precursor form of caspase 8 was cleaved to the 18-kDa fragment as early as 2 hours after treatment with rhTRAIL. Apoptosis induced by TRAIL was so extensive that most hepatocytes were killed by 6–10 hours. Proteolysis of all cellular proteins in the apoptotic cells made it difficult to detect any cleaved form of caspase 8 by 10 hours after the addition of TRAIL. Proteolysis was so extensive that even the signal for actin was diminished at 10 hours (Fig. 2). Cellular levels of the precursor form of caspase 10 (55 kDa) also decreased beginning at 2 hours after treatment with rhTRAIL, indicating they were cleaved during TRAIL-induced apoptosis. Caspase 9, which is activated after release of cytochrome c from mitochondria5, was also activated in hepatocytes exposed to rhTRAIL (Fig. 2). Cleavage of the precursor form of caspase 9 (45–48 kDa) to the 37-kDa fragment was evident at 2, 4 and 6 hours after exposure to rhTRAIL. The 37-kDa fragment of caspase 10 could no longer be detected 10 hours after treatment with TRAIL, probably reflecting extensive proteolysis of cellular proteins. Executioner or effector caspases 3 and 7 cleave cellular proteins such as poly (ADP-ribose) polymerase and DNA fragmentation factor, resulting in the characteristic physical changes of apoptosis, including DNA fragmentation and cell blebbing6–8. In normal human hepatocytes, exposure to TRAIL resulted in the activation of caspases 3 and 7 (Fig. 2). After normal human hepatocytes were treated for 2 hours with rhTRAIL, the precursor form of caspase 3 (32 kDa) was cleaved to the 19-kDa fragment. The precursor form of caspase 7 (35 kDa) was also cleaved to the catalytically active large subunit (19 kDa) after 2 hours of treatment with rhTRAIL. The activation of caspases 3 and 7 was maxNATURE MEDICINE • VOLUME 6 • NUMBER 5 • MAY 2000


ARTICLES

a

b

Control

4 hours

6 hours


c

d

Fig. 1 a, rhTRAIL-induced cytotoxicity and apoptosis. Morphology of control and apoptotic human hepatocytes. b, TRAIL-induced cell death in normal human hepatocytes. Cells were incubated with 200 ng/ml rhTRAIL for 2 h (L), 4 h (l), 6 h ( ) or 10 h ( ). The MTT assay was used to measure cell viability, as percent of control ( ). c, Sensitivity of hepatocytes from nonhuman species to rhTRAIL. Rhesus monkey hepatocytes (MKH), rat hepatocytes (RH), mouse hepatocytes (MH), human liver epithelial cells

(HLE), tumor cell lines (HepG2, AKN-1 and A431 cells) and human hepatocytes (HH) were incubated for 10 h with 200 ng/ml TRAIL (L) or 400 ng/ml TRAIL ( ). The MTT asay was used to measure cell viability, as percent of control (l). d, DNA fragmentation analysis. Cytosolic DNA from the human hepatocytes treated with TRAIL was separated by electrophoresis and visualized by ethidium bromide staining. Data are from two different human cases (representative of seven examined).

imal at 4–6 hours, and cellular levels of the active caspases and actin decreased thereafter, probably the result of cellular protein degradation accompanying the massive cell death seen at these later times. Restricted expression of TRAIL death receptors DR4 (ref. 9) and DR5 (ref. 10) and ‘decoy’ receptors DcR1 (refs. 11–13) and DcR2 (ref. 14) may regulate the sensitivity of cells to TRAIL. Apoptosis in normal human hepatocytes in culture might be due to the difference of expression of those receptors between tissue and cells in culture. We assessed the expression of TRAIL

receptors in liver tissue, hepatocytes in culture and a tumor cell line, HepG2, by RT–PCR (Fig. 3). Human liver tissue, hepatocytes, HepG2 cells, fetal liver tissue and fetal hepatocytes expressed both DR4 and DR5. These results indicate that the death receptors are expressed in liver tissue as well as isolated hepatocytes. The presence of DR4 and DR5 in liver tissue before cell isolation indicates that TRAIL-induced apoptosis is not an artifact of culture and that TRAIL-induced apoptosis in human liver could be expected. Expression of DcR1 could not be detected in HepG2 cells even after 40 cycles of PCR amplification. DcR1 was expressed in liver tissue and cultured hepatocytes. Expression of DcR1 was downregulated in some cultured hepatocytes; however, DcR2 was expressed in liver tissue, hepatocytes and HepG2 cells. Although it was originally identified based on its ability to induce apoptosis in tumor cells, tumor necrosis factor influences many biological activities, including the immune response15,16. The use of tumor necrosis factor or another apoptosis-inducing ligand, Fas, as anti-cancer agents is limited by the acute systemic toxicity, such as septic shock and fulminant hepatic failure16–18. TRAIL induces apoptosis in many tumor cell lines. As initial studies indicated TRAIL induces apoptosis in tumor cells, but not

Fig. 2 Activation of caspases in TRAIL-induced apoptosis in human hepatocytes. Except for the antibody against caspase 10, which detects only the precursor form (55 kDa) of caspase 10, antibodies against caspases detect both precursor and cleaved forms of caspases. Sizes (left margin): Caspase 3: precursor, 32 kDa; cleaved, 19 kDa. Caspase 7: precursor, 32 kDa; cleaved, 19 kDa. Caspase 8, precursor, 55 kDa; cleaved, 18 kDa. Caspase 9: precursor, 45–48 kDa; cleaved, 35 kDa. Antibody against actin (bottom) demonstrates equal loading of proteins. Data represent three separate experiments. Above gels, time after addition of rhTRAIL (in hours). NATURE MEDICINE • VOLUME 6 • NUMBER 45 • MAY 2000

565


ARTICLES

500 bp

1236 bp

species and the detection of all four TRAIL receptors in human liver tissue as well as cultured hepatocytes indicate that sensitivity to rhTRAIL is not an artifact of culture, but is inherent to humans. The data indicate that if TRAIL were used as an anti-cancer agent in human trials, considerable hepatotoxicity or fulminant hepatic failure could result. Moreover, the extrapolation of data from preclinical investigations in other species should be made with caution, and investigations with human cells should be included in the preclinical evaluation of therapeutic agents. Methods

796 bp

453 bp


838 bp

Human hepatocyte isolation and cell culture. Hepatocytes were isolated by collagenase digestion of human donor livers not used for transplantation19,20. Cells were plated in Dulbecco’s modified Eagle medium with 10% fetal bovine serum for 12–24 h, and the medium was changed to serumfree medium before cells were used for experiments. Human epithelial cells, the human epidermoid carcinoma cell line A431 (American Type Culture Collection, Rockville, Maryland) and human heptocellular carcinoma cell lines AKN-1 (ref. 21) and HepG2 (American Type Culture Collection, Rockville, Maryland) were maintained in Dulbecco’s modified Eagle medium with 10% fetal bovine serum. Production of recombinant TRAIL. The purification of rhTRAIL has been described22. rhTRAIL was further purified by gel filtration chromatography.

Fig. 3 RT–PCR analysis of the expression of TRAIL receptors in human liver tissue, human hepatocytes in culture, tumor cell line (HepG2), human fetal liver tissue (FT), and human fetal hepatocytes (FC). Right margin, amplification products.

in normal cells, it was proposed as a potential cancer therapeutic agent1,3. Preclinical studies have indicated that TRAIL may have little systemic toxicity3,4. Trimerized human TRAIL produces no cytotoxicity in normal mouse tissues when administered at doses that suppress growth of tumor cell lines3. In a study with nonhuman primates, injection of soluble human TRAIL did not cause toxicity to tissues or organs, apparently ‘clearing the way’ for Phase I studies in humans4. Here, rhTRAIL induced apoptosis in vitro in normal human hepatocytes isolated from 20 different individuals, and for each, apoptosis was induced in more than 60% of the cells. We confirmed previously published work, in that hepatocytes from rat, mouse and rhesus monkey were not sensitive to human TRAIL. These results indicate that there are considerable differences between species in their response to TRAIL. The insensitivity of rat, mouse and monkey hepatocytes to TRAIL-induced apoptosis indicates that hepatocyte isolation and culture does not somehow induce sensitivity to TRAIL. As with apoptosis induced by tumor necrosis factor and Fas ligand, several caspases were activated in TRAIL-induced apoptosis in human hepatocytes, including initiator caspases 8, 9 and 10 and executioner caspases 3 and 7. The differences between normal and tumor cells in their sensitivity to TRAIL were initially believed to be due to the different levels of expression of death receptors (DR4 and DR5) and decoy receptors (DcR1 and DcR2). However, the presence or absence of decoy receptors did not correlate with the sensitivity of hepatocytes to rhTRAIL here. Preclinical studies with mice and nonhuman primates have indicated that TRAIL does not induce substantial systemic toxicity3,4. Our results with cultured hepatocytes confirm the preclinical data and indicate that mouse, rat and monkey hepatocytes are not killed by exposure to rhTRAIL. However, human hepatocytes are very sensitive to rhTRAIL-induced apoptosis. The difference between human hepatocytes and those from the other 566

DNA fragmentation assay. The DNA fragmentation assay was done as described23. Cells treated with 200 ng/ml TRAIL were collected by centrifugation at 2,000g for 10 min at 4 °C. Cells were lysed with buffer (10 nM EDTA, 0.5% Triton-X100 and 20 nM Tris-HCl, pH 8.0) and incubated for 20 min on ice, then centrifuged at 500g for 10 min at 4 °C. Cytosolic DNA was extracted by phenol:chloroform (1:1) extraction of the supernatants. DNA was treated with 0.1 mg/ml RNase A for 30 min at 37 °C, and were separated by 1.2 % agarose gel electrophoresis and visualized with ethidium bromide staining. MTT assay. Cells were treated with 200 ng/ml TRAIL; 10% (volume/volume) of stock MTT solution (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (thiazolyl blue); Sigma) dissolved in medium at a concentration of 5 mg/ml was added to the medium directly, and cells were incubated for 30 min at 37°C, to allow cell-mediated reduction of MTT, to assay cell viability. The medium was aspirated and cells were washed with phosphate-buffered saline. Isopropanol (0.5 ml, equal to the volume of medium) was added to solublize the dye. Absorbance was measured at 570 nm. Western blot analysis. Cells treated with 200 ng/ml TRAIL were collected by centrifugation at 2,000g for 10 min. Cells were lysed with radioimmunoprecipitation assay buffer containing protease inhibitors. Total cellular proteins were separated by 15 % SDS–PAGE and transferred to PVDF membranes (NEN), blocked for 2 h at room temperature or overnight at 4 °C with 5% nonfat dry milk in blotto solution (20 mM Tris-HCl, pH 7.5, 150 mM NaCl and 0.1% Tween 20), and then were incubated for 2 h at room temperature or overnight at 4 °C with primary antibody. The membranes were washed three times for 10 min each with 1% nonfat dry milk in blotto solution, and were incubated for 1 h at room temperature with secondary antibody conjugated with horseradish peroxidase. The membranes were washed four times for 20 min each with blotto solution. Signals were visualized by enhanced chemiluminescence according to the manufacturer’s instructions (Renaissance chemiluminescence reagent; NEN). RT–PCR. Total RNA was isolated with RNAzol (Tel-Test, Friendswood, Texas) and 1 µg RNA was used for cDNA synthesis. The resulting RT products were amplified using the following conditions: 95 °C for 7 min, followed by 30 cycles (for DR4 and DcR2) or 40 cycles (for DR5 and DcR1) of 95 °C for 1 min, 57 °C for 1 min and 72 °C for 1 min, followed by a final cycle of 72 °C for 5 min. The amplification products were separated by 1 % agarose gel electrophoresis and visualized by ethidium bromide staining. The primers for DR4, DR5, DcR1 and DcR2 have been described 24 and were synthesized by Genosys (The Woodlands, Texas). NATURE MEDICINE • VOLUME 6 • NUMBER 5 • MAY 2000


ARTICLES Acknowledgments This work was supported by National Institutes of Health grant NO1-DK92310 to S.C.S and by RO1-GM-50441 to T.R.B.


RECEIVED 24 NOVEMBER 1999; ACCEPTED 25 FEBRUARY 2000 1. Pitti, R.M. et al. Induction of apoptosis by Apo-2 ligand, a new member of the tumor necrosis factor cytokine family. J. Biol. Chem. 271, 12687–12690 (1996). 2. Wiley, S.R. et al. Identification and characterization of a new member of the TNF family that induces apoptosis. Immunity 3, 673–682 (1995). 3. Walczak, H. et al. Tumoricidal activity of tumor necrosis factor-related apoptosisinducing ligand in vivo. Nature Med. 5, 157–163 (1999). 4. Ashkenazi, A. et al. Safety and antitumor activity of recombinant soluble Apo2 ligand. J. Clin. Invest. 104, 155–162 (1999). 5. Li, P. et al. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91, 479–489 (1997). 6. Germain, M. et al. Cleavage of automodified Poly(ADP-ribose) polymerase during apoptosis. Evidence for involvement of caspase-7. J. Biol. Chem. 274, 2837928384 (1999). 7. D’Amours, D., Germain, M., Orth, K., Dixit, V.M. & Poirier, G.G. Proteolysis of poly(ADP-ribose) polymerase by caspase 3: kinetics of cleavage of mono(ADP-ribosyl)ated and DNA-bound substrates. Radiat. Res. 150, 3–10 (1998). 8. Tang, D. & Kidd, V.J. Cleavage of DFF-45/ICAD by multiple caspases is essential for its function during apoptosis. J. Biol. Chem. 273, 28549–28552 (1998). 9. Pan, G. et al. The receptor for the cytotoxic ligand TRAIL. Science 276, 111-3 (1997). 10. Walczak, H. et al. TRAIL-R2: a novel apoptosis-mediating receptor for TRAIL. Embo. J. 16, 5386–5397 (1997). 11. Pan, G. et al. An antagonist decoy receptor and a death domain-containing receptor for TRAIL. Science 277, 815–818 (1997). 12. Sheridan, J.P. et al. Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science 277, 818–821 (1997).

NATURE MEDICINE • VOLUME 6 • NUMBER 45 • MAY 2000

13. Degli-Esposti, M.A. et al. Cloning and characterization of TRAIL-R3, a novel member of the emerging TRAIL receptor family. J. Exp. Med. 186, 1165–1170 (1997). 14. Pan, G., Ni, J., Yu, G., Wei, Y.F. & Dixit, V.M. TRUNDD, a new member of the TRAIL receptor family that antagonizes TRAIL signalling. FEBS Lett. 424, 41–55 (1998). 15. Fiers, W. Tumor necrosis factor. Characterization at the molecular, cellular and in vivo level. FEBS Lett. 285, 199–212 (1991). 16. Havell, E.A., Fiers, W. & North, R.J. The antitumor function of tumor necrosis factor (TNF), I. Therapeutic action of TNF against an established murine sarcoma is indirect, immunologically dependent, and limited by severe toxicity. J. Exp. Med. 167, 1067–1085 (1988). 17. Tanaka, M., Suda, T., Yatomi, T., Nakamura, N. & Nagata, S. Lethal effect of recombinant human Fas ligand in mice pretreated with Propionibacterium acnes. J. Immunol. 158, 2303–2309 (1997). 18. Ogasawara, J. et al. Lethal effect of the anti-Fas antibody in mice. Nature 364, 806–809 (1993); erratum 365(6446):568 (1993). 19. Strom, S.C. et al. Use of human hepatocytes to study P450 gene induction. Meth. Enzymol. 272, 388–401 (1996). 20. Strom, S.C., Dorko, K., Thompson, M.T., Pisarov, L.A. & Nussler, A.K. Ilots de Langerhans et Hepatocytes (eds. Franco, D., Boudjema, K. & Varet, B.) 195–205 (Institut National de la Sante et de la Recherche Medicale, Paris, 1998). 21. Nussler, A.K. et al. Isolation and characterization of a human hepatic epithelial-like cell line (AKN-1) from a normal liver. In Vitro Cell Dev. Biol. Anim. 35, 190–197 (1999). 22. Seol, D.W. & Billiar, T.R. A caspase-9 variant missing the catalytic site is an endogenous inhibitor of apoptosis. J. Biol. Chem. 274, 2072–2076 (1999). 23. Bohlinger, I. et al. DNA fragmentation in mouse organs during endotoxic shock. Am. J. Pathol. 149, 1381–1393 (1996). 24. Kothny-Wilkes, G. et al. Interleukin-1 protects transformed keratinocytes from tumor necrosis factor-related apoptosis-inducing ligand. J. Biol. Chem. 273, 29247-53 (1998).

567