- Email: [email protected]
BRE enhances in vivo growth of tumor cells
BBRC Biochemical and Biophysical Research Communications 326 (2005) 268–273 www.elsevier.com/locate/ybbrc
BRE enhances in vivo growth of tumor cells Ben Chung-Lap Chana, Qing Lia,1, Stephanie Ka-Yee Chowa, Arthur Kar-Keung Chinga, Choong Tsek Liewb, Pak-Leong Lima, Kenneth Ka-Ho Leec, John Yeuk-Hon Chand, Yiu-Loon Chuia,* a
b
Clinical Immunology Unit and Sir Y.K. Pao Centre for Cancer, Prince of Wales Hospital, Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China c Department of Anatomy, Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China d Department of Molecular Pathology, University of Texas, M.D. Anderson Cancer Center, Houston, TX, USA Received 3 November 2004 Available online 21 November 2004
Abstract Human BRE, a death receptor-associating intracellular protein, attenuates apoptotic response of human and mouse tumor cell lines to death receptor stimuli in vitro. In this report, we addressed whether the in vitro antiapoptotic effect of BRE could impact on tumor growth in vivo. We have shown that the mouse Lewis lung carcinoma D122 stable transfectants of human BRE expression vector developed into local tumor significantly faster than the stable transfectants of empty vector and parental D122, in both the syngeneic C57BL/6 host and nude mice. In vitro growth of the BRE stable transfectants was, however, not accelerated. No significant difference in metastasis between the transfectants and the parental D122 was detected. Thus, overexpression of BRE promotes local tumor growth but not metastasis. We conclude that the enhanced tumor growth is more likely due to the antiapoptotic activity of BRE than any direct effect of the protein on cell proliferation. 2004 Published by Elsevier Inc. Keywords: BRE; Apoptosis; Antiapoptotic protein; Cancer; Lewis lung carcinoma; D122; Tumorigenesis; Metastasis; Death receptor-associating protein; TNF-a
Human BRE (UniGene Cluster Hs:80426) was first proposed to be a stress responsive gene, as its transcription could be down-regulated by UV irradiation, retinoic acid (a cellular differentiation agent), and 4-nitroquinoline-1-oxide (a DNA-damaging compound) [1]. Subsequently, by yeast two-hybrid screens with cytoplasmic domain of TNF-R1 as the bait, BRE was identified as a binding partner. Since it was observed that overexpression of BRE in cell lines by transfection could repress TNF-a-induced activation of NF-jB, BRE was *
Corresponding author. Fax: +852 2645 0856. E-mail address: [email protected] (Y.-L. Chui). 1 Present address: Department of Medicine, State University of New York, College of Medicine, Syracuse, NY 13210, USA. 0006-291X/$ - see front matter 2004 Published by Elsevier Inc. doi:10.1016/j.bbrc.2004.11.013
also alternatively named as TNFRSF1A modulator [2]. BRE gene is expressed in multiple transcripts (at least 8) due to alternative splicing [3]. However, only one transcript, aa (renamed subsequently by GenBank as transcript variant 3), is predominantly expressed; the rest added together amounts to only between 20% and 30% of the total BRE transcripts, depending on whether or not the cells were stimulated by LPS. These transcripts could possibly encode a total of 3 protein isoforms, which differ from one another only at the C-terminus. The main transcript encodes a protein of 383 amino acids, BRE-a (renamed as isoform 2). It is this main isoform that was reported in the aforementioned studies [1,2] and in this report; BRE-a is henceforth simply addressed as BRE. The primary amino acid
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
sequence of BRE, which has no homology with any known protein, is highly conserved in the evolution of animal and plant species [3,4], suggesting that this ancient protein may have a unique but irreplaceable function in life since its advent in the single cellular ancestor. Between the BRE proteins of human and mouse, only three residues are different, two of which represent conservative substitutions [4]. Recently, it has been reported that BRE is a component of the multiprotein holoenzyme complex containing at least BRCA1, 2 and BARD1 in the nucleus. BRE together with another component BRCC36 was found to enhance in vitro ubiquitination activity (E3) of the BRCA1/BARD1 complex. It was also demonstrated by siRNA approach that BRE could play a pro-survival role in response to ionizing radiation [5]. We have shown that BRE resides in both the cytosolic and nuclear compartments, but with the majority located in the cytosol. BRE binds not only to TNF-R1, but also to Fas. Despite down-modulation of the prosurvival NF-jB activation by overexpression of BRE [2], apoptosis in response to death receptor stimulation in vitro was attenuated through blocking of the mitochondria-dependent branch of apoptosis. However, unlike the antiapoptotic Bcl-2 family members [6], BRE does not translocate to mitochondria to exert this inhibition; instead, it remains associated with the death-inducing signal complex (DISC) during apoptotic induction, and possibly suppresses the pro-mitochondrial apoptotic function of some phosphorylated proteins proximal to the DISC by ubiquitination and/or sumoylation. Furthermore, our data indicate no trafficking of BRE between the cytosolic and nuclear compartments during apoptotic induction, suggesting that BRE in the two compartments function independently [7]. As down-regulation of apoptosis has been associated with neoplastic development [8–11], we sought to confirm whether the antiapoptotic activity of BRE shown in vitro could be reflected by enhanced tumor growth in vivo. To address this issue, a well-established mouse tumor model, 3LL Lewis lung carcinoma [12], was chosen to test the effect of BRE overexpression on the tumor development.
Materials and methods Mice and cell lines. Male C57BL/6 and nude (BALB/c nude) mice 6–8 weeks old were obtained from the Laboratory Animal Services Centre of the Chinese University of Hong Kong, maintained and experimented according to the guidelines and protocols approved by the UniversityÕs Animal Experimentation Ethics Committee. The poorly immunogenic and highly metastatic D122 clone of 3LL Lewis lung carcinoma of C57BL/6 origin [12] was a generous gift from Prof. Lea Eisenbach at the department of Immunology of the Weizmann Institute of Science, Rehovot, Israel, and maintained in DMEM supplemented with 10% FCS. Generation and maintenance of D122 stable transfectants were as previously described [7].
269
Local tumor growth assay. C57BL/6 and nude mice arranged in groups were injected in the right footpad with 5 · 105 tumor cells in 50 ll PBS. Local tumor growth was monitored by measuring the maximal perpendicular diameters of the tumor with a caliper. Experiments with C57BL/6 mice as the tumor hosts were terminated at the time point when the mean tumor diameter of one of the groups reached around 20 mm. Experiment with the nude mice was terminated on day 30 after inoculation. Metastasis assay. C57BL/6 mice arranged in groups were injected with the tumor cells as above, but the tumor-bearing legs were amputated at the time when the footpad tumors reached between 6 and 8 mm in mean diameter. At day 25 after the amputation, all the surviving mice were sacrificed and the lung metastatic loads were determined by weighing the organ. Lungs harvested from mice which were moribund and euthanized before and around day 25 (i.e., from day 21 onwards) were also included in the analysis. All lung samples were formalin-fixed, paraffin-embedded, sectioned (8 lm), and stained with hematoxylin and eosin for examination of metastasis. Mice without injection of tumor cells were subjected to the same amputation protocol and their lung weights at day 25 were taken as the normal weight control. In vitro cell growth analysis. D122 or its transfectants were seeded at 1 · 104 cells/well in 24-well plate in 1 ml of culture medium. After 24, 48, and 72 h, cells were trypsinized and resuspended in ice-cold PBS. Viable cells were counted by trypan blue exclusion. Statistical analysis. Data are presented as means ± SE. One-way ANOVA with Bonferroni test was used to compare the transfected tumor groups with the parental D122 group. P values of <0.05 were considered statistically significant.
Results Overexpression of BRE promotes local tumor growth We have previously shown that overexpression of BRE in D122, a low immunogenic but highly metastatic clone of 3LL Lewis lung carcinoma, by stable transfection with a human BRE expression vector, attenuated the in vitro apoptotic response to TNF-a, in the presence of cycloheximide [7]. The same set of clones, plus one more empty vector transfectant as an extra negative control, was studied here for local tumor development upon footpad inoculation. Seven mice were included in each group. As shown in Fig. 1A, the two stable transfectants of human BRE (D122-p3BRE-a4 and a6) developed significantly faster than the parental D122 and stable transfectants of empty vector (D122-pcDNA2-A and B) in the syngeneic mouse strain, C57BL/6. To study whether the enhanced tumor growth by BRE overexpression could have resulted from better resistance to anti-tumor T cell response, the experiment was repeated in T cell-deficient nude mice. Similar result showing accelerated tumor growth by the two stable transfectants of human BRE was observed, suggesting that the tumor enhancing effect of BRE is unrelated to anti-tumor T cell response, if any, against D122 (Fig. 1B). To address whether the enhanced tumor growth by BRE overexpression could be due to any effect of the protein on cell proliferation, the transfectants and parental D122 were grown in tissue culture and counted
270
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
Fig. 1. Overexpression of BRE enhances D122 tumor growth in vivo but not cell proliferation in vitro. (A) C57BL/6 and (B) nude mice, both in five groups of seven animals, were injected with one of the following tumor cells in the right footpad: stable D122 transfectants of human BRE expression vector, D122-p3BRE-a4, and -a6; stable transfectants of vector only, D122-pcDNA3-A, and -B; and untransfected parental D122. Value at each indicated time point represents the mean tumor diameter of the group ± standard error (SE), as shown by one side of the error bar. Significant differences upon comparison with parental D122 are indicated for values <0.05 (*) <0.01 (**), or <0.005 (***). Experiments were performed three times with reproducible results. (C) In vitro growth of D122 and transfectants in culture wells as measured by the number of viable cells in triplicates at three consecutive time points starting from 1 · 104 cells/ml. Trypan blue exclusion showed >95% viability for all cultures throughout the experiment.
with trypan blue exclusion for a period of 72 h. The result as shown in Fig. 1C demonstrating retarded growth for all the transfectants compared with the parental D122 does not support the possibility that BRE overexpression could accelerate cell proliferation.
metastasis (Fig. 2B). Histological findings confirmed lung metastasis in all the samples of all the tumor groups (Fig. 3).
Discussion Overexpression of BRE does not significantly promote metastasis To address the possibility of whether BRE overexpression could also promote metastasis, C57BL/6 mice in 10 per group were inoculated with either one of the transfectants or the parental D122 into the right footpads, and the tumor-bearing legs were amputated as the tumors reached 6–8 mm diameter. Lung metastasis was then evaluated at 21–25 days afterwards. The metastatic load in the lungs was determined by the lung weight and metastasis was confirmed by histology. As shown in Fig. 2A, the local footpad tumors of the two transfectants of human BRE reached the point for amputation earlier than those of the empty vector control and parent D122. However, no statistically significant difference in lung weights between all the tumor groups was observed at 21–25 days after the amputation. Although the transfectant D122-p3BRE-a6 may seem to have marginally better metastasizing ability, the mundane performance of the other BRE transfectant, -a4, in metastasis failed to support a significant role of BRE in metastasis. Increased lung weights for all tumor groups compared with the normal lungs indicated
In this report we demonstrated that overexpression of BRE could promote tumor cell growth in vivo. This was shown by using footpad inoculation of 3LL Lewis lung carcinoma D122 clone and its stable transfectants of human BRE expression vector as the experimental model. D122 has low immunogenicity but is highly metastatic. Thus, the model can be used to assess tumor development in both the local tumorigenesis and metastasis [12]. Our findings indicate that stable transfection with human BRE enhances significantly local tumor growth only, but not the metastasis. In vitro proliferation of D122, however, is not accelerated by BRE overexpression. Using the same stable transfectants as reported here, we have previously shown that overexpression of BRE can attenuate the apoptotic response to a wide variety of death stimuli in vitro [7]. Taken together, our data indicate that the enhancement of local tumor growth by BRE overexpression is due to better survival of the tumor cells in vivo. The enhanced in vivo survival, however, does not lead to enhanced metastasis. Compared with local tumorigenesis, metastasis requires fulfillment of additional specific conditions. Successful metastasis embodies acquisition of the ability
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
Fig. 2. Overexpression of BRE promotes local tumor growth significantly but not metastasis. (A) C57BL/6 mice, in 4 groups of 10 animals, were injected with the tumor cells as in Fig. 1A. Value at each indicated time point represents the mean tumor diameter of the group ± standard error (SE), as shown by one side of the error bar. (B) Lung weight indicating the metastatic load measured at 25 days postamputation when the surviving mice were sacrificed or at 21–23 days when the mice were moribund. Value for each group is shown as mean ± SE. One mouse in the parental D122 group, three each of the D122-p3BRE-a4 and -a6 groups, and none in the empty vector control group were moribund at day 21–23. One of the parental D122 groups and two each of the D122-p3BRE-a4 and -a6 groups died between day 18 and 21, and were excluded from analysis, due to advanced state of decomposition. The P value between groups = 0.1823. Experiments were independently performed twice with reproducible results.
to leave the primary tumor site, enter and survive within the circulation, overcome the host defenses, extravasation, and grow in a distant organ as a vascularized tumor [13–15]. For D122, it has been shown that metastasis of the tumor cells is countered by a perforin-dependent cytotoxic mechanism mounted by the host. Local tumor growth, however, is only slightly affected by this anti-tumor cytotoxic mechanism [16]. Thus, failure of BRE overexpression to significantly enhance D122 metastasis suggests that this antiapoptotic protein confers little resistance to perforin-dependent apoptosis. Although dependence of mitochondrial perturbation by perforin/granzyme B-induced apoptosis has recently been shown in hematopoietic cells [17],
271
and that the antiapoptotic action of BRE is based on stabilization of the mitochondria [7], the issue of whether cell types other than the hematopoietic ones have the same mitochondrial dependence in perforin/ granzyme B-induced apoptosis remains unsettled [17]. Our present data therefore suggest that mitochondrial release of such proapoptotic factors as Smac/Diablo and HtrA2/Omi is not mandatory in perforin/granzyme B-induced apoptosis of D122. Local tumor growth of D122 has also been shown to be only slightly affected by Fas-dependent anti-tumor cytotoxicity [16]. In the same report, using specific antibody for systemic depletion of TNF-a, it was concluded that this cytokine plays no role in rejecting the D122 made immunogenic by transfection with CD80 and H-2Kb. However, such a conclusion of negative result requires control data to verify the effectiveness of the cytokine-depletion protocol, which was lacking in the report. An earlier paper using knock-out mice deficient in IFN-c, perforin or TNF-a to examine the role for each of the 3 factors in CD8 T cell immunity against an immunogenic but low metastatic A9 clone of 3LL Lewis lung carcinoma reported that only TNF-a is crucial in CD8 T cell-mediated elimination of the lung carcinoma [18]. In concordance with the aforementioned report, the paper also demonstrated minimal effect of the Fas system on controlling the growth of the 3LL carcinoma, despite the finding that D122, though not expressing Fas in vitro, does so in vivo [19]. Our previous study showing increased resistance to TNF-a-induced apoptosis in D122 overexpressing BRE is consistent with the view that this cytokine is important in anti-D122 response in vivo [7]. Furthermore, our observation of enhanced local tumor growth by the BRE transfectants also in the nude mice indicates that innate immunity plays a important role in controlling D122 local tumorigenesis. Many cell types of the innate immune system, such as macrophages and NK cells, are capable of producing TNF-a and have anti-tumor activity. Indeed, the general importance of innate immunity in immune surveillance against tumorigenesis has been demonstrated [20]. Results of this study suggest that the death receptorassociating antiapoptotic BRE may have a role in tumorigenesis. Indeed, silencer of death domain (SODD), a TNF-R1-associating protein that down-regulates the receptor activity and suppresses TNF-a-, as well as Fas-induced apoptosis, has been shown to be overexpressed in pancreatic cancer [21–23]. A recent report has also implied a link between BRE and breast cancer by virtue of its association with BRCA1 and 2 in a nuclear multiprotein holoenzyme complex [5]. Intriguingly, our online search of the SAGE Genie database of the Cancer Genome Anatomy Project revealed overexpression of BRE in ovary, peritoneum, pancreas, breast, lung, and brain cancers [24]. Definitive evidence
272
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
Fig. 3. Histological examination of lung metastasis. Hematoxylin and eosin staining of formalin-fixed, paraffin-embedded lung tissue sections of one representative sample from each tumor group and normal control. Metastatic tumors are indicated by asterisks.
for association between aberrant BRE expression and various cancers will have to await screening of well-annotated and matched normal and cancerous clinical samples. The weight of the data available to date suggests a high likelihood for the existence of such an association. Acknowledgments This work was supported by a direct grant for research of Project Code 2041040 from the Hong Kong Research Grants Council Direct Allocation to the Chinese University of Hong Kong. B.C.L.C. and Q.L. were, respectively, supported by a postdoctoral fellowship and a postgraduate studentship awarded by The Chinese University of Hong Kong. References [1] L. Li, H. Yoo, F.F. Becker, F. Ali-Osman, J.Y. Chan, Identification of a brain- and reproductive-organs-specific gene responsive to DNA damage and retinoic acid, Biochem. Biophys. Res. Commun. 206 (1995) 764–774. [2] C. Gu, A. Castellino, J.Y. Chan, M.V. Chao, BRE: a modulator of TNF-alpha action, FASEB J. 12 (1998) 1101–1108. [3] A.K.K. Ching, P.S. Li, Q. Li, B.C.L. Chan, J.Y.H. Chan, P.L. Lim, J.C. Pang, Y.L. Chui, Expression of human BRE in multiple isoforms, Biochem. Biophys. Res. Commun. 288 (2001) 535–545. [4] A.K.K. Ching, Q. Li, P.L. Lim, J.Y.H. Chan, Y.L. Chui, Expression of a conserved mouse stress-modulating gene, Bre: comparison with the human ortholog, DNA Cell Biol. 22 (2003) 497–504. [5] Y. Dong, M.A. Hakimi, X. Chen, E. Kumaraswamy, N.S. Cooch, A.K. Godwin, R. Shiekhattar, Regulation of BRCC, a holoenzyme complex containing BRCA1 and BRCA2, by a signalosome-
[6] [7]
[8]
[9] [10] [11] [12]
[13] [14]
[15] [16]
[17]
[18]
like subunit and its role in DNA repair, Mol. Cell 12 (2003) 1087– 1099. D.R. Green, G. Kroemer, The pathophysiology of mitochondrial cell death, Science 305 (2004) 626–629. Q. Li, A.K.K. Ching, B.C.L. Chan, S.K.Y. Chow, P.L. Lim, T.C.Y. Ho, W.K. Ip, C.K. Wong, C.W.K. Lam, K.K.H. Lee, J.Y.H. Chan, Y.L. Chui, A death receptor-associated antiapoptotic protein, BRE, inhibits mitochondrial apoptotic pathway, J. Biol. Chem. 2004 Oct 1 [Epub ahead of print]. B. Zhivotovsky, S. Orrenius, Defects in the apoptotic machinery of cancer cells: role in drug resistance, Semin. Cancer Biol. 13 (2003) 125–134. J.A. Hickman, Apoptosis and tumourigenesis, Curr. Opin. Genet. Dev. 12 (2002) 67–72. T. Jacks, R.A. Weinberg, Taking the study of cancer cell survival to a new dimension, Cell 111 (2002) 923–925. N.N. Danial, S.J. Korsmeyer, Cell death: critical control points, Cell 116 (2004) 205–219. O. Mandelboim, E. Vadai, M. Fridkin, A. Katz-Hillel, M. Feldman, G. Berke, L. Eisenbach, Regression of established murine carcinoma metastases following vaccination with tumourassociated antigen peptides, Nat. Med. 1 (1995) 1179–1183. P.S. Steeg, Metastasis suppressors alter the signal transduction of cancer cells, Nat. Rev. Cancer 3 (2003) 55–63. S. Ramaswamy, K.N. Ross, E.S. Lander, T.R. Golub, A molecular signature of metastasis in primary solid tumors, Nat. Genet. 33 (2003) 49–54. L.A. Liotta, E.C. Kohn, CancerÕs deadly signature, Nat. Genet. 33 (2003) 10–11. S.H. Lee, E. Bar-Haim, A. Machlenkin, O. Goldberger, I. Volovitz, E. Vadai, E. Tzehoval, L. Eisenbach, In vivo rejection of tumor cells dependent on CD8 cells that kill independently of perforin and FasL, Cancer Gene Ther. 11 (2004) 237–248. V.R. Sutton, M.E. Wowk, M. Cancilla, J.A. Trapani, Caspase activation by granzyme B is indirect, and caspase autoprocessing requires the release of proapoptotic mitochondrial factors, Immunity 18 (2003) 319–329. A. Pre´vost-Blondel, E. Roth, F.M. Rosenthal, H. Pircher, Crucial role of TNF-alpha in CD8 T cell-mediated elimination of 3LL-A9 Lewis lung carcinoma cells in vivo, J. Immunol. 164 (2000) 3645– 3651.
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273 [19] D. Rosen, J.H. Li, S. Keidar, I. Markon, R. Orda, G. Berke, Tumor immunity in perforin-deficient mice: a role for CD95 (Fas/APO-1), J. Immunol. 164 (2000) 3229– 3235. [20] V. Shankaran, H. Ikeda, A.T. Bruce, J.M. White, P.E. Swanson, L.J. Old, R.D. Schreiber, IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity, Nature 410 (2001) 1107–1111. [21] Y. Jiang, J.D. Woronicz, W. Liu, D.V. Goeddel, Prevention of constitutive TNF receptor 1 signaling by silencer of death domains, Science 283 (1999) 543–546.
273
[22] H. Eichholtz-Wirth, E. Fritz, L. Wolz, Overexpression of the Ôsilencer of death domain,Õ SODD/BAG-4, modulates both TNFR1- and CD95-dependent cell death pathways, Cancer Lett. 194 (2003) 81–89. [23] F. Ozawa, H. Friess, A. Zimmermann, J. Kleeff, M.W. Buchler, Enhanced expression of Silencer of death domains (SODD/BAG4) in pancreatic cancer, Biochem. Biophys. Res. Commun. 271 (2000) 409–413. [24] Available from: (Accessed 28 October, 2004).
BRE enhances in vivo growth of tumor cells Ben Chung-Lap Chana, Qing Lia,1, Stephanie Ka-Yee Chowa, Arthur Kar-Keung Chinga, Choong Tsek Liewb, Pak-Leong Lima, Kenneth Ka-Ho Leec, John Yeuk-Hon Chand, Yiu-Loon Chuia,* a
b
Clinical Immunology Unit and Sir Y.K. Pao Centre for Cancer, Prince of Wales Hospital, Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China c Department of Anatomy, Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China d Department of Molecular Pathology, University of Texas, M.D. Anderson Cancer Center, Houston, TX, USA Received 3 November 2004 Available online 21 November 2004
Abstract Human BRE, a death receptor-associating intracellular protein, attenuates apoptotic response of human and mouse tumor cell lines to death receptor stimuli in vitro. In this report, we addressed whether the in vitro antiapoptotic effect of BRE could impact on tumor growth in vivo. We have shown that the mouse Lewis lung carcinoma D122 stable transfectants of human BRE expression vector developed into local tumor significantly faster than the stable transfectants of empty vector and parental D122, in both the syngeneic C57BL/6 host and nude mice. In vitro growth of the BRE stable transfectants was, however, not accelerated. No significant difference in metastasis between the transfectants and the parental D122 was detected. Thus, overexpression of BRE promotes local tumor growth but not metastasis. We conclude that the enhanced tumor growth is more likely due to the antiapoptotic activity of BRE than any direct effect of the protein on cell proliferation. 2004 Published by Elsevier Inc. Keywords: BRE; Apoptosis; Antiapoptotic protein; Cancer; Lewis lung carcinoma; D122; Tumorigenesis; Metastasis; Death receptor-associating protein; TNF-a
Human BRE (UniGene Cluster Hs:80426) was first proposed to be a stress responsive gene, as its transcription could be down-regulated by UV irradiation, retinoic acid (a cellular differentiation agent), and 4-nitroquinoline-1-oxide (a DNA-damaging compound) [1]. Subsequently, by yeast two-hybrid screens with cytoplasmic domain of TNF-R1 as the bait, BRE was identified as a binding partner. Since it was observed that overexpression of BRE in cell lines by transfection could repress TNF-a-induced activation of NF-jB, BRE was *
Corresponding author. Fax: +852 2645 0856. E-mail address: [email protected] (Y.-L. Chui). 1 Present address: Department of Medicine, State University of New York, College of Medicine, Syracuse, NY 13210, USA. 0006-291X/$ - see front matter 2004 Published by Elsevier Inc. doi:10.1016/j.bbrc.2004.11.013
also alternatively named as TNFRSF1A modulator [2]. BRE gene is expressed in multiple transcripts (at least 8) due to alternative splicing [3]. However, only one transcript, aa (renamed subsequently by GenBank as transcript variant 3), is predominantly expressed; the rest added together amounts to only between 20% and 30% of the total BRE transcripts, depending on whether or not the cells were stimulated by LPS. These transcripts could possibly encode a total of 3 protein isoforms, which differ from one another only at the C-terminus. The main transcript encodes a protein of 383 amino acids, BRE-a (renamed as isoform 2). It is this main isoform that was reported in the aforementioned studies [1,2] and in this report; BRE-a is henceforth simply addressed as BRE. The primary amino acid
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
sequence of BRE, which has no homology with any known protein, is highly conserved in the evolution of animal and plant species [3,4], suggesting that this ancient protein may have a unique but irreplaceable function in life since its advent in the single cellular ancestor. Between the BRE proteins of human and mouse, only three residues are different, two of which represent conservative substitutions [4]. Recently, it has been reported that BRE is a component of the multiprotein holoenzyme complex containing at least BRCA1, 2 and BARD1 in the nucleus. BRE together with another component BRCC36 was found to enhance in vitro ubiquitination activity (E3) of the BRCA1/BARD1 complex. It was also demonstrated by siRNA approach that BRE could play a pro-survival role in response to ionizing radiation [5]. We have shown that BRE resides in both the cytosolic and nuclear compartments, but with the majority located in the cytosol. BRE binds not only to TNF-R1, but also to Fas. Despite down-modulation of the prosurvival NF-jB activation by overexpression of BRE [2], apoptosis in response to death receptor stimulation in vitro was attenuated through blocking of the mitochondria-dependent branch of apoptosis. However, unlike the antiapoptotic Bcl-2 family members [6], BRE does not translocate to mitochondria to exert this inhibition; instead, it remains associated with the death-inducing signal complex (DISC) during apoptotic induction, and possibly suppresses the pro-mitochondrial apoptotic function of some phosphorylated proteins proximal to the DISC by ubiquitination and/or sumoylation. Furthermore, our data indicate no trafficking of BRE between the cytosolic and nuclear compartments during apoptotic induction, suggesting that BRE in the two compartments function independently [7]. As down-regulation of apoptosis has been associated with neoplastic development [8–11], we sought to confirm whether the antiapoptotic activity of BRE shown in vitro could be reflected by enhanced tumor growth in vivo. To address this issue, a well-established mouse tumor model, 3LL Lewis lung carcinoma [12], was chosen to test the effect of BRE overexpression on the tumor development.
Materials and methods Mice and cell lines. Male C57BL/6 and nude (BALB/c nude) mice 6–8 weeks old were obtained from the Laboratory Animal Services Centre of the Chinese University of Hong Kong, maintained and experimented according to the guidelines and protocols approved by the UniversityÕs Animal Experimentation Ethics Committee. The poorly immunogenic and highly metastatic D122 clone of 3LL Lewis lung carcinoma of C57BL/6 origin [12] was a generous gift from Prof. Lea Eisenbach at the department of Immunology of the Weizmann Institute of Science, Rehovot, Israel, and maintained in DMEM supplemented with 10% FCS. Generation and maintenance of D122 stable transfectants were as previously described [7].
269
Local tumor growth assay. C57BL/6 and nude mice arranged in groups were injected in the right footpad with 5 · 105 tumor cells in 50 ll PBS. Local tumor growth was monitored by measuring the maximal perpendicular diameters of the tumor with a caliper. Experiments with C57BL/6 mice as the tumor hosts were terminated at the time point when the mean tumor diameter of one of the groups reached around 20 mm. Experiment with the nude mice was terminated on day 30 after inoculation. Metastasis assay. C57BL/6 mice arranged in groups were injected with the tumor cells as above, but the tumor-bearing legs were amputated at the time when the footpad tumors reached between 6 and 8 mm in mean diameter. At day 25 after the amputation, all the surviving mice were sacrificed and the lung metastatic loads were determined by weighing the organ. Lungs harvested from mice which were moribund and euthanized before and around day 25 (i.e., from day 21 onwards) were also included in the analysis. All lung samples were formalin-fixed, paraffin-embedded, sectioned (8 lm), and stained with hematoxylin and eosin for examination of metastasis. Mice without injection of tumor cells were subjected to the same amputation protocol and their lung weights at day 25 were taken as the normal weight control. In vitro cell growth analysis. D122 or its transfectants were seeded at 1 · 104 cells/well in 24-well plate in 1 ml of culture medium. After 24, 48, and 72 h, cells were trypsinized and resuspended in ice-cold PBS. Viable cells were counted by trypan blue exclusion. Statistical analysis. Data are presented as means ± SE. One-way ANOVA with Bonferroni test was used to compare the transfected tumor groups with the parental D122 group. P values of <0.05 were considered statistically significant.
Results Overexpression of BRE promotes local tumor growth We have previously shown that overexpression of BRE in D122, a low immunogenic but highly metastatic clone of 3LL Lewis lung carcinoma, by stable transfection with a human BRE expression vector, attenuated the in vitro apoptotic response to TNF-a, in the presence of cycloheximide [7]. The same set of clones, plus one more empty vector transfectant as an extra negative control, was studied here for local tumor development upon footpad inoculation. Seven mice were included in each group. As shown in Fig. 1A, the two stable transfectants of human BRE (D122-p3BRE-a4 and a6) developed significantly faster than the parental D122 and stable transfectants of empty vector (D122-pcDNA2-A and B) in the syngeneic mouse strain, C57BL/6. To study whether the enhanced tumor growth by BRE overexpression could have resulted from better resistance to anti-tumor T cell response, the experiment was repeated in T cell-deficient nude mice. Similar result showing accelerated tumor growth by the two stable transfectants of human BRE was observed, suggesting that the tumor enhancing effect of BRE is unrelated to anti-tumor T cell response, if any, against D122 (Fig. 1B). To address whether the enhanced tumor growth by BRE overexpression could be due to any effect of the protein on cell proliferation, the transfectants and parental D122 were grown in tissue culture and counted
270
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
Fig. 1. Overexpression of BRE enhances D122 tumor growth in vivo but not cell proliferation in vitro. (A) C57BL/6 and (B) nude mice, both in five groups of seven animals, were injected with one of the following tumor cells in the right footpad: stable D122 transfectants of human BRE expression vector, D122-p3BRE-a4, and -a6; stable transfectants of vector only, D122-pcDNA3-A, and -B; and untransfected parental D122. Value at each indicated time point represents the mean tumor diameter of the group ± standard error (SE), as shown by one side of the error bar. Significant differences upon comparison with parental D122 are indicated for values <0.05 (*) <0.01 (**), or <0.005 (***). Experiments were performed three times with reproducible results. (C) In vitro growth of D122 and transfectants in culture wells as measured by the number of viable cells in triplicates at three consecutive time points starting from 1 · 104 cells/ml. Trypan blue exclusion showed >95% viability for all cultures throughout the experiment.
with trypan blue exclusion for a period of 72 h. The result as shown in Fig. 1C demonstrating retarded growth for all the transfectants compared with the parental D122 does not support the possibility that BRE overexpression could accelerate cell proliferation.
metastasis (Fig. 2B). Histological findings confirmed lung metastasis in all the samples of all the tumor groups (Fig. 3).
Discussion Overexpression of BRE does not significantly promote metastasis To address the possibility of whether BRE overexpression could also promote metastasis, C57BL/6 mice in 10 per group were inoculated with either one of the transfectants or the parental D122 into the right footpads, and the tumor-bearing legs were amputated as the tumors reached 6–8 mm diameter. Lung metastasis was then evaluated at 21–25 days afterwards. The metastatic load in the lungs was determined by the lung weight and metastasis was confirmed by histology. As shown in Fig. 2A, the local footpad tumors of the two transfectants of human BRE reached the point for amputation earlier than those of the empty vector control and parent D122. However, no statistically significant difference in lung weights between all the tumor groups was observed at 21–25 days after the amputation. Although the transfectant D122-p3BRE-a6 may seem to have marginally better metastasizing ability, the mundane performance of the other BRE transfectant, -a4, in metastasis failed to support a significant role of BRE in metastasis. Increased lung weights for all tumor groups compared with the normal lungs indicated
In this report we demonstrated that overexpression of BRE could promote tumor cell growth in vivo. This was shown by using footpad inoculation of 3LL Lewis lung carcinoma D122 clone and its stable transfectants of human BRE expression vector as the experimental model. D122 has low immunogenicity but is highly metastatic. Thus, the model can be used to assess tumor development in both the local tumorigenesis and metastasis [12]. Our findings indicate that stable transfection with human BRE enhances significantly local tumor growth only, but not the metastasis. In vitro proliferation of D122, however, is not accelerated by BRE overexpression. Using the same stable transfectants as reported here, we have previously shown that overexpression of BRE can attenuate the apoptotic response to a wide variety of death stimuli in vitro [7]. Taken together, our data indicate that the enhancement of local tumor growth by BRE overexpression is due to better survival of the tumor cells in vivo. The enhanced in vivo survival, however, does not lead to enhanced metastasis. Compared with local tumorigenesis, metastasis requires fulfillment of additional specific conditions. Successful metastasis embodies acquisition of the ability
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
Fig. 2. Overexpression of BRE promotes local tumor growth significantly but not metastasis. (A) C57BL/6 mice, in 4 groups of 10 animals, were injected with the tumor cells as in Fig. 1A. Value at each indicated time point represents the mean tumor diameter of the group ± standard error (SE), as shown by one side of the error bar. (B) Lung weight indicating the metastatic load measured at 25 days postamputation when the surviving mice were sacrificed or at 21–23 days when the mice were moribund. Value for each group is shown as mean ± SE. One mouse in the parental D122 group, three each of the D122-p3BRE-a4 and -a6 groups, and none in the empty vector control group were moribund at day 21–23. One of the parental D122 groups and two each of the D122-p3BRE-a4 and -a6 groups died between day 18 and 21, and were excluded from analysis, due to advanced state of decomposition. The P value between groups = 0.1823. Experiments were independently performed twice with reproducible results.
to leave the primary tumor site, enter and survive within the circulation, overcome the host defenses, extravasation, and grow in a distant organ as a vascularized tumor [13–15]. For D122, it has been shown that metastasis of the tumor cells is countered by a perforin-dependent cytotoxic mechanism mounted by the host. Local tumor growth, however, is only slightly affected by this anti-tumor cytotoxic mechanism [16]. Thus, failure of BRE overexpression to significantly enhance D122 metastasis suggests that this antiapoptotic protein confers little resistance to perforin-dependent apoptosis. Although dependence of mitochondrial perturbation by perforin/granzyme B-induced apoptosis has recently been shown in hematopoietic cells [17],
271
and that the antiapoptotic action of BRE is based on stabilization of the mitochondria [7], the issue of whether cell types other than the hematopoietic ones have the same mitochondrial dependence in perforin/ granzyme B-induced apoptosis remains unsettled [17]. Our present data therefore suggest that mitochondrial release of such proapoptotic factors as Smac/Diablo and HtrA2/Omi is not mandatory in perforin/granzyme B-induced apoptosis of D122. Local tumor growth of D122 has also been shown to be only slightly affected by Fas-dependent anti-tumor cytotoxicity [16]. In the same report, using specific antibody for systemic depletion of TNF-a, it was concluded that this cytokine plays no role in rejecting the D122 made immunogenic by transfection with CD80 and H-2Kb. However, such a conclusion of negative result requires control data to verify the effectiveness of the cytokine-depletion protocol, which was lacking in the report. An earlier paper using knock-out mice deficient in IFN-c, perforin or TNF-a to examine the role for each of the 3 factors in CD8 T cell immunity against an immunogenic but low metastatic A9 clone of 3LL Lewis lung carcinoma reported that only TNF-a is crucial in CD8 T cell-mediated elimination of the lung carcinoma [18]. In concordance with the aforementioned report, the paper also demonstrated minimal effect of the Fas system on controlling the growth of the 3LL carcinoma, despite the finding that D122, though not expressing Fas in vitro, does so in vivo [19]. Our previous study showing increased resistance to TNF-a-induced apoptosis in D122 overexpressing BRE is consistent with the view that this cytokine is important in anti-D122 response in vivo [7]. Furthermore, our observation of enhanced local tumor growth by the BRE transfectants also in the nude mice indicates that innate immunity plays a important role in controlling D122 local tumorigenesis. Many cell types of the innate immune system, such as macrophages and NK cells, are capable of producing TNF-a and have anti-tumor activity. Indeed, the general importance of innate immunity in immune surveillance against tumorigenesis has been demonstrated [20]. Results of this study suggest that the death receptorassociating antiapoptotic BRE may have a role in tumorigenesis. Indeed, silencer of death domain (SODD), a TNF-R1-associating protein that down-regulates the receptor activity and suppresses TNF-a-, as well as Fas-induced apoptosis, has been shown to be overexpressed in pancreatic cancer [21–23]. A recent report has also implied a link between BRE and breast cancer by virtue of its association with BRCA1 and 2 in a nuclear multiprotein holoenzyme complex [5]. Intriguingly, our online search of the SAGE Genie database of the Cancer Genome Anatomy Project revealed overexpression of BRE in ovary, peritoneum, pancreas, breast, lung, and brain cancers [24]. Definitive evidence
272
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273
Fig. 3. Histological examination of lung metastasis. Hematoxylin and eosin staining of formalin-fixed, paraffin-embedded lung tissue sections of one representative sample from each tumor group and normal control. Metastatic tumors are indicated by asterisks.
for association between aberrant BRE expression and various cancers will have to await screening of well-annotated and matched normal and cancerous clinical samples. The weight of the data available to date suggests a high likelihood for the existence of such an association. Acknowledgments This work was supported by a direct grant for research of Project Code 2041040 from the Hong Kong Research Grants Council Direct Allocation to the Chinese University of Hong Kong. B.C.L.C. and Q.L. were, respectively, supported by a postdoctoral fellowship and a postgraduate studentship awarded by The Chinese University of Hong Kong. References [1] L. Li, H. Yoo, F.F. Becker, F. Ali-Osman, J.Y. Chan, Identification of a brain- and reproductive-organs-specific gene responsive to DNA damage and retinoic acid, Biochem. Biophys. Res. Commun. 206 (1995) 764–774. [2] C. Gu, A. Castellino, J.Y. Chan, M.V. Chao, BRE: a modulator of TNF-alpha action, FASEB J. 12 (1998) 1101–1108. [3] A.K.K. Ching, P.S. Li, Q. Li, B.C.L. Chan, J.Y.H. Chan, P.L. Lim, J.C. Pang, Y.L. Chui, Expression of human BRE in multiple isoforms, Biochem. Biophys. Res. Commun. 288 (2001) 535–545. [4] A.K.K. Ching, Q. Li, P.L. Lim, J.Y.H. Chan, Y.L. Chui, Expression of a conserved mouse stress-modulating gene, Bre: comparison with the human ortholog, DNA Cell Biol. 22 (2003) 497–504. [5] Y. Dong, M.A. Hakimi, X. Chen, E. Kumaraswamy, N.S. Cooch, A.K. Godwin, R. Shiekhattar, Regulation of BRCC, a holoenzyme complex containing BRCA1 and BRCA2, by a signalosome-
[6] [7]
[8]
[9] [10] [11] [12]
[13] [14]
[15] [16]
[17]
[18]
like subunit and its role in DNA repair, Mol. Cell 12 (2003) 1087– 1099. D.R. Green, G. Kroemer, The pathophysiology of mitochondrial cell death, Science 305 (2004) 626–629. Q. Li, A.K.K. Ching, B.C.L. Chan, S.K.Y. Chow, P.L. Lim, T.C.Y. Ho, W.K. Ip, C.K. Wong, C.W.K. Lam, K.K.H. Lee, J.Y.H. Chan, Y.L. Chui, A death receptor-associated antiapoptotic protein, BRE, inhibits mitochondrial apoptotic pathway, J. Biol. Chem. 2004 Oct 1 [Epub ahead of print]. B. Zhivotovsky, S. Orrenius, Defects in the apoptotic machinery of cancer cells: role in drug resistance, Semin. Cancer Biol. 13 (2003) 125–134. J.A. Hickman, Apoptosis and tumourigenesis, Curr. Opin. Genet. Dev. 12 (2002) 67–72. T. Jacks, R.A. Weinberg, Taking the study of cancer cell survival to a new dimension, Cell 111 (2002) 923–925. N.N. Danial, S.J. Korsmeyer, Cell death: critical control points, Cell 116 (2004) 205–219. O. Mandelboim, E. Vadai, M. Fridkin, A. Katz-Hillel, M. Feldman, G. Berke, L. Eisenbach, Regression of established murine carcinoma metastases following vaccination with tumourassociated antigen peptides, Nat. Med. 1 (1995) 1179–1183. P.S. Steeg, Metastasis suppressors alter the signal transduction of cancer cells, Nat. Rev. Cancer 3 (2003) 55–63. S. Ramaswamy, K.N. Ross, E.S. Lander, T.R. Golub, A molecular signature of metastasis in primary solid tumors, Nat. Genet. 33 (2003) 49–54. L.A. Liotta, E.C. Kohn, CancerÕs deadly signature, Nat. Genet. 33 (2003) 10–11. S.H. Lee, E. Bar-Haim, A. Machlenkin, O. Goldberger, I. Volovitz, E. Vadai, E. Tzehoval, L. Eisenbach, In vivo rejection of tumor cells dependent on CD8 cells that kill independently of perforin and FasL, Cancer Gene Ther. 11 (2004) 237–248. V.R. Sutton, M.E. Wowk, M. Cancilla, J.A. Trapani, Caspase activation by granzyme B is indirect, and caspase autoprocessing requires the release of proapoptotic mitochondrial factors, Immunity 18 (2003) 319–329. A. Pre´vost-Blondel, E. Roth, F.M. Rosenthal, H. Pircher, Crucial role of TNF-alpha in CD8 T cell-mediated elimination of 3LL-A9 Lewis lung carcinoma cells in vivo, J. Immunol. 164 (2000) 3645– 3651.
B.C.-L. Chan / Biochemical and Biophysical Research Communications 326 (2005) 268–273 [19] D. Rosen, J.H. Li, S. Keidar, I. Markon, R. Orda, G. Berke, Tumor immunity in perforin-deficient mice: a role for CD95 (Fas/APO-1), J. Immunol. 164 (2000) 3229– 3235. [20] V. Shankaran, H. Ikeda, A.T. Bruce, J.M. White, P.E. Swanson, L.J. Old, R.D. Schreiber, IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity, Nature 410 (2001) 1107–1111. [21] Y. Jiang, J.D. Woronicz, W. Liu, D.V. Goeddel, Prevention of constitutive TNF receptor 1 signaling by silencer of death domains, Science 283 (1999) 543–546.
273
[22] H. Eichholtz-Wirth, E. Fritz, L. Wolz, Overexpression of the Ôsilencer of death domain,Õ SODD/BAG-4, modulates both TNFR1- and CD95-dependent cell death pathways, Cancer Lett. 194 (2003) 81–89. [23] F. Ozawa, H. Friess, A. Zimmermann, J. Kleeff, M.W. Buchler, Enhanced expression of Silencer of death domains (SODD/BAG4) in pancreatic cancer, Biochem. Biophys. Res. Commun. 271 (2000) 409–413. [24] Available from: