• graft-versus-host disease;
  • bone marrow transplantation;
  • donor lymphocyte infusion;
  • Fas ligand


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The graft-versus-tumor (GVT) effect that occurs following allogeneic bone marrow transplantation (BMT) and donor lymphocyte infusion (DLI) is currently being subjected to intensive investigation because of clinical evidence for GVT efficacy against leukemia. In this report, we investigate the efficacy and molecular mechanisms of GVT against solid tumors, using a modification of the mouse parent-to-F1 BMT model. Mouse Colon26 cells in which tumor necrosis factor related apoptosis-inducing ligand (TRAIL) receptor expression was stably knocked down were transplanted to investigate the role of the TRAIL-TRAIL receptor system in the GVT effect. In addition, Fas ligand-(FasL) deficient mice on a C57BL6 (B6) background were used as donors, to determine the significance of the Fas-FasL system for the antitumor effect. The group that received B6 DLI followed by preconditioning with 950 rad irradiation underwent tumor reduction associated with the induction of IFN-γ, TRAIL and tumor-cell apoptosis. In vitro cultured Colon26 cells were resistant to TRAIL but susceptible to the combination of IFN-γ and TRAIL in a TRAIL-dose-dependent manner. The infusion of lymphocytes from FasL-defective donors reduced the tumor progression, although efficacy was decreased in the TRAIL receptor knockdown tumors but not in wild-type ones, compared with infusion of B6-derived lymphocytes. The findings indicate that GVT activity against subcutaneous colon tumors is efficiently induced by preconditioning with irradiation and allogeneic DLI, and that TRAIL and IFN-γ act cooperatively in the antitumor effect. © 2005 Wiley-Liss, Inc.

The combination of allogeneic bone marrow transplantation (BMT) and donor lymphocyte infusion (DLI) represents a potentially curative therapy for patients with hematological malignancies. The antileukemic effect of this approach has been substantiated by clinical evidence; much of the therapeutic potential is related to the graft-versus-leukemia (GVL) effect, in addition to the effects of high doses of chemoradiation therapy before transplantation. Since GVL effects are closely associated with graft-versus-host disease (GVHD), which is a life-threatening complication of BMT, the possibility of distinguishing GVHD from GVL has recently become the focus of intensive research.1, 2, 3, 4

In solid tumors, as opposed to hematological tumors, the malignant cells are enmeshed in a stroma that consists of a complex network of microvasculature and extracellular matrix. As the stroma prevents the effective priming of CTLs, solid cancer cells are often “immunologically ignored” and free from immunological rejection.5, 6 Although the feasibility of inducing a graft-versus-tumor (GVT) effect against solid cancers has already been suggested in some clinical studies,7, 8 the efficacy of this approach and the molecular mechanisms underlying it remain unclear. Previous reports have demonstrated that GVL activities are mediated by natural killer (NK) cells and alloreactive CTLs.9, 10 The cytotoxicity of CTLs is mediated through several effector systems: the Fas ligand–(FasL) Fas, perforin–granzyme and the tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL)-TRAIL receptor pathways. In particular, the TRAIL system is a tumor-specific cytotoxic pathway, as it preferentially induces apoptosis in a variety of cancer cells, with little effect on normal cells.11, 12, 13 TRAIL, a member of the TNF cytokine family, is involved in the cytotoxicity induced by T cells and natural killer (NK) cells.14, 15, 16, 17 TRAIL receptors transduce their death signals through the formation of a death-inducing signaling complex that includes the receptor, an adaptor molecule and caspase 8; the subsequent activation of caspase 3 results in apoptosis.18 Evidence for the safety and potential efficacy of TRAIL therapy against breast and colon cancer has been reported in a severe combined immunodeficiency mouse model.11, 13 As the effective application of this pathway might permit the separation of beneficial GVT effects from detrimental GVHD in transplantation therapies, it is important to unravel the contribution of TRAIL-induced cell death to the GVT effect against solid tumors.

In this study, we modified the mouse allogeneic BMT model established for experimental GVHD analysis to investigate the GVT effect against gastrointestinal tumors. The contribution of TRAIL-induced apoptosis was estimated by using cells in which the expression of TRAIL receptor was stably knocked down. In a mouse model of colon cancer based on subcutaneous transplantation of colon cancer cells, the combination of allogeneic DLI and irradiation preconditioning was effective in inhibiting tumor formation by inducing tumor apoptosis. The TRAIL-dependent cytotoxic system and the IFN-γ dependent immunological cascade contribute in a cooperative fashion to this antitumor effect.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References


All the C57BL/6 (B6, H-2b), BALB/c (H-2d) and BDF1 (H-2b/d) mice were purchased from SLC Corporation (Shizuoka, Japan). FasL-deficient mice (gld mice, B6 background) and their littermates were also from the SLC Corporation. The transgenic mice (B6 background) that ubiquitously express enhanced green fluorescent protein (GFP mice) were a generous gift from Dr. Masaru Okabe (Osaka University, Osaka, Japan).19 All procedures involving experimental animals were performed in accordance with protocols approved by the institutional committee for animal research of the University of Tokyo and complied with the US Public Health Service Policy on Human Care and Use of Laboratory Animals.


Colon26 is a mouse colon cancer cell line derived from BALB/c mice. To establish a Colon26 clone (Colon26-siTR) in which the expression of TRAIL receptor was stably suppressed, the short interfering RNA (siRNA) sequence 5′-ACAGACATCTAGCACGAC-3′ for mouse TRAIL receptor (GenBank accession no. AF176833) was determined using the original algorithm20, 21, 22, 23 and inserted into the pcPUR + U6i cassette vector, which contains the human U6 promoter. The sequence was designed so as not to suppress different TRAIL receptors. The puromycin-resistant clones (Colon26-siTR and control Colon26-WT) were selected as stable transfectants and the knockdown effect was verified by immunoblotting and immunocytochemistry with anti-mouse DR5 antibody for the wild-type receptor and anti-mouse DcR1 antibody for the decoy-type receptor (Cell Sciences, Norwood, MA).


Colon26-WT or Colon26-siTR cells (5 × 103) were prepared in 4-chambered wells and cultured for 36 hr. The cells were fixed in 2% paraformaldehyde for 10 min at room temperature. The cells were blocked with 5% dry-fat milk/PBS that included 10% goat serum for 1 hr at room temperature, and then incubated with primary antibody at 4°C overnight. The following primary antibodies were used: anti-mouse DR5 and anti-mouse DcR1 (Cell Sciences, Norwood, MA). The secondary antibodies were conjugated with fluorochrome Alexa 488 (Molecular Probes, Eugene, OR). The sections were counterstained with propidium iodide and observed under a confocal microscope (Leica Microsystems, Wetzlar, Germany).24

GVT model

Sex- and age-matched BDF1 recipient mice were lethally irradiated with a total dose of 950 rad (MBR-1520RB; Hitachi, Japan) and 5 × 106 bone marrow (BM) cells from B6, GFP or BALB/c mice and 3 × 107 donor splenocytes were injected into the tail veins of the irradiated recipients. For the purification of T cell-depleted BM cells (TCD-BM) from B6 mice, the BM cells were magnetically labeled with CD90 (Thy1.2) MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany) and separated using a MACS separation column according to the manufacturer's protocol. The recipient BM cells were sampled 4 weeks after the transplantation of green fluorescent protein (GFP)- positive BM cells, and chimerism was evaluated by flow cytometry (EPICS XL; Beckman Coulter, Fullerton, CA).

More than 90% of the cells in the recipient BM were GFP-positive, using this BMT protocol (data not shown). To evaluate the GVT effect after BMT, tumors were induced by subcutaneous injection of 1 × 107 murine Colon26 cells into the flanks of mice 2 days before BMT. The sizes of the resulting tumors were measured, and the volumes were calculated as [(length (mm) × width (mm)2)]/2. Survival was monitored daily, and body weight was determined every other day.


The livers, small intestines, kidneys and subcutaneous tumors were harvested from the recipients on days 8 and 14 post-BMT and processed for hematoxylin–eosin staining. The mice were anesthetized with pentobarbital sodium (120 mg/kg), and the tissues were harvested, fixed in 4% paraformaldehyde and embedded in paraffin. To permit effective visualization of GFP fluorescence, the tumor tissue samples were embedded using the Technovit catalyst system (Heraeus Kulzer GmbH, Wehrheim, Germany)25 before sectioning.

TUNEL assay

The TUNEL assay was conducted using the ApopTag kit (Intergen, Purchase, NY), according to the manufacturer's protocol, and the samples were observed under a confocal microscope (Leica Microsystems).24 For the semiquantitative study, the numbers of TUNEL-positive cells were calculated for 3 independent specimens from each of the groups.

Measurement of local blood perfusion in superficial tumor tissues

Tumor blood perfusion was measured using a laser Doppler perfusion imaging (LDPI) system (Moor Instruments, Devon, UK).26, 27 LDPI provides a noninvasive analysis of local blood perfusion in superficial tissues. The principle behind this technique is based on the change in wavelength (Doppler shift) of the laser light reflected from moving objects (the red blood cells in this case), while the wavelength of light reflected from stationary objects remains unchanged. The Doppler-shifted light is converted into an arbitrary perfusion signal, which is approximately proportional to the mean blood-cell velocity multiplied by the concentration of moving blood cells within the sampling volume. LDPI recordings were obtained for the mice on day 8 post-BMT. Each mouse was anesthetized 15 min before recording, to eliminate artifacts caused by body movements, and excess hair was removed from the skin covering the tumor, using a depilatory cream. The mouse was placed on a heating plate at 40°C, and the LDPI recording was made.26, 27

RNA, cDNA and real-time RT-PCR analysis

Total RNA samples from tumors were extracted using the ISOGEN reagent (Nippon Gene, Tokyo, Japan). The expression levels of RNAs for IFN-γ, IFN-inducible protein-10 (IP-10), monokine induced by IFN-γ (Mig), TRAIL and FasL were analyzed using real-time reverse transcriptase polymerase chain reaction (RT-PCR) (Applied Biosystems, Weiterstadt, Germany). An aliquot of total RNA (5 μg) was treated with DNAse I (Roche Diagnostics, Indianapolis, IN), and cDNA was prepared using the Improne II reverse transcription system (Promega, Madison, WI). Mouse hypoxanthine phosphoribosyltransferase (HPRT) mRNA served as the internal standard. The sequences of the PCR primers were as follows: for HPRT, 5′-GTTGGATACAGGCCAGACTTTGTTG-3′ and 5′-GATTCAACTTGCGCTCATCTTAGGC-3′; for IFN-γ, 5′-CAGCAACAGCAAGGCGAAA-3′ and 5′-CTGGACCTGTGGGTTGTTGAC-3′; for IP-10, 5′-CCTTCACCATGTGCCATGC-3′ and 5′-TCTTACATCTGAAATAAAAGAGCTCAGGT-3′; for Mig, 5′-GAGGAACCCTAGTGATAAGGAATGC-3′ and 5′-TCTTCAGTGTAGCAATGATTTCAGTTT-3′; for TRAIL, 5′-TCAGCACTTCAGGATGATGG-3′ and 5′-CACCAGCTGTTTGGTTCTCA-3′ and for FasL, 5′-GGCTGGGTGCCATGCA-3′ and 5′-GGCACTGCTGTCTACCCAGAA-3′.

Statistical analysis

The data are expressed as means ± SE. The levels of tumor growth in the groups were compared using two-way repeat measures ANOVA. The mortality rates in the groups were compared using the log rank test. The results for body-weight changes were analyzed using the 2-tailed Student's t-test. Values of p < 0.05 were considered to be statistically significant.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The combination of irradiation and B6-derived DLI suppresses the growth of subcutaneous Colon26 cancer cells

We modified the mouse model for allogeneic BMT and DLI with a full mismatch for major histocompatibility complex class I and II (B6 [H-2b] or BALB/c [H-2d] into BDF1 [H-2b/d]) to study the GVT effect against solid tumors. Our preliminary experiments demonstrated that BMT and DLI combined with 1,200 rad irradiation caused 100% mortality from GVHD within 30 days.28 On the other hand, the allogeneic DLI mice survived for at least 2 months under conditions of 950 rad. Since systemic GVHD could be enhanced in proportion to the irradiation dose,29 it was assumed GVHD would not be lethal with 950 rad irradiation. Therefore, in all the experiments, we used only 950 rad irradiation and estimated body-weight loss, survival rate and tumor growth.3 Transplantation of 3 × 106 donor BM cells and 3 × 107 splenic lymphocytes after 950 rad irradiation was sufficient to induce >90% engraftment of donor BM cells by day 30, while simultaneously avoiding death induced by myelosuppression.

Initially, we investigated the antitumor effect of DLI on the growth of subcutaneous Colon26 tumors inoculated into BDF1 mice (Fig. 1). Under our experimental conditions, subcutaneous Colon26 cells on a BALB/c background were established almost 100% of the time and progressed in BDF1 host mice. B6-derived DLI inhibited the growth of subcutaneous Colon26 tumors. However, this effect was observed only when DLI was combined with irradiation (p < 0.001). B6-derived DLI without irradiation did not suppress tumor progression; the tumors increased in size to the same extent as did the untreated tumors. Moreover, the TCD-BM and irradiation group showed similar initial growth suppression to the irradiation and DLI groups, which suggests that irradiation contributes to tumor reduction at the first progressive stage. Notably, although initial tumor growth was suppressed to the same degree in the group treated with BALB/c-derived DLI and irradiation as in the irradiated B6-derived DLI group, after day 10 the tumors increased in size to a greater extent in the BALB/c-derived DLI group than in the B6-derived DLI group (p < 0.001). These findings suggest that irradiation preconditioning and the immunological cytotoxicity activity of allogeneic DLI act synergistically.

thumbnail image

Figure 1. The combination of irradiation and B6-derived DLI inhibited the growth of Colon26 tumors. Colon26 cells (1 × 107) were inoculated subcutaneously in the flanks of recipient mice 2 days before BMT and DLI, and tumor size was measured every 2 days thereafter. B6-derived DLI inhibit the growth of subcutaneous Colon26 tumors only when combined with irradiation. Notably, in the group treated with BALB/c-derived DLI or TCD-BM with irradiation, the tumor size after day 8 increased much more than that in the B6-derived DLI group (*p < 0.001, n = 7). Similar results were obtained in 3 independent experiments.

Download figure to PowerPoint

Donor-derived lymphocytes infiltrate solid tumor tissues as well as target organs of GVHD

We have established a mouse model of BMT and DLI in which donor-derived cells are detectable by GFP expression. By day 14 post-BMT, the GFP-positive allogeneic donor lymphocytes had infiltrated the liver and intestine, but not the kidney following 950 rad irradiation (Figs. 2a2c). GFP-positive cells were distributed in the portal vein (PV) area of the liver, surrounding the intrahepatic bile ducts (IHBD; Fig. 2a, arrow), and in the intraepithelium of the intestine (Fig. 2b, arrow). Since the liver and intestine are well-known target organs of GVHD after allogeneic transplantation, these results are consistent with previous clinical findings. Furthermore, there was marked GFP-positive cell infiltration of the tumor tissues (Fig. 2d). TUNEL assays were performed on the tumor sections from day 14. As shown in Figures 2e and 2f, TUNEL-positive cells (red) were distributed throughout the tumor tissues. The TUNEL-positive apoptotic cells were predominantly GFP-negative, indicating that the apoptotic cells were derived from tumor tissues of the host and not from the donor lymphocytes. To compare semiquantitatively the ratios of apoptotic cells among the groups, the TUNEL-positive cells were counted in 3 separate specimens from the various groups (Fig. 2g). At day 14 in the irradiated B6-derived DLI groups, about 30% of the tumor cells were TUNEL-positive, while there were few apoptotic cells in the other groups (p* < 0.01). These findings are consistent with our hypothesis that the GVT effect on tumors requires both B6-derived DLI and irradiation (Fig. 1).

thumbnail image

Figure 2. Donor-derived lymphocyte infiltration followed by cancer cell apoptosis at day 14. By day 14 post-BMT, GFP-positive cells are distributed in the intrahepatic bile ducts ((a), arrow, PV, portal vein; IHBD, intrahepatic bile duct) and in the intraepithelium of the intestine ((b), arrow), but not in the kidney (c). Furthermore, GFP-positive cells have markedly infiltrated the tumor tissues (d) (bar = 80 μm). TUNEL-positive cells (red) are distributed in the tumor tissues treated with irradiation and B6-derived DLI; most of the TUNEL-positive apoptotic cells are GFP-negative (green) ((e) and (f)) (bar = 8 μm). (g) TUNEL-positive cells are evident only in the irradiated B6-derived DLI group (*p < 0.001, n = 3). Similar results were obtained in 3 independent experiments.

Download figure to PowerPoint

Irradiation plays a significant role in preconditioning for the GVT effect via cytokine induction

As shown in Figure 1, irradiation plays a preconditioning role in the GVT effect before day 10. Although irradiation is known to induce the apoptosis of vascular endothelial cells, ischemia-induced necrotic changes were not noted in the irradiated groups (Fig. 3a). Moreover, we monitored the changes in tumor blood flow using LDPI. Tumor blood flow was maintained at preirradiation levels on day 8 postirradiation (Fig. 3b).

thumbnail image

Figure 3. There were no obvious differences in tumor blood flow between the irradiated and nonirradiated groups at day 8 post-BMT. (a) Necrotic changes were not observed in either the irradiated or nonirradiated tumor tissues. (b) The effect of irradiation on tumor blood supply was investigated using LDPI. The upper panel shows the blood velocities in the subcutaneous tumors of mice. The bars in the lower panel indicate the intensities of arbitrary perfusion signals converted from Doppler-shifted light. The average tumor blood flows on day 8 (arrow) were not significantly different for the irradiated and nonirradiated groups (mean ± SE, n = 4). The data shown are representative of 3 independent experiments.

Download figure to PowerPoint

IFN-γ, which is a pleiotropic cytokine secreted by activated T cells and NK cells, plays a central role in both the innate and adaptive immune responses to transformed cells. It is known that irradiation induces the expression of IFN-γ in tumor tissues, which is followed by the activation of antigen-presenting cells. We found that IFN-γ mRNA expression was increased >50-fold (*p < 0.01) in the irradiated tumors as compared to that in the nonirradiated tumors (Fig. 4a). Similarly, the IFN-γ-dependent antiangiogenic factors Mig and IP-10, which have been reported to have chemotactic properties and to enhance leukocyte adhesion in vitro,30 were up-regulated >5.0-fold (*p < 0.01) and >4.3-fold (*p < 0.01), respectively, in the irradiated tumors (Fig. 4b). In a previous report, these factors were shown to lead to vascular remodeling in tumor tissues and the subsequent recruitment of cytotoxic lymphocytes.31

thumbnail image

Figure 4. The expression levels of IFN-γ and IFN-γ-inducible genes were upregulated by irradiation, prior to donor-cell infiltration. The expression levels of IFN-γ, Mig and IP-10 in the Colon26 tumors on day 8 postirradiation were analyzed by real-time RT-PCR. (a) The expression of IFN-γ mRNA was clearly induced in the irradiated tumors, as compared with the nonirradiated tumors (*p < 0.01, mean ± SE, n = 4). (b) Mig and IP-10 were also upregulated in the irradiated tumors (*p < 0.01, mean ± SE, n = 4). The data shown are representative of 3 independent experiments.

Download figure to PowerPoint

Taken together, these results suggest that irradiation induces the expression of IFN-γ and downstream molecules, which participate in preconditioning for the GVT effect.

Effective antitumor immunological cascades mediated by IFN-γ are continuously activated by B6-derived DLI in preirradiated mice

As the expression of IFN-γ was induced in the irradiated tumors by day 8 post-DLI, we estimated the IFN-γ expression levels in the tumors on day 14 using quantitative RT-PCR. We found that the expression of IFN-γ mRNA was maintained at a 20.0-fold (**p < 0.01) higher level in tumors treated with B6-derived DLI following irradiation, as compared to that in the untreated groups. Similarly, the expression patterns of Mig and IP-10 were 7.0-fold (*p < 0.05) and 15.0-fold (**p < 0.01) higher, respectively, in the groups that were treated with B6-derived DLI (Fig. 5a). These results suggest that the immunological cascade maintained by IFN-γ is continuously activated in tumors treated with allogeneic DLI following irradiation. The IFN-γ mRNA levels in the nonirradiated B6-derived DLI groups and BALB/c-derived groups were elevated, although these levels were lower than those of the B6-derived DLI groups (*p < 0.05 and **p < 0.01, respectively) (Fig. 5a).

thumbnail image

Figure 5. The expression levels of IFN-γ, FasL and TRAIL in the tumors were significantly upregulated on day 14 in the groups that received B6-derived DLI following irradiation. (a) The levels of IFN-γ, Mig and IP-10 mRNA on day 14 in the tumors (*p < 0.05, **p < 0.01, mean ± SE, n = 6). (b) The levels of TRAIL and FasL mRNA on day 14 in the tumors (*p < 0.05, **p < 0.01, mean ± SE, n = 6). The data shown are representative of 3 independent experiments.

Download figure to PowerPoint

We also examined the expression levels of 2 effector molecules of cytotoxic T cells, TRAIL and FasL, in the tumors. The mRNA levels for FasL and TRAIL were increased 8.3-fold (**p < 0.01) and 5.0-fold (*p < 0.05), respectively, at day 14 in tumors that were treated with irradiation plus B6-derived DLI, as compared to those in untreated tumors (Fig. 5b). These results appear to be consistent with those of a previous report indicating that TRAIL expression was increased on the surface of activated splenic T cells.32 These findings suggest that these 2 molecules function as effector molecules in the GVT effects seen for the B6-derived DLI groups. The expression levels of TRAIL and FasL may reflect the degree of cell recruitment to the tumor tissues.

TRAIL and FasL act cooperatively in the GVT effect against subcutaneous colon tumors

To elucidate the significance of the TRAIL-TRAIL receptor system in the production of GVT activity against solid tumors, Colon26-siTR cells, in which the wild-type TRAIL receptor gene is stably suppressed and other types of TRAIL receptor are not affected, were established (Fig. 6a). Using immunocytochemistry, the cell-surface expression of TRAIL receptors was detected in the Colon26-WT cells but was found to be decreased in the Colon26-siTR cells (Fig. 6a). As shown in Figure 6b, cultured Colon26-WT cells were resistant to TRAIL but susceptible to the combination of IFN-γ and TRAIL in a TRAIL-dose-dependent manner (*p < 0.01). Moreover, the effects did not appear again in the Colon26-siTR cells.

thumbnail image

Figure 6. Deficiencies in both the TRAIL and FasL systems result in reduction of the GVT effect against Colon26 tumors. (a) Establishment of the Colon26-siTR clone. Immunoblotting and immunocytochemistry clearly demonstrate that the expression of decoy-type TRAIL receptors was not affected; however, the expression of wild-type TRAIL receptors was reduced in these cells. (b) TRAIL has a dose-dependent effect on the growth of cultured Colon26-WT cells only in the presence of IFN-γ. Colon26-WT or Colon26-siTR cells (5 × 103) were grown in 4-well chambers for 24 hr before the addition of medium that contained recombinant mouse IFN-γ (10 μg/ml) and/or TRAIL (0, 1.0, 2.0 μg/ml). The numbers of viable cells were calculated by the TUNEL assay at 48 hr (*p < 0.05, n= 3). The results show the mean ± SE of 3 experiments. (c) Colon26-WT or Colon26-siTR cells (1 × 107) were inoculated subcutaneously into the flanks of recipient mice 2 days before BMT and DLI. B6-derived DLI were clearly effective at inhibiting the growth of Colon26-WT and Colon26-siTR tumors. The infusion of gld donor lymphocytes induced the GVT effect, although the inhibition of tumor growth was less pronounced for the Colon26-siTR tumors than that for the B6-derived DLI (*p < 0.01, **p < 0.001, n = 9). Similar results were obtained in 3 independent experiments. The tumor sizes of the dead mice were excluded from the curve. (d) The mice with Colon26-siTR tumors that received irradiation suffered weight loss by day 7, but those treated with B6-derived DLI recovered by day 20. (p < 0.01 vs. the BALB/c-derived DLI group, n = 9 each) (upper panel). In the untreated group and BALB/c-derived DLI group, the mice became cachexic, lost weight and eventually died as the tumors expanded (lower panel). Similar results were obtained in 3 independent experiments. (e) The survival time of the mice with Colon26-siTR tumors was significantly longer in the B6-derived DLI group than that in the BALB/c-derived DLI group (p < 0.05, n = 9) or untreated group (p < 0.05, n = 9). However, the gld-derived DLI group did not live longer than the BALB/c-derived DLI group (p = 0.0673, n = 9) or the untreated group (p = 0.1847, n = 9).

Download figure to PowerPoint

As shown in Figure 6c, subcutaneous Colon26-siTR tumors grew in the untreated group, while B6-derived DLI was clearly effective in inhibiting the growth of Colon26-siTR tumors, which suggests that TRAIL-dependent cell death is not essential for the production of the GVT effect against Colon26 cells. Given that previous reports have indicated that FasL expression in donor CTLs is not essential for the induction of the GVL effect against hematological malignancies,4 we investigated whether gld donor cells could provoke GVT activity against Colon26 tumors. As shown in Figure 6c, GVT activity against Colon26-WT tumors was unaffected by the lack of FasL in the donor cells, which indicates that, as has been shown for the GVL effect, FasL is not essential for the induction of the GVT effect. Interestingly, when gld-derived donor cells were transplanted into mice with Colon26-siTR tumors, the GVT effect was still observed (**p < 0.001), whereas the growth of the tumors was more rapid as compared with that of the B6-derived DLI group (*p < 0.01) (Fig. 6c).

In Figure 6d, the irradiated nontumor groups showed no significant reduction in body weight (upper panel), and the untreated groups demonstrated obvious body-weight losses concomitant with tumor progression (lower panel). The mice that received irradiation and transplantation without tumor inoculation survived for more than 2 months (Fig. 6e), while the untreated mice with tumors became cachexic and died following growth of the tumor (Fig. 6d upper panel, and Fig. 6e). In the irradiated DLI groups, the body-weight changes correlate with tumor growth. In Figure 6d, the mice with Colon26-siTR tumors that received irradiation suffered weight loss before day 7, but those treated with B6-derived DLI returned to their normal body weights by day 20 (Fig. 6d, upper panel). In contrast, treatment with BALB/c-derived DLI did not suppress Colon26-siTR tumor progression (Fig. 6c); the mice became cachexic and eventually died as the tumors expanded to the same extent as in the untreated mice (Fig. 6d, lower panel, and Fig. 6e). The mice that were treated with B6-derived DLI finally died from tumor expansion by day 30, although their survival times were significantly longer than those of the groups that did not receive DLI (p < 0.05) or that were treated with BALB/c-derived DLI (p < 0.05) (Fig. 6e). The group that was treated with gld-derived DLI did not live longer than the untreated groups (p = 0.1847) or the group treated with BALB/c-derived DLI (p = 0.0673) (Fig. 6e). Taken together, the survival time and body-weight change data correlate with tumor growth rates under our experimental conditions.

These data indicate that the FasL and TRAIL systems contribute in a cooperative fashion to the GVT effect against colon cancer tumors, and that TRAIL-induced cell death can compensate for the defect in FasL-dependent systems.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In this study, we modified a mouse allogeneic BMT model to study the GVT effect on solid tumors. The advantages of this BMT and DLI model are that GVL and GVHD activities are well defined in this model,4, 33 that it allows the use of various H-2b or H-2d cancer cells as target tumors, and that several transgenic mouse strains with the B6 background can be used as donors. In our experiments, GFP mice were used to permit the visualization of donor lymphocyte infiltration of target tissues. In this study, we demonstrated that irradiation and infusion of donor lymphocytes, which are allogeneic with respect to the tumor cells, produce a GVT effect against subcutaneous mouse colon tumors.

In addition to reports on GVL effects against leukemic cells, clinical evidence of the GVT effect against nonhematological tumors has been reported. However, since solid tumor cancers have characteristics that distinguish them from hematological malignancies and increase their resistance to rejection by CTLs, the clinical application of these immunotherapies seems to have stalled. A major reason for the lack of effectiveness of immunological therapies against solid tumors is that antigen presentation is insufficient, owing to aberrations in the abundant tumor vasculature. Remodeling of the tumor vasculature may be required before effective CTL priming and tissue infiltration to eliminate solid tumors occurs.34 In tumor tissues, IFN-γ induces the maturation of effector cells,35 the expression of molecules for antigen presentation36 and the secretion of angiostatic chemokines,37 including IP-10 and Mig, which react with the CXCR3 receptor, inhibit endothelial cell proliferation in vitro, and repress tumor vascularization in vivo.38 A recent report has described irradiation-induced vascular remodeling of tumor vessels via the expression of IFN-γ and IFN-γ-inducible molecules.31 Thus, the up-regulation of these molecules in the irradiated tumors seen in our experiments may also induce effective CTL recruitment during the effector phase via remodeling of the tumor vessels. Moreover, in the set of experiments using Colon26-siTR cells, there was a greater difference in the early GVT effects between the BALB/c and B6 DLI groups, as compared to the experiments using Colon26-WT cells (Fig. 6c), which suggests that the observed effect of irradiation is attributable to TRAIL (Fig. 1). In other words, preconditioning by irradiation may enhance the activation of the IFN-γ-dependent immunological cascade and the recruitment of TRAIL-expressing immune cells, which include NK cells.16

The TUNEL assay demonstrated the dose-dependent effect of TRAIL on Colon26-WT cells apoptosis, when used in combination with IFN-γ. The molecular mechanisms underlying the in vitro synergistic effects of IFN-γ and TRAIL remain to be elucidated. However, IFN-γ has been reported to influence cell sensitivity to TRAIL, for example, by regulating the expression of TRAIL receptors in cultured cells.39 Although the expression of TRAIL receptor in Colon26-WT cells was not enhanced by IFN-γ in our study (data not shown), IFN-γ treatment may influence the expression of other apoptosis-related molecules.40 In addition, since susceptibility to TRAIL therapy has been reported to be increased by the combination of TRAIL and IFN-γ,41, 42, 43, 44 the TRAIL-mediated GVT effect is expected to be enhanced by the observed induction of IFN-γ expression by prior irradiation. We propose that IFN-γ plays a role in TRAIL-induced tumor reduction through various mechanisms, which include extracellular antitumor effects, such as vascular remodeling and immune-cell activation, as well as the modulation of intracellular signal transduction, i.e., the upregulation of apoptosis-related molecules.

The cytotoxicity mediated by TRAIL is thought to be tumor-specific because TRAIL preferentially induces apoptotic cell death in a variety of transformed cells but not in normal cells.11, 12, 13 Evidence for the safety and potential efficacy of TRAIL therapy against breast and colon cancers has been reported for a severe combined immunodeficiency mouse model.11, 13 Trimerized recombinant TRAIL did not induce the apoptosis of normal human hepatocytes that were shown to be susceptible to another version of recombinant TRAIL.45, 46 Our results using TRAIL receptor knockdown cells indicate that the combination of TRAIL and FasL contributes to the antitumor effect. As our results suppose the possibility that the GVT effect could be substituted, to some extent, by TRAIL receptor-specific targeting, the specific method used for clinical application of recombinant TRAIL is of high importance. As TRAIL-dependent cell death is also increased by DNA-damaging agents, such as chemotherapeutic drugs,47 the development of molecular targeting protocols for TRAIL receptors is becoming increasingly important. New cancer treatment strategies might be developed using various combinatorial therapies that include the TRAIL system.

RNA interference (RNAi) is an evolutionarily conserved mechanism for gene silencing.48, 49 We have recently reported that plasmid vectors that express siRNAs can induce long-term, persistent silencing (stable RNAi expression20, 23, 50). This system has become a powerful tool for analyzing the effects of silencing endogenous genes. The deletion of various molecules expressed in donor or recipient cells is a tactic that can be used to address the molecular mechanisms involved in the GVT effect. Importantly, experiments designed to analyze the roles of ligand-receptor systems using gene-targeted mice or cells with genes silenced by stable RNAi may produce different findings, since cellular receptors are often composed of several distinct types, i.e., wild-type or decoy-type. For example, the mutation in the TRAIL knockout mice deletes the entire TRAIL-induced signal pathway, whereas the use of siRNA to knock down the expression of TRAIL receptors suppresses the function of each receptor type individually. Indeed, in contrast to the results of experiments using TRAIL knockout mice,32 knock down of the wild-type TRAIL receptor did not in itself diminish the GVT activity in our model. One possible explanation for this observation is that the perforin system may contribute more significantly to the GVT effect on solid tumors.32 Moreover, the human decoy-type TRAIL receptor TRAIL-R4 has been shown to activate the NFκB-dependent antiapoptosis pathway, in addition to interfering with the TRAIL signal.51 Assuming that a similar decoy-type receptor for mouse TRAIL exists but has not yet been reported, the TRAIL-induced NFκB-dependent antiapoptosis signal may be responsible for the observed decrease in GVT effects in our experiments. In the analysis of the GVT effect against cancer cells, various knockdown cells for each type of receptor are essential to analyze the contribution of each component to the overall antitumor effect.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We would like to thank Mitsuko Tsubouchi and the members of the Sata laboratory for technical assistance. K. Tateishi was supported by a grant from the Japan Health Science Foundation.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Kim YM, Sachs T, Asavaroengchai W, Bronson R, Sykes M. Graft-versus-host disease can be separated from graft-versus-lymphoma effects by control of lymphocyte trafficking with FTY720. J Clin Invest 2003; 111: 65969.
  • 2
    Cooke KR, Gerbitz A, Crawford JM, Teshima T, Hill GR, Tesolin A, Rossignol DP, Ferrara JL. LPS antagonism reduces graft-versus-host disease and preserves graft-versus-leukemia activity after experimental bone marrow transplantation. J Clin Invest 2001; 107: 15819.
  • 3
    Hill GR, Teshima T, Gerbitz A, Pan L, Cooke KR, Brinson YS, Crawford JM, Ferrara JL. Differential roles of IL-1 and TNF-α on graft-versus-host disease and graft versus leukemia. J Clin Invest 1999; 104: 45967.
  • 4
    Schmaltz C, Alpdogan O, Horndasch KJ, Muriglan SJ, Kappel BJ, Teshima T, Ferrara JL, Burakoff SJ, van den Brink MR. Differential use of Fas ligand and perforin cytotoxic pathways by donor T cells in graft-versus-host disease and graft-versus-leukemia effect. Blood 2001; 97: 288695.
  • 5
    Ochsenbein AF, Klenerman P, Karrer U, Ludewig B, Pericin M, Hengartner H, Zinkernagel RM. Immune surveillance against a solid tumor fails because of immunological ignorance. Proc Natl Acad Sci USA 1999; 96: 22338.
  • 6
    Spiotto MT, Yu P, Rowley DA, Nishimura MI, Meredith SC, Gajewski TF, Fu YX, Schreiber H. Increasing tumor antigen expression overcomes “ignorance” to solid tumors via crosspresentation by bone marrow-derived stromal cells. Immunity 2002; 17: 73747.
  • 7
    Eibl B, Schwaighofer H, Nachbaur D, Marth C, Gachter A, Knapp R, Bock G, Gassner C, Schiller L, Petersen F, Niederwieser D. Evidence for a graft-versus-tumor effect in a patient treated with marrow ablative chemotherapy and allogeneic bone marrow transplantation for breast cancer. Blood 1996; 88: 15018.
  • 8
    Childs R, Chernoff A, Contentin N, Bahceci E, Schrump D, Leitman S, Read EJ, Tisdale J, Dunbar C, Linehan WM, Young NS, Barrett AJ. Regression of metastatic renal-cell carcinoma after nonmyeloablative allogeneic peripheral-blood stem-cell transplantation. N Engl J Med 2000; 343: 7508.
  • 9
    Nash RA, Storb R. Graft-versus-host effect after allogeneic hematopoietic stem cell transplantation: GVHD and GVL. Curr Opin Immunol 1996; 8: 67480.
  • 10
    Parham P, McQueen KL. Alloreactive killer cells: hindrance and help for haematopoietic transplants. Nat Rev Immunol 2003; 3: 10822.
  • 11
    Sheridan JP, Marsters SA, Pitti RM, Gurney A, Skubatch M, Baldwin D, Ramakrishnan L, Gray CL, Baker K, Wood WI, Goddard AD, Godowski P, et al. Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science 1997; 277: 81821.
  • 12
    Ashkenazi A, Pai RC, Fong S, Leung S, Lawrence DA, Marsters SA, Blackie C, Chang L, McMurtrey AE, Hebert A, DeForge L, Koumenis IL, et al. Safety and antitumor activity of recombinant soluble Apo2 ligand. J Clin Invest 1999; 104: 15562.
  • 13
    Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Kubin M, Chin W, Jones J, Woodward A, Le T, Smith C, Smolak P, et al. Tumoricidal activity of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat Med 1999; 5: 15763.
  • 14
    Dranoff G. Cytokines in cancer pathogenesis and cancer therapy. Nat Rev Cancer 2004; 4: 1122.
  • 15
    Kayagaki N, Yamaguchi N, Nakayama M, Kawasaki A, Akiba H, Okumura K, Yagita H. Involvement of TNF-related apoptosis-inducing ligand in human CD4+ T cell-mediated cytotoxicity. J Immunol 1999; 162: 263947.
  • 16
    Kayagaki N, Yamaguchi N, Nakayama M, Takeda K, Akiba H, Tsutsui H, Okamura H, Nakanishi K, Okumura K, Yagita H. Expression and function of TNF-related apoptosis-inducing ligand on murine activated NK cells. J Immunol 1999; 163: 190613.
  • 17
    Thomas WD, Hersey P. TNF-related apoptosis-inducing ligand (TRAIL) induces apoptosis in Fas ligand-resistant melanoma cells and mediates CD4 T cell killing of target cells. J Immunol 1998; 161: 2195200.
  • 18
    Ashkenazi A, Dixit VM. Death receptors: signaling and modulation. Science 1998; 281(5381): 13058.
  • 19
    Okabe M, Ikawa M, Kominami K, Nakanishi T, Nishimune Y. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett 1997; 407: 31319.
  • 20
    Miyagishi M, Taira K. U6 promoter-driven siRNAs with four uridine 3′ overhangs efficiently suppress targeted gene expression in mammalian cells. Nat Biotechnol 2002; 20: 497500.
  • 21
    Schneider P, Olson D, Tardivel A, Browning B, Lugovskoy A, Gong D, Dobles M, Hertig S, Hofmann K, Van Vlijmen H, Hsu YM, Burkly LC, et al. Identification of a new murine tumor necrosis factor receptor locus that contains two novel murine receptors for tumor necrosis factor-related apoptosis-inducing ligand (TRAIL). J Biol Chem 2003; 278: 544454.
  • 22
    Wu GS, Burns TF, Zhan Y, Alnemri ES, El-Deiry WS. Molecular cloning and functional analysis of the mouse homologue of the KILLER/DR5 tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) death receptor. Cancer Res 1999; 59: 27705.
  • 23
    Imamura T, Kanai F, Kawakami T, Amarsanaa J, Ijichi H, Hoshida Y, Tanaka Y, Ikenoue T, Tateishi K, Kawabe T, Arakawa Y, Miyagishi M, et al. Proteomic analysis of the TGF-β signaling pathway in pancreatic carcinoma cells using stable RNA interference to silence Smad4 expression. Biochem Biophys Res Commun 2004; 318: 28996.
  • 24
    Tateishi K, Omata M, Tanaka K, Chiba T. The NEDD8 system is essential for cell cycle progression and morphogenetic pathway in mice. J Cell Biol 2001; 155: 5719.
  • 25
    Tanaka K, Sata M, Hirata Y, Nagai R. Diverse contribution of bone marrow cells to neointimal hyperplasia after mechanical vascular injuries. Circ Res 2003; 93: 78390.
  • 26
    Couffinhal T, Silver M, Zheng LP, Kearney M, Witzenbichler B, Isner JM. Mouse model of angiogenesis. Am J Pathol 1998; 152: 166779.
  • 27
    Kragh M, Quistorff B, Tenan M, Van Meir EG, Kristjansen PE. Overexpression of thrombospondin-1 reduces growth and vascular index but not perfusion in glioblastoma. Cancer Res 2002; 62: 11915.
  • 28
    Ohta M, Tateishi K, Kanai F, Ueha S, Guleng B, Washida M, Tanaka Y, Ijichi H, Ikenoue T, Sata M, Sudo T, Shiina S, et al. Reduced p38 mitogen-activated protein kinase in donor grafts accelerates acute intestinal graft-versus-host disease in mice. Eur J Immunol 2005; 35: 221021.
  • 29
    Hill GR, Crawford JM, Cooke KR, Brinson YS, Pan L, Ferrara JL. Total body irradiation and acute graft-versus-host disease: the role of gastrointestinal damage and inflammatory cytokines. Blood 1997; 90: 320413.
  • 30
    Piali L, Weber C, LaRosa G, Mackay CR, Springer TA, Clark-Lewis I, Moser B. The chemokine receptor CXCR3 mediates rapid and shear-resistant adhesion-induction of effector T lymphocytes by the chemokines IP10 and Mig. Eur J Immunol 1998; 28: 96172.
  • 31
    Ganss R, Ryschich E, Klar E, Arnold B, Hammerling GJ. Combination of T-cell therapy and trigger of inflammation induces remodeling of the vasculature and tumor eradication. Cancer Res 2002; 62: 146270.
  • 32
    Schmaltz C, Alpdogan O, Kappel BJ, Muriglan SJ, Rotolo JA, Ongchin J, Willis LM, Greenberg AS, Eng JM, Crawford JM, Murphy GF, Yagita H, et al. T cells require TRAIL for optimal graft-versus-tumor activity. Nat Med 2002; 8: 14337.
  • 33
    Fowler DH, Breglio J, Nagel G, Eckhaus MA, Gress RE. Allospecific CD8+ Tc1 and Tc2 populations in graft-versus-leukemia effect and graft-versus-host disease. J Immunol 1996; 157: 481121.
  • 34
    Singh S, Ross SR, Acena M, Rowley DA, Schreiber H. Stroma is critical for preventing or permitting immunological destruction of antigenic cancer cells. J Exp Med 1992; 175: 13946.
  • 35
    Fallarino F, Gajewski TF. Cutting edge: differentiation of antitumor CTL in vivo requires host expression of Stat1. J Immunol 1999; 163: 410913.
  • 36
    Boehm U, Klamp T, Groot M, Howard JC. Cellular responses to interferon-γ. Annu Rev Immunol 1997; 15: 74995.
  • 37
    Qin Z, Blankenstein T. CD4+ T cell-mediated tumor rejection involves inhibition of angiogenesis that is dependent on IFN-γ receptor expression by nonhematopoietic cells. Immunity 2000; 12: 67786.
  • 38
    Coughlin CM, Salhany KE, Gee MS, LaTemple DC, Kotenko S, Ma X, Gri G, Wysocka M, Kim JE, Liu L, Liao F, Farber JM, et al. Tumor cell responses to IFN-γ affect tumorigenicity and response to IL-12 therapy and antiangiogenesis. Immunity 1998; 9: 2534.
  • 39
    Meng RD, El-Deiry WS. p53-independent upregulation of KILLER/DR5 TRAIL receptor expression by glucocorticoids and interferon-γ. Exp Cell Res 2001; 262: 15469.
  • 40
    Chawla-Sarkar M, Lindner DJ, Liu YF, Williams BR, Sen GC, Silverman RH, Borden EC. Apoptosis and interferons: role of interferon-stimulated genes as mediators of apoptosis. Apoptosis 2003; 8: 23749.
  • 41
    Langaas V, Shahzidi S, Johnsen JI, Smedsrod B, Sveinbjornsson B. Interferon-γ modulates TRAIL-mediated apoptosis in human colon carcinoma cells. Anticancer Res 2001; 21: 37338.
  • 42
    Ruiz-Ruiz C, Lopez-Rivas A. Mitochondria-dependent and -independent mechanisms in tumour necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis are both regulated by interferon-γ in human breast tumour cells. Biochem J 2002; 365(Pt 3): 825–32.
  • 43
    Fulda S, Debatin KM. IFN-γ sensitizes for apoptosis by upregulating caspase-8 expression through the Stat1 pathway. Oncogene 2002; 21: 2295308.
  • 44
    Wang S, Boonman ZF, Li HC, He Y, Jager MJ, Toes RE, Niederkorn JY. Role of TRAIL and IFN-γ in CD4+ T cell-dependent tumor rejection in the anterior chamber of the eye. J Immunol 2003; 171: 278996.
  • 45
    Lawrence D, Shahrokh Z, Marsters S, Achilles K, Shih D, Mounho B, Hillan K, Totpal K, DeForge L, Schow P, Hooley J, Sherwood S, et al. Differential hepatocyte toxicity of recombinant Apo2L/TRAIL versions. Nat Med 2001; 7: 3835.
  • 46
    Jo M, Kim TH, Seol DW, Esplen JE, Dorko K, Billiar TR, Strom SC. Apoptosis induced in normal human hepatocytes by tumor necrosis factor-related apoptosis-inducing ligand. Nat Med 2000; 6: 5647.
  • 47
    Shankar S, Srivastava RK. Enhancement of therapeutic potential of TRAIL by cancer chemotherapy and irradiation: mechanisms and clinical implications. Drug Resist Updat 2004; 7: 13956.
  • 48
    Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 2001; 411: 4948.
  • 49
    Elbashir SM, Lendeckel W, Tuschl T. RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes Dev 2001; 15: 188200.
  • 50
    Brummelkamp TR, Bernards R, Agami R. A system for stable expression of short interfering RNAs in mammalian cells. Science 2002; 296: 5503.
  • 51
    Bernard D, Quatannens B, Vandenbunder B, Abbadie C. Rel/NF-κB transcription factors protect against tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL)-induced apoptosis by up-regulating the TRAIL decoy receptor DcR1. J Biol Chem 2001; 276: 273228.