Cancer Research Infection and Cancer: Biology, Therapeutics, and Prevention AACR Conference on Molecular Diagnostics - 2008
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 67, 4491-4498, May 1, 2007. doi: 10.1158/0008-5472.CAN-06-1627
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saudemont, A.
Right arrow Articles by Quesnel, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saudemont, A.
Right arrow Articles by Quesnel, B.

Immunology

Dormant Tumor Cells Develop Cross-Resistance to Apoptosis Induced by CTLs or Imatinib Mesylate via Methylation of Suppressor of Cytokine Signaling 1

Aurore Saudemont1,4, Abdelbasset Hamrouni1, Philippe Marchetti1, Jizhong Liu1, Nathalie Jouy2, Dominique Hetuin1, Francesco Colucci4 and Bruno Quesnel1,3

1 INSERM, U837, Institut de Recherche sur le Cancer de Lille, 2 Université Lille 2, Institut Fédératif de Recherche 114, and 3 Service des Maladies du Sang, Centre Hospitalier et Universitaire de Lille, Lille, France; and 4 Laboratory of Lymphocyte Signaling and Development, The Babraham Institute, Babraham, Cambridge, United Kingdom

Requests for reprints: Bruno Quesnel, Service des Maladies du Sang, Centre Hospitalier et Universitaire de Lille, Rue Polonovski, 59037, Lille, France. Phone: 33-3-20-44-66-40; Fax: 33-3-20-44-42-94; E-mail: brunoquesnel{at}hotmail.com.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
In the BCR/ABL DA1-3b mouse model of acute myelogenous leukemia, dormant tumor cells may persist in the host in a state of equilibrium with the CD8+ CTL-mediated immune response by actively inhibiting T cells. Dormant tumor cells also show a progressive decrease of suppressor of cytokine signaling 1 (SOCS1) gene expression and a deregulation of the Janus-activated kinase/signal transducers and activators of transcription (JAK/STAT) pathway due to methylation of the SOCS1 gene. Dormant tumor cells were more resistant to apoptosis induced by specific CTLs, but resistance decreased when SOCS1 expression was restored via demethylation or gene transfer. AG490 JAK2 inhibitor decreased the resistance of dormant tumor cells to CTLs, but MG132 proteasome inhibitor was effective only in SOCS1-transfected cells. Thus, SOCS1 regulation of the JAK/STAT pathways contributes to the resistance of tumor cells to CTL-mediated killing. Resistance of dormant tumor cells to apoptosis was also observed when induced by irradiation, cytarabine, or imatinib mesylate, but was reduced by SOCS1 gene transfer. This cross-resistance to apoptosis was induced by interleukin 3 (IL-3) overproduction by dormant tumor cells and was reversed with an anti–IL-3 antibody. Thus, tumor cells that remain dormant for long periods in the host in spite of a specific CTL immune response may deregulate their JAK/STAT pathways and develop cross-resistance to various treatments through an IL-3 autocrine loop. These data suggest possible new therapeutic targets to eradicate dormant tumor cells. [Cancer Res 2007;67(9):4491–8]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
In tumor dormancy, tumor cells persist in the host for a long time but do not grow (1). Tumor dormancy is commonly observed clinically and reflects the existence of residual tumor cells that can lead to relapse, sometimes decades after treatment (2–7). Understanding how malignant cells can persist in a dormant state may lead to new strategies to prevent such relapse.

Little is known about mechanisms involved in tumor dormancy. Dormancy may result from a balance between cell replication and cell death (1, 6, 7), as in the BCL1 mouse model of lymphoma, where an anti-idiotype immune response causes apoptosis (1, 8). We previously reported that in the DA1-3b/C3H mouse model of acute myelogenous leukemia (AML), vaccination of mice with leukemia cells that had been transduced with CD154, granulocyte macrophage colony-stimulating factor (GM-CSF) or interleukin 12 (IL-12) genes led to significant protection against leukemia and generated specific cytotoxic T cells (CTL) against leukemic cells (9, 10). In spite of this immune response, we detected several hundred persistent leukemic cells in long-term surviving mice (11). Persistent leukemic cells were more resistant to specific CTL-mediated killing because they expressed more B7-H1, and expression was proportional to the time they had persisted in the host. These data indicate that tumor dormancy may result from a balance between host immune response and active mechanisms developed by tumor cells to escape from cytotoxic cells. However, the low number of persistent cells suggests that they might also develop endogenous mechanisms to resist cell death.

A possible candidate for regulation of apoptosis in dormant tumor cells is suppressor of cytokine signaling 1 (SOCS1). SOCS1 is an inducible SH2-containing inhibitor of Janus-activated kinase (JAK) kinases and can suppress cytokine signaling (12). SOCS1 may also direct JAK proteins for ubiquitin-mediated degradation (13). In several tumor types, inactivation of SOCS1 by promoter methylation has been reported. Ectopic expression of SOCS1 abolishes proliferation that was mediated by a mutated KIT receptor, TEL-JAK2, or by v-ABL, suggesting that SOCS1 has a tumor-suppressor activity (14–16). Here, we investigated if SOCS1 might regulate apoptosis of dormant tumor cells. We show that SOCS1 is inactivated by DNA methylation, proportional to the time these dormant tumor cells have spent in the host. This inactivation leads to deregulation of JAK/signal transducers and activators of transcription (STAT) pathways and cross-resistance to apoptosis induced by CTLs and drugs through IL-3 overproduction.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mouse model of tumor dormancy and cell lines. The leukemic DA1-3b BCR/ABL-expressing cell line, the DA1-3b/C3H mouse model of tumor dormancy, and isolation of dormant tumor cells have been described previously (9, 11, 17). Briefly, C3H/Hej mice were injected intradermally with irradiated IL-12–transduced cells (DA1-3b/IL-12). Each mouse received three injections, 1 week apart, and immunity was challenged 7 days after the last injection, by i.p. injection of 104 leukemic DA1-3b cells. To analyze for persistent leukemic cells, we cultured 106 spleen cells per mouse and, after 2 or 3 weeks, obtained cell lines from mice killed at day 35 (cell line DA1-3b/d35), day 60 (DA1-3b/d60), day 90 (DA1-3b/d90), day 120 (DA1-3b/d120), and day 365 (DA1-3b/d365).

SOCS1 real-time PCR and SOCS1 quantitative real-time methylation-specific PCR. Quantification of SOCS1 mRNA was done by real-time PCR on an ABI PRISM 7700 system using TaqMan Master Mix 2x (Applied Biosystems). Data were normalized by 18s (PDAR 18S, Applied Biosystems). Primers and probes were designed with Primer Express software and purchased from Applied Biosystems (listed in Supplementary Table S1).

To quantify DNA methylation at the SOCS1 promoter, we developed a quantitative methylation-specific PCR (Q-MSP) technique adapted from previously published techniques (18). Bisulfite conversion of the genomic DNA was carried out using the CpGenome DNA Modification Kit (Intergen): 1 µg of DNA was treated following the manufacturer's recommendations and resuspended in a total volume of 25 µL. Three different sets of primers and probes were used to study promoter methylation of SOCS1 (listed in Supplementary Table S1). Q-MSP was done on an ABI PRISM 7700 system using the TaqMan Master Mix 2x and included water blanks and positive and negative controls. Quantification of methylated DNA and completion of the bisulfite modification were done as previously described (18).

Methylation of the SOCS3 promoter was analyzed by MSP (primers listed in Supplementary Table S1).

SOCS1 transfection and demethylation of dormant tumor cells. The pBLAST45-mSOCS1 expression plasmid and pBLAST45 control plasmid were purchased from Invivogen (Cayla). DA1-3b and DA1-3b/d365 cells were electroporated at 1500 µF and 250 V and further selected with blasticidin (Invivogen).

To analyze the re-expression of SOCS1 in persistent leukemic cells, DA1-3b, DA1-3b/d35, DA1-3b/d90, DA1-3b/d120, and DA1-3b/d365 cells were cultured for 10 days with 5 µmol/L 5-aza-2'-deoxycytidine (5-azadC) and 5 µmol/L trichostatin A (TSA; Sigma) in RPMI 1640 supplemented with 10% fetal bovine serum.

The immunophenotype of dormant tumor cells was analyzed by flow cytometry, with previously described specific monoclonal antibodies (mAb) (11, 17).

CTL assay. Specific CD8a+ CTLs were generated as previously described (9, 11, 17). Target cells were treated with 50 µmol/L AG490 for inhibition of JAK2, 20 µmol/L MG132 for proteasome inhibition, 20 µmol/L SN50 for inhibition of nuclear factor {kappa}B (NF-{kappa}B), or 1% DMSO (Sigma) as control, for 24 h at 37°C. To stimulate the JAK/STAT pathway, target cells were preincubated in 200 IU/mL IFN-{gamma}, 100 ng/mL IL-6, or 1 ng/mL IL-3 for 1 h before the CTL assay. Stimulated cells were also immunophenotyped using the panel of antibodies previously described (11, 17).

For apoptosis studies, N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone (z-VAD-fmk) was added to target cells at 100 µmol/L for 12 h. To inhibit granzyme B, 3,4-dichloroisocoumarin was used at 50 µmol/L for 16 h.

Flow-cytometric detection of intracellular phosphorylated STAT proteins. Expression of P-STAT1, P-STAT3, P-STAT5, and P-STAT6 was assessed by flow cytometry with specific mAbs (Cell Signaling Technology; ref. 19). Cells were resuspended in culture medium at 106 cells/mL and stimulated with IL-2 (200 IU/mL), IL-3 (200 IU/mL), IL-4 (10 ng/mL), IL-5 (30 IU/mL), IL-6 (10 ng/mL), IL-7 (50 ng/mL), IL-10 (100 IU/mL), IL-12 (20 IU/mL), IL-13 (50 ng/mL), IL-15 (10 ng/mL), GM-CSF (100 ng/mL), IFN-{gamma}(200 IU/mL), or tumor necrosis factor-{alpha} (TNF-{alpha}; 500 IU/mL) for 30 min. Immediately after stimulation, cells were fixed and then permeabilized (Intraprep permeabilization reagent, Beckman-Coulter) and incubated with the appropriate antibody for 30 min.

Apoptosis analysis. Quantification of the sub-G1 DNA content, analysis of depolarization of mitochondrial membranes, and production of reactive oxygen species (ROS) with DIOC6 (ref. 3; Bachem) and dihydroethidine (Bachem) were done as previously described (20). Phosphatidylserine exposure was measured with an FITC-annexin V labeling kit. Activated caspase expression was assessed by flow cytometry with FITC-labeled Val-Ala-Asp-fluoromethyl ketone (FITC-VAD-fmk; Bachem). About 106 cells were suspended in 1 mL PBS and incubated for 20 min with FITC-VAD-fmk at 5 µmol/L. Cells were washed twice, fixed with 0.5% paraformaldehyde, and analyzed. NF-{kappa}B binding activity in nuclear extracts was determined with the NF{kappa}B p50/p65 Transcription Factor Colorimetric Assay kit (Chemicon).

Cytokine dosage and blocking experiments. IL-2, IL-3, IL-4, IL-6, IL-10, IL-12, IL-15, IFN-{gamma}, TNF-{alpha}, granulocyte colony-stimulating factor (G-CSF), and GM-CSF production in the conditioned media (CM) of DA1-3b, DA1-3b/d365, and DA1-3b/d365-SOCS1 was analyzed by conventional ELISA methods. IL-3 blocking experiments with CM were done using an anti-mouse IL-3 mAb (clone MP2-8F8; Tebu) or its control isotype rat immunoglobulin G at a concentration of 2 µg/mL.

Statistical analyses. Statistical analyses were done using the Sigma Stat 3.0 software (SPSS Science).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Silencing of SOCS1 in dormant tumor cells and promoter methylation. We previously described the isolation of dormant tumor cells from C3H/Hej mice that had been vaccinated with irradiated DA1-3b/IL-12 cells and then challenged with wild-type DA1-3b leukemic cells expressing BCR/ABL (11). A few hundred dormant tumor cells were detected in the spleen and were isolated after 1, 2, 3, and 4 months and 1 year of complete remission. We established tumor dormancy cell lines DA1-3b/d35, DA1-3b/d60, DA1-3b/d90, DA1-3b/d120, and DA1-3b/d365, respectively. SOCS1 mRNA decreased over time spent in dormancy, with complete extinction after 1 year (Fig. 1A ). When these dormant tumor cells were incubated in 5-azadC, SOCS1 was partially re-expressed. Expression was considerably higher when cells were coincubated in both 5-azadC and TSA, indicating that promoter methylation and histone deacetylation co-operate to lock the gene. The level of methylated SOCS1 DNA increased proportionally with time spent in dormancy (Fig. 1B). Direct bisulfite sequencing of the SOCS1 promoter confirmed DNA methylation (Fig. 1C). We also detected methylated SOCS1 DNA in vivo in the spleen of mice in dormancy, ruling out that methylation could be acquired during adaptation of dormant tumor cells to in vitro culture (Supplementary Fig. S1). As a control, we analyzed the SOCS3 promoter, but it was never methylated, and its expression did not change (data not shown).


Figure 1
View larger version (37K):
[in this window]
[in a new window]

  		Figure 1. Quantitative analysis of methylation and expression of SOCS1 in dormant tumor cells. A, C3H/Hej mice were vaccinated with irradiated DA1-3b leukemic cells that had been transduced with IL-12 (DA1-3b/IL-12). The mice were subsequently challenged with 104 wild-type DA1-3b cells. {square}, expression of SOCS1 mRNA was measured by RQ-PCR in leukemic DA1-3b cells isolated from the spleen at selected times after challenge; Figure 1, same experiment after incubation of cells in 5-azadC and TSA. B, quantification of SOCS1 promoter methylation by Q-MSP, in DA1-3b cells from spleen, as above. C, sequences of bisulfite modified DNA from persistent DA1-3b leukemic cells isolated from mice after 1 mo (DA1-3b/d35) and 1 y (DA1-3b/d365). Yellow, CpG islets protected by methylation. D, survival curves of C3H/Hej mice injected i.p. with 105 DA1-3b/d365 cells transfected with a SOCS1 expression plasmid (DA1-3b/d365-pB.SOCS1) or with a control empty vector (DA1-3b/d365-pB; 10 mice per group).

 
Cells methylated on SOCS1 showed similar leukemogenicity to DA1-3b cells when injected into naïve mice. Transfection with SOCS1 significantly reduced the mortality of mice (Fig. 1D, P < 0.001). This shows that SOCS1 expression plays a role in the survival and/or proliferation of DA1-3b leukemic cells in vivo.

Deregulation of JAK/STAT pathways in dormant tumor cells. Because SOCS1 inhibits the JAK/STAT pathways, we measured the phosphorylation of STAT1, STAT3, STAT5, and STAT6 in dormant tumor cells after exposure to IL-2, IL-4, IL-5, IL-6, IL-7, IL-10, IL-12, IL-13, IL-15, GM-CSF, IFN-{gamma}, or TNF-{alpha}. In DA1-3b/d365 cells exposed to IFN-{gamma},STAT1 phosphorylation was increased compared with DA1-3b cells (Fig. 2 ). Transfection of a SOCS1 expression plasmid into the DA1-3b/d365 cells reversed this additional phosphorylation. IL-7 and TNF-{alpha}, IL-10, and GM-CSF induced additional phosphorylation of STAT3, STAT5, and STAT6, respectively, in dormant tumor cells (Supplementary Fig. S2).


Figure 2
View larger version (15K):
[in this window]
[in a new window]

  		Figure 2. STAT1 protein phosphorylation in dormant tumor cells. Flow-cytometric analysis of STAT1 phosphorylation in the following cells: DA1-3b, DA1-3b/d365, DA1-3b transfected with SOCS1 (DA1-3b-pB.SOCS1), and DA1-3b/365 transfected with SOCS1 (DA1-3b/d365-pB.SOCS1), with or without incubation with IFN-{gamma}.

 
Reduced sensitivity of dormant cells to apoptosis. We previously showed that dormant tumor cells are less sensitive to CTL-mediated lysis (11). Dormant tumor cells resist CTL-mediated lysis by overexpressing B7-H1 and B7.1, but blocking B7-H1 and B7.1 only partially restores their sensitivity. We hypothesized that SOCS1 extinction by promoter methylation might decrease sensitivity to CTL-mediated lysis and to apoptosis generally. Incubation in 5-azadC and TSA increased sensitivity of DA1-3b/d365 but not of DA1-3b cells (Fig. 3A–C ). A similar effect was obtained by transfection of a SOCS1 expression plasmid (Fig. 3D). Overexpression of SOCS1 in DA1-3b cells that already expressed SOCS1 also enhanced lysis. Preincubation with the JAK2 inhibitor AG490 increased sensitivity of DA1-3b/d365 cells but had little effect on DA1-3b cells (Fig. 4A ). In addition, preincubation of DA1-3b/d365 cells with IFN-{gamma}, but not DA1-3b (Fig. 4B) and DA1-3b/d365-SOCS1 cells decreased sensitivity. In DA1-3b/d365 cells, we verified that IFN-{gamma} did not induce B7-H1 expression that could inhibit CTLs. DA1-3b/d365 cells already expressed a high level of B7-H1 as previously reported, but IFN-{gamma} did not modify this expression (data not shown). Expression of the other costimulatory molecules B7.1, B7.2, B7-DC, and other cell surface molecules were not modified.


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Figure 3. CTL-mediated killing of dormant tumor cells. A, sensitivity of DA1-3b leukemic cells to CTL-mediated killing after incubation with 5-azadC and TSA. B, same experiment but with DA1-3b/d35 cells as a target. C, same experiment but with DA1-3b/d365 cells as a target. D, CTL-mediated killing of DA1-3b or DA1-3b/d365 cells transfected with SOCS1 (DA1-3b-pB.SOCS1 and DA1-3b/d365-pB.SOCS1) or with control empty vector (DA1-3b-pB and DA1-3b/d365-pB). All experiments were done in quadruplicate and repeated at least thrice. Points, mean; bars, SD.

 

Figure 4
View larger version (22K):
[in this window]
[in a new window]

  		Figure 4. JAK2 and proteasome inhibition in dormant tumor cells and CTL-mediated killing. A, sensitivity of DA1-3b leukemic cells and DA1-3b/d365 dormant tumor cells to CTL-mediated killing after incubation with AG490 JAK2 inhibitor. B, DA1-3b-pB, DA1-3b/SOCS1, DA1-3b/d365-pB, and DA1-3b/d365-SOCS1 cells were preincubated with 200 IU IFN-{gamma} for 1 h before CTL assay. C, DA1-3b-pB or DA1-3b-pB.SOCS1 cells were preincubated with MG132 proteasome inhibitor. D, same as (C), but with DA1-3b/d365-pB and DA1-3b/d365-pB.SOCS1 cells. All experiments were done in quadruplicate and repeated at least thrice. Points, mean; bars, SD.

 
These results suggest that extinction of SOCS1-induced resistance of dormant tumor cells is caused by deregulation of the JAK/STAT pathway. SOCS1 reportedly inhibits JAK proteins by targeting them to the proteasome pathway (21). Incubation of DA1-3b/d365 cells with the MG132 proteasome inhibitor had little effect on CTL-mediated lysis (Fig. 4C). However, DA1-3b/d365 cells transfected with a SOCS1 expression plasmid and DA1-3b cells showed a decrease in CTL-mediated lysis, suggesting that CTL-mediated killing requires SOCS1 expression (Fig. 4D).

To define the apoptotic pathway that CTL-mediated lysis induces in DA1-3b cells and DA1-3b/d365 cells, we blocked granzyme B using dichloroisocoumarin (Fig. 5A ), confirming that lysis involves the granzyme B/perforin pathway (11). The pan-caspase inhibitor z-VAD-fmk also blocked lysis (Fig. 5B). Detection of activated caspases by FITC-VAD-fmk showed reduced staining in dormant tumor cells when compared with DA1-3b cells (Fig. 5C).


Figure 5
View larger version (21K):
[in this window]
[in a new window]

  		Figure 5. Annexin V labeling, granzyme B, and caspase activities in dormant tumor cells exposed to CTL-mediated killing or therapeutic agents. A, sensitivity of DA1-3b/d365-pB or DA1-3b/d365-pB.SOCS1 leukemic cells to CTL-mediated killing after incubation with a granzyme B inhibitor, dichloroisocoumarin. B, same as (A), but after incubation with the pan-caspase inhibitor z-VAD-fmk. C, flow-cytometric analysis of caspase activity in DA1-3b and DA1-3b/d365 cells by FITC-VAD-fmk labeling. D, annexin V labeling in DA1-3b cells or DA1-3b/d365 cells exposed to specific CTLs, cytarabine, imatinib mesylate (10 µmol/L), and irradiation. All experiments were done in quadruplicate and repeated at least thrice. Points and columns, mean; bars, SD.

 
Recent studies showed that SOCS1 also regulates NF-{kappa}B signaling via p65/RelA (22, 23). As NF-{kappa}B activity tightly controls apoptosis, we preincubated DA1-3b, DA1-3b/d365, and DA1-3b/d365-SOCS1 cells with the NF-{kappa}B peptide inhibitor SN50 (24). No effect was seen on CTL-mediated lysis (Supplementary Fig. S3). We also determined p65/RelA binding activity on nuclear extracts from DA1-3b, DA1-3b/d365, and DA1-3b/d365-SOCS1 cells and found no differences, suggesting that reduced apoptosis in dormant tumor cells was not related to increased NF-{kappa}B activity (data not shown).

To determine if resistance to CTLs could also lead to drug resistance, we irradiated cells (100 Gy) or incubated them with 5 µmol/L 1-ß-D-arabinofuranosylcytosine or 10 µmol/L imatinib mesylate. Apoptosis, measured by annexin V labeling and the percentage of cells with sub-G1 DNA content, was reduced in dormant tumor cells (Fig. 5D; Supplementary Fig. S4). Their mitochondrial transmembrane potential was lost, and fewer ROS were generated (Fig. 6A ). z-VAD-fmk reduced all characteristics of apoptosis both in dormant tumor cells and in DA1-3b cells. Thus, dormant tumor cells had developed cross-resistance to caspase-dependent and mitochondrial-mediated apoptosis.


Figure 6
View larger version (23K):
[in this window]
[in a new window]

  		Figure 6. Apoptosis in dormant tumor cells and protective effects of IL-3 and CM. A, inner mitochondrial transmembrane potential and ROS measured by DIOC6 (3) and dihydroethidine-labeling, respectively, in DA1-3b cells or DA1-3b/d365 cells, exposed to specific CTLs, cytarabine, imatinib mesylate (10 µmol/L), and irradiation. B, annexin V labeling in DA1-3b-pB, DA1-3b/d365, or DA1-3b/d365-pB.SOCS1 cells treated with different doses of imatinib or control medium and after 1 h preincubation with IFN-{gamma}, SN50, or AG490. C, annexin V labeling of DA1-3b/cells exposed to imatinib mesylate, IL-3, anti–IL-3 blocking antibody (anti–IL-3 mAb), and CM from DA1-3b/d365 (d365-CM), and CTL activity against DA1-3b/d365 cells incubated with the anti–IL-3 mAb or control isotype. D, IL-3 dosage in CM from DA1-3b, DA1-3b/d365, and DA1-3b/d365-pB.SOCS1 cells. All experiments were done in quadruplicate and repeated at least thrice. Columns and points, mean; bars, SD.

 
We analyzed the effect of re-expressing SOCS1 on apoptosis induced by imatinib mesylate. DA1-3b/d365-SOCS1 cells showed increased apoptosis compared with DA1-3b/d365 cells, indicating that blocking SOCS1 may also lead to drug resistance (Fig. 6B). When the dose of imatinib was increased up to 20 µmol/L, we observed only a partial correction of sensitivity to apoptosis in DA1-3b/d365 cells. As observed with CTL-induced apoptosis, preincubation of DA1-3b/d365 cells with IFN-{gamma} reduced imatinib-induced apoptosis, and conversely, preincubation with the JAK2 inhibitor AG490 enhanced it (Fig. 6B). Addition of the NF-{kappa}B inhibitor SN50 had no significant effect.

Cross-resistance of dormant tumor cells through IL-3 overproduction. To test the hypothesis that silencing of SOCS1 would also modify the profile of cytokine produced by dormant tumor cells, which might then induce the cross-resistance to apoptosis of these cells, we first incubated DA1-3b cells with CM of DA1-3b/d365 (d365-CM) or DA1-3b/d365-pB.SOCS1 cells (d365-SOCS1-CM) and induced apoptosis with imatinib. CM from DA1-365 dormant tumor cells but not from DA1-3b/365-pBSOCS1 cells induced imatinib resistance in DA1-3b/cells (Fig. 6C; Supplementary Fig. S5B). Dosage of cytokines in the CM of DA1-3b, DA1-3b/d365, and DA1-3b/d365-pBSOCS1 cells did not reveal any production of IL-2, IL-4, IL-10, IL-12, IL-15, IFN-{gamma}, TNF-{alpha}, G-CSF, or GM-CSF. However, dormant tumor cells produced more IL-3 and IL-6 than DA1-3b and DA1-3b/d365-SOCS1 cells (Fig. 6D; Supplementary Fig. S5A). Adding recombinant IL-6 on DA1-3b had a modest effect on sensitivity to imatinib and did not add to the inhibitory effect of d365-CM (Supplementary Fig. S5B). However, adding recombinant IL-3 almost completely inhibited imatinib- and CTL-induced apoptosis (Fig. 6C; Supplementary Fig. S6). Moreover, addition of an anti–IL-3 antibody blocked the protective effect of DA1-3b/d365 CM on DA1-3b cells and restored the sensitivity of DA1-3b/365 dormant tumor cells to CTL-mediated lysis (Fig. 6C). Thus, the cross-resistance to apoptosis of dormant tumor cells is mediated through IL-3 overproduction induced by SOCS1 silencing.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Tumor dormancy results from the interaction between the host and tumor cells. In several experimental models, the immune response controls dormancy (1, 2, 8, 11, 17, 25–30). We previously observed in our mouse model of acute myeloid leukemia that dormant tumor cells progressively express more B7-H1 and B7.1, and that these molecules reduce CD8+ CTL-mediated lysis via interaction between B7.1 and CTLA-4 and between B7-H1 with a still unknown receptor (11). In other model systems, memory T cells or specific antibodies interact with dormant tumor cells (1, 8, 25, 31). These experimental models may be relevant for patients. For example, in a long-term survivor of melanoma, immune response shifted to escape variants of melanoma cells, after initial loss of antigen presentation by tumor cells, offering an example of both immune editing and immune adaptation, and resulting in equilibrium (32). Thus, dormant tumor cells may be selected by immune pressure. Here, we observed that dormant tumor cells progressively methylate the SOCS1 gene, ultimately silencing expression. Methylation of the SOCS1 gene has been reported in several human tumors, including lymphoid and myeloid hematologic malignancies (33–44). Our findings suggest a possible role in tumor dormancy. We showed a progressive increase of methylated SOCS1 DNA and a decrease of SOCS1 expression, suggesting that methylation confers a selective but slow growing advantage.

SOCS1 inhibits JAK kinase activities by direct interaction with the JAK activation loop and also targets JAKs to ubiquitin-mediated proteasomal degradation. In our model, silencing of SOCS1 increased phosphorylation of STAT proteins after exposure to cytokines. This suggests that in vivo dormant tumor cells, which are continuously exposed to various cytokines, could have enhanced JAK/STAT activity. To examine the consequence of this deregulation of JAK/STAT pathways, we measured the sensitivity of dormant tumor cells to CD8+ CTL-mediated lysis. As previously reported, dormant tumor cells showed reduced sensitivity (11). However, forcing the re-expression of SOCS1 by a demethylating agent and histone deacetylase inhibitor, or by transfection with SOCS1 cDNA, enhanced CTL-mediated lysis. Preincubation in AG490, an inhibitor of JAK2, also enhanced CTL-mediated lysis of dormant tumor cells, but preincubation in IFN-{gamma} decreased lysis. IFN-{gamma} can act via the JAK/STAT pathway and can be regulated by SOCS proteins. SOCS1 inhibits the JAK/STAT pathway in part by targeting JAK proteins to the proteasome. Blocking proteasomes with MG132 inhibited CTL-mediated lysis of DA1-3b cells and DA1-3b/d365 dormant tumor cells transfected with SOCS1. However, in dormant tumor cells that did not express SOCS1, the inhibitor had little effect. These data may suggest that sensitivity of tumor cells to CTLs depends on JAK/STAT pathway regulation by SOCS1, perhaps mediated via proteosomal degradation. In dormant tumor cells, extinction of SOCS1 regulation protects tumor cells from CTLs.

Constitutive activation of the JAK/STAT pathway is observed in many tumor types and involves many mechanisms, including mutation of cell surface receptors like FLT3 or KIT in AML; methylation of SOCS1, SOCS3, SHP1; TEL-JAK2 translocation in AML; and activation via BCR/ABL. Direct inactivation of SOCS1 by hepatitis C virus in hepatocytes, or by phosphorylation of SOCS1 protein by V-ABL that results in JAK/STAT deregulation, has also been reported (35, 45). JAK/STAT constitutive activation is involved in many characteristics of tumor cells, including invasion, metastasis, drug resistance, and proliferation. Constitutive activation could also contribute to the escape from immune surveillance and, thus, to the equilibrium between host and dormant tumor cells.

In addition to its regulatory role on the JAK/STAT pathway, SOCS1 has been reported to negatively regulate NF-{kappa}B via p65/RelA (22, 23). However, we did not find enhanced NF-{kappa}B activity in dormant tumor cells in vitro, but this does not rule out a possible deregulation of NF-{kappa}B in vivo when cells are exposed to an appropriate microenvironment.

Tumor dormancy is presumed to result from several mechanisms, including control of persistent tumor cells by host immune response, and modification of angiogenesis. Transient inactivation of oncogenes like MYC may also result in tumor cell differentiation and persistence of normal-appearing cells that recover their tumorigenic potential following oncogene reactivation (46, 47). Tumor dormancy in these models is induced by a loss of tumorigenicity. In contrast, our model shows a gain of function that results in more aggressive tumor cells, and the observed progressive methylation of SOCS1 suggests that persistent cells divide in vivo. These two models of tumor dormancy may not be mutually exclusive. Tumor cells may become dormant in vivo after oncogene inactivation that results in tumor cell differentiation. Deregulation of the JAK/STAT pathway and active suppression of T cells may be necessary for dormant tumor cells to resist apoptosis induced by CTLs or other factors.

Dormant tumor cells showed a decreased sensitivity to apoptosis induced by CTLs. CTLs, irradiation, cytarabine, and imatinib mesylate induced apoptosis via a caspase-dependent, mitochondria-mediated mechanism. Apoptosis was reduced when dormant tumor cells were exposed to these drugs. However, apoptosis increased when SOCS1 was re-expressed, and exposure of these cells to IFN{gamma} further decreased their sensitivity to imatinib. SOCS1 silencing induced an IL-3 overexpression in dormant tumor cells that induced the resistance of these cells to imatinib and CTL-mediated lysis, and anti–IL-3 blocking antibody was able to reverse this cross-resistance mechanism. We observed that CM of dormant tumor cells transferred the resistance to DA1-3b leukemic cells, suggesting also that a paracrine-mediated resistance effect may occur in vivo. We also found an overexpression of IL-6 induced by SOCS1 silencing. IL-6 did not mediate significant resistance to apoptosis. However, we cannot rule out that this overexpression has a role in vivo to allow persistence of dormant tumor cells. Thus, even if dormant tumor cells have only been selected by immune pressure, they develop a cross-resistance through an autocrine feedback loop both to immune response and to drugs. An IL-3 autocrine growth and survival loop has been observed in primary human chronic myelogenous leukemia CD34+ cells (48). Adaptive secretion of GM-CSF that mediate imatinib and nilotinib resistance through activation of the JAK2/STAT5 pathway in BCR/ABL-positive progenitors has recently been reported (49). In this study, CM from the LAMA-84 imatinib-resistant cell line induced resistance in imatinib-sensitive cells, and this effect was abrogated with AG490 and an anti–GM-CSF blocking antibody. AG490 has also been reported as an efficient inducer of apoptosis in imatinib-resistant cells (50). Thus, the cross-resistance to CTL and imatinib in BCR/ABL+ dormant tumor cells and resistance to imatinib in BCR/ABL+ progenitors may share common mechanisms.

These findings suggest that long-term persistence of dormant tumor cells may by itself generate more aggressive subclones, which could escape from immune response after years of persistence. It might also lead to resistance of relapsing tumors to various drugs. The resistance of dormant tumor cells to imatinib mesylate in our model may suggest that interaction between CTLs and BCR/ABL+ long-term persistent leukemic cells would lead to spontaneous resistance to tyrosine kinase inhibitors. The finding that, in addition to the IL-3 autocrine loop, in vitro exposure of these cells to IFN{gamma} can also decrease their sensitivity to imatinib further supports this hypothesis. New treatments that target the mechanisms that cause resistance would eradicate these persistent cells before they cause relapse.


   Acknowledgments
 
Grant support: Cancéropole Nord-Ouest, the Ligue Contre le Cancer (Comité du septentrion), the Association de Recherche sur le Cancer, the Groupement des Entreprises Françaises dans la Lutte Contre le Cancer, and the Fondation de France. A. Saudemont was recipient of a grant from the Region Nord-Pas-de-Calais/CHRU de Lille and is currently supported by Grant Research United Kingdom.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Received 5/ 3/06. Revised 2/ 3/07. Accepted 2/14/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Uhr JW, Marches R. Dormancy in a model of murine B cell lymphoma. Semin Cancer Biol 2001;11:277–83.[CrossRef][Medline]
  2. Davis TA, Maloney DG, Czerwinski DK, Liles TM, Levy R. Anti-idiotype antibodies can induce long-term complete remissions in non-Hodgkin's lymphoma without eradicating the malignant clone. Blood 1998;92:1184–90.[Abstract/Free Full Text]
  3. Chu S, Xu H, Shah NP, et al. Detection of BCR-ABL kinase mutations in CD34+ cells from chronic myelogenous leukemia patients in complete cytogenetic remission on imatinib mesylate treatment. Blood 2005;105:2093–8.[Abstract/Free Full Text]
  4. Holtz MS, Forman SJ, Bhatia R. Nonproliferating CML CD34+ progenitors are resistant to apoptosis induced by a wide range of proapoptotic stimuli. Leukemia 2005;19:1034–41.[CrossRef][Medline]
  5. Bhatia R, Holtz M, Niu N, et al. Persistence of malignant hematopoietic progenitors in chronic myelogenous leukemia patients in complete cytogenetic remission following imatinib mesylate treatment. Blood 2003;101:4701–7.[Abstract/Free Full Text]
  6. Allard WJ, Matera J, Miller MC, et al. Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with nonmalignant diseases. Clin Cancer Res 2004;10:6897–904.[Abstract/Free Full Text]
  7. Meng S, Tripathy D, Frenkel EP, et al. Circulating tumor cells in patients with breast cancer dormancy. Clin Cancer Res 2004;10:8152–62.[Abstract/Free Full Text]
  8. Vitetta ES, Tucker TF, Racila E, et al. Tumor dormancy and cell signaling. V. Regrowth of the BCL1 tumor after dormancy is established. Blood 1997;89:4425–36.[Abstract/Free Full Text]
  9. Saudemont A, Buffenoir G, Denys A, et al. Gene transfer of CD154 and IL12 cDNA induces an anti-leukemic immunity in a murine model of acute leukemia. Leukemia 2002;16:1637–44.[CrossRef][Medline]
  10. Vereecque R, Buffenoir G, Preudhomme C, et al. Gene transfer of GM-CSF, CD80 and CD154 cDNA enhances survival in a murine model of acute leukemia with persistence of a minimal residual disease. Gene Ther 2000;7:1312–6.[CrossRef][Medline]
  11. Saudemont A, Quesnel B. In a model of tumor dormancy, long-term persistent leukemic cells have increased B7-1 and B7.1 expression and resist CTL-mediated lysis. Blood 2004;104:2124–33.[Abstract/Free Full Text]
  12. Alexander WS, Hilton DJ. The role of suppressors of cytokine signaling (SOCS) proteins in regulation of the immune response. Annu Rev Immunol 2004;22:503–29.[CrossRef][Medline]
  13. Kile BT, Schulman BA, Alexander WS, Nicola NA, Martin HM, Hilton DJ. The SOCS box: a tale of destruction and degradation. Trends Biochem Sci 2002;27:235–41.[CrossRef][Medline]
  14. Rottapel R, Ilangumaran S, Neale C, et al. The tumor suppressor activity of SOCS-1. Oncogene 2002;21:4351–62.[CrossRef][Medline]
  15. Frantsve J, Schwaller J, Sternberg DW, Kutok J, Gilliland DG. Socs-1 inhibits TEL-JAK2–mediated transformation of hematopoietic cells through inhibition of JAK2 kinase activity and induction of proteasome-mediated degradation. Mol Cell Biol 2001;21:3547–57.[Abstract/Free Full Text]
  16. Kamizono S, Hanada T, Yasukawa H, et al. The SOCS box of SOCS-1 accelerates ubiquitin-dependent proteolysis of TEL-JAK2. J Biol Chem 2001;276:12530–8.[Abstract/Free Full Text]
  17. Saudemont A, Jouy N, Hetuin D, Quesnel B. NK cells that are activated by CXCL10 can kill dormant tumor cells that resist CTL-mediated lysis and can express B7-1 that stimulates T cells. Blood 2005;105:2428–35.
  18. Lo YM, Wong IH, Zhang J, Tein MS, Ng MH, Hjelm NM. Quantitative analysis of aberrant p16 methylation using real-time quantitative methylation-specific polymerase chain reaction. Cancer Res 1999;59:3899–903.[Abstract/Free Full Text]
  19. Ilangumaran S, Ramanathan S, Ning T, et al. Suppressor of cytokine signaling 1 attenuates IL-15 receptor signaling in CD8+ thymocytes. Blood 2003;102:4115–22.[Abstract/Free Full Text]
  20. Marchetti P, Castedo M, Susin SA, et al. Mitochondrial permeability transition is a central coordinating event of apoptosis. J Exp Med 1996;184:1155–60.[Abstract/Free Full Text]
  21. Zhang JG, Farley A, Nicholson SE, et al. The conserved SOCS box motif in suppressors of cytokine signaling binds to elongins B and C and may couple bound proteins to proteasomal degradation. Proc Natl Acad Sci U S A 1999;96:2071–6.[Abstract/Free Full Text]
  22. Mansell A, Smith R, Doyle SL, et al. Suppressor of cytokine signaling 1 negatively regulates Toll-like receptor signaling by mediating Mal degradation. Nat Immunol 2006;7:148–55.[CrossRef][Medline]
  23. Ryo A, Suizu F, Yoshida Y, et al. Regulation of NF-{kappa}B signaling by Pin1-dependent prolyl isomerization and ubiquitin-mediated proteolysis of p65/RelA. Mol Cell 2003;12:1413–26.[CrossRef][Medline]
  24. Lin YZ, Yao SY, Veach RA, Torgerson TR, Hawiger J. Inhibition of nuclear translocation of transcription factor NF-{kappa}B by a synthetic peptide containing a cell membrane-permeable motif and nuclear localization sequence. J Biol Chem 1995;270:14255–8.[Abstract/Free Full Text]
  25. Mahnke YD, Schwendemann J, Beckhove P, Schirrmacher V. Maintenance of long-term tumour-specific T-cell memory by residual dormant tumour cells. Immunology 2005;115:325–36.[CrossRef][Medline]
  26. Schirrmacher V. T-cell immunity in the induction and maintenance of a tumour dormant state. Semin Cancer Biol 2001;11:285–95.[CrossRef][Medline]
  27. Farrar JD, Katz KH, Windsor J, et al. Cancer dormancy. VII. A regulatory role for CD8+ T cells and IFN-{gamma} in establishing and maintaining the tumor-dormant state. J Immunol 1999;162:2842–9.[Abstract/Free Full Text]
  28. Yefenof E, Picker LJ, Scheuermann RH, Tucker TF, Vitetta ES, Uhr JW. Cancer dormancy: isolation and characterization of dormant lymphoma cells. Proc Natl Acad Sci U S A 1993;90:1829–33.[Abstract/Free Full Text]
  29. Muller M, Gounari F, Prifti S, Hacker HJ, Schirrmacher V, Khazaie K. EblacZ tumor dormancy in bone marrow and lymph nodes: active control of proliferating tumor cells by CD8+ immune T cells. Cancer Res 1998;58:5439–46.[Abstract/Free Full Text]
  30. Khazaie K, Prifti S, Beckhove P, et al. Persistence of dormant tumor cells in the bone marrow of tumor cell-vaccinated mice correlates with long-term immunological protection. Proc Natl Acad Sci U S A 1994;91:7430–4.[Abstract/Free Full Text]
  31. Schirrmacher V, Beckhove P, Choi C, Griesbach A, Mahnke Y. Tumor-immune memory T cells from the bone marrow exert GvL without GvH reactivity in advanced metastasized cancer. Int J Oncol 2005;27:1141–9.[Medline]
  32. Yamshchikov GV, Mullins DW, Chang CC, et al. Sequential immune escape and shifting of T cell responses in a long-term survivor of melanoma. J Immunol 2005;174:6863–71.[Abstract/Free Full Text]
  33. Reddy J, Shivapurkar N, Takahashi T, et al. Differential methylation of genes that regulate cytokine signaling in lymphoid and hematopoietic tumors. Oncogene 2005;24:732–6.[CrossRef][Medline]
  34. Galm O, Wilop S, Reichelt J, et al. DNA methylation changes in multiple myeloma. Leukemia 2004;18:1687–92.[CrossRef][Medline]
  35. Yoshida T, Ogata H, Kamio M, et al. SOCS1 is a suppressor of liver fibrosis and hepatitis-induced carcinogenesis. J Exp Med 2004;199:1701–7.[Abstract/Free Full Text]
  36. Sutherland KD, Lindeman GJ, Choong DY, et al. Differential hypermethylation of SOCS genes in ovarian and breast carcinomas. Oncogene 2004;23:7726–33.[CrossRef][Medline]
  37. Galm O, Yoshikawa H, Esteller M, Osieka R, Herman JG. SOCS-1, a negative regulator of cytokine signaling, is frequently silenced by methylation in multiple myeloma. Blood 2003;101:2784–8.[Abstract/Free Full Text]
  38. Chen CY, Tsay W, Tang JL, et al. SOCS1 methylation in patients with newly diagnosed acute myeloid leukemia. Genes Chromosomes Cancer 2003;37:300–5.[CrossRef][Medline]
  39. Melzner I, Moller P. Silencing of the SOCS-1 gene by CpG methylation? Blood 2003;102:1554–5.[Free Full Text]
  40. Zardo G, Tiirikainen MI, Hong C, et al. Integrated genomic and epigenomic analyses pinpoint biallelic gene inactivation in tumors. Nat Genet 2002;32:453–8.[CrossRef][Medline]
  41. Nagai H, Kim YS, Konishi N, et al. Combined hypermethylation and chromosome loss associated with inactivation of SSI-1/SOCS-1/JAB gene in human hepatocellular carcinomas. Cancer Lett 2002;186:59–65.[CrossRef][Medline]
  42. Depil S, Saudemont A, Quesnel B. SOCS-1 gene methylation is frequent but does not appear to have prognostic value in patients with multiple myeloma. Leukemia 2003;17:1678–9.[CrossRef][Medline]
  43. Johan MF, Bowen DT, Frew ME, Goodeve AC, Reilly JT. Aberrant methylation of the negative regulators RASSFIA, SHP-1 and SOCS-1 in myelodysplastic syndromes and acute myeloid leukaemia. Br J Haematol 2005;129:60–5.[CrossRef][Medline]
  44. Chim CS, Fung TK, Cheung WC, Liang R, Kwong YL. SOCS1 and SHP1 hypermethylation in multiple myeloma: implications for epigenetic activation of the Jak/STAT pathway. Blood 2004;103:4630–5.[Abstract/Free Full Text]
  45. Miyoshi H, Fujie H, Shintani Y, et al. Hepatitis C virus core protein exerts an inhibitory effect on suppressor of cytokine signaling (SOCS)-1 gene expression. J Hepatol 2005;43:757–63.[CrossRef][Medline]
  46. Shachaf CM, Felsher DW. Tumor dormancy and MYC inactivation: pushing cancer to the brink of normalcy. Cancer Res 2005;65:4471–4.[Abstract/Free Full Text]
  47. Shachaf CM, Kopelman AM, Arvanitis C, et al. MYC inactivation uncovers pluripotent differentiation and tumour dormancy in hepatocellular cancer. Nature 2004;431:1112–7.[CrossRef][Medline]
  48. Jiang X, Lopez A, Holyoake T, Eaves A, Eaves C. Autocrine production and action of IL-3 and granulocyte colony-stimulating factor in chronic myeloid leukemia. Proc Natl Acad Sci U S A 1999;96:12804–9.[Abstract/Free Full Text]
  49. Wang Y, Cai D, Brendel C, et al. Adaptive secretion of the granulocyte macrophage colony stimulating factor (GM-CSF) mediates imatinib- and nilotinib-resistance in BCR/ABL-positive progenitors via JAK-2/STAT-5 pathway activation. Blood 2006. Epub ahead of print.
  50. Samanta AK, Lin H, Sun T, Kantarjian H, Arlinghaus RB. Janus kinase 2: a critical target in chronic myelogenous leukemia. Cancer Res 2006;66:6468–72.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saudemont, A.
Right arrow Articles by Quesnel, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saudemont, A.
Right arrow Articles by Quesnel, B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online