Induction of class II major histocompatibility complex expression in human multiple myeloma cells by retinoid

Abstract

Class II major histocompatibility complex (MHC II) is normally silenced in plasma/multiple myeloma (MM) cells at the transcriptional level through downregulation of class II transactivator (CIITA), allowing MM cells to escape from immunological responses. Here we demonstrate that a retinoic acid receptor-α/β-selective retinoid Am80 (tamibarotene) could induce the expression of functional MHC II molecules in human MM cell lines. Am80 upregulated expression of the interferon regulatory factor-1 gene, followed by enhancement of CIITA expression. This is the first report demonstrating that retinoid can induce the expression of MHC II in terminally-differentiated plasma/MM cells.

Multiple myeloma (MM) is an intractable B-cell malignancy characterized by clonal proliferation of terminally-differentiated plasma cells in bone marrow.1,2 MM develops as a result of multistep tumorigenesis caused by chromosomal and genetic alterations.2 In addition to the low effectiveness of chemotherapy in MM, immunologic responses against MM cannot be detected because most MM cells do not express class II major histocompatibility complex (MHC II), allowing them to escape from direct recognition by idiotype-specific CD4- positive T cells.3,4 During normal B-cell differentiation to plasma cells, MHC II expression is silenced at the transcriptional level through downregulation of a transcriptional regulator, class II transactivator (CIITA).3,5 Transcription of CIITA is regulated by four distinct promoters, and promoter III has a binding site for a transcriptional repressor, positive regulatory domain I-binding factor 1/B-lymphocyte-induced maturation protein-1 (PRDI-BF1/Blimp-1), the expression of which is induced during terminal B-cell differentiation into plasma cells.6 Thus, stage-specific gene expression of PRDI-BF1 leads to silencing of MHC II expression in plasma cells. Promoter IV has binding sites for interferon-inducible factors such as interferon-regulatory factor-1 (IRF-1) and signal transducing activators of transcription 1 (STAT1), and it regulates interferon- associated expression of CIITA.7,8

Retinoids such as all trans-retinoic acid (ATRA) have been reported to inhibit the cell growth of a range of malignancies.9,10 We have previously demonstrated that a synthetic retinoid Am80 (tamibarotene) could inhibit MM cell growth through upregulation of p21 protein and downregulation of interleukin (IL)-6/IL-6 receptor.10 Am80 also inhibited angiogenesis in vitro and in vivo.10 Am80 is a retinoic acid receptor (RAR)-α/β-selective retinoid, which does not activate RAR-γ or retinoid X receptors (RXR), thus avoiding unfavorable adverse effects.11 In addition, Am80 has low affinity for cellular retinoic acid-binding protein (CRABP) and is active against CRABP-rich ATRA-resistant cells. In fact, Tobita et al. have reported that 14 of 24 patients with acute promyelocytic leukemia who relapsed after ATRA treatment attained complete remission following Am80 treatment and experienced fewer adverse effects.12 Thus, Am80 has additional therapeutic advantages compared to those of ATRA. In the present study, we performed gene expression profile analysis to examine the effects of Am80 on a human MM cell line.

Design and Methods

MHC II-negative human MM cell lines, U266, ILKM-2, ILKM-3, ILKM-8, NOP1, XG7, FR4, SK-MM1, SACHI, NCUMM-2 and NCUMM-3, a B-cell lymphoma cell line, Daudi, and a CD4-positive T-cell leukemia cell line, Jurkat, were used in this study.10,13,14 Retinoids were prepared as described previously.10 A cDNA microarray analysis was performed using IntelliGene HS Human expression chips containing 16,600 probes (Takara).15 Briefly, U266 cells were or were not treated 1 μM Am80 for 24 h, conditions in which Am80 shows approximately 5–10% growth inhibition on U266 cells as previously reported by us.10 After incubation, total RNA was purified and subjected to a microarray analysis. The results are presented as the ratio of gene expression in the Am80-treated sample to that in the untreated control (T/C ratio). The genes with a T/C ratio of ≥2.00 or ≤0.50 were selected. To validate the results of gene expression analysis, we performed quantitative reverse transcriptase polymerase chain reaction (RT-PCR) using Taqman Gene Expression Assays and the AB7300 System (Applied Biosystems). The expression value of each gene was normalized to that of β-actin.

To analyze MHC II expression, the cells were incubated with staining buffer containing the fluorescein isothiocyanate- conjugated mouse monoclonal antibody which can cross-react with HLA-DP, -DQ and -DR, or isotype-matched control antibody (BD Biosciences), followed by flow cytometry as described elsewhere.14 For interaction analysis between T-cell receptor (TCR) and MHC II, the reporter plasmid, 4κB-Luc,16 containing four nuclear factor-κB (NF-κB) binding sites, and the internal control plasmid, pRL-TK, expressing Renilla luciferase, were transfected into Jurkat cells using Nucleofector (Amaxa Bio systems). Twenty-four hours after transfection, Jurkat cells were cocultured with U266 or Daudi cells, which had or had not been pre-treated with 1 μM Am80 for 5 days. The cells were left unstimulated or stimulated with 300 ng/mL of staphylococcal toxin SEE (Toxin Technology). After 24 h incubation, luciferase activity was measured by a luminometer as described previously.16 In this system, when TCR signaling occurs, luciferase activity in Jurkat cells is induced through NF-κB activation.17

Results and Discussion

In order to examine the effects of Am80 on MM cells, we first performed gene expression profile analysis using a cDNA microarray containing 16,600 human genes. Table 1 shows the representative results of gene expression profiles after treatment of U266 cells with Am80. Am80 upregulated p21 expression and downregulated IL-6 expression, supporting the previous results.10 In addition, Am80 upregulated gene expressions of MHC II molecules including HLADRA, HLADMB, CD74, HLADQB1, HLADMA, HLADPA1 and HLADRB5, which prompted us to examine the effect of Am80 on the cell surface expression of MHC II. We used 11 MHC II-negative MM cell lines and a control MHC II-positive B-cell line, Daudi. As shown in Figure 1A, U266 cells did not express MHC II molecules on their surface. Am80 treatment induced the expression of these molecules in a time-dependent manner. On the other hand, in Daudi cells, constitutive expression of MHC II was detected but not modified by Am80 treatment. Among the MM cell lines examined, distinct induction of MHC II was observed in U266 and ILKM-2 cells (Figure 1B), both of which are IL-6-dependent cell lines. We then compared the effect of Am80 with that of other retinoids including ATRA and 9-cis retinoic acid (9-cis RA), which can bind to RAR-γ and RXR in addition to RAR-α/β. As shown in Figure 1C, ATRA and 9-cis RA also induced MHC II expression. There was no significant difference among these retinoids, suggesting that RARα/β is important for the induction of MHC II.

Figure 1.Induction of functional MHC II molecules by Am80. A. Induction of MHC II expression by Am80. U266 and Daudi cells were or were not treated with 1 μM Am80 for up to 5 days. MHC II expression on the cell surface was analyzed by flowcytometry using the monoclonal antibody which can cross-react with HLA-DP, -DQ and –DR. Each histogram shows the result of 20,000 viable cells counted. N.C., negative control stained with isotype-matched IgG. Control, untreated control. B. Effect of Am80 on induction of MHC II expression in a variety of MHC-negative MM cell lines. Eleven MHC II-negative MM cell lines were or were not treated with 1 μM Am80 for 5 days. The results are indicated as the percentage of MHC II-positive cells. Open bars and closed bars represent untreated control and Am80-treated samples, respectively. C. Induction of MHC II expression by retinoids. U266 cells were treated with Am80, ATRA or 9-cis retinoic acid (9-cis RA) for 5 days. MHC II expression on the cell surface was analyzed by flowcytometry. D. Functional interaction between MHC II-expressing U266 and Jurkat cells. The luciferase reporter plasmid containing four NF-κB-binding sites and the internal control plasmid were transfected into Jurkat cells, which were co-cultured with Daudi or U266 cells that had or had not been pretreated with 1 μM Am80 for 5 days. A superantigen SEE was added to the culture medium. The luciferase activity was normalized to that of an internal control and the results are indicated as fold-change compared with the activity of single-cultured Jurkat cells. The experiment was done in triplicate; column, mean; bars, standard deviation.

Table 1.Upregulated/downregulated genes after treating U266 cells with Am80.

Next, we examined whether the MHC II molecules induced by Am80 can functionally interact with TCR. We transfected the luciferase reporter plasmid into a CD4-positive T-cell line, Jurkat, and co-cultured them with U266 or Daudi cells, which were or were not pretreated with Am80. In order to bridge MHC II and TCR, the staphylococcal toxin SEE, a superantigen, was added. As shown in Figure 1D, in the presence of Daudi cells, luciferase activity was increased and driven by addition of SEE, whereas U266 cells could not stimulate luciferase activity without pre-treatment with Am80. Importantly, luciferase activity of U266 cells pre-treated with Am80 was increased by the addition of SEE. These results demonstrate that Am80 can induce the expression of functional MHC II molecules in U266 cells.

We then examined the effect of Am80 on CIITA gene expression to understand the mode of action of Am80 on MHC II induction. We used U266, Daudi and a MM cell line XG7, in which MHC II expression could not be induced by Am80 (Figure 1B). As shown in Figure 2A, PRDI-DF1 was highly expressed in U266 and XG7 cells, and CIITA expression was suppressed. Although Am80 treatment did not modify the expression of PRDI-DF1, Am80 induced a 4-fold increase in CIITA expression in U266 cells, but not in XG7 cells. In contrast, PRDI-DF1 was completely silenced in Daudi cells, and CIITA was highly expressed. Am80 treatment did not upregulate CIITA expression in Daudi cells. ATRA and 9-cis RA could induce expression of IRF-1 and CIITA in U266 cells (data not shown).

Figure 2.Induction of expression of IRF-1 and CIITA by Am80. A. The effect of Am80 on expression of the IRF-1, CIITA, PRDI-DF1 and STAT1 genes. U266, XG7 and Daudi cells were treated with Am80 (0-10 μM) for 24 h. Total RNA was isolated and subjected to quantitative RT-PCR using primers for IRF-1, STAT1, PRDI-DF1 or CIITA. Expression levels were normalized to the expression level of β-actin and indicated as fold-change compared to the expression of the untreated control. Experiments were done in triplicate; column, mean; bars, standard deviation. B. Temporal profile of gene expression after treatment with Am80. U266 cells were treated with 1 μM of Am80 for the indicated times (0–24 h).

We then examined the effect of Am80 on expression of the interferon-inducible genes such as IRF-1 and STAT1. In U266, but not XG7 or Daudi cells, Am80 induced a 1.6-fold increase in IRF-1 expression, whereas Am80 had no effect on STAT1 expression in either cell line (Figure 2A). The temporal profile of gene expression by U266 cells showed upregulation of IRF-1 at 1 h after the treatment with Am80 and upregulation of CIITA from 4 h after treatment (Figure 2B). IRF-1 has been reported to be induced by retinoid at the transcriptional level through the binding of RAR to the IRF-1 promoter.18,19 Although it remains unclear why Am80 could not induce the expression of IRF-1 in XG7 cells, these findings suggest that Am80 directly enhances the expression of IRF-1, which is followed by secondary induction of CIITA expression without changing the PRDI-DF1.

In conclusion, Am80 has an immunomodulatory effect in addition to anti-proliferative and anti-angiogenic activities. Although Am80 did not induce MHC II expression in all MM cell lines, it might enhance the immunogenicity of MM cells in vivo through the induction of MHC II expression. Since Am80 appears to be a safe and practical agent, which has many therapeutic advantages over other retinoids, it could be used as a chemopreventive agent from the earliest stage of MM.

Footnotes

• This work was supported in part by grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology, and the Ministry of Health, Labor and Welfare of Japan.
• We thank Drs. H. Kagechika (Tokyo Medical and Dental University, Tokyo, Japan) and K. Shudo (Research Foundation ITSUU Laboratory, Tokyo, Japan) for the kind gift of Am80.
• Author Contributions SI: directed this study and performed several experiments; SK: performed microarray analysis: RU: directed this study.
• Conflict of Interest The authors reported no potential conflicts of interest.

• Received May 18, 2006.
• Accepted August 3, 2006.