多发性骨髓瘤基因修饰瘤苗诱导体内抗肿瘤反应的实验研究

　　作者：任素萍,王立生,郭强,王华,贾向旭　　作者单位：北京放射与辐射医学研究所实验血液学研究室，北京 100850;1北京宣武医院血液科，北京 100053;2空军总医院血液科，北京100036

　　【摘要】本研究目的是评价NOD/SCID小鼠皮下移植瘤模型对多发性骨髓瘤基因修饰瘤苗引发体内抗肿瘤反应的效果。首先给NOD/SCID小鼠腹腔注射人外周血淋巴细胞以在其体内重建人的免疫系统，然后皮下接种γ-射线灭活的基因修饰骨髓瘤细胞sko-007(表达绿色荧光蛋白或者p53、GM-CSF和B7-1基因)，以PBS作为对照，最后植入活sko-007细胞进行攻击。结果发现，与对照组相比接种感染腺病毒Ad-p53/GM-CSF/B7-1的sko-007 细胞可以明显抑制移植瘤生长，病理分析显示移植瘤纤维组织增生伴弥漫性坏死增多，血管增生显著。免疫组织化学染色显示瘤灶内有人T淋巴细胞浸润。结论： p53、GM-CSF和B7-1基因修饰的骨髓瘤细胞能够诱导产生抗肿瘤免疫反应，有可能用于人类多发性骨髓瘤的免疫治疗。

　　【关键词】 多发性骨髓瘤

　　Multiple myeloma (MM) remains an incurable malignancy despite advances in chemotherapy. Myeloablative chemotherapy followed by allogeneic or autologous hematopoietic stem cell transplantation has increased the incidence of complete remission，but almost all patients achieving complete remission ultimately experience relapse[1-3]. Because these chemotherapies have only limited value，alternative strategies are needed to solve these problems. Immunotherapy may represent a means of maintaining complete remission.

　　Based on the fact that myeloma cells contain a multitude of tumor antigens that can effectively stimulate antitumor T cells，several investigators have repor- ted immunotherapeutic approaches via inoculating mye-

　　This project was supported in part by Chinese National Basic Research and Development Grants ’973’ (No. 2004CB518801 and No. 2002 CB713804)，Chinese High-Tech Program ’863’ (No. 2003AA216050) and Chinese National Science Foundation (No.30400189).

　　Corresponding author: WU Chu-Tse (吴祖泽)，Academician of Chinese Academy of Sciences，Professor of Department of Experimental Hematology，Beijing Institute of Radiation Medicine.

　　loma cell lysates，whole myeloma cells or genetically modified myeloma cell vaccines to augment the immunogenicity of myeloma cells. Numerous applications of genes encoding tumor suppressive proteins，cytokines and costimulatory molecules have been proposed in cancer (including MM) therapy[4-7]. Although the results of most preclinical investigations carried out in vitro and in mouse models were encouraging，the clinical evaluation is less satisfactory[8，9]. Further studies found that rather than the absence of tumor-specific antigen，myeloma cells escape from immune surveillance mainly by down-regulating the expression of costimulatory molecules and inhibiting induction and maturation of dendritic cells (DCs)[6，10-11]. Hence a combination of immune-stimulating genes should be more efficient than any single gene. In previous study，we have de-monstrated that whole myeloma cell vaccination co-transferred with human wild-type p53，GM-CSF and B7-1 genes mediated by recombinant adenovirus Ad-p53/GM-CSF/B7-1 induces allologous and autologous specific anti-tumor cytotoxicity in vitro[12]. In this study we investigated whether the generation of Ad-p53/GM-CSF/B7-1-mediated immunity is protective against subsequent tumor challenge.

　　NOD/SCID mice are the most immunodeficient of the SCID variants，and are the most supportive host for human stem cells[13]. Most SCID mouse myeloma models have used mouse or human MM cell lines. In 2 studies，fresh BMC from MM patient survived in conventional SCID mice，and in one group it was observed that human MM cell lines colonized in human fetal bone implanted subcutaneously in SCID mice[14-16]. However，all the mice myeloma models could not reflect the interaction between human immune system and human myeloma cells.

　　HuPBL-NOD/SCID mouse model has been applied in several other tumors[17-18]. In this study，we intraperitoneally injected human PBL into NOD/SCID mice to establish human immune system，innoculated the animals with genetically modified human myeloma cell vaccine，and then challenged subcutaneously the mice with live parental myeloma cells. The result showed that p53，GM-CSF and B7-1 gene-modified myeloma cell vaccine produces an in vivo antitumor immune response.

　　Materials and Methods

　　Animals，myeloma cells and peripheral blood lymphocytes (PBL)

　　In vivo experiments were performed in female NOD/SCID mice (from the Experimental Animal Center，Chinese Academy of Medical Sciences，Beijing)，aged 7 to 9 weeks，which were bred and maintained in specific pathogen-free conditions.

　　In this study，a human myeloma cell line sko-007，with human leukocyte antigen (HLA) A2 positive，was kindly provided by Professor Bei-Fen SHEN from Beijing Institute of Basic Medical Sciences and was cultured in RPMI 1640 (Sigma) supplemented with 10% FBS，100 U/ml penicillin，and 100 μg/ml streptomycin.

　　PBLs of normal donors in the Beitaiping Road 2# Hospital of Beijing，China，applied for establishment of human immune system in NOD/SCID mice，were isolated by Ficoll-Paque separation as described previously. Since sko-007 cells are HLA-A2 positive，PBLs with the same HLA antigen were selected and cultured in RPMI 1640 containing 15% FBS，5% human AB sera，50 U/ml IL-2，100 U/ml penicillin，and 100 μg /ml streptomycin. Cells were maintained in a humidified atmosphere containing 5% CO2 at 37℃. HLA-A2 gene of PBLs was amplified by polymerase chain reaction (PCR) as described previously[19]. Briefly，genomic DNA was extracted from PBLs of normal donors according to Wizard genomic DNA purification system (Promega，Madison，WI)，and HLA-A2 PCR was performed in a final volume of 50 μl with the use of 500 ng DNA，0.1 nmol/L MgCl2，0.01 nmol/L dNTP，0.01 nmol/L each primer，and 1 U Taq Polymerase (Promega) in PCR buffer. Five cycles，each consisting of 60 seconds at 96℃，60 seconds at 66℃，and 120 seconds at 72℃，followed by 25 cycles，each consisting of 60 seconds at 96℃，60 seconds at 56℃，and 120 seconds at 72℃，were performed in a Marstercycler Personal DNA Thermocycler (Eppendorf，Hamburg，Germany). The primers for HLA-A2 amplification were 5’-CCTCGTCCCCAGGCTCT-3’ (sense) and 5’-TGGCCCCTGGTACCCGT-3’ (antisense). β -actin was used as internal standard. The reaction products were electrophoretically separated through a 1.5% agarose gel and stained with ethidium bromide. The expected products generated by PCR were 813 base pairs (bp) and 396 bp for HLA-A2 and β-actin，respecti-vely.

　　Recombinant adenovirus and gene transfer

　　Ad-GFP (recombinant adenovirus expressing green fluorescence protein) was kindly provided by the Gene Therapy Unit，Baxter Healthcare Corp.，USA. Ad-p53/GM-CSF/B7-1 (recombinant adenovirus coexpressing human wild-type p53，GM-CSF and B7-1 proteins) was constructed by our department via homologous recombination in HEK293 cells (Ads E1-transformed human embryonal kidney cells). The inserted human wild-type B7-1 gene was driven by a Rous sarcoma virus (RSV) promoter，and p53 and GM-CSF genes，linked by internal ribosome entry site (IRES)，were driven by a cytomegalovirus (CMV) promo-ter[20]. These two kinds of recombinant adenovirus with high titer and purity were obtained by large-scale amplification in HEK293 cells and ultra-centrifugation in CsCl density gradient solution. The infection titers of Ad-GFP and Ad-p53/GM-CSF/B7-1 used in this study were 1×1011 pfu/ml and 5×1010 pfu/ml respectively.

　　To produce the myeloma cell vaccine，sko-007 cells were infected with Ad-GFP or Ad-p53/GM-CSF/B7-1 at a multiplicity of infection (MOI) of 200 for 2 hours. Culture medium was used for mock infection. After an additional 48 hours incubation at 37℃，5% CO2，transgenic expression of GFP，as well as B7-1，GM-CSF and p53 mediated by adenovirus were determined by flow cytometry，ELISA and Western blot，respectively，as previous described[12].

　　Establishment of human immune system and vaccine administration

　　Eighteen NOD/SCID mice were injected intraperitoneally with (3-4)×107 HLA-A2+ PBLs in 0.5 ml PBS，pH 7.4 on day 0 and then were randomly divided into 3 groups: control，Ad-GFP and Ad-p53/GM-CSF/B7-1 group. Each group of 6 huPBL-NOD/SCID mice was immunized twice subcutaneously on the abdomen with 1 ×106 irradiated Ad-p53/GM-CSF/B7-1- or Ad-GFP-infected sko-007 cells in 0.2 ml PBS or 0.2 ml PBS on days 7 and 14. Following vaccination，all animals received subcutaneous injection of 500 U recombinant human IL-2 per mouse for 3 times a week until sacrificed.

　　Tumor-challenge studies

　　On day 7 after the second injection，various treatment groups of mice were challenged subcutaneously on their backs with 5 × 106 live sko-007 cells. Tumor growth was monitored 2 to 3 times a week by measuring 2 maximum diameters of the tumor at the site of challenge with a vernier caliper and was reported as a mean of the 2 diameters. Mice were weighed using an electric scale and sacrificed by eyeball extirpation on day 66. The tumors were excised and weighed. The tumor volume was determined by measuring the length (a)，width (b) and thickness (c) using a vernier caliper and calculated by the formula: abcπ/6(mm2). The weight index was calculated as the weight ratio of tumor / mouse.

　　Processing of specimens for histopathology

　　When mice were killed，tumor tissues were removed from various treatment groups and fixed in 10% phosphate-buffered formalin for 24 to 48 hours，processed through graded alcohols，and embedded in paraffin. Serial sections of tumors were cut at various levels and stained with hematoxylin and eosin for histopathologic analysis. Additional sections were used for immunohistochemical staining for human T lymphocytes using monoclonal rabbit anti-human CD3 antibody (Zhongshan，Beijing). Subsequently，tissues were incubated with polyclonal biotin-conjugated goat anti-rabbit antibody (Zhongshan) followed by streptavidin- horseradish peroxidase (Zhongshan). The staining reaction was performed for 10 min with 3，3-diamino-benzidine-tetrahydrochloride (Zhongshan) in PBS (60 mg/100 ml). Finally，the tumor tissues were stained with hematoxylin to display the caryons.

　　Determination of human Ig levels

　　Levels of human IgG and IgA in the sera of huPBL-NOD/SCID mice were determined by agar diffusion assay. When mice were sacrificed，peripheral blood was harvested by eyeball extirpation. 100 μl sera were aspirated following centrifugation and mixed with 300 μl distilled water. 10 μl mixture of each sample was added to the assay hole of agar diffusion plate. After 24 hours incubation at room temperature，diameter of assay ring was tested and the content of human IgG or IgA was obtained according to standard curve.

　　Flow cytometry of human cells in mice spleens

　　When mice were sacrificed，spleens were removed and homogenized into a single-cell suspension. The spleen cells were washed twice in PBS，labeled with CD45-fluorescein isothiocyanate (FITC) (Becton Dickinson Bioscience PharMingen) for 30 min at 4°C，treated with 1×FACS Lysing solution (Becton Dickinson) for additional 15 minutes to lyse erythrocytes，and then washed twice before analysis. Cell surface immunofluorescence was measured using a FACSCalibur flow cytometer (Becton Dickinson Biosciences) and was analyzed with CellQuest software (Becton Dickinson Biosciences).

　　Statistical analysis

　　All data were presented as mean ± SD. SAS software (SAS，USA) was employed to determine the statistical significance of differences in weight index between samples using ANOVA analysis and DUNNETT t test，and P<0.05 was accepted as indicating significance.

　　Results

　　Discrimination of HLA-A2+ PBLs and reconstruction of human immune system in NOD/SCID mice

　　To efficiently present tumor antigens of HLA-A2 positive sko-007 cells，PBLs from HLA-A2+ normal donors were discriminated. As shown in Figure 1，3 of 6 DNA samples of PBLs were HLA-A2 positive，confirmed by PCR amplification. These PBLs from 3 donors were then inoculated to establish human immune system in 18 NOD/SCID mice by intraperitoneal injection.

　　To testify the availability of the huPBL-NOD/SCID mice model，human Igs in the sera and human leukocytes in the spleens were determined at the time when animals were sacrificed. Human IgG contents in control，Ad-GFP or Ad-p53/GM-CSF/B7-1 group were 5.73±3.14 g/L，5.42±2.11 g/L or 8.41±521 g/L，respectively (P>0.05)，and IgA contents were 1.29±0.30 g/L，1.39±0.46 g/L or 1.61±0.70 g/L，respectively (P>0.05). Human leukocytes were found in the spleens of mice (Figure 2). Our data indicated that human immune system was successfully established in NOD/SCID mice.

　　High expression of transgenes in sko-007 cells

　　Applying GFP as the reporter gene，several research groups，including us，have demonstrated efficient adenovirus-mediated gene transfer into myeloma cells. At a MOI of 200 pfu/cell，nearly all sko-007 cells were GFP-positive without obvious adenoviral toxicity (Figure 3A). Our previous study further indicated that human wild-type p53，GM-CSF，and B7-1 genes media- ted by the recombinant adenovirus Ad-p53/GM-CSF/B7-1 were highly expressed in three MM cell lines

　　(sko-007，U266 and RPMI8226) and primary myeloma cells，and that Ad-p53/GM-CSF/B7-1 had growth- inhibiting and apoptosis-inducing as well as immunogenicity-enhancing effect on myeloma cells[12]. In this study，we chose sko-007 to produce gene-modified MM vaccine and transplanted tumor in huPBL-NOD/SCID mice model. As shown in Figure 3B-D，the expressions of B7-1，GM-CSF and p53 on Ad-p53/GM-CSF/B7-1-infected sko-007 cells were determined by flow cytometry，ELISA and Western blot.

　　Generation of Ad-p53/GM-CSF/B7-1-mediated immunity is protective against subsequent tumor challenge

　　To examine whether the MM vaccine might affect its antitumor immunity in vivo，huPBL-NOD/SCID mice receiving 2 consecutive injections of irradiated (30 Gy) Ad-GFP- or Ad-p53/GM-CSF/B7-1-infected sko-007 cells or PBS were challenged with 5 ×106 sko-007 live tumor cells at day 7 after the second injection. Tumor growth was followed at the injection site. There were no differences between 3 groups within 40 days after tumor cell injection. At the time of sacrifice，most animals in control or Ad-GFP group (4 of 6 mice with PBS and 5 of 6 mice with Ad-GFP-infected sko-007，respectively) developed massive local tumors. In contrast，only 3 of 6 animals injected with Ad-p53/GM-CSF/B7-1-transferred sko-007 cells had measura-ble tumors (Figure 4). SAS analysis of the tumor weight index showed that Ad-p53/GM-CSF/B7-1-infected sko-007 vaccination significantly reduced tumor growth，compared with controls receiving Ad-GFP modified tumor cells or PBS (P<0.05 and P<0001，respectively) (Figure 4).

　　The in vivo antitumor activity of Ad-p53/GM-CSF/B7-1- transferred sko-007 cells was further stu-died by histopathologic analysis of implanted tumors of 3 groups. In PBS group，tumor cells were eugonic and showed typical morphologic features of neoplastic plasma cells，ie，irregular nuclear profiles with priminent nucleoli and abundant cytoplasma. Sheet necrotic areas were also identified (Figure 5A). Otherwise，in Ad-p53/GM-CSF/B7-1 group，tumor tissues displayed diffuse necrosis，mainly caused by apoptosis，accompanied with significant fibroplasias and blood vessel hyperplasia (Figure 5C). Tumor tissues of Ad-GFP group were similar to those of PBS group，accompanied with various degree of fibroplasias (Figure 5B). The results indicated that vaccination of mice with Ad-p53/GM-CSF/B7-1-transferred sko-007 cells displayed enhanced immunological protection compared to mice vaccinated with Ad-GFP- transferred sko-007 cell or PBS only. Immunostaining showed infiltration of human T cells in the tumor tissues，especially in the mice receiving Ad-p53/GM-CSF/B7-1-transferred sko-007 vaccination (Figure 6).

　　Discussion

　　Vaccination with whole tumor cells represents an attractive strategy for human cancer immunotherapy. Human tumors are extremely heterogeneous，making the identification of one or more ‘universal’ tumor antigens extremely difficult for many tumors. In addition，many potential tumor-associated antigens have not been

　　Transgenic expression of costimulatory molecules and/or cytokines is one means of enhancing the im-mune response to whole tumor cells. B7-1 and GM-CSF are particularly attractive for the immunotherapy of MM. It has been demonstrated in previous studies that myeloma cells lack the costimulatory molecule B7-1 and maturation of dendritic cells (DCs) is inhibited in MM patient，which thus may fail to induce an effective antitumor T-cell response[6，10]. Additionally，recent studies have shown that GM-CSF has the ability to promote maturation of precursor cells into DCs，and DCs cultured with apoptotic bodies stimulated significantly greater T cell proliferation than MM lysate-plused DCs or MM cells alone. Mature DCs then generate tumor antigen epitopes for cross-presentation on HLA-I class to stimulate CD8+ cytocoxic T cells，or for conventional presentation on HLA-II class to stimulate CD4+ helper T cells[10，21，22]. Considering that over-expression of exogenious P53 protein could induce apoptosis of p53 positive or negative myeloma cells，we proposed a new approach of co-transfer of immunomodulatory genes GM-CSF and B7-1 and apoptosis-inducing gene p53 to plasma cells as tumor vaccine might invoke an autologous immune response to tumor cells. Preliminary research suggested that by the introduction of wild-type p53，GM-CSF and B7-1 genes mediated by recombinant adenovirus Ad-p53/GM-CSF/B7-1，the growth of MM cells was inhibited via enhanced apoptosis and significant proliferation of autologous peripheral blood lymphocytes and specific cytotoxicity against autologous primary MM cells were induced in vitro[12].

　　In this report，we have shown that vaccination of MM cells modified by human wild-type p53，GM-CSF and B7-1 genes resulted in protection against challenge by additional local injection of nontransfected parental tumor cells in NOD/SCID mouse model. Tumor tissues increasingly displayed diffuse necrosis，accompanied with significant fibroplasias and blood vessel hyperplasia. Human T cells inhabited in the spleens and infiltrated the tumor tissues. This local effect is systemically immune mediated，because vaccination inhibited tumor growth from a subsequent tumor challenge at a distal site，which may particularly favor the antitumor responses required to treat the metastasis and spreading of human tumors. As showed in our previous study，this effect is probably mediated by the cytotoxicity of human lymphocytes inoculated intraperitoneally to NOD/SCID mice beforehand，which were activated by Ad-p53/GM-CSF/B7-1-transferred sko-007 cell vaccines.

　　In conclusion，we have shown that transgenic p53，GM-CSF and B7-1 expression induces a protective response against myeloma. The myeloma cell vaccine modified by these 3 genes may be of therapeutic value and can be considered for a phase I clinical trial of MM patients.

　　【参考文献】

　　1Kyle RA. The role of high-dose chemotherapy in the treatment of multiple myeloma: a controversy. Ann Oncol，2000;11(Suppl 1): 55-58

　　2Chiusolo P，Sica S，Piccirillo N，et al. Molecular and clinical following-up after stem cell transplantation for multiple myeloma. Ann Hematol，2001;80: 90-95

　　3Gahrton G，Bjorkstrand B. Progress in haematopoietic stem cell transplantation for multiple myeloma. J Intern Med，2000;248: 185-201

　　4Liu Q，Gazitt Y. Adenovirus-mediated delivery of p53 results in substantial apoptosis to myeloma cells and is not cytotoxic to flow-sorted CD34+ hematopoetic progenitor cells and normal lymphocytes. Exp Hematol，2000;28: 1354-1362

　　5Trudel S，Li Z，Dodgson C，et al. Adenovector engineered interleukin-2 expressing autologous plasma cell vaccination after high-dose chemotherapy for multiple myeloma —— a phase I study. Leukemia，2001;15: 846-854

　　6Tarte K，Zhang XG，Legouffe E，et al. Induced expression of B7-1 on myeloma cells following retroviral gene transfer results in tumor-specific recognition by cytotoxic T cells. J Immunol，1999;163: 514-524

　　7Dotti G，Savoldo B，Yotnda P，et al. Transgenic expression of CD40 ligand produces an in vivo antitumor immune response against both CD40+ and CD40- plasmacytoma cells. Blood，100: 200-207

　　8Gorschluter M，Ziske C，Glasmacher A，et al. Current clinical and laboratory strategies to augment the efficacy of immunotherapy in multiple myeloma. Clin Canc Res，2001;7: 2195-2204

　　9Ruffini PA，Biragyn A，Kwak LW. Recent advances in multiple myeloma immunotherapy. Biomed Pharmacother，2002;56: 129-132

　　10Hayashi T，Hideshima T，Akiyama M，et al. Ex vivo induction of multiple myeloma-specific cytotoxic T lymphocytes. Blood，2003;102: 1435-1442

　　11Brown RD，Pope B，Gibson J，et al. Dendritic cells from patients with myeloma are numerically normal but functionally defective due to TGFβ and IL-10 inhibition of the upregulation of CD80 expression after huCD40LT stimulation. Blood，2000;96: 161a

　　12Ren SP，Wu CT，Huang WR，et al. Adenoviral-mediated transfer of human wild-type p53，GM-CSF and B7-1 genes induces growth suppression and autologous anti-tumor cytotoxicity of multiple myeloma cells in vitro. Cancer Immuno Immunother，2006;55:375-385

　　13Wang JC，Lapidot T，Cashman JD，et al. High level engraftment of NOD/SCID mice by primitive normal and leukemic hematopoietic cells from patients with chronic myeloid leukemia in chronic phase. Blood，1998;91: 2406-2414

　　14Ahsmann EJ，van-Tol MJ，Oudeman-Gruber J，et al. The SCID mouse as a model for multiple myeloma. Br J Haematol，1995;89: 319-327

　　15Pilarski LM，Hipperson G，Seeberger K，et al. Myeloma progenitors in the blood of patients with aggressive or minimal disease: engraftment and self-renewal of primary human myeloma in the bone marrow of NOD/SCID mice. Blood，2000;95: 1056-1065

　　16Epstein J，Yaccoby S. Consequences of interactions between the bone marrow stroma and myeloma. Hematol J，2003;4: 310-314

　　17Nijmeijer BA，Willemze R，Falkenburg JH. An animal model for human cellular immunotherapy: specific eradication of human acute lymphoblastic leukemia by cytotoxic T lymphocytes in NOD/SCID mice. Blood，2002;100: 654-660

　　18Feuerer M，Beckhove P，Bai L，et al. Therapy of human tumors in NOD/SCID mice with patient-derived reactivated memory T cells from bone marrow. Nat Med，2001;7: 452-458

　　19任素萍，孔繁华，金荔等. HLA-A*02等位基因分型技术及汉HLA-A*02多态性研究. 解放军医学杂志，1999;24：180-184

　　20邱兆华，劳妙芬，吴祖泽. 共表达人p53、GM-CSF和B7-1基因的重组腺病毒的构建. 中国生物化学与分子生物学报，2001;17：329-334

　　21Pan PY，Li Y，Li Q，et al. In situ recruitment of antigen-presenting cells by intratumoral GM-CSF gene delivery. Cancer Immunol Immunother，2004;53: 17-25

　　22Mellman I，Steinman RM. Dendritic cells: specialized and regulated antigen processing machines. Cell，2001;106: 255-258