SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Malignant pleural mesothelioma is a refractory tumor with increasing incidence. In the present study, we established six mesothelioma cell lines possessing two allele deletions of the p16INK4A gene and one allele deletion of the neurofibromatosis type 2 gene, MM16, MM21, MM26, MM35, MM46 and MM56, from pleural effusion fluids or surgically resected tumors of Japanese patients. MM21, MM26 and MM46 cells failed to develop tumors in BALB/c-nude mice following subcutaneous inoculation. MM16 and MM35 cells slowly generated tumors at the site of subcutaneous inoculation in BALB/c-nude mice, but lost the expression of mesothelioma-related markers such as calretinin, D2-40 and Wilms’ tumor 1 in the subcutaneous tumors. On the other hand, MM56 cells rapidly generated tumors with the expression of calretinin and D2-40 in BALB/c-nude mice following subcutaneous inoculation. In addition, orthotopic implantation of MM56 cells into BALB/c-nude mice developed diffusely growing thoracic tumors by 3 weeks after implantation. Pleural effusions were observed in these mice 4 weeks after implantation. Thoracic tumors invaded aggressively into the chest wall 5 weeks after implantation and often metastasized into the lung, rib, peritoneum and pericardial cavity. On the pleural surface, MM56 cells were growing as single or multiple cell layers with the reactive mesothelium of recipient mice. These results indicate that MM56 cells can behave in a manner characteristic of human malignant pleural mesothelioma in the thoracic cavity of BALB/c-nude mice. The in vivo model using MM56 cells may be useful for studying the biological behavior of malignant pleural mesothelioma and developing its diagnostic and therapeutic strategies. (Cancer Sci 2011; 102: 648–655)

Malignant pleural mesothelioma (MPM), considered to be closely associated with asbestos exposure, is an aggressive tumor arising from mesothelial cells on the serosal surfaces of the thoracic cavity. Malignant pleural mesothelioma was once a rare disease, but its incidence is dramatically increasing worldwide. In Japan, it is expected to peak around 2025 as a result of widespread use of asbestos.(1) Malignant pleural mesothelioma is often diagnosed at an advanced stage and known to be resistant to conventional therapies. As a result it is associated with poor prognosis, with the median survival in the range of 9–17 months after the first diagnosis.(2) It is therefore important to establish a means for investigating the behaviors of MPM, leading to the development of early diagnosis and effective therapies.

Cell lines and animal models of human tumors are useful for studying the characteristics of tumors. Several MPM cell lines have been established(3–5) and animal models have been produced by inoculation of MPM cells or surgically resected MPM tissues into immunodeficient mice or rats.(6–9) Orthotopic implantation models are considered to be the most useful for studying the characteristics of MPM in vivo,(10–12) but most require a long period to develop MPM after implantation and often they have not reproduced the biological features of MPM well. In the pathological diagnosis of MPM, the distinction between MPM and reactive mesothelium (RM) is challenging because of the similar morphology and lack of reliable discriminating markers.(13,14) This problem may be resolved by establishing an experimental system that allows analysis of the morphological and immunohistological differences of MPM and RM on the pleural surface.

In the present study, we established six cell lines of MPM and found that one of them, termed MM56, exhibited the ability to reproduce the characteristic features of human MPM in BALB/c-nude mice. The in vivo model using MM56 cells might be useful for studying the biological behaviors of MPM and developing its diagnostic and therapeutic approaches.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Establishment of MPM cell lines.  Pleural effusion fluids and tumor tissues were obtained from Japanese patients diagnosed with MPM by histopathological examinations of pleural biopsy at the Hospital of Hyogo College of Medicine (Table 1). Cells in pleural effusion fluids were collected by centrifugation at 440g for 10 min and cultured in α-MEM (Invitrogen, Carlsbad, CA, USA) supplemented with 10% FBS (Equitech-Bio, Ingram, TX, USA), 100 U/mL penicillin (Sigma, St Louis, MO, USA) and 100 μg/mL streptomycin (Wako Pure Chemical Industries, Osaka, Japan). Adherent cells were cultured in a 37°C/5% CO2 humidified incubator to establish the MPM cell lines. Surgically resected tumors were cut into small pieces after removal of adipose tissues. Tissue pieces were plated on culture dishes, gently overlaid with culture medium and maintained in a 37°C/5% CO2 humidified incubator to establish the MPM cell lines.

Table 1.   Clinical characteristics and immunocyto/histochemical findings
Case/cell lineMM16MM21MM26MM35MM46MM56
  1. Pleural effusion fluids and tumor tissues were obtained from Japanese patients diagnosed with MPM by histopathological examinations of pleural biopsy. Immunocyto/histochemical findings of pleural effusions or tumor tissues, which are used as the sources for the establishment of cultured cell lines, are shown. B, biphasic type; CEA, carcinoembryonic antigen; CK, cytokeratin; E, epithelioid type; F, female; M, male; MPM, malignant pleural mesothelioma; ND, not determined; S, sarcomatoid type; TTF-1, thyroid transcription factor 1; WT-1, Wilms’ tumor 1.

SexMMFMMM
Age (years)654567676943
Asbestos exposure++++++
Histological typeBEEESE
Source for cultureTumorEffusionTumorEffusionEffusionTumor
Immunocyto/histochemical findings
 Calretinin+++++
 D2-40+ND+++
 WT-1ND+NDNDND
 CK5/6++++ND
 CK (CAM5.2)+NDNDNDNDND
 CK (AE1/AE3)+NDNDNDND+
 CEA
 TTF-1

This study was approved by the Ethics Committee of Hyogo College of Medicine and performed in accordance with the Declaration of Helsinki (1995) of the World Medical Association (as revised in Tokyo 2004). All patients provided written informed consent.

Cell proliferation assay.  Cells were plated in six-well plates 24 h before the cell proliferation assay. Adherent cells were harvested daily by 0.25% trypsin-EDTA treatment and the number of viable cells was calculated by the trypan blue exclusion method. The doubling time was determined at the log phase of cell proliferation.

Transmission electron microscopy.  Cells were harvested by 0.25% trypsin-EDTA treatment and fixed in 2.5% phosphate-buffered glutaraldehyde for 30 min and in 1% phosphate-buffered OsO4 for 15 min at room temperature. After dehydration, cells were embedded in epoxy resin and cut into 70-nm thickness. The sections were counterstained with uranil acetate and lead citrate and analyzed with a JEM-1220 electron microscope (JEOL, Tokyo, Japan).

Copy number analysis by real-time PCR.  The copy number of the p16INK4A and the neurofibromatosis type 2 (NF2) genes in the genomic DNA of MPM cells was analyzed by real-time PCR based on the dual-labeled fluorescent probe system, and determined by the comparative threshold cycle method using the RPS6 gene as a control and the genomic DNA of human pleural mesothelial cells, MeT-5A, as a reference. The following sets of forward primer, reverse primer and probe designed by FastPCR were used for real-time PCR: 5′-cgggcgactctggaggacgaagtt, 5′-agatctgtacgcgcgtggctcctc and 5′FAM-ctgggggcttgggaagcca-Eclipse for the p16INK4A gene; 5′-acctctttgatttggtgtgccgga, 5′-gtccattttgagccaggccactg and 5′FAM-cacaatcaaggacacagtggcct-Eclipse for the NF2 gene; and 5′-ggtttccccatgaagcagggtgtc, 5′-gagaacgctcagatttgcatccac and 5′FAM-tcagttcgtggttgcattgtgga-Eclipse for the RPS6 gene. In the comparative threshold cycle method, the calculated value of 1.0 indicated no deletion, that of 0.5 indicated a one allele deletion and that of 0 indicated a two allele deletion.

Implantation of MPM cells into mice.  Seven-week-old female BALB/c-nude mice were purchased from Japan SLC (Hamamatsu, Shizuoka, Japan) and maintained under specific pathogen-free conditions. The MPM cells (5 × 106 cells in 100 μL PBS) were injected subcutaneously into both flanks of BALB/c-nude mice, and the development of subcutaneous tumors was examined once a week. When tumors reached 1–2 cm in size, the mice were killed and the tumors were removed for histological examination. The tumor volume was calculated according to the formula: V (mm3) = ½ × A × B2, where A is the major axis and B is the minor axis.

For the orthotopic implantation model, MPM cells (3 × 106 cells in 100 μL PBS) were injected into the right thoracic cavity of BALB/c-nude mice through the intercostal space using a 29G needle after skin incision, as previously described.(10) After injection of the MPM cells, the incised skin was closed by 9 mm AUTOCLIP wound clips (Becton Dickinson, Sparks, MD, USA). The mice were killed and examined macroscopically for the development of tumors in the thoracic cavity on the scheduled day after inoculation or when they became moribund, whichever came first. The thoraces were removed from the mice for histological examination.

All experimental procedures were approved by the Animal Care Committee of Hyogo College of Medicine and performed in accordance with the criteria outlined in “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences (unpublished material).

Histocytological examinations.  Cultured cells were harvested by trypsin-EDTA treatment, collected by centrifugation, fixed in 10% formalin neutral buffer solution, embedded in paraffin and cut into 3-μm-thick sections to generate cell block specimens. Tissues removed from the mice were fixed in 10% formalin neutral buffer solution, embedded in paraffin and cut into 3-μm thickness. For the whole thoraces, decalcification was performed with EDTA before embedding in paraffin. Cells in pleural effusion fluids developed in mice orthotopically implanted with MPM cells were collected by centrifugation and subjected to preparation of cell block specimens or smear specimens. Paraffin-embedded specimens were stained with HE and smear specimens were stained with Papanicolaou.

Immunohistochemistry.  Specimens were heated to 98°C in Target Retrieval Solution (S1700; DakoCytomation, Glostrup, Denmark) or Target Retrieval Solution pH 9 (S2368; DakoCytomation), or incubated at room temperature with proteinase K (S3020; DakoCytomation) to facilitate antigen retrieval. Specimens of human origin were incubated with various mouse primary antibodies (Table 2) and sequentially with an anti-mouse immunoglobulin antibody using a ChemMate EnVision kit (DakoCytomation). Immunostaining with a rabbit anti-CD146 antibody (Table 2) was sequentially incubated with an anti-rabbit immunoglobulin using a ChemMate EnVision kit. For immunostaining of the mouse tissues with mouse primary antibodies, CAM5.2, calretinin and D2-40, a Histofine MOUSESTAIN kit (Nichirei Bioscience, Tokyo, Japan) was used to eliminate the background staining of endogenous mouse immunoglobulin. Immunoreacted cells were visualized with 3, 3′-diaminobenzidine, and the nuclei were lightly counterstained with hematoxylin.

Table 2.   Antibodies used in the present study
AntibodyCloneSourceDilutionRetrieval
  1. CD146 is a rabbit monoclonal antibody. Other antibodies used in the present study are mouse monoclonal antibodies. Source information with the manufacturers’ location is as follows: DakoCytomation, Glostrup, Denmark; Becton Dickinson, San Jose, CA, USA; and Epitomics, Burlingame, CA, USA. CEA, carcinoembryonic antigen; CK, cytokeratin; EMA, epithelial membrane antigen; ProK, proteinase K; TRS, Target Retrieval Solution; TTF-1, thyroid transcription factor 1; WT-1, Wilms’ tumor 1.

CalretininDAK-Calret1DakoCytomation1:100TRS, 20 min
D2-40D2-40DakoCytomation1:50TRS (pH 9), 20 min
WT-16F-H2DakoCytomation1:25TRS, 40 min; ProK, 5 min
CK 5/6D5/16 B4DakoCytomation1:50TRS (pH 9), 40 min
CK (CAM5.2)CAM5.2Becton DickinsonPre-dilutedProK, 5 min
CK (AE1/AE3)AE1/AE3DakoCytomation1:100TRS (pH 9), 20 min
CEAII-7DakoCytomation1:50TRS, 20 min
TTF-18G7G3/1DakoCytomation1:50TRS (pH 9), 40 min
EMAE29DakoCytomation1:40None
CD146EPR3208Epitomics1:200TRS, 20 min

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Establishment of MPM cell lines.  Six cell lines, designated MM16, MM21, MM26, MM35, MM46 and MM56, were established from the pleural effusion fluids or surgically resected tumors of untreated Japanese patients with MPM (Table 1). These cells grew as an adherent monolayer (Fig. 1a) with the doubling time of 12.4 ± 0.9 h for MM16, 13.9 ± 0.4 h for MM21, 15.7 ± 1.3 h for MM26, 32.8 ± 2.5 h for MM35, 11.5 ± 0.5 h for MM46 and 15.6 ± 1.8 h for MM56. Ultrastructural analysis revealed that these cells possessed numerous long, thin and irregular microvilli, which is characteristic of MPM, on the cell surface (Fig. 1b).

image

Figure 1.  Malignant pleural mesothelioma (MPM) cell lines, MM16, MM21, MM26, MM35, MM46 and MM56. (a) Phase-contrast microscopic analysis of cultured MPM cell lines. Scale bar, 50 μm. (b) Transmission electron microscopic analysis of cultured MPM cell lines. Scale bars, 2 μm. (c) HE staining of a cell block specimen and immunostaining of MM56 cells with antibodies to calretinin, Wilms’ tumor 1 (WT-1), cytokeratin (CK) 5/6 and CK (CAM5.2). Scale bar, 50 μm.

Download figure to PowerPoint

Immunostaining of the cell block specimens showed that all MM16, MM21, MM26, MM35, MM46 and MM56 cells were negative for carcinoma-related markers, carcinoembryonic antigen and thyroid transcription factor 1, but positive for both mesothelioma- and carcinoma-related markers, cytokeratin (CK) (clone CAM5.2) and CK (clone AE1/AE3). In addition, MM16, MM21, MM35 and MM56 cells were positive for mesothelioma-related markers, calretinin, D2-40, Wilms’ tumor 1 (WT-1) and CK5/6, MM26 cells for calretinin, D2-40 and WT-1, and MM46 cells for WT-1 (Table 3, Fig. 1c). These results indicate that all established cell lines originated from MPM and that MM16, MM21, MM35 and MM56 cells maintained the immunohistological phenotype of MPM well.

Table 3.   Immunostaining of MPM cell lines using cell block specimens
Cell linesMM16MM21MM26MM35MM46MM56
  1. Staining intensity was scored as follows: −, negative; ±, focal and very weak positive; 1+, weak positive; 2+, moderate positive; 3+, strong positive. CEA, carcinoembryonic antigen; CK, cytokeratin; MPM, malignant pleural mesothelioma; TTF-1, thyroid transcription factor 1; WT-1, Wilms’ tumor 1.

Mesothelioma-related markers
 Calretinin1+2+1+2+1+
 D2-40±3+3+2+1+
 WT-12+3+1+3+2+3+
 CK5/62+2+3+2+
Both mesothelioma- and carcinoma-related markers
 CK (CAM5.2)3+2+2+2+1+2+
 CK (AE1/AE3)3+3+3+3+3+3+
Carcinoma-related markers
 CEA
 TTF-1

To analyze the copy number of the p16INK4A and the NF2 genes in the established cell lines, real-time PCR with the comparative threshold cycle method was performed. The value of the p16INK4A and the NF2 genes was calculated as 0.01 and 0.53 for MM16, 0.02 and 0.39 for MM21, 0.01 and 0.53 for MM26, 0.01 and 0.52 for MM35, 0.00 and 0.22 for MM46, and 0.00 and 0.35 for MM56, respectively. These results indicated that two alleles of the p16INK4A gene and one allele of the NF2 gene were deleted in all established cell lines.

Tumorigenicity of the MPM cell lines in BALB/c-nude mice.  To examine the tumorigenic potential of the established cell lines, the MPM cells were injected subcutaneously into both flanks of the BALB/c-nude mice. The MM21, MM26 and MM46 cells failed to develop tumors at the site of inoculation. In contrast, the MM16 and MM56 cells formed tumors in all 12 mice (24 sites of both flanks) and the MM35 cells in nine of 12 mice (18 of 24 sites of flanks). The rate of tumor growth was examined in eight sites of four mice inoculated with MM16 and MM56 cells and in six sites of three mice inoculated with MM35 cells. MM56 tumors grew very aggressively, MM16 tumors grew moderately and MM35 tumors grew extremely slowly (Fig. 2a). The MM56 tumors were soft and rich in fluids, while MM16 and MM35 tumors were solid and firm. Immunostaining with anti-CAM5.2 antibody, which reacts to only human but not mouse CK, confirmed that MM16, MM35 and MM56 tumors arose from inoculated human MPM cells (Fig. 2b). The MM56 tumors retained immunoreactivity to mesothelioma-related markers, calretinin and D2-40 (Fig. 2c), whereas the MM16 and MM35 tumors lost immunoreactivity to these markers.

image

Figure 2.  Tumors developed from subcutaneously implanted MM16, MM35 and MM56 cells. (a) Growth of subcutaneous tumors. Each point represents the mean volume ± SEM of 6–8 tumors. (b) Macroscopic analysis and histological examination of subcutaneous tumors. The volume of MM16, MM35 and MM56 tumors was 3757 (after 9 weeks), 320 (17 weeks) and 3035 mm3 (7 weeks), respectively. MM16, MM35 and MM56 tumors were stained with HE and immunostained with anti-CAM5.2 antibody. Scale bars, 100 μm. Ep, epithelium; T, subcutaneous tumor. (c) Immunoreactivity of MM56 subcutaneous tumor to mesothelioma-related markers. A MM56 subcutaneous tumor was stained with HE and immunostained with antibodies to calretinin and D2-40. Scale bar, 100 μm.

Download figure to PowerPoint

Orthotopic implantation model of MPM cell lines in BALB/c-nude mice.  To generate an orthotopic implantation model of MPM cell lines, the MPM cells were inoculated orthotopically into the thoracic cavity of BALB/c-nude mice. As in subcutaneous implantation, MM21, MM26 and MM46 cells failed to develop thoracic tumors, but MM16, MM35 and MM56 cells were able to develop thoracic tumors. Orthotopic implantation of MM16 cells generated tumor nodules, but not diffusely growing tumors along the pleura, in the thoracic cavity 6 weeks after implantation. MM16 thoracic tumors lost expression of the mesothelioma-related markers (data not shown). Orthotopic implantation of the MM35 cells did not generate thoracic tumors in mice within 6 weeks after implantation. Some mice orthotopically inoculated with MM35 cells became moribund with pleural effusions and tumors in the thoracic cavities around 30 weeks after implantation (Fig. 3a,b). Cells in the pleural effusions of these mice grew in a papillary pattern and were CAM5.2 positive (Fig. 3c,d). Orthotopic implantation of MM56 cells generated thoracic tumors diffusely growing along the pleura and pleural effusions by 6 weeks after implantation (Fig. 3e,f). Cells in the pleural effusions showed MPM features with mirror ball-like cell clusters and multinucleated cells (Fig. 3g), and were CAM5.2 positive (data not shown). Histologically, CAM5.2-positive MM56 cells proliferated in the thoracic cavity (Fig. 3h,i) and grew along the parietal pleura of the anterior chest wall with invasion at the site of cell inoculation (Fig. 3j). Metastasis of MM56 thoracic tumors was often observed in the lung, rib, peritoneum and pericardial cavity throughout the experimental period (3–7 weeks) (Fig. 3k–r).

image

Figure 3.  Orthotopic implantation model of MM35 and MM56 cells. (a–d) A BALB/c-nude mouse 33 weeks after orthotopic inoculation with MM35 cells. (a) Pleural effusion fluids (700 μL) and MM35 thoracic tumors as observed from the abdominal cavity through the diaphragm. (b) MM35 thoracic tumors. (c) HE staining of a cell block specimen of pleural effusion. Scale bar, 100 μm. (d) Immunostaining of a cell block specimen of pleural effusion with anti-CAM5.2 antibody. (e–r) A BALB/c-nude mouse 6 weeks after orthotopic inoculation with MM56 cells. (e) MM56 thoracic tumors. (f) Diffuse growth involving the costal pleura of MM56 thoracic tumors. (g) Papanicolaou staining of a smear specimen of pleural effusion fluids (200 μL). Scale bar, 20 μm. (h) HE staining of the thorax. Scale bar, 2 mm. Ht, heart; Lu, lung; T, tumor. (i) Immunostaining of an adjacent specimen of (h) with anti-CAM5.2 antibody. (j) HE staining and immunostaining of the anterior chest wall shown in (f) with anti-CAM5.2 antibody. Arrows indicate tumor invasion at the inoculation site of the MM56 cells. Scale bar, 1 mm. (k–n) HE staining. Scale bars, 100 μm. (o–r) Immunostaining of an adjacent specimen of (k–n) with anti-CAM5.2 antibody. (k,o) Lung 3 weeks after implantation, (l,p) rib 5 weeks after implantation, (m,q) peritoneum 6 weeks after implantation, and (n,r) the pericardial cavity 7 weeks after implantation.

Download figure to PowerPoint

Utility of the MM56 orthotopic implantation model.  Orthotopic implantation of MM56 cells resulted in the generation of thoracic tumors 3 weeks after implantation and pleural effusions 4 weeks after implantation (Table 4). Mice often became moribund around 7 weeks after implantation. Histologically, CAM5.2-positive cell clusters were observed in the thoracic cavity and CAM5.2-positive single or multiple cell layers spread along the pleural surface 3 weeks after implantation (Fig. 4a,b). Aggressive invasion of CAM5.2-positive cells into the chest wall was observed 5 weeks after implantation (Fig. 4c,d). These results indicate that MM56 cells grow in mice in a manner characteristic of human MPM.

Table 4.   Tumorigenicity of the MM56 orthotopic implantation model
Weeks after inoculationIncidence
Thoracic tumorPleural effusion
  1. MM56 cells (3 × 106 cells in 100 μL PBS) were inoculated orthotopically into the right thoracic cavity of BALB/c-nude mice. The incidence of thoracic tumors and pleural effusions in recipient mice was examined at the indicated times.

34/40/4
44/43/4
55/53/5
65/55/5
72/21/2
image

Figure 4.  Utility of MM56 orthotopic implantation model. (a–d) Progression of MM56 cells in the thoracic cavity. (a,b) A BALB/c-nude mouse 3 weeks after orthotopic inoculation with MM56 cells. (c,d) A BALB/c-nude mouse 5 weeks after orthotopic inoculation with MM56 cells. (a,c) HE staining of the thoracic cavity. (b,d) Immunostaining of adjacent specimens of (a,c) with anti-CAM5.2 antibodies. CAM5.2-positive MM56 cells proliferate in the thoracic cavity and diffusely grow along the pleural surface (black arrows) 3 weeks after MM56 inoculation (a, b). In a mouse 5 weeks after MM56 inoculation, CAM5.2-positive MM56 cells invade aggressively into the musculature of the chest wall (asterisks) (c,d). Scale bars, 100 μm. (e–h) Discrimination between malignant pleural mesothelioma (MPM) and reactive mesothelium (RM) in a MM56 orthotopic implantation model 6 weeks after inoculation. (e) HE staining of the thoracic cavity. (f) High magnification of the boxed area in (e). (g) Immunostaining of an adjacent specimen of (e) with anti-CAM5.2 antibody. (h) Immunostaining of an adjacent specimen of (e) with anti-CD146 antibody. Scale bars, 100 μm. Black arrow heads (cell clusters) and black arrows (single or multiple cell layers) indicate growing CAM5.2- and CD146-positive MM56 cells. White arrows (a single cell layer) indicate CAM5.2- and CD146-negative RM of a recipient mouse.

Download figure to PowerPoint

To evaluate the utility of the MM56 orthotopic implantation model, thoracic tissues with MM56 cells were immunostained with anti-CD146 antibody that reacts to both human and mouse CD146, a marker discriminating between MPM (CD146-positive) and RM (CD146-negative).(15) As shown in Figure 4e–h, CAM5.2-positive MM56 cells growing in clusters in the thoracic cavity (black arrow heads) were positive for CD146. CAM5.2-positive MM56 single or multiple cell layers along the pleural surface (black arrows) were also positive for CD146, whereas CAM5.2-negative mouse single cell layers on the pleural surface (white arrows) were negative. These results indicate that the coexistence of MPM and RM on the pleural surface is reproduced in the MM56 orthotopic implantation model.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

In the present study we have established six mesothelioma cell lines, MM16, MM21, MM26, MM35, MM46 and MM56, from pleural effusion fluids or surgically resected tumors of untreated Japanese patients with MPM. The MM56 cell line was derived from a soft tumor of MPM consisting of polygonal cells partially with tubulopapillary or microcystic structures and expressing calretinin and D2-40. In addition, pleural effusion cytology of this MPM showed a cellular arrangement of sheets or mirror ball-like cell clusters. The MM56 cells formed subcutaneous tumors soft and rich in fluids and positive for calretinin and D2-40 in BALB/c-nude mice, indicating that the MM56 cells retain characteristic features of human MPM in BALB/c-nude mice. In addition, MM56 cells diffusely grew on the surface of parietal pleura with pleural effusions containing mirror ball-like cell clusters, invaded into the musculature of the chest wall and often metastasized into the lung, rib, peritoneum and pericardial cavity during 3–7 weeks after orthotopic implantation. Because MM56 cells could reproduce various stages of MPM from the earliest phases in situ to the advanced phases with metastasis within a short period in the BALB/c-nude mice, the MM56 orthotopic implantation model can be regarded as an excellent animal model for studying the biological behavior of human MPM in vivo and developing its diagnostic and therapeutic strategies.

Proliferation of atypical mesothelial cells on the pleural surface is considered as mesothelioma in situ, but it could only be diagnosed when accompanied by adjacent or subsequent invasive MPM.(16–19) Discrimination between early stage MPM and RM has been problematic due to morphological similarities and the lack of reliable discriminating markers.(13,14) On the other hand, we have recently reported that CD146 is a useful marker to discriminate MPM from RM in pleural effusion cytology.(15) In the MM56 orthotopic implantation model, the coexistence of MPM and RM on the pleural surface is reproduced, and immunostaining with anti-CD146 antibody is able to identify MM56-derived MPM cells. Therefore, the MM56 orthotopic implantation model should be useful in detecting in situ lesions and identifying additional markers that discriminate between early stage MPM and RM.

The MM35 cells were less useful, because they slowly produced subcutaneous tumors in BALB/c-nude mice and thoracic tumors in only two of five mice 30 weeks after inoculation in an orthotopic implantation model. However, cytological analysis of the pleural effusion fluids in the recipient mice revealed that the MM35 cells formed papillary cell clusters, one of the cytological features of MPM.(20,21) Isolation of sublines with high tumorigenicity and MPM characteristics from pleural effusion fluids of recipient mice may improve the usefulness of MM35 cells.

Most human MPM grow diffusely in the thoracic cavity and a few proliferate as a nodule. The former is classified as diffuse MPM and the latter is classified as localized MPM. In addition, some MPM develop as multiple nodules in the thoracic cavity. Biopsy tissues of MPM, from which the MM16 cell line was established, were histologically biphasic mesothelioma growing as multiple nodules, but MPM at autopsy was sarcomatoid mesothelioma with a loss of expression of the mesothelioma-related markers. The MM16 cell line kept the expression of the mesothelioma-related markers, calretinin, D2-40, WT-1 and CK5/6, but MM16 subcutaneous and thoracic tumors were negative for these markers. Orthotopic inoculation of MM16 cells generated tumor nodules, but not diffusely growing tumors along the pleura, in the thoracic cavity of the BALB/c-nude mice. The MM16 subcutaneous and thoracic implantation models might be useful to clarify the mechanism(s) by which MPM grows as nodules and changes the phenotype.

It has been reported that WT-1, CAM5.2 and AE1/AE3 are expressed in more than 80% of sarcomatoid mesothelioma and the combined use of these markers provides the highest sensitivity in the differentiation of sarcomatoid mesothelioma from true sarcoma.(22) Consistent with this report, the MM46 cell line established from the sarcomatoid mesothelioma was positive for WT-1, CAM5.2 and AE1/AE3. Although sarcomatoid mesothelioma is highly malignant in human MPM, the MM46 cells failed to develop tumors in the BALB/c-nude mice. Previously, Usami et al.(4) established an epithelioid mesothelioma cell line, Y-MESO-8A, and a sarcomatoid mesothelioma cell line, Y-MESO-8D, from a biphasic mesothelioma consisting of epithelioid and sarcomatoid components, and reported that Y-MESO-8A cells, but not Y-MESO-8D cells, had the potential for developing a subcutaneous tumor in BALB/c-nude mice. Therefore, in sarcomatoid mesothelioma, it is unlikely that a correlation is observed between malignancy in human and tumorigenicity in nude mice. The MM46 cells may be useful to study the characteristic features of sarcomatoid mesothelioma.

The p16INK4A gene is the most frequently inactivated tumor suppressor gene in human MPM, and ∼90% of MPM possesses deletions of the p16INK4A gene.(23) Abnormalities of the p16INK4A gene take part in the dysregulation of the cell cycle leading to malignant transformation of mesothelial cells. The NF2 tumor suppressor gene is also frequently mutated in MPM. Recently, Yokoyama et al.(24) reported that mutation of the NF2 gene promotes MPM proliferation through activation of YAP1, a transcriptional coactivator functionally inhibited by Merlin, a product of the NF2 gene. In all six cell lines established in the present study, two allele deletions of the p16INK4A gene and one allele deletion of the NF2 gene were found, which supports the theory that abnormalities of the p16INK4A and the NF2 genes participate in the development of MPM. On the other hand, there is a possibility that abnormalities other than the p16INK4A and the NF2 genes are related to the difference of tumorigenicity in nude mice among these six cell lines.

Novel biological therapies for MPM are under investigation.(2) Effects of various inhibitors against angiogenic factor, tyrosine kinase, ribonuclease, histone deacetylase and proteasome on MPM patients have been studied without much success. Mesothelin and CD26, mesothelioma-associated cell surface antigens, have been reported to be an immunotherapeutic target.(25–27) Recently, CD146 has been identified as the surface antigen recognized by an internalizing single chain antibody that can deliver liposome-encapsulated small molecule drugs into the cytoplasm of mesothelioma cells.(28) Our earlier study showed that CD146 was expressed in MPM but not RM,(15) supporting the theory that CD146 can be used as an effective immunotherapeutic target of MPM. The MM56 subcutaneous and orthotopic implantation models may serve as suitable in vivo systems to evaluate this possibility and develop preclinical strategies.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

This work was supported in part by Grants-in-Aid for Scientific Research and a Hitec Research Center Grant from the Ministry of Education, Science, Sports, Culture, and Technology of Japan, and Grants-in-Aid for Researchers, Hyogo College of Medicine. The authors thank Ms. Michiko Kakihana and Ms. Mio Ohkabe for their technical and secretarial assistance.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References