|Cancer||Cell death||Cell cycle||Cytoskeleton||Exo/endocytosis||Differentiation||Division||Organelles||Signalling||Stem cells||Trafficking|
Low-expression of E-cadherin in leukaemia cells causes loss of homophilic adhesion and promotes cell growth
Qing Rao, Ji‑Ying Wang, Jihong Meng, Kejing Tang, Yanzhong Wang, Min Wang, Haiyan Xing, Zheng Tian and Jianxiang Wang1
State Key Laboratory of Experimental Hematology, Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, 288 Nanjing Road, Tianjin 300020, Peoples Republic of China
E-cadherin (epithelial cadherin) belongs to the calcium-dependent adhesion molecule superfamily and is implicated in the interactions of haematopoietic progenitors and bone marrow stromal cells. Adhesion capacity to bone marrow stroma was impaired for leukaemia cells, suggesting that a breakdown of adhesive mechanisms governed by an adhesion molecule may exist in leukaemic microenvironment. We previously found that E-cadherin was low expressed in primary acute leukaemia cells compared with normal bone marrow mononuclear cells. In this study, we investigate the functional importance of low E-cadherin expression in leukaemia cell behaviours and investigate its effects in the abnormal interaction of leukaemic cells with stromal cells. After expression of E-cadherin was restored by a demethylating agent in leukaemia cells, E-cadherin-specific adhesion was enhanced. Additionally, siRNA (small interfering RNA)-mediated silencing of E-cadherin in Raji cells resulted in a reduction of cell homophilic adhesion and enhancement of cell proliferation and colony formation. These results suggest that low expression of E-cadherin contributes to the vigorous growth and transforming ability of leukaemic cells.
Key words: cell adhesion, cell growth, E-cadherin, leukaemia, RNA interference
Abbreviations: 5-Aza-CdR, 5-aza-2′-deoxycytidine, E-cadherin, epithelial cadherin, GAPDH, glyceraldehyde-3-phosphate dehydrogenase, MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide, SiCDH1, siRNA specific to CDH1, SiCon, control siRNA, siRNA, small interfering RNA
1To whom correspondence should be addressed (email email@example.com).
Adult normal haematopoiesis occurs within bone marrow microenvironment, which plays a crucial role in maintaining haematopoietic homoeostasis. The balance of signals generated in bone marrow microenvironment regulates haematopoietic progenitor proliferation and differentiation. The balance is achieved through the interaction between haematopoietic progenitor and bone marrow stromal cells that are mediated by specific molecules including cell adhesion molecules and cytokines (Zhu and Emerson, 2004; Taichman, 2005).
Adhesion molecules are implicated in the interaction of haematopoietic progenitors and bone marrow stromal cells (Li and Xie, 2005). E-cadherin (epithelial cadherin) belongs to the calcium-dependent adhesion molecule superfamily that initiates formation of cell–cell adherent junctions. E-cadherin is expressed predominantly on epithelial cells and plays a critical role in the establishment of normal tissue architecture (Gumbiner, 2005; Gumbiner et al., 1988). E-cadherin is also expressed on haematopoietic CD34+ stem cells, bone marrow stromal cells and erythroid progenitors and is functionally involved in the maturation of erythroid lineage. These suggest that E-cadherin may contribute to the haematopoietic process (Armeanu et al., 1995; Turel and Rao, 1998).
E-cadherin-mediated adhesion is accomplished by homophilic binding of extracellular domain of E-cadherin in a zipper-like fashion. The formation of cell–cell adhesion is ultimately dependent on direct binding of catenins (β-catenin and γ-catenin) to the intracellular domain of E-cadherin. As an essential mediator of Wnt signal transduction pathway, β-catenin is usually sequestered in the E-cadherin-adherent junctions. As a result, translocation of β-catenin from nucleus to plasma membrane leads to the inactivation of its target proto-oncogenes (Gottardi et al., 2001). Several studies have revealed that the loss of E-cadherin may result in the activation of Wnt signal pathway and contribute to early initiation stages of tumourigenesis (Cavallaro et al., 2002). Down-regulated or completely absent expression and mutation of E-cadherin gene has been observed in many carcinoma cells (Guilford et al., 1998).
Similar to normal haematopoiesis, leukaemogenesis arises in bone marrow. It is generally agreed that altered interactions between primitive progenitor cells and bone marrow stromal cells are essential for haematologic malignancies (Zhou et al., 2005; Williams and Cancelas, 2006). However, the role of the haematopoietic bone marrow microenvironment in malignant progression remains to be elucidated. There is some evidence demonstrating that abnormal adhesion creates a microenvironment that can prevent apoptosis of leukaemia cells and even promote leukaemia cells growth (Konopleva et al., 2002).
E-cadherin is one of the adhesion molecules which contribute to the interaction of haematopoietic progenitors with their niche. While the pivotal role of the loss of E-cadherin in tumourigenesis and carcinoma cell invasiveness has been well characterized, little is known about its function in the aberrant adhesion of haematopoietic progenitors and leukaemogenesis. Corn et al. (2000) have shown that methylation of E-cadherin occurs commonly in acute leukaemia. In our previous study, we found that E-cadherin is low expressed in primary leukaemic cells (Meng et al., 2008). Considering that loss of E-cadherin may play a role in leukaemogenesis, we now further analysed the function of loss of E-cadherin in the interaction of leukaemia cells with stromal cells and determined its potential role in the malignant behaviour of leukaemic cells.
2. Materials and methods
Recombinant human E-cadherin–Fc chimaera was purchased from R&D Systems. The SilencerTM siRNA Construction Kit was obtained from Ambion. OPTI-MEM medium was purchased from GIBCO. Mouse anti-E-cadherin antibody and peroxidase-conjugated anti-mouse IgG was the product of Santa Cruz Biotechnology. Anti-actin monoclonal antibody was from Sigma. SuperSignal chemiluminescent detection system was purchased from Pierce.
2.2. RNA isolation and SYBR Green quantitative real-time reverse transcription-PCR assay
Total RNA was extracted from leukaemia cell lines using RNAiso reagent (TaKaRa). cDNA was prepared with SuperScript III (Invitrogen) and used as templates for PCR. PCR core reagents and SYBR green (TaKaRa) were used with 10 μM of forward and reverse primers. The primers used for E-cadherin gene (CDH1) are forward TGAAGGTGACAGAGCCTCTGGAT, reverse TGGGTGAATTCGGGCTTGTT. Real-time quantitative PCR was performed with the Applied Biosystems 7500 Real-Time PCR System. Expression levels of the target genes were normalized against GAPDH (glyceraldehyde-3-phosphate dehydrogenase). All amplifications were done in triplicate, and at least three biological replicates were performed. The amplified products were also visualized by 2% agarose gel electrophoresis.
2.3. Indirect immunofluorescence staining of cell surface E-cadherin
Cell surface E-cadherin was determined by indirect immunofluorescence staining. After fixed cytocentrifuge slides were incubated with 5% goat serum, a mouse anti-E-cadherin monoclonal antibody was applied at 1:50 for 1 h at room temperature. The slides were then incubated with a FITC-conjugated goat anti-mouse IgG for 30 min and finally analysed with a laser scanning confocal fluorescence microscope (Leica TCS-4D DMIRBE; Leica).
2.4. Preparation of E-cadherin–Fc protein-coated plate and cell adhesion assay
Recombinant human E-cadherin–Fc chimera, the extracellular domain of human E-cadherin fused to the Fc region of human IgG1, was used in homophilic adhesion assay. For measuring homophilic E-cadherin interactions, microtitre plates were coated overnight at 4°C with 100 μl/well of E-cadherin–Fc (10 μg/ml) in PBS and blocked with 1% BSA. Human IgG1-coated plates were prepared in a similar manner and used in control.
The homophilic adhesion assay was performed as described in reference (Miki et al., 1993). Briefly, U937 or HL-60 cells were labelled with 1 mg/ml MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide] at 1×106 cells/ml at 37°C for 20 min. After extensive washing, cells were resuspended in growth medium containing 2 mM Ca2+ and added to each well coated with the chimaeric E-cadherin–Fc protein or human IgG (7.5×105 cells/well in 96-well cell culture plate). The plate was incubated at 37°C for 2 h. Any unbound cells were gently washed. After dimethylsulfoxide was added, the absorbance signal from the adherent cells was measured using a spectrophotometer.
2.5. Synthesis of siRNA (short interfering RNA) and siRNA transfection
Nucleotide sequence of 21 bp SiCDH1 (siRNA specific to CDH1) was designed by siRNA Target Finder (Ambion) and BLAST to determine specificity. Sense and antisense cDNA templates were synthesized (Invitrogen). Double-stranded, 21-nucleotide RNA (siRNA) molecules were synthesized with the SilencerTM siRNA Construction Kit according to the manufacturer's instructions. The sense and antisense sequences of siRNA to CDH1 were (SiCDH1) 5′-GCAGAATTGCTCACATTTCTT-3′ and 5′-GAAATGTGAGCAATTCTGCTT-3′, respectively. The SiCon (control siRNA) with a scrambled order was designed, and the sequences were 5′-TAGCGACTAAACACATCAACCTGTCTC-3′and 5′-TTGATGTGTTTAGTCGCTACCTGTCTC-3′, respectively.
One day before transfection, culture medium was completely replaced by fresh DMEM (Dulbecco's modified Eagle's medium) containing 10% serum without antibiotics. After being washed twice with OPTI-MEM, Raji cells were plated in a 24-well plate at 5×105 cells/well, and 293T cells were plated in a six-well plate at 5×105 cells/well in OPTI-MEM. Thirty nanomolar siRNA was transfected into Raji or 293T cells using Lipofectamine 2000 reagent following the manufacturer's protocol. The efficiency of siRNA-mediated protein suppression was determined by Western blot analysis.
2.6. Western blot
E-cadherin protein levels were determined by Western blot analysis. Total protein was obtained from cells using lysis buffer containing protease inhibitors. After being denatured and loaded onto SDS/polyacrylamide gels, protein samples were then transferred to nitrocellulose and probed with primary antibody. Mouse anti-E-cadherin antibody or anti-actin monoclonal antibody was used at 1:1000. Peroxidase-conjugated anti-mouse IgG at 1:2000 was then used as the secondary antibody. The immunoreactive proteins were visualized using the SuperSignal chemiluminescent detection system. Intensity of the bands was measured and analysed with a Multimage Light Cabinet Alpha Image densitometer (Alpha Innotech). Relative expression of E-cadherin was expressed as the fold of the intensity of E-cadherin compared with that of β-actin for each sample.
2.7. Adhesion-associated proliferation assay
Adhesion-associated proliferation assay was performed in 96-well plate precoated with 293T cells. A lymphoblast leukaemia cell line (Raji), which expresses E-cadherin, was used in this proliferation assay following siRNA treatment. After being treated with SiCDH1 or SiCon for 48 h, 2×104 Raji cells/well were seeded into 96-well plates, which were precoated with SiCDH1 or SiCon-treated 293T cells. Cells were then incubated at 37°C for an additional 24–72 h. At the end of incubation, Raji cells were removed to another 96-well plate, and the plate was incubated for 4 h in the presence of MTT (final concentration 0.5 mg/ml). After MTT solution was removed, 150 μl DMSO was added, and absorption was measured at wavelengths of 546 nm.
2.8. Colony-formation assay
Colony formation by Raji cells was carried out in triplicate in 100 μl essential medium containing 1% methylcellulose, 20% FCS (fetal calf serum), 5×10−4 M β-mercaptoethanol, 0.03% glutamine and 1×104 cells/ml in 96-well culture plates. Cells were incubated for 5 days in a humidified atmosphere with 5% CO2 at 37°C, and colonies were counted with a reversed microscope. Cell aggregates containing 40 or more cells were considered colonies. Each experiment was carried out in triplicate.
2.9. Statistical analysis
Independent pair t test was conducted to evaluate whether the differences in E-cadherin expression were statistically significant between different cell groups. Data were analysed using SPSS statistics software. P-values <0.05 were considered to be with statistically significant differences.
3.1. Restoration of E-cadherin expression in leukaemia cells induces cell homophilic adhesion
It has been reported that aberrant methylation of the E-cadherin promoter is a common mechanism for E-cadherin silence, and methylation of E-cadherin gene is also observed in leukaemia cell (Corn et al., 2000). To investigate the significance of underexpression of E-cadherin in leukaemia cells, we examined the CDH1 gene methylation patterns in a series of human leukaemia cell lines. We found that five of seven cell lines were fully methylated (data not shown). We then treated E-cadherin-silenced leukaemia cells with a demethylating agent, 5-Aza-CdR (5-aza-2′-deoxycytidine) to restore E-cadherin expression and explored its effects on E-cadherin-mediated cell adhesion.
Two E-cadherin-silenced leukaemia cell lines (HL-60 and U937) were chosen to further study the effect of E-cadherin on cell adhesion. After treatment with 5-Aza-CdR, E-cadherin mRNA (CDH1) was reexpressed in HL-60 and U937 cells (Figure 1A). Quantitative real-time PCR assay revealed that the expression level of CDH1 was increased significantly in a dose-dependent manner. Indirect immunofluorescence staining confirmed that the recovery of mRNA expression resulted in an increase in cell surface E-cadherin expression (Figure 1B).
E-cadherin-dependent homophilic adhesion assay was performed to determine whether E-cadherin reexpression could induce cell adhesion. After E-cadherin was restored by 5-Aza-CdR in HL-60 cell, cell adhesion with E-cadherin–Fc was enhanced (Figure 2). 5-Aza-CdR-treated cells showed a 1.4-fold increase in adhesion ability when compared with control cells. From the E-cadherin-blocking experiments, in the presence of calcium chelator EGTA or the block antibody against the extracellular domain of E-cadherin, adhesion ability of 5-Aza-CdR-treated cells was completely inhibited. Enhanced adhesion was not observed in 5-Aza-CdR-treated cells when cultured on human IgG-coated plates. Similarly, increased binding was also observed in U937 cells plated on E-cadherin–Fc following 5-Aza-CdR treatment (Figure 2). Statistical analysis showed that significant difference existed between 5-Aza-CdR-treatment compared with 5-Aza-CdR plus anti-E-cadherin antibody with P-values of 0.037 in HL-60 cells and 0.042 in U937 cells. These data confirmed that 5-Aza-CdR-induced cell binding was highly attributed to E-cadherin. The results demonstrate that restoration of E-cadherin expression in leukaemia cells leads to a reestablishment of E-cadherin-mediated cell adhesion. Low expression of E-cadherin in leukaemia cells may result in a reduction in E-cadherin-mediated cell homophilic adhesion.
3.2. E-cadherin-specific siRNA treatment reduces cell–cell adhesion
We then further determined if low expression of E-cadherin may result in abnormal interaction of leukaemic cell with stromal cell. RNA interference experiments were performed in E-cadherin-expressing Raji cells. To determine whether silencing of E-cadherin affects the interaction of Raji cells with stromal cells, we selected the E-cadherin-positive cell 293T, a human embryonic kidney cell line, to serve as stromal cells supporting Raji cell growth. E-cadherin-specific siRNA treatments were applied in both cell lines. After cells were transfected with SiCDH1, E-cadherin protein levels were markedly reduced in Raji and 293T cells (Figure 3A).
At 48 h after siRNA treatment, direct interaction of Raji cells with stromal cell was determined by plating Raji cells on 293T cells. Silencing of E-cadherin by siRNA in either Raji or 293T cells resulted in the adhesion inhibition in Raji cells (Figure 3B). The most significant inhibition was present in the group of both E-cadherin siRNA-treated cell lines with the inhibition rate of 63%. The results showed that E-cadherin-mediated homophilic adhesion was effectively blocked by E-cadherin siRNA treatment.
3.3. E-cadherin-specific siRNA treatment promotes proliferation of Raji cells on 293T stromal cells
We next examined the possibility that low expression of E-cadherin may provide a growth advantage to leukaemia cells through escaping from E-cadherin-mediated homophilic adhesion. After E-cadherin was down-regulated and E-cadherin-mediated homophilic adhesion was blocked, we determined the ability of Raji cells proliferation by culturing Raji cells on 293T stromal cells for 24–72 h. In MTT proliferation assay, SiCDH1-treated Raji cells, co-cultured with 293T cells, exhibited an increased proliferative capacity compared with SiCon-treated Raji cells plated on SiCon-treated 293T cells within 72 h. The most significant increase in cell proliferation was observed when Raji and 293T cells were both treated with SiCDH1 and were co-cultured for 48 h (P = 0.03) (Figure 4A). These data indicate that E-cadherin-mediated homophilic adhesion plays an important role in restricting cell proliferation.
3.4. E-cadherin-specific siRNA treatment promotes leukaemic cell colony formation
To investigate whether the deficiency in E-cadherin is involved in leukaemic cell growth and transforming ability, we then examined the effect of down-regulation of E-cadherin on leukaemic cell colony formation. The colony formation ability of Raji cells is presented and that the colony number of the cells treated with siCDH1 was higher than that of cells treated with siCon and untreated Raji cells with P-values of 0.005 and 0.001, respectively (Figure 4B). The data indicate that down-regulation of E-cadherin expression enhances leukaemia cell transformation. It could be inferred that low expression of E-cadherin may contribute to the transforming ability of leukaemia cells.
E-cadherin, which is one of the most important molecules involved in tissue morphogenesis and the maintenance of tissue integrity, also functions as a tumour suppressor protein (Gumbiner, 1996; Halbleib and Nelson, 2006). Numerous studies have reported a strong correlation between E-cadherin loss and the initiation of tumours (Berx and Van Roy, 2001). In our previous study, we measured the expression levels of E-cadherin in 91 leukaemia patients and found that the E-cadherin expression levels were decreased or absent in the majority of the leukaemia patients. To further investigate the significance of underexpression of E-cadherin in leukaemia, we treated E-cadherin-silenced leukaemia cells with a demethylating agent to restore E-cadherin expression and examined its effects on E-cadherin-mediated cell adhesion. After being treated with 5-Aza-CdR, E-cadherin-silenced leukaemia cells exhibited cell membrane E-cadherin expression, and the restoration of E-cadherin expression in leukaemia cells induces E-cadherin-specific adhesion. It suggests that low expression of E-cadherin in leukaemia cells may result in a reduction of cell homophilic adhesion.
Hypermethylation-associated silencing of numerous genes has been found in leukaemia cells; therefore, treatment with the demethylating agent in leukaemia cells may induce a number of gene expressions. Although a growth inhibition was observed in leukaemia cells after 5-Aza-CdR treatment (data not shown), we cannot conclude that the proliferation inhibition is caused by E-cadherin reexpression, since the proliferation assay is not E-cadherin specific. The role of E-cadherin in leukaemia cell proliferation was then determined by E-cadherin-specific RNA interference assay. To verify whether loss of E-cadherin plays a role in the proliferation of leukaemic cells and evaluate the influence of reduction in E-cadherin on the interaction of leukaemia cells with stromal cells, we used E-cadherin-specific siRNA to down-regulate E-cadherin expression in Raji cells and examined the effects on cell adhesion and proliferation. The results showed that knocking down of E-cadherin caused a significant reduction in cell–cell adhesion and promotion of cell growth, as well as an enhancement in colony formation. These results indicate that E-cadherin contributes to the homophilic adhesion and restricts cell proliferation via E-cadherin-mediated interaction with stromal cells. In leukaemia cells, low expression of E-cadherin may favour the vigorous growth and transforming ability of leukaemic cells through escaping the restraint caused by E-cadherin-mediated homophilic adhesion.
E-cadherin is a tumour suppressor protein with a well-established role in cell–cell adhesion. E-cadherin could inactivate Wnt signal pathway by binding and sequestering β-catenin from the nuclear signalling pool, which contributes to its tumour suppression function. In our previous study on leukaemic cell, we have found that cell membrane localization of β-catenin was correlated with E-cadherin expression and demonstrates that E-cadherin-mediated adhesions could recruit β-catenin to cell membrane (Rao et al., 2008). Loss of E-cadherin protein may contribute to aberrant nuclear localization of β-catenin and its target proto-oncogene activation, which may favour transforming ability of leukaemic cells.
More evidence suggests that cell adhesion plays a central role in the interaction between the developing haematopoietic cells with their microenvironment, which modulates the fate of haematopoietic stem cells (Kaplan et al., 2007). In our study, demethylation treatment as well as siRNA-mediated RNA interference, demonstrate that E-cadherin silence in leukaemia cells results in a loss of E-cadherin-mediated homophilic adhesion. Several studies have demonstrated that adhesion capacity to bone marrow stroma was impaired in leukaemia cells (Zhou et al., 2005). Abnormal interaction may, at least partially, cause the premature egress of haematopoietic cells in leukaemia, which is one of the hallmarks of leukaemia cells (Tavor et al., 2005). Alterations of adhesion behaviour of haematopoietic stem cells may alter the homoeostasis of haematopoietic cells with bone marrow microenvironment and ultimately lead to leukaemic transformation.
In summary, our study showed that low expression of E-cadherin caused the reduction of cell homophilic adhesion, therefore contributing to the vigorous growth and transformation of leukaemic cells. Our findings not only revealed the role of loss of E-cadherin expression in leukaemia cell behaviours, but also provide new insights into the importance of the abnormal interaction of leukaemic cells with bone marrow microenvironment in leukaemogenesis.
All authors participated in the design, interpretation of the studies and analysis of the data and review of the manuscript. Qing Rao designed and performed the research, analysed the results and wrote the manuscript. Ji-Ying Wang designed the small interfering RNA and analysed the data. Jihong Meng performed real-time quantitative PCR assay. Kejing Tang performed Western blot assay and immunofluorescence assay. Yanzhong Wang performed statistical analysis and Figure design. Min Wang assisted with the design of the study and with critical examination of the manuscript. Haiyan Xing and Zheng Tian contributed to siRNA transfection and cell proliferation analysis. Jianxiang Wang supervised the project, participated in the design and interpretation of the study and approved the final manuscript.
This work was supported by the
Corn, PG, Smith, BD, Ruckdeschel, ES, Douglas, D, Baylin, SB and Herman, JG (2000) E-cadherin expression is silenced by 5′ CpG island methylation in acute leukemia. Clin Cancer Res 6, 4243-8
Gottardi, CJ, Wong, E and Gumbiner, BM (2001) E-cadherin suppresses cellular transformation by inhibiting beta-catenin signaling in an adhesion-independent manner. J Cell Biol 153, 1049-60
Gumbiner, B, Stevenson, B and Grimaldi, A (1988) The role of the cell adhesion molecule uvomorulin in the formation and maintenance of the epithelial junctional complex. J Cell Biol 107, 1575-87
Konopleva, M, Konoplev, S, Hu, W, Zaritskey, AY, Afanasiev, BV and Andreeff, M (2002) Stromal cells prevent apoptosis of AML cells by up-regulation of anti-apoptotic proteins. Leukemia 16, 1713-24
Meng, J, Rao, Q, Wang, M, Wang, D, Li, M and Wang, Y (2008) Expression of E-cadherin in leukemia patients and the mechanisms of E-cadherin down-regulation in leukemic cell. Chinese J Pathol 24, 2044-7
Rao, Q, Wang, J, Meng, J, Tang, K, Tian, Z and Xing, H (2008) Leukemia cell surface expression of E-cadherin and its correlation with membrane localization of beta-catenin. Zhonghua Xue Ye Xue Za Zhi 29, 592-4
Tavor, S, Petit, I, Porozov, S, Goichberg, P, Avigdor, A and Sagiv, S (2005) Motility, proliferation, and egress to the circulation of human AML cells are elastase dependent in NOD/SCID chimeric mice. Blood 106, 2120-7
Turel, KR and Rao, SG (1998) Expression of the cell adhesion molecule E-cadherin by the human bone marrow stromal cells and its probable role in CD34(+) stem cell adhesion. Cell Biol Int 22, 641-8
Zhou, J, Mauerer, K, Farina, L and Gribben, JG (2005) The role of the tumor microenvironment in hematological malignancies and implication for therapy. Front Biosci 10, 1581-96
Received 17 June 2010/1 December 2010; accepted 14 December 2010
Published as Cell Biology International Immediate Publication 14 December 2010, doi:10.1042/CBI20100456
© The Author(s) Journal compilation © 2011 Portland Press Limited
ISSN Print: 1065-6995
ISSN Electronic: 1095-8355
Published by Portland Press Limited on behalf of the International Federation for Cell Biology (IFCB)
Figure 1 Restoration of E-cadherin expression in leukaemia cells treated with demethylating agent 5-Aza-CdR
Figure 2 Enhancement of E-cadherin-mediated cell adhesion by 5-Aza-CdR treatment in HL-60 or U937 cells
Figure 3 Effects of siRNA-mediated silencing of E-cadherin on the adhesion of Raji cells on 293T stromal cells