| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Institute of Immunology, University of Munich, 80336 Munich [E. P., K. W., S. O., A. Z., J. J., G. R., K. P.]; Institute of Pathology, Hospital Cottbus, 03048 Cottbus [P. S.]; and Laboratory of Immunogenetics, 80336 Munich [R. Z.], Germany
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
5 (eight of nine), 6 (five of nine), V (nine of nine), ß1 (nine of nine), and ß3 (nine of nine) integrin subunits and the Mr 67,000 laminin receptor (seven of nine). Contrary to our expectations, metastasis-promoting CD44 variant isoforms were only detected on two lines, whereas all cell lines expressed MUC18/melanoma cell adhesion molecule and intercellular adhesion molecule-1, two members of the immunoglobulin superfamily of adhesion molecules that are not frequently found on primary carcinoma cells. The consistent expression of various epithelial and tumor-associated antigens provides evidence that the established cell lines are derived from disseminated cancer cells present in the BM. The invasive phenotype of the immortalized cells was mirrored by their epithelial-mesenchymal transition and the expression of several metastasis-associated molecules, which might be potential candidates for novel therapeutic targets. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It would, therefore, be of great interest to identify the set of molecules that determine the affinity of epithelial tumor cells to BM. Immunocytochemical double marker analyses have indicated that CK+ cells in BM are a selected but still heterogeneous population of tumor cells with regard to the expression of cell adhesion receptors, oncogenes, proliferation markers, or MHC class I antigens (17, 18, 19)
. However, these evaluations were limited by the extremely low frequency of CK+ cells (
10-5–10-6). This obstacle has been overcome by the development of appropriate cell culture conditions and an experimental immortalization strategy that allows the generation of permanent cell lines from BM cancer cells (20)
. Here, we report the phenotypic characterization of nine of these unique cell lines with a focus on histogenetic marker molecules and metastasis-associated cell adhesion receptors. Our data indicate that these cell lines represent good working models of human BM micrometastases because they demonstrate many properties of in situ epithelial tumor cells. Implementation of these models into the development of new drugs might improve the efficacy of current anticancer therapy because micrometastatic tumor cells in BM should be better therapeutic targets than overt metastases, because these cells are easily accessible by i.v. injected drugs, and because the total tumor burden is significantly lower (i.e., in the range of 106–108 tumor cells per total BM; Ref. 21
).
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The BM cells were washed with Hank’s medium (Biochrom, Berlin, Germany) at 170 x g for 10 min, and MNCs were obtained by Ficoll density centrifugation at 1230 x g for 30 min. MNCs were washed at room temperature at 520 x g for 10 min, and an aliquot of 2 x 106 MNCs was then cytocentrifuged onto glass slides at room temperature at 130 x g for 5 min (106 cells per slide). After overnight air drying, slides were either stained immediately or stored at -80°C before use; after such preparation and storage, antigens were found to be preserved for at least 2 years4
. From each aspirate, one slide (106 MNCs) was examined, whereas one additional slide served as an immunoglobulin (immunoglobulin) isotype control.
|
Establishment of Tumor Cell Lines.
Between 8 x 106 and 150 x 106 MNCs per flask were initially plated in T25 culture flasks coated with extracellular matrix (Paesel & Lorei, Frankfurt, Germany; Table 1
). The extracellular matrix material is similar in organization and composition to naturally occurring basement membranes. Cells were incubated in 5–10% CO2 (incubator B 5061 EK/CO2; Heraeus, München, Germany). The culture medium contained RPMI 1640 supplemented with 10% FCS, 10 µg/ml transferrin, 5 µg/ml insulin, 2 mM glutamine, 10 ng/ml recombinant human epidermal growth factor (Boehringer Mannheim GmbH, Mannheim, Germany), synthetic androgen (R 1881, in case of prostate cells) and 10 ng/ml recombinant human basic fibroblast growth factor (PBH, Hannover, Germany). The cells were cultured under reduced oxygen, the medium was changed twice a week, and fresh growth factors were added to the cultures. At confluence, the adherent cells (including the epithelial tumor cells) were removed by trypsinization and passaged in new extracellular matrix-coated flasks.
The plasmid construct pUC In wt was cloned by integration of the SV40 genome fragments PstI/BstXI (2471 bp) and BstXI/BamHI (2226 bp) from the plasmid pSV In1 into the plasmid vector pUC12. The fragments from pSV In1 contain a disrupted origin of replication generated by an insertion of 1 bp in the center of the 27-bp palindrome, thereby destroying the BglI site. The plasmid contains the complete SV40 promoter/enhancer (nucleotides 1988–ori–2533). The pSV40 plasmid (5 µg per T25 culture flask; Ref. 20 ) was transfected at passage 2–4 when the cells were grown to 50–70% confluency using either microinjection, as described recently (20) , or 12.5 µl of the lipophilic transfection reagent lipofectin (Life Technologies, Inc., Eggenstein, Germany) in serum containing medium, according to the manufacturer’s instructions.
Immunocytochemical and Flow Cytometric Analyses of Cell Lines.
At different time points (Table 1
and Fig. 1
) adherent cells were detached by trypsin-EDTA treatment. The expression of SV40 T-Ag was monitored by immunocytochemical analysis (alkaline phosphatase anti-alkaline phosphatase alkaline technique, as described above), using anti-SV40 T-Ag mAbs Pab 101, 220, 416, and 419 (provided by Dr. Ellen Fanning, Vanderbilt University, Nashville, TN). For flow cytometric analysis, cells were fixed for 10 min at 4°C with either 80% methanol (staining of -smooth muscle actin, vimentin, E-cadherin, or androgen receptor) or 80% acetone (staining of CK) diluted in 19.8% PBS and 0.2% BSA; all other stainings were performed with unfixed cells. The primary antibodies were used at optimal concentrations of 2.5 to 10 µg/ml or as undiluted hybridoma supernatants. The following antibody clones were used: 1A4, anti-smooth muscle actin (DAKO); F39.4.1, anti-androgen receptor (Biogenex, San Ramon, CA); MEM48, anti-CD-18 (anti-ß2-integrin, Southern Biotechnology Associates, Birmingham, AL); 2A4, anti-CD29 (anti-ß1-integrin; Southern Biotechnology Associates); BIRMA-K3, anti-CD34 (DAKO); SAM-1, anti-CD49e (anti-5-integrin, Southern Biotechnology Associates); GoH3, anti-CD49f (anti-6-integrin, PharMingen, Hamburg, Germany); 23C6, anti-CD51 (anti-v-integrin, Southern Biotechnology Associates); BL-E6, anti-CD61 (anti-ß3-integrin; Sigma); 450–9D, anti-CD104 (anti-ß4-integrin, PharMingen); IOL44, anti-CD44s (Immunotech, Hamburg, Germany); VFF-7, anti-CD44v7+v8 (Bender, Vienna, Austria); VFF-8, anti-CD44v5 (Bender); VFF-18, anti-CD44v6 (Bender); 5B1, anti-CD45 (Boehringer Mannheim); A45B/B3, anti-pan CK (Micromet, Munich, Germany); CK2, anti-CK-18 (Boehringer Mannheim); anti-pan-CK, AB1/AE3 (DAKO); anti-CK5/6/8/18 LP34 S03 (Medac, Hamburg, Germany) AICD58, anti-CD58 (Immunotech); 50F11, anti-CD82 (PharMingen); D6.3.10, anti-desmoglein (Progen, Heidelberg, Germany); 6F9, anti-E-cadherin (provided by Dr. W. Birchmeier, Max-Delbrueck-Center, Berlin, Germany); 29.1.1, anti-epidermal growth factor receptor (Yeda, Rehovet, Israel); EN-4, anti-endothelial cells (Monosan, Uden, the Netherlands); erbB-7C1, anti-p185erbB2 (provided by J. P. Johnson); 4F9, anti-factor VIII (Immunotech); P3.58BA.3, anti-ICAM-1 (provided by J. P. Johnson); MLuc5, anti-Mr 67,000 laminin receptor (kindly provided by Dr. Maria Colnaghi, Institute for Experimental Cancer Research, Milan, Italy); 55–2LL10, anti-Lewis Y (provided by Dr. Peter Rieber, Institute of Immunology, Dresden, Germany); A76, anti-MUC1 (provided by Dr. U. Karsten, Max-Delbrueck-Center, Berlin, Germany); MUC18, MUC18BA.4; MAd5D7 and MAd1B4, anti-MUC18 (provided by J. P. Johnson); A67B/E3, anti-PSA (Connex, Munich, Germany); CYT-351, anti-PSM (Cytogen, Princeton, NJ); 1G11, anti-VCAM (Immunotech); and V9, anti-vimentin (DAKO). The antibody reaction was developed with a secondary FITC-conjugated rabbit antimouse immunoglobulin antibody (F0313; DAKO), used at a 1:50 dilution. Appropriate IgG and IgM isotype control antibodies were used to ensure specificity of staining. Undiluted culture supernatant of a hybridoma against murine H-2 (141-11, anti-H2; G. Hämmerling, DKFZ, German Cancer Research Centre, Heidelberg, Germany) served as unrelated IgM antibody control.
|
RT-PCR Analysis of MAGE mRNA Expression.
For amplification of MAGE, a PCR mixture (10 µl) was composed of 1 µl of cDNA, 1 µl of 10x PCR buffer [100 mM Tris (pH 8.3), 500 mM KCl, and 10 mM MgCl2], 40 µM dNTP, 0.4 µM each of the two primers, 5 µg of BSA (Boehringer Mannheim), and 0.6 units of Taq DNA-Polymerase (Boehringer Mannheim), overlaid with 12.5 µl of mineral oil. The primers for the MAGE genes were designed from previously published sequences (25)
and were selected to maximize mismatches between the different MAGE sequences, particularly in the 3' region, to avoid cross-amplification5
Specific PCR assays were possible for MAGE-1, -2, -4, and -12 because the percentage of identity of these genes varies between 64 and 85%. In contrast, the sequence of MAGE-6 was found to be 99% identical to that of MAGE-3. Therefore, our primers for MAGE-3/-6 hybridize to both genes, and PCR can detect MAGE-3 and/or MAGE-6 expression. To confirm the presence of cDNA template, a control amplification of cytosine-5 DNA methyltransferase was performed. Oligonucleotide primers were synthesized and purified at Genset (Paris, France). For the external fragment amplification of the MAGE, a PCR assay with the following cycling profile was used: denaturation at 94°C for 6 min, annealing at 60°C for 30 s, and extension at 72°C for 2 min for the first cycle; denaturation at 93°C for 40 s, annealing at 60°C for 30 s, and extension at 72°C for 20 s for 14 cycles; denaturation at 93°C for 40 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s for 50 cycles; terminal extension at 72°C for 2 min; and cooling at 30°C for 1 min. One µl of the reaction was then transferred into a second tube containing the PCR mixture described above. For the nested fragment amplification, 30 more cycles were run at 93°C (40 s), 58°C (30 s), and 72°C (30 s), with a final extension at 72°C for 2 min and cooling at 30°c for 1 min. The amplification was performed on a Hybaid Thermal Reactor (Biometra, Göttingen, Germany) using a plate control. Importantly, all pipetting was performed on ice under a laminar air-flow bench with filtered pipette tips; external and nested fragment amplification were performed in different rooms to avoid potential DNA cross-contamination. Negative control reactions (mock reactions) for the RNA preparation and RT-PCR were performed routinely. PCR products were analyzed on 1.8% agarose by gel electrophoresis and by direct visualization after ethidium bromide staining under UV light.
Isolation of Chromosomal DNA.
Genomic DNA of cell lines or peripheral blood lymphocytes was isolated following the protocol of Miller et al. (26)
. Genomic DNAs from cytospin preparations were isolated as followed: Hettich centrifuge carriers were cleaned with 2 M HCl, and the cytospins were fixed. The cells were overlaid with 80 µl of solution A [100 mM KCl, 10 mM Tris-HCl (pH 8.3), and 2.5 mM MgCl2], 80 µl of solution B [10 mM Tris-HCl (pH 8.3), 2.5 mM MgCl2, 1% Tween 20, and 1% NP40]. Proteinase K was added to a final concentration of 100 µg/ml, and centrifuge carriers were sealed with Parafilm and incubated for 2 h at 56°C. The liquid was transferred into an Eppendorf tube and extracted with phenol/chloroform, and the aqueous DNA solution was frozen at -20°C.
HLA-DRB1* Genotyping.
HLA-DRB1* typing was performed using either genomic DNA of cell lines and peripheral blood lymphocytes or DNA isolated from cytospin preparations of the patients, following the oligonucleotide typing system protocol described by Nevinny-Stickel et al. (27)
.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In seven patients, the initial concentration of CK+ tumor cells in BM was below 1 in 106 MNCs, which was the detection limit of our immunocytochemical assay (Table 1)
. Considering that we plated between 8 x 106 and 150 x 106 MNCs per culture, the estimated initial number of CK+ cells in each culture varied between <9 and 75 (Table 1)
. The establishment of a permanent cell line from these CK+ cells is exemplified for LC-M1 in (Fig. 1)
. CK+ cells started to grow exponentially and were transfected with SV40 T-Ag cDNA after the third culture passage. Crisis occurred in SV40-positive cells after a mean of 14 culture passages (range, 9–19 passages). The clones emerging from the crisis continued to grow exponentially and established permanent cell lines; all of these cells permanently expressed the SV40 T-Ag (Table 1)
. In contrast, BC-K1 and PC-H1, which were also expanded from BM but did not survive crisis, were negative for SV40 T-Ag (Table 1)
, indicating that the transfection was insufficient or only transient. Subsequent analyses were performed on SV40 T-Ag-positive cells after their crisis and on SV40 T-Ag-negative cells (PC-H1 and BC-K1) before their crisis (Table 1)
.
HLA-DRB1* Genotyping.
To verify the origin of our established cell lines from the respective patients and to rule out cross-contaminations in cell culture handling, HLA-DRB1* typing was performed using the genomic DNA of the cell lines in comparison to genomic DNA isolated from the original BM cytospins or, when available, DNA isolated from peripheral blood lymphocytes of the respective patient. In all cases, the established cell lines showed the same typing as their autologous peripheral blood lymphocytes (PC-R1) or BM cells (all other cell lines). The HLA-DRB1* type of all cell lines differed from each other, with the exception of CC-B1 and PC-H1. For DRB1*, the following typing results were obtained: PC-E1, 04/07; PC-H1, 03/6; PC-R1, 04/-; PC-S1, 04/12; BC-H1, 07/11; BC-K1, 03/14; LC-D1, 03/13; LC-M1, 09/11; and CC-B1, 03/6 (data not shown).
Histogenetic and Tumor-associated Antigens.
To confirm the epithelial nature of the established cell lines, we analyzed the expression of CKs as the major constituents of the epithelial cytoskeleton. As shown in Table 2
, all cell lines expressed detectable levels of CK if they were incubated with mAbs A45-B/B3 or CK2. However, staining for CK was only weak in two of these lines (BC-K1 and CC-B1) and heterogeneous in PC-E1 and BC-H1 cells. The presence of CK-negative subpopulations in these lines was confirmed by additional immunostaining with other anti-CK mAbs (AE1/AE3, LP34, and 5D3; data not shown). There was no significant difference between the staining pattern obtained with the broad spectrum anti-CK mAbs A45-B/B3 and mAb CK2 against CK component 18 (data not shown). Interestingly, Lehr and Pienta (28)
recently established a human BM endothelial cell line by transfection with SV40 T-Ag, which may serve as a control to this analysis. These cells lacked expression of CKs, indicating that this typical feature of epithelial tumor cells in BM is not induced by the SV40 oncogene. The epithelial nature of our cell lines was further supported by the consistent expression of E-cadherin, the most prominent epithelial cell adhesion molecule (Table 2)
.
|
. As positive controls, BM cells were stained under the same conditions, and all of the above antigens could be detected on subpopulations of these cells (data not shown). Interestingly, the mesenchymal cytoskeleton protein vimentin was coexpressed with CK in all cell lines (Table 2)
, indicating an epithelial to mesenchymal transition of micrometastatic cells.
The tumor origin of seven of nine cell lines was supported by immunodetection of the ErbB2 proto-oncogene gene product (Table 2)
, which usually results from an overexpression of the p185ErbB2 protein due to an amplification of the ErbB2 gene (29)
. Immunodetection of ErbB2 is a common feature of many epithelial tumors, including cancer of the breast (29, 30, 31)
, colon (32)
, lung (33)
, and prostate (34)
, 35)
. Using immunocytochemical double labeling, we showed previously that BM micrometastases express ErbB2 more frequently than the corresponding primary tumors; e.g., almost 70% of breast cancer patients presented with ErbB2+/CK+ cells in the BM, suggesting a selection bias for ErbB2+/CK+ cells in the BM environment (17)
.
Another common characteristic of cancer cells is the expression of genes of the MAGE family, which consists of at least 12 members (36
, 37)
. Here, we analyzed MAGE mRNA expression, using RT-PCR with primers specific for MAGE-1, -2, -3/-6, -4, and -12. As shown in Table 3
, at least one of these MAGE genes was expressed in all of eight cell lines analyzed (BC-K1 could not be analyzed because of the limited number of cells available). Two lines, PC-S1 and LC-M1, expressed all of the screened MAGE genes, whereas the expression pattern in the remaining lines was more heterogeneous. Interestingly, all of the cell lines expressed MHC class I molecules, as determined by staining with mAb W6/32 against a monomorphic determinant of these molecules (Table 3)
. The cell lines might, therefore, be able to present MAGE-derived peptides in the context of MHC class I molecules. To exclude that MAGE expression was only the result of our immortalization strategy, we analyzed BM cultures from seven cancer patients before transfection with SV40 T-Ag DNA. Consistent with the results on micrometastatic cell lines, MAGE mRNA expression was revealed by RT-PCR analysis in all of these cultures (data not shown).
|
. Three established American Type Culture Collection prostate cancer cell lines (Du145, PC-3, and LNCaP) were used as positive controls. The prostatic marker antigens were heterogeneously expressed in those lines as well as our micrometastatic cancer cell lines. PTI-1 mRNA was detected in all prostate carcinoma cell lines, indicating their prostatic origin. All micrometastatic cells stained negative for androgen receptor protein or PSA, whereas PSM seemed to be weakly expressed on a subpopulation of PC-E1 (5%) and PC-H1 (4%) cells, but this expression could not be confirmed at the mRNA level. However, RT-PCR analysis revealed that PC-S1 weakly expressed PSA mRNA and PCTA mRNA, whereas PC-E1 expressed androgen receptor mRNA (Table 4)
. Moreover, PC-R1 cells expressed mRNA encoding for human prostate-specific transglutaminase, a new marker for prostatic cells (38)
, whereas mRNA encoding kallikrein 2, which belongs to the same protease family as PSA, was only found in the control cell line LNCaP, not in any of our micrometastatic cell lines.
|
. Among the integrins, ß1 (CD29), ß3 (CD61), and V (CD51) integrin subunits stained positive on all cell lines. All lines with the exception of LC-M1 also expressed the 5 integrin (CD49e). On the other hand, the expression of 6 (CD49f) and ß4 (CD104) was more heterogeneous (expression in five and two cell lines, respectively), and none of the cells analyzed were positive for ß2 integrin (CD18).
|
, all cell lines expressed MUC18/MCAM (CD146), ICAM-1 (CD54), and LFA-3 (CD58). Flow cytometric analysis of 17 cell lines derived from primary tumors and metastases of patients with cancers of the breast, colon, prostate, and lung revealed that MUC18/MCAM was only detected in four cell lines (Table 6)
even in those cell lines, MUC18/MCAM was only present on subpopulations of tumor cells consisting of <10% to >50% of the total population. Thus, expression of MUC18/MCAM is not a common phenomenon of cultured epithelial tumor cells. MUC18/MCAM expression also appeared to be not only a consequence of the immortalization with SV40TAg, because both SV40TAg-negative cell lines (PC-H1 and BC-K1), which were not immortal, expressed MUC18/MCAM at the same level as the remaining SV40 T-Ag-positive cell lines (Table 5)
. Moreover, MUC18/MCAM was already expressed on LC-M1 tumor cells before transfection with SV40 T-Ag cDNA (Fig. 1)
. The intensity of MUC18/MCAM staining varied among the cell lines analyzed, with no clear differences between the different tumor entities (data not shown). MUC18/MCAM was not detectable by flow cytometric analysis on normal BM cells (data not shown). However, Filshie et al. (39)
recently demonstrated that MUC18 is expressed on a rare subpopulation (<1%), containing >90% of the stromal BM precursors.
|
. Similarly, CD44 variant isoforms CD44v5 and CD44v6, which have been implicated in metastasis formation (40)
were only detected on CC-B1 and PC-E1 cells, and CD44v7/8 were not expressed at all (Table 5)
, whereas CD44s, the standard form of CD44, was expressed on all cell lines analyzed (Table 5)
.
Among the molecules specifically mediating epithelial cell adhesion, the Mr 67,000 laminin receptor plays a prominent role in the interaction of tumor cells and the basement membrane (41)
. The Mr 67,000 receptor was present on eight cell lines (Table 5)
; however, in seven cell lines, the laminin receptor was only expressed on certain subsets of cells (Table 5)
. Further indication for a disturbance of epithelial cell adhesion was derived from analysis of homotypic adhesion receptors. Although E-cadherin was consistently present in all cell lines tested (Table 2)
, the desmoglein was only expressed in two lines (PC-R1 and PC-S1) and plakoglobin, the common intracytoplasmatic binding protein of both E-cadherin and desmoglein, was only found in two of eight cell lines analyzed (Table 5)
.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, indicating an epithelial to mesenchymal transition of micrometastatic cells that is associated with an increased in vitro motility, plasticity and growth rate of epithelial tumor cells (46
, 47)
. At present, the in situ role of an epithelial to mesenchymal transition in tumor progression and metastasis is still under debate (48
, 49)
. Indirect evidence has been derived from the clinical observation that neoexpression of vimentin on primary breast carcinomas is associated with an increased rate of metastatic relapse (50)
. Positive staining for the receptor tyrosine kinase p185ErbB2 in the majority of our cell lines might reflect a selective growth advantage of tumor cells expressing this growth factor receptor. ErbB2 expression is also a frequent characteristic of dormant BM carcinoma cells in situ (17) , which suggest that ErbB2-positive tumor cells have been selected during the dissemination process. One possible mechanism by which activated ErbB2 might support early dissemination of tumor cells is loss of epithelial cell adhesion through phosphorylation of catenins, the cytoplasmatic binding proteins of E-cadherin (51) . In addition, this analysis indicates that down-regulated expression of other homophilic adhesion receptors, such as desmoglein or plakoglobin, might also contribute to the loss of epithelial cell-cell adhesion required for the onset of tumor cell dissemination.
Originally described as melanoma-associated genes (MAGE), genes of the MAGE family are widely expressed in various epithelial tumors and cell lines derived thereof (36)
. With the exception of normal testis, MAGE antigens are not expressed in normal tissues (36)
. Thus, expression of these antigens on all of the micrometastatic cell lines analyzed provides further proof for their oncogenic nature (Table 3)
. Moreover, the concomitant expression of these antigens and MHC class I molecules on these cell lines (Table 3)
supports the view that these lines may be used to generate a tumor cell vaccine directed against minimal residual cancer (52)
. Because all of our cell lines consistently lacked a detectable expression of B7.1 protein (data not shown), it might be necessary to transfect these cells with this important costimulatory molecule together with cytokine genes, such as interleukin-2 or granulocyte macrophage colony-stimulating factor, to enhance the T cell-activating properties of such a future vaccine.
Among the cell surface molecules that mediate cell-substratum adhesion, integrins play a predominant role in cancer metastasis (53, 54, 55, 56)
. Our present data indicate that 5v, ß1, and ß3 integrin subunits are common characteristics of micrometastatic cells. Expression of the 5 subunit is associated with malignant progression in colon cancer (57)
, and it suppresses apoptosis in HT29 colon carcinoma cells (58)
. The 5ß1 fibronectin receptor is usually not expressed in nonmalignant tissues, but it plays an important role in the anchorage and regulation of fibronectin-mediated tumor cell adhesion in BM (59)
. Another important function of ß1 integrins is to suppress apoptosis in mammary epithelial cells through negative regulation of interleukin-1ß-converting enzyme expression (60)
. On the other hand, the V subunit is frequently expressed in epithelial cells (61)
, and in the form of an Vß3 heterodimer, it mediates osteoclast-bone recognition (62)
and is overexpressed by bone-residing breast cancer cells (63)
. Thus, it is conceivable that the observed pattern of integrin expression on micrometastatic cancer cells might favor their homing and survival in BM. Targeting of these receptors might be efficient to prevent homing of tumor cells reinfused into cancer patients during an autologous stem cell transplantation.
Besides integrins, the Mr 67,000 laminin receptor plays an additional role in tumor cell-substratum interaction. Expression of this receptor protein is associated with an increased rate of metastatic relapse in breast cancer (64, 65, 66)
. Indirect evidence that this observation might be due to an early onset of micrometastasis is provided by the consistent expression of the Mr 67,000 receptor on at least subsets of micrometastatic cancer cells. Interestingly, the Mr 67,000 receptor was coexpressed with the 6 integrin subunit in five of our cell lines, which might reflect the physical and regulatory association between both adhesion molecules (41)
.
Among the adhesion molecules of the immunoglobulin superfamily, the consistent expression of MUC18/MCAM on micrometastatic cells was unexpected because these molecules are usually undetectable on normal epithelial cells and infrequently expressed on primary carcinomas (67
, 68)
. Both MUC18/MCAM and ICAM-1 have been implicated in the metastasis of melanoma cells (68, 69, 70)
, but their role in carcinoma progression is still unclear. Neoexpression of ICAM-1 is found in 30% of epithelial tumors, whereas expression of MUC18/MCAM in primary carcinomas is even more infrequent (e.g., breast, 11%; colon, 0%; lung, 40%; and stomach, 7%). It might be speculated that a special selective pressure of the BM microenvironment could be responsible for the high frequency of MUC18/MCAM+ tumor cells seeding in this particular "soil." Evidence that the observed expression of MUC18/MCAM was not merely a cell culture artifact is derived from negative staining of several other cell lines established from primary or metastatic carcinomas and cultured under similar conditions as our micrometastatic cell lines. However, it cannot be excluded that MUC18/MCAM expression is up-regulated in cultured micrometastatic cells.
One of the prominent molecules involved in the interaction between cancer cells and the extracellular matrix is the glycoprotein CD44. In general, expression of CD44 variant isoforms has been positively correlated with human cancer progression [reviewed in Günthert (71) ]; however, in prostate cancer patients treated by radical prostatectomy, loss of CD44s and CD44v6 predicted an adverse prognosis (72) . Using mAbs against CD44v5–8, we observed staining of only two cell lines. Because CD44v6 expression on carcinoma cell lines cannot be easily modulated in vitro, it is unlikely that CD44v6 expression was down-regulated under our culture conditions (73) . The infrequent finding of the variant isoforms CD44v5–8 on our cell lines argues against an important role of CD44 as facilitator of BM micrometastasis, whereas the consistent expression of CD44 points to its potential role in cancer micrometastasis, as demonstrated recently in a murine fibrosarcoma model (74) .
Because of the few tumor cells initially present in each of the BM samples, we analyzed cell lines established from these cells. Although the in vitro expansion and subsequent immortalization with SV40 T-Ag cDNA might have altered the phenotype of tumor cells, we did not observe consistent differences between SV40 T-Ag-positive and -negative cells. Furthermore, the expression of CKs, ErbB2, and ICAM-1 antigens and the down-regulation of certain epithelial adhesion molecules (e.g., plakoglobin), which were observed on our cell lines, are also characteristics of in situ micrometastatic cancer cells (17 , 19 , 75) . This is consistent with a number of reports demonstrating retention of phenotypic properties following SV 40 T-Ag-induced immortalization of different epithelial cells (reviewed in Ref. 20 ). Interestingly, six of eight of our patients relapsed after a short postoperative period. Although this evaluation is very preliminary, it suggests that the common antigens found on our lines might be invasive markers.
The established cell lines might be useful models to study the biology of BM micrometastases. Moreover, these lines may help to identify potential target molecules for adjuvant anticancer therapy and to test new anticancer agents specifically directed against minimal residual cancer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by the Deutsche Forschungsgemeinschaft (Bonn, Germany). 
2 To whom requests for reprints should be addressed. Phone: 49 89 5996-685; Fax: 49 89 5996-685; E-mail: pantel{at}ifi.med.uni-muenchen.de
3 The abbreviations used are: mAb, monoclonal antibody; CK, cytokeratin; BM, bone marrow; MNC, mononucleated cell; T-Ag, large T antigen; ICAM-1, intercellular adhesion molecule-1; RT-PCR, reverse transcription-PCR; PSA, prostate-specific antigen; PSM, prostate-specific membrane antigen; PTI-1, prostate tumor inducer-1; MCAM, melanona cell adhesion molecule.
4 K. Pantel, unpublished data.
5 A. Zippelius, P. Kufer, R. Lutterbüse, G. Honold, M. W. Köllermann, R. Oberneder, G. Schlimok, K. Pantel, and G. Riethmüller. Mage gene expression in bone marrow and peripheral blood of cancer patients, manuscript in preparation.
Received 9/29/98. Accepted 10/28/98.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
6ß4 integrin. J. Biol. Chem., 272: 2342-2345, 1997.5ß1 integrin. Exp. Cell Res., 230: 111-120, 1997.[Medline]
Vß3 expression by bone-residing breast cancer metastases. Diagn. Mol. Pathol., 5: 127-135, 1996.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A Vincent-Salomon, F C Bidard, and J Y Pierga Bone marrow micrometastasis in breast cancer: review of detection methods, prognostic impact and biological issues J. Clin. Pathol., May 1, 2008; 61(5): 570 - 576. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Steinert, M. Hantschick, M. Vieth, I. Gastinger, F. Kuhnel, H. Lippert, and M. A. Reymond Influence of Subclinical Tumor Spreading on Survival After Curative Surgery for Colorectal Cancer Arch Surg, February 1, 2008; 143(2): 122 - 128. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Chia, N. Kusuma, R. Anderson, B. Parker, B. Bidwell, L. Zamurs, E. Nice, and N. Pouliot Evidence for a Role of Tumor-Derived Laminin-511 in the Metastatic Progression of Breast Cancer Am. J. Pathol., June 1, 2007; 170(6): 2135 - 2148. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J R Puddefoot, U K I Udeozo, S Barker, and G P Vinson The role of angiotensin II in the regulation of breast cancer cell adhesion and invasion. Endocr. Relat. Cancer, September 1, 2006; 13(3): 895 - 903. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. C. Sroka, M. E. Pennington, and A. E. Cress Synthetic D-amino acid peptide inhibits tumor cell motility on laminin-5 Carcinogenesis, September 1, 2006; 27(9): 1748 - 1757. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Yan and R. Shao Transduction of a Mesenchyme-specific Gene Periostin into 293T Cells Induces Cell Invasive Activity through Epithelial-Mesenchymal Transformation J. Biol. Chem., July 14, 2006; 281(28): 19700 - 19708. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Willipinski-Stapelfeldt, S. Riethdorf, V. Assmann, U. Woelfle, T. Rau, G. Sauter, J. Heukeshoven, and K. Pantel Changes in Cytoskeletal Protein Composition Indicative of an Epithelial-Mesenchymal Transition in Human Micrometastatic and Primary Breast Carcinoma Cells Clin. Cancer Res., November 15, 2005; 11(22): 8006 - 8014. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Dimitroff, L. Descheny, N. Trujillo, R. Kim, V. Nguyen, W. Huang, K. J. Pienta, J. L. Kutok, and M. A. Rubin Identification of Leukocyte E-Selectin Ligands, P-Selectin Glycoprotein Ligand-1 and E-Selectin Ligand-1, on Human Metastatic Prostate Tumor Cells Cancer Res., July 1, 2005; 65(13): 5750 - 5760. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Pantel Bone marrow: Homing organ for dormant micrometastatic cancer cells AACR Meeting Abstracts, April 1, 2005; 2005(1): 1477 - 1478. [Abstract]
|
|
|
|
|
|
D M Peehl Primary cell cultures as models of prostate cancer development Endocr. Relat. Cancer, March 1, 2005; 12(1): 19 - 47. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V Choesmel, F Foucault, J P Thiery, and N Blin Design of a real time quantitative PCR assay to assess global mRNA amplification of small size specimens for microarray hybridisation J. Clin. Pathol., December 1, 2004; 57(12): 1278 - 1287. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Dimitroff, M. Lechpammer, D. Long-Woodward, and J. L. Kutok Rolling of Human Bone-Metastatic Prostate Tumor Cells on Human Bone Marrow Endothelium under Shear Flow Is Mediated by E-Selectin Cancer Res., August 1, 2004; 64(15): 5261 - 5269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Shao, S. Bao, X. Bai, C. Blanchette, R. M. Anderson, T. Dang, M. L. Gishizky, J. R. Marks, and X.-F. Wang Acquired Expression of Periostin by Human Breast Cancers Promotes Tumor Angiogenesis through Up-Regulation of Vascular Endothelial Growth Factor Receptor 2 Expression Mol. Cell. Biol., May 1, 2004; 24(9): 3992 - 4003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Pantel, V. Muller, M. Auer, N. Nusser, N. Harbeck, and S. Braun Detection and Clinical Implications of Early Systemic Tumor Cell Dissemination in Breast Cancer Clin. Cancer Res., December 15, 2003; 9(17): 6326 - 6334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Abdel-Ghany, H.-C. Cheng, R. C. Elble, H. Lin, J. DiBiasio, and B. U. Pauli The Interacting Binding Domains of the {beta}4 Integrin and Calcium-activated Chloride Channels (CLCAs) in Metastasis J. Biol. Chem., December 5, 2003; 278(49): 49406 - 49416. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. L. Davis, F. Buerger, and A. E. Cress Differential Regulation of a Novel Variant of the {alpha}6 Integrin, {alpha}6p Cell Growth Differ., March 1, 2002; 13(3): 107 - 113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Otte, M. Zafrakas, L. Riethdorf, U. Pichlmeier, T. Loning, F. Janicke, and K. Pantel MAGE-A Gene Expression Pattern in Primary Breast Cancer Cancer Res., September 1, 2001; 61(18): 6682 - 6687. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Tomlinson, M. L. Alpaugh, and S. H. Barsky An Intact Overexpressed E-cadherin/{{alpha}},{beta}-Catenin Axis Characterizes the Lymphovascular Emboli of Inflammatory Breast Carcinoma Cancer Res., July 1, 2001; 61(13): 5231 - 5241. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. B. DeRoock, M. E. Pennington, T. C. Sroka, K. S. Lam, G. T. Bowden, E. L. Bair, and A. E. Cress Synthetic Peptides Inhibit Adhesion of Human Tumor Cells to Extracellular Matrix Proteins Cancer Res., April 1, 2001; 61(8): 3308 - 3313. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. Braun and K. Pantel Clinical Significance of Occult Metastatic Cells in Bone Marrow of Breast Cancer Patients Oncologist, April 1, 2001; 6(2): 125 - 132. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. Braun, B. S. Cevatli, C. Assemi, W. Janni, C. R.M. Kentenich, C. Schindlbeck, D. Rjosk, and F. Hepp Comparative Analysis of Micrometastasis to the Bone Marrow and Lymph Nodes of Node-Negative Breast Cancer Patients Receiving No Adjuvant Therapy J. Clin. Oncol., March 1, 2001; 19(5): 1468 - 1475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Braun, G. Schlimok, I. Heumos, G. Schaller, L. Riethdorf, G. Riethmüller, and K. Pantel erbB2 Overexpression on Occult Metastatic Cells in Bone Marrow Predicts Poor Clinical Outcome of Stage I-III Breast Cancer Patients Cancer Res., March 1, 2001; 61(5): 1890 - 1895. [Abstract] [Full Text]
|
|
|
|
|
|
H. Yonou, T. Yokose, T. Kamijo, N. Kanomata, T. Hasebe, K. Nagai, T. Hatano, Y. Ogawa, and A. Ochiai Establishment of a Novel Species- and Tissue-specific Metastasis Model of Human Prostate Cancer in Humanized Non-Obese Diabetic/Severe Combined Immunodeficient Mice Engrafted with Human Adult Lung and Bone Cancer Res., March 1, 2001; 61(5): 2177 - 2182. [Abstract] [Full Text]
|
|
|
|
|
|
S. Braun, C. Schindlbeck, F. Hepp, W. Janni, C. Kentenich, G. Riethmuller, and K. Pantel Occult Tumor Cells in Bone Marrow of Patients With Locoregionally Restricted Ovarian Cancer Predict Early Distant Metastatic Relapse J. Clin. Oncol., January 15, 2001; 19(2): 368 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Trefzer, C. Schlegel, W. Sterry, E. Spath-Schwalbe, K. Possinger, and C. Denkert Fulminant Intravascular Disseminating Malignant Melanoma Mimicking Acute Leukemia Blood, August 15, 1999; 94(4): 1483 - 1484. [Full Text] [PDF]
|
|
|
|
 |
|
 K. Pantel, R. J. Cote, and O. Fodstad Detection and Clinical Importance of Micrometastatic Disease J Natl Cancer Inst, July 7, 1999; 91(13): 1113 - 1124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Offner, W. Schmaus, K. Witter, G. B. Baretton, G. Schlimok, B. Passlick, G. Riethmuller, and K. Pantel p53 gene mutations are not required for early dissemination of cancer cells PNAS, June 8, 1999; 96(12): 6942 - 6946. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Davis, I. Rabinovitz, B. W. Futscher, M. Schnolzer, F. Burger, Y. Liu, M. Kulesz-Martin, and A. E. Cress Identification of a Novel Structural Variant of the alpha 6 Integrin J. Biol. Chem., July 6, 2001; 276(28): 26099 - 26106. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
