Several families of genes by and large located on the X chromosome encode proteins of unspecified function. Commonly known as cancer/testis (CT) antigens, they are considered, under normal conditions, only to be expressed in cells of the germ line and placenta. CT genes are also often expressed in cancer cells, hence their classification. Here we report that their expression in normal cells is wider spread and can be observed in cells with the potential for self-renewal and pleuripotency, namely, stem cells. Several CT genes and their products, CT antigens, including SSX, NY-ESO-1, and N-RAGE, were expressed in undifferentiated mesenchymal stem cells (MSCs) and down-regulated after osteocyte and adipocyte differentiation. To elucidate the possible overlapping function played by these genes in cancer and stem cells, a comparative analysis of the localization of their proteins was made. In addition, localization relative to other MSC markers was examined. This revealed that SSX localizes in the cytoplasm and overlap occurs in regions where matrix metalloproteinase 2 (MMP2) and vimentin accumulate. Nevertheless, it was found that no protein interactions between these molecules occur. Further investigation revealed that the migration of a melanoma cell line (DFW), which expresses SSX, MMP2, and vimentin, decreases when SSX is down-regulated. This decrease in cell migration was paralleled by a reduction in MMP2 levels. Analogous to this, SSX expression is down-regulated in MSCs after differentiation; concomitantly a reduction in MMP2 levels occurs. In addition, E-cadherin expression increases, mimicking a mesenchymal epithelial transition. These results afford SSX a functional role in normal stem cell migration and suggest a potentially similar function in cancer cell metastases.

A subgroup of tumor-specific antigens, commonly referred to as cancer/testis (CT) antigens, are expressed in several different histologic tumor types, yet there is no evidence of their presence being lineage specific (1, 2). The genes that encode these antigens are normally members of multigene families that map, typically but not exclusively, to the X chromosome. According to classification, the only normal cells that express CT genes are those of the germ line and placenta (3, 4). Nevertheless, for several CT genes, exceptions to these criteria have been described, another of which will be presented in this report.

CT antigens and genes were originally identified through a variety of methods. These include T-cell epitope labeling (5), serologic analysis of cDNA expression libraries (SEREX; refs. 6, 7), differential gene expression analysis (8), and with the aid of bioinformatics (9–11). To date, very little is known of the physiologic function(s) of these antigens or the mode of regulation of expression of their gene families (12). However, it has been shown that in the case of certain CT genes, reactivation of expression in cancerous cells can be due to a loss of epigenetic regulation, as occurs when methylated regions of chromatin are demethylated or when histones usually deacetylated are acetylated (13). Despite being expressed in several cancer types, relatively little is known of the role that CT antigens might play in cancer cell biology. Could the reactivation of their genes simply be an anomaly of deregulated expression or might they in one way or another contribute to the proliferative and metastatic potential of neoplasms? A fascinating analogy that has been raised is the seemingly similar and shared characteristics of stem cells and tumorigenic cells (14–16). This, combined with the belief that the expression of CT genes in normal adult tissue is initiated and restricted to the perpetually proliferating spermatogonial stem cells (13, 17), prompted us to investigate whether normal/healthy cells, capable of self-renewal and migration, might indeed also express these genes and their antigens.

Human postnatal bone marrow and fetal liver contain mesenchymal stem cells (MSCs), which are capable of replicating as undifferentiated cells that retain multilineage differentiation capabilities (18–22). MSCs have been characterized phenotypically as nonhematopoietic cells because they do not express the hematopoietic markers CD34, CD45, or CD14 (23). Upon proper induction, MSCs form cartilage, adipose, bone, and muscle, among other tissues, and promote the expansion of hematopoietic stem cells (18–22, 24, 25). Here we report that several CT genes and antigens are expressed in fetal and adult MSCs.

The CT genes examined in this study include NY-ESO-1 and members of the melanoma antigen family A (MAGE-A) and GAGE, RAGE, and SSX families. MAGE-A family members fall within a larger MAGE subgroup known as type I. All members of this subgroup, at least at the mRNA level, exhibit a cancer/testis-associated expression profile (10, 26–29). The GAGE family consists of at least nine members (30), and transcripts from these members have been discovered in Ewing's sarcoma, melanoma, lung carcinoma, and the placenta (31–34). Another highly immunogenic CT antigen, NY-ESO-1, has also been detected in numerous types of malignancies (35). The SSX family consists of nine homologues, only five of which have been reported to be expressed. SSX1, SSX2, and SSX4 are some of the most prevalently expressed CT genes in malignancies and are commonly fused to the SS18 gene in synovial sarcoma (36). SS18 associates with the human SNF/SWI complex, which has been characterized as a chromatin remodeling factor (37). The NH2 terminus of SSX harbors a putative Kruppel-associated box repression domain, whereas at the COOH terminus a novel repression domain has been identified (38). SSX has been observed to locate in the nucleus, where it colocalizes with the Polycomb group proteins (39). More recently SSX has been shown to interact with two cytoplasmic proteins, namely, RAB3IP, the human homologue of an interactor of the Ras-like GTPase Rab3A, and a novel protein, SSX2IP (40). Thorough reviews on the above-mentioned CT genes as well as others are provided by Scanlan et al. (4) and Zendman et al. (17).

Microarray analysis has recently revealed that several genes involved in cell adhesion, motility, and extracellular matrix interaction form part of the MSC transcriptome. These include vimentin, fibronectin, laminin, and matrix metalloproteinase 2 (MMP2) (41). Functionally, MMP2, also known as gelatinase and type IV collagenase, cleaves a number of cell matrix proteins, including gelatin, collagen, fibronectin, and proteoglycans. Numerous studies have shown a correlation between MMP2 expression and metastatic behavior (42). Vimentin is a component of the intermediate filaments of the cytoskeleton. Tumor cells that undergo an epithelial mesenchymal transition switch their production of intermediate filaments from keratin subtypes, which are expressed in fixed epithelial cells, to vimentin and keratin subtypes expressed by motile mesenchymal cells (42). Breast cancer cell lines that express vimentin (VIM+ breast cancer cells) have been shown to be highly invasive in vitro and highly metastatic in nude mice in comparison with vimentin-negative cell lines (42). In addition to being a prognostic factor in human breast cancer, it has been found that VIM+ breast cancer cells are capable of elaborating the activation of MMP2, necessary for epithelial mesenchymal transition (43, 44). The migratory ability MMP2 and vimentin provide metastatic cancer cells seems also to provide a mechanism by which MSCs can extravasate and migrate to other organs.

Tissue and Cell Samples. Melanoma cell lines were obtained from the ESTDAB repository (through Prof. Rolf Kiessling, Cancer Centrum Karolinska, Karolinska Institute, Sweden). They were cultured in RPMI 1640 with glutamine and 10% fetal bovine serum (FBS, Life Technologies, Inc., Gaithersburg, MD), and passaged on reaching confluence. Healthy testis and muscle tissue were provided by Prof. Olle Larsson (Department of Oncology and Pathology, Karolinska Hospital) and approved by the Ethics Committee at Karolinska Institute, Sweden. Cultured cell lines and tissues were processed for RNA and protein extraction. Normalized human fetal cDNA (Clontech, Palo Alto, CA) tested was isolated from tissues pooled from 10 to 22 male and female Caucasian fetuses ages 16 to 36 weeks. Bone marrow smears were prepared from freshly isolated, normal, adult, total bone marrow samples, with consent from donors.

Harvesting and Ex vivo Culture of Mesenchymal Stem Cells. To isolate human adult MSCs, bone marrow aspirates of 10 to 20 mL were taken from the iliac crest of three donors (MSC 26, 39, and 51; median age, 26; range, 4-58 years). To isolate human fetal MSCs, liver samples from four fetuses (MSC 1, 2, 11, and 17; median gestational age, 9 weeks; range, 7-10 weeks) aborted in the first trimester (patients had volunteered to donate fetal tissue) were collected. Use of bone marrow from healthy volunteers and the donated fetal tissue for the isolation and expansion of MSCs was approved by the Ethics Committee at Karolinska University Hospital. Informed consent was obtained from all donors.

MSCs were isolated and cultured according to methods previously reported (22). In brief, fetal livers were disintegrated by passage through a 100-μm nylon filter. Mononuclear cells from adult bone marrow and fetal livers were isolated from Percoll-separated bone marrow and resuspended in human MSC medium consisting of low-glucose DMEM (DMEM-LG, Life Technologies), supplemented with 10% FBS (Sigma, St. Louis, MO) and 1% antibiotic-antimycotic solution (Life Technologies) and plated at 4 × 103 cells/cm2 in Falcon flasks (Becton Dickinson). The serum lot used was selected based on optimal MSC growth with maximal retention of osteogenic and adipogenic differentiation potential (45). Cultures were maintained at 37°C in a humidified atmosphere containing 5% CO2 and harvested by treatment with trypsin and EDTA (Life Technologies). As determined by flow cytometry, cultured MSCs were uniformly positive for CD166, CD105, CD44, CD29, and CD73 and negative for hematopoietic markers CD14, CD34, and CD45.

Adipogenic and osteogenic differentiation was induced as previously described (19, 46). Briefly, for adipogenic differentiation, cells were seeded at a concentration of 2.1 × 104 cells/cm2 and induced by the addition of 1-methyl-3-isobutylxanthine (0.5 mmol/L), dexamethasone (1 μmol/L), insulin (10 μg/mL), and indomethacin (0.2 μmol/L) to induction medium (DMEM with high glucose and without FBS). The medium was replaced every 3 to 4 days for 15 days, altering between induction and supportive medium (DMEM with high glucose, 10% FBS, and 10 μg/mL insulin). Adipogenesis was measured by the accumulation of neutral lipids in fat vacuoles stained with Oil Red O. For osteogenic differentiation, 3.1 × 103 cells/cm2 were grown in MSC medium supplemented with dexamethasone (0.1 μmol/L), ascorbic acid (0.05 mmol/L), and glycerophosphate (10 mmol/L). The medium was replaced every 3 to 4 days for 15 days. Specimens were stained for mineral deposits with Alizarin red.

Primer Design. CT gene family sequences were obtained from the National Center for Biotechnology Information GenBank database and their open reading frame sequences were aligned. Conserved regions were selected and specific primers were designed to detect and amplify their respective members. For increased sensitivity, a nested PCR approach was designed. Sets of primers are not member specific, but rather family specific, thereby permitting the amplification of numerous homologues of a specific CTA family. Primer details are summarized in Table 1.

Table 1.

PCR primers designed and used to detect CT gene transcripts in tissues and cell lines

PrimerSpecificitySequenceT (°C)Size (bp)
SSXpan A SSX1 to -9 5′-CCT CAG ATG ACT TTC GGC AGG C-′3 65 635 
SSXpan B  5′-TCA CAT CTG GGG AGA GAG GAG G-′3   
SSXpan a  5′-TCA CAT CTG GGG AGA GAG GAG G-′3 65 345 
SSXpan b  5′-GCT ATG CAC CTG ATG ACG AGG G-′3   
MAGE A MAGE-A1, -2, -3, -5, -6, -12 5′-AGT CCT CAG GGA GCC TCC-′3 56 590 
MAGE B  5′-CTG CCG GTA CTC CAG GT-′3   
MAGE a  5′-CAC AAA GGC AGA AAT GCT GG-′3 57 288 
MAGE b  5′-ACT CAG CTC CTC CCA GAT TT-′3   
GAGE1, 2, 8 A GAGE1, -2, and -8 5′-GAC CAA GGC GCT ATG TAC-′3 54 256 
GAGE1, 2, 8 B  5′-GTC CAT CTC CTG CCC CTG-′3   
GAGE1, 2, 8 a  5′-CGT AGA GCC TCC TGA AAT-′3 52 230 
GAGE1, 2, 8 b  5′-CCA TCA GGA CCA TCT TCA-′3   
GAGE3-7b A GAGE3 to -7b 5′-GAC CAA GGC GCT ATG TAC-′3 54 284 
GAGE3-7b B  5′-GCG TTT TCA CCT CCT CTG-′3   
GAGE3-7b a  5′-GCC CGA GCA GTT CAG TG-′3 54 197 
GAGE3-7b b  5′-CCA TCA GGA CCA TCT TCA-′3   
RAGE A RAGE1, -2, and -3 5′-GAA CTG GAA TTG GAA GAA C-′3 51 7881 
RAGE B  5′-ATA TTA GCC CAG AAC ATC C-′3  8422 
    8893 
RAGE a  5′-GGG AGA CTG AAG GGT AGC-′3 54 7011 
RAGE b  5′-AAA CGG ACG CAG GCG CAT-′3  7372 
    7843 
NY-ESO-1 A NY-ESO-1 5′-CCT CGC CAT GCC TTT CGC G-′3 62 381 
NY-ESO-1 B  5′-CCG TCC TCC TCC AGG GA-′3   
NY-ESO-1 a  5′-CCC CAC CGC TCC CCG TG-′3 68 274 
NY-ESO-1 b  5′-CTG GCC ACT CGT GCT GGG A-′3   
PrimerSpecificitySequenceT (°C)Size (bp)
SSXpan A SSX1 to -9 5′-CCT CAG ATG ACT TTC GGC AGG C-′3 65 635 
SSXpan B  5′-TCA CAT CTG GGG AGA GAG GAG G-′3   
SSXpan a  5′-TCA CAT CTG GGG AGA GAG GAG G-′3 65 345 
SSXpan b  5′-GCT ATG CAC CTG ATG ACG AGG G-′3   
MAGE A MAGE-A1, -2, -3, -5, -6, -12 5′-AGT CCT CAG GGA GCC TCC-′3 56 590 
MAGE B  5′-CTG CCG GTA CTC CAG GT-′3   
MAGE a  5′-CAC AAA GGC AGA AAT GCT GG-′3 57 288 
MAGE b  5′-ACT CAG CTC CTC CCA GAT TT-′3   
GAGE1, 2, 8 A GAGE1, -2, and -8 5′-GAC CAA GGC GCT ATG TAC-′3 54 256 
GAGE1, 2, 8 B  5′-GTC CAT CTC CTG CCC CTG-′3   
GAGE1, 2, 8 a  5′-CGT AGA GCC TCC TGA AAT-′3 52 230 
GAGE1, 2, 8 b  5′-CCA TCA GGA CCA TCT TCA-′3   
GAGE3-7b A GAGE3 to -7b 5′-GAC CAA GGC GCT ATG TAC-′3 54 284 
GAGE3-7b B  5′-GCG TTT TCA CCT CCT CTG-′3   
GAGE3-7b a  5′-GCC CGA GCA GTT CAG TG-′3 54 197 
GAGE3-7b b  5′-CCA TCA GGA CCA TCT TCA-′3   
RAGE A RAGE1, -2, and -3 5′-GAA CTG GAA TTG GAA GAA C-′3 51 7881 
RAGE B  5′-ATA TTA GCC CAG AAC ATC C-′3  8422 
    8893 
RAGE a  5′-GGG AGA CTG AAG GGT AGC-′3 54 7011 
RAGE b  5′-AAA CGG ACG CAG GCG CAT-′3  7372 
    7843 
NY-ESO-1 A NY-ESO-1 5′-CCT CGC CAT GCC TTT CGC G-′3 62 381 
NY-ESO-1 B  5′-CCG TCC TCC TCC AGG GA-′3   
NY-ESO-1 a  5′-CCC CAC CGC TCC CCG TG-′3 68 274 
NY-ESO-1 b  5′-CTG GCC ACT CGT GCT GGG A-′3   

NOTE: PCR primers designed and used to detect CT gene transcripts in tissues and cell lines. The reaction protocol is described in Materials and Methods. A, outer forward primer; B, outer reverse primer; a, inner forward primer; b, inner reverse primer. T, PCR annealing temperature. Size, PCR product size. Superscript numbers 1, 2, and 3 refer to RAGE1, -2, and -3, respectively.

RNA Isolation and Reverse Transcription–PCR. RNA was isolated according to standard methods (Sigma). Four microliters of total extracted RNA from each sample were reverse-transcribed using random hexamers or oligo(dT)18 primers (75 ng/μL) and SuperScript II reverse transcriptase (Invitrogen, San Diego, CA), according to the manufacturer's instructions. Two microliters of first-strand cDNA was amplified by PCR with the aid of specific primers in a total reaction volume of 50 μL. One microliter of outer PCR product was then used in the subsequent nested PCR reaction (BioTaq DNA polymerase, Integrated DNA Technologies, Coralville, IA). Expected PCR product sizes for each set of primers are provided in Table 1. Integrity of the cDNA obtained from cell lines and tissues was verified by the amplification of β-actin cDNA. Outer PCR reactions were done as follows: a denaturation step at 94°C for 5 minutes, followed by 30 cycles of a denaturation step at 94°C for 30 seconds, an annealing step at a primer pair–specific temperature (see Table 1) lasting 30 seconds, and an elongation step at 72°C for 45 seconds. The final elongation step lasted 7 minutes. Inner PCR reactions were done similarly to the outer reaction methods, except that only 25 cycles of amplification were carried out. A 5-μL aliquot of each reaction was size fractionated on 1.5% agarose gel and visualized with ethidium bromide staining. Expected fragment sizes were determined by comparing them to a 100-bp DNA ladder (Invitrogen).

Immunofluorescence. MSCs and cell lines, cultured in chamber slides until 50% confluent, were fixed in cold, fresh acetone-methanol (1:1) for 15 minutes and then permeabilized in PBS-Triton X (0.1%). The preparations were then incubated in blocking buffer (bovine serum albumin 2%, Tween 20 0.2%, glycerol 5%, and sodium azide 0.02%) for 30 minutes before a 1-hour incubation with the following antigen-specific antibodies: N-18 (recognizes SSX1, SSX2, SSX4, and SSX5), N-12 (NY-ESO-1), T-20 (N-RAGE), and 6C1 (recognizes MAGE-1, -2, -3, -4, -6, -10, and -12) diluted 1:500 (Santa Cruz Biotechnology, Santa Cruz, CA). The following antibodies were used for the detection of adhesion molecules and matrix components: vimentin (Dakoppats, Glostrup, Denmark), fibronectin (BD Biosciences PharMingen, San Diego, CA), laminin (BD Biosciences PharMingen), and MMP2 (Oncogene Research Products, San Diego, CA). The cell preparations were washed at least thrice with PBS-Tween (0.05%), and then incubated with FITC-conjugated secondary antibodies (Dakoppats). When double immunofluorescence was done, FITC- and Texas Red–conjugated secondary antibodies (1:2000) were used. After washing off the excess of nonincorporated reagents with PBS-Tween, the cells were mounted and counterstained in 4′,6-diamidino-2-phenylindole–containing mounting solution (Vector Laboratories, Burlingame, CA), and visualized by confocal fluorescence microscopy using an Axioplan-2 Imaging system (Zeiss, Thornwood, NY). The images were captured with a cooled, charged-coupled device camera and analyzed with the Axiovision 3.1 software (Zeiss).

Immunoprecipitation and Immunoblotting. Cultured cells and tissue samples were lysed in lysis buffer [50 mmol/L Tris-HCl (pH 7.8), 300 mmol/L NaCl, 10 mmol/L NaF, 5 mmol/L EDTA, 1% Triton X-100, 0.2% sarcosyl, 1 mmol/L dithiothreitol, and 10% glycerol]. Samples were then sonicated for 1 minute (3 × 20 seconds), centrifuged for 10 minutes at 14,000 × g, and the resulting supernatant was collected for analysis. Approximately 15 μg of sample protein was subsequently electrophoresed on 12% NuPAGE Novex Bis-Tris Gels, using the SeeBlue Plus2 Prestained Standard as molecular weight marker. All procedures were carried out according to the manufacturer's specifications (Invitrogen). Protein G-coupled magnetic beads (Dynal Biotech, Smestad, Oslo, Norway) were used for the immunoprecipitation of SSX, MMP2 and vimentin, with the use of SSX1- [N-18 recognizes SSX1, 2, 4, and 5 (polyclonal raised in goat), Santa Cruz], MMP2- [clone Ab-8 (monoclonal raised in mouse), Oncogene Research Products], and vimentin- (monoclonal raised in mouse, Dakoppats) specific antibodies. Immunoblot detection of SSX, MMP2, and vimentin was carried out using a rabbit polyclonal antibody raised against a peptide spanning the SYT/SSX fusion gene (1:500 dilution; used to avoid overlapping detection of SSX and the immunoprecipitating immunoglobulin light chains; Innovagen, Lund, Sweden), and mouse monoclonal antibodies specific for MMP2 (1:400 dilution; Oncogene Research Products) and vimentin (1:400 dilution; Dakoppats), respectively. Negative controls included mouse and goat serum incubated with MSC extract, and antibodies raised in mouse without incubation with MSC extract. A horseradish peroxidase–coupled, species-specific antibody (1:5000 dilution; Pierce, Rockford, IL) was used as the secondary antibody. The blotted polyvinylidene difluoride membranes (Amersham Pharmacia Biotech, Piscataway, NJ) were incubated with enhanced chemiluminescence's substrate according to the manufacturer's instructions (Pierce), and subsequent detection was done using a charge-coupled device camera and a luminescent image analyser (LAS 1000 plus, Fuji Film, Edison, NJ).

Migration Assay. The role SSX might play in cell motility was assessed by comparing the migration of a melanoma cell line (DFW) which expresses SSX (DFW-SSX+) with an SSX knockdown DFW line (DFW-SSX−). Migration was calculated from the number of cells found to have passed through a 8-μm pore size polycarbonate membrane (Neuro Probe, Inc., Gaithersburg, MD) embedded with collagen (Vitrogen 100, Cohesion, Palo Alto, CA). Wells of a chemotaxis chamber (AP48, Neuro Probe, Inc.) were seeded with 5 × 105 cells. Cells were deprived of FBS 18 hours before starting the assay. Medium containing 5% FBS was used as an attractant; medium without FBS was used a negative control. The assay was stopped 6 hours after initiation. Four replicates of each sample were counted. The assay was done twice.

Cancer/Testis Genes Are Expressed in Mesenchymal Stem Cells and Bone Marrow Cells. Expression of the following CT gene families and their respective members was investigated: from the GAGE family, members 1, 2, 3, 4, 5, 6, 7a, 7b, and 8; from the MAGE-A family, members 1, 2, 3, 5, 6 and 12; from the RAGE family, members 1, 2, and 3; from the SSX family, members 1 to 9; and NY-ESO-1. The results are summarized in Table 2. All samples were tested on three independent occasions and scored as positive (+) if a sample yielded a PCR product of the correct size in any one of the tests. Testis tissue and two melanoma cell lines (Tu 379 and Bl-Mel) were used as positive controls and normal muscle tissue as a negative control in all PCR assays.

Table 2.

CT gene expression by RT-PCR in MSCs and bone marrow

CT gene familyTestisMelanoma
MuscleFetal MSCs
Adult MSCs
Bone marrow
Tu 379BL-Mel123412312
GAGE1, -2, -8 − − − − 
GAGE3 to -7b − − − − 
MAGE-A − − 
NY-ESO-1 − 
RAGE1, -2, -3 − − − 
SSX1 to -9 − 
CT gene familyTestisMelanoma
MuscleFetal MSCs
Adult MSCs
Bone marrow
Tu 379BL-Mel123412312
GAGE1, -2, -8 − − − − 
GAGE3 to -7b − − − − 
MAGE-A − − 
NY-ESO-1 − 
RAGE1, -2, -3 − − − 
SSX1 to -9 − 

NOTE: RT-PCR (see Material and Methods) was used to detect CT gene transcripts in testis (positive control), melanoma cell lines (Tu 379 and BL-Mel), muscle tissue (negative control), fetal MSCs (four samples), adult MSCs (three samples), and bone marrow (two samples). The table summarizes the results obtained from three independent tests. Samples were scored positive (+) if a band of the correct size was observed on any of the test occasions.

At least one member of the GAGE family was expressed in all MSC samples tested. Two MSC samples of fetal origin did not express GAGE1, -2, -8, or GAGE3 to -7b. None of the GAGE members were detected in bone marrow or in normal muscle. Testis and both melanoma lines were positive for GAGE1, -2, -8, and GAGE3 to -7b. The expression pattern of RAGE was similar to that of GAGE. RAGE family members were found expressed in MSCs, testis, and melanoma cell lines, but not in bone marrow or muscle. NY-ESO-1 and SSX were expressed in all fetal and adult MSCs, bone marrow, testis, and melanoma cell lines. Members of the MAGE-A family were detected in MSCs, bone marrow, testis, and Bl-Mel.

In summary; MAGE-A, NY-ESO, and SSX were expressed in fetal and adult MSCs as well as in bone marrow. Differences in the expression of GAGE family members in fetal and adult MSC were observed, with the expression of GAGE appearing to be more prevalent in adult MSC samples. GAGE and RAGE were expressed in MSCs but not in bone marrow.

Localization of Cancer/Testis Antigens in Mesenchymal Stem Cells and Bone Marrow Cells Differs from That Observed in Melanoma Cell Lines. The cellular localization of SSX, MAGE-1, NY-ESO-1, and N-RAGE was investigated by immunofluorescence in fetal and adult MSCs, bone marrow cells, and in two melanoma cell lines, Tu 379 and BL-Mel (Fig. 1A). A high percentage of MSC, Tu 379, and BL-Mel cells expressed the nuclear protein SSX, which localized both in the nuclei and in the cytoplasm of cells. Unexpectedly, in MSCs SSX seemed not only to locate in the nucleus but also the cytoplasm, exhibiting a finely granulated fibrous pattern. However, no firm conclusions could be drawn because a species-matched serum control exhibited significant levels of background with a similar pattern of fluorescence (Fig. 3A). Bone marrow smears inspected for SSX expression revealed that there were indeed SSX positive cells, albeit infrequent. Positive cells were nucleated CD34+ cells (hematopoietic stem cell; Fig. 1B). MAGE was expressed and localized in the cytoplasm and nuclei of adherent subpopulations of the melanoma cells, but was not detected in MSCs. N-RAGE was detected in both MSC and melanoma cell lines. N-RAGE exhibited a speckled pattern in the nuclei of MSC and BL-Mel cells. NY-ESO-1 was expressed in the cytoplasm and nuclei of both melanoma cell lines but seemed only to accumulate in the nucleoli of MSCs. Although bone marrow samples tested positive for MAGE-A and NY-ESO-1 transcripts (Table 2), their corresponding antigens could not be detected by immunofluorescence.

Figure 1.

Expression of CT antigens in MSCs and tumor cell lines, detected by immunofluorescence. Cells were grown in chamber slides, fixed and immunostained with CT antigen–specific antibodies as described in Materials and Methods. Negative controls, cells exposed to FITC conjugate in the absence of the primary antibody. Tu 379 and BL-Mel, melanoma cell lines. Cells are visualized using 40× and 63× objectives as indicated.

Figure 1.

Expression of CT antigens in MSCs and tumor cell lines, detected by immunofluorescence. Cells were grown in chamber slides, fixed and immunostained with CT antigen–specific antibodies as described in Materials and Methods. Negative controls, cells exposed to FITC conjugate in the absence of the primary antibody. Tu 379 and BL-Mel, melanoma cell lines. Cells are visualized using 40× and 63× objectives as indicated.

Close modal

SSX, N-RAGE, NY-ESO-1, and MAGE-A Are Down-regulated during Mesenchymal Stem Cell Differentiation. MSCs are capable of differentiating into various mesoderm-derived tissues; these include bone, cartilage and fat, none of which ordinarily express CT genes. We questioned whether CT antigen expression in MSCs is dependent on cells remaining undifferentiated. To investigate this, adult and fetal MSCs were cultivated in medium that induces adipocytic or osteocytic differentiation. After 15 days, differentiation had reached approximately 60%, as determined with the aid of Alizarin red and Oil Red O staining (Fig. 2A). Using immunofluorescence, we tested these samples for the presence of SSX, N-RAGE, and NY-ESO-1. At the end point of the MSC differentiation, assay cells had reached a high level of confluence and the differentiated cells were less amenable to preparation, making it difficult to observe the immunofluorescence pattern. Even so, it was evident that the differentiated cells exhibit a decrease in the prevalence of these CT antigens, and more so in the case of SSX (Fig. 2B). CT gene expression in differentiated samples was also analyzed by reverse transcription–PCR (RT-PCR). A quantitative method would have provided a more decisive result because the nested PCR approach used only gave inconclusive results. The differentiation experiments were carried out in triplicate, of which no samples could provide a repetitively negative or positive result. Our supposition is that due to incomplete differentiation (60%) of the MSC population, the RNA samples still contained small amounts of CT antigen transcripts, and although at lower levels than the undifferentiated sample, still on occasion yielded a RT-PCR product.

Figure 2.

CT antigen expression and MSC differentiation. A, histochemical [Oil Red O (adipocytes), Alazarin red (osteocytes)] verification of MSC differentiation. Adipogenic differentiation (Adipo diff.) was measured by the accumulation of neutral, lipid-filled vacuoles (red arrow). Osteogenic differentiation (Osteo diff.) was determined by the accumulation of mineral deposits (red arrows). B, expression of NY-ESO-1, N-RAGE, and SSX in MSCs visualized by immunofluorescence, with untreated (freshly plated MSCs) and control cells (MSCs cultivated for the time course of the experiment in the absence of differentiation factors). C, NY-ESO-1, RAGE1 to -3, SSX1 to -9, and β-actin transcripts in differentiated human fetal tissue. Lane 1, brain; lane 2, heart; lane 3, kidney; lane 4, lung; lane 5, liver; lane 6, smooth muscle; lane 7, spleen; lane 8, thymus. Positive controls: lane 9, mouse testis; lane 10, human cDNA control; lane 11, human testis. Lane 12, blank control; lane 13, 100-bp ladder.

Figure 2.

CT antigen expression and MSC differentiation. A, histochemical [Oil Red O (adipocytes), Alazarin red (osteocytes)] verification of MSC differentiation. Adipogenic differentiation (Adipo diff.) was measured by the accumulation of neutral, lipid-filled vacuoles (red arrow). Osteogenic differentiation (Osteo diff.) was determined by the accumulation of mineral deposits (red arrows). B, expression of NY-ESO-1, N-RAGE, and SSX in MSCs visualized by immunofluorescence, with untreated (freshly plated MSCs) and control cells (MSCs cultivated for the time course of the experiment in the absence of differentiation factors). C, NY-ESO-1, RAGE1 to -3, SSX1 to -9, and β-actin transcripts in differentiated human fetal tissue. Lane 1, brain; lane 2, heart; lane 3, kidney; lane 4, lung; lane 5, liver; lane 6, smooth muscle; lane 7, spleen; lane 8, thymus. Positive controls: lane 9, mouse testis; lane 10, human cDNA control; lane 11, human testis. Lane 12, blank control; lane 13, 100-bp ladder.

Close modal

CTAs Are Expressed in Early Stem Cell Progenitors but not in Differentiated Fetal Tissue. The down-regulation of SSX, NY-ESO-1 and N-RAGE in differentiated MSCs prompted us to investigate whether these genes are expressed in differentiated human fetal tissue. Tissue derived cDNA samples from fetal human brain, heart, kidney, liver, lung, muscle (smooth), spleen, and thymus were amplified by PCR using the corresponding primers (Table 1). Adult human testis and total human cDNA were included as positive controls (Fig. 2C). In addition, mouse testis was examined for orthologue transcripts of SSX, NY-ESO-1, and N-RAGE. Low levels of SSX expression was detected in the fetal liver and, in accordance with recently published results (47), SSX transcripts were also detected in mouse testis. Besides this, no transcripts from NY-ESO-1 or RAGE could be detected in differentiated human fetal tissue or in mouse testis.

SSX Partially Colocalizes but Does Not Interact with Matrix Metalloproteinase 2 and Vimentin in Mesenchymal Stem Cells. To further investigate localization of SSX in the cytoplasm of MSCs, an inspection as to where SSX located relative to other abundantly expressed MSC proteins (48) was conducted. Accordingly, double immunofluorescence images were made using SSX-specific antibodies and antibodies raised against cell adhesion (fibronectin and laminin), matrix (vimentin), or proteinase (MMP2) molecules. SSX was located in the cytoplasm of MSCs, exhibiting a finely granulated pattern. Yet, the pattern of fluorescence produced by a species-matched serum control was similar, although less intense, thus preventing us from forming any firm conclusions. Nevertheless, under close inspection slight differences between the two could be observed (Fig. 3A), of which the most discernible were the regions in which intense overlap between SSX and MMP2 and SSX and vimentin occur. Cells photographed for SSX and laminin localization revealed that the molecules overlap, yet no definitive colocalization was evident. The images of SSX relative to fibronectin unambiguously revealed that no colocalization between the molecules occurs.

Figure 3.

SSX and matrix proteins. A, localization of SSX relative to other MSC proteins [SSX (FITC), vimentin and MMP2 (Texas Red)] important for cellular matrix structure, adhesion, and migration. B, a coimmunoprecipitation assay of SSX, MMP2, and vimentin. Protein extracts from MSC 17 were immunoprecipitated with SSX- (S), MMP2- (M), and vimentin- (V) specific antibodies. The precipitates were resolved in 12% Bis/Tris gels and probed. SSX, MMP2, and vimentin precipitates were probed with SYT/SSX-specific antisera and MMP2- and vimentin-specific antibodies. Negative controls include mouse serum (m), goat serum (g), and antibodies only (a) (raised in mouse) C, migration of a melanoma cell line expressing SSX (DFW-SSX+) and not expressing SSX (DFW-SSX−). Values presented are those obtained from the averages of four measurements for each sample. The experiments were duplicated; thus, eight measurements were used for each sample. Bars, SD. D, Western blots illustrating the effect of MSC differentiation [undifferentiated (Und.), adipocyte (Adip.), osteocyte (Oste.)] on the expression of MMP2, vimentin, and E-cadherin. Included is the influence of SSX presence (SSX+) and absence (SSX−), in a melanoma cell line (DFW), on expression of the aforementioned molecules.

Figure 3.

SSX and matrix proteins. A, localization of SSX relative to other MSC proteins [SSX (FITC), vimentin and MMP2 (Texas Red)] important for cellular matrix structure, adhesion, and migration. B, a coimmunoprecipitation assay of SSX, MMP2, and vimentin. Protein extracts from MSC 17 were immunoprecipitated with SSX- (S), MMP2- (M), and vimentin- (V) specific antibodies. The precipitates were resolved in 12% Bis/Tris gels and probed. SSX, MMP2, and vimentin precipitates were probed with SYT/SSX-specific antisera and MMP2- and vimentin-specific antibodies. Negative controls include mouse serum (m), goat serum (g), and antibodies only (a) (raised in mouse) C, migration of a melanoma cell line expressing SSX (DFW-SSX+) and not expressing SSX (DFW-SSX−). Values presented are those obtained from the averages of four measurements for each sample. The experiments were duplicated; thus, eight measurements were used for each sample. Bars, SD. D, Western blots illustrating the effect of MSC differentiation [undifferentiated (Und.), adipocyte (Adip.), osteocyte (Oste.)] on the expression of MMP2, vimentin, and E-cadherin. Included is the influence of SSX presence (SSX+) and absence (SSX−), in a melanoma cell line (DFW), on expression of the aforementioned molecules.

Close modal

The connection between SSX in MSCs and proteins involved in migration was further investigated by attempting to identify protein interactions between SSX, MMP2, and vimentin. Immunoprecipitates from MSC protein extracts were obtained and separated by gel electrophoresis, transferred, and then probed with SSX-, MMP2- and vimentin-specific antibodies. The molecular weight of SSX and immunoglobulin light chains is 25 and 23 kDa, respectively, making the difference in migration of the two molecules difficult to distinguish. We therefore immunoprecipitated SSX with antibodies raised in goat and used SSX-specific antibodies raised in rabbit for immunoblot detection. This approach allowed us to detect SSX without interference from immunoglobulin light chains. Direct interactions between SSX and MMP2 and SSX and vimentin were not observed (Fig. 3B).

Reduced SSX Expression Impairs Migration and Matrix Metalloproteinase 2 Expression. We wished to determine whether SSX expression, being an effector of transcription, might play a role in maintaining a cell's ability to migrate. We thus used a melanoma cell line (DFW), known to express SSX, to determine if indeed this was the case. This was done by comparing the migration of an SSX-knockdown DFW line (DFW-SSX+) with the parent line (DFW-SSX+). We found that in DFW cells with diminished SSX expression the ability to migrate was reduced by 40% (Fig. 3C). This reduction in migration was accompanied by a decrease in MMP2 levels. Neither vimentin nor E-cadherin seems to be influenced by the absence of SSX in the melanoma cells (Fig 3D). Similarly, during MSC differentiation, except in the case of E-cadherin, SSX expression is also halted, MMP2 levels decrease, and vimentin levels remain constant.

One of the characteristics of CT genes is their restricted expression in testis, the ovary, and placenta. In testis and the ovary, CT antigen expression is primarily observed in immature cells of the germ line (spermatogonia and primary oocytes; 48, 49). In this investigation we show that several CT antigens known to be expressed in human cancers are expressed in human MSCs, consequently providing a link between the cellular processes of self-renewal and tumorigenicity. An analogy between stem cells and cancer cells, such as the capacity for self-renewal and proliferation, has long been described and has led to the notion of “cancer stem cells”. Whereas stem cells have the ability to generate normal tissue, cancer stem cells are thought to generate abnormally differentiated tissue with an unregulated rate of proliferation. The widespread and specific expression of CT genes in tumor cells raises the question of whether CT gene expression in these cells is an anomaly of reactivated expression, due possibly to a loss of epigenetic regulation that confers these cells a selective advantage, or could the presence of these CT gene–expressing cells be due to the clonal proliferation of an aberrant cancer stem cell. There is evidence for the latter alternative. For example, it has been reported that cancer cells can arise from gene mutations in normal stem cells (50–52). In addition, the existence of stem cells in tumor cell populations has been documented for leukemia (53, 54), epithelial cancers (52), and, recently, in brain tumors (55).

We have observed in MSCs that the expression of CT antigens, such as SSX, NY-ESO-1, and N-RAGE, are down-regulated as a result of differentiation into adipocytes and osteocytes. Thus, should mechanisms that regulate the normal differentiation and regeneration program become altered; it is plausible that these aberrant MSCs might give rise to preneoplastic stem cells. The distinctive ability of MSCs to migrate and intravasate into blood vessels would then prove advantageous for metastasis.

The cellular localization of CT antigens in MSCs also requires attention because this might be indicative of function. For instance, we have found that NY-ESO-1, previously reported to be a cytoplasmic protein (56), localizes in nucleoli-like structures of MSCs. Numerous nucleolar proteins have been shown to regulate cell proliferation and growth by controlling, among others, ribosomal biosynthesis and p53 function (57). An interesting study by Tsai and McKay (58) led to the identification of nucleostemin, a protein expressed in the nucleoli of central nervous system stem cells, embryonic stem cells, and tumors. Nucleostemin, similar to NY-ESO-1, is down-regulated when stem cells differentiate into dividing progenitors. In addition, it has been shown that nucleostemin effects proliferation of both stem and cancer cells by interacting with p53 in late S and G2 phases of the cell cycle. Thus, nucleolar localization of NY-ESO-1 may be indicative of novel cellular function(s), as well as an association with other nucleolar proteins, such as p53, p14ARF, or nucleostemin.

GAGE and RAGE expression is apparently restricted to MSCs because these genes were not detected in bone marrow, in which the stem cell component is primarily hematopoietic. Although not definitive, a difference in GAGE expression between fetal and adult MSCs was observed. GAGE was expressed in all adult MSC samples but not in all fetal samples. Fetal and adult MSCs possess the same phenotypic characteristics and differentiation potential; however, our previous work has shown that differences between the two cell types do indeed exist (22, 59). For example; adult MSCs proliferate less rapidly, express MHC class I molecules more abundantly, and seem to be more immunosuppressive than fetal MSCs (19, 46, 59–64).

In summary, it has been found that both adult and fetal MSCs express CT genes, and albeit at low levels, CT genes are expressed in cells of the adult bone marrow and fetal liver. Thus, in healthy individuals, these antigens are present in normal cells, cells that are associated with the immune system. In patients exhibiting CT antigen-specific humeral and T-cell responses, do the CT antigens expressed in the stem cells of these individuals remain undetected? If so, is this due to the unique immunoevasive properties attributed to MSCs? Note that the immunosuppressive properties harbored by MSCs may indeed lend themselves to the immunoevasive characteristics observed in cancer cells.

On initial inspection, SSX, previously reported to be a nuclear protein (65), seemed to localize primarily in the cytoplasm, with faint evidence of nuclear localization, and in the case of freshly plated MSCs, intense accumulation in regions that resemble philopodia. In addition, there is evidence that CD34+ cells also express SSX. Microarray analysis has recently revealed that a number of genes involved in cell adhesion, motility, and extracellular matrix interaction form part of the MSC transcriptome (41). These include vimentin, fibronectin, laminin, and MMP2. Thus, to further investigate the cytoplasmic localization of SSX in MSCs, an inspection about where SSX located relative to fibronectin, laminin, vimentin, and MMP2 was conducted. Immunofluorescence images made from a non–immune-serum control gave a similar, yet distinguishable, pattern of fluorescence to that of SSX. The most discernible differences (Fig. 3A) were observed in the SSX/MMP2 and SSX/vimentin images. Note also that the cells photographed for SSX/vimentin fluorescence were more rounded and did not share morphology entirely analogous to that seen in the SSX/MMP2 cells. The distinctive pattern of fluorescent overlap observed in the cytoplasm between SSX and MMP2 and SSX and vimentin raised the question about whether these molecules may indeed interact with each other. Consequently, an in vitro coimmunoprecipitation study for SSX and MMP2 and vimentin interaction was conducted. No interactions, however, were found. Nevertheless, the overlap in fluorescence between these molecules suggested that SSX might be implicated in contributing to cell migration. Accordingly, cell migration assays were done. The melanoma cell line (DFW), which expresses SSX, was used to determine if this was indeed the case, and it was found that DFW-SSX+ cells, when compared with DFW-SSX− cells, migrated more rapidly (Fig. 3B). This, combined with the previous finding that SSX does not directly interact with either of the investigated molecules required for migration, begged the question as to how impaired migration was brought about by reduced SSX expression. Consequently, the impact of SSX knockdown on MMP2, vimentin, and E-cadherin expression was investigated. It was found that neither vimentin nor E-cadherin were affected by a lack of SSX. Nevertheless, MMP2 was affected. Ablation of SSX was accompanied by a reduction in MMP2. The same molecules were probed for in undifferentiated MSCs and their differentiated progenitors. As seen in the immunofluorescence images of Fig. 2A, when MSCs differentiate, SSX expression is halted (Fig. 3D). Similarly in MMP2, once MSCs differentiate (lanes 2 and 3), a reduction in expression occurs. Vimentin levels remain constant both in the undifferentiated and differentiated MSCs; in contrast, as might be expected, the presence of E-cadherin is seen to increase once the cells have committed to a specific lineage. In conclusion, although SSX was observed to localize predominantly in the cytoplasm, most intensely in regions overlapping with MMP2 and vimentin accumulation, no interaction between these proteins was observed; instead, SSX seems to affect cell migration by affecting MMP2 expression. The analogy of characteristics shared by stem cells and cancer cells, combined with our identification of genes active in both unique cell types, has been rewarding, in that it has helped elucidate an additional function attributable to SSX. In addition, the relationship between MMP2 and SSX is a noteworthy observation, because it implicates SSX as participant in the processes of cell migration and, therefore, possibly also in tumor metastasis.

With regard to the therapeutic implications, if cancer stem cells are progenitors to metastatic cells, then by merging our understanding of stem cell characteristics with those of cancer development and progression, it might be possible to improve current cancer therapies. For example, by identifying and selecting molecules responsible for the induction of differentiation, or that target differentiation and/or epithelial mesenchymal transition–associated genes, such as CT genes, the treatment of cancers might be improved. CT antigens have received considerable attention as potential therapeutic targets in anticancer vaccine trials (12, 66). Whether these strategies will have deleterious side effects on the host stem cell population remains to be answered.

Grant support: Swedish Cancer Society (B. Brodin and K. Le Blanc), the Gustav V Foundation (B. Brodin), and the Swedish Cancer Foundation (K. Le Blanc).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Drs. Rolf Kiessling and Aniruddha Choudhury (Department of Immune and Gene Therapy, Cancer Centre Karolinska, Karolinska Institute, Sweden) for providing the melanoma cell lines, Berit Sundberg for technical assistance with the MSCs, and Niina Veitonmäki for technical assistance with the migration assays.

1
Boon T, Cerottini JC, Van den Eynde B, van der Bruggen P, Van Pel A. Tumor antigens recognized by T lymphocytes.
Annu Rev Immunol
1994
;
12
:
337
–65.
2
Chen YT, Scanlan MJ, Sahin U, et al. A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening.
Proc Natl Acad Sci U S A
1997
;
94
:
1914
–8.
3
Old LJ. Cancer/testis (CT) antigens—a new link between gametogenesis and cancer.
Cancer Immun
2001
;
30
:
1
–10.
4
Scanlan MJ, Simpson AJ, Old LJ. The cancer/testis genes: review, standardization, and commentary.
Cancer Immun
2004
;
4
:
1
.
5
van der Bruggen P, Traversari C, Chomez P, et al. A gene encoding an antigen recognized by cytolytic T-lymphocytes on a human melanoma.
Science
1991
;
254
:
1643
–7.
6
Gure AO, Tureci O, Sahin U, et al. SSX: a multigene family with several members transcribed in normal testis and human cancer.
Int J Cancer
1997
;
72
:
965
–71.
7
Jager D, Unkelbach M, Frei C, et al. Identification of tumor-restricted antigens NY-BR-1, SCP-1, and a new cancer/testis-like antigen NW-BR-3 by serological screening of a testicular library with breast cancer serum.
Cancer Immun
2002
;
2
:
5
.
8
Hubank M, Schatz DG. Identifying differences in mRNA expression by representational difference analysis of cDNA.
Nucleic Acids Res
1998
;
22
:
5640
–8.
9
Brinkmann U, Vasmatzis G, Lee B, Yerushalmi N, Essand M, Pastan I. PAGE-1, an X chromosome-linked GAGE-like gene that is expressed in normal and neoplastic prostate, testis, and uterus.
Proc Natl Acad Sci U S A
1998
;
95
:
10757
–62.
10
Lucas S, De Plaen E, Boon T. MAGE-B5, MAGE-B6, MAGE-C2, and MAGE-C3: four new members of the MAGE family with tumor-specific expression.
Int J Cancer
2000
;
87
:
55
–60.
11
Scanlan MJ, Altorki NK, Gure AO, et al. Expression of cancer-testis antigens in lung cancer: definition of bromodomain testis-specific gene (BRDT) as a new CT gene, CT9.
Cancer Lett
2000
;
150
:
155
–64.
12
Renkvist N, Castelli C, Robbins PF, Parmiani G. A listing of human tumor antigens recognized by T cells.
Cancer Immunol Immunother
2001
;
50
:
3
–15.
13
dos Santos NR, Torensma R, de Vries TJ, et al. Heterogeneous expression of the SSX cancer/testis antigens in human melanoma lesions and cell lines.
Cancer Res
2000
;
60
:
1654
–62.
14
Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells.
Nature
2001
;
414
:
105
–11.
15
Brickman JM, Burdon TG. Pluripotency and tumorigenicity.
Nat Genet
2002
;
32
:
557
–8.
16
Nakagawara A. Neural crest development and neuroblastoma: the genetic and biological link.
Prog Brain Res
2004
;
146
:
233
–42.
17
Zendman AJ, Ruiter DJ, Van Muijen GN. Cancer/testis-associated genes: identification, expression profile, and putative function.
J Cell Physiol
2003
;
194
:
272
–88.
18
Friedenstein AJ, Petrakova KV, Kuroksova AI. Heterotopic transplants of bone marrow: analysis of precursor cells for osteogenic and hematopoietic tissue.
Transplantation
1968
;
6
:
230
–47.
19
Pittenger MF, Mackay AM, Beck SC, et al. Multi-lineage potential of adult human mesenchymal stem cells.
Science
1999
;
284
:
143
–7.
20
Prockop DJ. Marrow stromal cells as stem cells for non-hematopoietic.
Science
1997
;
276
:
71
–4.
21
Campagnioli C, Roberts IA, Kumar S. Identification of mesenchymal stem/progenitor cells in human first-trimester fetal blood, liver, and bone marrow.
Blood
2001
;
98
:
2396
–402.
22
Götherström C, Ringdén O, Westgren M, Tammik C, Le Blanc K. Immunomodulatory effects of human fetal liver-derived mesenchymal stem cells.
Bone Marrow Transplant
2003
;
32
:
265
–72.
23
Haynesworth SE, Baber MA, Caplan AI. Cell surface antigens on human marrow-derived mesenchymal cells are detected by monoclonal antibodies.
Bone
1992
;
13
:
69
–80.
24
Noort WA, Kruisselbrink AB, Anker PS, et al. Mesenchymal stem cells promote engraftment of human umbilical cord blood-derived CD34+ cells in NOD/SCID mice.
Exp Hematol
2002
;
30
:
870
–8.
25
Angelopolou M, Novelli E, Grove J et al. Co-transplantation of human mesenchymal stem cells enhances human myelopoiesis and megakaryocytopoeisis in NOD/SCID mice.
Exp Hematol
2003
;
31
:
413
–20.
26
De Plaen E, Arden K, Traversari C, et al. Structure, chromosomal localization, and expression of 12 genes of the MAGE family.
Immunogenetics
1994
;
40
:
360
–9.
27
Muscatelli F, Walker AP, De Plaen E, Stafford AN, Monaco AP. Isolation and characterization of a MAGE gene family in the Xp21.3 region.
Proc Natl Acad Sci U S A
1995
;
92
:
4987
–91.
28
Lucas S, De Smet C, Arden KC, et al. Identification of a new MAGE gene with tumor-specific expression by representational difference analysis.
Cancer Res
1998
;
58
:
743
–52.
29
Chomez P, De Backer O, Bertrand M, De Plaen E, Boon T, Lucas S. An overview of the MAGE gene family with the identification of all human members of the family.
Cancer Res
2001
;
61
:
5544
–51.
30
De Backer O, Arden KC, Boretti M, et al. Characterization of the GAGE genes that are expressed in various human cancers and in normal testis.
Cancer Res
1999
;
59
:
3157
–65.
31
Brinkmann U, Vasmatzis G, Lee B, Pastan I. Novel genes in the PAGE and GAGE family of tumor antigens found by homology walking in the dbEST database.
Cancer Res
1999
;
59
:
1445
–8.
32
Liu XF, Helman LJ, Yeung C, Bera TK, Lee B, Pastan I. XAGE-1, a new gene that is frequently expressed in Ewing's sarcoma.
Cancer Res
2000
;
60
:
4752
–5.
33
Wang T, Fan L, Watanabe Y, et al. L552S, an alternatively spliced isoform of XAGE-1, is over-expressed in lung adenocarcinoma.
Oncogene
2001
;
20
:
7699
–709.
34
Zendman AJ, Van Kraats AA, Weidle UH, Ruiter DJ, Van Muijen GN. The XAGE family of cancer/testis-associated genes: alignment and expression profile in normal tissues, melanoma lesions and Ewing's sarcoma.
Int J Cancer
2002
;
99
:
361
–9.
35
Chen YT, Boyer AD, Viars CS, Tsang S, Old LJ, Arden KC. Genomic cloning and localization of CTAG, a gene encoding an autoimmunogenic cancer-testis antigen NY-ESO-1, to human chromosome Xq28.
Cytogenet Cell Genet
1997
;
79
:
237
–40.
36
Clark J, Rocques PJ, Crew AJ, et al. Identification of novel genes, SYT and SSX, involved in the t(X;18)(p11.2;q11.2) translocation found in human synovial sarcoma.
Nat Genet
1994
;
7
:
502
–8.
37
Kato H, Tjernberg A, Zhang W, et al. SYT associates with human SNF/SWI complexes and the C-terminal region of its fusion partner SSX1 targets histones.
J Biol Chem
2002
;
277
:
5498
–505.
38
Lim FL, Soulez M, Koczan D, Thiesen HJ, Knight JC. A KRAB-related domain and a novel transcription repression domain in proteins encoded by SSX genes that are disrupted in human sarcomas.
Oncogene
1998
;
17
:
2013
–8.
39
Soulez M, Saurin AJ, Freemont PS, Knight JC. SSX and the synovial-sarcoma-specific chimeric protein SYT-SSX co-localizes with the human Polycomb group complex.
Oncogene
1999
;
18
:
2739
–46.
40
de Bruijn DR, dos Santos NR, Kater-Baats E, et al. The cancer-related protein SSX2 interacts with the human homologue of a Ras-like GTPase interactor, RAB3IP, and a novel nuclear protein, SSX2IP.
Genes Chromosomes Cancer
2002
;
34
:
285
–98.
41
Silva WA Jr, Covas DT, Panepucci RA, et al. The profile of gene expression of human marrow mesenchymal stem cells.
Stem Cells
2003
;
21
:
661
–9.
42
dos Santos NR, de Bruijn DR, Kater-Baats E, Otte AP, van Kessel AG. Delineation of the protein domains responsible for SYT, SSX, and SYT-SSX nuclear localization.
Exp Cell Res
2000
;
256
:
192
–202.
43
Pulyaeva H, Bueno J, Polette M, et al. MT1-MMP correlates with MMP-2 activation potential seen after epithelial to mesenchymal transition in human breast carcinoma cells.
Clin Exp Metastasis
1997
;
15
:
111
–20.
44
Krishnamachary B, Berg-Dixon S, Kelly B, et al. Regulation of colon carcinoma cell invasion by hypoxia-inducible factor 1.
Cancer Res
2003
;
63
:
1138
–43.
45
Lennon DP, Haynesworth SE, Bruder SP, et al. Development of a serum screen for mesenchymal progenitor cells from bone marrow.
In Vitro Cell Dev Biol
1996
;
32
:
602
–11.
46
Le Blanc K, Tammik L, Zetterberg E, Rosendahl K, Ringdén O. HLA-expression and immunologic properties of undifferentiated and differentiated mesenchymal stem cells.
Exp Hematol
2003
;
31
:
890
–6.
47
Chen YT, Alpen B, Ono T, et al. Identification and characterization of mouse SSX genes: a multigene family on the X chromosome with restricted cancer/testis expression.
Genomics
2003
;
82
:
628
–36.
48
Jungbluth A, Busam K, Kolb D, et al. Expression of MAGE antigens in normal tissue and cancer.
Int J Cancer
2000
;
85
:
460
–5.
49
Jungbluth A, Busam K, Iversen K, et al. Cancer/testis antigens: structural and immunobiological properties.
Cancer Invest
2002
;
20
:
222
–36.
50
Sell S, Pierce GB. Maturation arrest of stem cell differentiation is a common pathway for the cellular origin of teratocarcinomas and epithelial cancers.
Lab Invest
1994
;
70
:
6
–22.
51
Passegue E, Jamieson CH, Ailles LE, Weissman IL. Normal and leukemic hematopoiesis: are leukemias a stem cell disorder or a reacquisition of stem cell characteristics?
Proc Natl Acad Sci U S A
2003
;
100
Suppl 1:
11842
–9.
52
Perez-Losada J, Balmain A. Stem-cell hierarchy in skin cancer.
Nat Rev Cancer
2003
;
3
:
434
–43.
53
Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell.
Nat Med
1997
;
3
:
730
–7.
54
Dick JE. Breast cancer stem cells revealed.
Proc Natl Acad Sci U S A
2003
;
100
:
3547
–9.
55
Singh SK, Clarke ID, Terasaki M, et al. Identification of a cancer stem cell in human brain tumors.
Cancer Res
2003
;
63
:
5821
–8.
56
Schultz-Thater E, Noppen C, Gudat F, et al. NY-ESO-1 tumor associated antigen is a cytoplasmic protein detectable by specific monoclonal antibodies in cell lines and clinical specimens.
Br J Cancer
2000
;
83
:
204
–8.
57
Drayton S, Peters G. Immortalisation and transformation revisited.
Curr Opin Genet Dev
2002
;
12
:
98
–104.
58
Tsai RY, McKay RD. A nucleolar mechanism controlling cell proliferation in stem cells and cancer cells.
Genes Dev
2002
;
16
:
2991
–3003.
59
Götherström C, Ringdén O, Tammik C, Zetterberg E, Westgren M, Le Blanc K. Immunological properties of human fetal mesenchymal stem cells.
Am J Obstet Gynecol
2004
;
190
:
238
–44.
60
Tse WT, Pendleton JD, Beyer WM, Egalka MC, Guinan EC. Suppression of allogeneic T-cell proliferation by human marrow stromal cells: implications in transplantation.
Transplantation
2003
;
75
:
389
–97.
61
Majumdar M, Keane-Moore M, Buyaner D. Characterization and functionality of cell surface molecules on human mesenchymal stem cells.
J Biomed Sci
2003
;
10
:
228
–41.
62
Rasmusson I, Ringdén O, Sundberg B, Le Blanc K. Mesenchymal stem cells inhibit the formation of cytotoxic T lymphocytes (CTL), but not activated CTL or natural killer (NK) cells.
Transplantation
2003
;
76
:
1208
–13.
63
Le Blanc K, Tammik L, Sundberg B, Haynesworth SE, Ringden O. Mesenchymal stem cells inhibit and stimulate mixed lymphocyte cultures and mitogenic responses independently of the major histocompatibility complex.
Scand J Immunol
2003
;
57
:
11
–20.
64
Bartholomew A, Sturgeon C, Siatskas M, et al. Mesenchymal stem cells suppress lymphocyte proliferation in vitro and prolong skin graft survival in vivo.
Exp Hematol
2002
;
30
:
42
–8.
65
Thompson EW, Price JT. Mechanisms of tumour invasion and metastasis: emerging targets for therapy.
Expert Opin Ther Targets
2002
;
6
:
217
–33.
66
Kirkin AF, Dzhandzhugazyan KN, Zeuthen J. Cancer/testis antigens: structural and immunobiological properties.
Cancer Invest
2002
;
20
:
222
–36.