Skip to main content
NCBI home page
Cancer Science logoLink to Cancer Science
. 2012 Dec 11;104(1):62–69. doi: 10.1111/cas.12038

HMMC‐1, a human monoclonal antibody to fucosylated core 1 O‐glycan, suppresses growth of uterine endometrial cancer cells

Fumiko Oikawa 1, Kyoko Kojima‐Aikawa 2,✉, Fumika Inoue 2, Atsushi Suzuki 1, Kyoko Tanaka 1, Eiichiro Tominaga 1, Daisuke Aoki 1,✉
PMCID: PMC7657104  PMID: 23035753

Abstract

HMMC‐1 is a human monoclonal antibody that reacts with a fucosylated and extended core 1 O‐glycan, Fucα1‐2Galβ1‐4GlcNAcβ1‐3Galβ1‐3GalNAc‐Ser/Thr, as an epitope. In the present study, we examined the effects of HMMC‐1 on cell proliferation of two human uterine endometrial cancer cell lines, HEC8 and HEC9, to investigate the role of glycoproteins bearing the HMMC‐1 epitope in cancer progression. HEC9 cells expressed high levels of the HMMC‐1 epitope, but HMMC‐1 reactivity was hardly detected in HEC8 cells. In a mouse model of lymph node metastasis using orthotopic implantation, HEC8 and HEC9 showed low (10%) and high (80%) metastatic potency, respectively. Growth of HEC9, but not HEC8, was remarkably inhibited by addition of HMMC‐1 to the culture medium. Cell cycle analysis and expression analysis showed that HMMC‐1 treatment increased the G1 phase population of HEC9 cells and induced cyclin‐dependent kinase inhibitors p16 and p21. Two glycoproteins, 97 and 137 kDa, with a strong reactivity to HMMC‐1 were purified, and the 97‐kDa glycoprotein was identified as CD166, an immunoglobulin superfamily cell adhesion molecule assumed to be involved in cancer metastasis. CD166 gene‐silencing dramatically reduced HMMC‐1 epitope expression and growth in HEC9 cells, indicating that CD166 is the primary glycoprotein presenting the HMMC‐1 epitope in HEC9 cells. Collectively, HMMC‐1 might arrest the cell cycle in the G1 phase by binding to O‐glycans on the CD166 expressed in HEC9 cells, raising the possibility that HMMC‐1 extensively inhibits invasive growth of HMMC‐1 epitope‐positive uterine endometrial cancer cells by targeting the cancer‐associated form of CD166. (Cancer Sci 2013; 104: 62–69)

Cells express various glycans attached to glycoproteins and glycolipids and form a glycocalyx on the outside of plasma membranes. It is well known that glycans are involved in various biological phenomena, such as intercellular recognition and signaling in physiological and pathological processes, and that glycosylation undergoes changes during malignant transformation.1, 2 Characteristic glycans specifically expressed in cancer cells are possible biomarkers for monitoring tumor development and progression. Thus, monoclonal antibodies specific to cancer‐related glycans are useful not only for diagnostic reagents but also molecular targeting therapy. Furthermore, research to investigate the functions of aberrant glycans that appear during malignant transformation by use of these monoclonal antibodies provides valuable information for understanding cancer biology.

Previously, we established the human monoclonal antibody HMMC‐1 (IgM)3 by immunizing a human immunoglobulin‐producing mouse4, 5 with cells of human uterine endometrial cancer cell line SNG‐S derived from SNG‐II cells.6 Immunohistochemical study showed that HMMC‐1 reacts strongly with specimens of uterine endometrial adenocarcinoma and epithelial ovarian cancer, but not those of normal endometrium, suggesting that HMMC‐1 detects tumor‐specific molecules. Then, we demonstrated that HMMC‐1 recognises a fucosylated and extended core 1 O‐glycan, Fucα1→2Galβ1→4GlcNAcβ1→3Galβ1→3GalNAcα1→Ser/Thr, and exhibits antibody‐dependent and complement‐mediated cytotoxicity in an in vitro assay system. This suggests that HMMC‐1 might be a promising candidate useful not only for cancer diagnosis but also cancer treatment. However, the direct effects of HMMC‐1 on malignant tumor properties and the molecules on cancer cells that HMMC‐1 targets have not been investigated.

In the present study, we found that HMMC‐1 possesses a suppressive effect on growth of cultured uterine endometrial cancer cells and identified the HMMC‐1‐reactive glycoprotein as CD166, a member of the immunoglobulin superfamily cell adhesion molecules.

Materials and Methods

HEC8 and HEC9 cells

Human uterine endometrial cancer cell lines HEC8 and HEC9 were cloned and established from HEC‐1‐A cells7 by limiting dilution. Both cell lines were maintained in RPMI‐1640 medium (Sigma‐Aldrich, St Louis, MO, USA) containing 10% fetal calf serum (FCS; Cansera International Inc., Etobicoke, ON, Canada) or serum‐free (ASF) medium (104N, Ajinomoto, Tokyo, Japan). Cells were cultured in a humidified 5% CO2 atmosphere at 37°C.

Western immunoblotting

Cells in culture plates were washed twice with ice‐cold PBS and solubilized in buffer A (50 mM Tris‐HCl, pH 8.0, 150 mM NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% Triton X‐100 and 1 mM phenylmethylsulfonylfluoride [PMSF]). After centrifugation at 65 000g for 20 min, the supernatant was subjected to 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS‐PAGE) using Laemmli's method.8 The proteins were electrotransferred to a polyvinylidene difluoride membrane (Immobilon‐P, Millipore Corp., Billerica, MA, USA). The membrane was incubated with Tris‐buffered saline (TBS) containing 2% skim milk and 0.1% Tween20 for 30 min, and then incubated overnight with HMMC‐1 hybridoma supernatant diluted with TBS containing 2% skim milk at 4°C. Protein bands were detected with horseradish peroxidase‐conjugated goat anti‐human IgM (Biosource, Camarillo, CA, USA) and ECL western blotting detection reagent (GE Healthcare Life Sciences KK, Tokyo, Japan). Anti‐p16 monoclonal antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), anti‐p21 monoclonal antibody (Abcam, Cambridge, UK) and anti‐p27 KIP1 monoclonal antibody (GeneTex Inc., Irvine, CA, USA) were used. For immunoprecipitation, 30 μL of Protein A‐Sepharose (Thermo Fisher Scientific Inc., Waltham, MA, USA) was incubated with 3 μg mouse anti‐CD166 monoclonal antibody (Abcam) for 90 min at 4°C. After being washed four times with ice‐cold TBS, the lysates were added and incubated overnight at 4°C. The precipitates were then washed four times with ice‐cold TBS and eluted in Laemmli sample buffer (2.5 concentration) at 98°C for 5 min. The supernatants were centrifuged and subjected to SDS‐PAGE and immunoblotting. Proteins in conditioned medium were precipitated using trichloroacetic acid (TCA) precipitation. Cells were cultured in a serum‐free medium (ASF medium, 104N), and the conditioned medium was collected. After centrifugation at 100 000g for 20 min, the supernatant was transferred to a plastic tube, mixed with 1/10 volume of 50% TCA, incubated for 10 min at 4°C and then centrifuged at 100 000g for 10 min. The pellet was washed with 80% acetone and centrifuged at 100 000g for 10 min at 4°C. The resultant pellet was dried, suspended in TBS, and subjected to SDS‐PAGE and immunoblotting.

Cell proliferation assay

HMMC‐1 was purified from hybridoma‐conditioned media using a Protein L‐Sepharose column (Thermo Fisher Scientific Inc., Waltham, MA, USA) according to the manufacturer's instructions. Then, 2 × 104 cells were plated in 35‐mm tissue culture plates in medium containing serum and treated with 5 μg/mL HMMC‐1 or human IgM (Chemicom International Inc., Temecula, CA, USA). The medium was changed every day during experiments. After treatment, the cells were harvested by trypsinization and cell numbers were counted using a hemocytometer.

Cell cycle analysis

Cell cycle distribution was analyzed by flow cytometry. Cells were treated with 10 μg/mL HMMC‐1 or human IgM for 24 h, and then harvested by trypsinization. After being washed with PBS, cells were fixed in 70% ethanol, centrifuged at 300g for 10 min, and washed with PBS. The cell pellets obtained were resuspended in PBS containing RNase A and propidium iodide (PI) and incubated for 30 min at room temperature in the dark. After being filtered through nylon mesh (40 μm), cells were applied to flow cytometry using an EPICS XL (Beckman Coulter, Brea, CA, USA).

RT‐PCR

Total RNA was isolated from cells using a RNeasy Plus mini kit (Qiagen, Hilden, Germany). cDNA was synthesized from total RNA using the PrimeScript RT‐PCR Kit (Takara, Tokyo, Japan). Polymerase chain reaction was performed using KOD plus polymerase (TOYOBO Co. Ltd, Osaka, Japan) and the sequence‐specific primers for: α1,2Fuc‐T1 (FUT1), 5′‐GCTAGTCTGTGTCCTCTCTG‐3′ (forward) and 5′‐CTCTCTGCGGATCTGTTCCCC‐3′ (reverse); β1,3Gal‐T5, 5′‐CAAACCAAGCCCAGAACCTG‐3′ (forward) and 5′‐TCAATCTCATCTTCGGGAAAGC‐3′ (reverse);9 β1,4Gal‐T1, 5′‐AATCGTGCTAAGCTCCTCAATGTTGGC‐3′ (forward) and 5′‐CTCGGTGTCCCGATGTCCACTGTGAT‐3′ (reverse);10 core 1 β1,3GlcNAc‐T, 5′‐CTGCTGGTGATCAAGTCCTC‐3′ (forward) and 5′‐GCATCCCACATGAGCAGCA‐3′ (reverse); p16, 5′‐CACTCTCACCCGACCCGT‐3′ (forward) and 5′‐GCATGGTTACTGCCTCTGGT‐3′ (reverse);11 p21, 5′‐GTCCGTCAGAACCCATGC‐3′ (forward) and 5′‐GGCGTTTGGAGTGGTAGAAA‐3′ (reverse);11 p27 5′‐AAATGTTTCAGACGGTTCCC‐3′ (forward) and 5′‐ACAGGATGTCCATTCCATGA‐3′ (reverse);11 CD166 variant1, 5′‐ATACCTTGCCGACTTGACGTACCT‐3′ (forward) and 5′‐CTCTGTTTTCATTAGCAGAGACATTC‐3′ (reverse); CD166 variant2, 5′‐ATACCTTGCCGACTTGACGTACCT‐3′ (forward) and 3′‐TCTGCCTCATCGTGTTCTGGAATAC‐5′ (reverse); and CD166 soluble, 5′‐ATACCTTGCCGACTTGACGTACCT‐3′ (forward) and 5′‐AAAGAACATGGTCTGGTACTGGCC‐3′ (reverse).12

Purification of glycoproteins reactive with HMMC‐1 and MALDI‐TOF mass spectrometry

HEC9 cells (36 10‐cm culture dishes) were solubilized in buffer B (50 mM Tris‐HCl, pH 8.0, 150 mM NaCl, 1% Triton X‐100) and centrifuged at 65 000g for 20 min. The supernatants were applied to a column (2.5 mL) of Ulex europaeus agglutinin I (UEA‐I)‐agarose (Seikagaku Kogyo, Tokyo, Japan). The column was washed with buffer B and then bound glycoproteins were eluted with 100 mM L‐fucose (Sigma‐Aldrich). The eluants were concentrated and then separated using SDS‐PAGE. The protein bands were stained with Coomassie brilliant blue (CBB) or subjected to immunoblotting with HMMC‐1. The protein bands that were confirmed to be reactive with HMMC‐1 by immunoblotting were excised from the gel and transferred to a plastic tube. Following in‐gel trypsin digestion, extraction and desalting of peptide fragments were performed as described by Shevchenko et al.13 The peptides were applied to AXIMA‐CFRplus MALDI‐TOF mass spectrometry (Shimadzu/Kratos, Kyoto, Japan) and the mass spectrum data was analyzed using MASCOT analysis.

Immunofluorescent microscopy

Cells were plated and grown to 70% confluence on 10‐mm microscope cover slips. After being washed with ice‐cold PBS, cells were fixed with 3% paraformaldehyde in PBS for 15 min at room temperature. Cells were incubated with PBS containing 1% bovine serum albumin for 30 min, and then incubated with HMMC‐1 or anti‐CD166 antibody overnight at 4°C. After being washed, cells were incubated with Alexa 488‐conjugated anti‐mouse IgG or Alexa 594‐conjugated anti‐human IgM (Molecular Probes, Eugene, OR, USA). The cells were observed using a confocal laser microscope (TCS SP5, Leica Microsystems, Heidelberg, Mannheim, Germany).

Gene silencing using siRNA

Cells (5 × 105) were plated at approximately 70% confluence in 35‐mm cell culture dishes and cultured overnight. They were transiently transfected with CD166 siRNA (25 nM, sense: 5′‐GAAGAGAAUGUUACAUUAATT‐3′, antisense: 5′‐UUAAUGUAACAUUCUCUUCAG‐3′) using Hiperfect transfection reagent (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions. Cells were harvested 72 h after transfection and subjected to immunoblotting with HMMC‐1 or anti‐CD166 antibody.

Paraaortic lymph node metastasis test using orthotopic implantation

A mouse model of in vivo lymph node metastasis was produced as described previously14 with the following modification. Balb/c nu/nu mice were anesthetized with avertin (2,2,2–tribromoethanol; Aldrich, Milwaukee, WI, USA). The lower abdomen was swabbed with 70% alcohol and an incision approximately 20 mm long was made in the skin of the left lower abdominal wall. Viable tumor cells (3 × 106) suspended in 50 μL of PBS were injected into the ipsilateral uterus horn using a 25‐gauge needle attached to a 1.0 mL syringe. The skin incision was closed with a soluble suture. After the animals had recovered from bradycardia and had stable spontaneous respiration, they were returned to their cages. Eight weeks after the inoculation of cells, the mice were killed and examined to determine whether they had developed metastasis in the lymph nodes.

Data presentation and statistical analysis

Statistical significance of the data was determined using the Student's t‐test with the Statcel statistical program (OMC, Saitama, Japan).

Results

Characteristics of HEC8 and HEC9 cells

HEC8 and HEC9 cell lines were established from the human uterine endometrial cancer cell line HEC‐1‐A, which had previously been derived from HEC‐1 cells by Kuramoto et al.7 In the present study, we first established several sub‐cell lines from HEC‐1‐A cells by single cell cloning and chose two cell lines, HEC8 and HEC9, which are remarkably different in HMMC‐1 reactivity. Immunoblot analysis showed that HEC9 cells strongly react with HMMC‐1, and proteins ranging in size from 90 to 200 kDa were detected. Major bands were estimated as 97, 137 and 200 kDa. On one hand, HEC8 was hardly reactive with HMMC‐1 (Fig. 1a). These cell lines are clearly distinct in morphological features, that is, the cell configuration of HEC8 is completely uniform, and large and spread in form. On the other hand, HEC9 is composed of cells having two different configurations, relatively large and small cells (Fig. 1b). In a previous study we showed that HMMC‐1 recognizes fucosylated and extended core 1 O‐glycan, Fucα1→2Galβ1→4GlcNAcβ1→3Galβ1→3GalNAcα1→Ser/Thr as an epitope structure, and α1,2 fucosyltransferase‐1 (α1,2Fuc‐T1) and core 1 β1,3‐N‐acetylglucosaminyltransferase (core 1 β1,3GlcNAc‐T) are responsible for expression of HMMC‐1 epitope glycans.3 Therefore, we analyzed expression profiles of several glycosyltranseferases, including α1,2Fuc‐T1 and core 1 β1,3GlcNAc‐T, that are involved in the biosynthesis of Fucα1→2Galβ1→4GlcNAcβ1→3Galβ1→3GalNAcα1→Ser/Thr by RT‐PCR. As shown in Figure 1(c), in comparison to HEC8, the expressions of core 1 β1,3GlcNAcT and β1,4galactosyltransferase‐1 (β1,4GalT1) by HEC9 were upregulated and those of α1,2FucT1 and β1,3galactosyltransferase‐5 (β1,3GalT5) were not changed. These results suggest that the difference between HEC8 and HEC9 in HMMC‐1 reactivity is mainly due to the difference in the expression levels of core 1 β1,3GlcNAcT and β1,4GalT1. We therefore considered that HEC8 and HEC9 are useful for comparative studies to investigate the effects of HMMC‐1 and the functional roles of HMMC‐1 epitope glycans expressed in endometrial cancer cells. The metastatic potentials of HEC8 and HEC9 cells were examined using a paraaortic lymph node metastasis test after orthotopic implantation, according to the method of Tamada et al.14 This model was conceived to correspond to the advanced clinical stage of gynecological carcinomas; thus, it would be useful in understanding the molecular biology of those carcinomas and in the development of therapeutic modalities against lymph node metastasis. As shown in Figure 1(d), implantation of HEC8 and HEC9 caused metastasis in 10% (1/10) and 80% (16/20), respectively, of lymph nodes.

Figure 1.

Figure 1

Open in a new tab

HEC8 and HEC9 cells. (a) Immunoblotting analysis of HEC8 and HEC9 cells with HMMC‐1. (b) Morphology of HEC8 and HEC9 cells (original magnification, × 20). (c) Expression profiles of glycosyltransferases in HEC8 and HEC9 cells. Glycosyltransferases related to biosynthesis of HMMC‐1 epitope structure, both fucosylated and extended core 1 O‐glycans, were analyzed using RT‐PCR. cDNA prepared from HEC8 and HEC9 cells were probed using primer sets specific for α1,2Fuc‐T1, β1,3Gal‐T5, β1,4Gal‐T1 and core 1 β1,3 GlcNAc‐T. The expression level of β‐actin was compared as an internal control. (d) Paraaortic lymph node metastasis due to orthotopic implantation of HEC8 and HEC9 cells. Viable tumor cells (3 × 106) suspended in 50 μL of phosphate‐buffered saline were injected into the ipsilateral uterus horn and formation of metastasis in lymph nodes was examined. Ten and 20 mice survived 8 weeks after the injection of HEC8 cells (white bar) and HEC9 cells (black bar), respectively.

Effects of HMMC‐1 on proliferation of uterine endometrial cancer cells

We examined the effects of HMMC‐1 on the cell growth of HEC8 and HEC9. When HEC9 cells were cultured in medium containing HMMC‐1 at a concentration of 5 μg/mL, their growth was reduced to approximately 50% during 7 days incubation compared with human IgM. HMMC‐1 did not induce apoptosis in HEC9 cells. No effect on cell growth was observed in HEC8 with HMMC‐1 treatment (Fig. 2). Cell cycle analysis using flow cytometry showed that HMMC‐1 treatment at a concentration of 10 μg/mL for 24 h increased the number of HEC9 cells in the G1 phase (Fig. 3a,b) compared with human IgM treatment. It is known that cell cycle progression is regulated by cyclin‐dependent kinases (CDK), which are controlled by CDK inhibitors. Therefore, we analyzed expression levels of the CDK inhibitors p16, p21 and p27, which are well known negative regulators of the G1/S transition.15, 16 Treatment of HEC9 cells with HMMC‐1 increased expression levels of p16 and p21, but not p27, suggesting that p16 and p21 are induced by HMMC‐1 and abrogate cell cycle progression (Fig. 3c,d).

Figure 2.

Figure 2

Open in a new tab

Effects of HMMC‐1 on proliferation of HEC8 and HEC9 cells. HMMC‐1 or human IgM (5 μg/mL) was added into the cell culture media of HEC8 (a) and HEC9 (b) cells and cell numbers were counted. Black and dotted bars show the results of the addition of HMMC‐1 and human IgM, respectively. White bars show cell growth in the medium alone. Data represent the mean ± SD (n = 3). *P < 0.05.

Figure 3.

Figure 3

Open in a new tab

Cell cycle analysis of HMMC‐1 treated cells. The number of cells in each phase of the cell cycle was analyzed using propidium iodide (PI) staining and subsequent flow cytometry. (a) HEC9 cells were treated for 24 h with HMMC‐1 or 10 μg/mL human IgM. (b) The cell cycle distribution was analyzed using Multicycle software. Values represent the mean ± SD of three independent experiments. P < 0.05. (c) Expression levels of p16, p21, p27 and β‐actin in HEC9 cells were examined using RT‐PCR after treatment with HMMC‐1 or 10 μg/mL human IgM for 24 h. (d) Immunodetection of p16, p21 and p27 in HEC9 cell lysate after treatment with HMMC‐1 or 10 μg/mL human IgM for 24 h.

Identification of glycoprotein reactive with HMMC‐1

For purification of the glycoprotein reactive with HMMC‐1, we first tried HMMC‐1 conjugated column chromatography, but sufficient protein for structural analysis was not obtained. Next, we used a column conjugated with UEA‐I, a lectin binding to fucosylated glycans, because glycoproteins reactive with HMMC‐1 were successfully detected with UEA‐I using lectin blotting, and two protein bands corresponding to 97 kDa and 137 kDa on SDS‐PAGE were purified using UEA‐I affinity chromatography. As shown in Figure 4, both protein bands strongly reacted with HMMC‐1; therefore, we performed proteolytic fragmentation of those protein bands using in‐gel digestion, and subsequently applied the peptide fragments to MALDI‐TOF mass spectrometry and MASCOT analysis. Table 1 shows information on the peptide fragments detected using MALDI‐TOF mass spectrometry: the 97 kDa and 137 kDa protein bands were deduced to be CD166 (activated leukocyte cell adhesion molecule [ALCAM]) and CD49c (VLA‐3 α subunit, integrin α3). To confirm the results of mass spectrometry, we performed immunoblotting of HEC9 cell lysates with anti‐CD166 antibody and anti‐CD49c antibody. The result showed that CD166, detected as the 97 kDa protein band, is abundantly expressed in HEC9 cells, but CD49c was not detected (data not shown). We carried out further tests to validate that the HMMC‐1 reactive glycoprotein is CD166. Anti‐CD166 antibody successfully immunoprecipitated the approximate 97 kDa protein from HEC8 and HEC9 cell lysates, and the only 97 kDa protein immunoprecipitated from HEC9 cell lysates was immunostained with HMMC‐1 (Fig. 5a). Immunofluorescence of HEC8 and HEC9 cells was examined using HMMC‐1 and anti‐CD166 antibody. HMMC‐1 strongly reacted with HEC9 but not HEC8, and this is consistent with the results of the immunoblotting (Fig. 1a). Both cells were intensely immunostained with anti‐CD166, and fluorescent signals of anti‐CD166 and HMMC‐1 were consistently colocalized in HEC9 cells (Fig. 5b). These results clearly indicate that HMMC‐1 reactive glycoprotein expressed in HEC9 cells is CD166. CD166 is a cell adhesion molecule and a transmembrane glycoprotein ubiquitously expressed in a variety of tissues. It is known that three alternative splicing isoforms, variant1, variant2 and soluble forms, are produced by the CD166 gene; furthermore, an extracellular domain of CD166 is shed by cleavage of matrix metalloproteinases, possibly by an ADAM17/TACE‐dependent pathway.17 We detected variant2 and the soluble variant of CD166 in HEC8 and HEC9 cells using RT‐PCR (Fig. 5c), but protein bands corresponding to the soluble variant were not detected using immunoblotting with anti‐CD166 and HMMC‐1 (data not shown). It is noteworthy that CD166 is partially released from HEC9 cells and that the shed CD166 is significantly modified by HMMC‐1 epitope glycans (Fig. 5d).

Figure 4.

Figure 4

Open in a new tab

SDS‐gel electrophoresis of HMMC‐1‐reactive proteins purified using UEA‐I column chromatography. HEC9 cell lysates were applied on a UEA‐I agarose column and adsorbed proteins were eluted with 100 mM Fuc. The eluates were concentrated and subjected to Coomassie brilliant blue staining and immunoblotting with HMMC‐1. Protein bands A and B (arrow) were subjected to in‐gel protease digestion and mass spectrometry.

Table 1.

HMMC‐1‐reactive proteins identified using MALDI‐TOF MS analysis

Band (kDa) Mass values (Da) Peptide sequence Protein Sequence positiona
97 1940.03 WKYEKPDGSPVFIAFR CD166 [56–71]
97 2343.03 WKYEKPDGSPVFIAFRSSTK CD166 [56–75]
97 1625.87 YEKPDGSPVFIAFR CD166 [58–71]
97 2029.00 YEKPDGSPVFIAFRSSTK CD166 [58–75]
97 1741.88 KSVQYDDVPEYKDR CD166 [76–89]
97 1613.78 SVQYDDVPEYKDR CD166 [77–89]
97 1486.87 VFKQPSKPEIVSK CD166 [130–142]
97 1191.66 ALFLETEQLK CD166 [143–152]
97 1319.76 ALFLETEQLKK CD166 [143–153]
97 1256.68 SSNTYTLTDVR CD166 [294–304]
97 1412.74 SSNTYTLTDVRR CD166 [294–305]
97 1554.89 ESLTLIVEGKPQIK CD166 [406–419]
97 934.47 KTDPSGLSK CD166 [423–431]
97 1655.84 HVNKDLGNMEENKK CD166 [560–573]
137 894.44 GPGPSRAPR CD49c [2–10]
137 1774.91 EAGNPGSLFGYSVALHR CD49c [44–60]
137 973.58 YLLLAGAPR CD49c [68–76]
137 1430.73 TGAVYLCPLTAHK CD49c [88–100]
137 2105.96 TGAVYLCPLTAHKDDCER CD49c [88–105]
137 854.46 VLVCAHR CD49c [137–143]
137 1724.86 YTQVLWSGSEDQRR CD49c [144–157]
137 968.48 GNSYMIQR CD49c [222–229]
137 1096.53 GNSYMIQRK CD49c [222–230]
137 2094.06 HRHMGAVFLLSQEAGGDLR CD49c [275–293]
137 1800.93 HMGAVFLLSQEAGGDLR CD49c [277–293]
137 1957.02 HMGAVFLLSQEAGGDLRR CD49c [277–294]
137 1083.51 VYIYHSSSK CD49c [399–407]
137 1602.95 GLLRQPQQVIHGEK CD49c [408–421]
137 1163.61 QPQQVIHGEK CD49c [412–421]
137 1146.71 ARPVINIVHK CD49c [462–471]
137 796.48 DRRPPR CD49c [523–528]
137 2133.92 FAGSESAVFHGFFSMPEMR CD49c [531–549]
137 2149.97 FAGSESAVFHGFFSMPEMR CD49c [531–549]
137 2165.91 FAGSESAVFHGFFSMPEMR CD49c [531–549]
137 771.31 MPDRPR CD49c [580–585]
137 1837.79 ECGPDNKCESNLQMR CD49c [614–628]
137 1135.60 AAFVSEQQQK CD49c [629–638]
137 1675.83 LQSFFGGTVMGESGMK CD49c [771–786]
137 1691.82 LQSFFGGTVMGESGMK CD49c [771–786]
137 1568.78 YYQIMPKYHAVR CD49c [1024–1035]
Open in a new tab
a

Sequence positions correspond to CD166 (accession number Q13740) and CD49c (accession number P26006).

Figure 5.

Figure 5

Open in a new tab

Detection of CD166 in HEC9 cells. (a) HEC9 cell lysates were immunoprecipitated with (lanes 1 and 2) or without (lane 3) anti‐CD166 antibody and the precipitates obtained were subjected to immunoblotting with HMMC‐1 (lane 1) and anti‐CD166 antibody (lanes 2 and 3) as described in the Materials and Methods. The protein band indicated by the arrow has immunoreactivity to both HMMC‐1 and anti‐CD166. The arrowheads indicate heavy and light chains of anti‐CD166 antibody added for immunoprecipitation. (b) Immunofluorescence analysis of HEC9 and HEC8 cells with anti‐CD166 (green) and HMMC‐1 (red). Scale bar, 20 μm. (c) Detection of CD166 gene expression using RT‐PCR analysis. S, soluble variant; 1, variant1; 2, variant2. (d) Cell lysates (Ly) and the conditioned medium (CM) were probed with anti‐CD166 antibody and HMMC‐1.

Silencing of CD166 expression by siRNA undergoes disappearance of HMMC‐1 reactivity in HEC9 cells

Transfection of CD166 siRNA significantly reduced CD166 expression in HEC9 cells, that is, the 97 kDa band of CD166 was hardly detected using immunoblotting (Fig. 6a). The cell morphology of HEC9 cells was not apparently changed by siRNA transfection (Fig. 6b). It was shown that besides the 97 kDa band, HMMC‐1 reactivity of the 137 and 200 kDa bands was remarkably reduced by CD166 siRNA. It is unexpected, but a possible explanation is that glycosylation in cells is complicated and cooperative processes by several glycosyltransferases to specific substrate proteins might be influenced by expression levels of not only glycosyltransferases but also substrate proteins. Because CD166 is the primary glycoprotein that undergoes HMMC‐1 epitope glycosylation, it was suspected that alteration of the CD166 expression level affects overall HMMC‐1 epitope glycosylation in HEC9 cells.

Figure 6.

Figure 6

Open in a new tab

Gene silencing of CD166 by siRNA. HEC9 cells were transfected by siRNA specific to CD166 and expression of CD166 was examined using immunoblotting (a). (b) Cell morphology (original magnification, ×20).

Effects of anti‐CD166 antibody and gene silencing of CD166 on proliferation of HEC9 cells

When HEC9 cells were cultured in medium containing anti‐CD166 antibody at a concentration of 5 μg/mL, their growth was remarkably reduced (Fig. 7a). Furthermore, it was revealed that CD166 gene silencing is effective in inhibiting cell growth of HEC9 (Fig. 7b).

Figure 7.

Figure 7

Open in a new tab

Effects of anti‐CD166 antibody and gene silencing of CD166 on proliferation of HEC9 cells. (a) Antibodies (5 μg/mL) were added into the cell culture media of HEC9 cells and cell numbers were counted. White, dotted and black bars show the results of the addition of human IgM, HMMC‐1 and anti‐CD166 antibody, respectively. Data represent the mean ± SD (n = 3). *P < 0.05. (b) Black bars show the results of cell growth of CD166 gene‐silencing HEC9 cells. Dotted and white bars show the results of the addition of HMMC‐1 (5 μg/mL) and medium alone, respectively. Data represent the mean ± SD (n = 3). *P < 0.05.

Discussion

In the present study, we found that HMMC‐1 suppressed cell growth of the human uterine endometrial cancer cell HEC9. A similar effect was observed in SNG‐S cells, a cell line used as an immunogen for HMMC‐1 production (data not shown), indicating that HMMC‐1 might be extensively effective for antibody therapy against HMMC‐1 epitope‐positive cancer cells. Regarding tumor cell suppression by anti‐glycan antibodies, Aixinjueluo and co‐workers18 have reported that anti‐GD2 (ganglioside, NeuAcα2‐8NeuAcα2‐3[GalNAcβ1‐4]Galβ1‐4GlcβCer) antibody suppresses the growth of cells of GD2‐positive small‐cell lung carcinoma and induces apoptosis with caspase activation. Therefore, we first considered whether HMMC‐1 induces apoptosis in a similar manner, but apoptotic cells were not detected during the 7‐day treatment with HMMC‐1 using the annexin V assay system (data not shown). Instead, it was found that HMMC‐1 induces cell cycle arrest in the G1 phase by upregulation of CDKI p16 and p21. It was reported that trastuzumab (Herceptin), a monoclonal antibody for human epidermal growth factor receptor 2 overexpressing in breast, ovarian, gastric and colon carcinomas, induces p27 upregulation and results in G1 arrest.19 In the case of growth suppression by HMMC‐1, the expression level of p27 did not change, suggesting that the signaling pathways induced by the antibody binding were different in HMMC‐1 and trastuzumab. G1 to S phase progression is regulated by INK4 family proteins, inhibitors of CDK 4 and 6, and p16 is one of the INK4 proteins.15 It has been shown that p16 accumulates during cell senescence and induces G1 arrest by associating with CDK 4 and 6 and promoting release and degradation of D‐type cyclins. Further studies including examination of whether HMMC‐1 induces senescence or terminal differentiation such as withdrawal to G0 phase are necessary to understand the detailed mechanisms of cell growth suppression by HMMC‐1.

We showed that HEC9 cells abundantly express HMMC‐1 reactive glycoproteins having molecular masses of 97, 137 and 200 kDa. We successively purified the 97‐ and 137‐kDa proteins and identified the 97‐kDa protein as CD166. CD166 is also known as ALCAM and it is one of the cell surface immunoglobulin superfamily members involved in cell–cell interactions through homophilic (CD166‐CD166) and heterophilic (CD166‐CD6) binding.20, 21, 22, 23 Twelve potential N‐glycosylation sites were found in the amino acid sequence of human CD166, and it has been reported that CD166 is actually N‐ and O‐glycosylated.17 CD166 expression is widely detected in proliferating or trafficking cells such as activated leukocytes, embryonal hematopoietic and endothelial cells, lung endothelial cells, endometrial cells and blastocysts,24, 25 and is involved in cell adhesion, growth and motility in normal tissues.21 Furthermore, CD166 has been evaluated in several malignancies including melanoma, prostate carcinoma, breast cancer, colorectal carcinoma and esophageal squamous cell carcinoma.26 In melanoma cells, CD166 is increased in the vertical phase cell growth,27 and in prostate cancer, CD166 is upregulated and downregulated in low‐grade and high‐grade tumors, respectively.28 It has been suggested that expression levels and localization sites of CD166 are associated with prognosis, patient survival29 and tumor stage/grade.30, 31 Many observations concerning CD166 expression in physiological and pathological conditions have been reported, but there has been little attention to the glycan structures of CD166. Sato et al. previously reported that CD166 expressed in Neuro2A cells possesses a unique disialic acid epitope on its O‐glycans and that the glycosylation is involved in neurite formation.32 Meanwhile, we showed that CD166 expressed in HEC9 cells undergoes modification of HMMC‐1 epitope glycans. It appears that CD166 is widely expressed in different cell types and tends to undergo glycosylation changes. It is likely that its glycans determine the functional roles in individual cell types.

MALDI‐TOF mass analysis showed the possibility that the 137‐kDa protein is CD49c. CD49c is also known as the VLA‐3 (α3β1 integrin) α subunit or integrin α3 subunit. VLA‐3 is a member of the integrin family of adhesion molecules and is expressed in many types of cancer cells and able to regulate cell functions associated with malignancy.33, 34, 35 It is possible that a set of adhesion molecules is specifically modified by HMMC‐1 epitope glycans. Because we failed to detect CD49c in HEC9 cell lysate using immunoblotting, further identification and confirmation of the 137‐kDa protein remains to be done.

Recently, it was revealed that specific glycan structures or glycosyltransferase genes control the metastasis of cancer. For example, I‐branching β1,6‐N‐acetylglucosaminyl transferase 2 (GCNT2), a member of the β1,6‐N‐acetylglucosaminyltransferase family that plays a critical role in glycosylation, is a novel metastasis‐related gene candidate.36 Expression of N‐acetylglucosaminyltransferase V in colorectal cancer correlates with metastasis and poor prognosis.37 It was shown that N‐acetylglucosaminyltransferase III contributes to TGF‐β‐induced epithelial‐to‐mesenchymal transition.38 In the present study, we was speculated that upregulation of core 1 β1,3GlcNAc‐T and β1,4Gal‐T1 and the resultant overexpression of HMMC‐1 epitope glycans in uterine endometrial cancer cells correlate with cancer metastasis. In the future, a possible experiment to confirm the hypothesis would be determination of the metastatic rate due to implantation of HEC9 cells with core 1 β1,3GlcNAc‐T and β1,4Gal‐T1 knockdowns.

Disclosure Statement

The authors have no conflict of interest.

Acknowledgments

This work was supported in part by a Grant‐in‐Aid for Scientific Research (B) 19390430 to D.A. from the Japan Society for the Promotion of Science and a research grant by Kyowa Hakko Kirin Co., Ltd.

(Cancer Sci, doi: 10.1111/cas.12038, 2012)

References

  • 1. Hakomori S. Aberrant glycosylation in cancer cell membranes as focused on glycolipids: overview and perspectives. Cancer Res 1985; 45: 2405–14. [PubMed] [Google Scholar]
  • 2. Kobata A, Amano J. Altered glycosylation of proteins produced by malignant cells, and application for the diagnosis and immunotherapy of tumors. Immunol Cell Biol 2005; 83: 429–39. [DOI] [PubMed] [Google Scholar]
  • 3. Nozawa S, Aoki D, Tsukazaki K et al HMMC‐1: a humanized monoclonal antibody with therapeutic potential against Müllerian duct‐related carcinomas. Clin Cancer Res 2004; 10: 7071–8. [DOI] [PubMed] [Google Scholar]
  • 4. Tomizuka K, Yoshida H, Uejima H, et al Functional expression and germline transmission of a human chromosome fragment in chimaeric mice. Nat Genet 1997; 16: 133–43. [DOI] [PubMed] [Google Scholar]
  • 5. Ishida I, Tomizuka K, Yoshida H et al Production of human monoclonal and polyclonal antibodies in TransChromo animals. Cloning Stem Cells 2002; 4: 91–102. [DOI] [PubMed] [Google Scholar]
  • 6. Nozawa S, Sakayori M, Ohta K et al A monoclonal antibody (MSN‐1) against a newly established uterine endometrial cancer cell line (SNG‐II) and its applications to immunohistochemistry and flow cytometry. Am J Obstet Gynecol 1989; 161: 1079–86. [DOI] [PubMed] [Google Scholar]
  • 7. Kuramoto H, Hamano M, Imai M. HEC‐1 cells. Hum Cell 2002; 15: 81–95. [DOI] [PubMed] [Google Scholar]
  • 8. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227: 680–5. [DOI] [PubMed] [Google Scholar]
  • 9. Hayashi N, Nakamori S, Okami J et al Association between expression levels of CA 19‐9 and N‐acetylglucosamine‐beta; 1,3‐galactosyltransferase 5 gene in human pancreatic cancer tissue. Pathobiology 2004; 71: 26–34. [DOI] [PubMed] [Google Scholar]
  • 10. Xu S, Zhu X, Zhang S et al Over‐expression of beta‐1,4‐galactosyltransferase I, II, and V in human astrocytoma. J Cancer Res Clin Oncol 2001; 127: 502–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Ohta K, Maruyama T, Uchida H et al Glycodelin blocks progression to S phase and inhibits cell growth: a possible progesterone‐induced regulator for endometrial epithelial cell growth. Mol Hum Reprod 2008; 14: 17–22. [DOI] [PubMed] [Google Scholar]
  • 12. Ikeda K, Quertermouns T. Molecular isolation and characterization of a soluble isoform of activated leukocyte cell adhesion molecule that modulates endothelial cell function. J Biol Chem 2004; 279: 55315–23. [DOI] [PubMed] [Google Scholar]
  • 13. Shevchenko A, Tomas H, Havlis J, Olsen JV, Mann M. In‐gel digestion for mass spectrometric characterization of proteins and proteomes. Nat Protoc 2006; 1: 2856–60. [DOI] [PubMed] [Google Scholar]
  • 14. Tamada Y, Aoki D, Nozawa S, Irimura T. Model for paraaortic lymph node metastasis produced by orthotopic implantation of ovarian carcinoma cells in athymic nude mice. Eur J Cancer 2004; 40: 158–63. [DOI] [PubMed] [Google Scholar]
  • 15. Shapiro GI. Cyclin‐dependent kinase pathways as targets for cancer treatment. J Clin Oncol 2006; 24: 1770–83. [DOI] [PubMed] [Google Scholar]
  • 16. Halevy O, Novitch BD, Spicer DB et al Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD. Science 1995; 267: 1018–21. [DOI] [PubMed] [Google Scholar]
  • 17. Micciche F, Da Riva L, Fabbi M et al Activated leukocyte cell adhesion molecular expression and shedding in thyroid tumors. PLoS ONE; 2011; 6: 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Aixinjueluo W, Furukawa K, Zhang Q et al Mechanisms for the apoptosis of small cell lung cancer cells induced by anti‐GD2 monoclonal antibodies. J Biol Chem 2005; 280: 29828–36. [DOI] [PubMed] [Google Scholar]
  • 19. Le XF, Claret FX, Lammayot A et al The role of cyclin‐dependent kinase inhibitor p27Kip1 in anti‐HER2 antibody‐induced G1 cell cycle arrest and tumor growth inhibition. J Biol Chem 2003; 278: 23441–50. [DOI] [PubMed] [Google Scholar]
  • 20. Patel DD, Wee SF, Whichard LP et al Identification and characterization of a 100‐kD ligand for CD6 on human thymic epithelial cells. J Exp Mel 1995; 181: 1563–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Bowen MA, Bajorath J, Siadak AW et al The amino‐terminal immunoglobulin‐like domain of activated leukocyte cell adhesion molecule binds specifically to the membrane‐proximal scavenger receptor cysteine‐rich domain of CD6 with a 1:1 stoichiometry. J Biol Chem 1996; 271: 17390–6. [DOI] [PubMed] [Google Scholar]
  • 22. Degen WG, van Kempen LC, Gijzen EG et al MEMD, a new cell adhesion molecule in metastasizing human melanoma cell lines, is identical to ALCAM (activated leukocyte cell adhesion molecule). Am J Pathol 1998; 152: 805–13. [PMC free article] [PubMed] [Google Scholar]
  • 23. Bowen MA, Aruffo AA, Bajorath J. Cell surface receptors and their ligands: in vitro analysis of CD6‐CD166 interactions. Proteins 2000; 40: 420–8. [PubMed] [Google Scholar]
  • 24. Cayrol R, Wosik K, Berard JL et al Activated leukocyte cell adhesion molecule promotes leukocyte trafficking into the central nervous system. Nat Immunol 2008; 9: 137–45. [DOI] [PubMed] [Google Scholar]
  • 25. Masedunskas A, King JA, Tan F et al Activated leukocyte cell adhesion molecule is a component of the endothelial junction involved in transendothelial monocyte migration. FEBS Lett 2006; 580: 2637–45. [DOI] [PubMed] [Google Scholar]
  • 26. Ofori‐Acquah SF, King JA. Activated leukocyte cell adhesion molecule: a new paradox in cancer. Transl Res 2008; 151: 122–8. [DOI] [PubMed] [Google Scholar]
  • 27. van Kempen LC, van denOord JJ, van Muijen GN, Weidle UH, Bloemers HP, Swart GW. Activated leukocyte cell adhesion molecule/CD166, a marker of tumor progression in primary malignant melanoma of the skin. Am J Pathol 2000; 156: 769–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Kristiansen G, Pilarsky C, Wissmann C et al ALCAM/CD166 is up‐regulated in low‐grade prostate cancer and progressively lost in high‐grade lesions. Prostate 2003; 54: 34–43. [DOI] [PubMed] [Google Scholar]
  • 29. King JA, Ofori‐Acquah SF, Stevens T, Al‐Mehdi AB, Fodstad O, Jiang WG. Activated leukocyte cell adhesion molecule in breast cancer: prognostic indicator. Breast Cancer Res 2004; 6: 478–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Tomita K, van Bokhoven A, Jansen CFJ et al Activated leukocyte cell adhesion molecule (ALCAM) expression is associated with a poor prognosis for bladder cancer patients. Uro Oncology 2003; 3: 121–9. [Google Scholar]
  • 31. Verma A, Shukla NK, Deo SV, Gupta SD, Ralhan R. MEMD/ALCAM: a potential marker for tumor invasion and nodal metastasis in esophageal squamous cell carcinoma. Oncology 2005; 68: 462–70. [DOI] [PubMed] [Google Scholar]
  • 32. Sato C, Matsuda T, Kitajima K. Neuronal differentiation‐dependent expression of the disialic acid epitope on CD166 and its involvement in neurite formation in neuro2A cells. J Biol Chem 2002; 277: 45299–305. [DOI] [PubMed] [Google Scholar]
  • 33. Elices MJ, Urry LA, Hemler ME. Receptor functions for the integrin VLA‐3: fibronectin, collagen, and laminin binding are differentially influenced by ARG‐GLY‐ASP peptide and by divalent cations. J Biol Chem 1991; 112: 169–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Natali PG, Nicotra MR, Bartolazzi A, Cavaliere R, Bigotti A. Integrin expression in cutaneous malignant melanoma: association of the alpha 3/beta 1 heterodimer with tumor progression. Int J Cancer 1993; 54: 68–72. [DOI] [PubMed] [Google Scholar]
  • 35. Mitchell K, Svenson KB, Longmate WM et al Suppression of integrin alpha3beta1 in breast cancer cells reduces cyclooxygenase‐2 gene expression and inhibits tumorigenesis, invasion, and cross‐talk to endothelial cells. Cancer Res 2010; 70: 6359–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Zhang H, Meng F, Wu S, et al Engagement of I‐branching β‐1,6‐N‐acetylglucosaminyltransferase 2 in breast cancer metastasis and TGF‐β signaling. Cancer Res 2011; 71: 4846–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Murata K, Miyoshi E, Kameyama M et al Expression of N‐acetylglucosaminyltransferase V in colorectal cancer correlates with metastasis and poor prognosis. Clin Cancer Res 2000; 6: 1772–7. [PubMed] [Google Scholar]
  • 38. Xu Q, Isaji T, Lu Y et al Roles of N‐acetylglucosaminyltransferase III in epithelial‐to‐mesenchymal transition induced by transforming growth factor β1 (TGF‐β1) in epithelial cell lines. J Biol Chem 2012; 287: 16563–74. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Cancer Science are provided here courtesy of Wiley

RESOURCES