Type VI collagen α1 chain polypeptide in non-triple helical form is an alternative gene product of COL6A1

Type VI collagen α1 chain polypeptide in non-triple helical form is an alternative gene product... Abstract Expression of type IV collagen α1 chain in non-triple helical form, NTH α1(IV), is observed in cultured human cells, human placenta and rabbit tissues. Biological functions of NTH α1(IV) are most likely to be distinct from type IV collagen, since their biochemical characteristics are quite different. To explore the biological functions of NTH α1(IV), we prepared some anti-NTH α1(IV) antibodies. In the course of characterization of these antibodies, one antibody, #141, bound to a polypeptide of 140 kDa in size in addition to NTH α1(IV). In this study, we show evidence that the 140 kDa polypeptide is a novel non-triple helical polypeptide of type VI collagen α1 chain encoded by COL6A1, or NTH α1(VI). Expression of NTH α1(VI) is observed in supernatants of several human cancer cell lines, suggesting that the NTH α1(VI) might be involved in tumourigenesis. Reactivity with lectins indicates that sugar chains of NTH α1(VI) are different from those of the α1(VI) chain in triple helical form of type VI collagen, suggesting a synthetic mechanism and a mode of action of NTH α1(VI) is different from type VI collagen. cancer, non-triple helical α1(IV) chain, non-triple helical α1(VI) chain, type IV collagen, type VI collagen Presence of type IV collagen α1 [α1(IV)] chain in non-triple helical form in cultured human cells (1), human placenta (2) and rabbit tissues (3) is reported and that is designated as NTH α1(IV) (3, 4). Cultured human cells secrete NTH α1(IV) under physiological conditions and the production of NTH α1(IV) depends on ascorbate. Under ascorbate-depleted condition, cells secrete NTH α1(IV) whose prolyl and lysyl residues have lower hydroxylation levels than type IV collagen or α1(IV) chain in triple helical form (4, 5). In addition, NTH α1(IV), but not the α1(IV) chain from the triple helical molecules, has specific sugar chains that are detectable by lectin Agaricus bisporus agglutinin (ABA) (2, 6). Namely, NTH α1(IV) has different chemical structure compared with the α1(IV) chain in triple helical form, suggesting that the biological function of NTH α1(IV) as well as regulation of synthesis and secretion is also different from that of the α1(IV) chain in triple helical form or type IV collagen. The NC1 domain derived from NTH α1(IV) shows tissue inhibitor of metalloproteinase-like activity against matrix metalloproteinase-9 (2). Sugiyama et al. reported that NTH α1(IV) expression is observed at a neovascular tip, where no type IV collagen is detected, in a rabbit angiogenic model, suggesting that physiological role of NTH α1(IV) is associated with angiogenesis or vascular system dynamics (3). Tumour cells produce different extracellular matrix or microenvironment presumably in favour of tumour formation and expansion; for example, peptides derived from type IV collagen promote the expansion and angiogenesis of tumour tissues (7). NTH α1(IV) expression is observed in various tumour cell cultures (M. Morita, manuscript in preparation). Although NTH α1(IV) could be involved in tumour angiogenesis, the biological function and the mode of action of NTH α1(IV) remain to be clarified. To explore the possible involvement of NTH α1(IV) in tumourigenesis, we have acquired some anti-NTH α1(IV) monoclonal antibodies that did not react with triple helical type IV collagen molecules (4). In the course of characterization of these antibodies, one of the antibodies, antibody #141, but not other monoclonal antibodies, reacted with a polypeptide of 140 kDa in size in addition to NTH α1(IV). We hypothesized that the 140 kDa polypeptide was a novel polypeptide that contains the epitope related to the NTH α1(IV). In this study, we characterized covalent structure of the 140 kDa polypeptide and identified that the 140 kDa polypeptide is a novel non-triple helical polypeptide of type VI collagen α1 [α1(VI)] chain encoded by COL6A1 named NTH α1(VI). We also showed that NTH α1(VI) expression was found in the supernatants of human liver cancer cell line HLF, human oesophageal cancer cell line TE8 and human lung cancer cell line NCI-H226. In addition, sugar chains of NTH α1(VI) were different from those of the α1(VI) chain in triple helical form from differential binding activities of lectins. Our results suggest that NTH α1(VI) is possibly involved in tumourigenesis and the mode of action and the biological function are distinct from those of the α1(VI) chain in triple helical form or type VI collagen. Here we report in detail. Materials and Methods Antibodies and lectins Anti-NTH α1(IV) monoclonal antibody #141 was prepared in our laboratory using purified NTH α1(IV) as an antigen (Nippon Kayaku) and partially characterized in our previous study (4). Anti-α1(IV) chain monoclonal antibody, JK132, was obtained using purified human collagen type IV as an antigen, as described previously (8). Anti-COL6A1 polyclonal antibody (N term) was purchased from Abgent (San Diego, CA). Horseradish peroxidase-labelled anti-mouse and anti-rabbit IgG antibodies were purchased from GE Healthcare (Little Chalfont, UK). Biotin-labelled lectin Lens culinaris agglutinin (LCA), Ricinus communis agglutinin I (RCA120) and wheat germ agglutinin (WGA) were purchased from J-OIL MILLS (Tokyo, Japan). LCA-conjugated agarose, RCA-conjugated agarose and WGA-conjugated agarose were purchased from J-OIL MILLS. Cell culture Human embryonic kidney cell line 293FT (Thermo Fisher Scientific, Waltham, MA) was cultivated in DMEM high glucose (Thermo Fisher Scientific) with 1 mM sodium pyruvate (Thermo Fisher Scientific), non-essential amino acids solution (Thermo Fisher Scientific) and 10% fetal bovine serum (FBS; Tissue Culture Biologicals, Tulare, CA). Human breast cancer cell line MCF-7 and human oesophageal cancer cell line TE8 purchased from Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University (Miyagi, Japan), human mesothelioma cell line MSTO-211H, human lung cancer cell line NCI-H226 and human lung cancer cell line NCI-H460 purchased from ATCC (Manassas, VA) and human liver cancer cell line HLF purchased from Japanese Collection of Research Bioresources Cell Bank (JCRB) (Tokyo, Japan) were cultivated in RPMI1640 (Thermo Fisher Scientific) with 10% FBS. Human fetal lung fibroblast TIG-1 cells were purchased from JCRB and cultured in DMEM (Thermo Fisher Scientific) with 10% FBS. All cell lines were cultured at 37°C in 5% CO2 atmosphere. Preparation of antibody-conjugated resins Antibody #141- or JK132-conjugated resin was prepared as follows. Two milligrams of antibody #141 or JK132 was added to 1 ml of Protein G Sepharose 4 Fast Flow (GE Healthcare) and each resin was incubated at 4°C for 1 h with gentle shaking. The resin was equilibrated with 0.2 M triethanolamine-HCl (pH 8.2) and incubated with 50 mM dimethyl pimelimidate at room temperature with gentle shaking to cross-link the resin and the antibody. After 1 h, the resin was equilibrated with 0.2 M ethanolamine-HCl (pH 8.2) and incubated at room temperature for 1 h with gentle shaking to block unreacted amines. After blocking, the resin was washed and equilibrated with phosphate-buffered saline (PBS; pH 7.4). Purification of the 140 kDa polypeptide by affinity chromatography on antibody #141-conjugated resin All purification procedures were performed at 4°C. The 140 kDa polypeptide was purified from conditioned medium of HLF cell culture as follows. One thousand and five hundred millilitres of the conditioned medium was passed through 6 ml of antibody #141-conjugated resin to adsorb polypeptide(s). After adsorption, the resin was washed with PBS and bound polypeptides including the 140 kDa polypeptide were eluted with 0.1 M glycine-HCl (pH 2.8). Finally, the 140 kDa polypeptide-containing fractions were collected, concentrated with Amicon Ultra-15 (Merck Millipore, Billerica, MA) and dialyzed against PBS. Extraction of human type VI collagen and its analysis with dot blotting The dermis-like structure formed by long-term (31 days) culture of human skin fibroblast, HF18, was used as starting material for type VI collagen preparation as reported previously (9). The lyophilized material (1.10 g) of the dermis-like structure was suspended in 5 ml of 50 mM Tris–HCl (pH 7.5) and 1 M NaCl with a protease inhibitor cocktail comprising 5 mM EDTA, 100 µM N-ethylmaleimide and 100 µM phenylmethylsulfonyl fluoride overnight. Isolation procedure was performed at 4°C. The supernatant was collected by centrifugation at 10,000 ×g for 30 min. The total protein concentration of the collected supernatant was estimated to be 0.1 mg/ml. SDS-PAGE analysis using CBB staining revealed that type VI, type I and type V collagens became visible after concentrating with Amicon Ultra-0.5 (Merck Millipore) by 10 times. Dot blotting was carried out as follows. Samples containing type VI collagen were subjected to heat denaturation at 90°C for 10 min with or without 0.02% 2-mercaptoethanol (2-ME) (denoted in Fig. 5A as Heated and reducing and Heated and non-reducing, respectively). Non-heated samples with or without 2-ME were also prepared. Samples containing indicated amount of type VI collagen were diluted with TBS (25 mM Tris–HCl, pH 7.5, 137 mM NaCl, 2.7 mM KCl) and blotted onto polyvinylidene difluoride (PVDF) membrane (Immobilon-P, Merck Millipore) using the filtration manifold system MilliBlot system (Merck Millipore). Each blot hole was washed with TBS once and the membrane was blocked with 1/4 dilution of Block Ace (DS Pharma Biochemical, Osaka, Japan). The membrane was incubated with antibody #141 and washed with Tris-buffered saline-Tween20 (20 mM Tris–HCl, pH 7.4, 150 mM NaCl, 0.05% Tween20). The bound antibodies were detected with horseradish peroxidase-labelled anti-mouse IgG antibody and visualized using Immobilon Western Chemiluminescent HRP Substrate (Merck Millipore) on an image analyser (ImageQuant LAS4000; GE Healthcare). Amino acid sequence analysis of the 140 kDa polypeptide Amino acid sequence analysis of the 140 kDa polypeptide was carried out in APRO Life Science Institute (Tokushima, Japan) as follows. An equal amount of 2× concentrated sample buffer with 2-ME (125 mM Tris–HCl, pH 6.8, 20% glycerol, 10% 2-ME, 4% SDS) was added to the purified 140 kDa polypeptide solution. The mixture was heated at 95°C for 5 min, loaded onto 7.5% SDS-PAGE gel and electrophoresed. After electrophoresis, the band of 140 kDa polypeptide was cut out from the gel and the gel was digested with endopeptidase Lys-C (Wako Pure Chemical Industries, Osaka, Japan) at 35°C for 20 h. Digested peptides were injected into an Aliance HPLC 2695 Separations Module System (Waters, Milford, MA) and separations were performed using a Symmetry C18 column (100 Å, 3.5 μm, 1.0 × 150 mm, Waters). The mobile phases consisted of A: 0.10% trifluoroacetic acid (TFA)-2% acetonitrile and B: 0.09% TFA-90% acetonitrile. Gradient conditions were 0% of mobile phase B for 6 min, 0–10% of mobile phase B in 5 min, 10–50% of mobile phase B in 75 min, 50–100% of mobile phase B in 5 min and 100% of mobile phase B for 5 min with a flow rate of 1 ml/min. Separated peptides were collected and analysed by a Procise 494 HT Protein Sequencing System (Thermo Fisher Scientific) according to the manufacturer’s instruction. Of note, peptides were detected at a UV wavelength of 210 nm and the reference wavelength was set at 280 nm. Preparation of mammalian expression vectors and transient expressions of those vectors in 293FT cells COL4A1 and COL6A1 coding regions were amplified from HLF cDNA by PCR with primers as described in Table I and each of coding regions was subcloned into pENTR vector (Thermo Fisher Scientific). For mammalian expression, COL4A1 or COL6A1 coding region was transferred to pcDNA3.2/V5-DEST vector (Thermo Fisher Scientific) by GATEWAY technology (Thermo Fisher Scientific) according to the manufacturer’s instruction. Table I. Primers used in this study Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  Table I. Primers used in this study Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  For transient expression, we transfected the indicated mammalian expression vector to 293FT cells with Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer’s instruction. After transfection, 293FT cells were cultivated at 37°C in 5% CO2 atmosphere for 3 days and the conditioned media were subjected to western blotting analysis. Immuno- or lectin precipitation Immuno- or lectin precipitation was carried out as described previously with slight modifications (4). Briefly, 10 µl of the indicated antibody-conjugated resin or lectin-conjugated agarose was added to 1 ml of the indicated conditioned medium and incubated at 4°C for 1 h followed by centrifugation. The resin and the agarose were washed with wash buffer (50 mM Tris–HCl, pH 7.4, 300 mM NaCl, 0.1% Triton X-100). After washing, precipitates were eluted with 2× concentrated sample buffer with 2-ME at 95°C for 5 min. Precipitates were subjected to western blotting analysis. Western blotting and CBB stain Western blotting was carried out as described previously with slight modifications (4). Briefly, an equal amount of 2× concentrated sample buffer with 2-ME was added to a sample and boiled at 95°C for 5 min. Of note, under non-reducing conditions, 2× concentrated sample buffer without 2-ME was used. The mixture was loaded onto 7.5% SDS-PAGE gel or 5–20% SuperSep Ace precast gel (Wako Pure Chemical Industries) and electrophoresed. The gel was transferred to PVDF membrane and blocked with Tris-buffered saline-Tween20 containing 5% (weight/volume) skim milk (BD, Franklin Lakes, NJ). The membrane was incubated with pre-determined antibodies or biotin-labelled lectins and washed with Tris-buffered saline-Tween20. The bound antibodies were detected with horseradish peroxidase-labelled anti-mouse or anti-rabbit IgG antibody. Of note, in the case of using a biotin-labelled lectin as a probe, Tris-buffered saline-Tween20 and horseradish peroxidase-conjugated streptavidin (GE Healthcare) were used for blocking and detection according to the manufacturer’s instruction. The membrane was developed with ECL (GE Healthcare) or ECL Prime (GE Healthcare) on a Hyperfilm ECL (GE Healthcare). In some experiments, we used the same membrane for detection with different antibodies. To reuse the membrane, the membrane was incubated at 55°C for 30 min in stripping buffer (62.5 mM Tris–HCl, pH 6.5, 100 mM 2-ME, 2% SDS) for stripping antibodies or lectins off and reprobed with the indicated antibody as described previously (10). Then, the membrane was subjected to western blotting. CBB stain was carried out with Rapid Stain CBB Kit (Nacalai Tesque, Kyoto, Japan) according to the manufacturer’s instruction. Results Antibody #141 has a higher affinity for 140 kDa polypeptide than that for NTH α1(IV) In the course of characterization of antibody #141, we found that antibody #141 detected both NTH α1(IV) and a band at 140 kDa (hereafter 140K) in supernatants of several cancer cell lines, human liver cancer cell line HLF, human oesophageal cancer cell line TE8 and human lung cancer cell line NCI-H226, under heated and non-reducing conditions (Fig. 1A), indicating that the 140K forms a non-disulphide bonded structure. Figure 1A also shows that JK132 bound to NTH α1(IV) (180 kDa) and two extra bands at 135 and 160 kDa. Iwata et al. reported the existence of 160 kDa of a short isoform of α1(IV) chain in addition to 180 kDa of α1(IV) chain. However, the band at 135 kDa (hereafter 135K) has not been reported (11). The 135K is slightly different from 140K in size, suggesting that the 140K and 135K are unknown isoforms of NTH α1(IV). Next, we carried out immuno-precipitation experiment. Figure 1B shows that antibody #141 could precipitate the 140K but not the 135K and JK132 could precipitate the 135K but not 140K, vice versa. Moreover, antibody #141 reactivity to the 140K is higher than the reactivity to NTH α1(IV) from the finding that antibody #141 precipitated larger amounts of the 140K than NTH α1(IV) (Fig. 1B). JK132 is reported to recognize the amino acid sequences contained in the region of 1165–1179 in human α1(IV) chain (1, 11) and antibody #141 was obtained in our laboratory using NTH α1(IV) as an antigen (4). Therefore, we hypothesized that the 140K and the 135K were novel isoforms of NTH α1(IV) and the differences of the affinities to the 140K between antibody #141 and JK132 were caused by the differences of recognition sequences between the two antibodies. Fig. 1 View largeDownload slide Recognition of a band at 140 kDa by antibody #141. (A) Representative image of a band at 140 kDa detected by antibody #141. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The conditioned medium was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with JK132 (IB: JK132) as described in Materials and Methods section. After JK132 detection, the membrane was stripped with the stripping buffer to remove JK132. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. (B) Representative image of precipitates of JK132 and antibody #141. Proteins in the conditioned medium of HLF cell culture were immuno-precipitated with JK132- or antibody #141-conjugated resin (IP JK132 or IP #141 in the figure). The precipitates were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions) and electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with antibody #141 (IB: #141) as described in Materials and Methods section. NTH α1(IV), the band at 135 kDa detected by JK132 (hereafter 135K), and the band at 140 kDa detected by antibody #141 (hereafter 140K) are indicated by filled arrowheads, an arrow, and an open arrowhead, respectively. Fig. 1 View largeDownload slide Recognition of a band at 140 kDa by antibody #141. (A) Representative image of a band at 140 kDa detected by antibody #141. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The conditioned medium was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with JK132 (IB: JK132) as described in Materials and Methods section. After JK132 detection, the membrane was stripped with the stripping buffer to remove JK132. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. (B) Representative image of precipitates of JK132 and antibody #141. Proteins in the conditioned medium of HLF cell culture were immuno-precipitated with JK132- or antibody #141-conjugated resin (IP JK132 or IP #141 in the figure). The precipitates were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions) and electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with antibody #141 (IB: #141) as described in Materials and Methods section. NTH α1(IV), the band at 135 kDa detected by JK132 (hereafter 135K), and the band at 140 kDa detected by antibody #141 (hereafter 140K) are indicated by filled arrowheads, an arrow, and an open arrowhead, respectively. Identification of the 140K as a polypeptide of α1(VI) chain To confirm the hypothesis, we tried to identify the 140K. The 140K was purified from the conditioned medium of HLF cell culture as described in Materials and Methods section. CBB staining shows that the purified sample contained the 140K as a major polypeptide with smaller polypeptides in less amounts (Fig. 2A). The 140K and one of the smaller polypeptides were immuno-stained with antibody #141 as shown in Fig. 2B. The purity of 140K was adequate for amino acid sequence analysis, but the N-terminal amino acid identification of the 140K failed. The failure of amino acid sequence analysis might be caused by some post translational modifications or formation of pyroglutamate at its N-terminal. Fig. 2 View largeDownload slide Evaluation of the 140K purity and antibody #141 reactivity. (A) Representative image of purified 140K. The 140K was purified from the conditioned medium of HLF cell culture as described in Materials and Methods section. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The 140K was electrophoresed onto 7.5% SDS-PAGE gel and analysed by CBB staining as described in Materials and Methods section. The indicated amount of BSA was loaded for estimation of the 140K quantity. (B) Antibody #141 reactivity toward purified 140K. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The mixture was electrophoresed and western blotting was carried out by probing with antibody #141(IB: #141) as described in Materials and Methods section. After the antibody #141 detection, the bound antibody was stripped off from the membrane and stained with CBB as described in Materials and Methods section. The 140K is shown by an open arrowhead. Fig. 2 View largeDownload slide Evaluation of the 140K purity and antibody #141 reactivity. (A) Representative image of purified 140K. The 140K was purified from the conditioned medium of HLF cell culture as described in Materials and Methods section. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The 140K was electrophoresed onto 7.5% SDS-PAGE gel and analysed by CBB staining as described in Materials and Methods section. The indicated amount of BSA was loaded for estimation of the 140K quantity. (B) Antibody #141 reactivity toward purified 140K. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The mixture was electrophoresed and western blotting was carried out by probing with antibody #141(IB: #141) as described in Materials and Methods section. After the antibody #141 detection, the bound antibody was stripped off from the membrane and stained with CBB as described in Materials and Methods section. The 140K is shown by an open arrowhead. Because the N-terminal of the 140K could not be analysed, internal peptide fragments of the 140K were generated by endopeptidase Lys-C digestion and N-terminal amino acid sequence analysis of the digested peptides was carried out. In brief, the digested peptides were separated by reverse-phased HPLC and the fractions of number 30, 40, 47 and 58 as shown in Fig. 3 were subjected to amino acid sequence analysis. Table II indicates that all the peptide fragments except for fraction number 40 were identified as those from COL6A1 (RefSeq accession number: NP_001839.2). Of note, fraction number 58 gave two sequences both of which matched sequences within COL6A1 and fraction number 40 could not be identified for unknown reasons. As all the peptides analysed could be ascribed to COL6A1 sequences, the 140K is likely a polypeptide of α1(VI) chain encoded by COL6A1. Table II. Amino acid sequences of the four peptides derived from purified 140K Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Table II. Amino acid sequences of the four peptides derived from purified 140K Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Fig. 3 View largeDownload slide Elution profile of peptides derived from endopeptidase Lys-C digested 140K. Purified 140K was digested with endopeptidase Lys-C and the digested peptides were separated by reverse-phased HPLC. The indicated fractions (marked by arrows with fraction numbers) were subjected to amino acid sequence analysis as described in Materials and Methods section. Blue line: absorbance at 210 nm for peptide detection. Red line: absorbance at 280 nm for reference. AU: arbitrary unit. Fig. 3 View largeDownload slide Elution profile of peptides derived from endopeptidase Lys-C digested 140K. Purified 140K was digested with endopeptidase Lys-C and the digested peptides were separated by reverse-phased HPLC. The indicated fractions (marked by arrows with fraction numbers) were subjected to amino acid sequence analysis as described in Materials and Methods section. Blue line: absorbance at 210 nm for peptide detection. Red line: absorbance at 280 nm for reference. AU: arbitrary unit. Although antibody #141 was obtained using NTH α1(IV) as an antigen (4), the amino acid sequence analysis indicated that the 140K is derived from COL6A1. To confirm the result of the amino acid sequence analysis, we transfected COL4A1 or COL6A1 expression vector to 293FT cells and determined whether antibody #141 recognized COL6A1. In addition, we checked that JK132 did not detect COL6A1, since JK132 recognizes the sequence contained in the region of 1165–1179 (KGEPGLPGRGFPGFPT) in COL4A1 (1, 11) and this sequence is not conserved in COL6A1. Figure 4 shows that antibody #141 recognizes both COL4A1 and COL6A1 under heated and reducing conditions, whereas JK132 recognizes only COL4A1 but not COL6A1. This result confirms that the 140K is a polypeptide of the α1(VI) chain in size. Fig. 4 View largeDownload slide Analysis of antibody #141 reactivity toward COL6A1. COL4A1 or COL6A1 expression vector was introduced into 293FT cells with Lipofectamine 2000 according to the manufacture’s instruction and the conditioned medium was harvested. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer with 2-ME and the mixture was boiled at 95°C (heated and reducing conditions). The mixture was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out as described in Materials and Methods section. Of note, the membrane was sequentially probed with the indicated antibody as follows; antibody #141 (IB: #141), JK132 (IB: JK132) and anti-COL6A1 polyclonal antibody (N-term) (IB: αCOL6A). After the detection, the first antibody was removed from the membrane. The membrane was reprobed with the indicated antibody and western blotting was carried out as described in Materials and Methods section. COL4A1 and COL6A1 are shown by filled arrowheads and open arrowheads, respectively. Fig. 4 View largeDownload slide Analysis of antibody #141 reactivity toward COL6A1. COL4A1 or COL6A1 expression vector was introduced into 293FT cells with Lipofectamine 2000 according to the manufacture’s instruction and the conditioned medium was harvested. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer with 2-ME and the mixture was boiled at 95°C (heated and reducing conditions). The mixture was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out as described in Materials and Methods section. Of note, the membrane was sequentially probed with the indicated antibody as follows; antibody #141 (IB: #141), JK132 (IB: JK132) and anti-COL6A1 polyclonal antibody (N-term) (IB: αCOL6A). After the detection, the first antibody was removed from the membrane. The membrane was reprobed with the indicated antibody and western blotting was carried out as described in Materials and Methods section. COL4A1 and COL6A1 are shown by filled arrowheads and open arrowheads, respectively. Fig. 5 View largeDownload slide Analysis of antibody #141 specificity. (A) Representative image of antibody #141 reactivity toward human type VI collagen. The indicated amount of human type VI collagen was blotted onto PVDF membrane. The membrane was probed with antibody #141 (IB: #141) and visualized as described in Materials and Methods section. (B) Confirmation of antibody #141 reactivity toward human α1(VI) chain under heated and reducing conditions. The partially purified type VI collagen sample was analysed by western blotting with antibody #141. The sample contained only the α1(VI) chain derived from type VI collagen as an antibody #141-reactive antigen. Of note, the conditioned medium of TIG-1 cell culture in the ascorbate-depleted medium was used as a positive control. The α1(IV) chain and the α1(VI) chain are shown by a filled arrowhead and an open arrowhead, respectively. HuVI: human type VI collagen. Fig. 5 View largeDownload slide Analysis of antibody #141 specificity. (A) Representative image of antibody #141 reactivity toward human type VI collagen. The indicated amount of human type VI collagen was blotted onto PVDF membrane. The membrane was probed with antibody #141 (IB: #141) and visualized as described in Materials and Methods section. (B) Confirmation of antibody #141 reactivity toward human α1(VI) chain under heated and reducing conditions. The partially purified type VI collagen sample was analysed by western blotting with antibody #141. The sample contained only the α1(VI) chain derived from type VI collagen as an antibody #141-reactive antigen. Of note, the conditioned medium of TIG-1 cell culture in the ascorbate-depleted medium was used as a positive control. The α1(IV) chain and the α1(VI) chain are shown by a filled arrowhead and an open arrowhead, respectively. HuVI: human type VI collagen. Conformation-dependent recognition of the α1(VI) chain with antibody #141 As shown above, antibody #141 recognizes a non-disulphide-bonded polypeptide of the α1(VI) chain as well as NTH αl(IV) (Fig. 1). The reactivity of this antibody to the α1(VI) chain could also be dependent on the conformation of type VI collagen. Therefore, we examined whether heat denaturation of triple helical type VI collagen molecule gave rise to the reactivity to antibody #141 by dot blot analysis, using partially purified human type VI collagen. Figure 5A indicates that antibody #141 bound to the α1(VI) chain derived from type VI collagen after heat denaturation under reducing or non-reducing condition. Positive reaction to the α1(VI) chain in native structure under non-heated and non-reducing conditions was negligibly small. Furthermore, antibody #141 hardly recognized the α1(VI) chain under non-heated and reducing conditions. The western blotting analysis using antibody #141 in Fig. 5B showed that the sample contained mainly one band at 140K corresponding for the α1(VI) chain probably derived from type VI collagen under heated and reducing condition. Since there were no visible band at 180 kDa of α1(IV) chain nor NTH α1(IV), the sample contained only the α1(VI) chain as an antibody #141-reactive antigen. These observations indicate that antibody #141 recognizes the α1(VI) chain in a conformation-dependent manner, suggesting that the specificity of antibody #141 to the 140K is caused by its conformation. Therefore, it is considered that the 140K is a novel non-triple helical polypeptide encoded by COL6A1 and we name it NTH α1(VI). Since NTH α1(VI) was detected in culture supernatant without reducing agents (Fig. 1A), it is not disulphide-bonded with α2(VI) or α3(VI) chain. Characterization of sugar chains of NTH α1(VI) To characterize post-translational modifications of NTH α1(VI), we examined the presence and types of sugar chains on NTH α1(VI) with several lectins. Figure 6A shows that LCA precipitated NTH α1(VI) s derived from HT1080 and Lu65A cells, whereas RCA120 reacted with the NTH α1(VI) recovered from HT1080 but not the NTH α1(VI) from Lu65A cells and WGA did not precipitate either NTH α1(VI). The reaction with RCA120 implies that NTH α1(VI) is differentially glycosylated between these cell lines. The three lectins recognize different moieties of sugar chains of the glycoprotein. (i) LCA binds to a core fucose of N-linked glycan (12, 13), (ii) RCA120 binds to a non-reducing terminal β-galactose (Gal) (14). (iii) WGA binds to a non-reducing terminal N-acetyl-d-glucosamine (GlcNAc) and its oligomer, such as N, N' diacetyl-chitobiose (GlcNAc-β-1, 4-GIcNAc), and a non-reducing terminal sialic acid (NeuAc) (15–17). Therefore, common characteristics of sugar chains of NTH α1(VI) are as follows; core-fucose structure and without GlcNAc and NeuAc at non-reducing terminal. To characterize the glycosylation status further, we examined lectin reactivity toward human type VI collagen. Figure 6B shows that LCA, RCA120 and WGA recognized human type VI collagen under heated and reducing conditions. Fujiwara et al. reported that sugar chains of type VI collagen have N-glycans with core-fucose structure and Gal or NeuAc, but not GlcNAc at non-reducing terminal (18). This report coincides with our result as shown in Fig. 6B, indicating that sugar chains of type VI collagen have core-fucose structure and Gal or NeuAc at non-reducing terminal. In contrast, Fig. 6A shows that the sugar chains of NTH α1(VI) do not have NeuAc at non-reducing terminal by WGA reactivity, suggesting that the sugar chains are different between NTH α1(VI) and the α1(VI) chain in triple helical form. Fig. 6 View largeDownload slide Characterization of sugar chains of NTH α1(VI) and type VI collagen. (A) Characterization of sugar chains of NTH α1(VI). Sugar chains of NTH α1(VI) from the conditioned medium of HT1080 and Lu65A cells were analysed using the indicated lectins. Briefly, 10 µl of the indicated lectin-conjugated agarose was added to 1 ml of conditioned medium from HT1080 or Lu65A cell culture. Precipitates including NTH α1(VI) were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions). The precipitates were separated onto 7.5% SDS-PAGE gel and NTH α1(VI) was detected with antibody #141 (IB: #141) as described in Materials and Methods section. (B) Characterization of sugar chains of type VI collagen. Ten microlitres of partially purified type VI collagen was denatured under heated and reducing conditions as similar to (a). Type VI collagen was separated onto 5–20% SuperSep Ace precast gel and sugar chains of type VI collagen were analysed using the indicated lectins as described in Materials and Methods section. After the analysis, the lectins bound on blotting membrane were removed. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. NTH α1(VI) in (A) and the α1(VI) chain in (B) are shown by open arrowheads. LCA, lectin Lens culinaris agglutinin; RCA120, lectin Ricinus communis agglutinin I; WGA, lectin wheat germ agglutinin; LP, lectin precipitation. Fig. 6 View largeDownload slide Characterization of sugar chains of NTH α1(VI) and type VI collagen. (A) Characterization of sugar chains of NTH α1(VI). Sugar chains of NTH α1(VI) from the conditioned medium of HT1080 and Lu65A cells were analysed using the indicated lectins. Briefly, 10 µl of the indicated lectin-conjugated agarose was added to 1 ml of conditioned medium from HT1080 or Lu65A cell culture. Precipitates including NTH α1(VI) were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions). The precipitates were separated onto 7.5% SDS-PAGE gel and NTH α1(VI) was detected with antibody #141 (IB: #141) as described in Materials and Methods section. (B) Characterization of sugar chains of type VI collagen. Ten microlitres of partially purified type VI collagen was denatured under heated and reducing conditions as similar to (a). Type VI collagen was separated onto 5–20% SuperSep Ace precast gel and sugar chains of type VI collagen were analysed using the indicated lectins as described in Materials and Methods section. After the analysis, the lectins bound on blotting membrane were removed. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. NTH α1(VI) in (A) and the α1(VI) chain in (B) are shown by open arrowheads. LCA, lectin Lens culinaris agglutinin; RCA120, lectin Ricinus communis agglutinin I; WGA, lectin wheat germ agglutinin; LP, lectin precipitation. Discussion In this study, we show that a novel non-triple helical collagen polypeptide NTH α1(VI) exists in several cancer cell lines (Figs 1A and 5). Furthermore, our results indicate that NTH α1(VI) is an alternative gene product of COL6A1 (Table II and Fig. 4). Another non-triple helical collagen polypeptide NTH α1(IV) is also reported (1–4). NTH α1(IV) is an alternative gene product of COL4A1 and shows a non-disulphide bonded structure (1). In addition, NTH α1(IV) is specifically recognized by anti-α1(IV) chain antibody JK132 (8), because an epitope of JK132 is hidden in triple helical conformation (1, 11), indicating that a higher order structure of NTH α1(IV) is different from the α1(IV) chain in triple helical form. Antibody #141 recognizes only non-triple helical form of α1(VI) chain (Fig. 5). Our result is similar to the reactivity of JK132 to NTH α1(IV), indicating that conformation of NTH α1(VI) is different from that of the α1(VI) chain in triple helical form. We show that antibody #141 recognizes NTH α1(IV) and NTH α1(VI) (Fig. 1). To clarify the specific reactivity of antibody #141 toward both polypeptides, we have been investigating the recognition sequence of antibody #141. We found that antibody #141 recognizes the region between 1055 and 1064 amino acid in COL4A1 and the region between 254 and 286 amino acid in COL6A1, suggesting that antibody #141 recognizes a common epitope in both molecules (T. Sato et al., manuscript in preparation). Therefore, it is considered that the specificity of antibody #141 toward NTH α1(IV) and NTH α1(VI) results from the recognition sequence. NTH α1(VI) has core-fucose structure as similar to type VI collagen (18), but the sugar chains of NTH α1(VI) do not have NeuAc at non-reducing terminal (Fig. 6). N-linked glycans are synthesized in endoplasmic reticulum through Golgi and have a common core pentasaccharide and multiple glycosylations are formed on the common core structure catalyzed by various enzymes (19–23). NeuAc is transferred to non-reducing terminal by the actions of sialyltransferases. It is also known that enzymatic specificities toward substrates are varied among sialyltransferases (22, 23). Differences of the non-reducing terminals between NTH α1(VI) and the α1(VI) chain may be caused by affinities of sialyltransferases to NTH α1(VI), since NTH α1(VI) has a non-triple helical structure (Fig. 5). However, structures of sugar chains attached to NTH α1(VI) are not fully determined yet. Further studies are needed. Type VI collagen is synthesized as a triple helical form (24–26). An α3(VI) chain is required for proper folding of type VI collagen. It has been discussed that in the absence of α3(VI) chain, α1(VI) and α2(VI) chains are retained within cells as unassembled polypeptides and degraded (27). NTH α1(VI) expression in supernatants of several cancer cell lines (Fig. 1A) raises a query whether or not α3(VI) chain is dispensable for NTH α1(VI) synthesis. The glycosylation status of NTH α1(VI) is different from type VI collagen (Fig. 6), suggesting that NTH α1(VI) is synthesized and secreted via different mechanism from the α1(VI) chain in triple helical form. It is reported that type VI collagen is involved in tumourigenesis, such as cancer cell growth (28, 29) and apoptosis (30, 31), and the mode of action of type VI collagen is mediated via interaction with β1 integrins or NG2/chondroitin sulfate proteoglycan (27, 32–34). As indicated, NTH α1(VI) expression is observed in several cancer cell lines (Fig. 1A), suggesting that NTH α1(VI) is also involved in tumourigenesis. The higher-order structure and the glycosylation status of NTH α1(VI) are different from type VI collagen (Figs 5 and 6). It is reported that higher-order structure of protein (e.g. type I collagen fibril and polymerized fibronectin) and glycosylation status (e.g. glycosylation defect in muscular dystrophy) influence biological functions (35–37), suggesting that the mode of action and the biological function of NTH α1(VI) are distinct from type VI collagen. Park et al. also reported that endotrophin (38), a cleaved product of COL6A3, promotes breast cancer growth, metastasis, epithelial mesenchymal transition and tumour angiogenesis (39). Because NTH α1(VI) is encoded by COL6A1 and the amino acid sequence of endotrophin is not conserved in COL6A1, the mode of action of NTH α1(VI) is also different from endotrophin. The mode of action and the biological function of NTH α1(VI) remain to be clarified and those are under investigation in our laboratory. In conclusion, we report a novel NTH form of collagen gene products as NTH α1(VI) encoded by COL6A1. Expression of NTH α1(VI) is observed in supernatants of several human cancer cell lines, suggesting that NTH α1(VI) is involved in tumourigenesis. Reactivity with lectins indicates that sugar chains of NTH α1(VI) are different from those of the α1(VI) chain in triple helical form, suggesting that the mechanism of synthesis, the mode of action and the biological function of NTH α1(VI) are different from those of the α1(VI) chain in triple helical form or type VI collagen as in the case of NTH α1(IV). Acknowledegement We thank Ms Chieko Seno for experimental support and Dr Keiko Sekine for technical advice. Funding This work was supported in part by a grant from the Strategic Research Foundation Grant-aided Project for Private Universities from the Ministry of Education, Culture, Sport, Science, and Technology, Japan, 2014-2018 [S1411005 to Y.I.]. Conflict of Interest None declared. References 1 Takahashi S., Yoshikawa K., Sasaki T., Takeda Y., Imamura Y., Sado Y., Hayashi T. ( 1999) Serum-dependent secretion of nondisulfide-bonded and unfolded type IV collagen α chains by cultured fetal lung fibroblasts. Connect. Tissue . 31, 161– 168 2 Kajimura D., Takahashi S., Yoshikawa K., Hattori S., Sado Y., Imamura Y., Hayashi T. ( 2004) Non-helical type IV collagen polypeptides in human placenta. Biochem. Biophys. Res. Commun  314, 11– 16 Google Scholar CrossRef Search ADS PubMed  3 Sugiyama H., Tokunaka K., Hayashi T., Imamura Y., Morita M., Yamato M. ( 2015) Non-triple helical form of type IV collagen α1 chain. Heliyon  1, e00051 Google Scholar CrossRef Search ADS PubMed  4 Morita M., Sugihara H., Tokunaka K., Tomura A., Saiga K., Sato T., Imamura Y., Hayashi T. ( 2017) Preparation and partial characterization of monoclonal antibodies specific for the nascent non-triple helical form of the type IV collagen alpha 1 chain. Biochem. Biophys. Rep  9, 128– 132 Google Scholar PubMed  5 Yoshikawa K., Takahashi S., Imamura Y., Sado Y., Hayashi T. ( 2001) Secretion of Non-helical collagenous pblypeptides of α1(IV) and α2(IV) chains upon depletion of ascorbate by cultured human cells. J. Biochem . 129, 929– 936 Google Scholar CrossRef Search ADS PubMed  6 Presant C.A., Kornfeld S. ( 1972) Characterization of the cell surface receptor for the Agaricus bisporus hemagglutinin. J. Biol. Chem . 247, 6937– 6945 Google Scholar PubMed  7 Lu P., Weaver V.M., Werb Z. ( 2012) The extracellular matrix: a dynamic niche in cancer progression. J. Cell Biol.  196, 395– 406 Google Scholar CrossRef Search ADS PubMed  8 Kino J., Adachi E., Yoshida T., Nakajima K., Hayashi T. ( 1988) Characterization of a monoclonal antibody against human placenta type IV collagen by immunoelectroblotting, antibody-coupled affinity chromatography, and immunohistochemical localization. J. Biochem . 103, 829– 835 Google Scholar CrossRef Search ADS PubMed  9 Hazeki N., Yamato M., Imamura Y., Sasaki T., Nakazato K., Yamamoto K., Konomi H., Hayashi T. ( 1998) Analysis of matrix protein components of the dermis-like structure formed in a long-term culture of human fibroblasts: type VI collagen is a major component. J. Biochem . 123, 587– 595 Google Scholar CrossRef Search ADS PubMed  10 Kaufmann S.H., Ewing C.M., Shaper J.H. ( 1987) The erasable Western blot. Anal. Biochem.  161, 89– 95 Google Scholar CrossRef Search ADS PubMed  11 Iwata M., Imamura Y., Sasaki T., Hayashi T. ( 1995) Evidence for a short form of alpha 1(IV) as a major polypeptide in bovine lens capsule. J. Biochem . 117, 1298– 1304 Google Scholar CrossRef Search ADS PubMed  12 Kornfeld K., Reitman M.L., Kornfeld R. ( 1981) The carbohydrate-binding specificity of pea and lentil lectins. Fucose is an important determinant. J. Biol. Chem . 256, 6633– 6640 Google Scholar PubMed  13 Tateno H., Nakamura-Tsuruta S., Hirabayashi J. ( 2009) Comparative analysis of core-fucose-binding lectins from Lens culinaris and Pisum sativum using frontal affinity chromatography. Glycobiology  19, 527– 536 Google Scholar CrossRef Search ADS PubMed  14 Baenziger J.U., Fiete D. ( 1979) Structural determinants of Ricinus communis agglutinin and toxin specificity for oligosaccharides. J. Biol. Chem . 254, 9795– 9799 Google Scholar PubMed  15 Wright C.S. ( 1987) Refinement of the crystal structure of wheat germ agglutinin isolectin 2 at 1.8 Å resolution. J. Mol. Biol.  194, 501– 529 Google Scholar CrossRef Search ADS PubMed  16 Wright C.S. ( 1992) Crystal structure of a wheat germ agglutinin/glycophorin-sialoglycopeptide receptor complex. Structural basis for cooperative lectin-cell binding. J. Biol. Chem . 267, 14345– 14352 Google Scholar PubMed  17 Wright C.S., Kellogg G.E. ( 1996) Differences in hydropathic properties of ligand binding at four independent sites in wheat germ agglutinin-oligosaccharide crystal complexes. Protein Sci.  5, 1466– 1476 Google Scholar CrossRef Search ADS PubMed  18 Fujiwara S., Shinkai H., Timpl R. ( 1991) Structure of N-linked oligosaccharide chains in the triple-helical domains of human type VI and mouse type IV collagen. Matrix  11, 307– 312 Google Scholar CrossRef Search ADS PubMed  19 Bieberich E. ( 2014) Synthesis, processing, and function of N-glycans in N-glycoproteins. Adv. Neurobiol.  9, 47– 70 Google Scholar CrossRef Search ADS PubMed  20 Schachter H. ( 2014) Complex N-glycans: the story of the “yellow brick road”. Glycoconj. J.  31, 1– 5 Google Scholar CrossRef Search ADS PubMed  21 Kizuka Y., Taniguchi N. ( 2016) Enzymes for N-glycan branching and their genetic and nongenetic regulation in cancer. Biomolecules  6, 25 Google Scholar CrossRef Search ADS   22 Harduin-Lepers A., Mollicone R., Delannoy P., Oriol R. ( 2005) The animal sialyltransferases and sialyltransferase-related genes: a phylogenetic approach. Glycobiology  15, 805– 817 Google Scholar CrossRef Search ADS PubMed  23 Bhide G.P., Colley K.J. ( 2017) Sialylation of N-glycans: mechanism, cellular compartmentalization and function. Histochem. Cell Biol.  147, 149– 174 Google Scholar CrossRef Search ADS PubMed  24 Trueb B., Winterhalter K.H. ( 1986) Type VI collagen is composed of a 200 kd subunit and two 140 kd subunits. EMBO J.  5, 2815– 2819 Google Scholar PubMed  25 Chu M.-L., Mann K., Deutzmann R., Pribula-Conway D., Hsu-Chen C.-C., Bernard M.P., Timpl R. ( 1987) Characterization of three constituent chains of collagen type VI by peptide sequences and cDNA clones. Eur. J. Biochem.  168, 309– 317 Google Scholar CrossRef Search ADS PubMed  26 Weil D., Mattei M.G., Passage E., N'Guyen V.C., Pribula-Conway D., Mann K., Deutzmann R., Timpl R., Chu M.L. ( 1988) Cloning and chromosomal localization of human genes encoding the three chains of type VI collagen. Am. J. Hum. Genet . 42, 435– 445 Google Scholar PubMed  27 Loeser R.F., Sadiev S., Tan L., Goldring M.B. ( 2000) Integrin expression by primary and immortalized human chondrocytes: evidence of a differential role for alpha1beta1 and alpha2beta1 integrins in mediating chondrocyte adhesion to types II and VI collagen. Osteoarthritis Cartilage  8, 96– 105 Google Scholar CrossRef Search ADS PubMed  28 Iyengar P., Espina V., Williams T.W., Lin Y., Berry D., Jelicks L.A., Lee H., Temple K., Graves R., Pollard J., Chopra N., Russell R.G., Sasisekharan R., Trock B.J., Lippman M., Calvert V.S., Petricoin E.F.III, Liotta L., Dadachova E., Pestell R.G., Lisanti M.P., Bonaldo P., Scherer P.E. ( 2005) Adipocyte-derived collagen VI affects early mammary tumor progression in vivo, demonstrating a critical interaction in the tumor/stroma microenvironment. J. Clin. Invest.  115, 1163– 1176 Google Scholar CrossRef Search ADS PubMed  29 You W.K., Bonaldo P., Stallcup W.B. ( 2012) Collagen VI ablation retards brain tumor progression due to deficits in assembly of the vascular basal lamina. Am. J. Pathol . 180, 1145– 1158 Google Scholar CrossRef Search ADS PubMed  30 Ruhl M., Sahin E., Johannsen M., Somasundaram R., Manski D., Riecken E.O., Schuppan D. ( 1999) Soluble collagen VI drives serum-starved fibroblasts through S phase and prevents apoptosis via down-regulation of bax. J. Biol. Chem.  274, 34361– 34368 Google Scholar CrossRef Search ADS PubMed  31 Howell S.J., Doane K.J. ( 1998) Type VI collagen increases cell survival and prevents anti-β1 integrin-mediated apoptosis. Exp. Cell Res . 241, 230– 241 Google Scholar CrossRef Search ADS PubMed  32 Burg M.A., Tillet E., Timpl R., Stallcup W.B. ( 1996) Binding of the NG2 proteoglycan to type VI collagen and other extracellular matrix molecules. J. Biol. Chem.  271, 26110– 26116 Google Scholar CrossRef Search ADS PubMed  33 Tulla M., Pentikainen O.T., Viitasalo T., Kapyla J., Impola U., Nykvist P., Nissinen L., Johnson M.S., Heino J. ( 2001) Selective binding of collagen subtypes by integrin alpha 1I, alpha 2I, and alpha 10I domains. J. Biol. Chem.  276, 48206– 48212 Google Scholar CrossRef Search ADS PubMed  34 Tillet E., Gential B., Garrone R., Stallcup W.B. ( 2002) NG2 proteoglycan mediates β1 integrin-independent cell adhesion and spreading on collagen VI. J. Cell. Biochem.  86, 726– 736 Google Scholar CrossRef Search ADS PubMed  35 Zigrino P., Drescher C., Mauch C. ( 2001) Collagen-induced proMMP-2 activation by MT1-MMP in human dermal fibroblasts and the possible role of α2β1 integrins. Eur. J. Cell Biol . 80, 68– 77 Google Scholar CrossRef Search ADS PubMed  36 Hocking D.C., Sottile J., Langenbach K.J. ( 2000) Stimulation of integrin-mediated cell contractility by fibronectin polymerization. J. Biol. Chem.  275, 10673– 10682 Google Scholar CrossRef Search ADS PubMed  37 Endo T. ( 2015) Glycobiology of alpha-dystroglycan and muscular dystrophy. J. Biochem.  157, 1– 12 Google Scholar CrossRef Search ADS PubMed  38 Aigner T., Hambach L., Söder S., Schlötzer-Schrehardt U., Pöschl E. ( 2002) The C5 domain of Col6A3 is cleaved off from the Col6 fibrils immediately after secretion. Biochem. Biophys. Res. Commun . 290, 743– 748 Google Scholar CrossRef Search ADS PubMed  39 Park J., Scherer P.E. ( 2012) Adipocyte-derived endotrophin promotes malignant tumor progression. J. Clin. Invest.  122, 4243– 4256 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations ABA Agaricus bisporus agglutinin AU arbitrary unit CBB Coomassie Brilliant Blue FBS fetal bovine serum Gal β-galactose IB immunoblotting IP immuno-precipitation LP lectin-precipitation LCA Lens culinaris agglutinin 2-ME 2-mercaptoethanol MW molecular weight GlcNAc N-acetyl-D-glucosamine GlcNAc-β-1,4-GIcNAc N,N' diacetyl-chitobiose NTH non-triple helical PBS phosphate-buffered saline PVDF polyvinylidene difluoride RCA120 Ricinus communis agglutinin I NeuAc sialic acid TFA trifluoroacetic acid WGA wheat germ agglutinin © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Biochemistry Oxford University Press

Type VI collagen α1 chain polypeptide in non-triple helical form is an alternative gene product of COL6A1

Loading next page...
 
/lp/ou_press/type-vi-collagen-1-chain-polypeptide-in-non-triple-helical-form-is-an-ebYqfJ2ASc
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
ISSN
0021-924X
eISSN
1756-2651
D.O.I.
10.1093/jb/mvy040
Publisher site
See Article on Publisher Site

Abstract

Abstract Expression of type IV collagen α1 chain in non-triple helical form, NTH α1(IV), is observed in cultured human cells, human placenta and rabbit tissues. Biological functions of NTH α1(IV) are most likely to be distinct from type IV collagen, since their biochemical characteristics are quite different. To explore the biological functions of NTH α1(IV), we prepared some anti-NTH α1(IV) antibodies. In the course of characterization of these antibodies, one antibody, #141, bound to a polypeptide of 140 kDa in size in addition to NTH α1(IV). In this study, we show evidence that the 140 kDa polypeptide is a novel non-triple helical polypeptide of type VI collagen α1 chain encoded by COL6A1, or NTH α1(VI). Expression of NTH α1(VI) is observed in supernatants of several human cancer cell lines, suggesting that the NTH α1(VI) might be involved in tumourigenesis. Reactivity with lectins indicates that sugar chains of NTH α1(VI) are different from those of the α1(VI) chain in triple helical form of type VI collagen, suggesting a synthetic mechanism and a mode of action of NTH α1(VI) is different from type VI collagen. cancer, non-triple helical α1(IV) chain, non-triple helical α1(VI) chain, type IV collagen, type VI collagen Presence of type IV collagen α1 [α1(IV)] chain in non-triple helical form in cultured human cells (1), human placenta (2) and rabbit tissues (3) is reported and that is designated as NTH α1(IV) (3, 4). Cultured human cells secrete NTH α1(IV) under physiological conditions and the production of NTH α1(IV) depends on ascorbate. Under ascorbate-depleted condition, cells secrete NTH α1(IV) whose prolyl and lysyl residues have lower hydroxylation levels than type IV collagen or α1(IV) chain in triple helical form (4, 5). In addition, NTH α1(IV), but not the α1(IV) chain from the triple helical molecules, has specific sugar chains that are detectable by lectin Agaricus bisporus agglutinin (ABA) (2, 6). Namely, NTH α1(IV) has different chemical structure compared with the α1(IV) chain in triple helical form, suggesting that the biological function of NTH α1(IV) as well as regulation of synthesis and secretion is also different from that of the α1(IV) chain in triple helical form or type IV collagen. The NC1 domain derived from NTH α1(IV) shows tissue inhibitor of metalloproteinase-like activity against matrix metalloproteinase-9 (2). Sugiyama et al. reported that NTH α1(IV) expression is observed at a neovascular tip, where no type IV collagen is detected, in a rabbit angiogenic model, suggesting that physiological role of NTH α1(IV) is associated with angiogenesis or vascular system dynamics (3). Tumour cells produce different extracellular matrix or microenvironment presumably in favour of tumour formation and expansion; for example, peptides derived from type IV collagen promote the expansion and angiogenesis of tumour tissues (7). NTH α1(IV) expression is observed in various tumour cell cultures (M. Morita, manuscript in preparation). Although NTH α1(IV) could be involved in tumour angiogenesis, the biological function and the mode of action of NTH α1(IV) remain to be clarified. To explore the possible involvement of NTH α1(IV) in tumourigenesis, we have acquired some anti-NTH α1(IV) monoclonal antibodies that did not react with triple helical type IV collagen molecules (4). In the course of characterization of these antibodies, one of the antibodies, antibody #141, but not other monoclonal antibodies, reacted with a polypeptide of 140 kDa in size in addition to NTH α1(IV). We hypothesized that the 140 kDa polypeptide was a novel polypeptide that contains the epitope related to the NTH α1(IV). In this study, we characterized covalent structure of the 140 kDa polypeptide and identified that the 140 kDa polypeptide is a novel non-triple helical polypeptide of type VI collagen α1 [α1(VI)] chain encoded by COL6A1 named NTH α1(VI). We also showed that NTH α1(VI) expression was found in the supernatants of human liver cancer cell line HLF, human oesophageal cancer cell line TE8 and human lung cancer cell line NCI-H226. In addition, sugar chains of NTH α1(VI) were different from those of the α1(VI) chain in triple helical form from differential binding activities of lectins. Our results suggest that NTH α1(VI) is possibly involved in tumourigenesis and the mode of action and the biological function are distinct from those of the α1(VI) chain in triple helical form or type VI collagen. Here we report in detail. Materials and Methods Antibodies and lectins Anti-NTH α1(IV) monoclonal antibody #141 was prepared in our laboratory using purified NTH α1(IV) as an antigen (Nippon Kayaku) and partially characterized in our previous study (4). Anti-α1(IV) chain monoclonal antibody, JK132, was obtained using purified human collagen type IV as an antigen, as described previously (8). Anti-COL6A1 polyclonal antibody (N term) was purchased from Abgent (San Diego, CA). Horseradish peroxidase-labelled anti-mouse and anti-rabbit IgG antibodies were purchased from GE Healthcare (Little Chalfont, UK). Biotin-labelled lectin Lens culinaris agglutinin (LCA), Ricinus communis agglutinin I (RCA120) and wheat germ agglutinin (WGA) were purchased from J-OIL MILLS (Tokyo, Japan). LCA-conjugated agarose, RCA-conjugated agarose and WGA-conjugated agarose were purchased from J-OIL MILLS. Cell culture Human embryonic kidney cell line 293FT (Thermo Fisher Scientific, Waltham, MA) was cultivated in DMEM high glucose (Thermo Fisher Scientific) with 1 mM sodium pyruvate (Thermo Fisher Scientific), non-essential amino acids solution (Thermo Fisher Scientific) and 10% fetal bovine serum (FBS; Tissue Culture Biologicals, Tulare, CA). Human breast cancer cell line MCF-7 and human oesophageal cancer cell line TE8 purchased from Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University (Miyagi, Japan), human mesothelioma cell line MSTO-211H, human lung cancer cell line NCI-H226 and human lung cancer cell line NCI-H460 purchased from ATCC (Manassas, VA) and human liver cancer cell line HLF purchased from Japanese Collection of Research Bioresources Cell Bank (JCRB) (Tokyo, Japan) were cultivated in RPMI1640 (Thermo Fisher Scientific) with 10% FBS. Human fetal lung fibroblast TIG-1 cells were purchased from JCRB and cultured in DMEM (Thermo Fisher Scientific) with 10% FBS. All cell lines were cultured at 37°C in 5% CO2 atmosphere. Preparation of antibody-conjugated resins Antibody #141- or JK132-conjugated resin was prepared as follows. Two milligrams of antibody #141 or JK132 was added to 1 ml of Protein G Sepharose 4 Fast Flow (GE Healthcare) and each resin was incubated at 4°C for 1 h with gentle shaking. The resin was equilibrated with 0.2 M triethanolamine-HCl (pH 8.2) and incubated with 50 mM dimethyl pimelimidate at room temperature with gentle shaking to cross-link the resin and the antibody. After 1 h, the resin was equilibrated with 0.2 M ethanolamine-HCl (pH 8.2) and incubated at room temperature for 1 h with gentle shaking to block unreacted amines. After blocking, the resin was washed and equilibrated with phosphate-buffered saline (PBS; pH 7.4). Purification of the 140 kDa polypeptide by affinity chromatography on antibody #141-conjugated resin All purification procedures were performed at 4°C. The 140 kDa polypeptide was purified from conditioned medium of HLF cell culture as follows. One thousand and five hundred millilitres of the conditioned medium was passed through 6 ml of antibody #141-conjugated resin to adsorb polypeptide(s). After adsorption, the resin was washed with PBS and bound polypeptides including the 140 kDa polypeptide were eluted with 0.1 M glycine-HCl (pH 2.8). Finally, the 140 kDa polypeptide-containing fractions were collected, concentrated with Amicon Ultra-15 (Merck Millipore, Billerica, MA) and dialyzed against PBS. Extraction of human type VI collagen and its analysis with dot blotting The dermis-like structure formed by long-term (31 days) culture of human skin fibroblast, HF18, was used as starting material for type VI collagen preparation as reported previously (9). The lyophilized material (1.10 g) of the dermis-like structure was suspended in 5 ml of 50 mM Tris–HCl (pH 7.5) and 1 M NaCl with a protease inhibitor cocktail comprising 5 mM EDTA, 100 µM N-ethylmaleimide and 100 µM phenylmethylsulfonyl fluoride overnight. Isolation procedure was performed at 4°C. The supernatant was collected by centrifugation at 10,000 ×g for 30 min. The total protein concentration of the collected supernatant was estimated to be 0.1 mg/ml. SDS-PAGE analysis using CBB staining revealed that type VI, type I and type V collagens became visible after concentrating with Amicon Ultra-0.5 (Merck Millipore) by 10 times. Dot blotting was carried out as follows. Samples containing type VI collagen were subjected to heat denaturation at 90°C for 10 min with or without 0.02% 2-mercaptoethanol (2-ME) (denoted in Fig. 5A as Heated and reducing and Heated and non-reducing, respectively). Non-heated samples with or without 2-ME were also prepared. Samples containing indicated amount of type VI collagen were diluted with TBS (25 mM Tris–HCl, pH 7.5, 137 mM NaCl, 2.7 mM KCl) and blotted onto polyvinylidene difluoride (PVDF) membrane (Immobilon-P, Merck Millipore) using the filtration manifold system MilliBlot system (Merck Millipore). Each blot hole was washed with TBS once and the membrane was blocked with 1/4 dilution of Block Ace (DS Pharma Biochemical, Osaka, Japan). The membrane was incubated with antibody #141 and washed with Tris-buffered saline-Tween20 (20 mM Tris–HCl, pH 7.4, 150 mM NaCl, 0.05% Tween20). The bound antibodies were detected with horseradish peroxidase-labelled anti-mouse IgG antibody and visualized using Immobilon Western Chemiluminescent HRP Substrate (Merck Millipore) on an image analyser (ImageQuant LAS4000; GE Healthcare). Amino acid sequence analysis of the 140 kDa polypeptide Amino acid sequence analysis of the 140 kDa polypeptide was carried out in APRO Life Science Institute (Tokushima, Japan) as follows. An equal amount of 2× concentrated sample buffer with 2-ME (125 mM Tris–HCl, pH 6.8, 20% glycerol, 10% 2-ME, 4% SDS) was added to the purified 140 kDa polypeptide solution. The mixture was heated at 95°C for 5 min, loaded onto 7.5% SDS-PAGE gel and electrophoresed. After electrophoresis, the band of 140 kDa polypeptide was cut out from the gel and the gel was digested with endopeptidase Lys-C (Wako Pure Chemical Industries, Osaka, Japan) at 35°C for 20 h. Digested peptides were injected into an Aliance HPLC 2695 Separations Module System (Waters, Milford, MA) and separations were performed using a Symmetry C18 column (100 Å, 3.5 μm, 1.0 × 150 mm, Waters). The mobile phases consisted of A: 0.10% trifluoroacetic acid (TFA)-2% acetonitrile and B: 0.09% TFA-90% acetonitrile. Gradient conditions were 0% of mobile phase B for 6 min, 0–10% of mobile phase B in 5 min, 10–50% of mobile phase B in 75 min, 50–100% of mobile phase B in 5 min and 100% of mobile phase B for 5 min with a flow rate of 1 ml/min. Separated peptides were collected and analysed by a Procise 494 HT Protein Sequencing System (Thermo Fisher Scientific) according to the manufacturer’s instruction. Of note, peptides were detected at a UV wavelength of 210 nm and the reference wavelength was set at 280 nm. Preparation of mammalian expression vectors and transient expressions of those vectors in 293FT cells COL4A1 and COL6A1 coding regions were amplified from HLF cDNA by PCR with primers as described in Table I and each of coding regions was subcloned into pENTR vector (Thermo Fisher Scientific). For mammalian expression, COL4A1 or COL6A1 coding region was transferred to pcDNA3.2/V5-DEST vector (Thermo Fisher Scientific) by GATEWAY technology (Thermo Fisher Scientific) according to the manufacturer’s instruction. Table I. Primers used in this study Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  Table I. Primers used in this study Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  Name  Sequence  COL4A1 Fwd  5′-GTCGACGCCACCATGGGGCCCCGGCTCAGCGTCTGGCTGCTG-3′  COL4A1 Rev  5′-GCGGCCGCTTATGTTCTTCTCATACAGACTTGGCAGCG-3′  COL6A1 Fwd  5′-GTCGACGCCACCATGAGGGCGGCCCGTGCTCTGCTGCCCCTG-3′  COL6A1 Rev  5′-GATATCTCTAGCCCAGCGCCACCTTCCTGGAGACTGT-3′  For transient expression, we transfected the indicated mammalian expression vector to 293FT cells with Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer’s instruction. After transfection, 293FT cells were cultivated at 37°C in 5% CO2 atmosphere for 3 days and the conditioned media were subjected to western blotting analysis. Immuno- or lectin precipitation Immuno- or lectin precipitation was carried out as described previously with slight modifications (4). Briefly, 10 µl of the indicated antibody-conjugated resin or lectin-conjugated agarose was added to 1 ml of the indicated conditioned medium and incubated at 4°C for 1 h followed by centrifugation. The resin and the agarose were washed with wash buffer (50 mM Tris–HCl, pH 7.4, 300 mM NaCl, 0.1% Triton X-100). After washing, precipitates were eluted with 2× concentrated sample buffer with 2-ME at 95°C for 5 min. Precipitates were subjected to western blotting analysis. Western blotting and CBB stain Western blotting was carried out as described previously with slight modifications (4). Briefly, an equal amount of 2× concentrated sample buffer with 2-ME was added to a sample and boiled at 95°C for 5 min. Of note, under non-reducing conditions, 2× concentrated sample buffer without 2-ME was used. The mixture was loaded onto 7.5% SDS-PAGE gel or 5–20% SuperSep Ace precast gel (Wako Pure Chemical Industries) and electrophoresed. The gel was transferred to PVDF membrane and blocked with Tris-buffered saline-Tween20 containing 5% (weight/volume) skim milk (BD, Franklin Lakes, NJ). The membrane was incubated with pre-determined antibodies or biotin-labelled lectins and washed with Tris-buffered saline-Tween20. The bound antibodies were detected with horseradish peroxidase-labelled anti-mouse or anti-rabbit IgG antibody. Of note, in the case of using a biotin-labelled lectin as a probe, Tris-buffered saline-Tween20 and horseradish peroxidase-conjugated streptavidin (GE Healthcare) were used for blocking and detection according to the manufacturer’s instruction. The membrane was developed with ECL (GE Healthcare) or ECL Prime (GE Healthcare) on a Hyperfilm ECL (GE Healthcare). In some experiments, we used the same membrane for detection with different antibodies. To reuse the membrane, the membrane was incubated at 55°C for 30 min in stripping buffer (62.5 mM Tris–HCl, pH 6.5, 100 mM 2-ME, 2% SDS) for stripping antibodies or lectins off and reprobed with the indicated antibody as described previously (10). Then, the membrane was subjected to western blotting. CBB stain was carried out with Rapid Stain CBB Kit (Nacalai Tesque, Kyoto, Japan) according to the manufacturer’s instruction. Results Antibody #141 has a higher affinity for 140 kDa polypeptide than that for NTH α1(IV) In the course of characterization of antibody #141, we found that antibody #141 detected both NTH α1(IV) and a band at 140 kDa (hereafter 140K) in supernatants of several cancer cell lines, human liver cancer cell line HLF, human oesophageal cancer cell line TE8 and human lung cancer cell line NCI-H226, under heated and non-reducing conditions (Fig. 1A), indicating that the 140K forms a non-disulphide bonded structure. Figure 1A also shows that JK132 bound to NTH α1(IV) (180 kDa) and two extra bands at 135 and 160 kDa. Iwata et al. reported the existence of 160 kDa of a short isoform of α1(IV) chain in addition to 180 kDa of α1(IV) chain. However, the band at 135 kDa (hereafter 135K) has not been reported (11). The 135K is slightly different from 140K in size, suggesting that the 140K and 135K are unknown isoforms of NTH α1(IV). Next, we carried out immuno-precipitation experiment. Figure 1B shows that antibody #141 could precipitate the 140K but not the 135K and JK132 could precipitate the 135K but not 140K, vice versa. Moreover, antibody #141 reactivity to the 140K is higher than the reactivity to NTH α1(IV) from the finding that antibody #141 precipitated larger amounts of the 140K than NTH α1(IV) (Fig. 1B). JK132 is reported to recognize the amino acid sequences contained in the region of 1165–1179 in human α1(IV) chain (1, 11) and antibody #141 was obtained in our laboratory using NTH α1(IV) as an antigen (4). Therefore, we hypothesized that the 140K and the 135K were novel isoforms of NTH α1(IV) and the differences of the affinities to the 140K between antibody #141 and JK132 were caused by the differences of recognition sequences between the two antibodies. Fig. 1 View largeDownload slide Recognition of a band at 140 kDa by antibody #141. (A) Representative image of a band at 140 kDa detected by antibody #141. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The conditioned medium was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with JK132 (IB: JK132) as described in Materials and Methods section. After JK132 detection, the membrane was stripped with the stripping buffer to remove JK132. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. (B) Representative image of precipitates of JK132 and antibody #141. Proteins in the conditioned medium of HLF cell culture were immuno-precipitated with JK132- or antibody #141-conjugated resin (IP JK132 or IP #141 in the figure). The precipitates were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions) and electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with antibody #141 (IB: #141) as described in Materials and Methods section. NTH α1(IV), the band at 135 kDa detected by JK132 (hereafter 135K), and the band at 140 kDa detected by antibody #141 (hereafter 140K) are indicated by filled arrowheads, an arrow, and an open arrowhead, respectively. Fig. 1 View largeDownload slide Recognition of a band at 140 kDa by antibody #141. (A) Representative image of a band at 140 kDa detected by antibody #141. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The conditioned medium was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with JK132 (IB: JK132) as described in Materials and Methods section. After JK132 detection, the membrane was stripped with the stripping buffer to remove JK132. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. (B) Representative image of precipitates of JK132 and antibody #141. Proteins in the conditioned medium of HLF cell culture were immuno-precipitated with JK132- or antibody #141-conjugated resin (IP JK132 or IP #141 in the figure). The precipitates were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions) and electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out by probing with antibody #141 (IB: #141) as described in Materials and Methods section. NTH α1(IV), the band at 135 kDa detected by JK132 (hereafter 135K), and the band at 140 kDa detected by antibody #141 (hereafter 140K) are indicated by filled arrowheads, an arrow, and an open arrowhead, respectively. Identification of the 140K as a polypeptide of α1(VI) chain To confirm the hypothesis, we tried to identify the 140K. The 140K was purified from the conditioned medium of HLF cell culture as described in Materials and Methods section. CBB staining shows that the purified sample contained the 140K as a major polypeptide with smaller polypeptides in less amounts (Fig. 2A). The 140K and one of the smaller polypeptides were immuno-stained with antibody #141 as shown in Fig. 2B. The purity of 140K was adequate for amino acid sequence analysis, but the N-terminal amino acid identification of the 140K failed. The failure of amino acid sequence analysis might be caused by some post translational modifications or formation of pyroglutamate at its N-terminal. Fig. 2 View largeDownload slide Evaluation of the 140K purity and antibody #141 reactivity. (A) Representative image of purified 140K. The 140K was purified from the conditioned medium of HLF cell culture as described in Materials and Methods section. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The 140K was electrophoresed onto 7.5% SDS-PAGE gel and analysed by CBB staining as described in Materials and Methods section. The indicated amount of BSA was loaded for estimation of the 140K quantity. (B) Antibody #141 reactivity toward purified 140K. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The mixture was electrophoresed and western blotting was carried out by probing with antibody #141(IB: #141) as described in Materials and Methods section. After the antibody #141 detection, the bound antibody was stripped off from the membrane and stained with CBB as described in Materials and Methods section. The 140K is shown by an open arrowhead. Fig. 2 View largeDownload slide Evaluation of the 140K purity and antibody #141 reactivity. (A) Representative image of purified 140K. The 140K was purified from the conditioned medium of HLF cell culture as described in Materials and Methods section. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The 140K was electrophoresed onto 7.5% SDS-PAGE gel and analysed by CBB staining as described in Materials and Methods section. The indicated amount of BSA was loaded for estimation of the 140K quantity. (B) Antibody #141 reactivity toward purified 140K. Ten microlitres of purified 140K was mixed with an equal amount of 2× concentrated sample buffer without 2-ME and the mixture was boiled at 95°C (heated and non-reducing conditions). The mixture was electrophoresed and western blotting was carried out by probing with antibody #141(IB: #141) as described in Materials and Methods section. After the antibody #141 detection, the bound antibody was stripped off from the membrane and stained with CBB as described in Materials and Methods section. The 140K is shown by an open arrowhead. Because the N-terminal of the 140K could not be analysed, internal peptide fragments of the 140K were generated by endopeptidase Lys-C digestion and N-terminal amino acid sequence analysis of the digested peptides was carried out. In brief, the digested peptides were separated by reverse-phased HPLC and the fractions of number 30, 40, 47 and 58 as shown in Fig. 3 were subjected to amino acid sequence analysis. Table II indicates that all the peptide fragments except for fraction number 40 were identified as those from COL6A1 (RefSeq accession number: NP_001839.2). Of note, fraction number 58 gave two sequences both of which matched sequences within COL6A1 and fraction number 40 could not be identified for unknown reasons. As all the peptides analysed could be ascribed to COL6A1 sequences, the 140K is likely a polypeptide of α1(VI) chain encoded by COL6A1. Table II. Amino acid sequences of the four peptides derived from purified 140K Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Table II. Amino acid sequences of the four peptides derived from purified 140K Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Fr. No.  Peptide sequence  Identified protein  Amino acid position  30  VIDRLSRDELVK  COL6A1  641–652  47  GLEQLLVGGS  COL6A1  136–145  58  TAEYDVAYGE  COL6A1  991–1000  58  DVFDFIPGSD  COL6A1  758–767  Fig. 3 View largeDownload slide Elution profile of peptides derived from endopeptidase Lys-C digested 140K. Purified 140K was digested with endopeptidase Lys-C and the digested peptides were separated by reverse-phased HPLC. The indicated fractions (marked by arrows with fraction numbers) were subjected to amino acid sequence analysis as described in Materials and Methods section. Blue line: absorbance at 210 nm for peptide detection. Red line: absorbance at 280 nm for reference. AU: arbitrary unit. Fig. 3 View largeDownload slide Elution profile of peptides derived from endopeptidase Lys-C digested 140K. Purified 140K was digested with endopeptidase Lys-C and the digested peptides were separated by reverse-phased HPLC. The indicated fractions (marked by arrows with fraction numbers) were subjected to amino acid sequence analysis as described in Materials and Methods section. Blue line: absorbance at 210 nm for peptide detection. Red line: absorbance at 280 nm for reference. AU: arbitrary unit. Although antibody #141 was obtained using NTH α1(IV) as an antigen (4), the amino acid sequence analysis indicated that the 140K is derived from COL6A1. To confirm the result of the amino acid sequence analysis, we transfected COL4A1 or COL6A1 expression vector to 293FT cells and determined whether antibody #141 recognized COL6A1. In addition, we checked that JK132 did not detect COL6A1, since JK132 recognizes the sequence contained in the region of 1165–1179 (KGEPGLPGRGFPGFPT) in COL4A1 (1, 11) and this sequence is not conserved in COL6A1. Figure 4 shows that antibody #141 recognizes both COL4A1 and COL6A1 under heated and reducing conditions, whereas JK132 recognizes only COL4A1 but not COL6A1. This result confirms that the 140K is a polypeptide of the α1(VI) chain in size. Fig. 4 View largeDownload slide Analysis of antibody #141 reactivity toward COL6A1. COL4A1 or COL6A1 expression vector was introduced into 293FT cells with Lipofectamine 2000 according to the manufacture’s instruction and the conditioned medium was harvested. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer with 2-ME and the mixture was boiled at 95°C (heated and reducing conditions). The mixture was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out as described in Materials and Methods section. Of note, the membrane was sequentially probed with the indicated antibody as follows; antibody #141 (IB: #141), JK132 (IB: JK132) and anti-COL6A1 polyclonal antibody (N-term) (IB: αCOL6A). After the detection, the first antibody was removed from the membrane. The membrane was reprobed with the indicated antibody and western blotting was carried out as described in Materials and Methods section. COL4A1 and COL6A1 are shown by filled arrowheads and open arrowheads, respectively. Fig. 4 View largeDownload slide Analysis of antibody #141 reactivity toward COL6A1. COL4A1 or COL6A1 expression vector was introduced into 293FT cells with Lipofectamine 2000 according to the manufacture’s instruction and the conditioned medium was harvested. Ten microlitres of the indicated conditioned medium was mixed with an equal amount of 2× concentrated sample buffer with 2-ME and the mixture was boiled at 95°C (heated and reducing conditions). The mixture was electrophoresed onto 7.5% SDS-PAGE gel. After electrophoresis, the gel was transferred to PVDF membrane and western blotting was carried out as described in Materials and Methods section. Of note, the membrane was sequentially probed with the indicated antibody as follows; antibody #141 (IB: #141), JK132 (IB: JK132) and anti-COL6A1 polyclonal antibody (N-term) (IB: αCOL6A). After the detection, the first antibody was removed from the membrane. The membrane was reprobed with the indicated antibody and western blotting was carried out as described in Materials and Methods section. COL4A1 and COL6A1 are shown by filled arrowheads and open arrowheads, respectively. Fig. 5 View largeDownload slide Analysis of antibody #141 specificity. (A) Representative image of antibody #141 reactivity toward human type VI collagen. The indicated amount of human type VI collagen was blotted onto PVDF membrane. The membrane was probed with antibody #141 (IB: #141) and visualized as described in Materials and Methods section. (B) Confirmation of antibody #141 reactivity toward human α1(VI) chain under heated and reducing conditions. The partially purified type VI collagen sample was analysed by western blotting with antibody #141. The sample contained only the α1(VI) chain derived from type VI collagen as an antibody #141-reactive antigen. Of note, the conditioned medium of TIG-1 cell culture in the ascorbate-depleted medium was used as a positive control. The α1(IV) chain and the α1(VI) chain are shown by a filled arrowhead and an open arrowhead, respectively. HuVI: human type VI collagen. Fig. 5 View largeDownload slide Analysis of antibody #141 specificity. (A) Representative image of antibody #141 reactivity toward human type VI collagen. The indicated amount of human type VI collagen was blotted onto PVDF membrane. The membrane was probed with antibody #141 (IB: #141) and visualized as described in Materials and Methods section. (B) Confirmation of antibody #141 reactivity toward human α1(VI) chain under heated and reducing conditions. The partially purified type VI collagen sample was analysed by western blotting with antibody #141. The sample contained only the α1(VI) chain derived from type VI collagen as an antibody #141-reactive antigen. Of note, the conditioned medium of TIG-1 cell culture in the ascorbate-depleted medium was used as a positive control. The α1(IV) chain and the α1(VI) chain are shown by a filled arrowhead and an open arrowhead, respectively. HuVI: human type VI collagen. Conformation-dependent recognition of the α1(VI) chain with antibody #141 As shown above, antibody #141 recognizes a non-disulphide-bonded polypeptide of the α1(VI) chain as well as NTH αl(IV) (Fig. 1). The reactivity of this antibody to the α1(VI) chain could also be dependent on the conformation of type VI collagen. Therefore, we examined whether heat denaturation of triple helical type VI collagen molecule gave rise to the reactivity to antibody #141 by dot blot analysis, using partially purified human type VI collagen. Figure 5A indicates that antibody #141 bound to the α1(VI) chain derived from type VI collagen after heat denaturation under reducing or non-reducing condition. Positive reaction to the α1(VI) chain in native structure under non-heated and non-reducing conditions was negligibly small. Furthermore, antibody #141 hardly recognized the α1(VI) chain under non-heated and reducing conditions. The western blotting analysis using antibody #141 in Fig. 5B showed that the sample contained mainly one band at 140K corresponding for the α1(VI) chain probably derived from type VI collagen under heated and reducing condition. Since there were no visible band at 180 kDa of α1(IV) chain nor NTH α1(IV), the sample contained only the α1(VI) chain as an antibody #141-reactive antigen. These observations indicate that antibody #141 recognizes the α1(VI) chain in a conformation-dependent manner, suggesting that the specificity of antibody #141 to the 140K is caused by its conformation. Therefore, it is considered that the 140K is a novel non-triple helical polypeptide encoded by COL6A1 and we name it NTH α1(VI). Since NTH α1(VI) was detected in culture supernatant without reducing agents (Fig. 1A), it is not disulphide-bonded with α2(VI) or α3(VI) chain. Characterization of sugar chains of NTH α1(VI) To characterize post-translational modifications of NTH α1(VI), we examined the presence and types of sugar chains on NTH α1(VI) with several lectins. Figure 6A shows that LCA precipitated NTH α1(VI) s derived from HT1080 and Lu65A cells, whereas RCA120 reacted with the NTH α1(VI) recovered from HT1080 but not the NTH α1(VI) from Lu65A cells and WGA did not precipitate either NTH α1(VI). The reaction with RCA120 implies that NTH α1(VI) is differentially glycosylated between these cell lines. The three lectins recognize different moieties of sugar chains of the glycoprotein. (i) LCA binds to a core fucose of N-linked glycan (12, 13), (ii) RCA120 binds to a non-reducing terminal β-galactose (Gal) (14). (iii) WGA binds to a non-reducing terminal N-acetyl-d-glucosamine (GlcNAc) and its oligomer, such as N, N' diacetyl-chitobiose (GlcNAc-β-1, 4-GIcNAc), and a non-reducing terminal sialic acid (NeuAc) (15–17). Therefore, common characteristics of sugar chains of NTH α1(VI) are as follows; core-fucose structure and without GlcNAc and NeuAc at non-reducing terminal. To characterize the glycosylation status further, we examined lectin reactivity toward human type VI collagen. Figure 6B shows that LCA, RCA120 and WGA recognized human type VI collagen under heated and reducing conditions. Fujiwara et al. reported that sugar chains of type VI collagen have N-glycans with core-fucose structure and Gal or NeuAc, but not GlcNAc at non-reducing terminal (18). This report coincides with our result as shown in Fig. 6B, indicating that sugar chains of type VI collagen have core-fucose structure and Gal or NeuAc at non-reducing terminal. In contrast, Fig. 6A shows that the sugar chains of NTH α1(VI) do not have NeuAc at non-reducing terminal by WGA reactivity, suggesting that the sugar chains are different between NTH α1(VI) and the α1(VI) chain in triple helical form. Fig. 6 View largeDownload slide Characterization of sugar chains of NTH α1(VI) and type VI collagen. (A) Characterization of sugar chains of NTH α1(VI). Sugar chains of NTH α1(VI) from the conditioned medium of HT1080 and Lu65A cells were analysed using the indicated lectins. Briefly, 10 µl of the indicated lectin-conjugated agarose was added to 1 ml of conditioned medium from HT1080 or Lu65A cell culture. Precipitates including NTH α1(VI) were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions). The precipitates were separated onto 7.5% SDS-PAGE gel and NTH α1(VI) was detected with antibody #141 (IB: #141) as described in Materials and Methods section. (B) Characterization of sugar chains of type VI collagen. Ten microlitres of partially purified type VI collagen was denatured under heated and reducing conditions as similar to (a). Type VI collagen was separated onto 5–20% SuperSep Ace precast gel and sugar chains of type VI collagen were analysed using the indicated lectins as described in Materials and Methods section. After the analysis, the lectins bound on blotting membrane were removed. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. NTH α1(VI) in (A) and the α1(VI) chain in (B) are shown by open arrowheads. LCA, lectin Lens culinaris agglutinin; RCA120, lectin Ricinus communis agglutinin I; WGA, lectin wheat germ agglutinin; LP, lectin precipitation. Fig. 6 View largeDownload slide Characterization of sugar chains of NTH α1(VI) and type VI collagen. (A) Characterization of sugar chains of NTH α1(VI). Sugar chains of NTH α1(VI) from the conditioned medium of HT1080 and Lu65A cells were analysed using the indicated lectins. Briefly, 10 µl of the indicated lectin-conjugated agarose was added to 1 ml of conditioned medium from HT1080 or Lu65A cell culture. Precipitates including NTH α1(VI) were eluted with 2× concentrated sample buffer containing 2-ME at 95°C (heated and reducing conditions). The precipitates were separated onto 7.5% SDS-PAGE gel and NTH α1(VI) was detected with antibody #141 (IB: #141) as described in Materials and Methods section. (B) Characterization of sugar chains of type VI collagen. Ten microlitres of partially purified type VI collagen was denatured under heated and reducing conditions as similar to (a). Type VI collagen was separated onto 5–20% SuperSep Ace precast gel and sugar chains of type VI collagen were analysed using the indicated lectins as described in Materials and Methods section. After the analysis, the lectins bound on blotting membrane were removed. The membrane was reprobed with antibody #141 (IB: #141) and western blotting was carried out as described in Materials and Methods section. NTH α1(VI) in (A) and the α1(VI) chain in (B) are shown by open arrowheads. LCA, lectin Lens culinaris agglutinin; RCA120, lectin Ricinus communis agglutinin I; WGA, lectin wheat germ agglutinin; LP, lectin precipitation. Discussion In this study, we show that a novel non-triple helical collagen polypeptide NTH α1(VI) exists in several cancer cell lines (Figs 1A and 5). Furthermore, our results indicate that NTH α1(VI) is an alternative gene product of COL6A1 (Table II and Fig. 4). Another non-triple helical collagen polypeptide NTH α1(IV) is also reported (1–4). NTH α1(IV) is an alternative gene product of COL4A1 and shows a non-disulphide bonded structure (1). In addition, NTH α1(IV) is specifically recognized by anti-α1(IV) chain antibody JK132 (8), because an epitope of JK132 is hidden in triple helical conformation (1, 11), indicating that a higher order structure of NTH α1(IV) is different from the α1(IV) chain in triple helical form. Antibody #141 recognizes only non-triple helical form of α1(VI) chain (Fig. 5). Our result is similar to the reactivity of JK132 to NTH α1(IV), indicating that conformation of NTH α1(VI) is different from that of the α1(VI) chain in triple helical form. We show that antibody #141 recognizes NTH α1(IV) and NTH α1(VI) (Fig. 1). To clarify the specific reactivity of antibody #141 toward both polypeptides, we have been investigating the recognition sequence of antibody #141. We found that antibody #141 recognizes the region between 1055 and 1064 amino acid in COL4A1 and the region between 254 and 286 amino acid in COL6A1, suggesting that antibody #141 recognizes a common epitope in both molecules (T. Sato et al., manuscript in preparation). Therefore, it is considered that the specificity of antibody #141 toward NTH α1(IV) and NTH α1(VI) results from the recognition sequence. NTH α1(VI) has core-fucose structure as similar to type VI collagen (18), but the sugar chains of NTH α1(VI) do not have NeuAc at non-reducing terminal (Fig. 6). N-linked glycans are synthesized in endoplasmic reticulum through Golgi and have a common core pentasaccharide and multiple glycosylations are formed on the common core structure catalyzed by various enzymes (19–23). NeuAc is transferred to non-reducing terminal by the actions of sialyltransferases. It is also known that enzymatic specificities toward substrates are varied among sialyltransferases (22, 23). Differences of the non-reducing terminals between NTH α1(VI) and the α1(VI) chain may be caused by affinities of sialyltransferases to NTH α1(VI), since NTH α1(VI) has a non-triple helical structure (Fig. 5). However, structures of sugar chains attached to NTH α1(VI) are not fully determined yet. Further studies are needed. Type VI collagen is synthesized as a triple helical form (24–26). An α3(VI) chain is required for proper folding of type VI collagen. It has been discussed that in the absence of α3(VI) chain, α1(VI) and α2(VI) chains are retained within cells as unassembled polypeptides and degraded (27). NTH α1(VI) expression in supernatants of several cancer cell lines (Fig. 1A) raises a query whether or not α3(VI) chain is dispensable for NTH α1(VI) synthesis. The glycosylation status of NTH α1(VI) is different from type VI collagen (Fig. 6), suggesting that NTH α1(VI) is synthesized and secreted via different mechanism from the α1(VI) chain in triple helical form. It is reported that type VI collagen is involved in tumourigenesis, such as cancer cell growth (28, 29) and apoptosis (30, 31), and the mode of action of type VI collagen is mediated via interaction with β1 integrins or NG2/chondroitin sulfate proteoglycan (27, 32–34). As indicated, NTH α1(VI) expression is observed in several cancer cell lines (Fig. 1A), suggesting that NTH α1(VI) is also involved in tumourigenesis. The higher-order structure and the glycosylation status of NTH α1(VI) are different from type VI collagen (Figs 5 and 6). It is reported that higher-order structure of protein (e.g. type I collagen fibril and polymerized fibronectin) and glycosylation status (e.g. glycosylation defect in muscular dystrophy) influence biological functions (35–37), suggesting that the mode of action and the biological function of NTH α1(VI) are distinct from type VI collagen. Park et al. also reported that endotrophin (38), a cleaved product of COL6A3, promotes breast cancer growth, metastasis, epithelial mesenchymal transition and tumour angiogenesis (39). Because NTH α1(VI) is encoded by COL6A1 and the amino acid sequence of endotrophin is not conserved in COL6A1, the mode of action of NTH α1(VI) is also different from endotrophin. The mode of action and the biological function of NTH α1(VI) remain to be clarified and those are under investigation in our laboratory. In conclusion, we report a novel NTH form of collagen gene products as NTH α1(VI) encoded by COL6A1. Expression of NTH α1(VI) is observed in supernatants of several human cancer cell lines, suggesting that NTH α1(VI) is involved in tumourigenesis. Reactivity with lectins indicates that sugar chains of NTH α1(VI) are different from those of the α1(VI) chain in triple helical form, suggesting that the mechanism of synthesis, the mode of action and the biological function of NTH α1(VI) are different from those of the α1(VI) chain in triple helical form or type VI collagen as in the case of NTH α1(IV). Acknowledegement We thank Ms Chieko Seno for experimental support and Dr Keiko Sekine for technical advice. Funding This work was supported in part by a grant from the Strategic Research Foundation Grant-aided Project for Private Universities from the Ministry of Education, Culture, Sport, Science, and Technology, Japan, 2014-2018 [S1411005 to Y.I.]. Conflict of Interest None declared. References 1 Takahashi S., Yoshikawa K., Sasaki T., Takeda Y., Imamura Y., Sado Y., Hayashi T. ( 1999) Serum-dependent secretion of nondisulfide-bonded and unfolded type IV collagen α chains by cultured fetal lung fibroblasts. Connect. Tissue . 31, 161– 168 2 Kajimura D., Takahashi S., Yoshikawa K., Hattori S., Sado Y., Imamura Y., Hayashi T. ( 2004) Non-helical type IV collagen polypeptides in human placenta. Biochem. Biophys. Res. Commun  314, 11– 16 Google Scholar CrossRef Search ADS PubMed  3 Sugiyama H., Tokunaka K., Hayashi T., Imamura Y., Morita M., Yamato M. ( 2015) Non-triple helical form of type IV collagen α1 chain. Heliyon  1, e00051 Google Scholar CrossRef Search ADS PubMed  4 Morita M., Sugihara H., Tokunaka K., Tomura A., Saiga K., Sato T., Imamura Y., Hayashi T. ( 2017) Preparation and partial characterization of monoclonal antibodies specific for the nascent non-triple helical form of the type IV collagen alpha 1 chain. Biochem. Biophys. Rep  9, 128– 132 Google Scholar PubMed  5 Yoshikawa K., Takahashi S., Imamura Y., Sado Y., Hayashi T. ( 2001) Secretion of Non-helical collagenous pblypeptides of α1(IV) and α2(IV) chains upon depletion of ascorbate by cultured human cells. J. Biochem . 129, 929– 936 Google Scholar CrossRef Search ADS PubMed  6 Presant C.A., Kornfeld S. ( 1972) Characterization of the cell surface receptor for the Agaricus bisporus hemagglutinin. J. Biol. Chem . 247, 6937– 6945 Google Scholar PubMed  7 Lu P., Weaver V.M., Werb Z. ( 2012) The extracellular matrix: a dynamic niche in cancer progression. J. Cell Biol.  196, 395– 406 Google Scholar CrossRef Search ADS PubMed  8 Kino J., Adachi E., Yoshida T., Nakajima K., Hayashi T. ( 1988) Characterization of a monoclonal antibody against human placenta type IV collagen by immunoelectroblotting, antibody-coupled affinity chromatography, and immunohistochemical localization. J. Biochem . 103, 829– 835 Google Scholar CrossRef Search ADS PubMed  9 Hazeki N., Yamato M., Imamura Y., Sasaki T., Nakazato K., Yamamoto K., Konomi H., Hayashi T. ( 1998) Analysis of matrix protein components of the dermis-like structure formed in a long-term culture of human fibroblasts: type VI collagen is a major component. J. Biochem . 123, 587– 595 Google Scholar CrossRef Search ADS PubMed  10 Kaufmann S.H., Ewing C.M., Shaper J.H. ( 1987) The erasable Western blot. Anal. Biochem.  161, 89– 95 Google Scholar CrossRef Search ADS PubMed  11 Iwata M., Imamura Y., Sasaki T., Hayashi T. ( 1995) Evidence for a short form of alpha 1(IV) as a major polypeptide in bovine lens capsule. J. Biochem . 117, 1298– 1304 Google Scholar CrossRef Search ADS PubMed  12 Kornfeld K., Reitman M.L., Kornfeld R. ( 1981) The carbohydrate-binding specificity of pea and lentil lectins. Fucose is an important determinant. J. Biol. Chem . 256, 6633– 6640 Google Scholar PubMed  13 Tateno H., Nakamura-Tsuruta S., Hirabayashi J. ( 2009) Comparative analysis of core-fucose-binding lectins from Lens culinaris and Pisum sativum using frontal affinity chromatography. Glycobiology  19, 527– 536 Google Scholar CrossRef Search ADS PubMed  14 Baenziger J.U., Fiete D. ( 1979) Structural determinants of Ricinus communis agglutinin and toxin specificity for oligosaccharides. J. Biol. Chem . 254, 9795– 9799 Google Scholar PubMed  15 Wright C.S. ( 1987) Refinement of the crystal structure of wheat germ agglutinin isolectin 2 at 1.8 Å resolution. J. Mol. Biol.  194, 501– 529 Google Scholar CrossRef Search ADS PubMed  16 Wright C.S. ( 1992) Crystal structure of a wheat germ agglutinin/glycophorin-sialoglycopeptide receptor complex. Structural basis for cooperative lectin-cell binding. J. Biol. Chem . 267, 14345– 14352 Google Scholar PubMed  17 Wright C.S., Kellogg G.E. ( 1996) Differences in hydropathic properties of ligand binding at four independent sites in wheat germ agglutinin-oligosaccharide crystal complexes. Protein Sci.  5, 1466– 1476 Google Scholar CrossRef Search ADS PubMed  18 Fujiwara S., Shinkai H., Timpl R. ( 1991) Structure of N-linked oligosaccharide chains in the triple-helical domains of human type VI and mouse type IV collagen. Matrix  11, 307– 312 Google Scholar CrossRef Search ADS PubMed  19 Bieberich E. ( 2014) Synthesis, processing, and function of N-glycans in N-glycoproteins. Adv. Neurobiol.  9, 47– 70 Google Scholar CrossRef Search ADS PubMed  20 Schachter H. ( 2014) Complex N-glycans: the story of the “yellow brick road”. Glycoconj. J.  31, 1– 5 Google Scholar CrossRef Search ADS PubMed  21 Kizuka Y., Taniguchi N. ( 2016) Enzymes for N-glycan branching and their genetic and nongenetic regulation in cancer. Biomolecules  6, 25 Google Scholar CrossRef Search ADS   22 Harduin-Lepers A., Mollicone R., Delannoy P., Oriol R. ( 2005) The animal sialyltransferases and sialyltransferase-related genes: a phylogenetic approach. Glycobiology  15, 805– 817 Google Scholar CrossRef Search ADS PubMed  23 Bhide G.P., Colley K.J. ( 2017) Sialylation of N-glycans: mechanism, cellular compartmentalization and function. Histochem. Cell Biol.  147, 149– 174 Google Scholar CrossRef Search ADS PubMed  24 Trueb B., Winterhalter K.H. ( 1986) Type VI collagen is composed of a 200 kd subunit and two 140 kd subunits. EMBO J.  5, 2815– 2819 Google Scholar PubMed  25 Chu M.-L., Mann K., Deutzmann R., Pribula-Conway D., Hsu-Chen C.-C., Bernard M.P., Timpl R. ( 1987) Characterization of three constituent chains of collagen type VI by peptide sequences and cDNA clones. Eur. J. Biochem.  168, 309– 317 Google Scholar CrossRef Search ADS PubMed  26 Weil D., Mattei M.G., Passage E., N'Guyen V.C., Pribula-Conway D., Mann K., Deutzmann R., Timpl R., Chu M.L. ( 1988) Cloning and chromosomal localization of human genes encoding the three chains of type VI collagen. Am. J. Hum. Genet . 42, 435– 445 Google Scholar PubMed  27 Loeser R.F., Sadiev S., Tan L., Goldring M.B. ( 2000) Integrin expression by primary and immortalized human chondrocytes: evidence of a differential role for alpha1beta1 and alpha2beta1 integrins in mediating chondrocyte adhesion to types II and VI collagen. Osteoarthritis Cartilage  8, 96– 105 Google Scholar CrossRef Search ADS PubMed  28 Iyengar P., Espina V., Williams T.W., Lin Y., Berry D., Jelicks L.A., Lee H., Temple K., Graves R., Pollard J., Chopra N., Russell R.G., Sasisekharan R., Trock B.J., Lippman M., Calvert V.S., Petricoin E.F.III, Liotta L., Dadachova E., Pestell R.G., Lisanti M.P., Bonaldo P., Scherer P.E. ( 2005) Adipocyte-derived collagen VI affects early mammary tumor progression in vivo, demonstrating a critical interaction in the tumor/stroma microenvironment. J. Clin. Invest.  115, 1163– 1176 Google Scholar CrossRef Search ADS PubMed  29 You W.K., Bonaldo P., Stallcup W.B. ( 2012) Collagen VI ablation retards brain tumor progression due to deficits in assembly of the vascular basal lamina. Am. J. Pathol . 180, 1145– 1158 Google Scholar CrossRef Search ADS PubMed  30 Ruhl M., Sahin E., Johannsen M., Somasundaram R., Manski D., Riecken E.O., Schuppan D. ( 1999) Soluble collagen VI drives serum-starved fibroblasts through S phase and prevents apoptosis via down-regulation of bax. J. Biol. Chem.  274, 34361– 34368 Google Scholar CrossRef Search ADS PubMed  31 Howell S.J., Doane K.J. ( 1998) Type VI collagen increases cell survival and prevents anti-β1 integrin-mediated apoptosis. Exp. Cell Res . 241, 230– 241 Google Scholar CrossRef Search ADS PubMed  32 Burg M.A., Tillet E., Timpl R., Stallcup W.B. ( 1996) Binding of the NG2 proteoglycan to type VI collagen and other extracellular matrix molecules. J. Biol. Chem.  271, 26110– 26116 Google Scholar CrossRef Search ADS PubMed  33 Tulla M., Pentikainen O.T., Viitasalo T., Kapyla J., Impola U., Nykvist P., Nissinen L., Johnson M.S., Heino J. ( 2001) Selective binding of collagen subtypes by integrin alpha 1I, alpha 2I, and alpha 10I domains. J. Biol. Chem.  276, 48206– 48212 Google Scholar CrossRef Search ADS PubMed  34 Tillet E., Gential B., Garrone R., Stallcup W.B. ( 2002) NG2 proteoglycan mediates β1 integrin-independent cell adhesion and spreading on collagen VI. J. Cell. Biochem.  86, 726– 736 Google Scholar CrossRef Search ADS PubMed  35 Zigrino P., Drescher C., Mauch C. ( 2001) Collagen-induced proMMP-2 activation by MT1-MMP in human dermal fibroblasts and the possible role of α2β1 integrins. Eur. J. Cell Biol . 80, 68– 77 Google Scholar CrossRef Search ADS PubMed  36 Hocking D.C., Sottile J., Langenbach K.J. ( 2000) Stimulation of integrin-mediated cell contractility by fibronectin polymerization. J. Biol. Chem.  275, 10673– 10682 Google Scholar CrossRef Search ADS PubMed  37 Endo T. ( 2015) Glycobiology of alpha-dystroglycan and muscular dystrophy. J. Biochem.  157, 1– 12 Google Scholar CrossRef Search ADS PubMed  38 Aigner T., Hambach L., Söder S., Schlötzer-Schrehardt U., Pöschl E. ( 2002) The C5 domain of Col6A3 is cleaved off from the Col6 fibrils immediately after secretion. Biochem. Biophys. Res. Commun . 290, 743– 748 Google Scholar CrossRef Search ADS PubMed  39 Park J., Scherer P.E. ( 2012) Adipocyte-derived endotrophin promotes malignant tumor progression. J. Clin. Invest.  122, 4243– 4256 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations ABA Agaricus bisporus agglutinin AU arbitrary unit CBB Coomassie Brilliant Blue FBS fetal bovine serum Gal β-galactose IB immunoblotting IP immuno-precipitation LP lectin-precipitation LCA Lens culinaris agglutinin 2-ME 2-mercaptoethanol MW molecular weight GlcNAc N-acetyl-D-glucosamine GlcNAc-β-1,4-GIcNAc N,N' diacetyl-chitobiose NTH non-triple helical PBS phosphate-buffered saline PVDF polyvinylidene difluoride RCA120 Ricinus communis agglutinin I NeuAc sialic acid TFA trifluoroacetic acid WGA wheat germ agglutinin © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The Journal of BiochemistryOxford University Press

Published: Apr 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off