Glycobiology Advance Access originally published online on April 13, 2006
Glycobiology 2006 16(8):729-735; doi:10.1093/glycob/cwj114
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Isolation and identification of nine sulfated glycosphingolipids containing two unique sulfated gangliosides from the African green monkey kidney cells, Verots S3, and their possible metabolic pathways
2 Research Center of Biomedical Analysis and Radioisotope and 3 Department of Biochemistry, Teikyo University School of Medicine, 2-11-1 Kaga Itabashi-ku, Tokyo 173-8605, Japan
1 To whom correspondence should be addressed; e-mail: yniimura{at}med.teikyo-u.ac.jp
Received on January 31, 2006; revised on April 6, 2006; accepted on April 7, 2006
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Verots S3 cells derived from the African green monkey kidney were revealed to contain nine types of sulfoglycolipids by incorporating [35S]sulfate. These sulfated glycolipids were separated by DEAE-Sephadex column chromatography and preparative thin-layer chromatography (TLC). The major sulfoglycolipids were characterized using TLC, gas–liquid chromatography (GLC), mass spectrometry, solvolysis, TLC immunostaining, and nuclear magnetic resonance spectra as follows: V1, SM4s (GalCer I3-sulfate); V2, SM3 (LacCer II3-sulfate); V3, SM2a (Gg3Cer II3-sulfate); V4, globopentaosyl ceramide sulfate (Gb5Cer V3-sulfate); V5, (Gg4Cer II3-sulfate, IV3-NeuAc); V6, SB1a (Gg4Cer II3, IV3-bis-sulfate); and V8, (Gg4Cer II3-NeuAc, IV3-sulfate). Both V5 and V8 were sulfated gangliosides comprising both N-acetyl neuraminic acid and sulfate, and this was the first report on V8. A minor component V7 was identified as SM1a (Gg4Cer II3-sulfate) based on its behavior in TLC, GLC, and liquid secondary ion mass spectroscopy. It was postulated that this substance was a precursor of V6 (SB1a) and V5 (Gg4Cer II3-sulfate, IV3-NeuAc), and to date, its presence has not been demonstrated in nature. Another minor component V9 was identified as glucosyl ceramide sulfate based on its migration in TLC and GLC. This renal cell line was shown to be an excellent model for studying the metabolism and function of sulfoglycolipids.
Key words: African green monkey kidney cell / biosynthetic pathways of sulfoglycolipids / sulfated ganglioside / sulfoglycolipids
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Various types of sulfoglycolipids are known to be ubiquitously distributed in the vertebrate kidney (Ishizuka, 1997
), and SM4s—a common sulfoglycolipid of the kidney—appears to be accumulated more abundantly in the medulla of the kidney than in the cortex (Karlsson et al., 1973). Osmolality is higher in the medulla than in the cortex, and this contributes to the reabsorption of water and urine concentration in the kidneys. The algorithm between the sulfoglycolipid concentration in the kidney and body weight suggested that sulfoglycolipids are important for water transport in the kidneys (Nagai et al., 1984). It has been shown that the Madin-Darby canine kidney (MDCK) cells synthesize the sulfated glycosphingolipids SM4s (GalCer I3-sulfate) and SM3 (LacCer II3-sulfate), and their synthesis depends on the osmolality of the medium (Niimura and Ishizuka, 1986, 1990). Furthermore, the hyperosmosis-resistant cells that were cloned from MDCK cells accumulate a higher concentration of these sulfoglycolipids (Niimura and Ishizuka, 1991). These results suggest that sulfoglycolipids in the renal epithelial cell membrane may contribute to the barrier against hypertonicity in osmoregulation in the counter current system. We surveyed the sulfoglycolipids in other renal cells and found that the Verots S3 cells derived from African green monkey kidneys synthesize many types of sulfated glycolipids. In this study, we characterized the sulfoglycolipids of the Verots S3 cells and described their possible metabolic pathways. |
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Metabolic labeling of sulfoglycolipids in Verots S3 cells and preparation of sulfoglycolipids
The rate of incorporation of [35S]sulfate for 24 h into the sulfolipid fraction of Verots S3 cells derived from African green monkey kidneys was 21,800 dpm/mg cell protein, and this was
11-fold of the rate in original Vero cells. The profiles of the sulfoglycolipids labeled with [35S]sulfate in the Verots S3 cells were investigated using two-dimensional thin-layer chromatography (2D-TLC) (Fig. 1).
|
Verots S3 cells have been shown to synthesize more than nine types of sulfoglycolipids that are designated as V1–V9. The amount of [35S]sulfate incorporated into these fractions was 8.1% (V1), 38.1% (V2), 2.6% (V3), 7.5% (V4), 5.4% (V5), 32.2% (V6), 0.5% (V7), 0.8% (V8), and 0.2% (V9) of the totally incorporated activity. Sulfoglycolipids labeled with [35S]sulfate were further separated using DEAE-Sephadex column chromatography and preparative TLC (Fig. 2).
|
A weak spot (V9) was not detected in this preparation; it was later detected on TLC using the solvent system chloroform : methanol : acetone : acetic acid : water (6 : 3 : 4 : 2 : 1, v/v). Each sulfoglycolipid was detected as a set of double bands that indicated the different compositions of the fatty acid species. V5, V6, and V8 were eluted in fractions derived from mono-sulfoglycolipid fractions (V1, V2, V3, V4, V7, and other bands).
Identification of major sulfoglycolipids (V1, V2, V3, and V6)
Sulfoglycolipids of the Verots cells labeled with radioactive sulfate were isolated using preparative TLC. Among these sulfated lipids, four major sulfoglycolipids (V1, V2, V3, and V6) showed behaviors similar to the standard SM4s, SM3, SM2a (Tadano and Ishizuka, 1982), and SB1a (Tadano et al., 1982), respectively, as observed on TLC and autoradiography (Fig. 3).
|
Simultaneously, the non-radioisotope (RI) preparation of the acidic glycolipids from mass-cultured cells (109 cells) was performed using DEAE-Sephadex column chromatography and preparative TLC to obtain each sulfoglycolipid that was detected using iodine vapor. The carbohydrate composition of major sulfoglycolipids was shown to be Gal for V1, Glc:Gal (1 : 1) for V2, Glc:Gal:GalNAc (1 : 1 : 1) for V3, and Glc:Gal:GalNAc (1 : 2 : 1) for V6. The mass spectrometric data of these sulfoglycolipids (summarized in Table I) show a structure comprising HSO3-Hex-Cer for V1, HSO3-Hex-Hex-Cer for V2, HexNAc-[HSO3]Hex-Hex-Cer for V3, and HSO3-Hex-HexNAc-[HSO3]Hex-Hex-Cer for V6.
|
[M-H]– ions that are composed of ceramide [d18:1/C24:0] and [d18:1/C24:1] were also detected in SM4s. Solvolysis of the labeled V6 revealed the product corresponding to SM1a on TLC (data not shown). Furthermore, TLC immunostaining revealed that V6 was reactive to anti-SB1a antibody (data not shown). On the basis of these results, the four major sulfoglycolipids were identified as SM4s (V1), SM3 (V2), SM2a (V3), and SB1a (V6).
Characterization of sulfoglycolipids V4 and V5
On the basis of the mass spectrum of V4 (Fig. 4), its core structure was demonstrated as SO3-Hex-HexNAc-Hex-Hex-HexCer. The major ceramides consisted of [d18:1/C24:0], [d18:1/C24:1], [d18:1/C22:0], and [d18:1/C16:0]. On the basis of the nuclear magnetic resonance (NMR) spectrum and comparison with data obtained previously from the human kidneys, V4 was concluded to be globopentaosyl ceramide V3-sulfate (Nagai et al., 1989) (Table II). V5 was eluted from the DEAE-Sephadex column from between the mono-sulfoglycolipid and bis-sulfoglycolipid fractions. On the basis of the mass spectrum of V5 (Fig. 5), the structure was suggested to be NeuAc-Hex-HexNAc-[HSO3]Hex-Hex-Cer.
|
|
|
The ceramide consisted of [d18:1/C24:0]. The treatment of V5 with neuraminidase in the presence or absence of cholate produced new bands that translocated to the region of SM1a (Fig. 6). This suggests that the NeuAc residue is linked to the terminal hexose of the sugar chain. The desialylated product comigrated with SM1a that was derived from SB1a on TLC (data not shown). Because the NMR spectrum of V5 produced signals identical to those produced by Gg4Cer II3-sulfate,IV3-NeuAc reported previously (Kasama et al., 2001) (Table II), the structure of V5 was identified as Gg4Cer II3-sulfate,IV3-NeuAc.
|
Minor components V7, V8, and V9
V7 migrated between V3 and V4 on TLC (Fig. 1). The two components of V7 separated by TLC were blotted on a membrane and were analyzed by mass spectrometry. The faster moving compound comprised [d18:1/C24:0] (Fig. 7), whereas the slower moving one comprised [d18:1/C16:0] (data not shown). The carbohydrate analysis of V7 showed Glc:Gal:GalNAc (1 : 2 : 1). Detection of a set of m/z values—1053 (or 941) and 622—attributable to a sulfate group suggests the presence of a sulfate ester bond in the second hexose of the sugar chain (Tadano-Aritomi et al., 1995). The core structure of Hex-HexNAc-(SO3-)Hex-Hex-Cer was concluded. Furthermore, V7 comigrated with SM1a that was derived from SB1a on TLC (Fig. 8).
|
|
Although the NMR spectrum was not measured because of lack of material, the structure of V7 can tentatively be assigned to SM1a based on its structure, which has not been reported before. However, this compound is a potential precursor of SB1a (V6) and Gg4Cer II3-sulfate,IV3-NeuAc (V5).
V8 was eluted more slowly from the DEAE-Sephadex column as compared with V5, and it moved faster than V5 on TLC. Its carbohydrate composition was Glc:Gal:GalNAc:NeuAc (1 : 2 : 1 : 1). The mass spectrum of V8 showed HSO3-Hex-HexNAc-[NeuAc] Hex-Hex-Cer as its core structure (Fig. 9), with the same molecular ion as that of V5; however, its pattern of fragmentation was different.
|
The NMR spectrum of V8 (Fig. 10) showed a spectrum similar to that of Gg4Cer II3-NeuGc,IV3-sulfate reported previously (Tadano-Aritomi et al., 1998) (Table II). The N-acetyl methyl signal of the NeuAc residue was observed at 1.88 ppm as a singlet.
|
On the basis of these results, the glycon structure of V8 was determined as Gg4Cer II3-NeuAc,IV3-sulfate (HSO3-3Galß1-3GalNAcß1-[NeuAc
2-3]4Galß1-4Glcß1-Cer).
The sulfolipid V9 moved faster than SM4s on TLC with the solvent system chloroform : methanol : acetone : acetic acid : water (6 : 3 : 4 : 2 : 1, v/v) (Iida et al., 1989) and was separated by performing TLC using the same solvent system. The gas–liquid chromatography (GLC) analysis of V9 showed glucose as the sole carbohydrate. On the basis of the behavior in TLC and the results of GLC, sulfoglycolipid V9 was hypothesized to be glucosyl ceramide sulfate. Because V9 was present in very low amounts, its structure could not be completely characterized.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
We investigated the sulfoglycolipids in the Verots S3 cells derived from the African green monkey kidneys by incorporating [35S]sulfate. These cells showed higher incorporation of [35S]sulfate into sulfoglycolipid fractions and different types of sulfoglycolipids, whereas the original Vero cells contained only SM4s and SM3 as the predominant sulfoglycolipids (unpublished data). We characterized the structures of nine types of sulfated glycolipids and presented V8 as the new glycon structure of Gg4Cer II3-NeuAc,IV3-sulfate, although a similar structure with N-glycolylneuraminic acid has been reported previously (Tadano-Aritomi et al., 1998
). Furthermore, the structure of V7 was identified as that of SM1a that has been reported to be the chemically degraded product of SB1a (Tadano et al., 1982) and has been speculated to be a precursor in the biosyntheses of SB1a and Gg4Cer II3-sulfate,IV3-NeuAc. The presence of both SB1a (V6) and Gg4Cer II3-sulfate, IV3-NeuAc (V5) in this cell strain strongly suggests the presence of SM1a as a precursor. The possible biosynthetic pathways of these sulfoglycolipids are shown in Fig. 11.
|
The sulfotransferase activity in this cell strain was estimated to be more elevated than in the original Vero cells. The sulfotransferase gene has been cloned from renal cancer cells (Honke et al., 1997). We believe the sulfotransferase in kidney may perform the sulfation in many kinds of sulfoglycolipid. Recently, it has been reported that the chicken influenza virus binds to the receptors of sialyl and sulfated glycon structures with a high affinity (Gambaryan et al., 2004). It is interesting that two types of sulfated gangliosides are exhibited in this cell line. The chemokine receptor functions of sulfated glycolipids have also been investigated (Sandhoff et al., 2005). The Verots S3 cells are classified as transformant cells that are transfected with the temperature-sensitive large T antigen gene, which affects multiple cellular processes (Ahuja et al., 2005). Although the relation of the expression of the large T antigen to sulfoglycolipids biosynthesis is not known, this cell strain is an excellent model for investigating the functions of sulfoglycolipids and various renal transports.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Cell culture and metabolic labeling
Verots S3 cells were purchased from the Riken Cell Bank (Tsukuba, Japan) and were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS). The cells were labeled with [35S]sulfate (370–740 kBq/mL) for 24 h. The lipids were extracted as described below. After 2D-TLC, the radioactivity incorporated into sulfoglycolipids was determined using an imaging analyzer (BAS-1500; Fuji Photo Film Co. Ltd. [Tokyo, Japan]). Total activity was measured using a liquid scintillation counter.
Preparation of sulfoglycolipids
Sulfoglycolipids were prepared from 109 cells (200 dishes, 15 cm in diameter) by using the chloroform–methanol extraction method, mild alkali treatment, and DEAE-Sephadex A-25 column chromatography, using the solvent system chloroform : methanol : 0.6 M ammonium acetate (30 : 60 : 8, v/v) and the solvent system chloroform : methanol : 1.2 M ammonium acetate (30 : 60 : 8, v/v) (Niimura and Ishizuka, 1990). TLC was performed on Silica Gel 60 high-performance TLC (HPTLC) plates (E. Merck, Darmstadt, Germany) with the solvent system chloroform : methanol : 0.2% CaCl2 (55 : 45 : 10, v/v) in the first step and then with the solvent system chloroform : methanol : acetone : acetic acid : water (5 : 2 : 4 : 2 : 1, v/v) in the second step. The glycolipids were visualized by using iodine vapors or by spraying the plate with orcinol–H2SO4 reagent and heating it for 5 min at 120°C.
Chemical analysis and solvolysis
Monosaccharides, fatty acids, and sphingoids were analyzed by GLC as described (Niimura et al., 1999). The sulfoglycolipids (V6, 2000 dpm) were incubated with 8 mM H2SO4 in dimethyl sulfoxide (Me2SO) : methanol (9 : 1) for 30 min at 70°C (Tadano et al., 1982). After neutralizing the solution with ammonia, the mixture was evaporated and applied to the DEAE-Sephadex A-25 column. After the column was washed with methanol and chloroform : methanol : water (30 : 60 : 8, v/v), elution was performed using chloroform : methanol : 0.6 M ammonium acetate (30 : 60 : 8, v/v).
Spectral analysis
Negative ion liquid secondary ion mass spectrometry (LSIMS) was performed using the Concept 1H spectrometer (Shimadzu/Kratos, Kyoto, Japan) fitted with a cesium-ion gun (Tadano-Aritomi et al., 1998). Approximately, 0.5 nmol of underivatized glycolipids was mixed with 1 µL triethanolamine as the matrix. The spectra were recorded at an accelerating voltage of 8 kV with a scan rate of 5 s/decade and at a resolution of 1000–2000.
For 1H-NMR spectroscopy measurements, purified glycolipids were treated repeatedly with 0.5-mL portions of CH3O[2H], followed by desiccation over P2O5 in vacuo in order to exchange labile protons with deutrons. Dried glycolipids were re-dissolved in 0.5 mL of [2H]Me2SO : [2H]2O (98 : 2, v/v) mixture. The spectra were recorded on a GX-400 400-MHz spectrometer from the Japan Electron Optical (JEO, Tokyo, Japan) at 60°C. The operation conditions for one-dimensional spectrum were as follows: frequency, 400 MHz; sweep width, 4 kHz; and sampling points, 16 k. All two-dimensional spectra were recorded with 512 x 2048 data points and a spectral width of 2500 Hz as described previously (Iida et al., 1989). Chemical shifts were indicated in terms of ppm based on the signal of tetramethylsilane (TMS) as an internal standard.
TLC immunostaining
The glycolipids were separated by TLC. After development, the plate was dried and then soaked in n-hexane containing 0.1% polyisobutylmethacrylate for 10 min. The plate was dried and then incubated with a 1 : 20 dilution of anti-SD1a monoclonal antibodies (Seikagaku Co., Tokyo, Japan) in PBS containing 1% egg albumin at 37°C for 30 min. After washing with PBS, sequential incubations were performed with 1 : 200 dilutions of peroxidase-conjugated goat anti-mouse IgM antibodies in PBS containing 1% egg albumin at 37°C for 30 min. After washing with PBS, the SB1a active glycolipid was visualized using 0.3% 4-chloro-1-naphthol.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
None declared.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
We thank Mrs Michiko Ogawa for providing technical assistance, Dr Keiko Tadano-Aritomi for the measurement of negative ion LSIMS, and Dr Naoko Tanaka for 1H-NMR spectroscopy.
|
Abbreviations |
|---|
The nomenclature system for lipids follows the recommendation of the Nomenclature Committee, International Union of the Pure and Applied Chemistry (IUPAC-IUB and Joint Commission on Biochemical Nomenclature [JCBN], 1999
); 1H-NMR, proton nuclear magnetic resonance; Cer, ceramide; d18:1, 4-sphingenine; GalNAc, N-acetyl galactosamine; GLC, gas–liquid chromatography; LSIMS, liquid secondary ion mass spectrometry; MDCK, Madin-Darby canine kidney cells; Me2SO, dimethyl sulfoxide; NeuAc, N-acetyl neuraminic acid; TLC, thin-layer chromatography |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Ahuja, D., Saenz-Robles, M.T., and Pipas, J.M. (2005) SV40 large T antigen targets multiple cellular pathways to elicit cellular transformation. Oncogene, 24, 7729–7745.[CrossRef][Web of Science][Medline]
Gambaryan, A.S., Tuzikov, A.B., Pazynina, G.V., Webster, R.G., Matrosovich, M.N., and Bovin, N.V. (2004) H5N1 chicken influenza viruses display a high binding affinity for Neu5Acalpha2-3Galbeta1-4(6-HSO3)GlcNAc-containing receptors. Virology, 326, 310–316.[CrossRef][Web of Science][Medline]
Honke, K., Tsuda, M., Hirahara, Y., Ishii, A., Makita, A., and Wada, Y. (1997) Molecular cloning and expression of cDNA encoding human, 3'-phosphoadenylylsulfate:galactosylceramide 3'-sulfotransferase. J. Biol. Chem. 272, 4864–4868.
Iida, N., Toida, T., Kushi, Y., Handa, S., Fredman, P., Svennerholm, L., and Ishizuka, I. (1989) A sulfated glucosylceramide from rat kidney. J. Biol. Chem., 264, 5974–5980.
Ishizuka, I. (1997) Chemistry and functional distribution of sulfoglycolipids. Prog. Lipid Res., 36, 245–319.[CrossRef][Web of Science][Medline]
IUPAC-IUB Joint Commission on Biochemical Nomenclature (JCBN). (1999) Nomenclature of glycolipids: Recommendations 1997. Glycoconj. J., 16, 1–6.
Karlsson, K.A., Samuelsson, B.E., and Steen, G.O. (1973) The sphingolipid composition of bovine kidney cortex, medulla and papilla. Biochim. Biophys. Acta, 316, 317–335.[Medline]
Kasama, T., Hamasaki, Y., Tanaka, N., Ishizuka, I., and Handa, S. (2001) Proceedings of the Japanese Conference on the Biochemistry of Lipids, 43, 151–154.
Nagai, K.I., Ishizuka, I., and Oda, S.I. (1984) Acidic glycolipids from kidney of suncus (Insectivora). J. Biochem., 95, 1501–1511.
Nagai, K., Roberts, D.D., Toida, T., Matsumoto, H., Kushi, Y., Handa, S. and Ishizuka, I. (1989) Mono-sulfated globotetraosylceramide from human kidney. J. Biochem., 106, 878–886.
Niimura, Y. and Ishizuka, I. (1986) Glycosphingolipid composition of a renal cell line (MDCK) and its ouabain-resistant mutant. J. Biochem., 100, 825–835.
Niimura, Y. and Ishizuka, I. (1990) Adaptive changes in sulfoglycolipids of kidney cell lines by culture in anisosmotic media. Biochim. Biophys. Acta, 1052, 248–254.[Medline]
Niimura, Y. and Ishizuka, I. (1991) Accumulation of sulfoglycolipids in hyperosmosis-resistant clones derived from the renal epithelial cell line MDCK (Madin-Darby canine kidney cell). Comp. Biochem. Physiol. B, 100, 535–541.[Medline]
Niimura, Y., Tomori, M., Tadano-Aritomi, K., Toida, T., and Ishizuka, I. (1999) The major acidic glycolipids from the kidney of the pacific salmon (Oncorhynchus keta): characterization of a novel ganglioside, fucosyl-N-acetylgalactosaminyl-GM1. J. Biochem., 126, 962–968.
Sandhoff, R., Grieshaber, H., Djafarzadeh, R., Sijmonsma, T.P., Proudfoot, A.E., Handel, T.M., Wiegandt, H., Nelson, P.J., and Grone, H.J. (2005) Chemokines bind to sulfatides as revealed by surface plasmon resonance. Biochim. Biophys. Acta, 1687, 52–63.[Medline]
Tadano, K. and Ishizuka, I. (1982) Isolation and characterization of the sulfated gangliotriaosylceramide from rat kidney. J. Biol. Chem., 257, 1482–1490.
Tadano, K., Ishizuka, I., Matsuo, M., and Matsumoto, S. (1982) Bis-sulfated gangliotetraosylceramide from rat kidney. J. Biol. Chem., 257, 13413–13420.
Tadano-Aritomi, K., Kubo, H., Ireland, P., Hikita, T., and Ishizuka, I. (1998) Isolation and characterization of a unique sulfated ganglioside, sulfated GM1a, from rat kidney. Glycobiology, 8, 341–350.
Tadano-Aritomi, K., Kubo, H., Ireland, P., Okuda, M., Kasama, T., Handa, S., and Ishizuka, I. (1995) Structural analysis of mono- and bis-sulfated glycosphingolipids by negative liquid secondary ion mass spectrometry with high- and low-energy collision-induced dissociation. Carbohydr. Res., 273, 41–52.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||