- Email: [email protected]
liquid interface culture system employing fluorous solvents
Journal of Fluorine Chemistry 188 (2016) 76–79
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u o r
Short Communication
Cellular glycosylation of amphiphilic saccharide primer in liquid/liquid interface culture system employing fluorous solvents Amane Sutoh, Maria Carmelita Z. Kasuya, Kenichi Hatanaka* Institute of Industrial Science, The University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8505, Japan
A R T I C L E I N F O
Article history: Received 14 April 2016 Received in revised form 20 June 2016 Accepted 21 June 2016 Available online 21 June 2016 Keywords: fluorous solvent liquid/liquid interface melanoma B16 cells saccharide primer ganglioside
A B S T R A C T
As fluorous compounds have extremely high oxygen-carrying capacity, liquid/liquid (culture medium/ fluorous solvent) interface culture system is one of their attractive applications in the field of biotechnology. However, it remains to be elucidated whether the metabolism of the cells cultured in such a system could work properly. To determine whether glycolipid metabolism of these cells are affected by such a low adhesive condition at the interface, we evaluated cellular glycosylation of a synthetic pseudoglycolipid by mouse melanoma B16 cells cultured in liquid/liquid interface system using three kinds of fluorous solvents. The results showed that the saccharide primer was sialylated by the cells when cultured at the interface using FluorinertTM FC-72 or perfluorodecalin, but not with dodecafluoroheptanol. The results of this work confirmed that glycolipid metabolism of cells cultured at the interface of aqueous/fluorous culture system functioned properly as well as normal aqueous culture system but depending on the kind of fluorous solvent used. Ganglioside biosynthesis was not compromised amidst an unusual environment such as the absence of a solid substrate for cell attachment and the presence of a perfluorinated solvent. ã 2016 Elsevier B.V. All rights reserved.
1. Introduction Fluorous solvents, i.e. highly fluorine-substituted liquid chemicals, have approximately 20 times higher O2 solubility than water [1] because of their relatively weak intermolecular interactions. Hence, they can be used as flexible O2 carriers to facilitate oxygen supply to cells or tissues [2]. As they are immiscible with water, several studies have addressed animal cell culture at the interface formed between fluorous solvent and overlying culture medium [3–6]. These studies have demonstrated that the anchoragedependent cells were able to grow and proliferate at the interface. Such culture system is expected to enhance the oxygen access to the cells and to overcome the poor oxygen supply, which is often considered to be a limiting factor of three-dimensional cell culture in vitro [5]. In addition, this simple culture method does not require trypsin treatment for subcultivation, thereby the adverse effects by trypsin [3] can be eliminated.
Abbreviations: DF-heptanol, dodecafluoroheptanol; PFD, perfluorodecalin; TIDF, DMEM/F12 containing Insulin-Transferrin-Selenium-X; HPTLC, high-performance thin layer chromatography. * Corresponding author. E-mail address: [email protected] (K. Hatanaka). http://dx.doi.org/10.1016/j.jfluchem.2016.06.011 0022-1139/ã 2016 Elsevier B.V. All rights reserved.
In previous studies [7], we evaluated morphology, growth and oxygen consumption of the cells cultured on the liquid/liquid interface, and reported that 2,2,3,3,4,4,5,5,6,6,7,7,-dodecafluoro-1heptanol (DF-heptanol) serves as an oxygen reservoir that could supply O2 to the culture medium. However, there is also an intriguing issue of whether metabolism of these cells are affected by such a low adhesive condition at the interface. Although some studies have reported on metabolic response of cultured cells to liquid/liquid interface system employing oxygenated perfluorocarbons [8], little is known about the effects of the system itself on cellular metabolism. In this research, we focused on oligosaccharide elongation, particularly biosynthesis of gangliosides by the cells at the liquid/ liquid interface. Gangliosides are sialic acid-containing glycosphingolipids, which are known to exist in the lipid raft on the surface of the plasma membrane and to be associated with a variety of biological phenomena, such as cell adhesion, signal transduction, malignant alteration of cancer cells, pathology of diseases, etc [9,10]. Unlike nucleic acids and proteins, they are not directly encoded in DNA and they are susceptible to the physiological conditions of the cells. Therefore, in order to elucidate the gap between genomic sequences and complicated biological phenomena, it is important to explore ganglioside behavior, which reflects physiological changes of the cells.
A. Sutoh et al. / Journal of Fluorine Chemistry 188 (2016) 76–79
We evaluated biosynthesis of ganglioside GM3 by mouse melanoma B16 cells at the culture medium/fluorous solvent interface using saccharide primer method. This method is an effective way to monitor biosynthesis of oligosaccharides without destroying the cells, in which the cells are incubated with saccharide primer, a synthetic pseudo-glycolipid, and the primers are taken into the cells and subsequently glycosylated by cellular glycosyltransferase [11,12]. Using this approach, previously we assessed the effects of some factors of cell culture conditions, e.g. xenobiotics [13] and the passage number of the cells [14], on cellular glycosylation. In this study, mouse melanoma B16 cells were cultured at the interface formed between fluorous solvents, such as DF-heptanol, FluorinertTM FC-72 (commercially available mixture of perfluorochemicals whose main constituent is perfluorohexane [15]) or perfluorodecalin (PFD), and culture medium containing dodecyl b-lactoside (Lac-C12) primer that mimics lactosylceramide, a precursor of a variety of glycolipids. How the cells behave in the presence of the saccharide primer during liquid/liquid interface culture was investigated and the effect of different fluorous solvents was compared. 2. Results 2.1. Morphology and viability of B16 cells at the liquid/liquid interface In order to investigate the effect of different fluorous solvents on cell morphology, B16 cells were observed using an optical microscope after 48 h of culture. Whereas the cells spread in polystyrene dish (Fig. 1A), the cells cultured in liquid/liquid system employing DF-heptanol remained rounded and existed in monolayer (Fig. 1B). As for the cells cultured with FC-72 or PFD, they were also rounded, but were grown as three-dimensional multicellular aggregates (Fig. 1C, D). From a macroscopic point of view (Fig. 1E), a sheet-like structure of the cells was observed when DF-heptanol was used. In addition, the medium overlying DF-heptanol was greatly blackened, which is considered to be due to melanin production. The number of the viable cells after 48 h for each condition is shown in Fig. 2. When FC-72 or PFD was used, there was no significant difference from control. By contrast, with DF-heptanol, the number of the viable cells was lower than control, indicating that DF-heptanol may not be suitable for this study because of its cytotoxicity. In the previous work using DF-heptanol as oxygen reservoir during biphasic culture [7], the cytotoxic effect of the fluorous solvent was not easily recognized because the medium used was supplemented with 10% fetal bovine serum (FBS) that promoted growth and proliferation. On the other hand, cell culture in the
77
presence of the saccharide primer as scaffold for glycosylation of cellular enzymes in this work required the use of culture medium without FBS to avoid the serum from interfering with ganglioside accumulation. This condition limited cell growth and proliferation. In both studies, cells thrived but the adverse effect of DF-heptanol on cell viability and its repercussion on glycosylation became more evident when compared with other fluorous solvents such as FC-72 and perfluorodecalin. 2.2. HPTLC analysis and identification of the sialylated product After incubation of B16 cells with Lac-C12 primer for 48 hours, the cells and the medium fractions were collected and analyzed by HPTLC. HPTLC results of the lipid extract from the cell homogenates and the culture medium are shown in Fig. 3A. The compounds containing sialic acid were visible as blue-violet bands [19]. In the presence of the Lac-C12 primer, there were new bands corresponding to a putative sialylated product, ganglioside GM3 analogue, in medium fraction when FC-72 or PFD was used (Lane 14, 15). Such a band could not be detected when DF-heptanol was used (Lane 13). It was shown that the primer was not taken into the cells cultured at medium/DF-heptanol interface (Lane 5). In order to confirm the structure of the products, they were scraped from the HPTLC plate, extracted with MeOH, and then identified from ESI-MS and enzyme assay. Mass spectrum of the product showed a peak at m/z 800.1. Fragmentation of the product showed peaks at m/z 508.9 and 289.7, corresponding to Lac-C12 primer and N-acetylneuraminic acid (NeuAc), respectively. Treatment with a2,3-sialidase showed that the product was hydrolyzed (Fig. 3B). Hence, it was confirmed that the product has a terminal a2-3 sialic acid moiety. From these results, the product was identified as a GM3 (NeuAc(a2-3)Lac-C12) analog. 2.3. Effects of the different fluorous solvents on the amount of sialylated primer The amounts of GM3 analog and endogenous GM3 in cell and medium fractions were quantitated by densitometric analysis (Fig. 4). The amounts of GM3 analog from the cells cultured in polystyrene flask, and liquid/liquid interface system using FC-72 and PFD were 11.87, 10.47 and 10.72 mmol/flask, respectively (SDs = 0.48, 1.80 and 1.76 mmol/flask, respectively). There were no significant differences among them. 3. Disucussion The present study demonstrated that Lac-C12 primer was sialylated by B16 cells when cultured at the interface using perfluoroalkanes, such as FC-72 or PFD, in the same way as when
Fig. 1. Morphology of B16 cells during (A) normal culture in polystyrene dish, and liquid/liquid interface culture employing (B) DF-heptanol, (C) FC-72 and (D) PFD. (E) The examples of liquid/liquid interface culture system after 48 h of culture, From left to right: normal culture, liquid/liquid interface culture employing DF-heptanol, FC-72, PFD.
78
A. Sutoh et al. / Journal of Fluorine Chemistry 188 (2016) 76–79
Fig. 2. The number of the viable cells after 48 h of incubation with/without Lac-C12 primer in liquid/liquid interface culture system. The data are presented as means SD (n = 3).
The findings of this research strongly support the usefulness of liquid/liquid interface culture system. Mouse melanoma B16 cells thrived at the interface between the culture medium layer and the fluorous solvent layer, and biosynthesis of glycolipids took place. This cell culture system has potential application to high-density culture of animal cells. By this method, the treatment of cells using the protease trypsin is not required for sub-cultivation and cells can be collected easily by pipetting and centrifugation. Moreover, the fluorous solvent can enhance oxygen supply to the cells. 4. Experimentation 4.1. Cell culture Mouse melanoma B16 cells were purchased from Riken Cell Bank (Tsukuba, Japan) and routinely maintained in Dulbecco’s Modified Eagle Medium: Nutrient Mixture F-12 (D-MEM/F-12; Gibco, Life Technologies Japan Ltd., Tokyo, Japan) supplemented with 10% FBS (JRH Biosciences, KS, USA) and 1% AntibioticAntimycotic solution (Nacalai Tesque, Kyoto, Japan). 4.2. Incubation of the cells with saccharide primer on the liquid/liquid interface
Fig. 3. HPTLC results. (A) Profile of lipids extracted from cell (Lanes 1-8) and culture medium (Lane 9-16) fractions after incubation of B16 cells with/without Lac-C12 primer. Lanes 1, 5, 9, 13: DF-heptanol; Lanes 2, 6, 10, 14: FC-72; Lanes 3, 7, 11, 15: PFD; Lanes 4, 8, 12, 16: control; Lane 17: standard endogenous GM3; Lane 18: Lac-C12 primer. (B) Profile before and after treatment with a 2,3-sialidase. Lane 1: medium fraction; Lanes 2, 3: the product without/with a 2,3-sialidase; Lane 4: Lac-C12 primer.
cultured on polystyrene dishes, but not with DF-heptanol. Also, the cells aggregated using perfluoroalkanes, but not in the case of DFheptanol. Although the cause for the difference requires further investigation, our data suggested that the cell membrane was adversely affected after exposure to fluorous alchohol and that uptake of the saccharide primer was inhibited. On the other hand, sialylation of the saccharide primer was not affected during liquid/ liquid interface culture employing perfluoroalkanes. This result indicates that all the steps of saccharide primer method, such as uptake of the primer, sialylation by sialyltransferase and release of the sialylated product functioned properly and this is the first report on ganglioside biosynthesis at the liquid/liquid interface.
Dodecyl b-lactoside (Lac-C12 primer) was synthesized as described previously [16]. In the saccharide primer method, serum-free medium TI-DF was used in order to avoid serum interfering with ganglioside accumulation [17]. TI-DF consists of D-MEM/F12 (20 mL) containing 1% Insulin-Transferrin-Selenium-X (ITS-X; Gibco) and 1% of Antibiotic-Antiomycotic solution. TI-DF (20 mL) was slowly overlaid onto three kinds of fluorous solvents (10 mL), such as 2,2,3,3,4,4,5,5,6,6,7,7,-dodecafluoro-1-heptanol (DF-heptanol; Tokyo Chemical Industry Co., Ltd., Tokyo, Japan), FluorinertTM FC-72 (3 M Japan Ltd., Tokyo, Japan) and perfluorodecalin (PFD; Tokyo Chemical Industry Co., Ltd.) in cell culture flasks (Nunc, Thermo Scientific Inc., MA, USA). B16 cells in dishes under subconfluent conditions were washed with TI-DF and harvested using 0.25% trypsin/EDTA. Then, 2 106 cells were seeded on the liquid/liquid interface with 50 mM of Lac-C12 primer. The flasks were incubated in a humidified atmosphere of 5% CO2 air at 37 C for 48 h. As control experiments, the cells were also cultured in TI-DF only with Lac-C12 primer. Besides, the cells were cultured at the interface between fluorous solvent and TI-DF without Lac-C12 primer. All the samples were prepared in triplicate.
A. Sutoh et al. / Journal of Fluorine Chemistry 188 (2016) 76–79
79
with resorcinol-HCl-Cu2+ reagent and heated at 120 C for 15 min. The densitometric analysis of the separated glycolipid bands were performed using ImageJ software (NIH, MD, USA). For each HPTLC plate, a standard curve was generated from the bands of reference ganglioside GM3 (Sigma-Aldrich, MO, USA) at various concentrations, and the amount of sialylated compounds in each lane was calculated. 4.5. Identification of glycosylated products ESI-MS analysis of the glycosylated products was performed on HCTultra (Bruker Daltonics, Bremen, Germany). Enzymatic digestion of the glycosylated products was carried out in 100 ml of AcONa buffer (pH 5.5) containing 0.5 mU of a2,3-sialidase from salmonella typhimurium LC2 (Takara Bio Inc., Shiga, Japan). The reaction was performed overnight at 37 C and then analyzed by HPTLC. Fig. 4. Amounts of sialylated primer and endogenous GM3 calculated from HPTLC (n = 3).
4.3. Lipid extraction After incubation for 48 h, the lower fluorous layer was removed, the cells were separated from the medium fraction by centrifugation and washed with PBS( ). To estimate the cell viability, the cells were stained with trypan blue and counted using Countess Automated Cell Counter (Invitrogen, Life Technologies, CA, USA). The lipids were extracted from the cell pellet with CHCl3-MeOH (2:1, v/v), then with CHCl3-2-propanol-water (7:11:2, v/v/v) in a sonicated bath. Lipids from the medium fraction were purified using reversed-phase SPE cartridge Sep-Pak C18 (Waters Corp., MA, USA) followed by anion-exchange SPE cartridge Inert-Sep SAX (GL Sciences Inc., Tokyo, Japan) [18]. 4.4. HPTLC analysis
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]
The products obtained from the culture medium fraction and the cell fraction were analyzed by high-performance thin layer chromatography (HPTLC Silica Gel 60 F-254; E. Merck KG, Darmstadt, Germany) with CHCl3-MeOH-0.25% aq. KCl solution (5:4:1, v/v/v) as developing solvent. The HPTLC plates were sprayed
[17] [18] [19]
K.C. Lowe, J. Fluor, Chem. 118 (2002) 19–26. M. Pilarek, Chem. Process Eng 35 (2014) 463–487. Y. Shiba, T. Oshima, M. Sato, Biotechnol. Bioeng. 57 (1998) 583–589. Y.J. Kwon, H. Yu, C.-A. Peng, Biotechnol. Bioeng. 72 (2001) 331–338. M. Pilarek, I. Grabowska, M.A. Ciemerych, K. Da˛bkowska, K.W. Szewczyk, Biotechnol. Lett. 35 (2013) 1387–1394. ski, J. Janicka, K. Janczyk-Ilach, M. Pilarek, I. Grabowska, I. senderek, M. Wojasin T. Ciach, Bioprocess Biosyst. Eng. 37 (2014) 1707–1715. H. Miyajima, M.C. Kasuya, K. Hatanaka, J. Fluor, Chem. 163 (2014) 46–49. J. Wardrop, M.R. Davey, J.B. Power, K.C. Lowe, Artif. Cells Blood Substit. Immobil. Biotechnol. 30 (2002) 63–70. S.-I. Hakomori, Curr. Opin. Hematol. 10 (2003) 16–24. T. Ariga, M.P. McDonald, R.K. Yu, J. Lipid, Res. 49 (2008) 1157–1175. Y. Miura, T. Yamagata, Biochem. Biophys. Res. Commun. 241 (1997) 698–703. T. Sato, K. Hatanaka, H. Hashimoto, T. Yamagata, Trends Glycosci. Glyc. 19 (2007) 1–17. T. Kato, M. Muraoka, K. Hatanaka, Anal. Biochem. 405 (2010) 103–108. Y. Shimura, J. Suzuki, M.C.Z. Kasuya, K. Matsuoka, K. Hatanaka, J. Biosci Bioeng, 114 (2012) 552–555. P. Adlercreutz, B. Mattiasson, J. Eur, Appl. Microbiol. Biotechnol. 16 (1982) 165–170. M.C.Z. Kasuya, M. Ikeda, K. Hashimoto, T. Sato, K. Hatanaka, J. Carbohydr. Chem. 24 (2005) 705–715. Y. Murozuka, M.C. Kasuya, M. Kobayashi, Y. Watanabe, T. Sato, K. Hatanaka, Chem. Biodivers 2 (2005) 1063–1078. T. Kato, R. Mitsumori, K. Hatanaka, Am J. Biochem. Biotechnol. 4 (2008) 385–392. L. Svennerholm, Biochim. Biophys. Acta 24 (1957) 604–611.
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u o r
Short Communication
Cellular glycosylation of amphiphilic saccharide primer in liquid/liquid interface culture system employing fluorous solvents Amane Sutoh, Maria Carmelita Z. Kasuya, Kenichi Hatanaka* Institute of Industrial Science, The University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8505, Japan
A R T I C L E I N F O
Article history: Received 14 April 2016 Received in revised form 20 June 2016 Accepted 21 June 2016 Available online 21 June 2016 Keywords: fluorous solvent liquid/liquid interface melanoma B16 cells saccharide primer ganglioside
A B S T R A C T
As fluorous compounds have extremely high oxygen-carrying capacity, liquid/liquid (culture medium/ fluorous solvent) interface culture system is one of their attractive applications in the field of biotechnology. However, it remains to be elucidated whether the metabolism of the cells cultured in such a system could work properly. To determine whether glycolipid metabolism of these cells are affected by such a low adhesive condition at the interface, we evaluated cellular glycosylation of a synthetic pseudoglycolipid by mouse melanoma B16 cells cultured in liquid/liquid interface system using three kinds of fluorous solvents. The results showed that the saccharide primer was sialylated by the cells when cultured at the interface using FluorinertTM FC-72 or perfluorodecalin, but not with dodecafluoroheptanol. The results of this work confirmed that glycolipid metabolism of cells cultured at the interface of aqueous/fluorous culture system functioned properly as well as normal aqueous culture system but depending on the kind of fluorous solvent used. Ganglioside biosynthesis was not compromised amidst an unusual environment such as the absence of a solid substrate for cell attachment and the presence of a perfluorinated solvent. ã 2016 Elsevier B.V. All rights reserved.
1. Introduction Fluorous solvents, i.e. highly fluorine-substituted liquid chemicals, have approximately 20 times higher O2 solubility than water [1] because of their relatively weak intermolecular interactions. Hence, they can be used as flexible O2 carriers to facilitate oxygen supply to cells or tissues [2]. As they are immiscible with water, several studies have addressed animal cell culture at the interface formed between fluorous solvent and overlying culture medium [3–6]. These studies have demonstrated that the anchoragedependent cells were able to grow and proliferate at the interface. Such culture system is expected to enhance the oxygen access to the cells and to overcome the poor oxygen supply, which is often considered to be a limiting factor of three-dimensional cell culture in vitro [5]. In addition, this simple culture method does not require trypsin treatment for subcultivation, thereby the adverse effects by trypsin [3] can be eliminated.
Abbreviations: DF-heptanol, dodecafluoroheptanol; PFD, perfluorodecalin; TIDF, DMEM/F12 containing Insulin-Transferrin-Selenium-X; HPTLC, high-performance thin layer chromatography. * Corresponding author. E-mail address: [email protected] (K. Hatanaka). http://dx.doi.org/10.1016/j.jfluchem.2016.06.011 0022-1139/ã 2016 Elsevier B.V. All rights reserved.
In previous studies [7], we evaluated morphology, growth and oxygen consumption of the cells cultured on the liquid/liquid interface, and reported that 2,2,3,3,4,4,5,5,6,6,7,7,-dodecafluoro-1heptanol (DF-heptanol) serves as an oxygen reservoir that could supply O2 to the culture medium. However, there is also an intriguing issue of whether metabolism of these cells are affected by such a low adhesive condition at the interface. Although some studies have reported on metabolic response of cultured cells to liquid/liquid interface system employing oxygenated perfluorocarbons [8], little is known about the effects of the system itself on cellular metabolism. In this research, we focused on oligosaccharide elongation, particularly biosynthesis of gangliosides by the cells at the liquid/ liquid interface. Gangliosides are sialic acid-containing glycosphingolipids, which are known to exist in the lipid raft on the surface of the plasma membrane and to be associated with a variety of biological phenomena, such as cell adhesion, signal transduction, malignant alteration of cancer cells, pathology of diseases, etc [9,10]. Unlike nucleic acids and proteins, they are not directly encoded in DNA and they are susceptible to the physiological conditions of the cells. Therefore, in order to elucidate the gap between genomic sequences and complicated biological phenomena, it is important to explore ganglioside behavior, which reflects physiological changes of the cells.
A. Sutoh et al. / Journal of Fluorine Chemistry 188 (2016) 76–79
We evaluated biosynthesis of ganglioside GM3 by mouse melanoma B16 cells at the culture medium/fluorous solvent interface using saccharide primer method. This method is an effective way to monitor biosynthesis of oligosaccharides without destroying the cells, in which the cells are incubated with saccharide primer, a synthetic pseudo-glycolipid, and the primers are taken into the cells and subsequently glycosylated by cellular glycosyltransferase [11,12]. Using this approach, previously we assessed the effects of some factors of cell culture conditions, e.g. xenobiotics [13] and the passage number of the cells [14], on cellular glycosylation. In this study, mouse melanoma B16 cells were cultured at the interface formed between fluorous solvents, such as DF-heptanol, FluorinertTM FC-72 (commercially available mixture of perfluorochemicals whose main constituent is perfluorohexane [15]) or perfluorodecalin (PFD), and culture medium containing dodecyl b-lactoside (Lac-C12) primer that mimics lactosylceramide, a precursor of a variety of glycolipids. How the cells behave in the presence of the saccharide primer during liquid/liquid interface culture was investigated and the effect of different fluorous solvents was compared. 2. Results 2.1. Morphology and viability of B16 cells at the liquid/liquid interface In order to investigate the effect of different fluorous solvents on cell morphology, B16 cells were observed using an optical microscope after 48 h of culture. Whereas the cells spread in polystyrene dish (Fig. 1A), the cells cultured in liquid/liquid system employing DF-heptanol remained rounded and existed in monolayer (Fig. 1B). As for the cells cultured with FC-72 or PFD, they were also rounded, but were grown as three-dimensional multicellular aggregates (Fig. 1C, D). From a macroscopic point of view (Fig. 1E), a sheet-like structure of the cells was observed when DF-heptanol was used. In addition, the medium overlying DF-heptanol was greatly blackened, which is considered to be due to melanin production. The number of the viable cells after 48 h for each condition is shown in Fig. 2. When FC-72 or PFD was used, there was no significant difference from control. By contrast, with DF-heptanol, the number of the viable cells was lower than control, indicating that DF-heptanol may not be suitable for this study because of its cytotoxicity. In the previous work using DF-heptanol as oxygen reservoir during biphasic culture [7], the cytotoxic effect of the fluorous solvent was not easily recognized because the medium used was supplemented with 10% fetal bovine serum (FBS) that promoted growth and proliferation. On the other hand, cell culture in the
77
presence of the saccharide primer as scaffold for glycosylation of cellular enzymes in this work required the use of culture medium without FBS to avoid the serum from interfering with ganglioside accumulation. This condition limited cell growth and proliferation. In both studies, cells thrived but the adverse effect of DF-heptanol on cell viability and its repercussion on glycosylation became more evident when compared with other fluorous solvents such as FC-72 and perfluorodecalin. 2.2. HPTLC analysis and identification of the sialylated product After incubation of B16 cells with Lac-C12 primer for 48 hours, the cells and the medium fractions were collected and analyzed by HPTLC. HPTLC results of the lipid extract from the cell homogenates and the culture medium are shown in Fig. 3A. The compounds containing sialic acid were visible as blue-violet bands [19]. In the presence of the Lac-C12 primer, there were new bands corresponding to a putative sialylated product, ganglioside GM3 analogue, in medium fraction when FC-72 or PFD was used (Lane 14, 15). Such a band could not be detected when DF-heptanol was used (Lane 13). It was shown that the primer was not taken into the cells cultured at medium/DF-heptanol interface (Lane 5). In order to confirm the structure of the products, they were scraped from the HPTLC plate, extracted with MeOH, and then identified from ESI-MS and enzyme assay. Mass spectrum of the product showed a peak at m/z 800.1. Fragmentation of the product showed peaks at m/z 508.9 and 289.7, corresponding to Lac-C12 primer and N-acetylneuraminic acid (NeuAc), respectively. Treatment with a2,3-sialidase showed that the product was hydrolyzed (Fig. 3B). Hence, it was confirmed that the product has a terminal a2-3 sialic acid moiety. From these results, the product was identified as a GM3 (NeuAc(a2-3)Lac-C12) analog. 2.3. Effects of the different fluorous solvents on the amount of sialylated primer The amounts of GM3 analog and endogenous GM3 in cell and medium fractions were quantitated by densitometric analysis (Fig. 4). The amounts of GM3 analog from the cells cultured in polystyrene flask, and liquid/liquid interface system using FC-72 and PFD were 11.87, 10.47 and 10.72 mmol/flask, respectively (SDs = 0.48, 1.80 and 1.76 mmol/flask, respectively). There were no significant differences among them. 3. Disucussion The present study demonstrated that Lac-C12 primer was sialylated by B16 cells when cultured at the interface using perfluoroalkanes, such as FC-72 or PFD, in the same way as when
Fig. 1. Morphology of B16 cells during (A) normal culture in polystyrene dish, and liquid/liquid interface culture employing (B) DF-heptanol, (C) FC-72 and (D) PFD. (E) The examples of liquid/liquid interface culture system after 48 h of culture, From left to right: normal culture, liquid/liquid interface culture employing DF-heptanol, FC-72, PFD.
78
A. Sutoh et al. / Journal of Fluorine Chemistry 188 (2016) 76–79
Fig. 2. The number of the viable cells after 48 h of incubation with/without Lac-C12 primer in liquid/liquid interface culture system. The data are presented as means SD (n = 3).
The findings of this research strongly support the usefulness of liquid/liquid interface culture system. Mouse melanoma B16 cells thrived at the interface between the culture medium layer and the fluorous solvent layer, and biosynthesis of glycolipids took place. This cell culture system has potential application to high-density culture of animal cells. By this method, the treatment of cells using the protease trypsin is not required for sub-cultivation and cells can be collected easily by pipetting and centrifugation. Moreover, the fluorous solvent can enhance oxygen supply to the cells. 4. Experimentation 4.1. Cell culture Mouse melanoma B16 cells were purchased from Riken Cell Bank (Tsukuba, Japan) and routinely maintained in Dulbecco’s Modified Eagle Medium: Nutrient Mixture F-12 (D-MEM/F-12; Gibco, Life Technologies Japan Ltd., Tokyo, Japan) supplemented with 10% FBS (JRH Biosciences, KS, USA) and 1% AntibioticAntimycotic solution (Nacalai Tesque, Kyoto, Japan). 4.2. Incubation of the cells with saccharide primer on the liquid/liquid interface
Fig. 3. HPTLC results. (A) Profile of lipids extracted from cell (Lanes 1-8) and culture medium (Lane 9-16) fractions after incubation of B16 cells with/without Lac-C12 primer. Lanes 1, 5, 9, 13: DF-heptanol; Lanes 2, 6, 10, 14: FC-72; Lanes 3, 7, 11, 15: PFD; Lanes 4, 8, 12, 16: control; Lane 17: standard endogenous GM3; Lane 18: Lac-C12 primer. (B) Profile before and after treatment with a 2,3-sialidase. Lane 1: medium fraction; Lanes 2, 3: the product without/with a 2,3-sialidase; Lane 4: Lac-C12 primer.
cultured on polystyrene dishes, but not with DF-heptanol. Also, the cells aggregated using perfluoroalkanes, but not in the case of DFheptanol. Although the cause for the difference requires further investigation, our data suggested that the cell membrane was adversely affected after exposure to fluorous alchohol and that uptake of the saccharide primer was inhibited. On the other hand, sialylation of the saccharide primer was not affected during liquid/ liquid interface culture employing perfluoroalkanes. This result indicates that all the steps of saccharide primer method, such as uptake of the primer, sialylation by sialyltransferase and release of the sialylated product functioned properly and this is the first report on ganglioside biosynthesis at the liquid/liquid interface.
Dodecyl b-lactoside (Lac-C12 primer) was synthesized as described previously [16]. In the saccharide primer method, serum-free medium TI-DF was used in order to avoid serum interfering with ganglioside accumulation [17]. TI-DF consists of D-MEM/F12 (20 mL) containing 1% Insulin-Transferrin-Selenium-X (ITS-X; Gibco) and 1% of Antibiotic-Antiomycotic solution. TI-DF (20 mL) was slowly overlaid onto three kinds of fluorous solvents (10 mL), such as 2,2,3,3,4,4,5,5,6,6,7,7,-dodecafluoro-1-heptanol (DF-heptanol; Tokyo Chemical Industry Co., Ltd., Tokyo, Japan), FluorinertTM FC-72 (3 M Japan Ltd., Tokyo, Japan) and perfluorodecalin (PFD; Tokyo Chemical Industry Co., Ltd.) in cell culture flasks (Nunc, Thermo Scientific Inc., MA, USA). B16 cells in dishes under subconfluent conditions were washed with TI-DF and harvested using 0.25% trypsin/EDTA. Then, 2 106 cells were seeded on the liquid/liquid interface with 50 mM of Lac-C12 primer. The flasks were incubated in a humidified atmosphere of 5% CO2 air at 37 C for 48 h. As control experiments, the cells were also cultured in TI-DF only with Lac-C12 primer. Besides, the cells were cultured at the interface between fluorous solvent and TI-DF without Lac-C12 primer. All the samples were prepared in triplicate.
A. Sutoh et al. / Journal of Fluorine Chemistry 188 (2016) 76–79
79
with resorcinol-HCl-Cu2+ reagent and heated at 120 C for 15 min. The densitometric analysis of the separated glycolipid bands were performed using ImageJ software (NIH, MD, USA). For each HPTLC plate, a standard curve was generated from the bands of reference ganglioside GM3 (Sigma-Aldrich, MO, USA) at various concentrations, and the amount of sialylated compounds in each lane was calculated. 4.5. Identification of glycosylated products ESI-MS analysis of the glycosylated products was performed on HCTultra (Bruker Daltonics, Bremen, Germany). Enzymatic digestion of the glycosylated products was carried out in 100 ml of AcONa buffer (pH 5.5) containing 0.5 mU of a2,3-sialidase from salmonella typhimurium LC2 (Takara Bio Inc., Shiga, Japan). The reaction was performed overnight at 37 C and then analyzed by HPTLC. Fig. 4. Amounts of sialylated primer and endogenous GM3 calculated from HPTLC (n = 3).
4.3. Lipid extraction After incubation for 48 h, the lower fluorous layer was removed, the cells were separated from the medium fraction by centrifugation and washed with PBS( ). To estimate the cell viability, the cells were stained with trypan blue and counted using Countess Automated Cell Counter (Invitrogen, Life Technologies, CA, USA). The lipids were extracted from the cell pellet with CHCl3-MeOH (2:1, v/v), then with CHCl3-2-propanol-water (7:11:2, v/v/v) in a sonicated bath. Lipids from the medium fraction were purified using reversed-phase SPE cartridge Sep-Pak C18 (Waters Corp., MA, USA) followed by anion-exchange SPE cartridge Inert-Sep SAX (GL Sciences Inc., Tokyo, Japan) [18]. 4.4. HPTLC analysis
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]
The products obtained from the culture medium fraction and the cell fraction were analyzed by high-performance thin layer chromatography (HPTLC Silica Gel 60 F-254; E. Merck KG, Darmstadt, Germany) with CHCl3-MeOH-0.25% aq. KCl solution (5:4:1, v/v/v) as developing solvent. The HPTLC plates were sprayed
[17] [18] [19]
K.C. Lowe, J. Fluor, Chem. 118 (2002) 19–26. M. Pilarek, Chem. Process Eng 35 (2014) 463–487. Y. Shiba, T. Oshima, M. Sato, Biotechnol. Bioeng. 57 (1998) 583–589. Y.J. Kwon, H. Yu, C.-A. Peng, Biotechnol. Bioeng. 72 (2001) 331–338. M. Pilarek, I. Grabowska, M.A. Ciemerych, K. Da˛bkowska, K.W. Szewczyk, Biotechnol. Lett. 35 (2013) 1387–1394. ski, J. Janicka, K. Janczyk-Ilach, M. Pilarek, I. Grabowska, I. senderek, M. Wojasin T. Ciach, Bioprocess Biosyst. Eng. 37 (2014) 1707–1715. H. Miyajima, M.C. Kasuya, K. Hatanaka, J. Fluor, Chem. 163 (2014) 46–49. J. Wardrop, M.R. Davey, J.B. Power, K.C. Lowe, Artif. Cells Blood Substit. Immobil. Biotechnol. 30 (2002) 63–70. S.-I. Hakomori, Curr. Opin. Hematol. 10 (2003) 16–24. T. Ariga, M.P. McDonald, R.K. Yu, J. Lipid, Res. 49 (2008) 1157–1175. Y. Miura, T. Yamagata, Biochem. Biophys. Res. Commun. 241 (1997) 698–703. T. Sato, K. Hatanaka, H. Hashimoto, T. Yamagata, Trends Glycosci. Glyc. 19 (2007) 1–17. T. Kato, M. Muraoka, K. Hatanaka, Anal. Biochem. 405 (2010) 103–108. Y. Shimura, J. Suzuki, M.C.Z. Kasuya, K. Matsuoka, K. Hatanaka, J. Biosci Bioeng, 114 (2012) 552–555. P. Adlercreutz, B. Mattiasson, J. Eur, Appl. Microbiol. Biotechnol. 16 (1982) 165–170. M.C.Z. Kasuya, M. Ikeda, K. Hashimoto, T. Sato, K. Hatanaka, J. Carbohydr. Chem. 24 (2005) 705–715. Y. Murozuka, M.C. Kasuya, M. Kobayashi, Y. Watanabe, T. Sato, K. Hatanaka, Chem. Biodivers 2 (2005) 1063–1078. T. Kato, R. Mitsumori, K. Hatanaka, Am J. Biochem. Biotechnol. 4 (2008) 385–392. L. Svennerholm, Biochim. Biophys. Acta 24 (1957) 604–611.