Glycobiology Advance Access originally published online on March 23, 2005
Glycobiology 2005 15(7):687-699; doi:10.1093/glycob/cwi055
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Role for up-regulated ganglioside biosynthesis and association of Src family kinases with microdomains in retinoic acid-induced differentiation of F9 embryonal carcinoma cells
3 Department of Biomembrane and Biofunctional Chemistry, Hokkaido University, Kita 21-Nishi 10, Kita-ku, Sapporo 001-0021, Japan; 4 Division of Project Research, Creative Research Initiative "Sousei" Hokkaido University, Kita 21-Nishi 10, Kita-ku, Sapporo 001-0021, Hokkaido, Japan; and 5 Biomembrane Research Group, Tokyo Metropolitan Institute of Gerontology, Tokyo 173-0015, Japan; 6 Core Research for Evaluational Science and Technology program (CREST), Japan Science and Technology Corporation (JST), Graduate School of Pharmaceutical Sciences, Frontier Research Center for Post-Genomic Science and Technology, Kita 21-Nishi 10, Kita-ku, Sapporo 001-0021, Japan
1 These authors contributed equally to this work.
2 To whom correspondence should be addressed; e-mail: inokuchi{at}kinou02.pharm.hokudai.ac.jp
Received on December 15, 2005; revised on March 15, 2005; accepted on March 16, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Mouse F9 embryonal carcinoma cells have been widely used as a model for studying the mechanism of embryonic differentiation, because they are similar to the inner cell mass of early mouse embryos and can differentiate into primitive endoderm (PrE) following retinoic acid (RA) treatment. During F9 cell differentiation, the carbohydrate chains of glycoproteins and their corresponding glycosyltransferases are known to undergo rapid changes. However, there have been no corresponding reports on the expression of gangliosides. We have developed a custom cDNA array that is highly sensitive for the genes responsible for sphingolipid (SL) biosynthesis and metabolism. Using this, we found that, of the 28 selected genes, 26 exhibited increased expression during F9 differentiation into PrE. Although neutral glycosphingolipids (GSLs) were expressed at similar levels before and after differentiation, a greater than 20-fold increase in total ganglioside content was evident in PrE. Glucosylceramide synthase inhibitors (D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol [D-PDMP] and its analog) depleted gangliosides and this resulted in delayed expression of Disabled-2 (Dab-2), suggesting the involvement of gangliosides in F9 cell differentiation. Disruption of cholesterol-enriched membrane microdomains by methyl-ß-cyclodextrin (MßCD) also delayed differentiation. Both MßCD and D-PDMP blocked the accumulation of Src family kinases (SFKs) to microdomains. However, D-PDMP did not block flotillin accumulation, yet MßCD did. Additionally, confocal laser microscopy revealed the formation of distinct functional microdomains integrating SFKs with gangliosides and cholesterol during PrE differentiation. Thus, we demonstrate the outstanding up-regulation of ganglioside biosynthesis and its importance in the formation of distinct microdomains incorporating SFKs with gangliosides during RA-induced differentiation of F9 cells.
Key words: differentiation / F9 embryonal carcinoma cell / gangliosides / microdomains / Src family kinsases
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Gangliosides, sialic acid-containing glycosphingolipids (GSLs), are found on the surface of the plasma membrane of all vertebrate cells and are involved in signal transduction, cell growth, and differentiation (Hakomori, 1981
). In the human promyelocytic leukemia cell line HL-60, exogenous GM3 induces differentiation into macrophage-like cells (Senoo and Momoi, 1985), while neolacto-series gangliosides induce differentiation into mature granulocytes (Nojiri et al., 1988). Recent approaches using the targeted disruption of genes active in the GSL biosynthesis pathways (Figure 1) have yielded some clues into the biological function of gangliosides. For instance, disruption of the gene for glucosylceramide synthase (GlcT), which catalyzes the addition of the initial glucose to ceramide, has shown that GSL synthesis is essential for development and differentiation in vivo (Yamashita et al., 1999). GlcT knockout mice die during midgastrulation (
embryonic day 7.5) by a major apoptotic process centered in the ectoderm (Yamashita et al., 1999). Moreover, GM2 synthase (GalNAcT) and GD3 synthase (GD3S) double knockout mice, which express only the ganglioside GM3, show refractory skin injury (Inoue et al., 2002), or exhibit lethal audiogenic seizures (Kawai et al., 2001), illustrating the fundamental importance of gangliosides in life.
|
|
Cellular membranes contain subdomains, which are called detergent resistant microdomains (DRMs) because they are detergent-insoluble and highly enriched in cholesterol and sphingolipids (SL) (GSLs and sphingomyelin) (Hakomori, 2000
; Simons and Toomre, 2000). Within the past decade, data have emerged from many laboratories implicating these lipid microdomains as critical for the proper compartmentalization of cellular signaling. In particular, DRMs are known to contain various signaling molecules such as glycosylphosphatidylinositol-anchored proteins and nonreceptor type protein kinases including Src family protein kinases (SFKs) (Simons and Ikonen, 1997; Varma and Mayor, 1998; Brown and London, 2000; Simons and Toomre, 2000). Additionally, cholesterol is thought to fill the space between the hydrocarbon chains of the SLs and to function as the glue that keeps the microdomain assembly together (Simons and Toomre, 2000). In fact, cholesterol depletion from cell membranes, using drugs such as methyl-ß-cyclodextrin (MßCD), generally results in a disruption of microdomain-mediated cellular functions (Nagafuku et al., 2003). Gangliosides, too, are important components of DRMs, and biological effects of gangliosides may, in fact, work through DRMs. However, detailed mechanisms of such actions remain to be investigated.
Mouse F9 embryonal carcinoma cells, which are derived from an induced murine teratocarcinoma, are widely used as model systems for studies on the early developmental stages of mammals. F9 cells can differentiate into primitive endoderm (PrE) in the presence of retinoic acid (RA) (Strickland and Mahdavi, 1978). During this differentiation, an apoptotic response and a dramatic decrease in the rate of proliferation occur (Sleigh, 1992; Atencia et al., 1994). Additionally, PrE cells become positive for several specific differentiation markers, such as tissue plasminogen activator, Type IV collagen, c-jun, cytokeratin ENDO A, Disabled-2 (Dab-2)/DOC-2, and follistatin (Strickland and Mahdavi, 1978; Duprey et al., 1985; Rickles et al., 1988; Bjersing et al., 1997; Cho et al., 1999). Regulation of genes responsible for glycoprotein synthesis (Cummings and Mattox, 1988) and sphingosine (Sph) 1-phosphate lyase (Kihara et al., 2003) during this differentiation has also been investigated; however, there have been no reports on ganglioside regulation. This may be because of the fact that levels of expression are very low in undifferentiated F9 cells.
Here, we show for the first time, in F9 cells treated with RA, marked increases in the genes responsible for SL biosynthesis and metabolism. Additionally, we found a greater than 20-fold increase in the total amounts of gangliosides in PrE. Using this model for differentiation, we demonstrate the importance of the formation of distinct microdomains that integrate SFKs with gangliosides and cholesterol during RA-induced differentiation of F9 cells.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
The expression of sphingolipid-related genes increases during RA-induced differentiation of F9 cells
In this study, our interest was focused on genes responsible for synthesizing and metabolizing SLs, especially gangliosides. The SL array included 29 probes to detect genes involved in the biosynthetic cascade that begins with a serine palmitoyl transferase or catabolic cascade (Table I). As shown in Figure 2, data from the SL array analysis revealed that most of the selected genes were up-regulated in the presence of RA, except for the Gb3 synthase gene (gene# 13), which was down-regulated, and the glucosylceramide (GlcCer) synthase gene (gene# 3), which remained unchanged. The genes involved in the synthesis, degradation, and modification of Sph, ceramide, and sphingomyelin, were all enhanced, including LCB1 (gene# 1), LCB2 (gene# 3), Sph kinases (gene# 16, 17, and 18), Sph-1 phosphate lyase (gene# 22), ceramidases (gene# 19 and 20), sphingomyelinases (gene# 23 and 24), and SPH-1P phosphatase (gene# 21). Such comprehensive changes suggest an acceleration of SL metabolism in the presence of RA.
|
Lactosylceramide (LacCer) acts as a precursor for a wide range of GSLs including the gangliosides. The LacCer synthase gene (gene# 5) and a possible candidate of LacCer synthase gene (gene# 4) were increased several-fold. In the ganglioside biosynthetic cascade, we examined 6 genes (gene# 6–11) responsible for the formation of a- and b-series gangliosides. In undifferentiated F9 cells, the expression of GM3 synthase (gene# 6) was very low and there were no detectable signals for GD3 synthase (gene# 7) or GQ1b synthase (gene# 11). On the other hand, the PrE F9 cells expressed these genes at high levels; however, catabolizing enzaymes series, glucocerebroside (gene# 25) and neuraminidase (gene# 27), were all up-regulated.
Comparison of SL expression between undifferentiated and differentiated F9 cells
In light of the data from the SL array analysis (Figure 2), the up-regulation of ganglioside biosynthesis in PrE cells was expected. Therefore, we first compared the gangliosides composition before and after differentiation, which was confirmed by the expression of the specific differentiation marker Dab-2 (Strickland and Mahdavi, 1978) 3 days after RA treatment (Figure 3A). Thin-layer chromatography (TLC) analysis of gangliosides revealed that undifferentiated F9 cells expressed detectable amounts of only GD1a, with no b-series gangliosides. However, following RA treatment, detectable bands appeared for GM3, GM2, GD3, GD1b, GT1b, and GQ1b (Figure 3B). The increase in total ganglioside levels in response to RA treatment was evident as early as Day 1, preceding the Dab-2 expression. Moreover, the ganglioside levels were elevated more than 20-fold at Day 4 (Figure 3C). However, the data from the SL array analysis revealed that the expression of neuraminidases in RA-treated cell increased (Figure 2), suggesting that the catabolism of gangliosides was up-regulated. Because ganglioside levels were greatly increased in RA/F9 cells, it is strongly suggested that the biosynthetic cascade of gangliosides is more activated than the catabolic cascade. The expression levels of GlcCer synthase gene were unchanged but those of glucocerebrosidase synthase gene were enhanced (Figure 2), suggesting increased degradation of GlcCer. However, the contents of GlcCer and LacCer slightly increased (1.3-fold) at Day 4 (Figure 3D). These results probably due to the increase of serine palmitoyl transferase activity, resulting in the enhanced supply of sphinganine, the precursor of all SLs and the GSLs (Figure 4). Sphingomyelin and cholesterol, which are involved in the formation of the microdomains, together with GSLs, increased about 1.4-fold and 2-fold, respectively at Day 4 (Figure 3E).
|
|
Gangliosides depletion delays RA-induced differentiation of F9 cells
We employed two inhibitors of glucosylceramide synthase (D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol [D-PDMP] and D-threo-1-phenyl-2-benzyloxycarbonylamino-3-pyrroridino-1-propanol [D-PBPP]) to deplete cellular gangliosides in F9 cells undergoing differentiation by RA. Both inhibitors were able to block the expression of gangliosides (Figure 5A). The morphology of the RA treated cells exhibited weakened adhesive ability in the presence of the inhibitors (Figure 5B). Moreover, the expression levels of Dab-2 at Day 3 and Day 4 were suppressed in these cells (Figure 5C). These data suggested the involvement of glycolipids, probably gangliosides in RA-induced differentiation of F9 cells.
|
To confirm whether the suppression of the differentiation by the inhibitors was due to a general and nonspecific suppression of cell growth, we exposed the cells to [3H]thymidine. The incorporation of thymidine, in either the presence or absence of RA, was not affected by these inhibitors (Figure 5D).
Lipid microdomains plays an important role in differentiation
It is known that gangliosides, as well as cholesterol and sphingomyelin, are fundamental architectural lipid components of membrane microdomains. The extraction of cholesterol from plasma membranes by the drug MßCD has been shown to result in the destruction of DRMs (Nagafuku et al., 2003). We examined the effect of this drug and the resulting cholesterol loss in RA-treated F9 cells. We decided the conditions for the MßCD treatment (3mM for 60 min), which decreased the cholesterol of the cells and did not suppress the growth of the cells, to exclude the possibility that the suppression of cell growth by MßCD might cause a nonspecific delay in the differentiation. The addition of MßCD reduced the cholesterol content (Figure 6B) and suppressed the expression of Dab-2 (Figure 6A), but did not affect the cell growth (Figure 6C). These results suggest that it takes at least 3 days for a microdomain to recover its function after depletion of cholesterol by MßCD because MßCD treatment caused the delay of F9 differentiation at Day 3 (Figure 6A), and in the process of F9 differentiation, microdomains play an important role.
|
Next, we compared functional proteins known to be associated with microdomains before and after differentiation. Cells were lysed in a buffer containing 0.1% Triton X-100, and the lysates were subjected to sucrose density centrifugation. Samples were separated into 12 fractions, from the top of the gradients, and subjected to western blotting using anti-SFK and antiflotillin antibodies; in addition, we confirmed that DRM isolation went well using an antibody for the transferrin receptor, which is nonraft protein (Figure 7C). As shown in Figure 7A, most of the SFKs were dispersed in the non-DRM fractions in undifferentiated F9 cells. However, upon RA treatment, the SFKs were translocated into the DRMs. Additionally, the expression of flotillin and its localization to DRMs was greatly increased in the RA/cells (Figure 7B). Both MßCD and D-PDMP blocked the localization of the SFKs to the DRMs (Figure 7A). However, flotillin localization in DRMs was not affected by D-PDMP but was inhibited by MßCD (Figure 7B). These results suggest that SFKs interact with gangliosides. Therefore, we investigated the interaction between SFKs and gangliosides by coimmunoprecipitation experiments. As shown in Figure 8, not control IgG but anti-SFKs antibody immunoprecipitated ganglioside GM1, indicating the association of SFKs and gangliosides. Moreover, flotillin, which translocated into DRMs independent of gangliosides, showed no association with ganglioside GM1.
|
|
An accumulation of cholesterol in the DRMs was observed in RA/F9 cells (Figure 7C), which was because of the increase of total cholesterol level (Figure 3D). To determine whether MßCD treatment affects the localization of gangliosides in the DRMs of RA/F9 cells, lipid extracts of the 12 fractions were examined for their reactivity to fluorescein-isothiocyamate (FITC)-labeled cholera toxin which binds to GM1. As shown in Figure 7D, MßCD did block the accumulation of gangliosides in the DRMs.
To confirm the results of DRMs analysis indicating the translocation of SFKs to microdomain in RA induced differentiation (Figure 7A), we performed a confocal laser microscopic analysis. GM1, visualized by staining with FITC-conjugated cholera toxin, was greatly increased after RA-induced differentiation (Figure 9A). This result correlated well with the TLC analysis (Figure 3A). To confirm that cholera toxin stained not intracellular but surface membrane, cells were treated with or without Triton X-100. As shown in Figure 9A, there is no difference on GM1 staining between cells with or without permeabilization. The spots stained by cholera toxin colocalized with the spots stained by biotinylated 
-toxin (BC), which selectively bind cholesterol-rich microdomains (Waheed et al., 2001), in the RA/F9 cells without permeabilization (Figure 9B, upper panel) and because our purpose in Figure 9 was to examine that SFKs translocate into lipid microdomains, we used BC. However, the extraction of cholesterol from plasma membranes with MßCD resulted in the marked decrease of BC reactivity with dispersed staining of GM1 (Figure 9B, lower panel). Colocalization of SFKs and cholesterol was examined by immunofluorescence and was observed in RA/F9 cells but not in undifferentiated cells (Figure 9C). And colocalization of SFKs and GM1 was also observed in RA/F9 cells (Figure 9D). Moreover, corresponding to Figure 7A, MßCD treatments did block the colocalization of SFKs and cholesterol (Figure 9C third lane panel), cholesterol and GM1 (Figure 9B, lower panel), or GM1 and SFKs (Figure 9D, middle panel). On the other hand, D-PDMP treatments interestingly dispersed staining of cholesterol and did block the colocalization of SFKs and cholesterol similar to MßCD treatments (Figure 9C, fourth lane panel). Regarding anti-SFKs antibody, we got the same data using the antibody that specifically react to c-Src, which are thought to gather into DRMs weakly (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
In this study, we developed a custom cDNA array that is highly sensitive to genes’ coding enzymes active in the biosynthesis and metabolism of SLs. We focused especially on the ganglioside related genes and investigated their expression levels during RA-induced differentiation of F9 cells into PrE (Table I and Figure 2). This novel, comprehensive analysis of SL related genes revealed that 27 of 29 genes increased during the PrE differentiation. In undifferentiated F9 cells, the expression of GM3 synthase was low, and there were no detectable signals for GD3 synthase or GQ1b synthase, correctly predicting an absence of b-series gangliosides. On the other hand, the differentiated F9 cells expressed genes responsible for the biosynthesis of a- and b-series gangliosides at high levels. The information from the SL array was supported by TLC analyses of GSLs, which exhibited an approximately 24-fold increase in total gangliosides during the differentiation (Figure 3). The expression pattern of gangliosides is known to change during the differentiation of P19 cells, another embryonal carcinoma cell line (Liour et al., 2000
; Osanai et al., 2003). Undifferentiated P19 cells, before RA treatment, express b-series gangliosides, such as GD3 and GT1b (Liour et al., 2000; Osanai et al., 2003), but F9 cells expressed only negligible levels of GD1a with no detectable amounts of the b-series gangliosides (Figure 3B). We thought that these differences were because of the F9 cell line originated from a 6 day embryo (Stevens, 1970) and the P19 cell line from a 7.5 day embryo (McBurney and Rogers, 1982), because early embryos express no gangliosides (Rahmann et al., 1994). However, mouse embryonic stem (ES) cells are more similar to P19 cells than to F9 cells (Minucci et al., 1996). Moreover the mechanism of F9 differentiation induced by RA is still not known in detail. Therefore, these differences might be because of the distinct characteristics between carcinoma cell lines. Nevertheless we consider that F9–RA model system is very useful to reveal the function of gangliosides because the outstanding up-regulation of ganglioside is observed during F9 differentiation induced by RA. A previous study, using cDNA microarrays, reported that only 5% of the genes among 8,900 genes were significantly induced during F9 cell differentiation (Harris and Childs, 2002). It is notable that the up-regulation of more than 90% genes responsible for SL biosynthesis and metabolism was observed.
To probe the roles of the increased gangliosides in PrE, we first used inhibitors of glucosylceramide synthase (D-PDMP and D-PBPP) to deplete the cellular gangliosides. Both inhibitors delayed RA-induced differentiation (Figure 5), suggesting that glycolipids, probably gangliosides, are involved in the differentiation. However, the inhibition of glucosylceramide synthesis should have suppressed also the synthesis of neutral glycolipids such as globo-series glycolipids. Thus, in this study we cannot discuss about the importance of neutral glycolipids; among them the expression of Forssman and SSEA-3 glycolipids are known to be changed during embryonal carcinoma (EC) cell differentiation (Willison et al., 1982; Damjanov et al., 1994).
Recent studies report that Dab-2 involves various signal transductions. For example, Dab-2 binds to the SH3 domains of Grab2 (Xu et al., 1998) and Dab-2 associates with myosin VI (Morris et al., 2002). Regarding to differentiation, Dab-2 is a negative regulator of Wnt signaling (Hocevar et al., 2003), and a blockade in Wnt signaling activates the differentiation of F9 cells (Shibamoto et al., 2004). Regarding to cell adhesion, Huang et al. reported that Dab-2 is a negative regulator of integrin alpha II 6 beta 3 that mediates cell adhesion (Huang et al., 2004). However, in F9 cells, regardless to the expression of Dab-2, the cell adhesion and aggregation are induced during the differentiation, suggesting the involvement of the other cell adhesion process.
Because gangliosides probably function through membrane microdomains (Iwabuchi et al., 1998b), we performed several critical experiments examining the involvement of microdomains in F9 differentiation. First, we found that the disruption of microdomains by MßCD delayed RA-induced differentiation (Figure 6). Total cholesterol levels did not decrease by MßCD very much (Figure 6B); however, the effectiveness of MßCD to deplete the cell surface cholesterol was confirmed by confocal microscopy using BC (Figure 9B). Second, as demonstrated in Figures 7 and 9, we found that, although SFKs in undifferentiated F9 cells were dispersed and did not colocalize with cholesterol, in F9 cells treated with RA, SFKs became concentrated in the DRMs and colocalized with cholesterol and GM1. Moreover, we could demonstrate here that the direct association between gangliosides and SFKs by coimmunoprecipitation (Figure 8). Third, we also revealed that the expression of flotillin increased during differentiation of F9 cells, and its accumulation to the DRMs was evident (Figure 7B). Finally, we considered that the increased gangliosides might cause the translocation of SFKs into the microdomains. Indeed, not only MßCD treatment but also ganglioside depletion by D-PDMP decreased SFK translocation into the microdomains in RA/F9 cells (Figure 7A) indicating the association of SFKs and gangliosides (Figure 8). In contrast, MßCD decreased the translocation of flotillin to the DRMs, but D-PDMP did not (Figure 7B), indicating that cholesterol, but not the gangliosides, is important for the association of flotillin with the microdomains. In bone marrow cells flotillin reportedly increases during osteoclast differentiation (Ha et al., 2003), but there has been no report regarding flotillin in F9 cells. This suggests a need for further investigation.
Despite the expression of cholesterol and neutral GSLs in undifferentiated F9 cells, only small amounts of SFKs were translocated into the DRMs (Figure 7A). Corresponding to these observations, SFKs and cholesterol did not colocalize in undifferentiated F9 cells, but did in PrE cells (Figure 9C). So, we conclude that neither cholesterol nor neutral GSLs are important in recruiting and integrating SFKs into microdomains, but, rather, it is the gangliosides that have significance in this role. Supporting these observations, we previously demonstrated that in 3LL mouse lung carcinoma cells, depletion of GSLs by D-PDMP resulted in the elimination of SFKs from the DRMs (Inokuchi et al., 2000). In rat brain an association of the Src family tyrosine kinase Lyn with the ganglioside GD3 has been reported (Kasahara et al., 1997), and GPI-anchored proteins, such as TAG-1, might contribute to this association (Kasahara et al., 2002). The approximate size of ganglioside-enriched microdomains are 500–900 nm, using choleratoxin subunit B to GM1 (Figure 9). A few recent studies (Munro, 2003; Glebov and Nichols, 2004) indicate that typical cholesterol-dependent microdomains termed "lipid raft" having GPI anchor, assumed to have a diameter of 10–20 nm, may be too small to be involved in signal transduction in contrast to major microdomains having sufficient size and organization of molecules. If "lipid raft" is defined by (1) resistance to 1% Triton X-100, (2) cholesterol-dependence, and (3) small size (10–20 nm diameter or less), then most of the ganglioside microdomains as described in this study are "nonlipid raft" because their size is presumed to be much larger, in analogy to glycolipid-enriched microdomains.
Our findings revealed a dramatic increase in various gangliosides in the RA–F9 differentiation system, and it will be most interesting to identify specific interactions and associations between each ganglioside and signaling molecule using this system. Moreover, allowing for the possibility of multiple microdomains, we must look more carefully into distinct functions within the microdomains, perhaps, those dependent on gangliosides and others on cholesterol. Conventional approaches, including confocal microscopy and immunoprecipitation using monoclonal antibodies which specifically recognize each ganglioside species, will be needed to thoroughly explore this issue. For instance, in mouse melanoma B16 cells a GSL signaling domain was separated from a caveolin-containing membrane fraction using an anti-GM3 monoclonal antibody (Iwabuchi et al., 1998a). Consistent with this, we showed for the first time here enlarging microdomains dependent on the increases in gangliosides during the RA-induced differentiation of F9 cells (Figure 9).
We are currently performing two approaches to identify specific functional proteins that directly bind to gangliosides to elucidate the detailed function of gangliosides in microdomains. One approach is a comprehensive analysis of molecules including SFKs that are selectively translocated into microdomains in response to ganglioside expression during differentiation by immunoprecipitation with antibodies selectively reacting to each ganglioside species, followed by two-dimensional electrophoresis and mass spectrometry (glycoproteomics). The second approach, once we find the proteins which interact, is microscopy analysis for microdomains using fluorescence correlation spectroscopy (FCS) and total internal reflection fluorescence microscopy (TIRFM) to obtain the direct evidence of ganglioside–protein interaction in living cells. We anticipate the advancement of practical and mechanistic understanding of ganglioside functions with these approaches.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Cell culture
Mouse F9 embryonal carcinoma cells were seeded on 0.1% gelatin-coated dishes and maintained in Dulbecco’s modified Eagle’s medium (D6429; Sigma, St. Louis, MO) containing 10% fetal bovine serum, 100 units/mL penicillin and 100 µg/mL streptomycin. The cells were induced to differentiate into PrE cells by the addition of 10 µM all trans-RA (Sigma) to the medium. Cellular GSLs were depleted using the glucosylceramide synthase inhibitors, D-PDMP (Inokuchi and Radin, 1987
) and D-PBPP (Jimbo et al., 2000). Each inhibitor was dissolved in H2O to make a 4 mM stock solution. F9 cells were treated with 20 µM D-PDMP or D-PBPP for the specified interval as described in each figure.
Sphingolipid DNA array (SL array)
Preparation of DNA array filters
To study the expression pattern of major SL-related genes in various cells and tissues with sufficient sensitivity and accuracy, we designed a nylon membrane based, custom SL array membrane. Gene probes were constructed using specific primers as indicated in Table I. Total RNA was collected from various mouse cell lines using an mRNA isolation kit (Trizol; Gibco Brl, Grand Island, NY) according to the manufacturer’s instructions. Total RNA was converted to cDNA using a reverse transcriptase, an oligo dT primer and random primers (8-mer). Polymerase chain reaction (PCR) was performed, using specific primers and cDNA as a template to generate fragments with a terminal adenine base. These DNA fragments were ligated to pGEM–T vector (Promega, Madison, WI) by T–A cloning using Ligation High (Toyobo Co., Osaka city, Japan). Ligated products were transfected into Escherichia coli (XL-1 Blue strain and J109) and plated on LB agar plates containing X-Gal, isopropyl-beta-D-thiogalactopyranoside (IPTG), and ampicillin.
After confirming the presence of inserted DNA, the colonies were propagated in LB liquid medium containing ampicillin. Plasmid DNA was purified from overnight cultures using BioRad’s quantum mini prep kit. The final concentration was determined using a spectrophotometer at A260. Plasmids containing different genes were used as templates and amplified by PCR using "KOD Dash" DNA polymerase (Toyobo Co.) and corresponding specific primers (Table I). Amplified PCR products were purified by Qiagen’s qiaquick PCR purification kit to remove enzyme, primers, free nucleic acids, and salts. The purified products were spotted on a positively charged nylon membrane using a mechanical spotting device. After spotting, the membrane was baked at 170°C to dry the membrane and denature the double stranded probe, followed by cross-linking using an ultraviolet (UV) cross linker.
The expression of SL-related genes are very low, so we had to first determine the optimum probe concentration which would maximally bind to the biotinylated samples and produce significantly detectable and measurable signals. The optimum concentration was determined by spotting different amounts of the DNA probes (5 ng, 20 ng, 25 ng, 40 ng, and 50 ng). We found that the most reliable results could be obtained using 20–40 ng per spot. ß-actin, G3PDH, and 
-tubulin G3PDH DNA (0.5 ng) were also spotted as house-keeping genes.
Detection
Total RNA from cells to be analyzed was freed from DNA by DNAse I treatment. The mRNA was isolated using magnetic beads (Toyobo Co.) that bind poly-A tailed mRNA. The purified mRNAs were converted by reverse transcription to double stranded cDNA, which was confirmed by PCR to carry the housekeeping genes. A short tail of deoxycytosine (dC) was attached to the 3' terminal of the cDNA. Finally, the poly-dC tailed cDNA was biotinylated using 16-UDP-biotin.
The membrane was hybridized for 12 h at 68°C with the biotinylated cDNA samples, using the commercial hybridization buffer Perfect Hyb, (Toyobo Co.). After hybridization, the membrane was washed with 2x SSC (3M sodium chloride, 0.3 M sodium citrate, pH 7) containing 0.1% sodium dodecyl sulphate (SDS), 3 times for 5 min each at 68°C, and then with 0.1x SSC/0.1% SDS, 3 times for 5 min each at 68°C. Finally, the membrane was brought to room temperature, blocked with blocking solution, and reacted with streptavidin alkaline phosphatase. The membrane was then soaked with dioxetane solution (Toyobo Co.) and incubated in the absence of light for 30 min. The developed membrane was scanned using a charge coupled digital (CCD) camera (LAS-3000, Fujifilm, Minamiashigara City, Japan) for 15–30 min. The image was analyzed using the image analysis software array gauge (Fujifilm).
Lipid analysis
Cells were washed twice with phosphate buffered saline (PBS), and lipids were extracted from the cell with chloroform/methanol (1:1 and 1:2 v/v, successively). Samples were equalized for protein concentrations, which were determined by a BCA kit (Pierce Chemical Company, Rockford, IL). For cholesterol analysis, lipids were separated on silica gel TLC plates (Merck, Darmstadt, Germany), with hexsane/diethyl ether/acetic acid (80:30:10) and detected by heating with cupric phosphoric acid. For sphingomyelin analysis the lipids were separated with chloroform/methanol/acetic acid/formic acid/water (48:12:7.2:2.4/1.2).
For neutral glycolipid analysis, the total lipid extract was dissolved in chloroform/methanol/water (30:60:8) and passed through a diethylaminoethyl cellulose-Sephadex A-25 column (0.8 x 4.5 cm; acetate form). Bound lipids were eluted with five volumes of chloroform/methanol/water (30:60:8), and solvents were completely evaporated. Esters were cleaved with methanolic 0.5 M NaOH for 1 h at 40°C. The solution was neutralized with 1 M acetic acid in methanol and diluted with 6 mL of 50 mM NaCl solution, then applied to a Sep–Pak C18 reverse-phase cartridge (Waters, Milford, Massachusetts). The cartridge was washed with 40 mL of water, and lipids were eluted with 10 mL of methanol and 10 mL of chloroform/methanol (1:1), successively. Neutral GSLs were separated by TLC using chloroform/methanol/12mM MgCl2 (65:25:4) and detected as above. For analysis of acidic glycolipids including gangliosides, the total lipid extract was further purified, separated by TLC using chloroform/methanol/0.2% CaCl2 (55:45:10), and then detected with orcinol-sulfuric acid reagent. The quantity of each lipid was measured with a dual-wavelength flying spot scanner (CS9300-PC, Shimadzu, Kyoto, Japan) in the reflectance mode, at 500 nm. For GM1 analysis from samples of flotation assay by sucrose density gradient, each fraction (10 µg protein) was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE) and transferred to polyvinylidene difluoride (PVDF) membrane (Immobilon Millipore Corp., Bedford, MA), followed by immunoblotting with cholera toxin B subunit-peroxidase conjugate (Sigma).
Serine palmitoyl transferase assay
Cells were lysed in buffer contained 50 mM 4-(2-hydroxyethyl)-1-piperazineethanesultonic acid (pH 7.5), 5 mM dithiothreitol, 50 mM ethylene diamine tetraacetic acid (EDTA; pH 8.0), 1 mM phenylmethylsulfonyl fluoride, 1x protease inhibitor mixture (CompleteTM; Nissui, Tokyo, Japan), while the reaction mixture contained 50 µM pyridoxal 5-phosphate, 50 µM palmitoyl CoA, 0.1% Triton X-100, 0.5 µCi [14C] L-serine, and 150 µg protein in 400 µL. The assay mix was incubated at 37°C for 30 min and lipids were extracted by the addition of 400 µL chloroform/methanol (1:2), vortexing, centrifuged for 3 min at 4000 x g, and the upper phase was removed. The organic phase was dried and suspended in chloroform/methanol (1:2). Lipids were separated by TLC plate with chloroform/methanol/2N NH3 aq., (40:10:1) and radioactive bands were visualized and quantified using a Fujex Bio-Imaging Analyzar, BAS2000 (Fuji Photo Film, Sendai, Japan).
[3H]Thymidine proliferation assay
Cells were cultured in 24-well plates for 3 or 4 days as described above. For the last 4 h, cells were incubated with 2 µCi [3H]thymidine and washed three times with PBS. Cells were treated with 1 mL of 10% (w/v) trichloroacetic acid, incubated for 20 min at 0°C, then washed three times with 5% (w/v) trichloroacetic acid. The precipitate was dissolved in 1 mL of 0.5 M NaOH solution, and the radioactivity was measured in a liquid scintillation counter.
Antibodies
Anti-Dab-2, antitransfferin, and anti-c-Src (SRC 2) antibodies were purchased from Transduction Laboratories (Lexington, KY) and Santa Cruz Biotechnologies, Inc. (Santa Cruz, CA). SRC 2 reacts with Src p60, Yes p62 and Fyn p59 but not with other Src gene family tyrosine kinases. Antiflotillin-1 antibody was purchased from BD Transduction Laboratories (Erembodegem, Belgium).
Immunoblotting
Cells were washed with PBS twice and lysed in buffer A contained PBS, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1x protease inhibitor mixture (CompleteTM), and 1 mM dithiothreitol, then sonicated. After centrifugation at 300 x g for 5 min at 4°C, the supernatant was treated with an equal volume of 10% (w/v) trichloroacetic acid and incubated for 20 min at 0°C. The precipitate was collected by centrifugation at 20,000 x g for 5 min at 4°C. The supernatant was discarded and the pellet was washed with acetone and resolubilized in buffer B (50 mM Tris–HCl [pH 8.0], 1 mM EDTA, 1% SDS). After quantification of protein concentrations using the BCA protein assay kit (Pierce Chemical Company), samples were diluted with equal volumes of 2 x SDS sample buffer contained 125 mM Tris–HCl (pH 6.8), 4% SDS, 20% glycerol, 10% 2-mercaptoethanol, and a trace amount of bromophenol blue. Proteins were separated by SDS–PAGE and transferred to PVDF membranes (Immobilon Millipore Corp.). Nonspecific binding sites were blocked by incubation in PBS containing 0.05% tween 20 and 5% dry milk. Immunoblots were then incubated with primary antibody (anti-Dab-2, anti-c-Src, and antiflotillin-1 antibodies, each diluted 1:1000) for 1 h and followed by secondary antibody [peroxidase-conjugated donkey anti-rabbit IgG F(ab')2 fragment or sheep anti-mouse IgG F(ab')2 fragment, both from Amersham Biosciences, Buckinghamshire, England, and diluted 1:5000) for 1 h. Labeling was detected by the enhanced chemiluminescence (ECL) or ECL-plus detection method (Amersham Biosciences).
Sucrose gradient centrifugation
F9 cells were washed with PBS and lysed on ice for 20 min in 2 mL of TNE buffer (10 mM Tris–HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA) containing 0.1% Triton X-100, 1x protease inhibitor mixture (Complete; Roche Molecular Biochemicals, Mannheim, Germany), 1 mM PMSF, and 2 mM orthovanadate. Lysates were centrifuged for 5 min at 1300 x g to remove nuclei and large cellular debris, and the supernatants were mixed with equal volumes of 85% (w/v) sucrose in TNE buffer. In an ultracentrifuge tube the mixed lysates were overlaid with 4 mL of 30% (w/v) sucrose in TNE buffer and then with 4 mL of 5% (w/v) sucrose in TNE buffer. The samples were centrifuged at 39,000 rpm for 18 h in an SW41 rotor (Beckman Instruments, Palo Alto, CA), and 1 mL fractions were collected from the top of the gradient and analyzed by immunoblotting and lipid analysis.
Immunoprecipitation
Cells were washed with PBS twice and lysed in IP buffer contained 50mM Tris–HCl (pH 7.5), 150 mM NaCl, 2 mM NaF, 1 mM EDTA, 1 mM ethyleneglycol bis (2-aminoethl ether) tetracetic acid (EGTA), 1 mM phenylmethylsulfonyl fluoride, 1x protease inhibitor mixture (CompleteTM), 1 mM dithiothreitol, and 1% Triton X-100 and sonicated. After centrifugation at 2000 x g for 3 min at 4°C, the supernatant was collected. After quantification of protein concentrations using a BCA protein assay kit (Pierce Chemical Company), samples were diluted with IP buffer. Aliquots (0.5 mL, 750 µg of protein) of the supernatants were precleared with 10 µL of protein A/G-Sepharose for 2 h at 4°C and then incubated with 5 µg anti-c-Src antibody or antiflotillin antibody for 2 h at 4°C and precipitated with 10 µL of protein A/G-Sepharose. Following immunoprecipitation, the beads were washed three times with PBS containing 0.05% tween 20 and lipids were extracted from immunoprecipitates using chloroform/methanol (1:2). The organic liquid containing lipid was dried and suspended in chloroform/methanol (1:2). And then TLC immunoblotting was performed by the method of Taki et al. (23), with slight modification as follows, lipids were separated on a high performance thin layer chromatography (HPTLC) plate with chloroform/methanol/0.2% CaCl2 (55:45:10) and immersed in a mixture of isopropyl alcohol/aqueous 0.2% calcium chloride/methanol (40:20:7) for 20 s. The plate was then covered with a PVDF membrane (Immobilon, Millipore, Bedford, MA) and a glass microfiber filter (Atto Instruments, Tokyo, Japan). This was then pressed (level 8) for 50 s with TLC thermal blotter (Atto Instruments) at 180°C, after which the PVDF membrane was separated from the plate and dried. The PVDF membrane was agitated in PBS containing 0.05% tween 20 and 5% dry milk for 1 h, and then, the membrane was incubated with cholera toxin B subunit-peroxidase conjugate (diluted 1:10,000 with PBS containing 0.05% tween 20 and 5% dry milk) at 4°C overnight. The membrane was washed with this buffered saline with 0.005% tween 20 and dipped and shaken in the secondary antibody, peroxidase-conjugated goat anti-mouse IgM, solution in 5% skim milk/TBS-T for 1 h. After washing with PBS containing 0.05% tween 20, labeling was detected by the ECL or ECL-plus detection method (Amersham Biosciences).
Confocal laser microscopy
The cells were washed twice with PBS and fixed with 3.7% formaldehyde in PBS at room temperature for 30 min and then treated with or without 0.5% Triton X-100/PBS for 5 min. Blocking was performed by incubation in PBS containing 10 mg/mL bovine serum albumin at room temperature for 30 min. To analyze cholesterol expression on the cell surface, biotinylated derivative (BC), of perfringolysin O (-toxin) was used as described previously (Waheed et al., 2001). Cells were incubated with biotinylated C (50 µg/mL) in PBS containing 10 mg/mL bovine serum albumin at room temperature for 1 h. After washing twice with PBS, the cells were stained with Alexa 488 or 594 avidin conjugate (diluted to 1:1000) at room temperature for 1 h.
To analyze GM1 expression, cells were incubated for 1 h at room temperature with FITC-cholera toxin B subunit-peroxidase conjugate (1:250, Sigma). Immunostaining of SFK was performed by incubating the cells with the anti-SFK antibody (1:100), followed by detection with Alexa 488 anti-rabbit IgG conjugate (1:200). To deplete cholesterol, cells were incubated in the presence of 3 mM MßCD for the last 1 h of cell cultures just before fixation. After washing three times with PBS, cells were mounted with Mowiol 4–88 (Calbiochem, San Diego, CA). Fluorescence microscopy was performed with a Zeiss Axioskop 2 plus (Carl Zeiss, Thornwood, NJ) microscope using an Axiocam CCD camera (Carl Zeiss).
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
This study was supported in part by grants-in-aid 15390019 (to J.I) and 15040202 (Scientific Research on Priority Area to J.I) from the Ministry of Education, Culture, Sports, Science and Technology. This work was also supported by the Mizutani Foundation for Glycoscience (to J.I).
|
Abbreviations |
|---|
Disabled-2, Dab-2; D-PBPP, D-threo-1-phenyl-2-benzyloxycarbonylamino-3-pyrroridino-1-propanol D-PDMP, D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol; DRMs, detergent resistant microdomains; EDTA, ethylene diamine tetraacetic acid; GSLs, glycosphingolipids; LacCer, lactosylceramide; MßCD, methyl-ß-cyclodextrin; PBS, phosphate buffered saline; PCR, polymerase chain reaction; PrE, primitive endoderm; PVDF, polyvinylidene difluoride, RA, retinoic acid; SDS, sodium dodecyl sulphate; SDS–PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; SFK, Src family kinase; SL, sphigolipid; Sph, sphingosine; SSC, 3M sodium chloride, 0.3 M sodium citrate, pH 7; TLC, thin-layer chromatography; TNE, 10 mM Tris–HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Atencia, R., Garcia-Sanz, M., Unda, F., and Arechaga, J. (1994) Apoptosis during retinoic acid-induced differentiation of F9 embryonal carcinoma cells. Exp. Cell Res., 214, 663–667.[CrossRef][ISI][Medline]
Bjersing, J.L., Brorsson, A., and Heby, O. (1997) Increased expression of c-jun, but not retinoic acid receptor beta, is associated with F9 teratocarcinoma stem cell differentiation induced by polyamine depletion. J. Cell Biochem., 67, 378–385.[CrossRef][ISI][Medline]
Brown, D.A. and London, E. (2000) Structure and function of sphingolipid- and cholesterol-rich membrane rafts. J. Biol. Chem., 275, 17221–17224.
Cho, S.Y., Lee, S.H., and Park, S.S. (1999) Differential expression of mouse Disabled 2 gene in retinoic acid-treated F9 embryonal carcinoma cells and early mouse embryos. Mol. Cells 9, 179–184.[ISI][Medline]
Cummings, R.D., and Mattox, S.A. (1988) Retinoic acid-induced differentiation of the mouse teratocarcinoma cell line F9 is accompanied by an increase in the activity of UDP-galactose: beta-D-galactosyl-alpha, 1,3-galactosyltransferase. J. Biol. Chem. 263, 511–519.
Damjanov, I., Zhu, Z.M., Andrews, P.W., and Fenderson, B.A. (1994) Embryonal carcinoma cells differentiate into parietal endoderm via an intermediate stage corresponding to primitive endoderm. In Vivo., 8, 967–973.[Medline]
Duprey, P., Morello, D., Vasseur, M., Babinet, C., Condamine, H., Brulet, P., and Jacob, F. (1985) Expression of the cytokeratin endo A gene during early mouse embryogenesis. Proc. Natl. Acad. Sci. U S A, 82, 8535–8539.
Glebov, O.O., and Nichols, B.J. (2004) Lipid raft proteins have a random distribution during localized activation of the T-cell receptor. Nat. Cell Biol., 6, 238–243.[ISI][Medline]
Ha, H., Kwak, H.B., Lee, S.K., Na, D.S., Rudd, C.E., Lee, Z.H., and Kim, H.H. (2003) Membrane rafts play a crucial role in receptor activator of nuclear factor kappaB signaling and osteoclast function. J. Biol. Chem., 278, 18573–18580.
Hakomori, S. (1981) Glycosphingolipids in cellular interaction, differentiation, and oncogenesis. Annu. Rev. Biochem., 50, 733–764.[CrossRef][ISI][Medline]
Hakomori, S.I. (2000) Cell adhesion/recognition and signal transduction through glycosphingolipid microdomain. Glycoconj. J., 17, 143–151.[CrossRef][ISI][Medline]
Harris, T.M. and Childs, G. (2002) Global gene expression patterns during differentiation of F9 embryonal carcinoma cells into parietal endoderm. Funct. Integr. Genomics, 2, 105–119.[CrossRef][Medline]
Hocevar, B.A., Mou, F., Rennolds, J.L., Morris, S.M., Cooper, J.A., and Howe, P.H. (2003) Regulation of the Wnt signaling pathway by disabled-2 (Dab2). EMBO J., 22, 3084–3094.[CrossRef][ISI][Medline]
Huang, C.L., Cheng, J.C., Liao, C.H., Stern, A., Hsieh, J.T., Wang, C.H., Hsu, H.L., and Tseng, C.P. (2004) Disabled-2 is a negative regulator of integrin alpha(IIb)beta(3)-mediated fibrinogen adhesion and cell signaling. J. Biol. Chem., 279, 42279–42289.
Inokuchi, J. and Radin, N.S. (1987) Preparation of the active isomer of, 1-phenyl-2-decanoylamino-3-morpholino-1-propanol, inhibitor of murine glucocerebroside synthetase. J. Lipid Res. 28, 565–571.[Abstract]
Inokuchi, J.I., Uemura, S., Kabayama, K., and Igarashi, Y. (2000) Glycosphingolipid deficiency affects functional microdomain formation in Lewis lung carcinoma cells. Glycoconj. J., 17, 239–245.[CrossRef][ISI][Medline]
Inoue, M., Fujii, Y., Furukawa, K., Okada, M., Okumura, K., Hayakawa, T., and Sugiura, Y. (2002) Refractory skin injury in complex knock-out mice expressing only the GM3 ganglioside. J. Biol. Chem., 277, 29881–29888.
Iwabuchi, K., Handa, K., and Hakomori, S. (1998a) Separation of "glycosphingolipid signaling domain" from caveolin-containing membrane fraction in mouse melanoma B16 cells and its role in cell adhesion coupled with signaling. J. Biol. Chem., 273, 33766–33773.
Iwabuchi, K., Yamamura, S., Prinetti, A., Handa, K., and Hakomori, S. (1998b) GM3-enriched microdomain involved in cell adhesion and signal transduction through carbohydrate–carbohydrate interaction in mouse melanoma B16 cells. J. Biol. Chem., 273, 9130–9138.
Jimbo, M., Yamagishi, K., Yamaki, T., Nunomura, K., Kabayama, K., Igarashi, Y., and Inokuchi, J.I. (2000) Development of a new inhibitor of glucosylceramide synthase. J. Biochem. (Tokyo), 127, 485–491.
Kasahara, K., Watanabe, K., Kozutsumi, Y., Oohira, A., Yamamoto, T., and Sanai, Y. (2002) Association of GPI-anchored protein TAG-1 with src-family kinase Lyn in lipid rafts of cerebellar granule cells. Neurochem. Res., 27, 823–829.[CrossRef][ISI][Medline]
Kasahara, K., Watanabe, Y., Yamamoto, T., and Sanai, Y. (1997) Association of Src family tyrosine kinase Lyn with ganglioside GD3 in rat brain. Possible regulation of Lyn by glycosphingolipid in caveolae-like domains. J. Biol. Chem., 272, 29947–29953.
Kawai, H., Allende, M.L., Wada, R., Kono, M., Sango, K., Deng, C., Miyakawa, T., Crawley, J.N., Werth, N., Bierfreund, U., and others (2001) Mice expressing only monosialoganglioside GM3 exhibit lethal audiogenic seizures. J. Biol. Chem., 276, 6885–6888.
Kihara, A., Ikeda, M., Kariya, Y., Lee, E.Y., Lee, Y.M., and Igarashi, Y. (2003) Sphingosine-1-phosphate lyase is involved in the differentiation of F9 embryonal carcinoma cells to primitive endoderm. J. Biol. Chem., 278, 14578–14585.
Liour, S.S., Kapitonov, D., and Yu, R.K. (2000) Expression of gangliosides in neuronal development of P19 embryonal carcinoma stem cells. J. Neurosci. Res., 62, 363–373.[CrossRef][ISI][Medline]
McBurney, M.W. and Rogers, B.J. (1982) Isolation of male embryonal carcinoma cells and their chromosome replication patterns. Dev. Biol., 89, 503–508.[CrossRef][ISI][Medline]
Minucci, S., Botquin, V., Yeom, Y.I., Dey, A., Sylvester, I., Zand, D.J., Ohbo, K., Ozato, K., and Scholer, H.R. (1996) Retinoic acid-mediated down-regulation of Oct3/4 coincides with the loss of promoter occupancy in vivo. EMBO J., 15, 888–899.[ISI][Medline]
Morris, S.M., Arden, S.D., Roberts, R.C., Kendrick-Jones, J., Cooper, J.A., Luzio, J.P., and Buss, F. (2002) Myosin VI binds to and localises with Dab2, potentially linking receptor mediated endocytosis and the actin cytoskeleton. Traffic, 3, 331–341.[CrossRef][ISI][Medline]
Munro, S. (2003) Lipid rafts: elusive or illusive? Cell, 115, 377–388.[CrossRef][ISI][Medline]
Nagafuku, M., Kabayama, K., Oka, D., Kato, A., Tani-ichi, S., Shimada, Y., Ohno-Iwashita, Y., Yamasaki, S., Saito, T., Iwabuchi, K., and others (2003) Reduction of glycosphingolipid levels in lipid rafts affects the expression state and function of glycosylphosphatidylinositol-anchored proteins but does not impair signal transduction via the T cell receptor. J. Biol. Chem., 278, 51920–51927.
Nojiri, H., Kitagawa, S., Nakamura, M., Kirito, K., Enomoto, Y., and Saito, M. (1988) Neolacto-series gangliosides induce granulocytic differentiation of human promyelocytic leukemia cell line HL-60. J. Biol. Chem., 263, 7443–7446.
Osanai, T., Kotani, M., Yuen, C.T., Kato, H., Sanai, Y., and Takeda, S. (2003) Immunohistochemical and biochemical analyses of GD3, GT1b, and GQ1b gangliosides during neural differentiation of P19 EC cells. FEBS Lett., 537, 73–78.[CrossRef][ISI][Medline]
Rahmann, H., Rosner, H., Kortje, K.H., Beitinger, H., and Seybold, V. (1994) Ca(2+)-ganglioside-interaction in neuronal differentiation and development. Prog. Brain Res., 101, 127–145.