University of Groningen

Regulation of polarity development in hepatocytesWouden, Johanna Margaretha van der

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2004

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Wouden, J. M. V. D. (2004). Regulation of polarity development in hepatocytes: a novel role for oncostatinM. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 09-11-2020

CHAPTER 3

Cell polarity development is controlled by sphinganine turnover, which modulates

membrane traffic via the cell-cycle regulatory p27kip1/CDK complex

Johanna M. van der Wouden1, Sven C.D. van IJzendoorn1, Delphine Théard and Dick

Hoekstra

Department of Membrane Cell Biology, Faculty of Medical Sciences,

University of Groningen, Groningen, the Netherlands

1These authors contributed equally to this work

part of this work has been submitted

Chapter 3

78

Abstract The molecular mechanisms that coordinate cell cycle control and cell

polarization are largely unknown. Sphingolipids play a crucial role in

both cell proliferation and polarization. Here, we demonstrate that

polarity-stimulating signaling cascades target dihydroceramide synthase

to reduce the level of free sphinganine. Sphinganine turnover, in turn,

modulates polarized endosomal membrane traffic via p27Kip1 and cyclin-

dependent kinase to control cell polarity development. Our data reveal a

novel mechanism that couples polarized membrane traffic and the

biogenesis of apical plasma membrane domains to the regulation of cell

cycle proteins in response to signaling-modulated sphingolipid

metabolism.

Chapter 3

79

Introduction Epithelial cell polarity depends on the structural asymmetry of apical and

basolateral membrane domains. Its establishment and maintenance

requires a careful orchestration of extra- and intracellular signals,

triggering crucial organelle- and membrane traffic-linked machineries, as

partly elucidated in systems as diverse as yeast, Drosophila and

mammalian cells. Impairment of these events may cause a loss of apical-

basolateral asymmetry and lead to deranged cell growth (Bilder et al.,

2000). Epithelial polarisation preferentially occurs during G1/G0 cell

cycle arrest. However, intracellular signal cascades that control

polarisation in response to cell cycle regulation are unknown.

Hepatocytes, like all epithelia, display distinct apical, bile canalicular (BC)

and basolateral, sinusoidal plasma membrane (PM) domains. Some of the

intracellular sites and molecular components of the mechanisms that

contribute to apical BC biogenesis in hepatocytes have been clarified.

These include E-cadherin (Matsui et al., 2002), protein kinase PK)A and

PKC activities (Zegers et al., 1998), and the sorting of specific

sphingolipids and proteins, mediated by the Golgi and endosomal

recycling system (Maier et la., 2002; Aït Slimane et al., 2002; Zegers and

Hoekstra, 1997). In particular the subapical compartment (SAC), the

hepatocyte equivalent of the common endosome in other epithelia, plays a

central role in polarized lipid and protein trafficking (van IJzendoorn et

al., 1998; Rahner et al., 2002; Wustner et al., 2001), and appears a target

for signals that promote apical-basolateral asymmetry (van IJzendoorn

and Hoekstra, 1999; van IJzendoorn et al., 2000).

Sphingolipids, together with cholesterol, cluster to form particularly

ordered membrane environments or rafts, in which apical proteins

and/or signal transduction molecules are co-assembled (Ait Slimane et

al., 2003; Simons and Ikonen, 1997). Therefore, the lateral dynamics and

organization of sphingolipids may play a prominent role in the regulation

of signalling cascades and cell polarity (Holthuis et al., 2001). Indeed, the

sphingoid base backbones of sphingolipids, i.e. sphingosine and

Chapter 3

80

sphinganine (dihydro-sphingosine), appear highly bioactive compounds

that may function as signalling molecules in the regulation of cell

proliferation, apoptosis, differentiation and membrane traffic (Shayman,

2000; Hannun et al., 2001; Friant et al., 2001). Not surprisingly,

alterations in the cellular levels of free sphingoid bases have been

implicated in various diseases, including hepatocellular carcinogenesis

(Spiegel and Merrill, 1996), which arises from dedifferentiation of mature

hepatocytes (Bralet et al., 2002; Gournay et al., 2002). However, no data

are available as to the involvement of free sphingoid bases in cell

polarisation. Here, we demonstrate for the first time that the sphingolipid

metabolite sphinganine controls membrane polarity development by

interfering with the regulation of cell cycle control via the cyclin-

dependent kinase inhibitor p27kip1.

Results Sphinganine Accumulation Blocks Polarity Development

To examine the involvement of sphingolipid metabolites in the process of

hepatocyte polarity development, we employed the mycotoxin fumonisin

B1 (FB1), which specifically inhibits dihydroceramide (DHC) synthase

activity in the sphingolipid biosynthetic pathway (Fig. 1a). As anticipated,

treatment of well-differentiated hepatoma HepG2 cells with FB1 resulted

in decreased levels of ceramides and complex sphingolipids and,

consequently, increased levels of the sphingoid bases sphinganine (SA)

and phyto-SA (Fig. 1b; Merrill, 1996). In order to examine potential

consequences on HepG2 cell polarity development, i.e. the biogenesis of

apical BC, cells were plated on coverslips in the presence of 0-30 µM FB1.

Each day a coverslip was taken and the ratio BC/100 cells was

determined as a measure for cell polarity (van IJzendoorn and Hoekstra,

2000). Control HepG2 cells rapidly acquired polarity reaching a

maximum ratio BC/100 cells 2 days after plating (Fig. 1c). By contrast,

cells grown in the presence of FB1 showed a dramatic inhibition of the

Chapter 3

81

Figure 1: Accummulation of SA inhibits HepG2 cell polarity development. (a) Scheme of the sphingolipid metabolic pathway and sites of action of distinct

sphingolipid synthesis inhibitors. (b) Effect of FB1 on sphingolipid profiles. Note

the decrease in ceramides and glycosphingolipids (circles) and the appearance of

free sphingoid bases (arrows) in FB1-treated cells. (c-f) Effect of 0-30 µM FB1

(c), 30 µM FB1 + 10 µM C6-DHC (e), or 0.5-2.0 µM SA (f) on cell polarity

development. (d) Fluorescence microscopical evaluation of cell polarity in control

or FB1-treated cells (nuclei in blue and BC in red). Note the absence of BC in

FB1-treated cultures. Bars 5 µm; * p<0.05 (student t-test).

Chapter 3

82

formation of BC in a dose-dependent manner, resulting in a reduction of

the number of polarised cells by ~75 % 2 d after plating (Fig. 1c,d).With

the concentrations of FB1 used, we did not observe significant effects on

cell viability (trypan blue exclusion; not shown), or overall morphology,

including cell attachment and spreading. Importantly, polarity

development was not restored when FB1-treated cells were replenished

with 10 µM C6-DHC (Fig. 1e) or short-chain ceramide (not shown). This

suggests that the accumulation of SA, rather than the inhibition of de

novo synthesis of ceramides and/or complex sphingolipids, was

responsible for the observed impediment of cell polarity development.

Indeed, the addition of 0.5-2.0 µM SA upon plating the cells inhibited

HepG2 cell polarity development in a dose-dependent manner (Fig. 1f),

thereby closely mimicking the effect of FB1 (Fig. 1c). Taken together, the

data suggest that SA accumulation inhibits the establishment of HepG2

cell polarity.

Sphinganine Turnover Promotes Polarity Development

To obtain further support for a direct correlation between SA level and

cell polarisation, we next investigated whether decreased levels of this

sphingolipid metabolite also affect cell polarity development. A specific

inhibitor of serine palmitoyltransferase, L-cycloserine (LCS) inhibits SA

biosynthesis (Fig. 1a). Treatment of HepG2 cells with 250 µM LCS, similar

to FB1, resulted in decreased levels of ceramides and complex

sphingolipids but, unlike FB1 (cf. Fig.1b), did not give rise to increased

levels of sphingoid bases (Fig. 2a). Rather, it is fair to conclude that the

amount of free sphingoid bases was also reduced, albeit below our

detection level. Interestingly however, in striking contrast to the effect of

FB1, LCS enhanced cell polarity development, in terms of both kinetics

and maximum polarity, evidenced by the steep increase of the number

(Fig. 2b), as well as the size (Fig. 2c) of the BCs. Similar results were

obtained with another specific inhibitor of serine palmitoyltransferase,

ISP-1/myriocin (unpublished data). The opposing effects of LCS and FB1

on polarity development, under the experimental conditions used, point

Chapter 3

83

Figure 2: LCS promotes HepG2 cell polarity development. (a) Effect of LCS

on sphingolipid profiles. Note the decrease in ceramides, glycosphingolipids and

sphingomyelin in LCS-treated cells (circles). (b,d) Effect of 50-250 µM LCS (b),

250 µM LCS + 10 µM C6-DHC (d), or 250 µM LCS + 0.5-2.0 µM SA (d) on cell

polarity development. (c) Fluorescence microscopical evaluation of cell polarity

in control or FB1-treated cells (nuclei in blue and BC in red). Note the increase

in BC circumference in LCS-treated cells. Bars 5 µm; * p<0.05 (student t-test).

to SA turnover, rather than to decreased levels of (dihydro)ceramide

and/or complex sphingolipids as a crucial parameter in polarity

development, a decrease promoting and an increase inhibiting polarity

development. Indeed, the polarity-stimulating effect of LCS treatment was

completely abolished in the presence of increasing concentrations of

Chapter 3

84

exogenous SA (Fig. 2d), whereas exogenous C6-DHC (Fig. 2d) or short-

chain ceramide were without effect (not shown).

Because these data indicate that the establishment of HepG2 cell polarity

is critically dependent on the level of SA, we next investigated whether

DHC synthase, the predominant enzyme in SA turnover, functions as a

target for cellular signals that contribute to polarity development. For

this, we took into account our previous observation that the stable cAMP

analog dibutyryl cAMP (dbcAMP), via activation of PKA, stimulates HepG2

cell polarity development (Zegers and Hoekstra, 1997; van IJzendoorn

and Hoekstra, 2000a, 2000b). Control and dbcAMP-treated HepG2 cells

were broken, and incubated with 3H-SA and stearyl CoA. The activity of

DHC synthase was significantly upregulated in dbcAMP-treated cells

(approx. 1.5-fold), as evidenced by the increased production of

radiolabeled (dihydro)ceramide, whereas treatment with FB1 blocked the

activity of the enzyme (Fig. 3a). Importantly, dbcAMP failed to stimulate

polarity development in DHC synthase-inhibited or SA-treated cells (Fig.

3b). We conclude that alterations in the pool of free SA, regulated by

signal-mediated mediated modulation of acyl CoA-dependent DHC

synthase activity is a crucial parameter in cAMP/PKA-stimulated HepG2

cell polarity development.

Figure 3: (a) dbcAMP stimulates FB1-

sensitive acyl-CoA-dependent DHC

synthase activity. (b) Increased levels of free

SA block dbcAMP-stimulated polarity

development in HepG2 cells.

Chapter 3

85

Sphinganine Interferes with Polarised Membrane Traffic, exiting

from SAC

cAMP/PKA activity is required for HepG2 cell polarity development (van

IJzendoorn and Hoekstra 2000b) and, moreover, mediates the

stimulating effects of interleukin-6 family cytokines on BC biogenesis

(van der Wouden et al., 2002). We previously identified a specific BC

(apical PM)-directed traffic pathway originating from the subapical

compartment (SAC) in HepG2 cells as a primary target for cAMP/PKA

signalling (van IJzendoorn and Hoekstra, 1999; van IJzendoorn et al.,

2000). Thus, cAMP/PKA activates a specific SAC-to-BC vesicular traffic

route, which is followed by distinct, fluorescently labelled lipids including

sphingomyelin (van IJzendoorn and Hoekstra, 1999; van IJzendoorn et

al., 2000) and galactosylceramide (GalCer; Fig. 4). Inhibition of this route

effectively prohibits polarity development (van IJzendoorn and Hoekstra,

2000a). Since the accumulation of SA prevented dbcAMP-stimulated

development of HepG2 cell polarity (Fig. 4b), we next investigated whether

the observed effect was directly correlated with a frustration of

cAMP/PKA signaling to redirect and stimulate SAC-to-BC membrane

trafficking.

To this end, cells were pretreated with FB1, SA or buffer, and SAC was

preloaded with fluorescent GalCer as described in detail elsewhere (van

IJzendoorn and Hoekstra 2000a, 2000b; van der Wouden et al., 2002).

Cells were subsequently treated with dbcAMP or buffer at 4 ºC, warmed

to 37ºC, and lipid flow from SAC to either apical (BC) or basolateral

membrane was followed for 20 min. In control cells, C6NBD-GalCer

disappeared during this time interval from the bile canalicular pole (BCP),

which constitutes SAC (where the lipid is located at t=0) and BC. Indeed,

the percentage of BCP that contained C6NBD-GalCer decreased in time

(Fig. 4a: 1, 2). The relative distribution of the lipid analogue in the

remaining, faintly-labeled C6NBD-GalCer-positive BCP shows that the

probe predominantly resided in SAC and little if any movement from SAC

to BC could be detected (Fig. 4b and c: 1, 2). Rather, these data indicate

Chapter 3

86

Figure 4: FB1 interferes with polarized membrane trafficking from SAC. SAC of control (a-c: 1-3) or FB1-treated (a-c: 4, 5) HepG2 cells were preloaded

with C6NBD-GalCer, as described in detail elsewhere (van Ijzendoorn and

Hoekstra, 2000a, 2000b; van der Wouden et al., 2002). Following a subsequent

chase, the percentage of C6NBD-GalCer positive BCP, comprising BC and/or

SAC, was determined (a), as well as the corresponding relative distribution of

the fluorescent probe in the BCP, i.e., in BC, SAC or both (BC+SAC) (b). 1.

Before the chase (t=o); 2. 20 min chase in Hank’s balanced salt solution (HBSS)

(control cells) ; 3. 20 min chase in HBSS + 1 mM dbcAMP (control cells); 4, 20

min chase in HBSS + 1 mM dbcAMP (FB1-treated cells); 5. 20 min chase in

HBSS + 1 mM dbcAMP + 5 µM roscovitine (FB1-treated cells). Data are

expressed as mean +/- SD of at least three independent experiments carried out

in duplicate. (c) Illustrative images showing the distribution of the fluorescent

probe in the BCP (i.e. BC alone, BC+SAC, or SAC alone). Asterisks: BC; arrows:

SAC. Lower panels are phase contrast images to upper panels. Bars 5 µm.

Chapter 3

87

that the probe was transported from SAC to the basolateral domain of the

cells, consistent with previous observations (van IJzendoorn and

Hoekstra, 2000a, 2000b). By contrast, in dbcAMP-treated cells, C6NBD-

GalCer remained associated with the BCP region during the 20 min

chase, as evidenced by the unchanged percentage of BCP that contained

C6NBD-GalCer (Fig. 4a: 1, 3). Analysis of the corresponding relative

distribution of the probe in the BCP reveals a flow of the lipid analogue

from SAC to BC (Fig. 4b and c: 1, 3; cf. van IJzendoorn and Hoekstra,

2000a, 2000b; van der Wouden et al., 2002). In striking contrast,

dbcAMP failed to target C6NBD-GalCer from SAC to BC in cells in which

SA turnover was inhibited (Fig. 4b and c: 1, 4), but not in cells pretreated

with LCS, in which the SA levels are decreased (unpublished data, cf.

fig.4 1 vs 3). Thus, in cells with elevated SA levels, C6NBD-GalCer

remained associated with the BCP during the 20 min chase (Fig. 4a: 1, 4),

and analysis of the relative distribution of the lipid analogue in the

labeled BCP revealed that most C6NBD-GalCer localized to SAC (Fig. 4b

and c: 1, 4), indicating entrapment of the lipid in SAC. These data

strongly suggest that the turnover of SA is a prerequisite for cAMP/PKA

signaling to redirect polarized membrane traffic from SAC and, in this

way, to stimulate polarity development.

Sphinganine Modulates Transport and Polarity Development Via

p27Kip1/CDK

Both cAMP/PKA signaling and sphingoid bases have been implicated in

cell growth control. Indeed, as shown by BrdU incorporation (Fig. 5a), in

time, cell polarity-stimulating (db)cAMP/ PKA signaling delayed G1-S-

phase transition of HepG2 cells. By contrast, the inhibition of SA

turnover stimulated S-phase entry. Entry of cells into S-phase is typically

regulated by the concerted action of cyclin-dependent kinase (CDK) and

its functional inhibitor p27kip1. During the establishment of HepG2 cell

polarity (0-48 h), p27kip1 expression increased, while under dbcAMP/PKA-

mediated polarity promoting conditions its expression was strongly

enhanced (Fig. 5b). It has been reported that elevated p27kip1 levels in

Chapter 3

88

Figure 5: Effect of dbcAMP and SA on cell cycle. (a) dbcAMP and FB1 inhibit

and advance S-phase entry, as evidenced by the number of cell that had

incorporated BrdU. (b) Time-dependent expression of p27kip1 in control and

dbcAMP-treated cells. (c) Effect of FB1 on p27kip1 expression. (d) Roscovitine

restores the ability of dbcAMP to stimulate polarity development in FB1-treated

cells. Error bars represent SD of at least three independent experiments carried

out in duplicate.

HepG2 cells delay S-phase entry causing cell cycle arrest in G1 (Klausen

et al., 2000), and its role in the differentiation of a variety of epithelial

cells has been proposed (Quaroni et al., 2001; Boriello et al., 2000;

Deschênes et al., 2001). In contrast, reduced p27kip1 levels, which is

accompanied by increased CDK activity (Li et al., 2002), play a critical

role in the pathogenesis of many human cancers, including

hepatocellular carcinogenesis, which arises from dedifferentiation of

hepatocytes. In order to investigate whether the accumulation of free SA

affects these cell cycle regulatory proteins and, hence, their interference

with HepG2 cell polarity development, we first examined the effect of SA

Chapter 3

89

on the expression of p27kip1. As shown in Fig. 5c, inhibition of SA

turnover dramatically reduced p27kip1 expression by ~75 %, which could

not be restored by the exogenous addition of C6-DHC (unpublished data),

thus implicating a direct role of free sphingoid base accumulation. Note

that the reduced expression of p27kip1 is in agreement with the observed

stimulatory effect of free SA on S-phase entry (c.f. Fig. 5a). In order to

examine the involvement of CDK activity as a downstream effector of

p27kip1, we next tested the effect of roscovitine, a specific CDK inhibitor,

on the ability of dbcAMP/PKA to stimulate apical PM-directed traffic and

polarity development in DHC synthase-inhibited cells. We observed that

direct short-term inhibition of CDK activity in HepG2 cells with

roscovitine, thereby antagonizing the effect of down-regulated p27kip1

levels in FB1-treated cells, restored the ability of dbcAMP/PKA signaling

to reactivate C6NBD-GalCer trafficking from SAC and to promote its

targeting to BC (Fig. 4b and c: 1, 5). Indeed, the percentage of C6NBD-

GalCer-labeled BCP remained constant during the chase (Fig. 4a: 1, 5),

and analysis of the relative distribution of the probe in the labeled BCP

clearly revealed a flow of the lipid analogue from SAC to BC (Fig. 4b and

c: 1, 5). Importantly, short-term inhibition of CDK activity by roscovitine

also restored the ability of dbcAMP to stimulate polarity development,

despite an inhibition of DHC synthase activity (Fig. 4c). Together, these

data indicate that SA-controlled modulation of the expression and activity

of the cell cycle proteins p27kip1 and CDK interferes with the regulation of

apical membrane traffic from SAC and, hence, polarity development.

Chapter 3

90

Discussion Cell cycle control and cell polarization are key issues in tissue

development, and resolving the mechanism that co-ordinates these

events is a fundamental challenge in developmental and cell biology.

Thus, while epithelial cells preferentially differentiate and acquire a

polarized phenotype during G1 phase arrest, a loss of epithelial

architecture, i.e. PM depolarization, often occurs during carcinogenesis,

when cell cycle progression is deregulated (Bilder and Perrimon, 2000;

Wodarz, 2000). Here, a novel mechanism is described, identifying at the

molecular level a link between the regulation of sphingolipid metabolism,

cell cycle control and cell polarization, as expressed by biogenesis of the

apical membrane domain, triggered by membrane trafficking from the

subapical compartment, SAC (van IJzendoorn and Hoekstra, 1999; van

IJzendoorn et al., 2000). Moreover, the stimulatory effect of the

cAMP/PKA signaling pathway on cell polarity development (Zegers and

Hoekstra, 1997; van IJzendoorn and Hoekstra, 2000b) includes a cell

cycle control modulation, exerted via controlling the turnover of the

sphingolipid metabolite, sphinganine. The inability to maintain high

p27kip1 levels and to suppress CDK activity precludes the establishment

of an epithelial phenotype, which can be attributed to an inhibitory effect

of CDK activity on cAMP/PKA-stimulated, apical PM directed trafficking

from SAC (Fig. 6). These data also support a role for CDK in endosome

dynamics, as previously reported (Vergès et al., 1997; Gaulin et al., 2000;

Tuomikoski et al., 1989).

Intriguingly, also yeast cell polarization is tightly regulated in the cell

cycle and involves CDK (Madden and Snyder, 1998), the activity of which

is linked to Sic1p (yeast counterpart of p27kip1) and Rho1p (Drgonova et

al., 1999). Furthermore CDK activity plays a crucial role in maintaining

the polarized localization of apical components during mitosis in

Drosophila neuroblasts, which is required for asymmetric cell division

(Tio et al., 2001). In fact, a striking parallel exists between the

mechanisms that control asymmetric division in neuroblasts, and

Chapter 3

91

Figure 6: Working model for the functional integration of cAMP/PKA signalling, sphingolipid metabolism, cell cycle control, membrane traffic and cell polarity. (see text for explanation)

polarity as established in epithelial cells (Wodarz et al., 2001). Given its

functioning in cellular systems ranging from yeast and Drosophila

neuroblasts to polarized hepatocytes, it is tempting to suggest that as a

general molecular principle, CDK and its regulators may link cell cycle

control to the generation and/or maintenance of the asymmetric, apical

localization of molecules.

DHC synthase-controlled SA turnover appears a key event in cAMP/PKA-

modulated expression and activity of p27kip1 and CDK, membrane traffic,

and the establishment of polarity, all of which appear intimately

connected. The molecular mechanism by which cAMP/PKA signaling

increases DHC synthase activity remains to be investigated. Possibly, PKA

directly phosphorylates DHC synthase or, alternatively, may affect its

localization, as demonstrated for other ER/Golgi enzymes, thus affecting

substrate clearance. A diminished DHC synthase activity, as observed in

FB1-treated cells results in a diminished expression of p27kip1. Possibly,

high levels of free SA promote p27kip1 degradation, similar to the protein

degradation-promoting activities of sphingoid bases in yeast (Chung et

Chapter 3

92

al., 2000). Decreased p27kip1 levels prevent the downregulation of CDK

activity, which leads to G1-phase shortening and S-phase advance,

typically observed in many cancers. It is of interest to note that the

longevity-assurance gene (LAG1) in yeast encodes an essential component

of the acyl-CoA-dependent, FB1-sensitive DHC synthase (Schorling et al.,

2001). Deletion of LAG1 extends life span, i.e. increases the number of

times an individual yeast cell divides before it polarizes to form a bud

(Jazwinski and Conzelmann, 2002). Moreover, overexpression of LAG1

mediates resistance to FB1 and a human homologue of LAG1, LASS2,

inhibits colony formation of human hepatocellular carcinoma cells,

suggesting a role for LAG1/LASS2 in regulating cell growth (Pan et al.,

2001). Our data also support and provide a mechanism for the

observation that FB1 stimulates DNA synthesis, i.e., via accumulation of

sphingoid bases, in this manner displaying mitogenic activity (Schroeder

et al., 1994; Merrill et al., 2001). Moreover, the perturbing effects of SA on

cell cycle proteins such as p27kip1 and CDK and, consequently, on apical

membrane traffic and the establishment of epithelial architecture, as

shown here, provide a molecular mechanism by which the non-genotoxic

mycotoxin FB1 exerts its (pre)neoplastic hepatocarcinogenic effects

(Gelderblom et al., 2001).

Methods De novo sphingolipid biosynthesis

Control, FB1-, or LCS-treated HepG2 cells were labeled with 14C-serine, scraped

and sphingolipid profiles were determined by thin-layer chromotography (TLC)

and autoradiography as described (Veldman et al., 1998).

Determination of HepG2 cell polarity development

HepG2 cell polarity development was determined by fluorescent microscopical

(Olympus AX70) analysis of the ratio BC/100 cells as described (van IJzendoorn

and Hoekstra, 2000a, 2000b). Data are expressed as mean +/- SD of at least

three independent experiments, carried out in duplicate.

Chapter 3

93

DHC synthase assay

Cells, control and treated with dbcAMP for 30 min, were scraped, and

homogenized in sucrose buffer by sonication. DHC synthase activity was

measured using 3H-SA and stearyl-CoA as substrate in the presence or absence

of FB1. Lipids were extracted with CHCl3/methanol and separated by TLC.

Radiolabeled SA and (dihydro)ceramide were scraped from the TLC plates and

quantified in a liquid scintilation counter (Veldman et al., 1998). Data are

expressed as percentage of control and standardised to protein content.

Determination of p27kip1 expression

Control cells and cells treated with FB1 for 3 days were scraped. Proteins from

whole cell lysates were separated by SDS-PAGE and transferred to PVDF

membranes (van der Wouden et al., 2002). PVDF membranes were probed

sequentially with monoclonal anti-p27kip1 antibodies (Transduction Laboratories)

and HRP-conjugated secondary antibodies (Amersham). p27kip1 was visualised

by ECL and quantified using Scion Image software.

Acknowledgments S.v.IJ. is supported by the Royal Dutch Academy of Sciences (KNAW). We thank Karin

Klappe for help with the DHC synthase assay.

94