1797
Journal of Cell Science 112, 1797-1801 (1999)
Printed in Great Britain © The Company of Biologists Limited 1999
JCS9895
Apical secretion of chondroitin sulphate in polarized Madin-Darby canine
kidney (MDCK) cells
Svein Olav Kolset1,*, Tram Thu Vuong2 and Kristian Prydz2
1Institute for Nutrition Research, University of Oslo, Box 1046 Blindern, 0316 Oslo, Norway
2Department of Biochemistry, University of Oslo, Box 1041 Blindern, 0316 Oslo, Norway
*Author for correspondence (e-mail: [email protected])
Accepted 1 March; published on WWW 11 May 1999
SUMMARY
Sugar moieties have been shown to contain sufficient and
necessary information to target examples of secreted and
transmembrane glycoproteins to the apical surface of
epithelial MDCK cells. We have investigated if the sugar
chains of proteoglycans, the glycosaminoglycans, also
contain structural determinants for apical transport. Here
we show that although 75% of the proteoglycan secretion
from MDCK cell monolayers is into the basolateral
medium, 75% of the proteoglycans of the chondroitin
sulphate type are secreted apically. The sorting information
in the chondroitin sulphate proteoglycans is localized to the
sugar chains, since protein-free chondroitin sulphate
chains, initiated on hexyl β-D-thioxyloside, were also
predominantly secreted to the apical medium.
INTRODUCTION
that would influence the secretion of proteoglycans to either
the apical or the basolateral side of filter-grown MDCK cell
monolayers.
Here we show that GAG chains of the CS type on PGs
produced by MDCK cells contain apical sorting information.
Polarized MDCK II monolayers secreted CSPGs into the apical
medium, while most of the PGs secreted, which were of the
HS type, were found in the basolateral medium. Furthermore,
protein-free CS chains, initiated on hexyl β-D-thioxyloside,
were also predominantly secreted to the apical medium,
demonstrating that the apical sorting information in CSPG is
localized to the GAG chain.
Epithelial cells express and secrete proteoglycans (PGs) that
contribute to cell adhesion and the construction of the
basement membrane (Caplan et al., 1987). The major portion
of these proteoglycans is of the heparan sulphate (HS) type,
but several epithelial cell types also express chondroitin
sulphate proteoglycans (CSPGs) (Couchman et al., 1996;
Svennevig et al., 1995). Both HSPG and CSPG synthesis is
completed in the Golgi apparatus, where the linear
glycosaminoglycan (GAG) chains are polymerized and
sulphated. The biological activities of the PGs are to a large
extent dependent on the type of repeating disaccharides in the
GAG chains and the pattern of sulphation in these units. HSGAGs contain repeating disaccharides of N-acetylglucosamine and hexuronic acid, while the corresponding unit
in CS-GAGs is N-acetyl-galactosamine and hexuronic acid.
Increasing evidence has accumulated, mainly from
studies with epithelial MDCK cells, that both N-linked (Urban
et al., 1987; Scheiffele et al., 1995; Gut et al., 1998) and
O-linked (Yeaman et al., 1997) glycoproteins may contain
apical sorting information. Other structures have also been
implicated in apical sorting, particularly the anchors of
glycosylphosphatidylinositol (GPI)-linked proteins (RodriguezBoulan and Powell, 1992). Structures or signals that mediate
basolateral sorting are found in the cytoplasmic tail of several
transmembrane proteins (Matter and Mellman, 1994). Since
few studies have addressed the question of PG sorting in
polarized epithelial cells (Caplan et al., 1987; Mertens et al.,
1996; Stow et al., 1991), we were interested in whether the two
main classes of GAG chains, HS and CS, contain information
Key words: Sorting, Polarized cell, Proteoglycan, MDCK cell,
Chondroitin sulphate
MATERIALS AND METHODS
Cell culture
MDCK II wild-type cells and cells transfected with inserts coding for
mouse syndecan-1 with a deleted cytoplasmic tail (MDCK-TL)
(Miettinen et al., 1994) were cultured on polycarbonate filters at a
density of 106 cells per filter in Dulbecco’s modified Eagle’s medium
with 5% fetal calf serum, antibiotics and L-glutamine (All from BioWhittaker, Verviers, Belgium). Established epithelial polarized
monolayers were demonstrated by measuring the transepithelial
resistance with Millicell-ERS equipment (Millipore Corp., Bedford,
MA, USA). MDCK-TL cells express four times more PGs than the
MDCK II wild-type cells. HSPG represent 80-90% of the total PG
expression in MDCK-TL cells and 70% in MDCK II wild-type cells,
the remainder being mainly CSPG (S. O. Kolset and K. Prydz,
unpublished observation). The cells were labelled for 20 hours, in
sulphate-free RPMI 1640 with 2% FCS and glutamine, without
penicillin or streptomycin sulphate, with 0.1 mCi/ml [35S]sodium
1798 S. O. Kolset, T. T. Vuong and Kristian Prydz
sulphate (Amersham, UK) in both the apical (1.5 ml) and basolateral
(2.5 ml) media. MDCK-TL cells were labelled in the absence and
presence of 0.1 mM hexyl-β-D-thioxyloside (HX-xyl).
Table 1. The secretion of 35S-labelled macromolecules and
35S-labelled proteoglycans in polarized MDCK II cells
Gel electrophoresis
Medium fractions were collected and cell fractions solubilized in 4 M
guanidine/2% Triton X-100 in 0.05 M sodium acetate buffer, pH 6.0
and chromatographed on Sephadex G-50 fine (Amersham Pharmacia,
Uppsala, Sweden) columns. The 35S-labelled macromolecules
recovered were boiled and subjected to SDS-PAGE under reducing
conditions, using 4%-20% polyacrylamide gels (Novex, Encinitas,
CA, USA). 14C-labelled rainbow molecular mass standards were from
Amersham. Samples were treated with nitrous acid to degrade heparan
sulphate-related structures (Shively and Conrad, 1976) before SDSPAGE. The gels were dried, treated with Amplify (Amersham) and
subjected to autoradiography with Fuji Medical X-ray film (Fuji,
Japan).
Medium compartment
Gel chromatography
Some samples were treated with chondroitinase ABC (Seikagaku
Kogyo Co., Tokyo, Japan) to degrade chondroitin sulphate, or nitrous
acid and enzyme in combination, and analyzed by Sepharose CL-6B
(Pharmacia) gel chromatography. The column was run in 0.05 M TrisHCl, pH 8.0 with 0.15 M NaCl and 0.1% Triton X-100. Markers for
the total (Vt) and void volume (V0) of the column were Dextran Blue
and dinitrophenyl alanine (both from Sigma). To determine the levels
of 35S-labelled proteoglycans secreted to the apical and basolateral
media, 35S-labelled macromolecules were subjected to DEAE
(Pharmacia) ion-exchange chromatography. Material binding to the
column and eluting at high salt concentrations is defined as
proteoglycans, as has previously been described (Svennevig et al.,
1995).
Apical
Basolateral
35S-labelled
macromolecules
16.6±0.1
83.4±2.6
35S-labelled
proteoglycans
24±3.8
76±17.4
The amount of [35S]sodium sulphate incorporated into macromolecules was
measured after Sephadex G-50 gel chromatography. Two parallel cell cultures
were used to harvest apical and basolateral medium fractions (n=2). 35Slabelled proteoglycan levels were determined after DEAE ion exchange
chromatography of 35S-labelled macromolecules from the two medium
fractions.
Values are % of total label recovered.
proteoglycans in the apical and the basolateral media were
determined. The basolateral medium contained almost
exclusively HSPG, since the high molecular mass 35S-labelled
macromolecules were completely depolymerised after nitrous
acid treatment (Figs 1, right panel, 2, upper panel). In contrast,
the apical medium contained a proteoglycan form which was
resistant to this treatment (Figs 1, right panel, 2, middle panel).
Immune precipitation
cells were solubilized in 0.05 M Tris-HCl, pH 7.5 with
1% Nonidet P-40, 2 mM EDTA, 0.15 M NaCl and 35 µg/ml of PMSF.
The apical and basolateral media (non-adherent cells removed) were
used directly. The fractions were incubated overnight at 4°C with the
monoclonal antibody 281-2 against mouse syndecan-1, treated with
protein A-Sepharose (Pharmacia) for 2 hours at 4°C. Finally, the
beads were washed, boiled in sample buffer and the released products
analyzed by SDS-PAGE. Dried gels were analysed by scanning
densitometry (Molecular Dynamics, Sunnyvale, CA, USA).
35S-labelled
RESULTS
To study the secretion of PGs from epithelial MDCK cells, we
incubated confluent monolayers of filter-grown cells in the
presence of [35S]sodium sulphate to incorporate radioactive
label into the GAG chains. Scanning densitometry analyses
showed that the basolateral medium contained 76% and the
apical medium 24% of the secreted [35S]sodium sulphatelabelled proteoglycans (smear of higher molecular mass than
the 97 kDa marker in left panel of Fig. 1; Table 1). Several
sulphated proteins are also released into the apical and
basolateral media. These are separated by SDS-PAGE and have
molecular masses below the 97 kDa standard (Fig. 1). MDCK
cells have previously been demonstrated to synthesize both
heparan sulphate proteoglycans (HSPGs) and chondroitin
sulphate proteoglycans (CSPGs) (Svennevig et al., 1995). The
two types of GAG chains in these PGs may be selectively
degraded, HS by HNO2 (nitrous acid) treatment and CS and
dermatan sulphate by chondroitinase ABC treatment. At the
end of the labelling period the types of GAG chains in secreted
Fig. 1. SDS-PAGE of 35S-labelled macromolecules from MDCK-II
cells. 35S-labelled macromolecules were isolated from MDCK-II
cells and subjected to SDS-PAGE before (control) and after nitrous
acid treatment (HNO2). Api, apical medium; Baso, basolateral
medium; Cell, cell fraction. The migration distances of 14C-labelled
molecular mass standards (in kDa) are indicated. PGs can be seen as
a smear (upper and lower boundaries marked with asterisks),
particularly in the basolateral fraction, with molecular mass above
the 97 kDa standard. Sulphated proteins can be seen below the 97
kDa standard. The basolateral fraction contains almost exclusively
HSPG, which is degraded after HNO2 treatment. CSPG, which is
resistant to HNO2 treatment, can only be seen in the apical fraction.
The CSPG component (also seen as a smear) is indicated with an
arrowhead.
Proteoglycan sorting in polarized cells 1799
Fig. 2. Sepharose CL-6B gel chromatography. 35S-labelled
macromolecules from basolateral medium (upper panel) and apical
medium (two lower panels) were chromatographed after nitrous acid
treatment, which degrades HS (open circles), or both nitrous acid and
chondroitinase ABC treatment (degrades both HS and CS) (closed
circles). Untreated controls (closed squares) are shown in the upper
and middle panels. Fractions were collected and counted for
radioactivity. The basolateral fration contains only HSPG, which is
degraded with HNO2 to mostly di-and tetrasaccharides and free
[35S]sodium sulphate, and eluted in the vt fraction of the column
(upper panel). The apical fraction contains material which is resistant
to HNO2 (open circles) and eluted in the vo fractions (the middle
panel). However, negligible material of a parallel sample was
resistant (eluting in the vo fractions) to a combination of HNO2 and
chondrotinase ABC, but was eluted as completely depolymerised
products in the vt of the column, as shown in the lower panel. The
HNO2 resistant material eluting in the vo fractions in the middle
panel is, accordingly CSPG (see also Fig. 1, arrowhead).
When nitrous acid was combined with chondroitinase ABC
treatment the apical high molecular mass 35S-labelled
macromolecules were completely depolymerised and eluted in
retarded positions from the column (Fig. 2, lower panel). The
nitrous acid-resistant PG, which is only found in the apical
medium (Fig. 1, arrowhead) is, accordingly of the CSPG type.
By using scanning densitometry of the gels the amount of
CSPG in the apical medium was found to be 75% of the total
35S-labelled CSPG expressed by the MDCK II cells (Table 2).
In contrast the major part of HSPG secreted could be recovered
from the basolateral medium (Table 2), in accordance with
what would be expected for bulk flow secretion.
Xylosides consist of a hydrophobic compound coupled to
xylose, the sugar that is transferred during PG biosynthesis to
serine when GAG chain synthesis is initiated. Xylosides added
to the medium of cultured cells gain access to the enzymes
involved in GAG synthesis due to their hydrophobic moiety.
The GAG chains that develop from the xyloside are uncoupled
from protein cores and are almost always of CS nature. To
investigate if the apical sorting information was localized to
the glycan chains of CSPGs or to the protein cores, we used
hexyl-β-D-thioxyloside (HX-xyl), known to initiate the
synthesis of free CS chains (Kolset et al., 1990). In these
experiments we used MDCK-TL cells, transfected with the
gene for mouse syndecan-1 without the cytoplasmic tail. These
cells express almost exclusively HSPG (Miettinen et al., 1994).
The formation of CS-xylosides will mostly compete with the
synthesis of endogenous CSPGs and to a much lesser extent
with HSPGs (see results on syndecan-1 expression below).
Furthermore, in MDCK-TL cells syndecan-1 is secreted in
approximately equal amounts to the apical and basolateral
media (Miettinen et al., 1994). This provides us with two
advantages. Firstly, we can follow the effect of xyloside on
secretion of one particluar proteoglycan species to both sides
of the MDCK cell monolayer. Secondly, we can conclude that
the CS chains we detect are initiated on the xyloside that was
added, rather than derived and released from endogenous
CSPG after completed biosynthesis. The cells were treated
with 0.1 mM HX-xyl, a concentration shown in pilot
experiments to give maximum stimulation of GAG synthesis.
Untreated MDCK-TL cells incorporated [35S]sodium sulphate
mainly into HSPG. After HX-xyl treatment, secretion of free
CS chains (8-30 kDa) to the apical medium could be detected
(Fig. 3). Only a minor fraction of these CS chains could be
seen in the basolateral medium. These CS chains were
completely depolymerised after chondroitinase ABC
treatment, as demonstrated by a shift in elution position
compared to untreated material, after Sepharose CL-6B gel
chromatography (not shown). Quantitation of both the apical
and basolateral fractions from xyloside-treated cells by gel
chromatography showed that more than 75% of the xylosideTable 2. The secretion CSPG and HSPG into the apical
and basolateral medium of MDCK II cells
% of proteoglycan secreted
Medium compartment
Apical
Basolateral
35S-labelled
CSPG
75.2±10.9 (n=5)
22.8±11.3 (n=5)
35S-labelled
HSPG*
21±3.3 (n=2)
79±18.0 (n=2)
Apical and basolateral media were harvested from MDCK II cells labelled
20 hours with [35S]sodium sulphate. After gel chromatography the apical and
basolateral media were treated with nitrous acid and then analyzed by SDSPAGE. The amount of 35S-labelled CSPG in the upper part of the gels was
measured by scanning densitometry.
*Based partly on data published by Svennevig et al. (1995).
1800 S. O. Kolset, T. T. Vuong and Kristian Prydz
Fig. 3. SDS-PAGE of 35S-labelled macromolecules from MDCK-TL
cells. 35S-labelled macromolecules were isolated from MDCK-TL
cells cultured with and without 0.1 mM HX-xyl, labelled with
[35S]sodium sulphate for 20 hours and subjected to SDS-PAGE. For
quantitative comparisons, equal percentage volumes of the various
fractions were loaded onto the gels. Material expressed after xyloside
treatment, indicated by asterisks, can only be seen in the medium
fractions below the 46 kDa standard. The majority of the GAG
chains initiated with xyloside can be recovered from the apical
medium. The xyloside-initiated material in both the apical and
basolateral media is chondroitin sulphate.
initiated CS chains (three separate experiments) were in the
apical medium. Immunoprecipitations with a syndecan-1
monoclonal antibody showed that the level of synthesis and
secretion pattern of syndecan-1 in MDCK-TL cells was not
affected by xyloside treatment.
DISCUSSION
The data presented show that CSPG expressed by wild-type
MDCK II cells and CS initiated on HX-xyl in MDCK-TL cells
were both secreted from the apical surface of MDCK cell
monolayers, to a greater extent than or similarly to gp 80
(clusterin) (not shown). The latter is considered a typical
marker of apical protein secretion in MDCK cells. 75% of
secreted gp 80 has been recovered from the apical medium in
polarized MDCK cell cultures (Urban et al., 1987). It is
conceivable that the two endogenously synthesized and
secreted molecules gp 80 and CSPG may contain apical sorting
information. This information could be confined to the glycan
moieties in both cases, in agreement with the accumulating
evidence for glycan involvement in apical sorting for both
secretory and transmembrane proteins (Scheiffele et al., 1995;
Urban et al., 1987; Yeaman et al., 1997). This argument is
strengthened by the fact that removal of glycans results in more
randomised transport of proteins to the apical and basolateral
domains (Scheiffele et al., 1995; Urban et al., 1987).
Sphingolipid- and cholesterol-rich rafts in the trans-Golgi
network have been postulated to be the proper membrane
environment for molecules with apical sorting signals to
accumulate and become incorporated into apical vesicles.
Insolubility of rafts after detergent extraction of MDCK cells
was first shown by Skibbens et al. (1989). This insolubility has
later been one experimental approach used to demonstrate the
localization of proteins to rafts (Keller and Simons, 1997). The
resulting glycolipid-rich complexes are enriched in GPIanchored proteins, influenza haemagglutinin (Sheiffele et al.,
1997) and membrane-integrated proteins like VIP21 (caveolin1; Kurzchalia et al., 1992) and VIP36 (Fiedler et al., 1994).
However, not all proteins transported apically have been
unequivocally detected in glycolipid-rich complexes (Graichen
et al., 1996; Keller and Simons, 1998). The association of
secreted proteins with detergent-insoluble complexes has been
particularly difficult to demonstrate and thus the general
concept of rafts as mediators of apical sorting remains to be
established.
A sorting mechanism that may co-operate with, or function
independently of, rafts is the recognition of glycans by
molecules with lectin activity. Both ERGIC-53 (Itin et al.,
1996) and VIP36 (Fiedler and Simons, 1996) are molecules
localized to the secretory pathway with lectin activity, but their
biological role in this context is unknown. Interestingly, VIP36
has been reported to bind N-acetyl-galactosamine (Fiedler and
Simons, 1996), the amino sugar found in the repeating
disaccharide unit of CSPG and not HSPG. It remains to be
shown if VIP36 plays a role in the sorting of CSPG.
An early study of protein sorting in MDCK cells included a
study of the polarity of secretion of a particular HSPG, which
was mainly secreted basolaterally (Caplan et al., 1987). This is
in good agreement with our finding that most of the secreted
PGs are recovered from the basolateral medium and that these
are of the HS type. The observed distribution of HSPG between
the two media, however, may not be a result of intracellular
sorting, because the ratio is within the range of what is expected
for bulk transport without signal recognition. HS chains have
been shown to negatively influence the ability of GPI-anchors
to sort the proteoglycan glypican to the apical surface in both
MDCK cells and CaCo-2 cells (Mertens et al., 1996). While a
variant of glypican without HS-chains was transported
predominantly apically, reintroduction of HS-chains shifted the
localization to a more basolateral distribution.
It is not clear if the HS chains can be recognized as
basolateral sorting information or if their presence masks the
apical sorting information in the GPI-anchored proteoglycan
(Mertens et al., 1996). The data presented here show that CS
chains will direct CSPGs to the apical side of MDCK II cells.
The sorting and correct tissue distribution of CSPGs is likely
to be of biological importance. It has been shown that the large
CSPG versican is located on the apical surface in some
glandular epithelia (Bode-Lesniewska et al., 1996).
Furthermore, parasite adhesion to throphoblastic villi has been
shown to depend on the presence of CS (Fried and Duffy,
1996). In addition, thrombomodulin, located on the apical
surface of endothelial cells and important for the regulation of
coagulation, has been shown to contain CS (Bourin et al.,
1990).
Proteoglycan sorting in polarized cells 1801
This work was supported by the Norwegian Cancer Society, Novo
Nordisk, Nansen-fondet, Blix-fondet and the Norwegian Research
Council. Hexyl-β-D-thioxyloside was a kind gift from Dr S. Suzuki,
Aichi Medical University, Japan. MDCK-TL cells were kindly
provided by Dr M. Jalkanen, Center for Biotechnology, Turku, Finland.
REFERENCES
Bode-Lesniewska, B., Dours-Zimmermann, M. T., Odermatt, B. F., Briner,
J., Heitz, P. U. and Zimmermann, D. R. (1996). Distribution of the large
aggregating proteoglycan versican in adult‡human tissue. J. Histochem.
Cytochem. 44, 303-312.
Bourin, M.-C., Lundgren-Åkerlund, E. and Lindahl, U. (1990). Isolation
and characterization of the glycosaminoglycan component of rabbit
thrombomodulin proteoglycan. J. Biol. Chem. 265, 15424-15431.
Caplan, M. J., Stow, J. L., Newman, A. P., Madri, J., Anderson, H. C.,
Farquhar, M. G., Palade, G. E. and Jamieson, J. D. (1987). Dependence
on pH of polarized sorting of secreted proteins. Nature 329, 632-635.
Couchman, J. R., Kapoor, R., Sthanam, M. and Wu, R. R. (1996). Perlecan
and basement membrane-chondroitin sulphate proteoglycan (bamacan) are
two basement membrane chondroitin/dermatan sulphate proteoglycans in
the Engelbreth-Holm-Swarm tumor matrix. J. Biol. Chem. 271, 9595-9602.
Fiedler, K., Parton, R. G., Kellner, R., Etzold, T. and Simons, K. (1994).
VIP36, a novel component of glycolipid rafts and exocytic carrier vesicles
in epithelial cells. EMBO J. 13, 1729-1740.
Fiedler, K. and Simons, K. (1996). Characterization of VIP36, an animal
lectin homologous to leguminous lectins. J. Cell Sci. 109, 271-276.
Fried, M. and Duffy, P. E. (1996). Adherence of plasmodium falciparum to
chondroitin sulphate A in the human placenta. Science 272, 1502-1504.
Graichen, R., Losch, A., Appel, D. and Koch-Brandt, C. (1996). Glycolipidindependent sorting of a secretory glycoprotein to the apical surface of
polarized epithelial cells. J. Biol. Chem. 271, 15854-15857.
Gut, A., Kappeler, F., Hyka, N., Balda, M. S., Hauri, H.-P. and Matter, K.
(1998). Carbohydrate-mediated Golgi to cell surface transport and apical
targeting of membrane proteins. EMBO J. 17, 1919-1929.
Itin, C., Roche, A. C., Monsigny, M. and Hauri, H.-P. (1996). ERGIC-53 is
a functional mannose-selective and calcium-dependent human homologue
of leguminous lectins. Mol. Biol. Cell 7, 483-493.
Keller, P. and Simons, K. (1997). Post-Golgi biosynthetic trafficking. J. Cell
Sci. 110, 3001-3009.
Keller, P. and Simons, K. (1998). Cholesterol is required for surface transport
of influenza virus haemagglutinin. J. Cell Biol. 140, 1357-1367.
Kolset, S.O., Sakurai, K., Ivhed, I., Øvervatn, A. and Suzuki, S. (1990).
The effect of β-D-xylosides on the proliferation and proteoglycan
biosynthesis of monoblastic U937 cells. Biochem. J. 265, 637-645.
Kurzchalia, T., Dupree, P., Parton, R. G., Kellner, R., Virta, H., Lehnert,
M. and Simons, K. (1992). VIP21, A 21-kD membrane protein is an
integral component of trans-Golgi-network-derived transport vesicles. J.
Cell Biol. 118, 1003-1014.
Matter, K. and Mellman, I. (1994). Mechanisms in cell polarity: Sorting and
transport in epithelial cells. Curr. Opin. Cell Biol. 6, 545-554.
Mertens, G. M., Van der Schueren, B., van den Berghe, H. and David, G.
(1996). Heparan sulphate expression in polarized epithelial cells: The apical
sorting of glypican (GPI-anchored proteoglycan) is inversely related to its
heparan sulphate content. J. Cell Biol. 132, 487-497.
Miettinen, H. M., Edwards, S. N. and Jalkanen, M. (1994). Analysis of
transport and targeting of syndecan-1: Effect of cytoplasmic tail deletions.
Mol. Biol. Cell 5, 1325-1339.
Rodriguez-Boulan, E. and Powell, S. K. (1992). Polarity of epithelial and
neuronal cells. Annu. Rev. Cell Biol. 8, 395-427.
Scheiffele, P., Peränen, J. and Simons, K. (1995). N-glycans as apical sorting
signals in epithelial cells. Nature 378, 96-98.
Scheiffele, P., Roth, M. G. and Simons, K. (1997). Interaction of influenza
virus haemagglutinin with sphingolipid-cholesterol membrane domains via
its transmembrane domain. EMBO J. 16, 5501-5508.
Shively, J. E. and Conrad, H. E. (1976). Formation of anhydrosugars in the
chemical depolymerization of heparin. Biochemistry 15, 3932-3942
Skibbens J. E., Roth M. G. and Matlin K. S. (1989). Differential
extractability of influenza virus hemagglutinin during intracellular transport
in polarized epithelial cells and nonpolar fibroblasts. J. Cell Biol. 108, 82132.
Stow, J. L., deAlmeida, J. B., Narula, N., Holtzman, E. J., Ercolani, L. and
Ausiello, D. A. (1991). A heterotrimeric G protein, G alpha i-3, on Golgi
membranes regulates the secretion of a heparan sulphate proteoglycan in
LLC-PK1 epithelial cells. J. Cell Biol. 114, 1113-1124.
Svennevig, K., Prydz, K. and Kolset, S. O. (1995). Proteoglycans in
polarized epithelial Madin-Darby canine kidney cells. Biochem. J. 311, 881888.
Urban, J., Parczyk, K., Leutz, A., Kayne, M. and Kondor-Koch, C. (1987).
Constitutive apical secretion of an 80-kD sulphated glycoprotein complex
in the polarized epithelial Madin-Darby canine kidney cell line. J. Cell Biol.
105, 2735-2743.
Yeaman, C., Le Gall, A. H., Baldwin, A. N., Monlauzeur, L., Le Bivic, A.
and Rodriguez-Boulan, E. (1997). The O-glycosylated stalk domain is
required for apical sorting of neurotrophin receptors in polarized MDCK
cells. J. Cell Biol. 139, 929-940.