Vimentin affects localization and activity of sodium-glucose cotransporter SGLT1 in membrane rafts
Isabelle Runembert1,
Guillaume Queffeulou1,
Pierre Federici2,
François Vrtovsnik1,
Emma Colucci-Guyon3,
Charles Babinet3,
Pascale Briand2,
Germain Trugnan4,
Gérard Friedlander1 and
Fabiola Terzi1,*
1 INSERM U426 and Department Physiology, Faculté
de Médecine Xavier Bichat, IFR 02,
Université Paris 7, Paris, France
2 INSERM U380, Institut Cochin de
Génétique
Moléculaire, Paris, France
3 URA 1960 CNRS, Institut Pasteur, Paris, France
4 INSERM U538, Faculté de
Médecine Saint-Antoine, Paris, France
View larger version (22K):
[in a new window]
|
Fig. 1. Sodium-glucose cotransport activity of proximal tubular cells in primary
culture from vimentin-null mice (Vim-/-; closed circles)
and from wild-type littermates (Vim+/+; open circles). The
uptake of methyl- -D-glucopyranoside (MGP) was evaluated in the presence
of [14C]-MGP (0.5 µCi/ml) and appropriate concentrations of MGP.
Na-dependent glucose uptake (insert) was calculated as the difference between
MGP uptakes measured in the presence of sodium or glucamine. Eadie-Hofstee
plot shows the Vmax and the Km of the MGP
transport system of the two cell types. Data are means±s.e.m. of four
separate cultures, experiments were performed in triplicate. Statistical
analysis: ANOVA, Vim-/- vs Vim+/+
cells, P<0.005.
|
|
View larger version (46K):
[in a new window]
|
Fig. 2. SGLT1 and SGLT2 mRNA and SGLT1 protein expression of proximal tubular cells
in primary culture from vimentin-null mice (Vim-/-) and
from wild-type littermates (Vim+/+). (A) Northern blot
analysis. Total RNA was extracted from Vim+/+ and
Vim-/- cultured cells and from whole kidney (K) of
Vim+/+ mice using RNAzol kit. cDNA probes, labeled by a
random priming method, were: the rat SGLT1, the rat SGLT2 and the mouse GAPDH.
Blots are representative samples from six animals and six separate cultures.
(B) Western blot analysis of brush border membranes (BBM). BBM were prepared
by MgCl2 precipitation and differential centrifugation procedures.
Proteins were immunoblotted with a rabbit polyclonal anti-SGLT1 antibody and a
rabbit polyclonal anti-5'-nucleotidase antibody. Blots are
representative samples from five separate cultures. (C) Western blot analysis
of biotinylated proteins extracted from BBM. For specific cell surface
biotinylation experiments, cells were incubated twice consecutively with
NHS-ss-biotin, BBM were prepared and the biotinylated antigens were recovered
with streptavidin agarose beads. Then, proteins were immunoblotted with a
rabbit polyclonal anti-SGLT1 antibody. Blots are representative samples from
two separate cultures. Statistical analysis: no difference was observed
between Vim+/+ and Vim-/- cells for
any of the parameters.
|
|
View larger version (79K):
[in a new window]
|
Fig. 3. SGLT1 protein expression in detergent-resistant membranes (DRM) of proximal
tubular cells in primary culture from wild-type animals. (A) Cell membranes
from wild-type cells were solubilized in Triton X-100, DRM were purified on a
sucrose flotation gradient and an aliquot of each 1 ml gradient fraction
(lanes 1-8=5-30% sucrose; lanes 9-12=40% sucrose) was analyzed by western
blotting. A rabbit polyclonal anti-SGLT1 antibody and a rabbit polyclonal
anti-caveolin antibody were used. (B) SGLT1 protein localization in a
Vim+/+ cell after 4% formaldehyde and ice-cold methanol
fixation (left panel) or antibody crosslinking (right panel). A rabbit
polyclonal anti-SGLT1 antibody was used, followed by a secondary
FITC-conjugated antibody. (C) SGLT1 (left panel) and 5'-nucleotidase
(5'-Nu, middle panel) crosslinking in Vim+/+ cells.
An overlay of SGLT1 and 5'-nucleotidase images is shown in the right
panel. A rabbit polyclonal anti-SGLT1 antibody and a mouse monoclonal
anti-5'-nucleotidase antibody were used, followed by the specific
secondary FITC- and TRITC-conjugated antibodies. Finally, cells were fixed in
4% formaldehyde and ice-cold methanol.
|
|
View larger version (40K):
[in a new window]
|
Fig. 4. Effect of vimentin gene inactivation on SGLT1 expression in
detergent-resistant membranes (DRM) of proximal tubular cells in primary
culture from vimentin-null mice (Vim-/-; hatched bars) and
from wild-type littermates (Vim+/+; open bars). Cells were
solubilized in Triton X-100, DRM were purified on a sucrose flotation gradient
and fractions 4-7 were pooled and immunoblotted with a rabbit polyclonal
anti-SGLT1 antibody, a rabbit polyclonal anti-5'-nucleotidase antibody
and a rabbit polyclonal anti-caveolin antibody. Blots are representative
samples from three separate cultures. Data are means±s.e.m. ANOVA:
Vim-/- vs Vim+/+ cells,
**P<0.005.
|
|
View larger version (22K):
[in a new window]
|
Fig. 7. Vimentin protein expression in detergent-resistant membranes (DRM) of
proximal tubular cells in primary culture from wild-type animals. Cell
membranes were solubilized in Triton X-100, then DRM were purified on a
sucrose flotation gradient. (A) An aliquot of each 1 ml gradient fraction
(lanes 1-8=5-30% sucrose; lanes 9-12=40% sucrose) was collected and analyzed
by western blotting using a mouse monoclonal anti-vimentin antibody. (B)
Fractions 4-7 of DRM were pooled, washed, resuspended in TNE buffer containing
1% Triton X-100 and a second sucrose gradient was performed as described
above, followed by western blot analysis. A mouse monoclonal antivimentin
antibody and a rabbit polyclonal anti-caveolin antibody were used. (C) The
pooled 4-7 DRM fractions were immunoprecipitated with a rabbit polyclonal
anti-caveolin antibody (lines 1 and 2) or non-immune rabbit serum (line 3).
The immunoprecipitates were analyzed by western blotting using either a rabbit
polyclonal anti-caveolin antibody (left and right) or a mouse monoclonal
anti-vimentin antibody (middle). (D) Effect of methyl-ß-cyclodextrin
(MCD) on DRM vimentin expression. Cells were treated or not with MCD (10 mM at
37°C for 2 hours), then DRM were prepared as described above, followed by
western blotting using a mouse monoclonal anti-vimentin antibody. All blots
are representative samples from three separate cultures.
|
|
View larger version (123K):
[in a new window]
|
Fig. 8. Vimentin protein localization in proximal tubular cells in primary culture
from wild-type animals. (A-C) Cells were fixed in 4% formaldehyde and ice-cold
methanol, then incubated with a rabbit polyclonal anti-vimentin antibody (A)
and a mouse monoclonal anti-5'-nucleotidase antibody (B), followed by
the specific secondary TRITC-and FITC-conjugated antibodies. The overlay is
shown in panel C. (D-F) Cells were fixed in 4% formaldehyde and ice-cold
methanol, then incubated with a mouse monoclonal antivimentin anti-vimentin
antibody (D) and a rabbit polyclonal anti-SGLT1 antibody (E), followed by the
specific secondary TRITC- and FITC-conjugated antibodies. The overlay is shown
in F.
|
|
© The Company of Biologists Ltd 2002
P<0.05,
