First published online September 18, 2007
doi: 10.1242/10.1242/jcs.007948
Journal of Cell Science 120, 3309-3320 (2007)
Published by The Company of Biologists 2007
Distinct v-SNAREs regulate direct and indirect apical delivery in polarized epithelial cells
Thomas Pocard1,
André Le Bivic2,
Thierry Galli3,4 and
Chiara Zurzolo1,5,*
1 Unité de Trafic Membranaire et Pathogenèse, Institut Pasteur, 75724, Paris CEDEX 15, France
2 UMR 6212 CNRS/Université Aix Marseille II, IBDML, case 907, Faculté des Sciences de Luminy, 13288, Marseille CEDEX 09, France
3 Membrane Traffic in Neuronal and Epithelial Morphogenesis, INSERM Avenir Team, 75005, Paris, France
4 Institut Jacque Monod, CNRS UMR7592, Universities Paris 6&7, 75005, Paris, France
5 Dipartimento di Biologia e Patologia Cellulare e Molecolare, Università degli Studi di Napoli Federico II, 80131 Napoli, Italy
View larger version (61K):
[in this window]
[in a new window]
|
Fig. 2. DPPIV but not Ag 35/40 kDa is mislocalized in fully polarized stable FRT clones lacking TI-VAMP. FRT cells stably expressing PLAP (FRT PLAP), and FRT cells stably expressing PLAP and a vector where the siRNA targeting rat TI-VAMP was introduced (FRT si TI-VAMP::PLAP), were grown on filters for 4 days. (A) Cell extracts (40 µg) from FRT PLAP and FRT si TI-VAMP::PLAP were analyzed by SDS-PAGE and western blotting with antibodies against PLAP (lanes two and four) and human TI-VAMP (lanes one and three). Note that the two FRT cells express similar amounts of PLAP, whereas TI-VAMP is constitutively knocked-down in FRT si TI-VAMP::PLAP compared with FRT PLAP. (B) Confocal Z and X-Y sections (top and bottom) of FRT PLAP and FRT si TI-VAMP::PLAP cells were labeled with antibodies against rat DPPIV (left panel) and Ag 35/40 kDa (right panel) and secondary antibodies coupled to FITC. Note that DPPIV is only mislocalized in FRT si TI-VAMP::PLAP cells, whereas the basolateral localization of Ag 35/40 kDa is unchanged in both cell lines. Bars, 10 µm.
|
|
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 3. PLAP is mislocalized in fully polarized FRT cells stably lacking TI-VAMP. (A) Confocal Z and X-Y sections (top and bottom) of FRT cells stably expressing PLAP, and FRT cells stably expressing si TI-VAMP::PLAP, grown on filters for 4 days and labeled with antibody against PLAP and secondary antibodies coupled to FITC. PLAP is mislocalized to the lateral membrane of FRT si TI-VAMP::PLAP (right panel). (B) FRT PLAP and FRT si TI-VAMP::PLAP cells were grown on filters for 4 days and labeled with LC-biotin (Biot) added to the apical (Ap) or basolateral (Bl) surface. After lysis, PLAP was immunoprecipitated with a specific antibody, run on SDS-PAGE and revealed with HRP-streptavidin (lane one and panel three). One tenth of the immunoprecipitate (Tot) was run on SDS-PAGE and assayed with an antibody against PLAP (panels two and four). Bars, 10 µm. IgG, immunoglobulin G.
|
|
View larger version (33K):
[in this window]
[in a new window]
|
Fig. 4. Transient and stable loss of TI-VAMP does not affect the establishment of a polarized monolayer on filters. FRT cells stably expressing PLAP, FRT cells stably expressing PLAP electroporated with 100 pmol of siRNA targeting rat TI-VAMP or against human  -globin, and FRT si TI-VAMP::PLAP cells, were grown on filters for 4 days. (A) Transepithelial resistance (TER) was measured every day. Note that there is no difference in the establishment of polarity between control (siRNA beta glo) and TI-VAMP transient-knockdown cells (siRNA TI-VAMP) and between control (FRT PLAP) and TI-VAMP stable-knockdown cells (FRT si TI-VAMP::PLAP). (B) Confocal X-Y and Z sections at the level of tight junctions of control (siRNA beta glo), TI-VAMP transient-knockdown (siRNA TIVAMP) and TI-VAMP stable-knockdown cells show that the integrity of the tight junctions is preserved in each case. Bars, 10 µm.
|
|
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 5. Only PLAP but not DPPIV is mislocalized in Caco2 cells after transient RNAi of TI-VAMP. Caco2 cells were electroporated with 60 pmol of siRNA targeting human TI-VAMP or against human -globin and grown on filters for 4 days. (A) Cell extracts (40 µg) from control (siRNA beta glob) and TI-VAMP transient-knockdown (siRNA TI-VAMP) cells were analyzed by SDS-PAGE and western blotting with antibodies against human calreticulin (top panel, internal control for protein loading) and human TI-VAMP (bottom panel). TI-VAMP expression was decreased by more than 70% after transient RNAi of Caco2 cells on filters in three different experiments. (B,C) Confocal Z and X-Y sections (top and bottom) of control (siRNA beta glob) and TI-VAMP transient-knockdown (siRNA TI-VAMP) Caco2 cells. (B) Cells were labeled with an antibody against PLAP and a secondary antibody coupled to FITC. Compared with its apical localization in Caco2 cells (left panel), PLAP is mislocalized to the lateral membrane in Caco2 TI-VAMP transient-knockdown cells (right panel). (C) Cells were labeled with antibodies against human DPPIV and secondary antibodies coupled to FITC. Note that, compared with FRT cells, the apical localization of DPPIV is unaffected by the transient knockdown of TI-VAMP, as is the basolateral localization of Ag525 (B, right panel). Bars, 10 µm.
|
|
View larger version (70K):
[in this window]
[in a new window]
|
Fig. 6. VAMP8 but not VAMP3 is mislocalized in Caco2 cells after transient RNAi of TI-VAMP. Caco2 cells were electroporated with 60 pmol of siRNA targeting human TI-VAMP or against human -globin and grown on filters for 4 days. (A) Confocal Z and X-Y sections (top and bottom) of control (siRNA beta glob) and TI-VAMP transient-knockdown (siRNA TI-VAMP) Caco2 cells. Cells were labeled with antibodies against human VAMP3 (left panel) and VAMP8 (right panel) and secondary antibodies coupled to FITC. Note that the lateral localization of VAMP3 is unaffected by the transient knockdown of TI-VAMP. Compared with its vesicular localization in control Caco2 cells, VAMP8 seems to be less concentrated in vesicular structures and slightly mislocalized to the lateral membrane in Caco2 TI-VAMP transient-knockdown cells. (B) Digitized images of confocal X-Y sections of control (siRNA beta glob) and TI-VAMP (siRNA TI-VAMP) Caco2 cells. Cells were labeled with antibody against human VAMP8 and secondary antibody coupled to FITC. Fluorescent images were converted and analyzed quantitatively using ImageJ® software (see Materials and Methods). A statistical analysis of the area of each fluorescent object was performed (data not shown). We found fewer objects in the case of TI-VAMP knockdown compared with the control (130 versus 164 in the control). We also found that there were more objects of small area (less than five pixels) and fewer objects of large area (greater than ten pixels) in the case of the TI-VAMP knockdown. Bar, 10 µm.
|
|
View larger version (51K):
[in this window]
[in a new window]
|
Fig. 7. DPPIV but not PLAP is mislocalized in Caco2 cells after transient RNAi of VAMP8. Caco2 cells were electroporated with 60 pmol of siRNA targeting human VAMP8 or against human -globin and grown on filters for 4 days. (A) Cell extracts (40 µg) from control (siRNA beta glob) and VAMP8 transient-knockdown (siRNA VAMP8) cells were analyzed by SDS-PAGE and western blotting with antibodies against human calreticulin (top panel, internal control for protein loading) and human VAMP8 (bottom panel). VAMP8 expression decreased by more than 80% after transient RNAi of Caco2 cells on filters in three different experiments. (B,C) Confocal Z and X-Y sections (top and bottom) of control (siRNA beta glob) and VAMP8 transient-knockdown (siRNA VAMP8) Caco2 cells. (B) Cells were labeled with an antibody against PLAP and a secondary antibody coupled to FITC. Compared with its apical localization in Caco2 cells (left panel), PLAP is unaffected by the transient knockdown of VAMP8. (C) Cells were labeled with antibodies against human DPPIV (left panel) and secondary antibodies coupled to FITC. Note that, compared with FRT cells, the apical DPPIV is also mislocalized to the lateral membrane in Caco2 VAMP8 transient-knockdown cells (right panel), whereas the basolateral localization of Ag525 is unaffected (B, right panel). Bars, 10 µm.
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 8. Model for v-SNARE specificity in the different pathways to the plasma membrane of polarized epithelial cells. Upon sorting at the level of the TGN, proteins are addressed to the apical or basolateral plasma membrane via at least three different pathways: the direct apical pathway (bold arrowhead), the direct basolateral pathway (dashed arrow) and the transcytotic pathway (open arrowhead). Post-TGN carriers could contain different v-SNAREs (e.g. TI-VAMP, VAMP8, VAMP3). In this case, the specificity of fusion would arise from the pairing of each v-SNARE with the appropriate t-SNARE and from the stoichiometry of the v-SNAREs present on each vesicle (the predominant VAMP is indicated by underlining and alternatives are indicated by parentheses). The specificity of sorting could be dependent on the functional interaction between a v-SNARE and a specific adaptor (e.g. AP-1B for recycling endosomes to the basolateral plasma membrane in the direct basolateral pathway). Alternatively, each v-SNARE might be actively sorted into different post-TGN cargos (in this case, only the SNARE indicated by underlining would be present in the vesicle). T-J, tight junction.
|
|
© The Company of Biologists Ltd 2007
88% in three different experiments. (B,C) Confocal Z and X-Y sections (top and bottom) of control (siRNA beta glob) and TI-VAMP transient-knockdown (siRNA TI-VAMP) cells. (B) Cells were labeled with antibodies against rat DPPIV (left panel) and Ag 35/40 kDa (right panel), and secondary antibodies were coupled to FITC. Note that the apical localization of DPPIV in controls switches in TI-VAMP transient-knockdown cells to the lateral membrane (Z section and X-Y section bottom panel), whereas the basolateral localization of Ag 35/40 kDa is unchanged. (C) Cells were labeled with antibodies against PLAP and secondary antibodies coupled to FITC. Note that, as for DPPIV, PLAP is mislocalized to the lateral membrane. Bars, 10 µm.