JCS035584 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig S1. Specificity of anti-Y359-P, anti-Y904-P and anti-AE1Ct. (A-C) Reactivity of antibodies to dot blots containing phosphorylated and nonphosphorylated N- and C-terminal AE1 peptides. Phosphorylated and nonphosphorylated peptides of the C-terminus and N-terminus of kAE1, were spotted on to nitrocellulose membrane at the concentrations indicated and then western blotted using anti-Y359-P, anti-Y904-P or anti-AE1Ct. (A) The anti-Y359-P antibody is specific for the phosphorylated N-terminus of kAE1. (B) The anti-Y904-P antibody is specific for the phosphorylated C-terminus of kAE1 and crossreactivity only occurs at very high concentrations of peptide. (C) The binding of the anti-AE1Ct antibody (raised to C-terminus residues 881-911) is unaffected by phosphorylation. (D) Representative western blots of AE1 monoclonal Bric170 immunoprecipitates from MDCKI cells stably expressing full-length kAE1, kAE1Y359A, kAE1Y904F, NtΔ1-360AE1, NtΔ1-338AE1 and kAE1R901Stop incubated in the presence or absence of 200 µM pervanadate for 30 minutes. The membranes were sequentially probed with anti-Y904-P, anti-AE1Ct and then anti-Y359-P, stripping the membranes after each antibody application. Anti-Y359-P did not detect pervanadate-treated Y359A or NtΔAE1-360 but did detect phosphorylation of full-length, NtΔAE1-340, kAE1Y904F and kAE1R901Stop in pervanadate-treated cells. The anti-Y904-P antibody detected phosphorylated kAE1 in all pervanadate-treated samples except cells expressing kAE1Y904F and kAE1R901Stop, which lack the Y904 residue. Neither anti-Y359-P nor anti-Y904-P reacted with kAE1 proteins in the absence of pervanadate. The blot shown is representative of at least two independent experiments.

• Supplemental Figure S2 -

  Fig. S2. Sequence comparison of N- and C-terminal regions of kAE1 from multiple species. An alignment of the N-terminal sequence around Y359 and the C-terminal sequence around Y904 was generated with Clustal W (http://www.ebi.ac.uk/clustalw/). The Y904 (highlighted in red) was conserved in all the species but the Y359 (highlighted in red) was only found in humans, mouse, dog and horse. The sheep AE1 protein sequence was kindly provided by Lesley Bruce, (Bristol Institute of Transfusion Sciences, Southmead Hospital, Bristol, UK).

• Supplemental Figure S3 -

  Fig. S3. NtΔ1-338AE1 localisation with intracellular markers. Nonpolarised MDCKI cells stably NtΔ1-338AE1 were fixed as described in the Materials and Methods. Cells were immunostained with Bric170 and rabbit antibodies to TGN38, furin, LAMP-1 or calnexin. The primary antibodies were detected using Alexa Fluor 488 goat anti-mouse and Alexa Fluor 594 goat anti-rabbit secondary antibodies. The Bric170 image is shown in A,D,G,J and the intracellular markers TGN38, Furin, Lamp-1 and calnexin in B,E,H and K. The corresponding merged images (C,F,I and L) show an overlap between NtΔ1-338AE1 detected with Bric170 and the TGN markers TGN38 and furin, (C and F) but not LAMP-1 or calnexin (I and L). Scale bars are shown in Panels A,D, G and J.

• Supplemental Figure S4 -

  Fig. S4. Densitometry comparison of pervanadate-induced phosphorylation of kAE1 residues Y359 and Y904. We analysed four representative pervanadate-treated time course western blots by densitometry. Densitometry data for Y359 and Y904 phosphorylation bands were corrected for kAE1 loading differences and then averaged across the four repeats for each time point. Standard error bars are shown. The graph shows that pervanadate induces the phosphorylation of Y359 and Y904 at a similar rate and to comparable levels.

• Supplemental Figure S5 -

  Fig. S5. Sodium bicarbonate does not induce global phosphorylation in the cell compared with pervanadate. MDCKI-kAE1 cells were cultured in three 10 cm² dishes until confluent. The medium of one dish was replaced for 1 hour with medium containing 514 mM sodium bicarbonate, one dish was treated with 200 µM pervanandate for 30 minutes and a control plate was in normal medium. The dishes were lysed in 1.5 ml SDS-PAGE sample buffer and 50 µl cell lysate was separated by SDS-PAGE 8% (w/v) gel and then blotted on to PVDF membrane. Duplicate blots were probed with two different anti-phosphotyrosine antibodies: 4G10 (Millipore, Watford, UK) PY350 (Santa Cruz, CA). Blots were also probed with an AE1 specific antibody to locate the position of the kAE1 band (not shown) and the position of kAE1 is marked. Two different exposures are shown (15minutes top panels and 1minutes bottom panels). Pervanadate caused the strong phosphorylation of a range of cellular proteins, whereas sodium bicarbonate only caused a marginal increase in cellular phosphorylation (visible on the longer film exposure) compared with untreated cells.

• Supplemental Figure S6 -

  Fig. S6. Bicarbonate-induced phosphorylation. MDCKI-kAE1 cells were cultured in 10 cm² dishes until confluent. kAE1 expression was enhanced with an overnight sodium butyrate treatment. All sodium bicarbonate medium was preincubated at 5% CO₂, 37°C for 1 hour to equilibrate. Dishes were treated as described and then immediately lysed and kAE1 immunoprecipitated with a monoclonal Bric170 as outlined in materials and methods. Eluates were separated on an 8% SDS-PAGE gel and then blotted on to PVDF membrane. Blots were probed with rabbit phosphospecific antibodies to anti-Y359-P, stripped and probed with a anti-AE1ct to show kAE1 protein loading and then probed with a anti-Y904-P. Alternatively, in some cases two immunoprecipitations were conducted and the blots probed with either anti-Y359-P or anti-Y904-P and then stripped and probed with anti-AE1ct. (A) Effects of various sodium bicarbonate concentrations on Y359 and Y904 phosphorylation. The medium of six dishes was replaced for 1 hour with medium containing the indicated concentrations of NaHCO₃ and a control plate was left in normal medium. Phosphorylation of Y359 and Y904 can be observed at 300 mM sodium bicarbonate, and it increases with sodium bicarbonate concentration. The strongest signal was achieved with 514 mM NaHCO₃. (B) Sodium-bicarbonate-induced kAE1 phosphorylation is inhibited by Src kinase family inhibitors PP1 and PP2. Three dishes were pretreated for 10 minutes with protein tyrosine kinase inhibitors, one dish each with: PP1 (25 µM), PP2 (25 µM) or PP3 (25 µM) (an inactive analogue of PP2). The medium of these dishes was replaced for 1 hour with medium containing 514 mM sodium bicarbonate and the respective inhibitor. One dish was treated with only 514 mM sodium bicarbonate and one dish was treated with normal medium. Similarly to pervanadate-induced phosphorylation, sodium-bicarbonate-induced phosphorylation was inhibited by PP1 and PP2 but not by the inactive analogue PP3. (C) AE1 inhibitor DIDS does not inhibit sodium-bicarbonate-induced kAE1 phosphorylation. Three dishes were pretreated for 20 minutes with 1 mM DIDS. The medium of these dishes was replaced for 1 hour with medium containing 514 mM sodium bicarbonate and 1mM DIDS. The medium of further three dishes was replaced for 1 hour with medium containing 514 mM sodium bicarbonate, and a control plate was left in normal medium. The densitometry data is shown in the graph in D. This compares the three repeats shown with and without DIDS for phosphorylated Y359 or Y904 corrected for kAE1 protein levels. Standard error bars are shown. This shows that phosphorylation of Y359 and Y904 induced by 514 mM sodium bicarbonate is not inhibited by the anion exchanger inhibitor DIDS.

• Supplemental Figure S7 -

  Fig. S7. Colocalisation of internalised kAE1 with intracellular markers. MDCKI-kAE1 cells were grown on coverslips, cooled on ice and bound to FITC-Bric6. Excess antibody was removed by washing, and then internalisation was allowed to occur at 37°C (A,G,M). A matching set of coverslips was treated in parallel with 200 µM pervanadate as described in the Materials and Methods (D,J,P). Cells were immunostained with the indicated intracellular marker and imaged using confocal microscopy. Pervanadate treatment internalised substantial amounts of FITC-Bric6 compared with the control cells. After 30 minutes, the majority of the internalised FITC-Bric6 did not overlap with early endosomal marker EEA1, late endosomal marker LAMP-1 or the TGN-recycling compartment marker Furin. Therefore, after 30 minutes in the presence or absence of pervanadate, internalised kAE1 resides in an unidentified intermediate compartment. Scale bars: 7.5 µm (A-F) and 15 µm (G-R).

• Supplemental Figure S8 -

  Fig. S8. Hypertonicity blocks pervanadate-induced kAE1 internalisation in polarised cells. MDCKI cells stably expressing kAE1 were seeded at high density on filters and allowed to polarise. 200 µM pervanadate (PV) was added to the cells for the times indicated in A, C and D or 514 mM NaHCO₃ for 1 hour in B. kAE1 was detected using AE1 monoclonal antibody Bric170 and a suitable secondary antibody. The upper micrographs are focal planes taken parallel to the epithelium (X-Y) near the centre of the cells for kAE1. The lower micrographs show focal planes perpendicular to the epithelium (X-Z) along the area indicated by the white line in the X-Y image. kAE1 was internalised in cells treated with pervanadate (A) but not after 1 hour of bicarbonate treatment (B). The pervanadate-induced internalisation was blocked in cells exposed to 514 mM NaCl (C) or 1 M sucrose (D). Note also that D shows that 1 M sucrose induces the most obvious cell shrinkage of the polarised monolayer. Scale bar: 30 µm.