PDX1 phosphorylation maps to a C-terminal GSK3 consensus site. A, schematic of PDX1 protein highlighting the ERK (Ser-61 and -66) and newly identified GSK3 (Ser-268 and -272) site (top panel). Also shown is a comparison of the PDX1 phosphorylation site with known GSK3 sites (bottom panel). B, PDX1 one-dimensional autoradiography (FLAG-pPDX) and Western immunoblotting (IB; alkaline phosphatase-based detection (top panel)). Lower panels show PDX1 tryptic peptides derived from FLAG-PDX-WT (lane 1 and panel i), FLAG-PDX-S272A (lane 2 and panel ii), and FLAG-PDX-S268A (lane 3 and panel iii). C, one-dimensional autoradiography (GST pPDX1), Coomassie staining (GST PDX1), and phosphopeptide maps of bacterially expressed GST-PDX1 WT (lane 1 and panel i), GST-PDX1-S272A (lane 2 and panel ii), and GST-PDX1-S268A (lane 3 and panel iii) phosphorylated in vitro using purified GSK3β and [γ-³²P]ATP. Note loss of spot B in PDX-S272A mutant (panel ii in B and C) and loss of both spots A and B in PDX-S268A mutant (panel iii in B and C). D, tryptic maps of PDX1 phosphopeptides from in vitro GSK3-labeled GST-PDX1 (panel i), in vivo labeled FLAG-PDX1-WT (panel ii), or a mixture of equal counts from both (panel iii). Note the co-migration of spot A and B from the two reactions in panel iii. Open arrowheads indicate points of reference from in vitro maps and closed arrowheads from in vivo maps.

AKT and GSK3β regulate PDX1 stability via the newly identified C-terminal phosphorylation site. A, Western blotting of anti-HA immunoprecipitates from Min6 cells expressing HA-PDX1-WT (lanes 1–4) or HA-PDX1-S268A (lanes 5–8) in the presence of control vector (lanes 1 and 5) or kinase-dead AKT2 (AKT2-KD, lanes 2 and 6), constitutively active GSK3β (GSK-CA, lanes 3 and 7), or AKT2 (lanes 4 and 8). Densitometric quantification shown as fold difference in PDX1 protein levels normalized to HA-PDX1-WT (gray bars) or HA-PDX1-S268A (black bars) in the presence of control vector. C–J, HA immunoprecipitates from Min6 cells expressing HA-PDX1-WT or HA-PDX1-S268A alone (C and D), or in the presence of GSK3β (E–G) or AKT2 (H–J) following a time-dependent cycloheximide (Cx) chase were probed using anti-PDX1 antibodies. Tubulin (Tub) was used as a control. Densitometric quantification (D, F, and G, and I and J) demonstrates the mean ± S.D. (n = 3) of protein loss at each time point (*, p ≤ 0.05 between treatments at the indicated time point).

Mapping of glucose-regulated PDX1 phosphorylation sites using metabolically labeled rat islets and Min6 cells. A and B, PDX1 immunoprecipitates from [³²P]orthophosphate-labeled rat islets (A) and Min6 cells (B) treated with 4 mm glucose (LG) or 24 mm glucose (HG) as indicated were resolved using SDS-PAGE. Phosphorylated PDX1 (pPDX) was identified by autoradiography, and tryptic peptide mapping was performed on the excised labeled bands. Two prominent phosphopeptides labeled as spots A and B showed a higher incorporation of phosphate in the presence of low glucose compared with high glucose (panels i and ii). C, phosphopeptide A and B localize to the C terminus of PDX1. FLAG immunoprecipitates from [³²P]orthophosphate-labeled Min6 cells were transfected with FLAG-PDX1-WT (panel i) or FLAG-PDX1-C-term (amino acids 213–284) (panel ii). Phosphoamino acid analysis revealed spots A (panel iii) and B (panel iv) to be phosphoserines (pS), as inferred by the co-localization of counts/min from the phosphopeptide hydrolysates with unlabeled ninhydrin-stained phosphoamino acid standards (indicated by circles). Loading dye migration is represented by X (xylene cyanol) and D (dinitrophenyl-lysine); loading origin is marked by the closed circle; top edge of the map marks chromatography front.

AKT and GSK3β kinases regulate the phosphorylation and total protein levels of PDX1. A, one-dimensional autoradiography (FLAG pPDX), Western immunoblotting (IB), and phosphopeptide maps of FLAG-PDX1 immunoprecipitates from Min6 cells overexpressing AKT-KD in the absence (lane 1 and panel i) or presence of GSK3β inhibitor (lane 2 and panel ii) or overexpressing the vector control (lane 3 and panel iii) or Myr-AKT (lane 4 and panel iv). B–E, Western blots and densitometry of PDX1 protein levels from extracts of mouse (B and C) and human islets (D and E) in the presence of vehicle (NT, lane 1), AKT inhibitor (AKT-I, lane 2), GSK3β inhibitor (GSK-I, lane 3), or both (AKT-I+ GSK-I, lane 4). Histone H3 was used as a loading control. Densitometric data from three independent experiments were normalized to vehicle-treated controls (NT) and presented as the mean ± S.D. fold difference (*, p < 0.05 compared with untreated islets).

Glucose treatment modulates activity of AKT and GSK kinases to regulate half-life of PDX1 protein. A, Western blotting of protein extracts from Min6 cells incubated in 4 mm glucose (L, lane 1), 24 mm glucose (H, lanes 2 and 3), or 4 mm glucose + 100 nm insulin (lanes 4 and 5) in the absence (lanes 2 and 4) or presence of AKT inhibitor (lanes 3 and 5). Cellular extracts were probed as indicated. B, Min6 cells transfected with HA-PDX1-WT (lanes 1–4), along with control vector (lanes 1–3) or AKT2 (lane 3), or with HA-PDX1-S268A (lanes 5 and 6) were incubated in 4 mm glucose (L, lanes 1, 3, 4, and 5), in the presence or absence of GSK3β inhibitor (GSK-I) or 24 mm glucose (H, lanes 2 and 6), followed by immunoprecipitation using anti-HA antibodies and Western blotting for PDX1 protein. Tubulin was used as a control. C, densitometric quantitation of bands (n = 3) was normalized to PDX1-WT in 4 mm glucose (LG) and presented as the mean ± S.D. fold difference (*, p < 0.05 compared with PDX1-WT in 4 mm glucose; ns, not significant). D–F, time-dependent cycloheximide chase of HA-immunoprecipitates from Min6 cells transfected with either HA-PDX1-WT or HA-PDX1-S268A and incubated in 4 mm glucose (LG) or 24 mm glucose (HG) as indicated. Tubulin (Tub) was used as a control. t½ was determined from densitometric quantification (E and F) of time-dependent protein loss presented as the mean ± S.D. (n = 3). Closed and open circles, PDX1-WT in HG and LG, respectively; closed squares, PDX1-S268A in LG; *, p < 0.05 between treatments at the indicated time point.