Supplementary materilas and methods

Histological studies

One part of regenerating pancreases were fixed in aqueous Bouin’s solution for 24h and embedded in paraplast (Paraplast Plus, Labonord, Templemars, France). Each paraffin block containing the remnant pancreases was serially sectioned (5 µm). Pancreas sections were selected at fixed intervals throughout the block (every 35th section). For morphometric analysis sections were stained for insulin using a horseradish peroxidase conjugated secondary antibody as previously described (Movassat et al. 1997).

Proliferation studies

For proliferation studies, BrdU staining was performed using a cell proliferation kit (RPN Cell Proliferation Kit, GE Healthcare Europe, Orsay, France) according to manufacturer’s instructions (Movassat et al, 1997). Sections were then double-stained for insulin or amylase or CK20, followed by incubation with appropriate alkalin phosphatase-conjugated secondary antibodies

To estimate the pancreatic cell replication rate, total beta or ductal cells were counted in at least 4 sections per pancreas, using an Olympus BX40 microscope (40x objective). Then BrdU-positive beta cells or BrdU+ ductal cells, were counted. Results were expressed as the percentage of BrdU+ beta or BrdU+ ductal cells respectively. For acinar replication study, the total number of amylase+ cells (acinar cells) was counted in 6 random fields per section, each field containing at least 350 acinar cells. The number of BrdU+ /amylase+ cells was counted in the same fields and the results were expressed as the percentage of BrdU+ /amylase+ over total amylase+ cells.

Immunostaining procedures for nuclear transcription factors

For PDX1, Ngn3 and Sox9 stainings, slides were subjected to antigen retrieval procedure. They were incubated for 15 min in boiling citric acid buffer (10 mM, pH 6.0) and allowed to cool down at room temperature for an hour. Slides were then washed in TBS and incubated with goat serum for 1 hour at room temperature. Sections were then incubated with primary antibodies (anti- PDX1, or anti-Ngn3 or anti-Sox9) (supplementary table 2) diluted in TBS containing 0.1% Triton X-100 (TTBS) overnight at 4°C. Slides were washed in TBS and then incubated with the appropriate secondary antibodies conjugated with horseradish peroxidase for 1 hour at room temperature. These secondary antibodies were revealed with DAB substrat kit (Vectors Laboratories, Burlingame, UK). This was followed by a second immunostaining using CK20 or amylase or insulin, using appropriate alkalin phosphatase-conjugated secondary antibodies. The secondary antibodies were revealed with an alkaline phosphatase substrate kit (Vectors Laboratories, Burlingame, UK).

Cell apoptosis

In order to assess cell death by apoptosis, we performed a TUNEL assay using ApopTag peroxidase in situ apoptosis detection kit (Millipore, Molsheim, France) according to manufacturer’s instructions.TUNEL staining was followed by insulin or amylase or CK20 staining to identify apoptotic beta cells, acinar cells and ductal cells respectively. The number of apoptotic beta cells was determined by counting TUNEL+/insulin+ cells as well as the total insulin+ cells, in at least 4 sections per pancreas. The results were expressed as the percentage of TUNEL+/insulin+ cells over the total number of beta cells counted. For acinar and ductal cell apoptosis the total number of TUNEL+/amylase+ or TUNEL+/CK20+ cells were counted and expressed respectively as the number of apoptotic acinar or apoptotic ductal cells per µm2 of pancreatic tissue section.

Tissue lysates

Whole pancreas and liver lysates were prepared by sonication of tissues in a lysis buffer (1 mM NaF, 1% NP 40, 0.25% sodium deoxycholate, 1 mM Na3VO4, 1 mM EDTA, 150 mM NaCl, 1 mM PMSF, aprotinin and leupeptin at 10 µg/mL). The concentration of the protein in the lysates was determined by the bicinconinic acid protein assay (Uptima, Interchim, Montluçon, France).

Gel electrophoresis and Western Blotting

20 µg of proteins were resolved on 10% Tris-HCl acrylamide gel and then transferred to immobilon PVDF membranes (Millipore, Molsheim, France). Membranes were blocked for non specific binding with 5% nonfat dry instant milk in Tris buffered saline-Tween (TTBS) and incubated with anti GSK3 antibody overnight at 4°C. After washes in TTBS, membranes were incubated with appropriate horseradish peroxidase-conjugated secondary antibody. After washes in TTBS, separated proteins were visualized by an ECL kit (GE Healthcare Europe, Orsay, France) for 5 min and light emission was captured on X-ray film (GE Healthcare Europe, Orsay, France). Membranes were then stripped using Re-Blot solution (Millipore, Molsheim, France) and then re-probed with actin antibody.