Fig 1.
Schematic representation of proinsulin synthesis, followed by either secretion or degradation. (1) Proinsulin molecules co-translationally translocate into the ER lumen. (2) Properly folded proinsulin molecules traffic to the secretory granules. Misfolded or abundant proinsulin molecules dislocate across the ER membrane (3) into the cytosol, potentially assisted by HRD1, Derlin-1 or Derlin-2 and p97 (inset). Cytosolic proinsulin is then degraded into smaller peptides by the proteasome. Proinsulin derived peptides are reimported into the ER lumen via TAP and (4) loaded onto MHC class I molecules. These traffic to the plasma membrane and (5) present the proinsulin-derived peptides to CD8+ T-cells.
Fig 2.
Generation of preproinsulin expressing K562 cells.
(A) K562 cells were retrovirally transduced to express preproinsulin-IRES-GFP and sorted for the GFP positive population. Flow cytometry analysis of wild-type preproinsulin-expressing cells before transduction (left panel), after retroviral transduction but before sorting (middle panel) and after sorting (right panel). Sorting yielded a cell population that was approximately 95% GFP positive. (B) Human pancreatic islets cells and preproinsulin-expressing K562 cells were lysed and proteins were separated on 12% Nu-PAGE. Proinsulin levels were analyzed by Western blot. The number of cells used to prepare the lysates is indicated. (C) Schematic representation of the preproinsulin molecule including the three disulfide bonds. The epitopes eluted from MHC class I molecules are depicted in red.
Table 1.
Epitopes eluted from MHC class I.
Fig 3.
Proinsulin is dislocated into the cytosol and degraded by the proteasome.
(A) K562 cells stably expressing preproinsulin were pulse labeled with 35S-methionine and cysteine for 15 minutes and chased for the indicated times. Immunoprecipitated proinsulin was analyzed using 15% SDS-PAGE. Quantification of the pulse chase experiment is shown below. Gel is representative for four experiments. (B) Preproinsulin-expressing K562 cells were treated for 3 hours with either DMSO, 100 μM ALLN or 10 μM Lactacystin (Lc). Cells were lysed and proteins were separated using 12% Nu-PAGE. Proinsulin levels were analyzed by Western blot. Transferrin receptor (TFR) was blotted as a loading control. Gel is representative for four experiments. (C) K562 cells stably expressing preproinsulin were treated for 3 hours with 100 μM ALLN and treated with Streptolysin-O to permeabilize the plasma membrane. After separation of the cytosol (supernatant) from the cell (pellet), protein levels for GFP (top panel), calreticulin (CRT, middle panel) and proinsulin (PI) were analyzed using 15% SDS-PAGE and Western blot. Gels are representative for three experiments.
Fig 4.
Derlin2, p97 and HRD1 knockdown increases proinsulin steady state levels.
(A) Schematic representation of the experimental setup for the knockdown of ERAD proteins with shRNAs. Preproinsulin-expressing K562 cells were transduced to express the respective shRNAs together with mOrange from a bicistronic lentiviral expression vector. mOrange expression levels were analyzed by flow cytometry either before transduction or on day 3 and day 7 after transduction; the latter includes 4 days of selection with puromycin. Flow cytrometry analysis of a representative transduction of K562 cells is shown. (B) Proinsulin-expressing K562 cells were transduced with the indicated shRNAs. Seven days after transduction, cell lysates were prepared and loaded onto 12% Nu-PAGE; Derlin-1, Derlin-2 and p97 protein levels were analyzed by Western blot. Actin was included as a loading control. Gels are representative for three different experiments. (C) K562 cells were transduced as described for B and proinsulin levels were analyzed by Western blot. Actin was included as a loading control. Gels are representative for three different experiments.
Fig 5.
Derlin-2 overexpression decreases proinsulin steady state levels.
K562 cells stably expressing preproinsulin were transduced with cDNA to overexpress Derlin-1 or Derlin-2 from a lentiviral expression vector. Seven days post transduction and selection, cell lysates were prepared and separated on 12% Nu-PAGE. Protein levels were analyzed by Western blot using antibodies against the indicated proteins and quantification of PI levels is shown. Gels are representative for three different experiments.
Fig 6.
Derlin-2 depletion delays insulin degradation.
(A) K562 cells stably expressing preproinsulin were transduced to express either nonsense shRNA (left) or an shRNA targeting Derlin-1 (right). After pulse-labeling for 15 minutes with 35S-methionine and cysteine, cells were chased for the indicated times. Proinsulin was immunoprecipitated from the lysates and analyzed using 15% SDS-PAGE. Quantification of the pulse chase experiment is shown on the right. Gels are representative for two different experiments. (B) Similar as described for A but with two shRNAs targeting Derlin-2. Quantification of the pulse chase experiment is shown on the right.