Fig 2: Localization of endogenous CREG1. (A) LO2 cells were coimmunostained for CREG1 and the early endosomal marker EEA1, the late endosomal marker RAB7, and LAMP1. Insets are 3× magnifications of the boxed regions. (B) E7.5 mouse embryos were coimmunostained for CREG1 and SNX1 (sorting nexin 1; an early endosome marker), RAB7, and LAMP2. The panels on the right are magnified images of the boxed regions in the images on the left. Arrowheads indicate colocalization of CREG1 with the organelle markers (in yellow). (C) E16.5 mouse embryos were coimmunostained for CREG1 and CDH1/E-cadherin (cadherin 1); the nucleus was counterstained with DAPI. The right panel is a confocal image with the optical section thickness of 0.5 µm. (D) Confocal images of coimmunostaining of E16.5 visceral yolk sac for CREG1 and immunoglobulin G (IgG). CREG1 was localized with internalized IgG on the apical side of the endoderm. (E) Confocal images show colocalization of CREG1 with the lysosomal marker LAMP2 toward the basal side of the visceral endoderm

Fig 3: The effect of inhibition of macropinocytosis and dynamin-dependent endocytic trafficking on the expression of lysosomal proteins. (A) LO2 hepatocytes and HEK293 cells were transfected with pCMV6-CREG1 or the vector alone and cultured for 10 days. Cells were coimmunostained for CREG1 and IGF2R. (B) LO2 cells were immunoprecipitated (IP) with anti-CREG1 monoclonal antibody or pre-immune IgG. Immunoblotting was used to detect for IGF2R. (C) HEK293 cells stably transfected with pCMV6-CREG1 (T) or the vector alone (C) were immunoprecipitated with anti-IGF2R polyclonal antibody and immunoblots were probed for CREG1 and IGF2R. (D) Confluent LO2 cells were cultured in RPMI-1640 medium for 24 h. Conditioned media were immunoprecipitated with HEXA antibody or pre-immune IgG. Immunoblots were analyzed with the same antibody. Cell lysates were also analyzed by immunoblotting. (E) The percentage of HEXA in the conditioned media were plotted. N = 3, P = 0.785. (F) LO2 cells stably transfected with pCMV6-CREG1 or the empty vector were pre-incubated with the macropinocytosis inhibitor 5-(n-ethyl-n-isopropyl)amiloride (EIPA), dynasore (dyn, an inhibitor for dynamin-dependent endocytosis), or the vehicle control for 30 min. The cells were then labeled with TRITC-dextran (Dex) or Alexa Fluor 488-TF for 30 min. Treatment of the cells with EIPA and dynasore inhibited the internalization of TRITIC-Dex and Alexa Fluor 488-TF, respectively. (G) LO2 cells were treated with EIPA, dynasore, or the vehicle control (C) for 24 h and then harvested for immunoblot analysis. GAPDH served as a loading control. (H) The blots were quantified by densitometry and plotted as a ratio to GAPDH. N = 3, *P < 0.05 vs pCMV6-C, #P < 0.05 vs pCMV6-CREG1-C

Fig 4: CREG1 is localized at the endosomal-lysosomal compartment. (A) Tissues were isolated from C57BL/6J mice and analyzed by immunoblotting for CREG1. Ponceau S staining for proteins on PVDF membranes revealed protein loading for each lane. WAT: white adipose tissue. The spleen lane in the right panel was cropped from the same blot as those of leukocyte and bone marrow and separated with a black line. In the right panel, protein samples that slightly migrated into separating gels were transferred to PVDF membranes, stained with Ponceau S, and served as loading controls. (B) LO2 human hepatocytes were transfected with C-terminally MYC-tagged human CREG1 cDNA in the pCMV6 vector. Ten-days after transfection, cultures containing both transfected and untransfected cells were immunostained with antibodies to MYC-tag, human CREG1, and markers for subcellular organelles. The nucleus was counterstained with DAPI. Asterisk symbols indicate the transfected cells. (C) LO2 cells stably transfected with pCMV6-CREG1 or pCMV6 alone were subjected to immunoblotting using antibodies against the C-terminal region of CREG1 (ab191909), MYC-tag, the CREG1 N-terminal region (ab59719), and the 30R epitope (sc-100,695). ACTB/β-actin served as a loading control

Fig 5: CREG1 promotes endocytic trafficking. (A) LO2 cells stably transfected with human CREG1 cDNA (pCMV6-CREG1) or the vector alone were incubated with 100 µg/ml TRITIC-dextran for 30 min. The cells were then fixed with 3% paraformaldehyde and internalized TRITIC-dextran was visualized by fluorescence microscopy. (B) Fluorescence intensity of internalized TRITIC-dextran was quantified using ImageJ software. N = 30 for each group, *P < 0.001. (C) LO2 cells stably transfected with pCMV6-CREG1 or the empty vector were incubated with 50 µg/ml Alexa Fluor 488-TF (transferrin) for 30 min. Endocytosis of Alexa Fluor 488-TF was visualized by fluorescence microscopy. (D) Fluorescence intensity of internalized Alexa Fluor 488-TF was quantified and plotted. N = 30 for each group, *P < 0.001. (E) Mixed populations of CREG1-transfected and untransfected LO2 cells were incubated with 50 µg/ml Alexa Fluor 488-TF for 30 min. Cells were fixed and stained for CREG1. Arrows indicate colocalization of CREG1 with internalized Alexa Fluor 488-TF in the transfected cells. (F) LO2 cells were incubated with 50 µg/ml Alexa Fluor 488-TF at 4°C for 15 min, washed, and chased at 37°C for 30 or 60 min. The cells were then fixed and immunostained for RAB5. Arrowheads point to the accumulation of Alexa Fluor 488-TF at a subplasma membrane region. Arrows indicate colocalization of Alexa Fluor 488-TF with RAB5. (G) AML12 mouse hepatocytes were stably transfected with control or Creg1 shRNA (CREG1 knockdown, CREG1 KD). The cells were analyzed by immunoblotting for CREG1. ACTB served as a loading control. (H) CREG1 knockdown and control cells were incubated with 200 µg/ml FITC-ALB (albumin), fixed, and photographed by fluorescence microscopy. (I) Fluorescence intensity of internalized FITC-ALB was quantified. N = 30 for each group, *P < 0.001. (J) AML12 cells were incubated with 50 µg/ml Alexa Fluor 488-TF for 30 min and photographed by fluorescence microscopy. (K) Internalized Alexa Fluor 488-TF was quantified and plotted. N = 30 for each group, *P < 0.001