Cell lines, culture conditions, and reagents

3T3-L1 pre-adipocytes (cat# CL-173) and adipose-derived mesenchymal stem cells (adMSC, cat# PCS-500-011) were obtained from ATCC. The murine pre-adipocytes were maintained in high glucose (4.5 g/L) Dulbecco’s Modified Eagle Media, DMEM, (ThermoFisher Scientific) supplemented with 10% heat-inactivated bovine calf serum (R&D Systems-Atlanta Biologicals) and antibiotic–antimycotic (ThermoFisher Scientific) whereas the human pre-adipocytes were maintained in mesenchymal stem cell growth media (cat# PCS-500-030 and PCS-500-040, ATCC). K8484 and Panc10.05 cell lines have been previously described (Manley et al. 2022). Acetyl Coenzyme A, Nitro Blue Tetrazolium (NBT), N-Acetyl Cysteine (NAC) and H2O2 were purchased from Sigma-Aldrich. 2X SYBR green master mix (cat# K1070) was purchased from APExBIO (Houston, TX). SHP099 (cat. # S8278) and CPI-613 (cat# S2776; PDHA1 inhibitor) were from Selleckchem. For drug/compound treatment, a stock solution of 40 mM of SHP099 or 200 mM PDHA1 inhibitor was prepared in DMSO and further diluted in water (1:1 ratio) before the desired final concentrations were reconstituted in the media; appropriate volume of 50% DMSO (in water) was used as vehicle control throughout the experiments. NAC was initially dissolved in water. Where the same vehicle treatment control was used for NAC, SHP099 and PDHA1 inhibitor, NAC stock solution was diluted in DMSO (1:1 ratio) before the final desired concentrations were reconstituted in the culture media. The treatment media was replenished every 36–48 h.

Adipogenesis stimulation

For induction of adipogenesis, 2 days over-confluent pre-adipocytes were switched from their normal growth media to differentiation media consisting of high glucose DMEM supplemented with 10% FBS, Insulin (cat# I0516, Sigma; 1 µg/ml for 3T3-L1, 3 µg/ml for adMSC), Dexamethasone (cat# 11015 Cayman Chemicals; 10 µg/ml), Rosiglitazone (cat# 71740, Cayman Chemicals; 3 µg/ml) and IBMX (cat# I5879, Sigma; 10 µg/ml) for 2 days (4 days for adMSC) before being maintained in DMEM with 10% FBS containing Insulin only (1 µg/ml); the media was changed every two days thereafter.

Collection of conditioned media (CM)

At indicated time points, adipocyte culture media was replaced with serum-free media or, where treatments took place, the cells were first washed with 1X PBS, and cultured for 24 h. This media was then supplemented with fresh serum-free media at 1:1 ratio for downstream applications.

Immunoblotting

Immunoblotting has been previously described (Messaggio et al. 2017). Protein samples were quantified using BCA (Pierce). 20–30 µg of protein were loaded per lane. Antibodies against pSHP2 [Y542 (cat# 3751)], SHP2 (cat# 3397), PDHA1 (cat# 3205), HK1 (cat# 2024), Lamin B1 (cat# 12586), p-AMPKα [Thr 172 (cat# 2535)], UCP1 (cat#14670), Perilipin-1 (cat# 9349), beta-actin (cat# 4970) and GAPDH (cat# 5174) were purchased from cell signaling technologies. Glut1 was from Invitrogen (cat# PA1-46152). Tubulin (cat# E7-s) was purchased from the Developmental Studies Hybridoma Bank (DSHB, Iowa City, IA).

Oil red O staining

For Oil Red O staining, the cells were rinsed with 1X PBS, fixed in formalin for 45 min, rinsed again with 60% Isopropanol before being stained with Oil Red O (cat# A12989, Alfa Aesar; 5 mg/ml) in 60% Isopropanol for 15 min. The stained cells were washed in 60% Isopropanol at least 3 times, 15 s each, and left in distilled water for visualization under the microscope.

Mitochondrial isolation and immunoprecipitation (IP) of SHP2

The protocol used to isolate the mitochondria has been previously established and reported (Cai et al. 2020). In brief, cells were lysed in MTiso-buffer [3 mM HEPES (pH 7.4), 210 mM mannitol, 70 mM sucrose and 0.2 mM EGTA] by Dounce homogenization [about 50 strokes (up and down)]. The homogenate was piled up on 340 mM sucrose solution in a 50 ml-tube, and centrifuged at 500×g (10 min, 4 °C) to remove nuclei and unbroken cells as pellet. Then, the supernatant was collected in 1.5 ml-tubes and centrifuged at 10,000×g (10 min, 4 °C) to isolate mitochondria as pellet. IP has been previously described (Olou et al.2017) with a modification. Briefly, the isolated mitochondrial pellet was lysed in CHAPS buffer {0.3% of 3-(3-cholamidopropyl)-dimethylammonio)-1-propanesulfonate, (CHAPS), 20 mM Tris–HCL (pH7.4), 120 mM NaCl, 10% glycerol and 5 mM EDTA) and the lysate was incubated at 4 degrees overnight with SHP2 or IgG antibodies crosslinked to the Dynabeads (Thermofisher).

In-silico protein–protein interaction modeling

The protein structures of SHP2 (PDB ID: 2SHP) and PDHA1 (PDB ID: 1NI4) were downloaded from the protein data bank. These PDB files were prepared for docking by removing ligands, water molecules and additional chain of amino acids. Chain A was selected for both proteins. We further used GrammX protein–protein docking webserver v.1.2.0 (http://vakser.compbio.ku.edu/resources/gramm/grammx/) (Tovchigrechko and Vakser 2005, 2006) to predict the binding between SHP2 and PDHA1 by using default settings. About 10 models of the protein–protein complex were studied. The best predicted model was selected based on the number of hydrogen bonds, bond distance and visualized using Pymol (Alexander et al. 2011).

Cell survival by MTT assay

2000 cells were seeded in replicate in a 96 well plate, exposed to conditioned media (CM) and incubated with MTT (0.5 mg/ml) reagent for 1.5 h. Media was removed, then crystals were dissolved in DMSO before absorbance was read at 570 nm.

Immunofluorescence (IF)

The cells were seeded and stimulated with differentiation factors in an IF chamber (Fisher Scientific). For IF, the incubation chamber was removed and the slide, with the cells, was washed 3 times with PBS, followed by (1) cells fixation for 30 min at room temperature (RT) in 2% buffered formalin in PBS and (2) 3 washes (5 min each with PBS). The cells were then permeabilized with 0.1% Triton in TBS for 30 min followed by blocking for 1 h at RT using Donkey-serum IF buffer (10 mM Tris pH 7.4, 0.1 M MgCl2, 0.5% Tween20, 2% BSA, 5% donkey Sera). The cells were stained with primary antibodies (1:100) in blocking buffer (diluted in TBS, 1:1 ratio) overnight at 4 degrees Celsius. The next day, the cells were washed 3 times (5 min each) with 0.05% Triton in TBS before being stained with 2° antibodies (Alexa-fluor dye, Invitrogen; 1:2000) and Hoescht nuclear dye (1:2000) in 0.05% Triton-TBS for 1 h at RT on a shaker. Lastly, the slide was washed 4 times (10 min each) in 0.05% Triton-TBS before being mounted and visualized under fluorescence microscope. SHP2 antibodies used were raised in mouse (cat# LF-MA0186, Thermo Fisher) while PDHA1 antibodies (cat# 3205, cell signaling) were from rabbit origin.

ROS assay by nitroblue tetrazolium (NBT) and dihydroethidium (DHE)

NBT (cat# N6876, Sigma) is a yellow powder which, in the presence of ROS, gets reduced to dark-blue crystals, therefore allowing for a visualization of ROS. For the assay (Furukawa et al. 2004), cells were incubated with 0.2% NBT in PBS for 1.5 h and switched to PBS. The crystals were visualized and imaged under the microscope, then dissolved in acetic acid (50%) and the absorbance was subsequently read at 560 nm. DHE assay was performed according to the manufacturer’s protocol (Abcam, cat# ab236206).

Glycerol assay

For glycerol release, 23-day mature adipocytes were treated in phenol red-free serum-free media. The media was collected and incubated with assay reagents according to the manufacture’s protocol (Neogen, cat# K-GCROL).

Lactate assay

This assay has been previously described (Manley et al. 2022). In brief, cells were grown in 96 well pate and the media were collected for measurement of secreted lactate according to the manufacturer protocol (Cayman Chemical; cat# 600,450).

Transwell migration and invasion assays

PDAC cells (50 K) were seeded in the upper chamber (24-well inserts, 8 µm pore size; Corning Incorporated cat# 3422) of the transwell in serum-free media. 600 µL of conditioned media were pipetted into the bottom chamber of the plate and made a contact with the bottom of the insert. The insert was coated (invasion) or not (migration) with Matrigel and allowed to solidify in the incubator before the cells were seeded. Following the assay, cotton-tipped applicator was used to remove cells, from upper surface of the transwell, that did not migrate/invade. Then the migrated or invader cells were first fixed in 70% ethanol and stained with 0.2% crystal violet for 10 min each. The staining was dissolved in 50% acetic acid and absorbance was read at 590 nm.

Cell cycle analysis

250 K cells were seeded in 6-well plate, synchronized in serum-free media for 24 h and then exposed to conditioned media for 18 h. This was followed by Propidium Iodide (PI) staining protocol, including incubation with RNAse, and cell cycle phases were evaluated by flow cytometry. The analysis was performed using the FlowJo software. The percentage of cells in each phase of the cell cycle were determined by PI intensity and quantified.

ELISA

For the quantitative murine IL-6 assay, the Quantikine ELISA immunoassay kit (R & D System, cat# M6000B) was used. In brief, following treatment of the 23-day differentiated adipocytes in serum-free, the media was collected and used for the assay according to the manufacturer’s protocol.

IL-6 neutralization

The levels of IL-6 in the media were first determined by quantitative ELISA as described above. The media was then incubated with IL-6 antibodies (cat# 16706181; Thermo Scientific, 50 ng/mL) overnight and used to grow PDAC cells; IgG antibodies were used as control.

RNA isolation and qPCR

Protocols have been previously described (Messaggio et al. 2017). The qPCR program was according to the three-step method in the protocol from the master mix vendor (APExBIO) and was as follow: initial denaturation (hold; 1 cycle): 95 °C for 2 min; 40 cycles of denaturation (95 °C for 15 s), annealing (50–60 °C for 30 s) and extension (72 °C for 30 s) followed by 1 cycle of 95 °C for 15 s, 60 °C for 1 min and 95 °C for 15 s. The primers used (below) are specific to mouse species:

PDHA1 forward: TGATCCGCCTTTAGCTCCATC; Reverse: TGTGACCTTCATCGGCTAGAA.

Ptpn11 (SHP2): cat# QT00103362, Qiagen.

2-NBDG (2-deoxy-2-[(7-nitro-2,1,3-benzoxadiazol-4-yl) amino]-D-glucose) uptake

For 2NBDG uptake assay, adipocytes were differentiated in a 96-well plate, then glucose and serum starved for 40 min prior to incubation with 2NBDG (200 µM, Cayman Chemical, Cat# 11046) in PBS for 80 min. At endpoint, cells were washed with PBS and 2NBDG content was measured via plate reading at excitation 475 nm, emission 550.

Lentiviral transfection for SHP2 knockdown

For the generation of SHP2 knockdown cells, short hairpin RNA (shRNA) constructs, with scrambled SCR or targeting two independent regions of SHP2 mRNA, were obtained from Millipore Sigma (Burlington, MA USA). These constructs were used to generate packaged lentiviruses by transfecting the constructs with packaging constructs (pCMV-dR8.2 #8455 and pCMV-VSV-G #8454, addgene) into HEK293T cells to produce viral supernatants which were subsequently used to transduce indicated cells. Successfully transfected cells were selected with puromycin treatment (2 µg/mL) until complete cell death was evident in un-infected cells.

Diet-induced obesity in mice

Methods have been previously described (Mendonsa et al. 2015). In brief, the mice were subjected to either a standard chow diet or high fat diet (TD88137 Teklad, 42% calories from fat, 42.7% from carbohydrates, and 15.2% from protein) for three months at which time the body weight was determined and tissues collected. Animals were handled in accordance with the Institutional Animal Care and Use Committee (IACUC) of the University of Kansas Medical Center.

Patient samples

Adipose tissue samples were obtained from de-identified pre-operatively consented patients at the University of Kansas Medical Center via the Biospecimen Repository. The patient’s body mass index (BMI) was across the spectrum and categorized as follows: normal (18–24), overweight (25–29) and obese (over 30).

Statistical analysis

For assessment of statistical significance, Ordinary one-way analysis of variance (ANOVA) was performed with Sidak’s multiple comparisons test/Dunnett’s multiple comparisons test, or where there are just two sample groups, a t-test was performed. The tests were performed using GraphPad Prism9 software. p < 0.05 was considered significant.