Targeting Galectin 3 illuminates its contributions to the pathology of uterine serous carcinoma

Background Uterine serous cancer (USC) comprises around 10% of all uterine cancers. However, USC accounts for approximately 40% of uterine cancer deaths, which is attributed to tumor aggressiveness and limited effective treatment. Galectin 3 (Gal3) has been implicated in promoting aggressive features in some malignancies. However, Gal3’s role in promoting USC pathology is lacking. Methods We explored the relationship between LGALS3 levels and prognosis in USC patients using TCGA database, and examined the association between Gal3 levels in primary USC tumors and clinical-pathological features. CRISPR/Cas9-mediated Gal3-knockout (KO) and GB1107, inhibitor of Gal3, were employed to evaluate Gal3’s impact on cell function. Results TCGA analysis revealed a worse prognosis for USC patients with high LGALS3. Patients with no-to-low Gal3 expression in primary tumors exhibited reduced clinical-pathological tumor progression. Gal3-KO and GB1107 reduced cell proliferation, stemness, adhesion, migration, and or invasion properties of USC lines. Furthermore, Gal3-positive conditioned media (CM) stimulated vascular tubal formation and branching and transition of fibroblast to cancer-associated fibroblast compared to Gal3-negative CM. Xenograft models emphasized the significance of Gal3 loss with fewer and smaller tumors compared to controls. Moreover, GB1107 impeded the growth of USC patient-derived organoids. Conclusion These findings suggest inhibiting Gal3 may benefit USC patients.

Formalin-fixed paraffin-embedded tissue representing the primary tumor of from each patient.All patients were confirmed to have uterine serous cancer.The FFPE blocks were sliced to a thickness of 5μm using a microtome, and glass slides were prepared.Tissue sections were subjected to hematoxylin and eosin (H&E) staining and Gal3 immunohistochemistry (IHC) using the same method described before (1).Anti-Gal3 goat polyclonal IgG (R&D Systems, Minneapolis, MN, USA, catalog number: AF1197, 1:100) was validated using Gal3 positive cells and Gal3 knockout (Gal3-KO) cell line (supplementary Fig. S1A-B).Two certified pathologists blinded to the original pathology report and diagnosis, scored Gal3 expression by the percentage of Gal3-positive tumor cells as previously reported (2).The clinical information collected included age at diagnosis, body mass index (BMI), stage, degree of myometrium invasion (<1/2 or ≥1/2), lymphovascular space invasion (LVSI), cervical involvement, serosal invasion, presence of extrauterine lesions (ovarian, omentum, vaginal, parametrium, pelvic or para-aortic lymph nodes, other organs).
The patient derived organoids were maintained as previously described (6).

Generation of Gal3-KO Cells and Knockdown Cells
Oligonucleotides targeting LGALS3 were synthesized, annealed and ligated into the single guide RNA scaffold of the LentiCRISPRv2 (AddGene, Watertown, MA, USA, catalog number: 52961) via the BsmBI sites, according to the method previously described (7).Two different oligo pairs were chosen for construct synthesis targeting the following sequences in exon 3 (sgRNA1) or 4 (sgRNA2) of LGALS3: sgRNA1 GTCTACCCAGGGCCACCCAG and sgRNA2 GCTGATAACAATTCTGGGCA. GGTCTCCCTGAAGCCCTCTGCA.For sgRNA2: forward; GCCTTATCTCTTTGGCCCCTGGG, reverse; CCTGTTTGCATTGGGCTTCACCG).PCR products were sent to AZENTA for Sanger sequence (South Plainfield, NJ, USA).All primers were synthesized by IDT (Coralville, IA, USA).Gal-3 KO cells obtained using sgRNA1 in ARK1 and sgRNA2 in ARK2 and Gal3 control cells (Gal3-CTRL) generated using the same method without gRNA were used in subsequent experiments.

Successful construct generation was verified by
The SKOV3 Gal3 knockdown (Gal3-KD) cell line was generated utilizing the shRNA system and the materials and methodology previously described (8).

Conditioned Media (CM) for add back experiments.
ARK2 Gal3-CTRL and ARK2 Gal3-KO cells were counted and plated at concentrations estimated to reach 70-80% confluence after three days.The day after plating, the complete media was changed out to media containing 10% or 2% FBS.After 48 hours, the conditioned medium was collected and centrifuged at 3500 rpm for 5 minutes.The supernatant was filtered with a 0.2 μm surfactant free cellulose acetate (SFCA) filter (Thermo Scientific, Carlsbad, CA, USA, catalog number: 09-740-105) and applied to the further experiments.
Extracellular vesicles (EVs) extraction: EVs were isolated from 10 ml of ARK2 culture media with sub-confluent condition on a 10 cm dish using Amicon® Ultra-15 Centrifugal Filter Unit (Millipore Sigma, Burlington, MA, USA, catalog number: UFC910024).Debris was excluded from the media by centrifugation and a 0.22 μm sterile syringe filter system.Then, the EVs were isolated by centrifugation (at 3,000 × g for 30 min, rinsed with filtered DPBS, and spun at 3,000 × g for 30 min).All procedures were performed at 4℃.The concentrated EVs were diluted with lysis buffer and subjected to sonification and subjected to ELISA.
ELISA assay: Gal3 concentrations in conditioned media and EVs were measured using the Human Galectin-3 Quantikine ELISA Kit (R&D Systems, catalog number: DGAL30), following the manufacture's protocol.

To investigate the distribution of the Galectin 3
Immunocytochemistry (ICC): Ten thousand cells of ARK1 and ARK2 were seeded on Falcon 4well culture slides (Corning Inc., Kennebunk, ME, catalog number: 354114) in 300 µL of complete media.After overnight incubation, the cells were fixed with 4% paraformaldehyde (PFA) (Biotium, Fremont, CA, USA, catalog number: 22023) at room temperature (RT) for 15 minutes.Subsequently, the cells were rinsed three times with PBS and permeabilized using 0.1% Triton X (Sigma-Aldrich, St. Louis, MO, USA, catalog number: X100), followed by another three rinses.The cells were then incubated with 3% bovine serum albumin (BSA) (Sigma-Aldrich, catalog number: A3294) at RT for 1 hour.Primary antibodies, Anti-Gal3 goat polyclonal IgG (R&D Systems, catalog number: AF1197, 1:300) or Normal Goat IgG Control (R&D Systems, catalog number: AB-108-c), were added to the wells and incubated at RT for 1 hour.After three rinses with PBS and PBST, the cells were incubated with the appropriate secondary antibody, Donkey Anti-Goat IgG H&L (Alexa Fluor® 488) (Abcam, Cambridge, UK, catalog number: ab150129, 1:200), for 1 hour at RT in the dark.Following three rinses with PBS and PBST, the cells were stained with Phalloidin (Invitrogen, 1:400 Alexa Fluor™ 555 Phalloidin, catalog number: A34055) if required, at RT for 1 hour.The cells were then rinsed twice with PBS and mounted with DAPI (Vector Laboratories, Inc., Burlingame, CA, USA, catalog number: H-1200).Finally, the cells were observed and imaged under a 60x magnification using the FV3000 confocal laser scanning microscope (Olympus, Tokyo, Japan).

Assessment of cell viability and proliferation.
Cell counts: 0.1 x 10 6 cells/well were seeded on the 6-well plates (Corning Inc., catalog number: 3516).Cell counts were determined with a TC20 Automated Cell Counter (Bio-Rad Inc, Hercules, CA, USA) at the indicated time points (time zero = O/N incubation, 24, 48, and 72 hours after O/N incubation) in triplicate.Dead cells were excluded using the trypan blue staining (Life Technologies, catalog number: 15250061).
Cell death assessment: 0.1 x 10 6 cells (ARK1 and ARK2 CTL and Gal3-KO) were seeded per well on 6-well plates (Corning Inc., catalog number: 3516) in the complete media.After 24 and 72 hours, all cells (those that were attached and those in the supernatant) were collected and the baseline level of apoptosis and necrosis were assessed by FITC Annexin V Apoptosis Detection Kit with 7-AAD (BioLegend, San Diego, CA., catalog number: 640922).At a minimum, twenty thousand cells per technical replicate was evaluated with a flow cytometer (Gallios Flow Cytometer, Beckman-Coulter, France).The results were analyzed using FlowJo™ v10.8 Software (BD Life Sciences).
MTT assay to measure the chemosensitivity: Cells were harvested at sub-confluency and plated in 96-well plates (Corning Inc., catalog number: 3596) at 1,000 cells per well in 100 μL of media per well.After overnight incubation, 150 μL of the complete media containing each concentration (0, 1, 2.5, 5, 10, 20 µM) of carboplatin (Sigma-Aldrich, St. Louis, MO, USA, catalog number: C2538-250MG) was added to each well.Cell viability was assessed by MTT assay 72 hours after adding the media, as previously described (9).BH3 profiling: To ascertain the degree of involvement of galectin 3 in cell priming towards apoptosis, we conducted BH3 profiling as described previously (10).
Cell cycle assay: Cells were seeded at 0.5 x 10 6 cells per well on the 6 well plate (Corning Inc., catalog number: 3516) and incubated overnight.The cells were rinsed with serum-free medium, and then incubated with serum-free media for 24 hours.After which, the media was removed and the complete media was replaced, and the plates were incubated for an additional 6 hours.The cells were trypsinized, collected and processed using the Cell Cycle Phase Determination Kit (Cayman Chemical Company, Ann Arbor, MI, USA, catalog number: 10009349) as per manufactures instructions and assessed by flow cytometry.At least 20,000 events for each end point measure were evaluated.The FlowJo™ v10.8 Software (BD Life Sciences) cell cycle analysis module (Dean-Jett-Fox model) was used to analyze the percentage of population for each of the subG1, G0/G1 phase, S phase, and G2/M phase.

Investigating the impact of loss/inhibition of Gal3 on EGFR signaling
Wild type ARK1 and ARK2 cells were seeded and cultured in a complete media with 10% FBS.
When cells reached sub confluence, the media was replaced with complete media containing vehicle (Dimethyl sulfoxide: DMSO) (Fisher Scientific Co, catalog number: D136-1) or 10 µM of GB1107 (MedChemExpress, Monmouth Junction, NJ, USA, catalog number: HY-114409), a Gal3 small molecule inhibitor (SMI).The cells were then maintained in culture for 12 or 72 hours.Cell lysates were recovered and were utilized for immunoblot analysis.
Assessment of Gal3 on EGF/EGFR signaling: Wild type ARK1 and ARK2 cells were cultured in serum-depleted media for 6 hours.The cells were then treated with 20 ng/ml of EGF (Life Technologies, catalog number: PHG0311) + vehicle or 10 µM of GB1107.After 30 minutes, cells were harvested in a lysate buffer, and immunoblotting was run for analysis of EGFR/pEGFR and ERK1/2/pERK1/2.

Assessment of cancer stem cell properties and cancer stem cell markers.
Colony forming assay: Five hundred cells per well were seeded into 6-well plates (Corning Inc., catalog number: 3516) and the medium was replaced every two days (day 3, day 6).At the end of the culture period the cells were fixed with methanol and stained with 0.5% crystal violet (GFS chemicals, Columbus, OH, USA, catalog number: 31201).Images of each well were captured, and the average size of the colonies and the average number of colonies were determined by Fiji software version 2.3.0 (11).To examine the add-back effect of Gal3, the colony-forming assay was also conducted by incorporating CM mixed with complete media in a 1:1 ratio instead of using complete media.To further support Gal3 contributes to colony forming ability, 10 µM of GB1107 or vehicle (DMSO) was added to the CM from ARK2 Gal3-CTRL cells and the colony-forming assay was conducted in ARK1 and ARK2 Gal3-KO cell lines as described above.
Sphere forming assay: Media for the sphere-forming assay included DMEM/F12 (Life Technologies, catalog number: 11330032), 1% Pen/Strep, 10 ng/mL of bFGF recombinant human protein (Life Technologies, catalog number: 13256-029), 20 ng/mL of EGF (Life Technologies, catalog number: PHG0311), and B27 (Life Technologies, catalog number: 17504044).Cells were collected and resuspended in sphere-forming media.4 × 10 4 cells (ARK1) or 2 × 10 4 cells (ARK2) per well with 2 mL of the media were seeded in low attachment plates (Corning Inc., catalog number: CLS3473).Cells were cultured for 14 days adding 100 μL of the media with growth factors every two days.After the 14-day-incubation, tumor spheres in the wells were imaged under 40× magnification of EVOS M5000 Imaging System (Thermo Fisher Scientific, San Jose, CA, USA).The average number of the tumor spheres was calculated.
Evaluation of cancer stem cell markers: ARK1 and ARK2 Gal3-CTRL and Gal3-KO cells were collected under sub-confluent status (roughly 70 to 80%) to analyze the percent of cells displaying the cancer stem cell markers (CD44, CD117, CD133, and aldehyde dehydrogenase (ALDH) activity).For assessment of CD44, CD117, and CD133 populations in the control and Gal3-KO cells, the isotype cell lines were harvested, rinsed twice with the staining buffer (BioLegend, catalog number: 420201) and stained with anti-CD44 antibody conjugated with APC (BD Biosciences, catalog number: 559942, 2:100), anti-CD117 antibody conjugated with PE (BD Biosciences, catalog number: 555714, 5:100) or anti-CD133 antibody conjugated with PE (Miltenyi Biotec, Bergisch, Gladbach, Germany, catalog number: 130-113-748, 2:100) at 4°C for 30 minutes.The isotype control for CD44 (BioLegend, catalog number: 401210), CD117 (BioLegend, catalog number 400114), and CD133 (BioLegend, catalog number: 401208) were used as the negative controls.The cells were also stained with DAPI (BioLegend, catalog number: 422801) to distinguish between live dead cells.At least 20,000 live singlet cells per technical replicate were assessed by flow cytometry the results were analyzed using FlowJo™ v10.8 Software (BD Life Sciences).ALDH activity was evaluated by using the ALDEFLUOR Kit (StemCell Technologies, Durham, NC, catalog number: 01700) in accordance with manufacturer's instructions.Briefly, cells were harvested, washed, counted, and suspended in ALDEFLUOR buffer at 1 × 10 6 cells/mL.Cells were divided into two aliquots and incubated with 1.5 mmol/L ALDEFLUOR substrate in the presence or absence of the ALDH inhibitor N,N-diethylaminobenzaldehyde (DEAB).Following a 45-minute incubation at 37°C, the endpoints were assessed via flow cytometer as described above.
Investigating the impact of Gal3 inhibition on Notch1 signaling: Wild type ARK1 and ARK2 cells were seeded and cultured in the organoid culture condition (6).Cells were cultured in the organoid culture media with 10 µM of GB1107 or vehicle (DMSO) for 7 days changing media every other.After 7-days in culture, the cells were collected, and the lysates were utilized for immunoblot analysis.

Exploring the role of Gal3 in metastatic potential
Fibronectin adhesion assay: The fibronectin-coated 96 well-plate was made by adding 80 μL of fibronectin solution (1 μg/mL, Sigma-Aldrich, catalog number: F0556-100UL) and incubating for 1 hour at RT.After removing the solution, wells were rinsed with PBS three times.Ten thousand of ARK1 and ARK2 Gal3-CTRL and Gal3-KO cells per well were seeded with 100 μL of the complete media and incubated for 1 hour at 37℃, 5% CO2.After incubation, wells were rinsed gently with PBS, and cells were fixed with methanol for 15 minutes at RT. Cells were then stained with crystal violet for 20 minutes at RT.After staining, the wells were washed with dH2O, and an image recorded of the areas at the center of each well using a EVOS M5000 Imaging System (Thermo Fisher Scientific) at 40× magnification.The number of adherent cells was counted by Fiji software version 2.3.0 (11).To confirm if the pharmacologic inhibition has a similar effect as Gal3-KO in cell adhesion ability, wild type ARK1 and ARK2 cells were cultured in the media containing 10 µM of GB1107 or vehicle (DMSO) for 72 hours (changed media every other day).Then, cells were collected and submitted to the fibronectin adhesion assay, where DMSO or 10 µM of GB1107 were added to the media.
Trans-well migration assay: After a 24 hour-culture in non-serum containing medium, the cells were trypsinized and resuspended in serum-free medium.In each insert, 1 × 10 5 cells of ARK1 cells in 200 μL of serum-free medium or 0.5 × 10 5 cells of ARK2 were placed in the upper chamber (Corning Inc., catalog number: 353097) and 500 µL of the complete medium with 10% FBS was added to the lower chamber as a chemoattractant.Cells that migrated across the bottom membrane of the insert (Corning Inc., catalog number: 3470) were fixed with methanol and stained with crystal violet staining solution (GFS chemicals Columbus, catalog number: 31201) after 8 hours (ARK1) and 24 hours (ARK2) incubation.Four images were captured at distinct positions on each insert membrane using Nikon ECLIPSE Ni-U (Nikon, Tokyo, Japan) with a magnification of 40×.The Fiji software version 2.3.0 was employed to quantify the number of migrated cells (11).To confirm if the pharmacologic inhibition has a similar effect as Gal3-KO in cell migration ability, wild type ARK1 and ARK2 cells were cultured in the media containing 10 µM of GB1107 or vehicle (DMSO) for 72 hours (changed media every other day).Then, the wild type ARK1 and ARK2 cells were collected and submitted to the transwell migration assay, where DMSO or 10 µM of GB1107 were added to the media.
Invasion assay: Corning BioCoat Matrigel Invasion Chambers with 8.0 µm PET membrane 24well plates were used to determine Gal3 influence on tumor cell invasion (Corning Inc., catalog number: 354480).After 24 hour-culture with non-serum medium, the cells were trypsinized and resuspended in serum-free medium.In each insert, 1 × 10 5 cells in 200 µL of serum-free medium were placed in the upper chamber and 500 µL of the complete medium with FBS was added to the lower chamber as a chemoattractant.Cells that invaded and migrated across the Matrigel™ layer were imaged and analyzed as described above.To confirm if the pharmacologic inhibition has a similar effect as Gal3-KO in cell invasion ability, wild type ARK1 and ARK2 cells were cultured in the media containing 10 µM of GB1107 or vehicle (DMSO) for 72 hours (changed media every other day).Then, cells were collected and submitted to the Matrigel invasion assay, where DMSO or 10 µM of GB1107 were added to the media.

Determining the impact of Gal3 positive and negative tumor cell condition medium on tumor microenvironment.
Angiogenesis invasion assay using HUVECs: HUVECs were cultured as described above.
Biocoat Matrigel Invasion Chambers (Corning Inc., catalog number: 354480) with an 8.0-µm pore polyester membrane were used.The Matrigel™ in each invasion chamber was primed with 200 µL of CM form ARK2 Gal3-CTRL or CM from ARK2 Gal3-KO for 24 hours at 37℃ in a 5% CO2 incubator.Fifty thousand HUVECs were mixed in 200 µL of endothelial basal culture medium with no growth supplement and plated on top of the primed Matrigel™ after the CM was aspirated.Subsequently, 500 µL of endothelial culture medium with growth medium supplement was placed in the lower reservoir below the invasion chamber as an attractant.The samples were incubated for 16 hours, cells that invaded and migrated across the Matrigel™ layer were imaged and analyzed as described above.
Tube-forming assay using HUVECs: HUVECs were cultured as described above.Forty-eight well plates (Thermo Fisher Scientific, catalog number: 150687) were used, and Cultrex with reduced growth factor (RGF) and basement membrane extract (BME) (R&D Systems, catalog number: 3536-005-02) mixed with CM form ARK2 Gal3-CTRL or ARK2 Gal3-KO (1:1 ratio) were plated in each experimental well.Fifty thousand HUVECs were seeded on top of the RGF BME and incubated at 37 ℃, 5% CO2 for 6 hours.Tubes were assessed at 6 hours across all conditions with the EVOS M5000 Imaging System (Thermo Fisher Scientific) and quantified with Wimasis image analysis software (https://www.wimasis.com/en/WimTube).Four representative images were taken per well under 40× magnification.Total tube length and number of branch points were used as the endpoints.
For the add-back experiment using rhGal3 (R&D Systems, catalog number: 1154-GA-050), 5 μg/mL of rhGal3 diluted with PBS containing 0.1% BSA or PBS containing 0.1% BSA was applied to the BME mixed with CM from Gal3-KO and run the same assay.
To further support the impact of Gal3 in tube forming, 10 µM of GB1107 or vehicle (DMSO) was applied to the BME mixed with CM from ARK2 Gal3-CTRL cells and the tube forming assay was conducted as described above.
Assessment of Gal3 influence on the transition of fibroblasts to cancer associated fibroblast (CAFs): IMR90 cells were seeded on Falcon 4 well culture slide (Corning Inc., catalog number: 354114) with 10,000 cells per chamber for ICC or on a Falcon 6 cm dish (Corning Inc., catalog number 353002) with 150,000 cells per dish.After overnight incubation, cells were treated with a 1:1 mixture of the complete media described above and 10% CM (from ARK2 Gal3-CTRL or ARK2 Gal3-KO).After 48 hours of incubation, treated cells were either assessed via ICC or lysates were subjected to western blot.A positive control included cells treated with 10 ng/mL of Transforming Growth Factor-β1 (TGF-β1) (PeproTech, Rocky Hill, NJ, USA, catalog number: 100-21) and the negative controls was vehicle (PBS) alone.ICC was conducted using the same method described before.The Human/Mouse/Rat alpha-Smooth Muscle Actin Antibody (R&D Systems, catalog number: MAB1420) was employed as the primary antibody at a dilution of 1:100, while the same concentration of the Mouse IgG2A Isotype Control (R&D Systems, catalog number: MAB003) was utilized as the isotype control.The F(ab')2-Goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody (Invitrogen, catalog number: A-11017) was employed as the secondary antibody at a dilution of 1:1000.Phalloidin was not utilized in this assay.
After fibroblasts were found to be Gal3-positive, we exposed them to 10 ng/ml TGF-β1 for 48 hours which resulted in reduced Gal3 expression compared with those exposed with vehicle as evidenced by immunoblotting.We then investigated the impact of pharmacological inhibition of Gal3 in fibroblasts and any exogenous Gal3 in CM.To accomplish this, fibroblasts were cultured in complete media with DMSO (vehicle) or 10 µM GB1107, or 1:1 mixed media with Gal3positive CM with DMSO (vehicle) or 10 µM GB1107.After 48 hours the expression level of αSMA was assessed by immunoblotting as the indicator of conversion to CAF phenotype.In vivo limiting dilution tumorigenic assay: ARK1 Gal3-CTRL and Gal3-KO cells were collected, counted, aliquoted and resuspended in 1:1 PBS: Cultrex Basement Membrane Extract, Type 2 (R&D Systems, catalog number: 3532-010-02) (100 µL of PBS: 100 µL of Cultrex Basement Membrane Extract) solution.Then, 1 × 10 4 , 5 × 10 4 , 1 × 10 5 , and 2 × 10 5 cells were injected subcutaneously in the dorsal flank of each mouse using 27-gauge needle.Each dilution group had four mice.Mice were palpated for tumor every two days starting on day 3 post injection to determine time to onset.Once tumors were evident, the date was noted, and mice were weighed and palpated twice a week to monitor tumor growth.Mice were to be euthanized when the tumors reached >15 mm or if the mouse lost more than 15% bodyweight or appeared in distress as indicated by ruffled fur or hunched posture or difficulty moving around the cage.

Confirmation of loss of Gal3 effect using in vivo models and patient derived organoids
Mice were euthanized in accordance with our IACUC protocol.After euthanizing the mice, tumors were collected, weighed and processed for further investigation.
Intraperitoneal tumor implantation model: The intraperitoneal (IP) orthotopic model was incorporated to better appreciate the impact of loss of Gal3 on tumor cell implantation, establishment, and potential growth of 1 million of ARK1 Gal3-CTRL or ARK1 Gal3-KO cells injected in a volume of 200 µL PBS into the peritoneal cavity.Sixty days post injection the mice were euthanized.The peritoneal cavity was imaged to document intraperitoneal disseminated lesions, the lesions were harvested, and total tumor weight was determined.H&E staining and IHC: Xenograft tumor samples were processed, embedded in paraffin and 5 μm sections were prepared.Slides were subjected to H&E, and IHC for Ki67 and phospho-Histone H3 (pHH3).Ki67 (1:100, DAKO, Carpinteria, CA, USA, catalog number: M7240) immunostaining was performed as previously described (1).IHC for pHH3 (1:100, Cell Signaling Technology, Danvers, MA, USA, catalog number: 9701) was performed following the manufacturer's instructions.Simultaneously, staining without primary antibodies was performed (Supplementary Fig. S1C-D.Briefly, after antigen unmasking with 10 mM sodium citrate buffer and quenching the endogenous peroxidase with 3% hydrogen peroxide (Fisher Scientific Co, Pittsburgh, PA, USA, catalog number: H325) in dH2O, nonspecific background was blocked with 6% BSA in 0.1% PBS-T.Primary antibody was diluted in 6% BSA and incubated 1 hour at RT.After washes, sections were incubated with secondary antibody (anti-rabbit IgG conjugated with HRP, 1:50, Santa Cruz Biotechnology, Dallas, TX, USA, catalog number: sc-2357), washed and incubated with a DAB substrate (DAKO, catalog number: SK-4100) and counterstained with hematoxylin.
Confirmation of the effects of Galectin 3 small-molecule inhibitors: The impact of Gal3 smallmolecule inhibitors (SMIs), specifically TD139 (Selleck Chemicals, Houston, TX, USA, catalog number: S0471) and GB1107, on cell viability was assessed through MTT and cell death analysis.MTT assay was performed using the previously described methodology, investigating the effects of TD139 and GB1107 at concentrations of 0, 1, 5, 10, 25, and 50 µM for 72-hour incubation.
For cell death analysis, a concentration of 10 µM of each Gal3-SMI was utilized, followed by a 72-hour incubation period and subsequent analysis same as described previously.
The off-target effects of these Gal3-SMIs were confirmed by conducting a similar MTT assay on both Gal3-CTRL and Gal3-KO cells of ARK1 and ARK2.
Colony formation assay was performed in the presence or absence of 10 µM of each Gal3-SMI for 72 hours.The culture medium containing Gal3-SMIs was changed out every other day.The other procedure was same as described.
Investigation of Gal3 pharmaceutical inhibition effect using patient-derived organoids: Organoids were cultivated on the 48-well plate (Corning Inc., catalog number: 3548) using the methodology explicated above.The organoids were seeded in Matrigel (Corning Inc., 356231) and subsequently cultured in the organoid culture medium, which contained 10 µM of GB1107 or the vehicle (DMSO).The media was replenished every other day, and the entire area occupied by the organoids was quantified observed using an Incucyte Live Cell System and Incucyte's automated Organoid Software Analysis Module (Essen BioScience, Ann Arbor, MI, USA).

Immunoblot analysis
Immunoblot analysis was performed as previously described (1) The dilution for all primary antibodies was 1:1000 except for Galectin 3 (1:2000).After incubation with the primary antibody, blots were incubated with the corresponding secondary antibody at RT for 1 hour.Then blots were developed using a chemiluminescent detection reagent (ECL reagent.Prometheus ProSignal, Genesee Scientific, San Diego, CA, USA, catalog number: 20-301B or Clarity Max™ Western ECL Substrate: Bio Rad Inc, catalog number: 1705062).Each blot was stripped using Re-Blot Plus Strong Antibody Stripping Solution (Millipore Sigma, catalog number: 2504) when the same membrane was used to reprobe with another primary antibody.Images were acquired using Bio Rad Chemi Doc Imaging System (Bio-Rad Inc) and analyzed using Bio Rad Image Lab Software (ver.6.1.0)(Bio-Rad Inc).
Sanger sequencing of the sgRNA region.Transfections of ARK1 and ARK2 lines were performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA, catalog number: 11668019) with co-transfection of pLenti-C-mGFP-P2A-Puro Lentiviral Gene Expression Vector (OriGene Technologies, Inc. Rockville, MD, USA, catalog number: PS100093) to detect the cells which were transfected.The GFP positive cells were isolated by flow cytometry (FACSAria™ Fusion Flow Cytometer, BD Life Sciences, San Jose, CA, USA) and single cells were plated in individual well in 96 well plate.Lysates of individual CRISPR-Cas9-edited clones were evaluated by western blot for Gal3 expression.Biallelic knockout through indel generation was confirmed in each line following PCR amplification of genomic DNA extracted using DNeasy Blood & Tissue Kits (QIAGEN, Hilden, Germany, catalog number: 69506) with primers flanking the guide RNA target of the LGALS3 gene (For sgRNA1: forward; CTCAAGGATGGCCTGGCGCATG, reverse; Animal protocols: Animal procedures were approved by the Massachusetts General Hospital Institution Animal Care and Use Committee and were performed according to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (Protocol number: 2017N000236).Female NSG (NOD scid gamma) mouse, 6 -8 weeks age, were obtained from Steele Laboratories at MGH.The number of mice used in our in vivo experiments was based on preliminary experiments conducted using these specific cell lines.The experiments did not require blinding and randomization.