Cell culture and cell lines
The following CRC cell lines were obtained from the American Type Culture Collection (ATCC): HCT116 (ATCC, CCL-247), DLD1 (ATCC, CCL-221), RKO (ATCC, CRL-2577), HCT8 (ATCC, CCL-244), NCI-H508 (ATCC, CCL-253), Colo205 (ATCC, CCL-222), SW620 (ATCC, CCL-227), SW48 (ATCC, CCL-231) and HCT15 (ATCC, CCL-225). The DiFi, GP5D and HT55 cell lines were provided by J. Engelman, and the VACO432 cell line was provided by S. Markowitz. Short tandem repeat profiling was performed to authenticate all cell lines. Cells were cultured in high-glucose DMEM (Fisher Scientific, 11-965-118) supplemented with 10% FBS (Fisher Scientific, 26-140-079). All cultures were maintained at 37 °C in a humidified atmosphere containing 5% CO2.
In vitro sugar experiments
Three media formulations were used for in vitro sugar experiments. The Glu + Fru condition contained 10 mM glucose and 10 mM fructose. The Glu condition contained 20 mM glucose, and the Fru condition contained 20 mM fructose. All media were prepared using glucose-free DMEM (Fisher Scientific, 11-966-025) supplemented with 10% dialysed FBS (Fisher Scientific, SH30079.03). d-Glucose (Sigma-Aldrich, G8270) and d-fructose (Sigma-Aldrich, F0127) were added to achieve the desired concentrations. For isotope-labelling metabolomics, media were prepared by replacing d-glucose with [U-13C6] d-glucose (Cambridge Isotope Laboratories, CLM-1396-10) or d-fructose with [U-13C6] d-fructose (Cambridge Isotope Laboratories, CLM-1553-PK).
Migration and invasion assays using transwell
The migration and invasion assays were performed as previously described44. For the migration assay, 100 μl of CRC cells (6 × 105 for DLD1, 8 × 105 for HCT116, 4–5 × 105 for HCT8, 5–8 × 105 for RKO and 5 × 105 for the remaining nine CRC cell lines) in serum-free assay medium was seeded into the cell culture insert (upper chamber) of a transwell (Fisher Scientific, 07-200-150). The lower chamber was filled with 600 μl of assay medium containing 10% dialysed FBS (Fisher Scientific, SH30079.03). For chemical treatments, compounds were added to both the upper and lower chambers. After incubation for 48–72 h (48–72 h for DLD1 and HCT116, 48 h for HCT8, 60–72 h for RKO and 48 h for the remaining nine CRC cell lines), the bottom surfaces of the inserts were stained with 0.1% crystal violet (Sigma-Aldrich, C0775) in 20% methanol for 10 min. Non-migrated cells on the upper side of the membrane were removed with cotton swabs before analysis.
For the invasion assay, 40 μl of 15% Matrigel (Fisher Scientific, CB40230C) in serum-free assay medium was added to the insert, and the plate was incubated at 37 °C for 30 min before cell seeding. Cells were seeded using the same numbers as in the migration assay, and the same assay medium was added. The incubation time for the invasion assay was 24–48 h longer than for the migration assay. After incubation, the bottom surfaces of the inserts were stained with 0.1% crystal violet (Sigma-Aldrich, C0775) in 20% methanol for 10 min, and non-invaded cells on the upper side of the membrane were removed with cotton swabs before analysis.
Unless otherwise specified, the assay medium used in both migration and invasion assays was glucose-free DMEM supplemented with either 10 mM glucose and 10 mM fructose (Glu + Fru condition) or 20 mM glucose (Glu condition), and 10% dialysed FBS. The compounds used in these assays included fluvastatin (Sigma-Aldrich, PHR1620-1G), FK866 (APExBIO, A4381), (±)-mevalonolactone (a cell-permeable form of mevalonate; Sigma-Aldrich, M4667-1G), metformin (Cayman Chemical, 13118), α-ketobutyrate (Sigma-Aldrich, K401-5G) and nicotinamide (Sigma-Aldrich, N0636). All chemical doses used for migration assays were optimized to minimize cytotoxicity, and cell numbers were measured in parallel using the SYBR Green assay under the same treatment conditions. Although nicotinamide and mevalonolactone did not induce cytotoxicity, FK866, metformin and fluvastatin showed mild cytotoxic effects in certain cell lines, even at optimized concentrations. Therefore, migration data were normalized to cell number to account for potential proliferation differences, as presented in the Extended Data. Significant suppression of migration persisted after normalization, confirming that reduced motility was not solely caused by effects on cell viability. After drying the membranes, images were acquired by microscopy and quantified by measuring the stained area using ImageJ (v.1.54 f).
Cell growth and viability assays
For the growth assay, CRC cells were seeded into 96-well plates (4,000 cells per well) and cultured for 1–5 days. After incubation, either the SYBR Green I assay or CellTiter-Glo Luminescent Cell Viability Assay (Promega, G7571) was used to quantify cell growth, depending on the experiment. For the SYBR Green I assay, the medium was removed and 50 μl of 0.2% SDS was added to each well, followed by 150 μl of diluted SYBR Green I solution (1:750 in water) (Fisher Scientific, S-7567). Fluorescence was measured using a plate reader (BioTek, Synergy H1) at excitation 485 nm and emission 528 nm. For experiments using CellTiter-Glo, the assay was performed according to the manufacturer’s instructions.
For cell viability measurements in transwell-based migration or invasion assays, the same number of cells used in the transwell inserts was seeded into 24-well plates. Cells were cultured under identical conditions (100 μl serum-free assay medium and 600 μl assay medium with 10% dialysed FBS) and for the same durations as the corresponding migration or invasion assays. After incubation, 200 μl of 0.2% SDS was added to each well. Cell lysates were sonicated for five cycles (2 s on, 2 s off, 30% amplitude) using a probe sonicator (Active Motif EpiShear, Q120AM) to homogenize the lysate. Then, 50 μl of the lysate was transferred to a new well, and 150 μl of diluted SYBR Green I solution (1:750 in water) was added. Fluorescence was measured as described above using the same plate reader.
Colony formation assay
The clonogenic ability of CRC cells was measured using a previously described protocol45. In brief, 500–2,000 CRC cells per well were seeded into six-well plates under Glu + Fru or Glu conditions and cultured for 10–14 days. The medium was then removed, and each well was washed twice with 2 ml of 1× PBS. Cells were fixed with 400 μl of methanol at room temperature (20–25 °C) for 20 min. After removing the methanol, 1 ml of 0.5% crystal violet (Sigma-Aldrich, C0775) in 25% methanol was added to each well and incubated for 40 min. The stain was then removed, and the wells were washed with deionized water. Plates were dried overnight at room temperature, and individual colonies in each well were manually counted after image acquisition and marking using ImageJ (v.1.54f).
Laboratory animals and sugar treatments
Male NU/J (athymic) mice (7 weeks old ) were purchased from The Jackson Laboratory (strain no. 002019), housed in a modified barrier facility and used for caecum orthotopic injection of CRC cells. Male and female NOD scid gamma (NSG) mice were purchased from The Jackson Laboratory (strain no. 005557) and housed in a high-barrier facility optimized for immunodeficient mice. Parental and F1 generation NSG mice were used for breeding, and F1 and F2 generation male mice were used for colon mucosal and splenic injection of CRC cells. All mice were housed in a controlled environment (22 ± 1 °C, 60–70% humidity, 12 h light–dark cycle) and fed ad libitum with standard chow (Picolab Rodent Diet 20, 5053).
For sugar treatment experiments, mice were given ad libitum access to either tap water, glucose solution (100 g glucose in 400 ml autoclaved tap water; 25% w/v) or Glu + Fru solution (45 g glucose + 55 g fructose in 400 ml autoclaved tap water; 25% w/v). These special water bottles were changed twice per week. In separate studies, mice received daily oral gavage with Glu + Fru (45 mg glucose + 55 mg fructose in 400 μl tap water), glucose alone (100 mg in 400 μl tap water) or tap water alone (400 μl). All tap water used for mouse treatments was autoclaved.
All animal experiments were approved by the Institutional Animal Care and Use Committees (IACUC) of Baylor College of Medicine (BCM) and the MD Anderson Cancer Center. The maximal tumour burden permitted by the IACUC (≤5 mm diameter for oral, head, neck and osseous tumours) was not exceeded in this study.
Caecum injection mouse model
The caecum orthotopic injection of CRC cells into the NU/J mice was conducted as previously described46. Male NU/J mice (8 weeks old) with similar body weights were randomly assigned to experimental groups. Before surgery, mice were administered preemptive analgesia with buprenorphine ER (ZooPharm LLC, Rx 255459) and meloxicam (Covetrus, 11695-6936-1), and anaesthetized with 3% isoflurane (Covetrus, 11695-6777-2) in 2 l min−1 oxygen. The abdominal surgical site was sterilized, and a 1–2 cm incision was made in the skin and abdominal wall musculature to expose the caecum. A single-cell suspension of 2 × 106 CRC cells in 30 μl of 100% prechilled Matrigel (Corning, CB40230C) was carefully injected into the caecal wall using 29-gauge insulin syringes (BD Medical, 324702). Excess tumour cells were gently removed using saline-moistened sterile gauze. The caecum was returned to the abdominal cavity, and the incision was closed using absorbable polyglycolic acid 4-0 suture (Ethicon, J303H) and wound clips (CellPoint Scientific, 201-1000). Analgesic drugs were administered for 3 days postoperatively, and wound clips were removed after 10 days. Body weight and food and water consumption were recorded weekly. After 4–6 weeks, all mice from the same cohort were killed on the same date, and necropsies were performed to examine tumour metastasis and collect tissues.
Intrasplenic injection mouse model
Intrasplenic injection of CRC cells was conducted following the previously described protocol47. NSG mice (8 weeks old) with similar body weights were randomly assigned to experimental groups. Before surgery, mice were administered preemptive analgesia with buprenorphine ER (ZooPharm LLC, Rx 255459) and meloxicam (Covetrus, 11695-6936-1), and anaesthetized with 3% isoflurane (Covetrus, 11695-6777-2) in 2 l min−1 oxygen. The left abdominal side, previously shaved, was sterilized using three alternating scrubs with 5% povidone–iodine solution (Veterinary Betadine, 67618-154-16) and alcohol prep pads (Covidien, 6818). A 1 cm incision was made in the skin and abdominal wall musculature at the estimated location of the spleen. The spleen was exposed and gently ligated with absorbable polyglycolic acid 4-0 suture. A single-cell suspension of 7 × 105 CRC cells in 30 μl of PBS was injected into the spleen using a 30-gauge syringe (BD, 305106). The spleen was then isolated with the suture and removed using a low-temperature cautery (World Precision Instruments, 500390). The abdominal wall was closed with absorbable suture (Ethicon, J303H), and the skin was closed with wound clips (CellPoint Scientific, 201-1000). Analgesic drugs were administered for 3 days postoperatively, and wound clips were removed after 10 days. Body weight and food and water consumption were recorded weekly. The Lumina In Vivo Imaging System (IVIS) was used to monitor the bioluminescence of labelled cells each week. After 4–6 weeks, all mice from the same cohort were killed on the same date, and necropsies were performed to examine tumour metastasis and collect tissues.
Colonic mucosal injection mouse model
The colonoscopy-based colonic mucosal injection of CRC cells was performed as previously described18. First, the colonoscopy system—consisting of a miniature endoscope (1.9 mm outer diameter), a xenon light source, a triple-chip camera and an air pump (Karl Storz, Tuttlingen, Germany)—was set up. NSG mice (8 weeks old) with similar body weights were randomly assigned to experimental groups. Before the experiment, mice were anaesthetized using 3% isoflurane (Covetrus, 11695-6777-2) in 2 l min−1 oxygen, and the colonoscope was inserted about 2 cm into the rectum. A sterile 30-gauge needle attached to plastic tubing (provided with the colonoscope) was inserted into the working channel, and the needle was advanced into the colonic wall at a 45-degree angle. A single-cell suspension of 1 × 105 CRC cells in 40 μl of PBS was injected into the colonic wall (20 μl per site, two sites per mouse). Body weight and food and water consumption were recorded weekly. Colonoscopy was used weekly to visualize colonic tumours after cell implantation. After 4 weeks, all mice from the same cohort were killed on the same date, and necropsies were performed to examine tumour metastasis and collect tissues.
Measurement of in vivo luminescence in mice using IVIS
Mice implanted with luciferase-expressing cancer cells were intraperitoneally injected with 100 μl of 15 mg ml−1 luciferin (Gold Biotechnology, LUCK-1G) per mouse. Then, 7 min later, mice were anaesthetized using the XGI-8 Gas Anesthesia System (PerkinElmer, XGI-8) with 3% isoflurane (Covetrus, 11695-6777-2) in 2 l min−1 oxygen. Then, 10 min after luciferin injection, mice were transferred to the Lumina IVIS (PerkinElmer). Images with exposure times of 1 s, 5 s, 10 s, 30 s, 1 min and 2 min were acquired according to the equipment manual and used for further analysis. For liver imaging, 10 min after luciferin injection, mice were killed using CO2, and livers were immediately collected and placed in the Lumina IVIS. Images with exposure times of 2 s, 5 s and 10 s were recorded and used for further analysis. Total luminescence values were quantified using Aura Software (Spectral Instruments Imaging, Aura Software, v.4.0.7).
Drug treatments of mice
In the FK866 treatment experiment, mice were treated with 10 mg kg−1 FK866 (APExBIO, A4381) in PBS (Corning, MT 21-040-CV) containing 2% dimethylsulfoxide (DMSO) (Sigma-Aldrich, D4540) by intraperitoneal injection. In the simvastatin treatment experiment, mice were treated with 30 mg kg−1 day−1 simvastatin (Selleckchem, S1792) in PBS containing 10% DMSO by oral gavage. During the drug treatment period, mice had ad libitum access to the Glu + Fru solution (25% w/v, 45:55 ratio) in their drinking water.
H&E and immunohistochemistry staining
Mouse tissues were placed in tissue cassettes (Thermo, 22-272417) and fixed in 4% paraformaldehyde in PBS (Santa Cruz, sc-281692) for 24 h at 4 °C. Tissues were then transferred to 70% ethanol (VWR, 89125-188) before processing. Tissue processing, embedding, sectioning and H&E staining were performed by the BCM Breast Cancer Histology Core at BCM. Stained slides were scanned by HistoWiz for bright-field whole-slide scanning.
Human tissue slides for H&E and SORD immunohistochemistry (IHC) staining were obtained from the Human Tissue Acquisition and Pathology (HTAP) Services at BCM. All original tumour tissues and matched normal tissues (not all samples included matched normal tissues) were collected at the Michael E. DeBakey Veterans Affairs Hospital, Ben Taub General Hospital and Baylor St. Luke’s Medical Center. Tissue acquisition was based on a system designed to maximize the number of tissues available from BCM-affiliated institutions. The process began with a patient tissue advocate who identified all tissues with the potential to be banked for HTAP each day. A redundant consenting protocol, which included multiple attempts to obtain patient consent, was used. After collection, tissues were barcoded, organized, catalogued and banked in the tissue database. H&E and IHC staining of human tissues were performed by the BCM Breast Cancer Histology Core. The SORD antibody (Sigma-Aldrich, HPA040621) used for IHC staining was validated on SORD KO and SORD-proficient xenograft mouse tissues before application on human slides to ensure antibody specificity. Stained slides were scanned by HistoWiz for bright-field whole-slide scanning.
Human colon tumour organoid and normal colon organoid culture
Human colon tumour organoids were provided by M. L. Martin at Weill Cornell Medicine (the Englander Institute for Precision Medicine), and human normal colon organoids (also called enteroids) were acquired from BCM (Gastrointestinal Experimental Model Systems Core). All 3D cultures of organoids were maintained in 15 μl per drop, totalling five drops (75 μl per well) of Matrigel (Gibco, CB40230C) in 12-well plates (Life Technologies, 150628) with 1 ml of the following medium in a 37 °C, 5% CO2 humidified incubator. The medium was changed every 2 days, and organoids and enteroids were passaged at a 1:6 ratio when crowded.
The medium for culturing human colon tumour organoids contained the following components: Advanced DMEM/F12 (Gibco, 12634010), B27 Supplement (Invitrogen, 17504001) at 1× concentration, l-glutamine (Gibco, 25030164) at 2 mM, HEPES (Amresco, J848) at 10 mM, Primocin (Invitrogen, NC9141851) at 100 μg ml−1, human recombinant EGF (R&D Systems, 236-EG-200) at 50 ng ml−1, Gastrin I (Sigma-Aldrich, SCP0152) at 10 nM, Y-27632 (cAMP inhibitor; Selleck Chemicals, S1049) at 10 μM, A-83-01 (TGF-β inhibitor; R&D Systems, 2939) at 500 nM, SB202190 (MAPK inhibitor; Selleck Chemicals, S1077) at 3 μM, nicotinamide (Sigma-Aldrich, N3376) at 10 mM, N-acetylcysteine (Sigma-Aldrich, A9165) at 1.25 mM, prostaglandin E2 (Sigma-Aldrich, P0409) at 10 nM, Noggin (10% conditioned medium), R-spondin (10% conditioned medium) and penicillin–streptomycin (Gibco, 15140-122) at 1× concentration.
The medium for culturing human colon organoids contained the following components: Advanced DMEM/F12 (Gibco, 12634010), B27 Supplement (Invitrogen, 17504001) at 1× concentration, l-glutamine (Gibco, 25030164) at 2 mM, HEPES (Amresco, J848) at 10 mM, Primocin (Invitrogen, NC9141851) at 100 μg ml−1, human recombinant EGF (R&D Systems, 236-EG-200) at 50 ng ml−1, Gastrin I (Sigma-Aldrich, SCP0152) at 10 nM, Y-27632 (cAMP inhibitor; Selleck Chemicals, S1049) at 10 μM, A-83-01 (TGF-β inhibitor; R&D Systems, 2939) at 500 nM, SB202190 (MAPK inhibitor; Selleck Chemicals, S1077) at 10 μM, nicotinamide (Sigma-Aldrich, N3376) at 10 mM, N-acetylcysteine (Sigma-Aldrich, A9165) at 1 mM, prostaglandin E2 (Sigma-Aldrich, P0409) at 10 nM, penicillin–streptomycin (Gibco, 15140-122) at 1× concentration, Noggin conditioned medium (10%), R-spondin conditioned medium (10%) and WNT-3A conditioned medium (30%).
Human enteroid 2D monolayer differentiation
The collagen IV stock solution was prepared by dissolving collagen IV (1 mg ml−1; Sigma-Aldrich, C5533-5MG) in 100 mM acetic acid (0.6% acetic acid in H2O). A 24-well plate was coated with 500 μl per well of the collagen working solution (prepared by diluting the collagen IV stock 1:30 (v/v) in ice-cold water) and incubated at 37 °C for 1.5 h. Three-dimensional cultured enteroids were dissociated using 500 μl per well of 0.5 mM EDTA (Fisher, BP24821) diluted in PBS. Enteroids from one well of a 12-well plate (approximately 75 μl) were used to seed one well of a 24-well plate. After centrifugation at 300g at 4 °C for 5 min, the pellet was resuspended in 500 μl of 0.05% trypsin and 0.5 mM EDTA and incubated at 37 °C for 5 min. Then, 1 ml of DMEM/F12 supplemented with 10% FBS was added, and the mixture was pipetted 60 times to dissociate the cells. Following centrifugation at 400g at 4 °C for 5 min, the enteroid pellet was resuspended in 500 μl of enteroid medium and seeded onto the coated plate. After 1 day, the medium was replaced with differentiation medium lacking R-spondin, WNT-3A, nicotinamide and SB202190, and the percentage of Noggin conditioned medium was reduced to 5%. The medium was refreshed every 2 days.
Reverse transcription and qPCR
RNA from organoids, enteroids and 2D-differentiated enteroids was extracted using the RNeasy Plus Mini Kit (Qiagen, 74136), and cDNA was synthesized using the SuperScript IV VILO Kit (Fisher, 11756050) following the manufacturer’s instructions. Real-time qPCR was performed using amfiSure qGreen Q-PCR Master Mix (GenDEPOT, Q5600-010) on a CFX96 Real-Time System (Bio-Rad). Primers for qPCR included ACTB1 F, GCAAAGACCTGTACGCCAAC; ACTB1 R, ACATCTGCTGGAAGGTGGAC; SORD F, GGCTCTGAGATGACCACCGT; SORD R, GGTCACACTTGAGCATGATTTTCA; LGR5 F, CCTGCTTGACTTTGAGGAAGACC; and LGR5 R, CCAGCCATCAAGCAGGTGTTCA. All primers were ordered from Thermo Fisher Scientific.
Plasmids for gene KO and knockdown of CRC cell lines and LbNOX
CRISPR–Cas9-mediated genome editing was used to generate SORD KO cell lines48. The gRNA sequences (sgSORD-1, AGCAAAGTGACCATCCCGAT; sgSORD-2, TTGTTTAGGGCCAATCGGGA; sgAKR1B1-1, TCAGGTCGCTGAGTGTCTTC; sgAKR1B1-2, TCCCATACCTTAAAGCCAGT) were cloned into the lentiCRISPRv2-puro plasmid (Addgene, 98290) using the restriction enzyme BsmBI (Thermo, FD0454). The plasmids were then packaged into lentivirus and transduced into CRC cell lines.
Pre-designed short hairpin RNA (shRNA) plasmids for SORD knockdown—including SORD shRNA1 (TRC Clone ID, TRCN0000028069; target sequence, GAGAACTATCCTATCCCTGAA), SORD shRNA2 (TRC Clone ID, TRCN0000028052; target sequence, GCCAATCGGGATGGTCACTTT) and SORD shRNA3 (TRC Clone ID, TRCN0000028106; target sequence, CGTCCAAGTCTGTGAATGTAA)—were purchased from MilliporeSigma. These plasmids were also packaged into lentivirus and transduced into CRC cell lines.
The LbNOX open reading frame was amplified from pUC57-LbNOX (Addgene, 75285) by PCR20,21 using the following primers: F, TCTAGAATGAAGGTCACCGTGGTCGG; R, GTTCGAATTACTTGTCATCGTCATCCTTGTAA. The open reading frame was then cloned into the pCDH-EF1-MCS-IRES-copGFP plasmid (System Biosciences, CD530A-2) for LbNOX overexpression. Plasmids were transfected into cell lines using Lipofectamine (Invitrogen, 100022050 and 100022057). GFP-positive cells were sorted by flow cytometry (Beckman, CytoFLEX SRT) and used for subsequent experiments.
Lentivirus package
Plasmids were co-transfected with d8.2 (Addgene, 8455) and VSVG (Addgene, 8454) at a ratio of 1:1:0.1 into HEK293T cells using threefold (m/m) polyethylenimine (Sigma-Aldrich, 764965). After 2 days of culture at 37 °C with 5% CO2, the medium was collected, filtered through a 0.45 μm filter (Millipore, SLHVM33RS) and stored at −80 °C before use. HCT116, DLD1, HCT8 and RKO CRC cell lines were infected with the packaged lentivirus, and positive cells were selected with puromycin (InvivoGen, ant-pr-5) at the following concentrations: 2 μg ml−1 for DLD1, 1 μg ml−1 for HCT116, 15 μg ml−1 for HCT8 and 1 μg ml−1 for RKO.
Immunoblotting
Proteins were extracted from tissues or cells using RIPA buffer (Cell Signaling Technology, 9806S) containing a protease inhibitor cocktail (GenDEPOT, P3100-001), separated on 4–20% Criterion TGX Stain-Free Gels (Bio-Rad, 5678094 and 5678095) and transferred to 0.45 μm PVDF membranes (MilliporeSigma, IPVH00010). Membranes were probed with primary antibodies at 4 °C overnight and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies at room temperature for 1 h. Protein bands were visualized using ECL western blotting substrates (Bio-Rad, 1705060 and 1705062) and imaged using the ChemiDoc Imaging System (Bio-Rad).
The antibodies used included SORD (1:1,000; Proteintech, 15881-1-AP), AKR1B1 (1:1,000; Proteintech, 15439-1-AP), β-actin (1:2,000; Cell Signaling Technology, 3700), goat anti-rabbit IgG(H+L)-HRP (1:5,000; GenDEPOT, SA002) and goat anti-mouse IgG(H+L)-HRP (1:5,000, GenDEPOT, SA001).
NAD⁺/NADH ratio measurement by enzymatic assay
Cells were plated in six-well dishes with standard DMEM + 10% FBS and cultured overnight. Treatments were applied the following day and continued for 24 h. Cells were then washed with PBS and extracted in 200 μl of ice-cold lysis buffer (1% dodecyltrimethylammonium bromide (Sigma-Aldrich, D8638) in 0.2 N NaOH), diluted 1:1 with PBS and centrifuged at 15,000g for 5 min. For NADH measurement, 20 μl of lysate was transferred to PCR tubes and incubated at 75 °C for 30 min. For NAD⁺ measurement, 20 μl of lysate was transferred to PCR tubes, then 20 μl of lysis buffer and 20 μl of 0.4 N HCl were added, followed by incubation at 60 °C for 20 min. After the respective incubations, samples were cooled to room temperature and quenched with the appropriate neutralizing solution: 20 μl of 0.25 M Tris in 0.2 N HCl for NADH and 20 μl of 0.5 M Tris base for NAD⁺. Following sample preparation, enzyme-linked luminescence-based detection of NAD⁺ and NADH was performed according to the manufacturer’s instructions using the NAD⁺/NADH-Glo Assay Kit (Promega, G9071).
Peredox NADH biosensor to measure NAD+/NADH ratio
The cytosolic Peredox NADH biosensor plasmid (Addgene, 32383) was purchased from Addgene and transiently transfected into CRC cell lines using Lipofectamine 3000 (Thermo Fisher, L3000008) 24 h before sugar treatment. Transfected cells were incubated under Glu + Fru conditions for 24 h. Green (excitation, 395/25 nm; emission, 525/34 nm) and red (excitation, 555/15 nm, emission, 605/52 nm) fluorescence channels were acquired using the Eclipse Ti2 inverted Nikon microscope equipped with the Photometrics Prime 95B Scientific CMOS monochrome camera, Lumencor Spectra X light engine and OKO Labs micro-incubator, located at the Basic Sciences Research Building Microscopy Core, Department of Genetics, MD Anderson Cancer Center. Fluorescence intensities were quantified using NIS-Elements Imaging Software (v.5.42.03) as previously described19,49
Metabolomics and isotope-labelled metabolomics
Carbon metabolism metabolites, including intermediates from glycolysis and the TCA cycle as well as sugars such as glucose, fructose and sorbitol, were detected by LC–MS using a previously established protocol50,51. For isotope-labelling experiments, cells were treated with glucose-free DMEM (Gibco, 11966025) supplemented with either unlabelled d-fructose (Sigma-Aldrich, F0127) and d-glucose (Sigma-Aldrich, G8270) or 13C6 d-fructose (Cambridge Isotope Laboratories, CLM-1553-PK) and 13C6 d-glucose (Cambridge Isotope Laboratories, CLM-1396-10). Cell extracts were prepared by adding 400 μl of 40:40:20 acetonitrile (ACN):methanol:water containing 0.5% formic acid to each well of a six-well plate, followed by neutralization with 35.2 μl of 15% ammonium bicarbonate. After centrifugation at 20,000g for 15 min at 4 °C, the supernatants were transferred to sample vials (Thermo, 2-SVWGK) for LC–MS analysis. High-performance LC was performed using an XBridge BEH Amide XP Column (Waters, 186006724). Mobile phase A consisted of 5% ACN with 20 mM ammonium acetate and 20 mM ammonium hydroxide (pH 9.75–9.85); mobile phase B was 100% ACN. The column temperature was maintained at 25 °C, and the flow rate was 0.15 ml min−1. The LC gradient was as follows: 0 min, 85% B; 2 min, 85% B; 3 min, 80% B; 5 min, 80% B; 6 min, 75% B; 7 min, 75% B; 8 min, 70% B; 9 min, 70% B; 10 min, 50% B; 12 min, 50% B; 13 min, 25% B; 16 min, 25% B; 18 min, 0% B; 23 min, 0% B; 24 min, 85% B; and 30 min, 85% B. MS was performed on a Q Exactive Orbitrap Mass Spectrometer (Thermo) in both positive and negative ion modes. Data were processed and analysed using TraceFinder software (Thermo, OPTON-31001).
Untargeted global metabolomics analysis using LC–tandem MS (MS/MS), measuring more than 700 metabolites, was conducted as previously described9. Specifically, 600 μl of −20 °C 3 mM antioxidant monobromobimane (Fisher Scientific, M20381) in methanol (80:20, v/v) was added to cell samples and incubated for 2 h at room temperature, followed by 1 h of incubation at 0 °C. Samples were then centrifuged for 15 min at 16,000g, and supernatants were transferred to clean tubes. This extraction step was repeated two additional times, and the combined supernatants were dried using a SpeedVac (Savant) and stored at −80 °C until analysis. For normalization, post-extraction tissue or tumour pellets were solubilized in 800 μl of 0.2 M aqueous NaOH at 95 °C for 60 min, and protein content was quantified using the BCA Assay (Thermo, PI23227). For metabolite analysis, dried extracts were reconstituted in ACN (70:30, v/v) containing 0.025% acetic acid, and 3 μl of the solution was injected for LC–MS. Metabolite profiling was performed using an Agilent 1200 LC system coupled to an Agilent 6230 time-of-flight mass analyser. Chromatographic separation was achieved using aqueous normal-phase gradient separation on a Diamond Hydride column (Microsolv). Mobile phase A consisted of 6 μM EDTA and 0.025% acetic acid in isopropanol (50:50, v/v); mobile phase B consisted of 6 μM EDTA and 5 mM ammonium acetate in ACN (90:10, v/v). The following LC gradient was applied: 0–1.0 min, 99% B; 1.0–15.0 min, 20% B; 15.1–29.0 min, 0% B; and 29.1–37.0 min, 99% B. Both positive and negative ion mass spectra were acquired in 2 GHz (extended dynamic range) mode at 1.41 spectra per second over a mass/charge range of 40–1400 m/z. Data were saved in both centroid and profile modes using Agilent MassHunter Workstation B.06.00 Data Acquisition Software. Raw data files were analysed using Mass Profiler Professional (Agilent, v.B.14.5) and MassHunter Profinder (Agilent, v.B.08.00).
For targeted metabolomics, cell lysates were extracted twice with 500 μl of 80% ice-cold methanol. After centrifugation at 20,000g for 20 min at 4 °C, the supernatants were dried and reconstituted in 15 μl of LC–MS-grade water, then transferred to sample vials (Thermo, 2-SVWGK). LC–MS/MS and data analysis for untargeted polar metabolomics were performed by the BIDMC Mass Spectrometry Core Facility at Beth Israel Deaconess Medical Center, as previously described52. Absolute quantification of mevalonate pathway metabolites was performed by Creative Proteomics using LC–MS. In brief, samples were extracted with 80% methanol as described above, reconstituted in 20 μl of high-performance LC-grade water and 5–7 μl was injected into a hybrid 6500 QTRAP triple quadrupole mass spectrometer (AB SCIEX) coupled to a Prominence UFLC HPLC system (Shimadzu). Metabolites were analysed by selected reaction monitoring (SRM), targeting a panel of 300 endogenous water-soluble metabolites. Some metabolites were detected in both positive and negative ion modes, yielding a total of 311 SRM transitions using polarity switching. The electrospray ionization voltage was +4950 V in positive mode and −4,500 V in negative mode. The dwell time was 3 ms per SRM transition, and total cycle time was 1.55 s. Approximately 9–12 data points were acquired per detected metabolite.
For mevalonate pathway metabolite measurements, cells cultured in 10 cm dishes were washed with 5 ml cold PBS and extracted with 1 ml of 80% methanol, then placed on ice for 1 h. The extracts were centrifuged at 16,000g for 15 min, and 100 μl of the supernatant was stored for analysis. The remaining 900 μl was dried using a SpeedVac (Eppendorf, Vacufuge) and reconstituted in 80 μl of 80% methanol. For LC–MS analysis, 5 μl of both unconcentrated and concentrated samples was injected. LC was performed using an XBridge BEH Amide XP Column (Waters, 186006724). Mobile phase A consisted of 5% ACN with 20 mM ammonium acetate and 20 mM ammonium hydroxide (pH 9.75–9.85); mobile phase B was 100% ACN. The column was maintained at 25 °C, and the flow rate was 0.15 ml min−1. The LC gradient was as follows: 0 min, 85% B; 2 min, 85% B; 3 min, 80% B; 5 min, 80% B; 6 min, 75% B; 7 min, 75% B; 8 min, 70% B; 9 min, 70% B; 10 min, 50% B; 12 min, 50% B; 13 min, 25% B; 16 min, 25% B; 18 min, 0% B; 23 min, 0% B; 24 min, 85% B; and 30 min, 85% B. MS was performed using a Q Exactive Orbitrap Mass Spectrometer (Thermo) in negative ion mode. Data were processed and analysed using TraceFinder software (Thermo, OPTON-31001).
For the measurement of sorbitol in both cells and medium at different time points, cells were cultured in six-well plates under the indicated conditions, as shown in the figures. At the end of treatment, 1 ml of medium was mixed with 3 ml of methanol and centrifuged at 16,000g for 20 min. The supernatants were then dried using a SpeedVac (Eppendorf, Vacufuge) and reconstituted in 400 μl of 80% methanol per sample by 10 min of water-bath sonication (Branson, 2800). Samples were centrifuged at 16,000g for 20 min, and 5 μl of the supernatant was used for LC–MS injection. Meanwhile, the remaining medium was removed, and cells were washed with 2 ml of cold PBS, then extracted with 1 ml of 80% methanol and placed on ice for 1 h. The extracts were centrifuged at 16,000g for 15 min, and 5 μl of the supernatant was used for LC–MS injection. LC was performed using an XBridge BEH Amide XP column (Waters, 186006724) under negative ion mode, as described above for mevalonate pathway metabolite measurements. MS was carried out using a Q Exactive Orbitrap mass spectrometer system (Thermo), and data were processed and analysed using TraceFinder software (Thermo, OPTON-31001).
RNA-seq and data analysis
Total RNA was extracted from SORD WT (Cas9-expressing control cells) and KO cells derived from HCT116 and DLD1 using the RNeasy Plus Mini Kit (Qiagen, 74136). Before RNA extraction, cells were cultured under Glu + Fru conditions for 48 h. Sample quality control, RNA-seq and data analysis were performed by Novogene. In brief, after isolation, RNA integrity and quantification were assessed using the RNA Nano 6000 Assay Kit with the Bioanalyzer 2100 system (Agilent Technologies). Qualified samples had an RNA Integrity Number value of >10. Libraries were generated using the NEBNext Ultra RNA Library Prep Kit for Illumina (New England Biolabs, E7530L) following the manufacturer’s protocols, with index codes added. Clustering of the index-coded samples was performed on a cBot Cluster Generation System using the PE Cluster Kit cBot-HS (Illumina), according to the manufacturer’s instructions. After cluster generation, library preparations were sequenced on an Illumina platform to generate paired-end reads. Following quality control checks for Q20, Q30 and GC content, the paired-end clean reads were aligned to the reference genome using STAR (Spliced Transcripts Alignment to a Reference). Gene-level quantification was performed with FeatureCounts, and differential expression analysis was conducted using the DESeq2 R package. Differentially expressed genes were identified using the edgeR R package; those with a false discovery rate of <0.05 and |log2(fold change)| > 0.5 were used for pathway analysis with Qiagen Ingenuity Pathway Analysis. Heatmaps were generated using Qlucore (v.3.9). The R code used to process the RNA-seq data is available upon request.
Analysis of data from public databases
Gene expression data from human CRCs and adjacent normal tissues (GSE41258, GSE14297, GSE49355 and GSE35834) were obtained from the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo). Data from The Cancer Genome Atlas Pan-Cancer dataset were obtained from the UCSC Xena database. OncoGEO B37 and B38 datasets were obtained from QIAGEN OmicSoft Lands. The expression levels of SORD and AKR1B1 in normal tissue, primary tumour and metastatic tumour groups were extracted from these datasets and compared using unpaired t-tests or one-way ANOVA. Analysis of scRNA-seq data was performed using the Human Colon Cancer Atlas (c295) dataset via the Single Cell Portal (https://singlecell.broadinstitute.org/single_cell)
Statistics and reproducibility
All animals were randomly grouped. No statistical methods were used to pre-determine sample sizes, but our sample sizes are similar to those reported in previous publications9,33. Data collection and analysis in mice were not conducted blind to the experimental conditions. In the animal experiments, mice that died prematurely or failed to develop primary colon tumours (owing to injection issues) were excluded from analysis. In Extended Data Fig. 3a,g (public gene expression data), one outlier data point (|x − μ| > 6σ) per figure was excluded for graphical purposes, and this exclusion did not affect the P value or interpretation. No data were excluded from other analyses.
All data are shown as means; error bars, s.e.m. Data distribution was assumed to be normal, but this was not formally tested. When comparing means between two groups, a two-tailed unpaired t-test was used after confirming that the data were sampled from a Gaussian distribution using the D’Agostino–Pearson normality test. When comparing means across more than two groups, a one-way ANOVA was performed using Prism 10 (GraphPad). For comparing the effects of genotype and treatment, two-way ANOVA was conducted with post hoc comparisons using Holm’s multiple comparisons test in Prism 10 (GraphPad). Statistical significance is shown as *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001. The Venn diagrams in Fig. 2 and Extended Data Fig. 3 were generated in https://bioinformatics.psb.ugent.be/webtools/Venn/.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Source link