Targeted LC-MS/MS Profiling of Central Carbon Metabolism Intermediates in Cultured HepG2 Cells Using HILIC Separation and MRM

Six-well plates, with 4-6 biological replicate wells per condition (separate wells, same passage) and 3 independent experimental days. Each plate includes a process blank (medium-only extraction) and a pooled QC sample injected every 8-10 runs to monitor instrument drift. Cells are seeded at 5 x 10^5 cells/well and harvested at ~80% confluence. A single metabolite-quenching time point is used; for flux studies, ^13C6-glucose labeling time courses (0, 5, 15, 30, 60 min) replace the static design. Inject samples in randomized order, bracketed by QCs and solvent blanks.

BSL-1; HepG2 is a non-pathogenic human-derived line (handle human cell lines with universal precautions). Methanol and acetonitrile are flammable and toxic — handle in a fume hood, keep away from ignition sources, wear nitrile gloves, lab coat, and safety glasses. Ammonium hydroxide is corrosive and an inhalation hazard. Liquid nitrogen/dry ice: use cryo-gloves and ensure ventilation to prevent asphyxiation. Collect organic solvent waste in a dedicated flammable-solvent container.

Process blank: extraction solvent carried through the full workflow with no cells, to flag background/carryover. Vehicle control: cells treated with the carrier solvent at matched concentration. Pooled QC: equal-volume mix of all samples injected repeatedly to assess analytical reproducibility (target CV <20% for reliable metabolites). Stable-isotope internal standards spiked at quenching control for extraction efficiency and ion-suppression. A calibration curve of authentic standards (8-point, 1 nM-100 µM) anchors quantification. Optionally include a known metabolic inhibitor (e.g., 2-deoxyglucose) as a positive biological control that predictably depletes downstream glycolytic intermediates.

Well-behaved data show sharp, symmetric chromatographic peaks with retention-time drift <0.1 min across the batch, QC CV <20% for the majority of metabolites, and internal-standard recovery 70-120%. Typical intracellular lactate dominates (high abundance); TCA intermediates like citrate and malate are mid-abundance; α-KG and PEP are lower. A glycolytic shift shows elevated lactate/G6P/FBP and depressed TCA pools. Calibration curves should be linear (R² > 0.99) over the working range.

• HepG2 cells (ATCC HB-8065), passage <20, mycoplasma-negative
• DMEM (Gibco 11965, 25 mM glucose) + 10% FBS + 1% pen/strep
• LC-MS grade methanol, acetonitrile, water (Fisher Optima)
• Extraction solvent: 80:20 methanol:water (v/v), pre-chilled to -80 °C
• Ammonium acetate and ammonium hydroxide for HILIC mobile phase (LC-MS grade)
• Stable-isotope internal standard mix (e.g., U-^13C labeled metabolite library, Cambridge Isotope Laboratories) or single-compound IS (^13C3-lactate, ^13C6-citrate)
• Dry ice / liquid nitrogen for quenching
• 96-well or vial autosampler plates, 0.22 µm filter, speedvac/lyophilizer
• Authentic metabolite standards for MRM optimization and calibration curves

To extract polar metabolites from cultured HepG2 hepatocellular carcinoma cells and quantify ~25 central carbon metabolism intermediates (glycolysis, pentose phosphate pathway, TCA cycle) by reversed-phase/HILIC liquid chromatography coupled to triple-quadrupole tandem mass spectrometry in multiple reaction monitoring (MRM) mode. The protocol delivers relative and, where stable-isotope standards are available, absolute quantification of metabolite pool sizes normalized to cell number or protein.

1. Grow HepG2 to ~80% confluence in 6-well plates. Pre-chill extraction solvent (-80 °C), centrifuge, and dry-ice bath.
2. Quench: aspirate medium rapidly, place plate on dry ice, and immediately add 1 mL ice-cold 80:20 MeOH:water containing internal standards per well. (For adherent quenching, wash once with ice-cold saline first — note this may lose volatile pools; for flux work, quench directly without wash.)
3. Scrape cells in the cold solvent, transfer to pre-chilled tubes, and incubate 20 min at -80 °C.
4. Vortex, centrifuge 15,000 x g, 10 min, 4 °C to pellet protein/debris. Transfer supernatant to a fresh tube.
5. Dry supernatant under nitrogen or in a speedvac (no heat) to completeness.
6. Reconstitute in 50-100 µL 50:50 acetonitrile:water, vortex, centrifuge, and transfer to autosampler vials.
7. Inject 5 µL onto a HILIC column (e.g., 2.1 x 100 mm amide, 3 µm) at 0.3 mL/min with a 95%-40% acetonitrile gradient (aqueous = 10 mM ammonium acetate, pH 9), 20 min run.
8. Acquire negative-mode MRM transitions for each target; include retention-time scheduling.
9. Run calibration standards, QC pool every 8-10 injections, and blanks bracketing the batch.

Independent variable: the treatment condition (and labeling time for flux). Dependent variables: peak-area ratios and concentrations of each targeted metabolite (nmol per 10^6 cells or per mg protein), and ^13C labeling fractions if applicable. Controlled variables: cell confluence at harvest, quenching speed/temperature, extraction solvent ratio and volume, internal standard amount, injection volume, column temperature, gradient, and instrument tune state.

We hypothesize that a glycolytic perturbation (e.g., a treatment shifting cells toward the Warburg phenotype) will increase intracellular lactate and glycolytic intermediates (G6P, FBP, PEP) by >50% while reducing TCA intermediates (citrate, α-KG, malate), relative to vehicle controls. Conversely, an OXPHOS-promoting treatment will elevate TCA intermediate pools without raising lactate.

Process raw files in vendor software (e.g., MassHunter Quantitative, Skyline, or MAVEN/El-MAVEN for open tools). Integrate each MRM peak, divide analyte area by its internal-standard area, and back-calculate concentration from the calibration curve. Normalize to cell count or protein per well. For isotope tracing, correct for natural abundance (e.g., IsoCor) and compute mass-isotopomer distributions and labeling fractions. Filter metabolites with QC CV >30% or below LOD. Generate a metabolite-by-sample matrix for downstream stats.

1. Poor or shifting retention on HILIC: column not equilibrated or mobile-phase pH drift — equilibrate ≥10 column volumes and prepare fresh ammonium acetate buffer. 2. Low signal/ion suppression: matrix overload — dilute reconstituted extract or reduce injection volume; verify internal-standard recovery. 3. High QC CV: inconsistent quenching/drying — speed up quench and avoid heating during evaporation. 4. Peak splitting/fronting: poor reconstitution solvent strength — match initial mobile-phase composition (high organic). 5. Carryover in blanks: increase needle wash and add a strong-wash injection between samples.

Use biological replicate as the experimental unit (n=4-6 per condition across 3 days). Log-transform peak-area ratios to stabilize variance. For two groups, apply Welch's t-test per metabolite; for multiple conditions, one-way ANOVA. Correct for multiple comparisons across the metabolite panel using Benjamini-Hochberg FDR (q < 0.05). Set alpha = 0.05. Report fold-change and FDR-adjusted p-values; for global structure, PCA and hierarchical clustering of z-scored data.