Metabolomic Diversity of Human Milk Cells over the Course of Lactation

Study design

A pilot study for the analysis of the metabolome of the cell fraction of human milk was conducted. The study included fifteen HM samples collected at different postnatal ages and labeled as colostrum (n=1,1-3 days), transitional (n=3, 4-14 days) or mature (n=11, >14 days) milk. The cellular fraction was studied with microscopy and staining techniques, and their metabolic fingerprint was analyzed by ultra-performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry (UPLC–QqTOF-MS) in the positive and negative electrospray ionization modes.

Sample processing & metabolomic analysis

A 15 mL aliquot of fresh HM was centrifuged during 20 min at 800 x g and 4 °C. The supernatant was discarded, and the pellet was resuspended in PBS and centrifuged during 5 min at 800 x g and 4 °C. The washing step was repeated. For cell lysis and metabolite extraction, 300 µL of lysis solution (i.e., CH₃OH:H₂O 3:1, v/v) containing the internal standards (IS) L-tryptophan-D₅ (≥99%), caffeine-D₉ (≥99%), and L-phenylalanine-D₅ (≥99%) at 0.25 μM each were added to the cell pellet. The mixture was transferred to a microcentrifuge tube placed on ice and 200 µL of the lysis solution were additionally employed to ensure quantitative transfer of cells. After three freeze-thaw cycles, samples were stored at -80 °C until further processing.

After thawing, the samples were centrifuged at 15493 x g and 4 °C for 5 minutes. The supernatant was transferred to a microcentrifuge tube and evaporated under vacuum at 30 °C on a miVac centrifugal vacuum concentrator (Genevac LTD, Ipswich, UK). Dry residues were resuspended in 70 µL of H₂O:CH₃CN (85:15 v/v, 0.1% v/v HCOOH) and homogenized, followed by a centrifugation at 15493 x g and 4 °C for 5 minutes to remove cell debris. Supernatants were transferred to microcentrifuge tubes and stored at -80 °C until analysis. A pooled quality control (QC) sample was prepared by mixing 10 µL of each sample extract. A blank extract was prepared following the same procedure as described for samples, but without cells. In addition, the supernatant obtained from the second washing step of one sample was processed as described for samples and analyzed.

Metabolomic analysis of HM cell extracts was carried out by UPLC–QqTOF-MS as described elsewhere [1]. Briefly, separations were performed on a Synergi^TM Hydro-RP 80 Å LC C₁₈ column (150 × 2 mm, 4 μm, Phenomenex, Torrance, USA) in a 1290 Infinity UPLC system running the following binary gradient with solvent A (H₂O, 0.1% v/v HCOOH) and solvent B (CH₃CN, 0.1% v/v HCOOH) as mobile phase components: 1% B for 2 min, linear gradient from 1 to 80% B in 8 min, from 80 to 98% B in 0.1 min,  98% B for 1.9 min, return to initial conditions in 0.1 min, and column equilibration with 1% B during 2.9 min. The flow rate was set to 0.4 mL min^-1 and column and autosampler temperatures were set at 40 and 4 °C, respectively.

For MS detection an Agilent 6550 Spectrometer iFunnel quadrupole time-of-flight (QqTOF) MS system was used operating in ESI⁺ and ESI^- modes. Full scan MS data was acquired between 70 and 1500 m/z using the following ionization parameters: gas T, 200 ⁰C; drying gas, 14 L min^-1; nebulizer, 30 psi; sheath gas T, 350 ⁰C; sheath gas flow, 10 L min^-1. Mass reference standards were introduced into the source for automatic MS spectra recalibration during analysis via a reference sprayer valve using the 149.02332 (background contaminant), 121.050873 (purine), and 922.009798 (HP-0921) m/z as references in ESI⁺, and 119.036 (purine) and 980.0163 ([HP-0921 + CH₃COOH-H]^-) m/z in ESI^-. MS² data were acquired using iterated data dependent acquisition as described elsewhere [2] using centroid mode at a rate of 5 Hz in the extended dynamic range mode (2 GHz), a collision energy set to 20 V, medium isolation window (~4 amu), MS² fragmentation with automated selection of five precursor ions per cycle, and an exclusion window of 0.15 min after two consecutive selections of the same precursor. UPLC-MS data acquisition was carried out employing MassHunter Workstation (version B.07.00) from Agilent.

The analytical batch included an initial system suitability check (2 μM IS solution) followed by 9 QCs replicates for system conditioning and MS² data acquisition. HM sample extracts were injected in random order. The QC was injected once at the beginning, twice at the end, and during the batch for the assessment of instrumental performance [3]. The blank extract was injected twice and the blank extract from cell supernatant once at the end of the measurement sequence to identify signals from other than biological origin, and possible carryover [4].

Data description

From de ESI+ and ESI- analyses:

• 21 .mzXML files including MS1 data of 15 samples, 4 QCs and 2 blanks
• 9 .ms2 files including MS/MS data of QC injections
• 9 .mgf files including MS/MS data of QC injections
• .xlsx file with metadata (results from the cytochemical and immunocytochemical characterization of HM cells have been included)