MSACL 2018 US Abstract

Topic: Troubleshooting

Fat Loving Vitamins the Struggle is Real

Matthew Crawford (Presenter)
LabCorp

Authorship: Matthew Crawford, Erin Fagan, Kyle Cahill, and Russell Grant
Laboratory Corporations of America Inc

Short Abstract

Transitioning lipophilic vitamins A (retinol), E (α- and γ/β- tocopherol), and β-carotene onto a high throughput LC-MS/MS platform provided a gamut of challenges within the laboratory. Application of troubleshooting tools such as; pairwise split sample stress testing, adsorptive loss contact studies, spike order preparative loss determination, IS binding equivalency, phospholipid elution characterization and ionization mode switching, variance reduction of precursor ion population, transition ratio monitoring for interferent identification, echo summing, pre-validation critical assay component assessment and enhancements in chromatographic efficiency were used for resolution of pre-validation challenges. Spoiler alert: We were not 100% successful.

Long Abstract

Problem

Accurately, precisely and selectively measuring lipophilic vitamins using LC-MS/MS.

Method Information

100 L of serum or plasma precipitated with either organic or acid solutions

Aria Transcend TLX-4 operating with Agilent 1200SL pumps

AB Sciex 5500 Triple Quadrupole Mass Spectrometer with Pos/Neg APCI switching

Troubleshooting Steps

Calibration:

Parallel stock and calibrator preparations (light versus dark, amber versus clear, with or without antioxidants), adsorptive loss evaluation (repeated contact with of glass pipets/class A glassware contact studies) and effect of temperature (ambient to -70C) were evaluated in a pairwise manner. Final conditions confirmed with NIST SRM materials (bias < 5%).

Ionization:

Positive ion mode APCI led to radical cation [M.]+ and protonated [M+H]+ precursor ions for tocopherols (efficiency loss) together with multiple chromatographic peaks (very surprising, un resolved with solution and LC development) together with ionization variance between calibrators and samples (IS recovery variance due to co-elution with phospholipids, confirmed with SRM and post column infusion). Switching acquisition modes from positive (retinol) to negative (tocopherols, only [M-H]- precursors, no ionization effect from co-eluting phopsholipids) then back to positive (β-carotene), ameliorated the matrix effect, selectivity and response function variance observed.

Sample preparation:

Recovery losses of ~80% were observed for β-carotene using SLE. Spike order experiments using D4- β-carotene as a proxy indicated losses during evaporation/reconstitution (impact of light, container materials, passivating materials, solvents and temperature were attempted unsuccessfully) Multistep IS addition (miscible/soluble/precipitation solvent) studies followed to establish process for IS binding equivalency prior to protein precipitation.

Chromatography:

Pre-validation evaluation of different LC column lots indicated drastic changes for β-carotene (development column had become passivated), resulting in additional LC and solvent development efforts (unsuccessful).

Single sample preparation for two LC-separations:

Brute force chromatographic redevelopment was followed including rescreening of solvents and columns (functionalities) together with increasing chromatographic efficiency (theoretical plates, temperature, linear velocity and peak capacity) generated reproducible β-carotene analysis and resolution from α-carotene (~30% circulating, confirmed through transition ratio and retention time alignment with standards)) however an unknown interferent was observed (peak asymmetry in ~20% of samples) that could not be resolved effectively.

Outcome

Vitamins A/E have been successfully validated and launched clinically. β-carotene was removed from the assay and development continues.


References & Acknowledgements:


Financial Disclosure

DescriptionY/NSource
Grantsno
SalaryyesLabCorp
Board Memberno
Stockno
Expensesno

IP Royalty: no

Planning to mention or discuss specific products or technology of the company(ies) listed above:

no