| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
1 APHP, Hôpital Trousseau, Service de Biochimie, Paris, France
2 Unité Mixte de Recherche, Centre National, de la Recherche Scientifique 8612, Faculté de Pharmacie, Université Paris-Sud 11, Châtenay-Malabry, France
aAddress correspondence to this author at: UMR CNRS 8612, Faculté de Pharmacie, Université Paris-Sud 11, 22 Rue J-B Clément, 922696 Châtenay-Malabry, France. Fax 33-146835409; e-mail fathi.moussq{at}cep.u-psud.fr.
To the Editor:
Riboflavin (RF; vitamin B2) participates in redox reactions in 2 coenzyme forms, flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD) (1). Quantification of these 3 flavins in plasma and erythrocytes is essential for studying vitamin B2 metabolism (1)(2). We describe here an HPLC method for quantifying these flavins in both matrices.
Riboflavin and FMN sodium salt were from Sigma, FAD disodium salt and acetonitrile were from Merck, and the internal standard galactoflavin was from Merck & Dohme. All chemicals were analytical grade. For quality control, we used 3 ClinChek serum and whole-blood controls (Recipe), reconstituted and stored at –80 °C. Aliquots of aqueous (9 g/L NaCl) flavin stock solutions (50 µmol/L) were stored at –20 °C in the dark. Stock solution concentrations were verified by measurement of absorbances at 450 nm (3).
The HPLC system consisted of a Thermoquest SP 8700 X pump, Rheodyne Model 7125 100-µL loop injector, Jasco FP-920 fluorescence detector (excitation, 456 nm; emission, 512 nm), and Shimadzu Model C-R1B Chromatopac Integrator. A Phenomenex Onyx Monolithic C18 column [50 x 4.6 mm (i.d.)] protected by a 2 x 4.6 mm (i.d.) C18 precolumn was used at ambient temperature. The mobile phase was a mixture of acetonitrile and 0.15 mol/L potassium phosphate buffer, pH 2.4 (6:94 by volume); the flow rate was 3.0 mL/min. The identities and purities of the flavin peaks were verified against retention times, supplementation with pure material, and fluorescence characteristics (see the file on experimental conditions in the Data Supplement that accompanies the online version of this letter at http://www.clinchem.org/content/vol52/issue5/).
Venous blood samples from 61 fasting volunteers (35 women and 26 men), ages 24–55 years (median, 26 years), were collected into EDTA Vacutainer Tubes (Becton Dickinson) and immediately placed on ice. Aliquots (50 µL) of whole blood were processed or stored at –80 °C until use. The remaining sample was centrifuged (10 min at 2000g and 4 °C), and 50 µL of plasma was processed or stored at –80 °C. Further sampling and manipulations were performed under subdued lighting.
To prepare the samples for analysis, we vortex-mixed 50 µL of plasma, 50 µL of 25 nmol/L aqueous galactoflavin, and 100 µL of 100 g/L trichloroacetic acid for 2 min in a 1.5-mL polypropylene microcentrifuge tube. After centrifugation (5 min at 2000g), the supernatant was decanted into a second tube and neutralized with 20 µL of 2.0 mol/L K3PO4. For analysis of whole blood, we used 50 µL of sample, 200 µL of internal standard, and 250 µL of 100 g/L trichloroacetic acid and injected 100 µL of the resulting mixture into the HPLC.
We performed calibration with pooled EDTA-plasma enriched with 10 nmol/L RF, 10 nmol/L FMN, and 60 nmol/L FAD or an individual EDTA–whole blood enriched with 10 nmol/L RF, 20 nmol/L FMN, and 300 nmol/L FAD. Aliquots of these enriched samples were stored at –20 °C until analysis.
We calculated concentrations by the peak-area ratios of each flavin vs the internal standard. The basal flavin contents of the plasma and whole blood used for calibration were determined with the standard-addition method (4) (see the experimental section in the online Data Supplement). We tested linearity at concentrations greater than and less than endogenous vitamer concentrations (see the experimental section in the online Data Supplement).
Chromatograms of a plasma and a whole-blood sample before and after addition of the riboflavin vitamers are shown in Fig. 1
. Separation on monolithic porous silica columns allowed a shorter run time without loss of performance. Between-day CVs (n = 30) of the retention times were <4%.
|
The limits of detection (signal-to-noise ratio = 5) were 1.0 and 2.3 nmol/L for FAD, 0.3 and 0.8 nmol/L for FMN, and 0.5 and 1.3 nmol/L for RF in plasma and whole blood.
The recovery and precision data for the method are summarized in Table 1 of the online Data Supplement. Within- and between-day CVs were 3%–7% and 6%–10%, respectively. The stability of the vitamers and the effects of hemolysis were checked (5), and the obtained results were in agreement with published results(5). The frequency distributions of the concentrations of the 3 flavins in plasma and whole blood from our volunteers agree with those obtained with capillary electrophoresis(5)(6), HPLC (3)(7)(8), and liquid chromatography–tandem mass spectrometry(9).
In whole blood, the concentrations of FAD and RF were correlated (r = 0.34; P = 0.007), the concentrations of FAD and FMN (r = 0.06; P = 0.63) and of RF and FMN (r = 0.09; P = 0.5) were not (see Fig. 1 in the online Data Supplement).
Our simple and sensitive HPLC method for measuring RF and its coenzyme forms in plasma and whole blood allows use of a 50-µL sample with a single deproteinization step for sample preparation and analysis.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
