Quantification of Hepcidin-25 in Human Serum by Isotope Dilution Micro-HPLC–Tandem Mass Spectrometry

Hepcidin-25, a liver-produced peptide hormone, was initially isolated from human urine and blood ultrafiltrate. Hepcidin-25 is thought to be the central regulator of iron metabolism (1). Iron deficiency is associated with low hepcidin-25 concentrations and anemia of chronic disease with high concentrations, but the true diagnostic value of hepcidin-25 is still under investigation. A recently published review(2) stated that only semiquantitative methods for comparative studies based on mass spectrometry (MS) have been used for the determination of hepcidin in serum and urine. Isotope-dilution MS is generally accepted as yielding high analytical specificity and accuracy. We report an isotope-dilution–micro-HPLC–tandem MS (MS/MS) method that allows the quantification of hepcidin-25 present at less than nanomole-per-liter concentrations. During sample preparation, hepcidin-25 undergoes strong nonspecific binding to surfaces. A stable isotope-labeled internal standard helps to compensate for any matrix effects. Micro-HPLC-MS/MS with monolithic capillary columns ensures highest resolution and limits of quantification (3). Native hepcidin-25 (Mr 2789) was purchased from Bachem AG. The internal standard, hepcidin-25 [DTHFPI(13C615N1)CI(13C615N1)FCCGCCHRSKCGMCCKT-OH] was synthesized with isotopically labeled isoleucines by use of the FastMoc/tBu-strategy. Renaturation and purification were performed according to a previously reported method (4). Correct folding was verified by testing bioactivity with an in vitro ferroportin internalization assay(1). Blood samples were collected and anonymized in-house according to the Roche Diagnostics policy, and informed consent was obtained from all sample donors; results were not used for regulatory purposes. Calibrators with concentrations of 0.1–100.0 nmol/L were prepared from hepcidin-25. For sample preparation we added 5 μL concentrated formic acid and 50 μL internal standard solution (1450 nmol/L in water, 0.04% acetic acid) to 45 μL human serum or calibrator. Samples were ultrafiltered with a Microcon Ultracel YM-10 filter (Millipore). The flow-through was transferred into HPLC vials. A Thermo Electron Quantum-Ultra triple-quadrupole mass spectrometer, Dionex Ultimate 3000 micro-HPLC, and PS-DVB Monolithic 200-μm internal diameter × 5 cm column (P/N 161409) were used for online micro-HPLC-MS/MS. A 2-μL sample was injected. The mobile phases were (eleunt A) 1% formic acid/0.025% trifluoroacetic acid in water and (eleunt B) 1% formic acid/0.025% trifluoroacetic acid in acetonitrile, flow rate 3 μL/min, with a linear gradient from 0% to 80% eleunt B during 7 min and then held at 80% eleunt B until minute 11. We used microelectrospray ionization in the positive mode; recorded selected reaction-monitoring transitions were m/z 930.8→ 1145.5 and m/z 935.5→1152.6, collision energy 33 V, and argon collision gas 2.0 mTorr. Carryover was excluded by blank injections. Samples were kept in long-term storage at −80 °C and were stable for at least 2 days at 6 °C. Processed samples were stable for 48 h at 6 °C. We observed no ion suppression attributable to changing elution conditions and monitoring-signal intensities; postcolumn infusion experiments are not compatible with the used microflow ion-source and could not be performed. We verified linearity and recovery by using spiked human serum with 7 concentrations from 0.1 to 80 nmoL/L. A correlation coefficient (r2) of 0.9968 was observed between expected and observed hepcidin-25, and recoveries of 95%–100% were calculated. To determine the total imprecision of the method we used 3 native samples containing 3 different concentrations, measured in triplicate on 6 different days. The CVs were 21% (0.3 nmol/L), 17% (1.2 nmol/L), and 4.8% (5.9 nmol/L), with a limit of quantification of 0.3 nmol/L. At a concentration of 0.1 nmol/L the signal:noise ratio was 10:1. With samples from 5 of the donors, no differences in the measured values were observed for results from 6 sampling devices (BD Serum SST II, BD Serum Z, BD Serum SST, Sarstedt Serum, Sarstedt Serum Gel, and Sarstedt Serum EDTA). Hepcidin-25 concentrations from 10 apparently healthy fasting persons drawn in the morning on 4 consecutive days were measured (Table 1 ). Observed hepcidin-25 concentrations were in the lower nanomole-per-liter range. Concentrations were higher in males than females. Day-to-day variation of hepcidin-25 values (nmol/L) for serum samples colected in the morning from apparently healthy fasting individuals. F, female; M, male. Day-to-day variation of hepcidin-25 values (nmol/L) for serum samples colected in the morning from apparently healthy fasting individuals. F, female; M, male. The concentrations we observed were in agreement with results of the ferroportin internalization assay. Ferroportin internalization was detected at concentrations below 1 nmol/L hepcidin-25, and 50% internalization occurred at 4.6 nmol/L. These results indicate that the measured hepcidin-25 serum concentrations were within a concentration interval relevant for regulation of iron homeostasis. In 1 individual on 2 days we performed monitoring of the intraday variation in hepcidin-25. We observed a clear increase in hepcidin-25 concentrations from morning to evening (morning 1.8/2.4, noon 2.1/3.9, afternoon 2.6/7.4, and evening 4.5/7.8 nmol/L). This observation is in good agreement with the other reported data (5). Preliminary data from tumor patients and patients with anemia of chronic disease show hepcidin-25 concentrations up to 70 nmol/L. The micro-HPLC-MS/MS method described here allows quantification of hepcidin-25 at the low concentrations that are observed in human serum. Future clinical validation studies will be necessary to evaluate the influence of disease status and the usefulness of hepcidin measurement in differential diagnoses. Grant/Funding Support: None declared. Financial Disclosures: None declared.