Determination of plasma adenosine concentration with an isocratic reversed-phase HPLC method

Research output: Contribution to journalArticle


Endogenous adenosine released by many tissues both under physiological and pathophysiological conditions can be formed extracellularly and intracellularly by dephosphorilation of 5'-AMP. Adenosine has a wide spectrum of biological effects on various tissues, it is regarded as the primary mediator of the autoregulation of coronary vascular bed. The physiological actions of adenosine are related to its interstitial concentration, in humans, however only determination of plasma adenosine level is feasible. Measurement of plasma adenosine concentration is a technical challenge for two reasons: 1. a sensitive HPLC-technique is required due to the low adenosine concentrations (nM); 2. a suitable 'stop-solution' is needed to prevent the extremely rapid elimination and formation of adenosine in blood sample. In addition, either a peak-shifting or a subtraction HPLC approach is needed to separate adenosine from co-eluting plasma peaks. This study was designed the address the above issues. We compared stop-solutions of different compositions and found, that the stop-solution containing 400 μM dipyridamole and 10 μM of EHNA was suitable to prevent the change of adenosine concentration after sampling. We developed an isocratic elution with subtraction method to avoid the problems that arise with the peak shifting method and gradient elution. Immediately after sampling (7.5 ml of blood was collected into a heparinised syringe containing 2.5 ml of ice-cold stop-solution) blood was centrifuged, plasma was removed and deproteinized in a 25.000 molecular weight cut off membrane cones (Amicon, Danvers, MA model, 1 CF25). The ultrafiltered sample was divided portions and injected directly (unknown) and after adenosine-deaminase treatment (blank). Chromatographic analysis was carried out with an LKB high-performance liquid chromatograph, which consisted of an LKB 2249 pump, an LKB 2154 sample injection valve (50 μl) and an LKB 2141 dual-wave-length UV-detector. Separation was performed on a 3 μm Microspher C18 (100 mm x 4.6 mm I.D.) column with 5 μm Chromsep C18 guard column (10 mm x 3 mm) by isocratic mobile phase, at flow rate 1.0 ml/min. The mobile phase was a mixture of an ionpair buffer (consisting of 0.1 M Na2HPO4 and 0.01 M TBAH; pH-5.0) and acetonitrile (v/v; 98:2). The UV-detector recorded absorbance at 260 nm. Mobile phase was filtered through a 0.45 μm filter and degassed by helium before the analysis. Our method offers low baseline-noise and short run time (4 min) which is helpful to overcome low adenosine level. In addition, despite the co-eluting peaks, identifying adenosine by subtraction of a blank (adenosine deaminase treated sample) chromatogram from paired unknowns yields considerable safety due to the low variance of retention time. With our method the recovery of adenosine was 93 ± 4.5% (mean ± SE) from human plasma (n = 6) and intra- and interassay coefficients of variation were 6.5% and 7.2%, respectively (n = 6 in both cases). The mean level of plasma adenosine of nine healthy volunteers (three males and six females aged between 23 and 50) was 59 ± 8 nM (mean ± SE, n = 9). The levels of plasma adenosine were measured in one female healthy volunteer (aged 25) on four consecutive days resulting in adenosine concentration of 20, 26, 33 and 70 nM.

Original languageEnglish
Pages (from-to)208-212
Number of pages5
JournalActa pharmaceutica Hungarica
Issue number6
Publication statusPublished - Jan 1 1994

ASJC Scopus subject areas

  • Pharmaceutical Science

Fingerprint Dive into the research topics of 'Determination of plasma adenosine concentration with an isocratic reversed-phase HPLC method'. Together they form a unique fingerprint.

  • Cite this