S-Nitrosoalbumin Plasma Levels in Health and Disease: Facts or Artifacts? Value of Analytical Chemistry in Nitric Oxide Clinical Research
To the Editor:
The discovery of the endothelium-derived relaxing factor (EDRF), its identification as nitric oxide (·NO), and the recognition of its multiple biological functions, especially in the cardiovascular system, are fascinating scientific achievements of the last two decades, an effort that was awarded the Nobel Prize for Medicine in 1998.
Curiously, no other small molecule like ·NO challenges so many scientists from so very different disciplines. In the literature, there is no further example for the development and application of so wide a spectrum of analytical approaches and methods in recent years that has yielded highly divergent values, often within a range of three orders of magnitude, and has consequently led to numerous deceptive conclusions.
In 1992, Stamler et al1 reported for the first time that ·NO circulates in plasma of healthy humans primarily as S-nitrosoalbumin (SNALB; 7000 nmol/L, n=5). Mainly on the basis of this finding, Stamler’s group1 suggested that SNALB may be a physiological reservoir of ·NO, by which ·NO-related actions such as vasodilation are regulated in humans. This highly interesting finding has initiated much scientific work in this area.
Until today, however, there is no solid confirmation, perhaps with a single exception,2 of Stamler’s originally reported values for endogenous normal plasma levels of SNALB, not even by Stamler himself, who communicated3 in 1997 that normal SNALB plasma levels may be much lower, ie 200 to 1000 nmol/L.
In 1999, for the first time, we questioned4 Stamler’s findings on endogenous normal SNALB plasma levels. By means of a fully validated, accurate, and artifact-free GC-MS method, which involves—as the sole method in this area—use of 15N-labeled SNALB (ie, S15NALB) as internal standard and affinity-column extraction of SNALB and S15NALB from plasma, we found that SNALB indeed exists physiologically in plasma of humans (181 nmol/L, n=23, healthy volunteers; 161 nmol/L, n=40, patients with hepatic diseases), but at concentrations severalfold smaller than Stamler’s originally reported1 and even revised values.3 We have confirmed these values by using cysteine/Cu2+ instead of HgCl2 (authors’ unpublished results, 2002).
In the meantime, many other groups, eg Marley et al,5 Cannon et al,6 Moriel et al,7 and Rossi et al,8 have also reported that normal plasma levels of endogenous SNALB are considerably lower (ie, 28 to 250 nmol/L)4–8⇓⇓⇓⇓ even when compared with Stamler’s newer data.3
The sole group that confirmed Stamler’s originally reported SNALB plasma levels is Tyurin’s group, whose study was recently published in Circulation Research.2 Tyurin’s group has, moreover, found that SNALB levels may significantly differ in health (4200 nmol/L, nonpregnant; 5100 nmol/L, normal pregnancy) and disease (6300 nmol/L, preeclampsia). Highly elevated S-nitrosoprotein plasma concentrations were also found in hypercholesterolemia (540 nmol/L) compared with normolipidemia (250 nmol/L).7
The conclusions by Tyurin’s2 as well as by Moriel’s7 groups are, in our opinion, not justified, both from the analytical and the biological points of view. The pathophysiology (eg, hypertension) of preeclampsia and hypercholesterolemia is closely related to vascular endothelial dysfunction, ie, to a relative deficiency of available ·NO.9 This most likely results from the inhibition of ·NO synthase (·NOS) activity by endogenous ·NOS inhibitors, notably asymmetric dimethylarginine (ADMA), the plasma levels of which are evidently elevated in preeclampsia and hypercholesterolemia with concomitantly decreased l-arginine plasma levels.9 Two diametrically opposite explanations were given by Tyurin’s group2 (ie, accumulation of SNALB due to decreased ascorbate levels in preeclampsia) and Moriel’s group7 (ie, enhanced S-nitrosothiol synthesis due to increased ceruloplasmin levels in hypercholesterolemia). However, there is no evidence that ascorbate may catalyze ·NO release from SNALB in vivo. Moriel’s7 conclusion contradicts the elevated plasma nitrate levels measured by the authors themselves (ie, 40 versus 19 μmol/L in control subjects), suggesting increased ·NO synthesis. Therefore, Tyurin’s2 and Moriel’s7 findings are not supportive of Stamler’s1 hypothesis. The mechanisms and the biochemical factors known to be involved in in vitro formation of SNALB and ·NO release from SNALB (eg, ascorbate, thiols, ceruloplasmin, and Cu2+) remain to be established in vivo.
Highly divergent values for endogenous plasma levels of SNALB and many other members of the l-arginine/·NO pathway raise numerous questions. The most serious question concerns, in our opinion, the stepmother importance attributed to analytical chemistry by researchers, authors, editors, reviewers, and publishers of scientific journals, as well as by commercial companies providing assay kits. Despite the damning evidence for the need of reliable, unquestionable quantitative analytical approaches in this field of research, choice of analytical quantitative methods is directed to simplicity, rapidity, and commercial availability rather than to reliability. This practice does not actually promote science and must therefore be changed by all participants.
- ↵Stamler JS, Jaraki O, Osborne J, Simon DI, Keaney J, Vita J, Singel D, Valeri CR, Loscalzo J. Nitric oxide circulates in mammalian plasma primarily as an S-nitroso adduct of serum albumin. Proc Natl Acad Sci U S A. 1992; 89: 7674–7677.
- ↵Tyurin VA, Liu SX, Tyurina Y, Sussman NB, Hubel CA, Roberts JM, Taylor RN, Kagan VE. Elevated levels of S-nitrosoalbumin in preeclampsia plasma. Circ Res. 2001; 88: 1210–1215.
- ↵Tsikas D, Sandmann J, Gutzki FM, Stichtenoth DO, Frölich JC. Measurement of S-nitrosoalbumin by gas chromatography-mass spectrometry, II: quantitative determination of S-nitrosoalbumin in human plasma using S-[15N]nitrosoalbumin as internal standard. J Chromatogr B Biomed Sci Appl. 1999; 726: 13–24.