Printer Friendly

Effects of uremic toxins and fatty acids on serum protein binding of furosemide: possible mechanism of the binding defect in uremia.

Furosemide, an anthranilic acid derivative, is one of the most potent diuretic drugs commercially available [1, 2]. In comparison with other types of diuretics, furosemide causes diuresis, which is accompanied by an increase in renal blood flow and glomerular filtration rate. As a result, furosemide is frequently used in diuretic therapy for patients with renal dysfunction (glomerular filtration rate <20 mL/min) and for elderly patients. The changes in pharmacological and toxicological response to furosemide frequently seen in these patients probably result from a complex set of events, for which the relevant mechanism is still unknown [3].

Even when concentrations of albumin (the major carrier protein for furosemide in blood) are within the reference range, a decrease in serum binding of furosemide frequently occurs in patients with renal failure [4]. Because furosemide usually is almost entirely serum protein-bound in healthy subjects [4, 5], this increase in its unbound fraction in renal failure may affect the pharmacokinetics of the drug and hence its diuretic action or toxicity in uremic patients.

Currently, the most widely accepted explanation for the drug-binding defect in uremic serum is the accumulation of endogenous binding inhibitors, often referred to as "uremic toxins" [4, 6-10]. In earlier studies, Ikeda et al. [11] showed that indoxyl sulfate (IS) (3), one of these uremic toxins, competitively inhibits the binding of furosemide to bovine serum albumin. They concluded that indole sulfate represents one of the major inducers of defective furosemide binding to serum in uremia. In our laboratory, however, preliminary experiments with human serum suggested that a substantial proportion of the uremic binding defect cannot be explained by IS alone.

The purpose of present study was to elucidate the molecular mechanism of interaction between furosemide and uremic toxin(s), as related to the binding of furosemide to human serum albumin (HSA), and to identify the major inhibitors of this binding. We also investigated the influence of long-chain fatty acids on the furosemide binding to serum in the presence of uremic toxin, given that the concentrations of fatty acids, which modulate drug binding, are considerably increased in patients with chronic renal dysfunction, especially patients undergoing hemodialysis therapy.

Materials and Methods


HSA (essentially fatty acid-free) was purchased from Sigma Chemical Co., as were oleate, linoleate, and stearate. Furosemide was a gift from Hoechst Japan. IS, indole-3-acetate (IA), and hippurate (HA) were obtained as pure substances from Nacalai Tesque (Kyoto, Japan). CMPF was kindly provided by A. Takadate (Daiichi College of Pharmaceutical Sciences, Fukuoka, Japan). All other chemicals were of analytical grade. All solutions were prepared in deionized and distilled water. Phosphate buffer (67 mmol/L, pH 7.4), prepared with dibasic and monobasic sodium phosphate, was used exclusively in this study.

Uremic serum was pooled from 8 men with chronic renal dysfunction (mean [+ or -] SD age 56.2 [+ or -] 12.5 years, creatinine concentration 114 [+ or -] 15 mg/L, blood urea N 652 [+ or -] 98 mg/L). This serum pool contained CMPF, IS, IA, and HA at mean concentrations of 274, 152, 45, and 357 [micro]mol/L, respectively; nonesterified (free) fatty acid (FFA) was at a mean concentration of 0.52 mmol/L. "Normal" (nonuremic) pooled serum, prepared from blood samples obtained from 5 healthy men (age 28.4 [+ or -] 6.3 years) with healthy renal functions (mean creatinine and blood urea N concentrations of 14 and 123 mg/L, respectively), contained CMPF, IS, IA, and HA at mean concentrations of 4.0,1.2, 13.3, and 13.9 [micro]mol/L, respectively, and FFA at 0.41 mmol/L. All subjects were withdrawn from furosemide treatment for at least 2 days and from any other medication for >12 h before blood sampling.


In vitro protein binding of furosemide was determined in the healthy volunteers' and renal failure patients' sera and in isolated HSA solution at 25 [+ or -] 1 [degrees]C by the following procedures. Serum protein binding was measured in the presence of furosemide at 6.6 mg/L (20 [micro]mol/L), corresponding approximately to the mean maximum concentration seen after administration of 80 mg of furosemide by intravenous infusion over 1 h. To assess the per unit inhibitory strength of uremic toxin toward furosemide serum binding, each toxin was individually added to the nonuremic serum pool to a final concentration of 300 [micro]mol/L. This concentration was selected on the basis of the concentration of CMPF observed in uremic failure patients (our preliminary experiments had shown that CMPF had the most potential for inhibiting furosemide binding to HSA).

In addition, sera contrived to resemble renal patients' sera were prepared by adding to the healthy volunteers' sera IS, IA, HA, and CMPF in concentrations corresponding to renal uremic conditions. To study HSA binding, we added furosemide to isolated HSA solution to give final furosemide concentrations of 10 to 55 [micro]mol/L--multiples of the molar concentration of the HSA. The unbound species were isolated by ultrafiltration (Tosoh Plastic, Kanazawa, Japan) of 0.9- or 1.35-mL aliquots in a prewarmed (25 [degrees]C) centrifuge at 3000 or 5000g for 15 min. Adsorption of furosemide or uremic toxins onto the filtration membrane and apparatus was negligible.

The free concentration of ligand was determined by HPLC with the following columns (all from Cica Merck): Superspher 100 RP-18e for furosemide, LiChrosorb RP-select B for IS and IA, and LiChrosorb RP-18 for CMPF. The mobile phases consisted of distilled water/acetonitrile/methanol (491:9:4 by vol) for furosemide; 35 mmol/L phosphate buffer (pH 4.6)/acetonitrile/methanol (56:29:15 by vol) for IS and IA; and 11 mmol/L phosphate buffer (pH 5.0)/acetonitrile/methanol (10:5:1 by vol) for CMPF. For all separations, the flow rate was 1 mL/min. The UV detector was set at 285 nm for furosemide, IS, and IA and at 240 nm for CMPF. The unbound concentrations of HA were estimated by a previously described HPLC method [12]. The CVs of these assays were <5% for all ligands. Statistical significance of binding data was evaluated by one-way analysis of variance (ANOVA).

Albumin concentrations were determined with the BCA Protein Assay Kit from Pierce Chemical Co. FFA concentrations were determined with an assay kit from Wako Pure Chemical Co.


Binding parameters were estimated by fitting the experimental data to the following equation by using a nonlinear least-squares computer program (MULTI) [13].


where r is the number of moles of bound drug per protein molecule; [[D.sub.b]] and [[D.sub.f]] are the bound and unbound drug concentrations, respectively; [[P.sub.t]] is the total protein concentration; and [K.sub.i] and [N.sub.i] are the binding constant and the number of binding sites for the ith class of binding sites, respectively. The simultaneous binding of two ligands was analyzed by a previously reported method [12].



where [K.sub.A] and [K.sub.B] are the binding constants of ligand A and B, [[A.sub.f]] and [[B.sub.f]] are the free concentrations of ligand A and B, and [[A.sub.b]] and [[B.sub.b]] are the bound concentrations of ligand A and B, respectively. [chi] is a coupling constant, [K.sub.BA] is the binding constant of ligand A in the presence of ligand B, and [K.sub.AB] is the binding constant of ligand B in the presence of ligand A. Using these equations, we can calculate the theoretical values of [chi]. The interaction mode of the ligands on a macromolecule can be evaluated by the sign and magnitude of the value of [chi]. For example, if ligand A and B are independently bound to protein, [chi] is equal to 1. [chi] >1 and 0< [chi] <1 indicate cooperative and anticooperative interaction between ligands, respectively. Competitive displacement between ligands is indicated by [chi] = 0.

Inhibition of furosemide binding by uremic toxin or fatty acid was estimated from monitored changes of the free ligand fraction, [Florin], which was calculated as:

[Florin] = [[D.sub.f]]/[[D.sub.f]] + [[D.sub.b]] (4)



To examine IS, IA, HA, and CMPF as potential inhibitors of furosemide serum binding, we measured the free fraction of furosemide at therapeutic concentration (6.6 mg/L) in serum obtained from healthy subjects with or without uremic toxins at concentrations of 300 [micro]mol/L. As shown in Fig. 1, all the uremic toxins tested showed statistically significant inhibitory effects on the binding of furosemide to serum protein: CMPF showed the greatest effect, ~3 times more potent than the other three toxins.

The free fractions of furosemide in pooled serum obtained from renal failure patients were compared with those in artificial uremic serum, which was prepared from the nonuremic serum pool as followed. First, the HSA concentration of the nonuremic pool was adjusted [diluted with isotonic saline (9 g/L NaCl)] to mimic the albumin concentration of the uremic pool. Then, the adjusted nonuremic pool was supplemented with IA, IS, HA, and CMPF to mimic the concentrations of these compounds in the uremic pool. As shown in Fig. 2 (columns A and B), the free fractions of drug were comparable between these two systems. This result indicated that IS, IA, HA, and CMPF concentrations reasonably accounted for the changes observed in the free furosemide concentration in uremia. Furthermore, >80% of this binding defect could be reproduced by the addition of CMPF alone (Fig. 2, column C).


The mechanisms of interactions between furosemide and uremic toxins with respect to HSA were analyzed with a theoretical model for the simultaneous binding of two ligands, as described in Materials and Methods. Furosemide binding to HSA was significantly decreased, as it is for whole serum, by the presence of all four uremic toxins. The inhibition behaviors of IS, IA, and HA were qualitatively similar to each other, whereas the effect of CMPF was much stronger. Fig. 3 shows the results for CMPF and IS as typical for these two groups. For the CMPF-furosemide interactions, the experimental values fit well with the theoretical curve, which is based on the assumption that furosemide and CMPF compete at a common primary binding site on the albumin molecule. In contrast, the extent of mutual displacement between furosemide and an indole ring-containing uremic toxin and HA, with respect to binding to site II, was less than that expected for a competitive mechanism, indicating an anticooperative interaction between furosemide and site II-bound uremic toxins. This type of antagonistic binding was further quantified in terms of coupling constant ([chi]). As shown in Fig. 3B, binding isotherms constructed by using the [chi] values of 0.60 for the furosemide--IS system were in good agreement with the experimental data. Similar results were also obtained for IA and HA by using [chi] = 0.75 and 0. 56, respectively (data not shown).




Among fatty acids, oleate is a most abundant in human serum; linoleate and stearate are also present in substantial amounts. Therefore, we compared the effect of these fatty acids on the serum protein binding of furosemide in the presence of uremic toxin. Fig. 4 shows the effects of oleate on the free fraction of furosemide in serum with or without CMPF. For [oleate] / [HSA] ratios [less than or equal to]4, the free fraction of furosemide in serum was decreased in the absence of CMPF. Interestingly, further addition of oleate to serum ([oleate] / [HSA] ratios >4) led to significant (P <0.01) increases in the unbound fraction of furosemide. Similar complicated effects related to oleate binding have also been observed for warfarin, which is a typical marker ligand for site I binding [14].

As shown above, the binding of furosemide to serum was greatly inhibited by the presence of CMPF. The presence of oleate at concentrations up to double the albumin concentration had no significant effect on the increased free fraction of furosemide caused by CMPF binding. However, when [oleate] / [HSA] exceeded 4, the free fraction of furosemide in serum containing CMPF was considerably greater than that observed at low oleate concentration.



To elucidate the role of CMPF on the complicated inhibition behaviors on furosemide binding observed in serum containing oleate, furosemide, and CMPF, we examined the influence of oleate on the binding of CMPF to serum protein. As shown in Fig. 5, the free fraction of CMPF was considerably increased by adding oleate to serum.

Like oleate, qualitatively similar effects on the furosemide binding in the presence of CMPF were observed for both linoleate and stearate (Fig. 6) The effect of fatty acids on the free fraction of furosemide in the presence of CMPF is in the order of oleate > linoleate > stearate.


Among the endogenous solutes retained in uremia, IS, IA, HA, and CMPF are recognized as the major inhibitors of drug binding in serum in uremia. In patients with renal dysfunction, these compounds accumulate in the serum to concentrations as great as 202, 101, 883, and 370 [micro]mol/L, respectively [8-10,15,16]. As we have shown here, impaired serum protein binding of furosemide in renal failure can be reproduced by adding IS, IA, HA, and CMPF to nonuremic serum. Therefore, it is unlikely that other known uremic substances, e.g., guanidine, methylguanidine, and guanidinosuccinic acid, or other unidentified uremic compounds of low molecular mass can displace strongly albumin-bound drugs from binding sites responsible for the binding defect of furosemide in uremic sera. Moreover, the binding results indicate that IS, IA, HA, and CMPF inhibit the binding of furosemide to serum, although CMPF is much more potent in this respect than the others. To account for the mechanisms that govern such specificity in inhibitory potency, we carefully examined the relationship between the furosemide binding site(s) and the uremic toxin binding sites. Recently, the primary binding site of uremic toxins used in this study was reevaluated and identified by means of a theoretical model for the simultaneous binding of two ligands [12]. The location of the primary binding site for CMPF on HSA was found to be within specialized cavities in subdomain IIA, corresponding to site I, one of the major drug-binding areas. In contrast, uremic toxins that contain an indole ring and HA were primarily bound to site II, another major drug-binding area, which is located in subdomain IIIA of the HSA structure [17]. Some studies have suggested that, as with CMPF, the primary binding site for furosemide is located within site I [5]. These findings suggest that furosemide and CMPF may share a common high-affinity binding site. In fact, the present data demonstrate that the inhibitory effect of CMPF on furosemide binding to HSA is competitive, whereas moderate antagonism (anticooperative interaction) in binding to HSA is observed for furosemide and the uremic toxins that are bound to site II. On the basis of its concentrations in uremic serum [15,16], its binding affinity to HSA [12,18], and the inhibition characteristics found in the present experiments, we think it reasonable to conclude that CMPF represents the compound that is primarily responsible for the impaired binding of furosemide to uremic serum. In addition, uremic toxins that contain an indole ring and HA are also minor contributors to this furosemide-binding defect.



Our conclusions are not in agreement with the data reported by Ikeda's group, which concluded that IS is the major serum binding inhibitor of furosemide in uremia because its serum concentrations were increased in rabbits with experimentally induced acute renal failure and because it inhibited the interaction of furosemide with bovine serum albumin via competitive inhibition [11]. This apparent discrepancy may have several sources: the fact that this group failed to consider CMPF as a potential protein-binding inhibitor in uremia, species difference in ligand specificity of binding sites between bovine and human albumin, and differences in the fatty acid contents of the albumin preparation used.

Many patients with chronic renal dysfunction are involved in hemodialysis treatment, to remove accumulated waste substances. Heparin, frequently used as an anticoagulant during this therapy, is known to enhance the concentrations of fatty acids in blood up to fatty acid to albumin molar ratios as great as 6 to 7, via activation of lipoprotein lipase [19]. Such increases in fatty acids are often associated with the modulation of ligand binding to albumin, for which both competitive and allosteric effects have been reported [19,20]. Very recently, we found an interesting phenomenon in which the heparin-induced increase in fatty acids triggered a reaction that led to an enhancement of the preferential increase of the free fraction of pharmacologically active ketoprofen enantiomer during hemodialysis [21]. For this reason, we also examined the effect of unsaturated fatty acid (oleate), polyunsaturated fatty acid (linoleate), and saturated fatty acid (stearate) on the binding of furosemide to serum in the presence of CMPF. Among these fatty acids, oleate is the most potent with respect to furosemide binding, both in the absence and presence of CMPF. Compared with the furosemide serum binding at a low molar ratio of oleate to albumin (comparable with that observed in resting healthy subjects), the binding of furosemide to serum protein was inhibited when four or more oleate molecules were bound to one albumin molecule. Under such circumstances, oleates occupy not only high-affinity sites but also low-affinity sites, including site I [22]. As a result, at high concentrations, oleates compete with the furosemide bound to site I.


This oleate-induced increase in the unbound furosemide fraction is further promoted by the binding of CMPF to HSA. Interestingly, the magnitude of the inhibition of furosemide binding to serum observed in the oleate-CMPF-furosemide system is significantly larger than that obtained by assuming the independent inhibitory potency between CMPF and oleate on the binding of furosemide to serum. Analysis of the relationship between CMPF and oleate on HSA binding demonstrates that the free fraction of CMPF is considerably increased by the binding of oleate to its low-affinity sites. Thus, the synergistic response that occurs between CMPF and oleate on the inhibition of furosemide-protein binding is accounted for by a cascade mechanism of furosemide displacement in the oleate-CMPF-furosemide system, in which oleate also indirectly inhibits the binding of furosemide by transiently increasing the concentrations of unbound CMPF that competitively displace furosemide. Although CMPF and furosemide compete with each other for HSA binding, certain differences exist in the oleate binding response between the two ligands--possibly because of slight differences in the position of their binding sites in site I, given that site I consists of several subsites that overlap one another [23]. On the basis of these findings, we conclude that the combination of direct and cascade effects of oleate and CMPF result in the noticeable inhibition of furosemide binding in serum. Although to a lesser degree, linoleate and stearate exhibit a similar cascade effect on the furosemide binding in serum in the presence of CMPF. At the moment, the precise mechanism of the differences in the cascade effect of oleate, linoleate, and stearate is unclear, especially because these fatty acids are likely to bind to the same binding sites. Perhaps they differ in binding constants and allosteric effector activities.

Because CMPF is not removed by hemodialysis treatment [24], the further enhanced binding defect of furosemide in serum may occur in patients with renal insufficiency during hemodialysis therapy. In fact, the heparin-induced cascade effect of fatty acids on the serum binding of ketoprofen, which is bound to site II on HSA, was also observed for the fatty acid-indole uremic toxin-ketoprofen system [21]. Considering the findings to date, it is reasonable to postulate the hemodialysis-induced cascade interaction model in fatty acid-uremic toxin-drug systems, in which a transient increase in the concentrations of long-chain fatty acids could produce a cascade displacement of both site I- and II-bound drugs by their competitive inhibitors, namely, CMPF and uremic toxins that contain an indole ring (see Fig. 7).

In summary, accumulation of CMPF in patients with renal failure appears to account for a substantial portion of the impaired serum protein binding of furosemide observed in such patients. This CMPF-induced furosemide-binding defect may be further stimulated by the binding of high concentrations of long-chain fatty acids to HSA via a cascade mechanism. Recent studies have shown that CMPF is likely to possess inhibitory potency for the renal organic anion transport system [7,24]. Because furosemide is known to undergo active tubular secretion [25], CMPF may also interact with furosemide at the renal anion-transport system, which would serve to reduce the renal clearance of furosemide. Consequently, interactions of CMPF and furosemide with respect to serum protein binding and renal excretion may increase the free fraction of furosemide in serum of patients with renal insufficiency.

Received December 2, 1996; revision accepted July 7, 1997.


[1.] Brater DC. Diuretics. In: Williams RL, Brater DC, Mordenti J, eds. Rational therapeutics: a clinical pharmacologic guide for the health professional. New York: Marcel Dekker, 1989:269-315.

[2.] Uchino K, Isozaki S, Amano J, Tanaka N, Saitoh Y, Nakagawa F, et al. Clinical pharmacokinetics and diuretic effect of furosemide in plain tablet and retard capsule with normal subjects and cirrhotic patients. J Pharmacobio-Dyn 1983;6:684-91.

[3.] Barter DC. Clinical pharmacology of loop diuretics in health and disease. Eur Heart J 1992;13:10-4.

[4.] Gulyassy PF, Depner TA. Impaired binding of drugs and endogenous ligands in renal diseases. Am J Kidney Dis 1983;2:578601.

[5.] Kragh-Hansen U. Molecular aspects of ligand binding to serum albumin. Pharmacol Rev 1981;33:17-53.

[6.] Lindup WE. Plasma protein binding of drug--some basic and clinical aspects. In: Bridges JW, Chasseaud LF, Gibson CG. eds. Progress in drug metabolism, Vol. 10. London: Taylor and Francis, 1987:141-85.

[7.] Lindup WE, Henderson SJ, Barker CE. Drug binding in renal disease. In: Belpaire F, Bogaert M, Tillement JP, Verbeeck E, eds. Plasma binding of drug and its consequence. Ghent: Academic Press, 1991:103-20.

[8.] McNamara PJ, Lalka D, Gibaldi M. Endogenous accumulation products and serum protein binding in uremia. J Lab Clin Med 1981;98:730-40.

[9.] Bowmer CJ, Lindup WE. Decreased drug binding in uremia: effect of indoxyl sulfate and other endogenous substances on the binding of drugs and dyes to human albumin. Biochem Pharmacol 1982;31:319-23.

[10.] Liebich HM, Pickert A, Tetschner B. Gas chromatographic and gas chromatographic-mass spectrometric analysis of organic acids in plasma of patients with chronic I failure. J Chromatogr 1984;289: 259-66.

[11.] Ikeda K, Yoshitomi H, Nakayama T, Goto S, Kimura T. Plasma protein binding of furosemide in renal failure rabbits-investigation of endogenous protein binding inhibitors. J Pharm Pharmacol 1984;36:663-7.

[12.] Sakai T, Takadate A, Otagiri M. Characterization for binding site of uremic toxins on human serum albumin. Biol Pharm Bull 1995; 18:1755-61.

[13.] Yamaoka K, Tanigawara Y, Nakagawa T, Uno T. A pharmacokinetic analysis program (MULTI) for micro-computer. J Pharmacobio-Dyn 1981;4:879-85.

[14.] Chakrabarti SK. Cooperativity of warfarin binding with human serum albumin induced by free fatty acid anion. Biochem PharmaCol 1978;27:739-43.

[15.] Henderson SJ, Lindup WE. Interaction of 3-carboxyl-4-methyl-5-propyl-2-furan propanoic acid, an inhibitor of plasma protein binding in uremia, with human albumin. Biochem Pharmacol 1990;40:2543-8.

[16.] Lim CF, Bernard BF, de Jong M, Docter R, Krenning EP, Hennenyann G. A furan fatty acid and indoxyl sulfate are the putative inhibitors of thyroxine hepatocyte transport in uremia. J Clin Endocrinol Metab 1993;76:318-24.

[17.] Carter DC, Ho JX. Structure of serum albumin. Adv Protein Chem 1994;45:153-203.

[18.] Sjoholm I, Kober A, Odar-Cederlof I, Boaga O. Protein binding of drug in uremic and normal serum: the role of endogenous binding inhibitors. Biochem Pharmacol 1976;25:1205-13.

[19.] Naranjo CA, Sellers EM. Fatty acids modulation of drug binding to plasma proteins. In: Reidenberg MM, Erill S, eds. Drug-protein binding. New York: Praeger Scientific, 1986:233-51.

[20.] Birkett DJ, Myers SP, Hagedom J. Effect of fatty acids on the binding of drugs and bilirubin to human serum albumin. Adv Pharmacol Ther 1979;7:125-34.

[21.] Sakai T, Maruyama T, Imamura H, Shimada H, Otagiri M. Mechanism of stereoselective serum binding of ketoprofen after hemodialysis. J Pharmacol Exp Ther 1996;278:786-92.

[22.] Fehske KJ, Muller WE, Wollert U. The location of drug binding sites in human serum albumin. Biochem Pharmacol 1981;30:687-92.

[23.] Yamasaki K, Maruyama T, Kragh-Hansen U, Otagiri M. Characterization of site I on human serum albumin: concept about the structure of drug binding site. Biochim Biophys Acta 1996;1295: 147-57.

[24.] Niwa T, Takeda N, Maeda K, Shibata M, Tatematsu A. Accumulation of furancarboxylic acids in uremic serum as inhibitors of drug binding. Clin Chim Acta 1988;173:127-38.

[25.] Van Ginneken CAM, Russel FGM. Saturable pharmacokinetics in the renal excretion of drugs. Clin Pharmacokinet 1989;16:38-54.

(3) Nonstandard abbreviations: IS, indoxyl sulfate; CMPF, 3-carboxy-4-methyl-5-propyl-2-furanpropionate; IA, indole-3-acetate; HA, hippurate; HSA, human serum albumin; FFA, free fatty acids.


(1) Department of Pharmaceutics, Faculty of Pharmaceutical Sciences, Kumamoto University, 5-1 Oe-honmachi, Kumamoto 862, Japan.

(2) Department of Pharmacy, Miyazaki Medical College Hospital, Kiyatakecho, Miyazaki 889-16, Japan.

* Author for correspondence. Fax +81-96-362-7690; e-mail
COPYRIGHT 1997 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Drug Monitoring and Toxicology
Author:Takamura, Norito; Maruyama, Toru; Otagiri, Masaki
Publication:Clinical Chemistry
Date:Dec 1, 1997
Previous Article:Quantification of parathyroid hormone-related protein mRNA by competitive PCR and time-resolved lanthanide fluorometry.
Next Article:Development and validation of sensitive method for determination of serum cotinine in smokers and nonsmokers by liquid chromatography/atmospheric...

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters