Eur J Clin Pharmacol (1984) 27:459-463 European Journal of Clinical Pharmacology © Springer-Verlag 1984 Naloxone Protein Binding in Adult and Foetal Plasma L. A. Asali and K. F. Brown 1 1Department of Pharmacy, University of Sydney, Sydney, NSW, and Perinatal Pharmacology Laboratory, Foundation 41, Sydney, Australia Summary. Binding of naloxone hydrochloride was determined at 37 °C, by equilibrium dialysis against 0.067 M phosphate buffer, pH 7.4, in plasma ob- tained from 18 healthy adults, and 18 samples of um- bilical cord venous (foetal) plasma. The percentage free fraction (% free) in plasma was independent of naloxone concentration (9 ng/ml to 2.5 gg/ml). Per- cent free naloxone in adult (2 =54.0) was lower (p<0.01) than in foetal (,~ =61.5) plasma. In buf- fered solutions of purified HSA, %free naloxone (2 = 68.7) was independent of HSA concentration over the range 3.0 g/dl to 5.5 g/dl. Adult plasma con- centrations of cq-acid glycoprotein (al-AGP) and fl- lipoprotein were higher (p < 0.01) than foetal concen- trations. Furthermore %free naloxone in foetal plasma decreased with the in-vitro addition of puri- fied aI-AGP. It is suggested that qualitative differ- ences in adult and foetal albumin and quantitative differences in plasma levels of al-AGP and perhaps fl-lipoprotein are responsible for naloxone plasma binding differences between adults and the newborn. Key words: naloxone, plasma protein binding; al- bumin, alphal-acid glycoprotein, beta-lipoprotein, adult plasma, foetal/umbilical plasma Naloxone, a narcotic antagonist without agonist ac- tion, has become the drug of choice in the treatment of narcotic overdose [1], in the management of nar- cotic-induced depression in the newborn shortly af- ter birth [2] and in adults post-anaesthesia [3]. More recently, higher doses (0.5 to 1.0 mg/kg) of naloxone have been administered in the management of shock [4] and neonatal apnoea of prematurity [5]. Despite the widespread use of naloxone in adults and the newborn, no information is available on the binding of naloxone to plasma proteins. Albumin is the major binding protein for many acidic drugs. However, basic drugs may also exhibit high affinities for other plasma proteins including al-acid glycoprotein (al-AGP), lipoproteins and y- globulin. A preliminary study [6] of naloxone bind- ing to purified a1-AGP (50 mg/dl), reported that less than 20% of the drug was bound. However the con- centration of naloxone was not specified but was re- ported to be a realistic clinical level. The fractions of various acidic and basic drugs bound to the proteins in umbilical cord plasma have often been reported to be less than in adult plasma [7, 8]. Thus, the objectives of this study were firstly, to determine the extent of naloxone binding in plasma of healthy adults and in foetal cord plasma and to es- tablish the various plasma proteins responsible for variations in naloxone binding. Secondly, to estab- lish whether the binding of naloxone is linear over a wide range of plasma concentrations. Materials, Subjects and Methods Naloxone hydrochloride and naltrexone hydrochlo- ride were gifts from Endo Laboratory (Garden City, NY, USA). Aqueous standard solutions of these compounds were stored at 4°C. Purified al-AGP and human serum albumin (HSA; Fraction V, elec- trophoretic purity, 100%) were purchased from Beh- ringwerke AG (Marburg, FRG). All other chemicals were of analytical reagent grade. Binding of naloxone in plasma was determined in 18 healthy adult volunteers comprising 7 males and 11 females (age range 22 to 42 years) and in a
`
`Nalox1204
`Nalox-1 Pharmaceuticals, LLC
`Page 1 of 5
`
`

`

`460 L. A. Asali and K. F. Brown: Naloxone Plasma Protein Binding further 3 parturient women (maternal plasma), at de- livery. Binding of naloxone was also studied in the umbilical cord, venous plasma (foetal plasma) of 18 neonatal subjects. Venous blood samples were collected, by single venipuncture of an antecubital vein from the adult volunteers and from the clamped umbilical cords, in- to plastic tubes (Disposable Products, Sydney, Aus- tralia) containing ammonium heparin, 125 IU/a 0 ml. Plasma was harvested after centrifugation for 5 rain at 1000 g and stored frozen at - 18 °C to - 22 °C. Im- mediately prior to plasma binding determination, plasma was thawed at room temperature and centri- fuged to remove fibrous material. Naloxone HC1 (200ng/10~tl, equivalent to 180 ng/10 ~1 naloxone base) was dissolved in Soren- son's phosphate buffer, 0.067 M, pH 7.4, containing 3.5 g/dl HSA. 10 ~tl of this solution were added to 0.99 ml of HSA solution or plasma resulting in a final concentration of 180 ng/ml naloxone base. Binding of naloxone (180ng/ml) in solutions containing a range of concentrations of purified HSA (0.5 to 5.5 g/dl) in Sorenson's buffer was determined. In ad- dition, binding of a range of naloxone concentra- tions (18 ng/ml to 3.6 p.g/ml) to HSA (3.5 g/dl) was also determined. Furthermore, binding of naloxone to adult and foetal whole plasma samples was deter- mined using similar methodology. Binding of naloxone was determined by equili- brium dialysis in 1 ml, PTFE half dialysis cells (Spec- trum Medical Industries, Inc., California, USA). One ml aliquots of HSA solution or plasma were dialysed across a semipermeable membrane (Spectrapor II, Spectrum Medical Industries, Inc., California, USA) against an equal volume of Sorenson's phosphate buffer at 37°C by rotating for 1 h. Preliminary studies had confirmed that equilibrium was achieved within that period. At equilibrium, naloxone concen- trations in aliquots of the protein containing com- partment (300 gl) and buffer containing compart- ment (750 ~tl), were determined by the HPLC meth- od of Asali et al. [9] using naltrexone HC1 (50 ng/ 50 gl) as internal standard. The percentage of free naloxone (% free) was calculated as the ratio of nal- oxone concentrations in the buffer and protein con- taining compartments. The reproducibility of the binding methodology was established using HSA 3.5g/dl and found to be good (C.V.=3.8%, n=8, mean %free = 68.8%). Total protein and albumin concentrations in adult and foetal plasma were measured by Rapid Stat Kit (Pierce, Rockford, Ill, USA). Concentrations of a 1-AGP and fl-lipoprotein were measured by radi- al immunodiffusion (M-partigen, Behringwerke AG, Marburg, FRG). Statistical Analysis Unless otherwise stated, data are expressed as mean+SD. Statistical comparisons between the binding in plasma samples from adult and foetal groups were made using the Mann-Whitney U Test [10]. Correlations between endogenous plasma con- stituents and the plasma binding of naloxone were performed by linear, partial and multiple regression analyses [111. Results Influence of Variation in HSA Concentration The influence of HSA concentration on the binding of naloxone (180ng/ml) was determined (Fig.l). The %free naloxone remained constant at 68.7%, in- dependent of HSA concentrations between 3.0 to 5.5 g/dl. At 0.5 g/dl and 1.5 g/dl HSA, the %free were higher (94.6% and 84.5% respectively) as ex- pected. The constancy of %free with increasing HSA concentration above 3.0 g/dl reflects the fact that, at fixed total drug concentration the free concentration diminishes as the number of available sites increases. Influence of Variation in Naloxone Concentration The %free naloxone in 3.5g/dl HSA solution (68.7+0.865%), in plasma from an adult (55.9+ 0.748%) and a sample of umbilical venous plasma 100 LIJ UJ n~ 90 LL 70 000 • I I I I 2 3 s 6 Fig. 1. Variation in %free naloxone with HSA concentration r / o~ // oo i i J i // I I o.2 o.4 o.6 NALOXONE (Ng/ml) Fig. 2. %Free naloxone as a function of naloxone total concentra- tion in (D), 3.5g/dl HSA; (O), foetal plasma: and (O), adult plasma
`
`Nalox1204
`Nalox-1 Pharmaceuticals, LLC
`Page 2 of 5
`
`

`

`L. A. Asali and K. F. Brown: Naloxone Plasma Protein Binding 65 tu 6o ILl r~ It. 50 45 O o 35 SO 55 50 45 FOETAL ADULT L__p <o.o~ I Fig.3. %Free naloxone in foetal (umbilical venous) and adult plasma, spiked with an initial concentration of 180 ng/ml nalox- one. Arrows and bars indicate means and standard deviations Table 1. Plasma protein concentration in foetal, adult and mater- nal plasmas and naloxone %free Foetal 1 Adult 1 Maternal %Free 61.5 (4- 2.13) 2 54.1 (+ 4.40) 53.2 Albumin 3.83 (+ 0.361) 4.60 (4- 0.699) 3.32 [g/dl] al-AGP 15.4 (+ 7.75) 49.8 (+14.8) 56.6 [mg/dl] fl-Lipoprotein 85.6 (+25.9) 248 (+46.6) 223 [mg/dl] Total protein 5.70 (_+ 0.595) 6.56 (_ 1.09) 6.31 [g/dll n 18 18 3 1 All differences between adult and foetal mean values are signifi- cant (p<0.01); 2 figures in parentheses are SD (61.1+1.38%) was constant, independent of nalox- one concentration over the range of initial concen- trations from 18 ng/ml to 3.6 ~tg/ml (Fig.2). The range of equilibrium, naloxone concentrations (9 ng/ml to 2.25 ~tg/ml) is typical of the plasma levels encountered after high doses of naloxone (of the or- der of 0.5-1.0 mg/kg) are administered to neonates for the treatment of apnoea of prematurity (unpub- lished data). Binding of Naloxone to Adult and Foetal Plasma Naloxone binding was greater (p< 0.01) in both an adult and a single foetal plasma sample, than in HSA; indicating that significant non-albumin bind- ing of naloxone occurs in whole plasma (Fig. 2). Individual data for the binding of naloxone in adult and foetal plasma are compared in Fig. 3. The mean values of %free were higher (p< 0.01, Table 1) 461 in the foetal plasma samples (61.5 _+ 2.13%) than in the adult samples (54.0 _+ 4.40%). Plasma concentrations of albumin, at-AGP, fl- lipoprotein and total protein in adult samples were consistently higher (p<0.01) than in the samples of cord plasma (Table 1). A weak bivariate correlation was observed be- tween the values of %free naloxone and pooled val- ues of plasma albumin concentration (r=-0.465, p< 0.01, n=39) including adult, foetal and 3 mater- nal samples collected at delivery (Fig. 4). However stronger correlations (Figs. 5 and 6) were observed between %free naloxone and values of plasma al-AGP concentration (r= - 0.762, p< 0.01, n = 39), and plasma fl-lipoprotein concentration (r= -0.731, 65 Ixl A(~ Oh O • • • Q@ 45 • I I I I I I 4/3 4 5 ALBUMIN (gldl) Fig.4. Relationship between °free naloxone and plasma albu- min concentrations in foetal (©), adult (•) and maternal samples (0). Solid line, pooled data; Y= -3.4207 X+71.53; r=0.4651 ; p < 0.01. - - - -, foetal plasma; Y = - 0.5108 X + 63.46; r=-0.0866; p>0.05. - ...... , adult and maternal plasmas; Y = - 1.3504 X + 59.77; p > 0.05 65 0 mruJLu %•N 1.1_ • • 0"~ 4~ • s'o i;o #.-A 6 P(mg/dl) Fig.& Relationship between %free naloxone and pooled plasma al-AGP concentrations in foetal (©), adult (O) and maternal (0) samples. Y= -0.1815 X+63.63; r= -0.8089; p< 0.01. 651 08 • ,) • ' ' 2;o ' 3;o 8-LIPOPROTEIN (mg/dll Fig.& Relationship between %free naloxone and pooled plasma- lipoprotein concentrations in foetal (©), adult (O) and maternal (0) samples. Y= -0.0441 X+64.67; r= -0.7310; p<0.01
`
`Nalox1204
`Nalox-1 Pharmaceuticals, LLC
`Page 3 of 5
`
`

`

`462 L. A. Asali and K. F. Brown: Naloxone Plasma Protein Binding Table 2. Percentage of free naloxone in umbilical cord plasma before and after the in-vitro addition of purified al-AGP (Y~/0f~ = - 0.1937 X~-AGP + 64.02; r = - 0.9394) Cord Plasma a1-AGP %Free naloxone plasma concentration [mg/dl] Initial Final Initial Final 1 28.0 57.4 59.6 49.7 2 18.5 50.0 60.7 54.0 18.5 67.8 60.7 49.8 18.5 70.5 60.7 51.9 18.5 73.2 60.7 51.7 18.5 84.6 60.7 47.6 p<0.01, n=38, one female subject was excluded from the adult group as an outlier). To establish whether or not the correlation be- tween %free naloxone and at-AGP is real, concen- tration of at-AGP in two samples of cord plasma were increased by in-vitro addition of the purified protein to produce final concentrations equivalent to adult levels (Table 2). The percentage of free nalox- one decreased consistently with progressive in- creases in avAGP concentration (Y°/0free=-0.194 Xa,- ACP + 64.02; r= -- 0.939). The simultaneous influences of the concentra- tions of plasma al-AGP, albumin and fl-lipoprotein were examined by multiple linear regression analy- ses. When the combined effects of variation in al-AGP and fl-lipoprotein concentrations, on %free naloxone were examined, a value for multiple R 2 of 0.5592 was obtained. When the additional influence of variation in albumin was included with the two variables, multiple R 2 increased only marginally (RZ= 0.591). Discussion Although the plasma protein binding of naloxone is relatively weak, the %free in adult plasma (54.0%) was lower than that in cord plasma (61.5%). At HSA concentrations ranging from 3.0g/dl to 5.5g/dl, %free naloxone was 68.7% (Fig. 1). Thus, it is obvi- ous that plasma albumin is the major binding con- stituent in plasma but that binding also occurs to plasma constituents other than albumin. Despite the fact that albumin concentrations in adult plasma are higher than in foetal plasma, the %free naloxone in HSA solutions was independent of variation in HSA concentration over the range 3.0 g/dl to 5.5 g/all. This implies that the difference between foetal and adult plasma binding is unrelated to variation in albumin concentration. The weak bi- variate correlation between %free naloxone and plasma albumin concentration (r= - 0.465) of foetal, adult and maternal samples is therefore probably fortuitous. The relationship between %free naloxone and plasma albumin was further examined using partial correlation analysis where a1-AGP was held constant. Under these conditions, the linear correla- tion between %free naloxone and plasma albumin disappears (r= +0.228, p>0.05, n=39). This con- firmed that the correlation between %free and plas- ma albumin concentration is spurious and results from the interdependence of albumin and aI-AGP concentrations, which were correlated linearly (r= 0.455, p< 0.01, n= 39). It has been reported that albumin isolated from human cord blood (Alb-F) is different in several physicochemical respects [12] and in amino acid se- quence and content [13] from albumin isolated from adults (Alb-A). Both Alb-F and Alb-A were shown to coexist in roughly equal amounts in the serum of cord blood [14]. Foetal albumin is gradually replaced by Alb-A during ontogenic development during the first 4 to 5 months after birth. Thus, consistent with the findings for diazepam plasma binding [14] it would appear that there are qualitative differences between adult and foetal albumin with respect to naloxone plasma binding. Nation [15] reported that there was no significant difference between plasma albumin concentration in cord plasma and maternal plasma collected at deliv- ery. This is supported in this study in that albumin concentrations determined in 3 samples of maternal plasma, collected at delivery (Table 1) were similar in magnitude to the cord levels and lower than the con- centrations in the 18 healthy adults. Notwithstanding the lower maternal albumin concentrations com- pared with the adult plasma, the binding of naloxone was similar in both maternal and healthy adult plas- ma; and higher than in the cord plasma (Fig.3) which had albumin concentrations of similar magni- tude (Fig. 4). These findings would appear to provide strong evidence that the lower binding of naloxone in umbilical cord plasma relative to either maternal or healthy adult plasma is partly attributable to the existence of qualitative differences between adult and foetal albumins as suggested previously [12-14]. Thus the apparent weak correlation, described above, between %free naloxone with albumin con- centration data, pooled from both adult and cord plasmas, is not valid; since the cord and adult albu- mins belong to different populations. As expected correlations could not be demonstrated between %free naloxone and albumin concentration data (Fig.3) from either cord plasma (r=-0.0866; p> 0.05, n=18) or adult and maternal plasmas (r= -0.263,p> 0.05, n=21).
`
`Nalox1204
`Nalox-1 Pharmaceuticals, LLC
`Page 4 of 5
`
`

`

`k A. Asali and K. F. Brown: Naloxone Plasma Protein Binding 463 The finding of a positive correlation between %free naloxone and the acute phase protein a t-AGP (r= - 0.762) suggests that al-AGP is involved in nal- oxone binding as reported previously [6]. In proof of these proposals, elevation of the concentration of aI-AGP in 2 samples of cord plasma to adult con- centration levels by in-vitro addition of purified a t-AGP increased naloxone binding (Table 2) to a value within the adult range. Thus, c~1-AGP concen- tration is an important determinant of naloxone plasma binding, and is partially responsible for the fact that cord plasma binds naloxone to a lesser ex- tent than adult plasma. Partial regression analyses, correcting for the effects of albumin and fl-lipopro- tein, indicated that plasma a1-AGP accounted for the largest proportion (R 2--- 0.248) of the overall var- iability in %free naloxone. BickeI [16] showed that the binding affinity and capacity of lipoproteins for basic drugs including chlorpromazine and imipramine was at least as high as that of albumin, while Nilson [17] demonstrated that approximately 31% of quinidine plasma binding was associated with lipoproteins. Thus, the linear correlation observed between naloxone %free and plasma fl-lipoprotein concentration (r=-0.731) in this study suggests that the lower concentrations of fl-lipoprotein observed in the foetal group samples might also partially explain the lower binding of nal- oxone in cord plasma compared with that in the adult group. Partial regression analysis between %free naloxone and plasma fl-lipoprotein concentra- tion, with plasma albumin and al-AGP concentra- tions held fixed, indicate that variation in fl-lipopro- tein concentration accounts for only 14% of the variability in naloxone %free: Using multiple regression analysis it was shown that quantitative differences in both czl-AGP and fl- lipoprotein together account for only 55.9% of the overall variability in naloxone binding. Thus, it seems likely that qualitative differences in the bind- ing affinities between adult and foetal albumins and perhaps other variables, including the globulin frac- tion [18] may- account for the remaining proportion of the variability in naloxone binding. Acknowledgements. The authors gratefully acknowl- edge the receipt of a University of Sydney Postgrad- uate Award for LAA. The assistance of J. French in collecting blood samples is gratefully acknowledged. References 1. Simons PS (1973) The treatment of methadone poisoning with naloxone. J Pediatr 83 : 846 2. Wiener PC, Hogg MIJ, Rosen M (1977) Effect of naloxone on pethidine induced neonatal depression. Br Med J 2:228-231 3. Davis GK, Tolhurst-Cleaver CL, James TL (1980) Respiratory depression after intrathecal narcotics. Anaesthesia 35: 1080-1083 4. Amir S (1982) Opiate antagonists improve survival in anaphy- lactic shock. Eur J Pharmacol 80 [1]: 161 5. Burnard ED, John E, Henderson-Smart D, 'Ibdd DA (1983) Naloxone and recurrent apnoea of prematurity. Intensive Care Newborn 4:127-142 6. Romach MK, Piafsky KM, Abel JG, Khouw V, Sellers EM (1981) Methadone binding ro orosomucoid (avacid glycopro- tein): determinant of free fraction in plasma. Clin Pharmacol Ther 29 [2]: 211-217 7. Kurz H, Mauser-Ganshorn A, Stickel HH (1977) Differences in the binding of drugs to plasma proteins from newborn and adult man I. Europ J Clin Pharmacol 11 : 463-467 8. Piafsky KM, Woolner EA (1982) The binding of basic drugs to al-acid glycoprotein in cord serum. J Pediatr 100 [5]: 820-822 9. Asali LA, Nation RL, Brown KF (t983) Determination of nal- oxone in blood by high-performance-liquid-chromatography. J Chromatogr 278:329-335 10. Freund JE (1974) Modern elementary statistics. Prentice-Hail International: London 11. Nie NH, Hull CH, Jenkins JG, Steinbrenner K, Bent DH (t975) Statistical package for the social sciences. Mc Gregor & Werner, New York, USA 12. Miyoshi K, Saijo K, Kotani Y, Kashiwagi T, Kawai H (1966) Characteristic properties of fetal human albumin (Alb E) in isomerization equilibrium. Tokushima J Exp Med 13:121-128 13. Wallace S (1977) Altered plasma albumin in the newborn in- fant. Br J Clin Pharmacol 4:82-84 14. Ridd M J, Brown KF, Nation RL, Collier CB (1983) Differen- tial transplacental binding of diazepam: causes and implica- tions. Eur J Clin Pharmacol 24:595-601 15. Nation RL (1981) Meperidine binding in maternal and fetal plasma. Clin Pharmacol Ther 29 [4]: 472-479 16. Bickel MH (1975) Binding of chlorpmmazine and imipramine to red cells, albumin, lipoproteins and blood components. J Pharm Pharmacol 27:733-738 17. Nilson OG, Storstein L, Jacobsen S (1977) Effect of heparin and fatty acids on the binding of quinidine and warfarin in plasma. Biochem Pharmacol 26:22%235 18. Kurz H, Michels H, Stickel HH (1977) Differences in the bind- ing of drugs to plasma proteins from newborn and adult man II. Eur J Clin Pharmacol 11 : 469-472 Received: May 18, 1984 accepted: August 24, 1984 Kenneth F. Brown, Ph.D. The University of Sydney Department of Pharmacy NSW 2006, Sydney, Australia
`
`Nalox1204
`Nalox-1 Pharmaceuticals, LLC
`Page 5 of 5
`
`