Plasma protein binding in PET
The binding of PET radiopharmaceuticals to plasma proteins, in part, directs their biodistribution and pharmacokinetics, and affects the measurable parameters, like volume of distribution, binding potential, and net influx rate. These effects can be account for in the analysis, if the fraction of free radioligand in plasma can be reliably measured; this however is difficult, especially when the free fraction is very small.
All radiopharmaceuticals are susceptible to some degree of association with the components of the blood. Most commonly, a loose and rapidly exchanging lipophilic or hydrogen-bonding interaction takes place with blood proteins or the membranes of blood cells, especially erythrocytes. High lipophilicity often means high plasma protein binding (Zoghbi et al., 2012), which reduces glomerular filtration and increases circulation half-life. High plasma protein binding increases plasma-to-blood ratio, which is important when Blood TAC is converted to plasma TAC or vice versa. On the other hand, plasma-to-blood ratio can sometimes be used to measure the fraction of not protein bound and/or non-metabolized radiopharmaceutical in plasma (Salmon et al., 1990).
Tightly protein-bound radiopharmaceutical is practically non-diffusible, or at least less likely to partition across capillary membranes into tissues than the free radiopharmaceutical in plasma, during the short residence time of blood in the capillaries. Usually the equilibrium is rapid between bound and free radiopharmaceutical in the plasma, and therefore most of the protein-bound radiopharmaceutical is be available to a tissue even during a single capillary pass, affecting only the distribution between the compartments. During the typical PET scan also protein-bound radiopharmaceutical is expected to become available to tissue (Berridge, 2009). If equilibrium is not reached, that would decrease the transport rate constant from plasma to tissue (Berridge, 2009). Morgan and Huang (1993) have simulated the effect of plasma protein binding on volume of distribution, capillary clearance and elimination rate, and Buck and Burger (1996) the effects in PET studies. If the plasma protein binding process of the radioligand is so slow that equilibrium has not been achieved before the radiopharmaceutical reaches the target tissue, but fast enough to proceed markedly during the study, then the transfer rate constant K1 would become time-dependent, possibly causing severe biases in analysis (Buck & Burger, 1996).
Human serum albumin (HSA) is responsible for the majority of known drug-protein binding interactions in human blood, partly because of its abundance (55-60% of total serum protein, with plasma concentration of ∼0.6 mM or 30-45 g/L), and partly because of its molecular structure, including hydrophobic binding pockets, enabling it to reversibly bind various ligands, and some even with high affinity (Kragh-Hansen et al., 2002). Albumin mainly binds acidic and neutral compounds. Binding capacity of albumin is considered to be non-saturable.
Albumin is synthesized by specialized hepatocytes, and its half-life in serum is almost three weeks in humans, but less than two days in rodents. Most of the albumin is in extravascular pool, including lymphatic system, but intra- and extravascular albumin are in constant exchange. Plasma albumin concentration decreases by age, leading to different free fractions of certain drugs (Viani et al., 1992). Albumin is not glycosylated; its molecular weight is 66.5 kDa (585 amino acids), and size about 7.2 nm. Neonatal Fc receptors (FcRn) recognize albumin, transporting it across membranes, and protecting it from catabolism by endothelial cells and enabling its reabsorption by proximal tubules in the kidneys. FcRn is also found on epithelial cells of the liver, spleen, proximal small intestine, and lungs, and on white blood cells. Aspirin causes irreversible acetylation of albumin, which changes the affinity of albumin to some substrates. Salicylates and other drugs can also reversibly inhibit binding of radioligands to albumin by saturating specific binding pockets.
PET radiopharmaceutical binding affinity to nonhuman serum albumins is variable, and not predictable based on drug affinities for HSA (Robertson et al, 1990; Mathias et al., 1995; Basken et al., 2008). This may be one reason for the different applicability of radiopharmaceuticals in human and animal PET studies. Strong binding to albumin decreases the radiopharmaceuticals renal clearance. Albumin can however be labelled specifically to determine the plasma volume in tissue, and to detect vascular leakage.
Also some other plasma proteins have high affinity for certain substrates. For example, transferrin binds cationic iron, and also has a strong affinity for 68Ga3+. Basic compounds, lipophospholipids, and biliverdin are mainly bound and transported by α1-AGP, which is smaller than albumin but highly glycosylated; its half-life in plasma is only few days.
Tumour cells can trap albumin and other plasma proteins to be used as nutrients. Albumin is also accumulated at the sites of inflammation. Leakiness of blood vessels at affected sites may also increase the uptake of protein-bound PET radiopharmaceuticals (EPR effect).
Proteins involved in the complement system are abundant in plasma (∼3 g/L), and can interact with radioligands nonspecifically, but most importantly can lead to covalent opsonization of nanoparticles. Orosomucoid, or α1acid glycoprotein (AGP), is found in plasma in concentrations 0.5-1.5 g/L. It can bind certain drugs with high or moderate affinity. Immunoglobulin G is another abundant protein in serum. Fibrinogen is normally present in plasma in concentrations 15-40 mg/L.
Accurate measurement of plasma protein binding inside the region-of-interest in vivo during the PET study is impossible. Measurement from blood samples collected during the PET study are hampered by low radioactivity concentration, and label-carrying metabolites may cause bias in results (Yacobi and Levy, 1975). In vitro measurements will not provide the correct answer, because binding is dependent on pH (Hinderling and Hartmann, 2005; Kochansky et al., 2008), temperature, and other conditions, all of which are unknown in the region-of-interest. Since plasma is buffered by CO2 carbonic anhydrase system, pH in plasma samples will rise during storage, leading to overestimation of plasma protein binding (Kochansky et al., 2008; Ye et al., 2016).
Methods for analysis of plasma samples, including ultrafiltration and plasma/RBC partitioning, have been critically reviewed by Rowland (1980) and Wright et al (1996). Schuhmacher et al (2004) have improved the RBC partitioning method by replacing the RBCs with solid-supported lipid membranes. HPFA (Shibukawa et al., 1999) was recently applied in the protein binding measurements in PET; results were in good agreement with ultrafiltration, and were not affected by nonspecific binding (Amini et al., 2014).
Microdialysis techniques could be used to measure in vivo the concentration of unbound drugs (Tsai, 2003), or radiopharmaceuticals.
Amini N, Nakao R, Schou M, Halldin C. Determination of plasma protein binding of positron emission tomography radioligands by high-performance frontal analysis. J Pharm Biomed Anal. 2014; 98C: 140-143. doi: 10.1016/j.jpba.2014.05.024.
Banker MJ, Clark TH. Plasma/serum protein binding determinations. Curr Drug Metab. 2008; 9(9): 854-859. PMID: 18991581.
Berridge MS. The importance of kinetic enhancement. J Nucl Med. 2009; 50(8): 1203-1204. doi: 10.2967/jnumed.108.060905.
Buck A, Burger C. Effect of intravascular ligand binding on parameter estimates derived from tracer kinetic modelling. Eur J Nucl Med. 1996; 23(4): 422-430. doi: 10.1007/BF01247371.
Evans G (ed). A Handbook of Bioanalysis and Drug Metabolism. CRC Press, 2004.
Gillette JR. Overview of drug-protein binding. Ann N Y Acad Sci. 1973; 226: 6-17. doi: 10.1111/j.1749-6632.1973.tb20464.x.
Kochansky CJ, McMasters DR, Lu P, Koeplinger KA, Kerr HH, Shou M, Korzekwa KR. Impact of pH on plasma protein binding in equilibrium dialysis. Mol Pharm. 2008; 5(3): 438-448. doi: 10.1021/mp800004s.
Kragh-Hansen U. Molecular aspects of ligand binding to serum albumin. Pharmacol Rev. 1981; 33: 17-53. PMID: 7027277.
Mandula H, Parepally JMR, Feng R, Smith QR. Role of site-specific binding to plasma albumin in drug availability to brain. J Pharmacol Exp Ther. 2006; 317(2): 667-675. doi: 10.1124/jpet.105.097402.
Morgan DJ, Huang JL. Effect of plasma protein binding on kinetics of capillary uptake and efflux. Pharmaceutical Res. 1993; 10(2): 300-304. doi: 10.1023/A:1018959415963.
Nakao R, Amini N, Halldin C. Simultaneous determination of protein-free and total positron emission tomography radioligand concentrations in plasma using high-performance frontal analysis followed by mixed micellar liquid chromatography: application to [11C]PBR28 in human plasma. Anal Chem. 2013; 85(18): 8728-8734. doi: 10.1021/ac401742v.
Peters T. Serum albumin. Adv Protein Chem. 1985; 37: 161-245. PMID: 3904348.
Peters T Jr. All About Albumin. Biochemistry, Genetics, and Medical Applications. Academic Press, 1995. ISBN 978-0-12-552110-9.
Riant P, Urien S, Albergens E, Renouard A, Tillement JP. Effects of the binding of imipramine to erythrocytes and plasma proteins on its transport through the rat blood-brain barrier. J Neurochem. 1988; 51: 421-425. doi: 10.1111/j.1471-4159.1988.tb01055.x.
Rowland M. Plasma protein binding and therapeutic drug monitoring. Ther Drug Monitoring 1980; 2: 29-37. PMID: 6762701.
Schuhmacher J, Kohlsdorfer C, Bühner K, Brandenburger T, Kruk R. High-throughput determination of the free fraction of drugs strongly bound to plasma proteins. J Pharm Sci. 2004; 93(4): 816-830. doi: 10.1002/jps.10588.
Shibukawa A, Kuroda Y, Nakagawa T. High-performance frontal analysis for drug-protein binding study. J Pharm Biomed Anal. 1999; 18(6): 1047-1055. PMID: 9925341.
Smith QR, Fisher C, Allen DD. The Role of Plasma Protein Binding in Drug Delivery to Brain. In: Blood-Brain Barrier. Eds. Kobiler D, Lustig S, Shapira S. Springer, 2001, pp 311-321.
Sorger D, Becker GA, Hauber K, Schildan A, Patt M, Birkenmeier G, Otto A, Meyer P, Kluge M, Schliebs R, Sabri O. Binding properties of the cerebral α4β2 nicotinic acetylcholine receptor ligand 2-[18F]fluoro-A-85380 to plasma proteins. Nucl Med Biol. 2006; 33: 899-906.
Vraka C, Mijailovic S, Fröhlich V, Zeilinger M, Klebermass E-M, Wadsak W, Wagner K-H, Hacker M, Mitterhauser M. Expanding logP: present possibilities. Nucl Med Biol. 2018; 58: 20-32. doi: 10.1016/j.nucmedbio.2017.11.007.
Wright JD, Boudinot FD, Ujhelyi MR. Measurement and analysis of unbound drug concentrations. Clin Pharmacokinet. 1996; 30(6): 445-462. doi: 00003088-199630060-00003.
Ye M, Nagar S, Korzekwa K. A physiologically based pharmacokinetic model to predict the pharmacokinetics of highly protein-bound drugs and the impact of errors in plasma protein binding. Biopharm Drug Dispos. 2016; 37(3): 123-141. doi: 10.1002/bdd.1996.
Zoghbi SS, Anderson KB, Jenko KJ, Luckenbaugh DA, Innis RB, Pike VW. On quantitative relationships between drug-like compound lipophilicity and plasma free fraction in monkey and human. J Pharm Sci. 2012; 101(3): 1028-1039. doi: 10.1002/jps.22822.
Updated at: 2019-06-14
Created at: 2008-04-02
Written by: Vesa Oikonen, Päivi Marjamäki