Computational Approaches to Lipophilicity

In the design of new compounds as well as the design of experimental procedures an a priori calculation log P or log D values may be very useful. Methods may be based on the summation of fragmental [40-42], or atomic contributions [43-45], or a combination [46, 47]. Reviews on various methods can be found in references [40, 48-51]. Further approaches based on the used of structural features have been suggested [48,

52]. Atomic and fragmentai methods suffer from the problem that not all contributions may be parameterized. This leads to the observation that for a typical pharmaceutical file about 25 % of the compounds cannot be computed. Recent efforts have tried to improve the "missing value" problem [53].

Molecular lipophilicity potential (MLP) has been developed as a tool in 3D-QSAR, for the visualization of lipophilicity distribution on a molecular surface and as an additional field in CoMFA studies [49]. MLP can also be used to estimate conformation-dependent log P values.

Membrane Systems to Study Drug Behaviour

In order to overcome the limitations of octanol other solvent systems have been suggested. Rather than a simple organic solvent, actual membrane systems have also been utilized. For instance the distribution of molecules has been studied between unilamellar vesicles of dimyristoylphosphatidylcholine and aqueous buffers. These systems allow the interaction of molecules to be studied within the whole membrane which includes the charged polar head group area (hydrated) and the highly lipophilic carbon chain region. Such studies indicate that for amine compounds ionized at physiological pH, partitioning into the membrane is highly favoured and independent of the degree of ionization. This is believed to be due to electrostatic interactions with the charged phospholipid head group. This property is not shared

Fig. 1.9 Structures of charge neutral (phosphatidylcholine) and acidic (phosphatidylser-ine) phospholipids together with the moderately lipophilic and basic drug chlorphenter-mine. The groupings R1 and R2 refer to the acyl chains of the lipid portions.

1.7 Membrane Systems to Study Drug Behaviour 111

with acidic compounds even for the "electronically neutral" phosphatidylcholine [54]. Such ionic interactions between basic drugs are even more favoured for membranes containing "acidic" phospholipids such as phosphatidylserine [55]. The structures of these two phospholipids are shown in Figure 1.9 below together with the structure of the basic drug chlorphentermine.

Table 1.1 shows the preferential binding of chlorphentermine to phosphatidyl-choline-containing membranes, the phospholipid with overall acidic charge. These systems predict the actual affinity of the compound for the membrane, rather than its ability to cross the membrane. Membrane affinity, and hence tissue affinity, is particularly important in the persistence of drugs within the body, a topic which will be covered in Section 4.2.

Tab. 1.1: Affinity (k) and capacity (moles drug/moles lipid) of chlorphentermine for liposomes prepared from phosphatidylcholine and phosphatidylserine.


k (10-4) M







1 Pliska V, Testa B, Van de Waterbeemd H (Eds). Lipophilicity in Drug Action and Toxicology. VCH, Weinheim, 1996.

2 Van de Waterbeemd H, Carter RE, Grassy G, Kubinyi H, Martin YC, Tute MS, Willett P, Pure Chem. 1997, 69, 1137-1152; Ann. Rep. Med. Chem. 1998, 33, 397-409.

3 Dearden JC, Bresnen GM, Quant. Struct. Activity Relat. 1988, 7, 133-144.

4 Hersey A, Hill AP, Hyde RM, Livingstone DJ, Quant. Struct. Activity Relat. 1989, 8, 288-296.

5 Hansch C, Leo A, Substituent Constants for Correlation Analysis in Chemistry and Biology. Wiley Interscience, New York, 1979.

6 Hansch C, Leo A, Hoekman D, Exploring QSAR. Hydrophobic, Electronic, and Steric Constants. ACS, Washington, 1995.

7 Hansch C, Leo A, Exploring QSAR. Fundamentals and Applications in Chemistry and Biology. ACS, Washington, 1995.

8 Fujita T, Iwasa J, Hansch C, J. Amer. Chem. Soc. 1964, 86, 5175-5180.

9 Rekker RF, De Kort HM, Eur. J. Med. Chem. 1979, 14, 479-488.

10 Rekker RF, Mannhold R, Calculation of Drug Lipophilicity. VCH, Weinheim, 1992.

11 Leo A, Abraham DJ, Proteins: Struct. Funct. Gen. 1988, 2, 130-152.

12 Leo A, Hansch C, Elkins D, Chem. Rev. 1971, 71, 525-616.

13 Manners CN, Payling DW, Smith DA, Xenobiotica 1988, 18, 331-350.

14 Kulagowski JJ, Baker R, Curtis NR, Lee-son PD, Mawer IM, Moseley AM, Ridgill MP, Rowley M, Stansfield I, Foster AC, Gromwood S, Hill RG, Kemp JA, Marshall GR, Saywell KL, Tricklebank MD, J. Med. Chem. 1994, 37, 1402-1405.

15 Palm K, Luthman K, Ros J, Grasjo J, Ar-tursson P, J. Pharmacol. Exp. Ther. 1999, 291, 435-443.

16 Reymond F, Carrupt PA, Testa B, Girault HH, Chem. Eur. J. 1999, 5, 39-47.

17 Smith RN, Hansch C, Ames MM, J. Pharm. Sci. 1975, 64, 599-605.

18 Avdeef A, In: Lipophilicity in Drug Action and Toxicology (Eds Pliska V, Testa B, Van de Waterbeemd H), pp. 109-139. VCH, Weinheim, 1996.

19 Young RC, Mitchell RC, Brown ThH, Ganellin CR, Griffiths R, Jones M, Rana KK, Saunders D, Smith IR, Sore NE, Wilks TJ, J. Med. Chem. 1988, 31, 656-671.

20 El Tayar N, Tsai RS, Testa B, Carrupt PA. Leo A, J. Pharm. Sci. 1991, 80, 590-598.

21 Abraham MH, Chadha HS, Whiting GS, Mitchell RC, J. Pharm. Sci. 1994, 83, 1085-1100.

22 Van de Waterbeemd H, Testa B, Adv. Drug Res. 1987, 16, 85-225.

23 Van de Waterbeemd H, In: Methods for Assessing Oral Drug Absorption

(Ed. Dressman J), Marcel Dekker, New York, 2000, pp. 31-49.

24 Abraham MH, Chadra HS, Martins F, Mitchell RC, Pestic. Sci. 1999, 55, 78-99.

25 Van de Waterbeemd H, Camenisch G, Folkers G, Raevsky OA, Quant. Struct. Activity Relat. 1996, 15, 480-490.

26 Van de Waterbeemd H, Camenisch G, Folkers G, Chretien JR, Raevsky OA, J. Drug Target. 1998, 6, 151-165.

27 Ter Laak AM, Tsai RS, Donne-Op den Kelder GM, Carrupt PA, Testa B, Eur. J. Pharm. Sci. 1994, 2, 373-384.

28 N'Goka V, Schlewer G, Linget JM, Chambon J-P, Wermuth C-G, J. Med. Chem. 1991, 34, 2547-2557.

29 Hamilton HW, Steinbaugh BA, Stewart BH, Chan OH, Schmid HL, Schroeder R, Ryan MJ, Keiser J, Taylor MD, Blankley CJ, Kaltenbronn JS, Wright J, Hicks J, J. Med. Chem. 1995, 38,1446-1455.

30 Camenisch G, Folkers G, Van de Waterbeemd H, Pharm. Acta Helv. 1996, 71, 309-327.

31 Camenisch G, Folkers G, Van de Waterbeemd H, Eur. J. Pharm. Sci. 1998, 6, 321-329.

32 Van de Waterbeemd H, Eur. J. Pharm. Sci. 1997, 2 (Suppl.), S26-S27.

1.1 Physicochemistry and Pharmacokinetics 13

33 Leahy DE, Taylor PJ, Wait AR, Quant. Struct. Activity Relat. 1989, 8, 17-31.

34 Leahy DE, Morris JJ, Taylor PJ, Wait AR, J. Chem. Soc. Perkin Trans. 1992, 2, 723-731.

35 Suzuki N, Yoshida Y, Watarai H, Bull. Chem. Soc. Jpn. 1982, 55, 121-125.

36 Paterson DA, Conradi RA, Hilgers AR, Vidmar ThJ, Burton PhS, Quant. Struct. Activity Relat. 1994, 13, 4-10.

37 Yang CY, Cai SJ, Liu H, Pidgeon Ch, Adv. Drug Del. Rev. 1996, 23, 229-256.

38 Ottiger C, Wunderli-Allenspach H, Pharm. Res. 1999, 16, 643-650.

39 Balon K, Riebesehl BU, Muller BW, Pharm. Res. 1999, 16, 882-888.

41 Mannhold R, Rekker R.F, Dross K, Bijloo G, De Vries G, Quant. Struct. Activity Relat. 1998, 17, 517-536.

42 Rekker RF, Mannhold R, Bijloo G, De Vries G, Dross K, Quant. Struct. Activity Relat. 1998, 17, 537-548.

43 Kellogg GE, Joshi GS, Abraham DJ, Med. Chem. Res. 1992, 1, 444-453.

44 Viswanadhan VN, Ghose AK, Revankar GR, Robins RK, J. Chem. Inf. Comput. Sci. 1989, 29, 163-172.

45 Ghose AK, Crippen GM, J. Chem. Inf. Comput. Sci. 1987, 27, 21-35.

46 Meylan WM, Howard PH, J. Pharm. Soc. 1995, 84, 83-92.

47 Spessard GO, J. Chem. Inf. Comput. Sci. 1998, 38, 55-57.

48 Buchwald P, Bodor N, Curr. Med. Chem. 1998, 5, 353-380.

49 Carrupt PA, Testa B, Gaillard P, Rev. Comput. Chem. 1997, 11, 241-315.

50 Van de Waterbeemd H, Mannhold R, Quant. Struct. Activity Relat. 1996, 15, 410-412.

51 Mannhold R, Van de Waterbeemd H, J. Comput-Aid. Mol. Deg. 2001, 15, 337-354.

52 Moriguchi I, Hirono S, Nakagome I, Hirano H, Chem. Pharm. Bull. 1994, 42, 976-978.

53 Leo A, ACS Meeting, Anaheim, 1999.

54 Austin RP, Davis AM, Manners CN, J. Pharm. Sci. 1995, 84, 1180-1183.

55 Lullman H, Wehling M, Biochem. Pharmacol. 1979, 28, 3409-3415.

Pharmacokinetics and Metabolism in Drug Design 15 Edited by D. A. Smith, H. van de Waterbeemd, D. K. Walker, R. Mannhold, H. Kubinyi, H. Timmerman I

Copyright © 2001 Wiley-VCH Verlag GmbH ISBNs: 3-527-30197-6 (Hardcover); 3-527-60021-3 (Electronic)



ADME Absorption, distribution, metabolism and excretion

CNS Central nervous system

CYP2D6 Cytochrome P450 2D6 enzyme

GIT Gastrointestinal tract i.v. Intravenous

PET Positive emission tomography



Average amount of drug in the body over a dosing interval


Maximum amount of drug in the body over a

dosing interval


Minimum amount of drug in the body over a

dosing interval


Area under plasma concentration time curve


Initial concentration after i. v. dose


Average plasma concentration at steady state


Free (unbound) plasma concentration


Initial free (unbound) plasma concentration


Steady state concentration




Unbound clearance


Hepatic clearance


Intrinsic clearance


Intrinsic clearance of unbound drug


Oral clearance


Plasma clearance


Renal clearance


Systemic clearance






Fractional response


Maximum response

F Fraction of dose reaching systemic circulation (bioavailability)

fu Fraction of drug unbound

Ka Affinity constant

Kb Dissociation constant for a competitive antagonist

Kd Dissociation constant kei Elimination rate constant

Km Affinity constant (concentration at 50 % Vmax)

ko Infusion rate k+j Receptor on rate k_j Receptor off rate

L Ligand log D7a Distribution coefficient (octanol/buffer) at pH 7.4

ln2 Natural logarithm of 2 (i. e. 0.693)

pA2 Affinity of antagonist for a receptor (= - log10[KB])

Q Blood flow

R Receptor

RL Receptor ligand complex

RO Receptor occupancy s Substrate concentration t time after drug administration

T Dosing interval tj/2 Elimination half-life

Vd Volume of distribution

Vd(f| Apparent volume of distribution of free (unbound) drug

Vmax Maximum rate of reaction (Michaelis-Menten enzyme kinetics)

£ Dosing interval in terms of half-life (= T/t1/2)

Setting the Scene

Pharmacokinetics is the study of the time course of a drug within the body and incorporates the processes of absorption, distribution, metabolism and excretion (ADME). In general, pharmacokinetic parameters are derived from the measurement of drug concentrations in blood or plasma. The simplest pharmacokinetic concept is that based on total drug in plasma. However, drug molecules may be bound to a greater or lesser extent to the proteins present within the plasma, thus free drug levels may be vastly different from those of total drug levels. Blood or plasma are the traditionally sampled matrices due to (a) convenience and (b) to the fact that the concentrations in the circulation will be in some form of equilibrium with the tissues of the body. Because of analytical difficulties (separation, sensitivity) it is usually the total drug that is measured and used in pharmacokinetic evaluation. Such measurements and analysis are adequate for understanding a single drug in a single species in a number of different situations since both protein binding and the resultant unbound fraction are approximately constant under these conditions. When species or

2.2 Intravenous Administration: Volume of Distribution 117

drugs are compared, certain difficulties arise in the use of total drug and unbound (free) drug is a more useful measure (see below).

Intravenous Administration: Volume of Distribution

When a drug is administered intravenously into the circulation the compound undergoes distribution into tissues etc. and clearance. For a drug that undergoes rapid distribution a simple model can explain the three important pharmacokinetic terms: volume of distribution, clearance and half-life.

Volume of distribution (Vd) is a theoretical concept that connects the administered dose with the actual initial concentration (C0) present in the circulation. The relationship is shown below:

For a drug that is confined solely to the circulation (blood volume is 80 mL kg-1) the volume of distribution will be 0.08 L kg-1. Distribution into total body water (800 mL kg-1) results in a volume of distribution of 0.8 L kg 1. Beyond these values the number has only a mathematical importance. For instance a volume of distribution of 2 L kg-1 means only, that less than 5 % of the drug is present in the circulation. The drug may be generally distributed to many tissues and organs or concentrated in only a few.

For different molecules, the apparent volume of distribution may range from about 0.04 L kg-1 to more than 20 L kg-1. High molecular weight dyes, such as indo-cyanine green, are restricted to the circulating plasma after intravenous administration and thus exhibit a volume of distribution of about 0.04 L kg-1. For this reason such compounds are used to estimate plasma volume [1] and hepatic blood flow [2]. Certain ions, such as chloride and bromide, rapidly distribute throughout extracellular fluid, but do not readily cross cell membranes and therefore exhibit a volume of distribution of about 0.4 L kg-1 which is equivalent to the extracellular water volume [3]. Neutral lipid-soluble substances can distribute rapidly throughout intracellular and extracellular water. For this reason antipyrine has been used as a marker of total body water volume and exhibits a volume of distribution of about 0.7 L kg-1 [4]. Compounds which bind more favourably to tissue proteins than to plasma proteins can exhibit apparent volumes of distribution far in excess of the body water volume. This is because the apparent volume is dependent on the ratio of free drug fractions in the plasma and tissue compartments [5]. High tissue affinity is most commonly observed with basic drugs and can lead to apparent volumes of distribution up to 21 L kg-1 for the primary amine-containing calcium channel blocker, amlodipine [6].

Intravenous Administration: Clearance

Clearance of drug occurs by the perfusion of blood to the organs of extraction. Extraction (E) refers to the proportion of drug presented to the organ which is removed irreversibly (excreted) or altered to a different chemical form (metabolism). Clearance (Cl) is therefore related to the flow of blood through the organ (Q) and is expressed by the formula:

The organs of extraction are generally the liver (hepatic clearance - metabolism and biliary excretion; ClH) and the kidney (renal excretion, ClR) and the values can be summed together to give an overall value for systemic clearance (ClS):

Extraction is the ratio of the clearance process compared to the overall disappearance of the compound from the organ. The clearance process is termed intrinsic clearance Cli, the other component of disappearance is the blood flow (Q) from the organ. This is shown in Figure 2.1 below.

Fig. 2.1 Schematic illustrating hepatic extraction with Q, blood flow and Cl^ intrinsic clearance (metabolism).

Combining Eqs. (2.2) and (2.3) with the scheme in Figure 2.1 gives the general equation for clearance:

Where Cl = Cls if only one organ is involved in drug clearance. Within this equation Cli is the intrinsic clearance based on total drug concentrations and therefore includes drug bound to protein. Lipophilic drugs bind to the constituents of plasma (principally albumin) and in some cases to erythrocytes. It is a major assumption, supported by a considerable amount of experimental data, that only the unbound (free) drug can be cleared. The intrinsic clearance (Cli) can be further defined as:

Where Cliu is the intrinsic clearance of free drug, i. e. unrestricted by either flow or binding, and fu is the fraction of drug unbound in blood or plasma.

Inspection of the above equation indicates for compounds with low intrinsic clearance compared to blood flow, Q and (Cli + Q) effectively cancel and Cl (or Cls) approximates to Cli. Conversely, when intrinsic clearance is high relative to blood flow,

2.4 Intravenous Administration: Clearance and Half-life 19

Fig. 2.2 Inter-relationship between various terms of drug clearance used within pharmacokinetic analysis.

2.4 Intravenous Administration: Clearance and Half-life 19

Fig. 2.2 Inter-relationship between various terms of drug clearance used within pharmacokinetic analysis.

Cli and (Cli + Q) effectively cancel and Cl (or ClS) is equal to blood flow (Q). The implications of this on drugs cleared by metabolism is that the systemic clearance of low clearance drugs are sensitive to changes in metabolism rate whereas that of high clearance drugs are sensitive to changes in blood flow.

It is important to recognize the distinction between the various terms used for drug clearance and the inter-relationship between these. Essentially intrinsic clearance values are independent of flow through the organ of clearance, whilst unbound clearance terms are independent of binding. These relationships are illustrated in Figure 2.2.

Intravenous Administration: Clearance and Half-life

Clearance is related to the concentrations present in blood after administration of a drug by the equation:

where AUC is the area under the plasma concentration time curve. Clearance is a constant with units often given as mL min-1 or mL min-1 kg-1 body weight. These values refer to the volume of blood totally cleared of drug per unit time. Hepatic blood flow values are 100, 50 and 25 mL min-1 kg-1 in rat, dog and man respectively. Blood clearance values approaching these indicate that hepatic extraction is very high (rapid metabolism).

Blood arriving at an organ of extraction normally contains only a fraction of the total drug present in the body. The flow through the major extraction organs, the liver and kidneys, is about 3 % of the total blood volume per minute, however, for many drugs, distribution out of the blood into the tissues will have occurred. The duration of the drug in the body is therefore the relationship between the clearance (blood flow through the organs of extraction and their extraction efficiency) and the amount of the dose of drug actually in the circulation (blood). The amount of drug in the circulation is related to the volume of distribution and therefore to the elimination rate constant (kel) which is given by the relationship:

The elimination rate constant can be described as a proportional rate constant. An elimination rate constant of 0.1 h-1 means that 10 % of the drug is removed per hour.

The elimination rate constant and half-life (fy2), the time taken for the drug concentration present in the circulation to decline to 50 % of the current value, are related by the equation:

Half-life reflects how often a drug needs to be administered. To maintain concentrations with minimal peak and trough levels over a dosing interval a rule of thumb is that the dosing interval should equal the drug half-life. Thus for once-a-day administration a 24-h half-life is required. This will provide a peak-to-trough variation in plasma concentration of approximately two-fold. In practice the tolerance in peak-to-trough variation in plasma concentration will depend on the therapeutic index of a given drug and dosing intervals of two to three half-lives are not uncommon.

The importance of these equations is that drugs can have different half-lives due either to changes in clearance or changes in volume (see Section 2.7). This is illustrated in Figure 2.3 for a simple single compartment pharmacokinetic model where the half-life is doubled either by reducing clearance to 50 % or by doubling the volume of distribution.



Fig. 2.3 Effect of clearance and volume of distribution on half-life for a simple single compartment pharmacokinetic model.


Intravenous Administration: Infusion

With linear kinetics, providing an intravenous infusion is maintained long enough, a situation will arise when the rate of drug infused = rate of drug eliminated. The

2.5 Intravenous Administration: Infusion 21

Fig. 2.4 Plasma concentration profile observed after intravenous infusion.

Fig. 2.4 Plasma concentration profile observed after intravenous infusion.

plasma or blood concentrations will remain constant and be described as "steady state". The plasma concentration profile following intravenous infusion is illustrated in Figure 2.4.

The steady state concentration(Css) is defined by the equation:

where ko is the infusion rate and Clp is the plasma (or blood) clearance. The equation which governs the rise in plasma concentration is shown below where the plasma concentration (Cp) may be determined at any time (t).

Thus the time taken to reach steady state is dependent on kel. The larger kel (shorter the half-life) the more rapidly the drug will attain steady state. As a guide 87 % of steady state is attained when a drug is infused for a period equal to three half-lives. Decline from steady state will be as described above, so a short half-life drug will rapidly attain steady state during infusion and rapidly disappear following the cessation of infusion.

Increasing the infusion rate will mean the concentrations will climb until a new steady state value is obtained. Thus doubling the infusion rate doubles the steady state plasma concentration as illustrated in Figure 2.5.

Oral Administration

When a drug is administered orally, it has to be absorbed across the membranes of the gastrointestinal tract. Incomplete absorption lowers the proportion of the dose able to reach the systemic circulation. The blood supply to the gastrointestinal tract (GIT) is drained via the hepatic portal vein which passes through the liver on its passage back to the heart and lungs. Transport of the drug from the gastrointestinal tract to the systemic circulation will mean the entire absorbed dose has to pass through the liver.

On this "first-pass" the entire dose is subjected to liver extraction and the fraction of the dose reaching the systemic circulation (F) can be substantially reduced (even for completely absorbed drugs) as shown in the following equation:

Again E is the same concept as that shown in Figure 2.1. This phenomenon is termed the first-pass effect, or pre-systemic metabolism, and is a major factor in reducing the bioavailability of lipophilic drugs. From the concept of extraction shown in Figure 2.1, rapidly metabolized drugs, with high Cli values, will have high extraction and high first-pass effects. An example of this type of drug is the lipophilic calcium channel blocker, felodipine. This compound has an hepatic extraction of about 0.80, leading to oral systemic drug exposure (AUC) of only about one-fifth of that observed after intravenous administration [7]. Conversely, slowly metabolized drugs, with low Cli values, will have low extraction and show small and insignificant firstpass effects. The class III anti-dysrhythmic drug, dofetilide, provides such an example. Hepatic extraction of this compound is only about 0.07, leading to similar systemic exposure (AUC) after oral and intravenous doses [8].

Fig. 2.6 Schematic illustrating the disposition of a drug after oral administration.

2.7 Repeated Doses 23

A complication of this can be additional first-pass effects caused by metabolism by the gastrointestinal tract itself. In the most extreme cases, such as midazolam, extraction by the gut wall may be as high as 0.38 to 0.54 and comparable to that of the liver itself [9].

The previous equations referring to intravenously administered drugs (e.g. Eq. 2.6) can be modified to apply to the oral situation:

Where Clo is the oral clearance and F indicates the fraction absorbed and escaping hepatic first-pass effects. Referring back to the intravenous equation we can calculate F or absolute bioavailability by administering a drug intravenously and orally and measuring drug concentrations to derive the respective AUCs. When the same dose of drug is given then:

The estimation of systemic clearance together with this value gives valuable information about the behaviour of a drug. High clearance drugs with values approaching hepatic blood flow will indicate hepatic extraction (metabolism) as a reason for low bioavailability. In contrast poor absorption will probably be the problem in low clearance drugs which show low bioavailabilities.

Repeated Doses

When oral doses are administered far apart in time they behave independently. This is usually not the desired profile if we assume that a certain concentration is needed to maintain efficacy and if a certain concentration is exceeded side-effects will occur. Giving doses of the drug sufficiently close together so that the following doses are administered prior to the full elimination of the preceding dose means that some accumulation will occur, moreover a smoothing out of the plasma concentration profile will occur. This is illustrated in Figure 2.7.

Ultimately if doses are given very close together then the effect is that of intravenous infusion and a steady state occurs. In fact for any drug an average steady state value (Cavss) can be calculated:

where T represents the interval between doses, Dose is the size of a single administered dose and Cl is clearance. Note that F ■ Dose/T in this equation is actually the dosing rate as for intravenous infusion.

The same relationship to kd and half-life also apply, so that as with intravenous infusion 87.5 % of the final steady state concentration is achieved following administration of the drug for three half-lives.

This equation can be rewritten to indicate the amount of drug in the body by substitution of kel for Clo. Since kel = 0.693/t1/2 the following equation emerges:

where A^ is the average amount of drug in the body over the dosing interval. By relating this to each dose an accumulation ratio (Rac) can be calculated:

The maximum and minimum amounts in the body (Amax and Amin respectively) are defined by:

where £ = T/t1/2 or the dosing interval defined in terms of half-life. These equations mean that for a drug given once a day with a 24-h half-life then a steady state will be largely achieved by 3-4 days. In addition, the amount of drug in the body (or the plasma concentration) will be approximately 1.4 times that of a single dose and that this will fluctuate between approximately twice the single dose and equivalent to the dose.

Development ofthe Unbound (Free) Drug Model

As outlined earlier, pharmacokinetics based on total drug concentrations proves useful in many situations, but is limited when data from a series of compounds are compared. This is the normal situation in a drug discovery programme and alternative presentations of pharmacokinetic information need to be explored. Since the medicinal chemist is trying to link compound potency in in vitro systems (receptor binding, etc.) with behaviour in vivo; it is important to find ways to unify the observations.

Measurement of the unbound drug present in the circulation and basing pharmacokinetic estimates on this, allows the in vitro and in vivo data to be rationalized.

2.9 Unbound Drug and Drug Action 25

Fig. 2.8 Schematic illustrating equilibrium between drug and receptor.

2.9 Unbound Drug and Drug Action 25

Fig. 2.8 Schematic illustrating equilibrium between drug and receptor.

The first and possibly the simplest biological test for a drug is the in vitro assessment of affinity for its target. Such experiments can be shown schematically as illustrated in Figure 2.8 in which the drug is added to the aqueous buffer surrounding the receptor (or cell or tissue) and the total drug added is assumed to be in aqueous solution and in equilibrium with the receptor.

Unbound Drug and Drug Action

The biological or functional response to receptor activation can be assumed to be directly proportional to the number of receptors (R) occupied by a given ligand (L) at equilibrium. This assumption is termed the occupancy theory of drug response. The equation describing this phenomenon was proposed as:

where EF is the fractional response, EM is the maximal response, [RL] is the concentration of receptor-ligand complex and [R]T is the total receptor concentration. At equilibrium, R + L ^ RL, such that the affinity constant KA can be defined as Ka - [RL]/[L][R]. This is the same equation as that derived from Langmuir's saturation isotherm, which derives from the law of mass action. It is possible to describe the occupancy theory in the following way:

• the receptor/ligand (RL) complex is reversible

• association is a bimolecular process

• dissociation is a monomolecular process

• all receptors of a given type are equivalent and behave independently of one another

• the concentration of ligand is greatly in excess of the receptor and therefore the binding of the ligand to the receptor does not alter the free (F) concentration of the ligand

• the response elicited by receptor occupation is directly proportional to the number of receptors occupied by the ligand

The equilibrium dissociation constant Kd gives a measure of the affinity of the ligand for the receptor.

Kd can also be defined by the two microconstants for rate on and off k+1 and k_1 so that Kd - k_1/k+1, where Kd is the concentration of the ligand (L) that occupies 50 % of the available receptors.

Antagonist ligands occupy the receptor without eliciting a response, thus preventing agonist ligands from producing their effects. Since this interaction is usually competitive in nature, an agonist can overcome the antagonist effects as its concentration is increased. The competitive nature of this interaction allows the determination of a pA2 value, the affinity of an antagonist for a receptor as shown below.

where KB = the dissociation constant for a competitive antagonist and is the ligand concentration that occupies 50 % of the receptors.

We thus have a series of unbound drug affinity measures relating to the action of the drug. The values are those typically obtained by the pharmacologist and form the basis of the structure-activity relationships which the medicinal chemist will work on. It is possible to extend this model to provide a pharmacokinetic phase as shown in Figure 2.9.

Here we assume that:

• free drug is in equilibrium across the system

• only free drug can exert pharmacological activity (see above)

• drug is reversibly bound to tissues and blood

• only free drug can be cleared

To examine the validity of this model, data from a number of 7 transmembrane (7TM) receptor antagonists (antimuscarinics, antihistaminics, p-adrenoceptor blockers etc.) were examined. The KB values for these drugs were compared to their free (unbound) plasma concentration. To simplify the analysis the plasma concentration data was taken from patients at steady state on therapeutic doses. Steady state means that the dosing rate (rate in) is balanced by the clearance rate (rate out). This concept is exactly as described earlier for intravenous infusion, however the steady state is an average of the various peaks and troughs that occur in a normal dosage regimen. The relationship between the values was very close and the in vitro potency values can be adjusted to 75 % receptor occupancy (RO) rather than 50% using Eq. (2.21) shown below (where the ligand concentration is represented by L):

2.10 Unbound Drug Model and Barriers to Equilibrium 27

Fig. 2.10 Correlation of in vitro potency with plasma free drug concentration required for efficacy.

2.10 Unbound Drug Model and Barriers to Equilibrium 27

Fig. 2.10 Correlation of in vitro potency with plasma free drug concentration required for efficacy.

When this relationship is plotted, a 1:1 relationship is seen as shown in Figure 2.10. Thus the free concentration present in plasma is that actually seen at the receptor. Moreover, the in vitro values (KB) determined from receptor binding actually represent the concentration required in the patient for optimum efficacy.

We can thus see that the traditional indicators of potency that drive synthetic chemistry, such as pA2 values, can have direct relevance to the plasma concentration (free) required to elicit the desired response. If we extend this example further it is unlikely that in all cases there is a simple direct equilibrium for all compounds between the free drug in plasma and the aqueous media bathing the receptor. The concentration of the free drug in the plasma is in direct equilibrium with the interstitial fluid bathing most cells of the body, since the capillary walls contain sufficient numbers of pores to allow the rapid passage of relatively small molecules, regardless of physicochemistry. Most receptor targets are accessed extracellularly. We can expect therefore that all drugs, regardless of their physicochemistry, will be in direct equilibrium at these targets, with the free drug in plasma. For instance the G-protein-coupled receptors have a binding site which is accessible to hydrophilic molecules.

This is exemplified by the endogenous agonists of these receptors that are usually hydrophilic by nature. Adrenalin, dopamine and histamine are representative and have log D74 values of - 2.6, - 2.4 and - 2.9 respectively.

The antagonists included in Figure 2.10 range in log D7 4 value. For example, within the p-adrenoceptor antagonists the range is from - 1.9 for atenolol to 1.1 for propranolol. This range indicates the ease of passage from the circulation to the receptor site for both hydrophilic and lipophilic drugs.


Unbound Drug Model and Barriers to Equilibrium

In some cases barriers such as the blood-brain barrier exist, in other cases the target is intracellular. Here the model has to be extended to place the receptor in a biophase

Fig. 2.11 Schematic pharmacodynamic/pharmacokinetic model incorporating a biophase and a drug binding compartment.

(Figure 2.11). The model also includes a "compartment" for the drug that is re-versibly bound to tissues and blood. The significance of this will be explored after further examination of the role of the biophase.

Aqueous channels are much fewer in number in the capillaries of the brain (blood-brain barrier) and rapid transfer into the brain fluids requires molecules to traverse the lipid cores of the membranes. Actual passage into cells, like crossing the blood-brain barrier, also requires molecules to traverse the lipid core of the membrane due to the relative paucity of aqueous channels. Distribution of a drug to the target, whether a cell membrane receptor in the CNS or an intracellular enzyme or receptor, therefore critically determines the range of physicochemical properties available for the drug discoverer to exploit. The access of the CNS to drugs is illustrated by reference to a series of dopamine D2 antagonists. Here receptor occupancy can be measured by the use of PET scanning and this "direct measure" of receptor occupancy can be compared with theoretical occupation calculated from the free drug plasma concentrations in the same experiment. Figure 2.12 shows this comparison for the lipophilic antagonists remoxipride, haloperidol and thioridazine and the hydrophilic compound sulpiride.

100 I 90

Was this article helpful?

0 0

Post a comment