Write a review paper comparing the two articles from journal?

A u

M D

a

A R R A A

K P I C P E

1

i p I m p s w o s e d m t a a r 4 c o e p i

0 d

Journal of Pharmaceutical and Biomedical Analysis 55 (2011) 385–390

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j p b a

kinetic method for the determination of plasma protein binding of compounds nstable in plasma: Specific application to enalapril

ark C. Wenlock ∗, Patrick Barton, Rupert P. Austin epartment of Medicinal Chemistry, AstraZeneca R&D Charnwood, Loughborough LE11 5RH, UK

r t i c l e i n f o

rticle history: eceived 6 January 2011 eceived in revised form 1 February 2011 ccepted 2 February 2011

a b s t r a c t

Traditional methods for the determination of plasma protein binding (PPB), such as equilibrium dialysis and ultrafiltration, normally operate on a timescale ranging from tens of minutes to several hours and are not suitable for measuring compounds that have significant chemical degradation on this timescale. One such compound is enalapril. Although stable in human plasma enalapril is subject to rapid esterase-

vailable online 2 March 2011

eywords: lasma protein binding nitial rates harcoal binding

catalyzed hydrolysis in rat plasma. A method has been developed which allows the extent of rat PPB of enalapril to be determined from initial rates kinetics of the adsorption of the unstable compound to dextran coated charcoal (DCC). The method has been applied to stable compounds, and the results are consistent with those from traditional equilibrium dialysis experiments. The experimental method is simple to run, requires no specialized equipment, and can potentially be applied to other compounds

lasma

lasma instability nalapril

that show instability in p

. Introduction

The extent of binding of a drug to plasma proteins is an mportant property which has a large influence on the efficacy, harmacokinetics and toxicology of the compound in vivo [1–4].

t is a property which undergoes much measurement and opti- ization during the drug discovery process, but has suffered in the

ast because the experimental methodology was very labor inten- ive and lacked automation [5]. This has recently been addressed ith the development of higher throughput technologies based

n multi-well equilibrium dialysis [4,6–8] and ultrafiltration [9] ystems, and also with the use of mixtures of compounds in ach incubation [4,9,10], facilitated by modern mass spectrometry etectors with high sensitivity and fast scanning rates. These new ethods are becoming more widespread and will greatly facili-

ate the optimization cycle in drug discovery. However, the routine pplication of these methods is not suitable for compounds that re chemically unstable in plasma, particularly when the chemical eaction is fast compared with the long equilibration time (typically h or more) of the experiment. Chemical instability of research ompounds in plasma is not an uncommon phenomenon, and is

ften a consequence of hydrolysis of ester groups catalyzed by sterases in the plasma [11]. Plasma instability is not necessarily a roperty which will render a drug unsuitable for use. If a compound

s much more unstable in plasma than in other tissues but has a high

∗ Corresponding author. Tel.: +44 1509 645310; fax: +44 1509 645576. E-mail address: [email protected] (M.C. Wenlock).

731-7085/$ – see front matter © 2011 Elsevier B.V. All rights reserved. oi:10.1016/j.jpba.2011.02.006

where traditional experimental techniques are unsuitable. © 2011 Elsevier B.V. All rights reserved.

volume of distribution, then it can have a much longer pharmacoki- netic terminal half life than the half life in plasma in vitro. Hence plasma instability will not necessarily lead to poor pharmacoki- netics due to short half life. Furthermore, in developing effective prodrugs and antedrugs therapies, plasma instability can be a pur- posefully designed feature where upon systematic exposure a drug is either activated (prodrug) or deactivated (antedrug) [12–14]. In order to gain greater understanding of the efficacy, pharmacoki- netics and toxicology of plasma unstable compounds, it is valuable to generate a good estimate of the free concentration of the com- pound in plasma, and a determination of the extent of PPB will be a key component of the free concentration estimate.

Enalapril is a prodrug that contains an ester group that is hydrolyzed by esterases to enalaprilat an angiotensin-converting enzyme inhibitor [15]. Enalapril displays very little hydrolysis in human plasma but rapid hydrolysis in rat plasma [16,17]. Hence equilibrium dialysis could be employed to determine the extent of PPB in human plasma but not in rat plasma. Therefore to measure the rat PPB of enalapril an experimental PPB method that could operate on a short timescale was developed by the modification of existing methods based on adsorption of compounds to DCC [18–20]. The use of DCC in plasma binding determinations is based on the fact that compounds will adsorb more slowly onto DCC in the presence of plasma than in the absence of plasma due to the low-

ered free concentration of compound in plasma. The first reported DCC method required determination of the full time course of adsorption of the drug to DCC both in the presence and absence of plasma [18,19]. Nonlinear curve fitting of the data to a derived kinetic model then allowed the extraction of the extent of plasma

3 ical an

b w o t d a t r a 3 h t fi a c m c a m c e c w

2

2

p g C e A 3 ( C A ( p c

2

I A e m M a u s C d a o

2

o c d c 1

86 M.C. Wenlock et al. / Journal of Pharmaceut

inding from the kinetic data. The DCC absorption kinetic method as modified to an equilibrium method where only the final extent

f adsorption to DCC in the presence and absence of plasma needs o be determined [20]. None of these methods account for the egradation of a compound within plasma and they are unsuit- ble for those compounds that are very unstable in plasma due to he time required for the DCC binding process to reach equilib- ium. Of the 3 reported methodologies the shortest time course, nd hence exposure of a compound to plasma, is approximately 0 min [18] and this will be unsuitable for compounds with plasma alf-lives of <30 min as significant decomposition would occur on his timescale. The original kinetic method [18] has been modi- ed to only consider the initial rate of DCC adsorption rather than nalysis of the full time course, and the chemical degradation pro- ess has further been incorporated into the kinetic modeling. This ethodology benefits greatly from experimental simplicity and

an be applied to compounds where the plasma half life is only few minutes. To validate this initial rates methodology the PPB easurements for compounds that were stable in plasma were

ompared to the measurements obtained from using a standard quilibrium dialysis methodology. This included the rat PPB of 3 ompounds and the human PPB of enalapril. The rat PPB of enalapril as then estimated using the initial rates methodology.

. Materials and methods

.1. Materials

Potassium dihydrogenphosphate, disodium hydrogenphos- hate, sodium chloride, formic acid, enalapril, verapamil, HPLC rade acetonitrile and DCC were purchased from Sigma-Aldrich ompany (Dorset, UK). Warfarin was purchased from Fisher Sci- ntific (Leicestershire, UK). Sildenafil was obtained from the straZeneca compound collection. Frozen human (pooled from donors), rat (Sprague–Dawley), dog (Beagle) and guinea pig

Dunkin–Hartley) plasmas were sampled and processed by the linical Pharmacology Unit and Animal Units at AstraZeneca R&D lderley Park (Cheshire, UK). Isotonic phosphate buffered saline

buffer) at pH 7.4 was prepared from potassium dihydrogen- hosphate (1.77 g), disodium hydrogenphosphate (7.67 g), sodium hloride (4.38 g), and water (1 l).

.2. Instrumentation

Centrifugations were carried out using a Heraeus Biofuge Fresco. ncubations were carried out in a Heraeus B15 incubator at 37 ◦C.

Dianorm® system with cells of 1 ml volume was used for quilibrium dialysis experiments, along with Diachema cellulose embranes with molecular weight cut off of 5 kDa (Dianorm, unich, Germany). All HPLC analyses were carried out using Waters 2777 auto-sampler, a Waters 2690 separations mod-

le and a Waters Quattro Ultima mass spectrometer using a elected ion recording quantitation method. Waters symmetry 8 5 �m × 3.9 mm × 20 mm columns were used along with a gra- ient of acetonitrile-aqueous (0.1%) formic acid (1:99, v/v) to cetonitrile-aqueous (0.1%) formic acid (99:1, v/v) at a flow rate f 2 ml/min over 5 min.

.3. PPB using equilibrium dialysis

To one compartment of each of the dialysis cells were added 1 ml

f plasma and 10 �l of a solution of the compound of interest at a oncentration of 2 mM in DMSO. The other compartment of each ialysis cell was filled with 1 ml buffer. The cells were then sealed, lamped to the Dianorm unit, and rotated in a water bath at 37 ◦C for 8 h. The dialysis cells were then emptied and the plasma and buffer

d Biomedical Analysis 55 (2011) 385–390

compartments solutions were treated in the following way such that the samples for HPLC/MS analysis were all present in an iden- tical matrix of 6-fold diluted plasma. 100 �l of the plasma solution from the dialysis cell was added to 500 �l buffer. 500 �l of the buffer solution from the dialysis cell was added to 100 �l blank plasma. Four standards covering a 100-fold range in concentration were prepared for each compound using the 2 mM DMSO stock solution and 6-fold diluted plasma. The 6-fold diluted plasma samples were then directly injected into the HPLC/MS system for analysis. The plasma and buffer compartment concentrations were interpolated from the 4 point calibration line derived from the standards. These interpolated concentrations were then multiplied by the necessary factors to account for the sample dilutions prior to analysis, finally giving the concentration in plasma compartment of the dialysis cell ([Drug]plasma cell) and concentration in the buffer compartment of the dialysis cell ([Drug]buffer cell). The percent bound was then calculated using Eq. (1), where the factor of 1.05 accounts for the small dilution of the plasma which takes place through the osmotic volume shift during the dialysis experiment [21].

% Bound

= 100× 1.05 × ([Drug]plasma cell − [Drug]buffer cell)

1.05×([Drug]plasma cell − [Drug]buffer cell) + [Drug]buffer cell (1)

2.4. Kinetics of degradation in plasma

The reactions were initiated by addition of a 2 mM solution of the compound of interest in DMSO (50 �l) to plasma of the relevant species (5 ml), with incubation at 37 ◦C. Aliquots of the solution (250 �l) were removed at timed intervals and added to acetonitrile (500 �l) and vortex mixed to quench the reaction and precipi- tate the plasma proteins. These solutions were then centrifuged at 11,000 × g for 5 min before quantitation of the supernatants by HPLC/MS. It was assumed that the degradation of the compound in plasma followed pseudo first order kinetics, and this process is described by Eq. (2)

−d[Drug]plasma dt

= k′[Drug]plasma (2)

where k′ is the pseudo first order rate constant. [Drug]plasma is the concentration of drug in plasma. k′ was then derived from the slope of a plot of ln(MS response) against time.

2.5. Kinetics for DCC adsorption

Fig. 1 shows the kinetic system in question. The drug undergoes reversible binding with plasma proteins and with DCC. The drug can also undergo irreversible chemical degradation in the plasma. If we first consider the situation where degradation does not occur then according to this scheme, the rate of loss of free drug concentration in the plasma, [Drug]free, is given by Eq. (3)

−d[Drug]free dt

= [Drug]free ∑

i

ki,on[P]i − ∑

i

ki,off[Drug]i,bound

+ k1[Drug]free[DCC]plasma − k−1[Drug]DCC (3)

where the summations are over all of the binding sites on each

of the proteins in the plasma, [P]i is the concentration of each of the protein binding sites, ki,on is the rate constant for binding of the drug to each of the binding sites, [Drug]i,bound is the concentration of bound drug at each of the binding sites, ki,off is the rate constant for dissociation of the drug from each of the binding sites, [DCC]plasma

M.C. Wenlock et al. / Journal of Pharmaceutical an

Fig. 1. Kinetic scheme for the adsorption of a plasma unstable compound to DCC i c o

i c a f b

g

w [ e a t s f b t fi t t m

w p (

�

n plasma. k1 is the rate constant for drug adsorption to DCC and k−1 is the rate onstant for drug desorption from DCC. k2 is the rate constant for the degradation f free drug in plasma.

s the concentration of DCC within the plasma, [Drug]DCC is the con- entration of the drug–DCC complex, k1 is the rate constant for drug dsorption to DCC, and k−1 is the rate constant for drug desorption rom DCC. Similarly, the rate of loss of plasma bound drug is given y Eq. (4)

−d[Drug]bound dt

= − ∑

i

d[Drug]i,bound dt

= ∑

i

ki,off[Drug]i,bound − [Drug]free ∑

i

ki,on[P]i (4)

The rate of change of total drug in plasma, [Drug]plasma, is then iven by the sum of Eqs. (3) and (4)

−d[Drug]plasma dt

= − (

d[Drug]bound dt

+ d[Drug]free dt

)

= k1 fu[Drug]plasma[DCC]plasma − k−1[Drug]DCC (5)

here fu is the free fraction of drug in plasma which is equal to Drug]free/[Drug]plasma. The [Drug]plasma used in all plasma kinetic xperiments is 20 �M (see experimental description in Sections 2.4 nd 2.6), and the main contribution to the PPB is most likely due o the binding to albumin (present at approximately 600 �M), con- equently fu is considered to be constant [22]. The substitution of u[Drug]plasma for [Drug]free in Eq. (5) assumes rapid PPB equili- ration which is reasonable considering typical binding kinetics o albumin [23]. If we now consider the situation where pseudo rst order kinetic degradation takes place of free drug from within he plasma then Eq. (5) needs an additional term. Eq. (6) contains he additional term k′[Drug]plasma (see Eq. (2)) which can be deter-

ined experimentally from plasma degradation experiments

−d[Drug]plasma dt

= k1 fu[Drug]plasma[DCC]plasma − k−1[Drug]DCC +k′[Drug]plasma (6)

With respect to the kinetic system displayed in Fig. 1, k′ = k2 fu here k2 is the rate constant for the degradation of free drug from lasma. If we now consider only the initial rate of the reaction

where [Drug]DCC = 0), we can derive Eq. (7)

plasma = − [

d[Drug]plasma dt

]t=0

= k1 fu[Drug]t=0plasma[DCC] t=0 plasma + k′[Drug]

t=0 plasma (7)

d Biomedical Analysis 55 (2011) 385–390 387

If we now consider the drug binding to DCC in buffer, rather than in plasma, then the kinetics will be described by Eq. (8)

− d[Drug]buffer dt

= k1[Drug]buffer[DCC]buffer (8)

Hence the initial rate will be given by Eq. (9)

�buffer = − [

d[Drug]buffer dt

]t=0 = k1[Drug]t=0buffer[DCC]

t=0 buffer (9)

Combination of Eqs. (7) and (9) then gives to Eq. (10)

�plasma �buffer

= k1 fu[Drug]

t=0 plasma[DCC]

t=0 plasma + k′[Drug]

t=0 plasma

k1[Drug] t=0 buffer[DCC]

t=0 buffer

(10)

If the same initial concentration of drug is used (see experi- mental description in Section 2.6), in all the kinetic studies (i.e. [Drug]plasma = [Drug]buffer), then Eq. (10) can be simplified to Eq. (11)

�plasma �buffer

= fu[DCC]t=0plasma

[DCC]t=0buffer + k

′

k1[DCC] t=0 buffer

(11)

A rearranged version of Eq. (9) (for k1), can be substituted into Eq. (11) and subsequent rearrangement will finally give an expression for fu, Eq. (12)

fu = �plasma − k′[Drug]t=0buffer

�buffer

[DCC]t=0buffer [DCC]t=0plasma

(12)

Hence fu can be determined from the rates of binding of the com- pound to DCC in the presence and absence of plasma, along with a determination of the rate of degradation in plasma. For compounds that are stable in plasma (i.e. k′ = 0), Eq. (12) reduces to the more simple form given by Eq. (13)

fu = �plasma �buffer

[DCC]t=0buffer [DCC]t=0plasma

(13)

2.6. DCC adsorption methodology

For each compound two initial rate experiments were carried out. One experiment contains a solution of the compound in buffer and DCC, and the other experiment contains the compound in plasma and DCC. Into 10 centrifuge tubes was placed buffer or plasma (1980 �l) along with DCC. The DCC concentrations that were employed ranged from 0.05 to 5 mg/ml and each tube was incubated at 37 ◦C and stirred magnetically. A further incubation, with the DCC excluded, was also prepared in order to generate the MS response of compound at zero time. The DCC concentration was dependent on the compound and if the experiment involves buffer or plasma.

The adsorption process was initiated by the sequential addition of a 2 mM solution of the compound of interest in DMSO (20 �l) at timed intervals. The time intervals were further apart at the begin- ning of the experiment than at the end, since the tubes were all centrifuged together at the end of the incubation time at 11,000 × g for 15 s to sediment the DCC. Hence the last few reactions to be initiated became the samples with the shortest incubation times. For a buffer experiment the supernatants were then quantified by HPLC/MS without further preparation. For a plasma experi- ment, supernatant (250 �l) was added to acetonitrile (500 �l) and vortexed in order to quench the reactions and precipitate the pro-

teins, followed by further centrifugation at 11,000 × g for 2 min. The final supernatant was then quantified by HPLC/MS. The incubations were followed for a maximum of 10 min. The data were then ana- lyzed by plotting concentration versus time and carrying out a least squares fit of the data to a quadratic equation using Microsoft Excel.

388 M.C. Wenlock et al. / Journal of Pharmaceutical and Biomedical Analysis 55 (2011) 385–390

Table 1 Kinetics of adsorption to DCC and equilibrium dialysis data on compounds stable in rat plasma.

Compound 10−8 �buffer (mol dm−3 s−1 )

10−8 �plasma (mol dm−3 s−1 )

% Bound by charcoal binding kinetic method

% Bound by equilibrium dialysis method

Sildenafil 2.64a ± 0.48 3.48b ± 0.72 86.8 ± 3.6 86.1 ± 3.4 Verapamil 1.49a ± x0.13 3.14b ± 0.34 78.9 ± 2.9 81.7 ± 4.5 Warfarin 18.2b ± 2.9 1.17c ± 0.34 99.4 ± 0.21 99.4 ± 0.20

Errors show standard deviation from 3 repeat measurements. I

T t o

3

s a p t T E r y d r p T b e n D a i c a d k i T e T p n t a i t y p a

T K

E I

nitial drug concentration = 20 �M in all experiments. a [DCC] = 0.05 mg/ml. b [DCC] = 0.5 mg/ml. c [DCC] = 5.0 mg/ml.

he initial rate of loss of compound was then derived by differen- iation of the derived quadratic equation in order to find the slope f the curve at t = 0.

. Results

Initial studies were carried out using 3 compounds that are table in plasma. The rates of adsorption of sildenafil, verapamil nd warfarin to DCC were determined in both buffer and rat lasma according to the method described in the experimental sec- ion. The observed rates, from triplicate experiments, are given in able 1 along with the percent bound to plasma calculated using q. (13). The binding of these 3 compounds to the same batch of at plasma was also determined using a standard equilibrium dial- sis approach. These data are also shown in Table 1, and values etermined from the two different approaches are consistent. The ate of degradation of enalapril was determined in human and rat lasma and the pseudo first order rate constants, k′, are given in able 2. k′ for enalapril degradation in rat plasma was found to e 4.9 × 10−4 s−1 which corresponds to a half life of 24 min. As xpected significant degradation was observed in rat plasma but ot in human plasma [16,17]. The rate of adsorption of enalapril to CC was then determined in buffer, rat plasma and human plasma, nd the observed rates are given in Table 2. Since enalapril is stable n human plasma, the binding to human plasma using DCC kinetics an be calculated using Eq. (13), and the result is given in Table 2 long with the value determined using conventional equilibrium ialysis. The short rat plasma half life of enalapril complicates the inetics of adsorption to DCC, since the observed loss of compound s due to both adsorption to DCC and plasma induced degradation. he modified kinetic model is therefore required for deriving the xtent of plasma binding from the kinetic data with use of Eq. (12). his method leads to a percent bound value of 50.3 ± 12.2 in rat lasma (Table 2). Clearly, the binding of enalapril to rat plasma was ot determined using equilibrium dialysis since the rate of degrada- ion is too fast compared to the timescale of a dialysis experiment, nd hence the experiment would not reach equilibrium. However,

n addition to its stability in human plasma, enalapril was also found o be stable in dog and guinea pig plasma, and equilibrium dial- sis was therefore used to determine the percent bound to dog lasma as 51.6 ± 9.4 and the percent bound to guinea pig plasma s 56.7 ± 7.3.

able 2 inetics of adsorption to DCC and equilibrium dialysis data for enalapril.

Species 10−4 k′ (s−1 ) 10−8 �buffer (mol dm−3 s−1 )

10−8 �p (mol dm

Human No reaction 6.8a ± 1.1 2.8a ± 0 Rat 4.9 ± 0.8 6.8a ± 1.1 4.4a ± 0

rrors show standard deviation from 3 repeat measurements. nitial drug concentration = 20 �M in all experiments.

a [DCC] = 0.25 mg/ml.

Fig. 2. Adsorption of verapamil to 0.05 mg/ml DCC in buffer (�), and to 0.5 mg/ml DCC in rat plasma (�).

4. Discussion

4.1. DCC concentration

In order to validate the use of initial rates kinetics of adsorption to DCC as a method of plasma binding determination, 3 plasma stable drugs were chosen for study. The 3 compounds (sildenafil, verapamil and warfarin) were selected since they exhibit a range of extent of plasma binding and charge type (predominantly neutral, positively charged, and negatively charged at pH 7.4, respectively). The rate of adsorption of these compounds to DCC in both buffer and rat plasma was determined, and some example data is shown for verapamil in Fig. 2. An important aspect of the DCC kinetic plasma binding method is the selection of suitable DCC concentrations in the buffer and plasma experiments. For more highly plasma bound compounds it is necessary to use a higher DCC concentration in plasma than in buffer. This is because the low free concentration of compound in plasma will lead to a very low rate of DCC adsorption, which will in turn lead to an imprecise rate measurement when using a short timescale experiment, or will necessitate a very long

timescale experiment for a precise rate to be determined. A suit- able rate for measurement on a convenient timescale can easily be achieved by using a higher DCC concentration in plasma than in buffer. Under the experimental conditions chosen, the rate of adsorption of verapamil to DCC is actually faster in plasma than in

lasma −3 s−1 )

% Bound by charcoal binding kinetic method

% Bound by equilibrium dialysis method

.4 59.6 ± 8.9 64.4 ± 7.6

.6 50.3 ± 12.2 Not determined

ical and Biomedical Analysis 55 (2011) 385–390 389

b f o i f (

4 f

d e l a e c f a t t

4 o

m T h o i d p e s r m a t t b c

i f t a p

a w b r t t T p t e a F r w i

M.C. Wenlock et al. / Journal of Pharmaceut

uffer (Fig. 2). This is because the concentration of DCC was 10- old higher in plasma than in buffer while the free concentration f verapamil in plasma is 5-fold lower than that in buffer (since it s ∼80% bound in rat plasma). This results in the approximately 2- old higher rate of adsorption to DCC in rat plasma than in buffer Table 1).

.2. Validation of the initial rates DCC adsorption methodology or plasma stable compounds

There is good correspondence between the plasma binding data etermined using DCC adsorption and the data determined using quilibrium dialysis for the 3 plasma stable compounds. Unpub- ished data on a range of proprietary compounds also shows good greement between the initial rates DCC adsorption method and quilibrium dialysis. Due to their plasma stability over the time ourse of the experiment these 3 compounds provide validation or Eq. (13). This should be expected since published methods have lready been validated which differ only in that they either utilize he whole time course of compound adsorption or they simply use he final equilibrium concentrations [16–18].

.3. Application of the modified method to determine the rat PPB f enalapril

The difference between Eq. (12) and Eq. (13) is that the for- er accounts for chemical degradation in plasma of a compound.

his modified equation only needs to be applied when the plasma alf life of the compound is rapid compared to the time course f the experiment. Extensive validation of this modified method s difficult since a traditional slow method such as equilibrium ialysis cannot be used to measure the plasma binding of a com- ound which exhibits fairly rapid degradation in plasma. However, nalapril is a compound that is useful for this problem since it is table in human plasma but undergoes rapid ester hydrolysis in at plasma. Therefore the binding to human plasma can be deter- ined using both DCC adsorption kinetics and equilibrium dialysis,

nd the binding to rat plasma can be determined using DCC adsorp- ion kinetics. Assuming that there is little interspecies difference in he plasma binding of enalapril, it is reasonable to expect similar inding to rat plasma, to that observed in human plasma where the ompound is stable.

Using the initial rates DCC adsorption method the percent bound n human plasma for enalapril was 59.6 ± 8.9. This result compares avorably with the equilibrium dialysis value of 64.4 ± 7.6, and with he previously published value of 50% [24], again confirming the pplicability of the initial rates DCC adsorption method to com- ounds that are stable in plasma.

The kinetics of DCC adsorption was determined in rat plasma nd in buffer, and some of these data are shown in Fig. 3 along ith the kinetics of degradation in rat plasma. Since low plasma

inding was expected, the same DCC concentration was used in at plasma and in buffer. Consequently it can be seen in Fig. 3 that he rate of DCC adsorption is slower in plasma than in buffer, and hat these rates are slower than the rate of degradation in plasma. hese three rates are then used to calculate the percent bound to rat lasma as 50.3 ± 12.2 (Table 2). It follows from the observation that here are only very small interspecies variations in the PPB value of nalapril in plasmas where it is chemically stable (i.e. human, dog

nd guinea pig), that this rat PPB value shows good correspondence. urthermore, the observation that the percent bound of enalapril to at plasma is slightly less than that to human plasma is consistent ith what would typically be expected for interspecies differences

n PPB [4].

Fig. 3. Degradation of enalapril in rat plasma (�). Adsorption of enalapril to 0.25 mg/ml DCC in buffer (�), and to 0.25 mg/ml DCC in rat plasma (�).

4.4. Further work and considerations

Even though enalapril has a rat plasma half life of 24 min the initial rates DCC adsorption method along with Eq. (12) enables the rat PPB to be determined. Furthermore, unpublished work has been conducted to determine the PPB of a wide range of propri- etary compounds with plasma half lives as short as 2 min, where convenient rates of DCC adsorption were achieved through the suit- able manipulation of the concentration of DCC used. An assumption that is made regarding the degradation of compound in plasma is that the observed plasma degradation follows pseudo first order kinetics but if this is not the case then Eq. (12) is not applicable.

This method could in principle be applied to compounds exhibit- ing even more rapid degradation through the use of dilute plasma along with the method of Wan et al. [10] for the extrapolation of the data back to that in undiluted plasma. However, for compounds that predominantly bind to �1-acid glycoprotein (AAG), which is present at approximately 10–30 �M in plasma, it is important that the AAG concentration in the diluted plasma is at least 10 times greater than the compound concentration [4]. The determination of binding to individual plasma proteins, i.e. serum albumin, may be an alternative strategy for estimating the PPB for a plasma unstable compound. However, preparations of serum albumin may still con- tain significant amounts of the enzymes, i.e. esterases, which could still cause a compound to significantly degrade over the experi- ment’s time course. With respect to compounds that are substrates from esterases, the use of inhibitors could also be considered to minimize any degradation [25]. Practically, it may be difficult to obtain enough data points to fit a quadratic curve to extract the rate constants from the DCC absorption kinetics using the described methodology. In such cases, the rate constant may be extracted by fitting a linear line but it must be acknowledged that the associated experimental error to the PPB measurement will probably increase. Interestingly, when using just the first two time points to extract the DCC absorption initial rates in buffer and plasma, the rat PPB measurement for enalapril is 50.4% which is almost identical to the value determined using the more formal methodology.

5. Conclusions

Enalapril displays rapid rat plasma instability that therefore prevents its PPB being determined using conventional methods.

A simple initial rates DCC adsorption method that accounts for enalapril’s plasma instability has been developed and can be car- ried out without the use of very specialized equipment. Experiences with proprietary compounds with very rapid plasma instability have shown this method to have wider applicability. Although more

3 ical an

c n d

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

90 M.C. Wenlock et al. / Journal of Pharmaceut

ompounds are needed to properly validate this methodology it onetheless should be of interest to groups looking for a method of etermining the PPB of plasma labile prodrugs and antedrugs.

eferences

[1] S. Schmidt, D. Gonzalez, H. Derendorf, Significance of protein binding in phar- macokinetics and pharmacodynamics, J. Pharm. Sci. 99 (2010) 1107–1122.

[2] L.M. Berezhkovskiy, On the influence of protein binding on pharmacological activity of drugs, J. Pharm. Sci. 99 (2010) 2153–2165.

[3] G.L. Trainor, The importance of plasma protein binding in drug discovery, Expert Opin. Drug Discov. 2 (2007) 51–64.

[4] R.E. Fessey, R.P. Austin, P. Barton, A.M. Davis, M.C. Wenlock, The role of plasma protein binding in drug discovery, in: B. Testa, S.D. Kramer, H. Wunderli- Allenspach, G. Folkers (Eds.), Pharmacokinetic Profiling in Drug Research: Biological, Physicochemical, and Computational Strategies, Wiley-VCH, Wein- heim, 2006, pp. 119–141.

[5] H.G. Weder, J. Schildknecht, P. Kesselring, A new equilibrium dialyzing system, Am. Lab. 3 (1971) 17–21.

[6] I. Kariv, H. Cao, K.R. Oldenburg, Development of a high throughput equilibrium dialysis method, J. Pharm. Sci. 90 (2001) 580–587.

[7] M.J. Banker, T.H. Clark, J.A. Williams, Development and validation of a 96-well equilibrium dialysis apparatus for measuring plasma protein binding, J. Pharm. Sci. 92 (2003) 967–974.

[8] N.J. Waters, R. Jones, G. Williams, B. Sohal, Validation of a rapid equilibrium dialysis approach for the measurement of plasma protein binding, J. Pharm. Sci. 97 (2008) 4586–4995.

[9] E.N. Fung, Y.H. Chen, Y.Y. Lau, Semi-automatic high-throughput determination of plasma protein binding using a 96-well plate filtrate assembly and fast liq- uid chromatography-tandem mass spectrometry, J. Chromatogr. B 795 (2003) 187–194.

10] H. Wan, M. Rehngren, High-throughput screening of protein binding by equilib-

rium dialysis combined with liquid chromatography and mass spectrometry, J. Chromatogr. A 1102 (2006) 125–134.

11] L. Di, E.H. Kerns, Solution stability – plasma, gastrointestinal, bioassay, Curr. Drug Metab. 9 (2008) 860–868.

12] H.J. Lee, Y. Zhengqing, K. Dong-Hoon, H.M. Mclean, Recent advances in prodrugs and antedrugs, Curr. Opin. Drug Discov. Dev. 1 (1998) 235–244.

[

[

d Biomedical Analysis 55 (2011) 385–390

13] P. Ettmayer, G.L. Amidon, B. Clement, B. Testa, Lessons learned from marketed and investigational prodrugs, J. Med. Chem. 47 (2004) 2393–2404.

14] M.O.F. Khan, K.K. Park, H.J. Lee, Antedrugs: an approach to safer drugs, Curr. Med. Chem. 12 (2005) 2227–2239.

15] P. Li, P.S. Callery, L.S. Gan, S.K. Balani, Esterase inhibition attribute of grape- fruit juice leading to a new drug interaction, Drug Metab. Dispos. 35 (2007) 1023–1031.

16] D.J. Tocco, F.A. Deluna, A.E.W. Duncan, T.C. Vassil, E.H. Ulm, The physiological disposition and metabolism of enalapril maleate in laboratory animals, Drug Metab. Dispos. 10 (1982) 15–19.

17] K.S. Pang, W.F. Cherry, J.A. Terrell, E.H. Ulm, Disposition of enalapril and its diacid metabolite, enalaprilat, in a perfused rat liver preparation, Drug Metab. Dispos. 12 (1984) 309–313.

18] J. Yuan, D.C. Yang, J. Birkmeier, J. Stolzenbach, Determination of protein bind- ing by in vitro charcoal adsorption, J. Pharmacokinet. Biopharm. 23 (1995) 41–55.

19] M. Khurana, J.K. Paliwal, V.P. Kamboj, R.C. Gupta, Binding of centchroman with human serum as determined by charcoal adsorption method, Int. J. Pharm. 192 (1999) 109–114.

20] L.M. Berezhkovskiy, Determination of drug binding to plasma proteins using competitive equilibrium binding to dextran-coated charcoal, J. Pharmacokinet. Pharmacodyn. 33 (2006) 595–608.

21] F.D. Boudinot, W.J. Jusko, Fluid shifts and other factors affecting plasma pro- tein binding of prednisolone by equilibrium dialysis, J. Pharm. Sci. 73 (1984) 774–780.

22] L.M. Berezhkovskiy, On the calculation of the concentration dependence of drug binding to plasma proteins with multiple binding sites of different affinities: determination of the possible variation of the unbound drug fraction and cal- culation of the number of binding sites of the protein, J. Pharm. Sci. 96 (2006) 249–257.

23] U. Kragh-HANSEN, S.O. Brennan, M. Galliano, O. Sugita, Binding of warfarin, salicylate, and diazepam to genetic variants of human serum albumin with known mutations, Mol. Pharmacol. 37 (1990) 238–242.

24] S. Vozeh, W. Taeschner, M. Wenk, Pharmacokinetic drug data, in: Clini- cal Pharmacokinetics Drug Data Handbook, ADIS Press, Auckland, NZ, 1990, pp. 1–38.

25] E.N. Fung, N. Zheng, M.E. Arnold, J. Zeng, Effective screening approach to select esterase inhibitors used for stabilizing ester-containing prodrugs analyzed by LC–MS/MS, Bioanalysis 2 (2010) 733–743.