DMD #39354dmd.aspetjournals.org/content/dmd/early/2011/05/24/... · 5/24/2011 · James A McLure, Donald J Birkett, David J Elliot, J Andrew Williams, Andrew Rowland and John O Miners

DMD #39354

1

Application of the fluorescent probe 1-anilinonaphthalene-8-sulfonate (ANS) to the

measurement of the non-specific binding of drugs to human liver microsomes.

James A McLure, Donald J Birkett, David J Elliot, J Andrew Williams, Andrew Rowland and

John O Miners

Department of Clinical Pharmacology, Flinders University School of Medicine, Adelaide,

Australia (JAM, DJB, DJE, AR, JOM)

Translational Oncology, Pfizer Global Research and Development, La Jolla, CA, USA

(JAW)

Current address: Metabolic Research Unit, School of Medicine, Deakin University, Geelong,

Australia (JAM)

DMD Fast Forward. Published on May 24, 2011 as doi:10.1124/dmd.111.039354

Copyright 2011 by the American Society for Pharmacology and Experimental Therapeutics.

This article has not been copyedited and formatted. The final version may differ from this version.DMD Fast Forward. Published on May 24, 2011 as DOI: 10.1124/dmd.111.039354

at ASPE

T Journals on A

pril 21, 2021dm

d.aspetjournals.orgD

ownloaded from

http://dmd.aspetjournals.org/

DMD #39354

2

Running title: ANS fluorescence as a marker of drug microsomal binding.

Address for correspondence: Professor JO Miners, Department of Clinical Pharmacology,

Flinders University School of Medicine, Flinders Medical Centre, Bedford Park, SA 5042,

Australia. Telephone: 61-8-82044131. Fax: 61-8-82045114. Email:

[email protected]

Number of text pages (excluding Tables): 24

Number of Tables: 7

Number of figures: 2

Number of references: 32

Abstract: 250 words

Introduction: 748 words

Discussion: 1129 words

ABBREVIATIONS

ANS, 1-anilinonaphthalene-8-sulfonate; fu(mic) , fraction of drug unbound in the incubation

mixture; HLM, human liver microsomes; NSB, non-specific binding; PB, phosphate buffer

(0.1M, pH 7.4).


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

3

ABSTRACT

The fluorescence of 1-anilinonaphthalene-8-sulfonate (ANS) in the presence of human liver

microsomes (HLM) is altered by drugs that bind non-specifically to the lipid bilayer. The

present study characterised the relationship between the non-specific binding (NSB) of drugs

to HLM as measured by equilibrium dialysis and the magnitude of the change in baseline

ANS fluorescence. Fraction unbound in incubations of HLM (fu(mic)) was determined for

sixteen drugs (12 bases, 3 acids and 1 neutral) with logP values in the range 0.1 to 6.7 at three

concentrations (100, 200 and 500 µM). Changes in ANS fluorescence induced by each of the

drugs in the presence of HLM were measured by spectrofluorometry. Values of fu(mic)

determined by equilibrium dialysis ranged from 0.08 to 1.0. Although NSB of the basic drugs

tended to increase with increasing logP, exceptions occurred. Basic drugs generally caused an

increase in ANS fluorescence, while the acidic and neutral drugs resulted in a decrease in

ANS fluorescence. There were highly significant (p < 0.001) linear relationships between the

modulus (absolute value) of the increment/decrement in ANS fluorescence and both fu(mic) (r

= 0.90 to 0.96) and log(1-fu(mic)/ fu(mic)) (r = 0.85 to 0.92) at the three drug concentrations.

Agreement between measured fu(mic) and that predicted by ANS fluorescence was very good

(< 10% variance) for a validation set of six compounds. The ANS fluorescence method

provides an accurate measure of the NSB of drugs to HLM. Physicochemical determinants

other than log P and charge type influence the NSB of drugs to HLM.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

4

The application of in vitro approaches to predict drug disposition parameters in humans has

found widespread acceptance over the last two decades (Houston, 1994; Ito et al., 1998;

Obach, 1999; Riley et al., 2005; Hosea et al., 2009; Lave et al., 2009). In particular, in vitro –

in vivo extrapolation has been employed to predict clearance and drug-drug interaction

potential for hepatically cleared compounds (Riley et al., 2005; Miners et al., 2006 and 2010;

Lave et al., 2009). Most commonly, in vitro intrinsic clearance for a pathway (CLint),

typically determined using human liver microsomes (HLM) or hepatocytes as the enzyme

source, is extrapolated to hepatic clearance and extraction ratio using appropriate scaling and

physiological modelling procedures. CLint may be calculated from measurement of product

formation under initial rate conditions (as Vmax/Km), or from the rate of substrate depletion.

Similarly, extrapolation of an inhibitor constant (Ki) determined in vitro allows prediction of

the magnitude of an inhibitory drug-drug interaction (Ito et al., 1998 and 2004; Miners et al.,

2010).

However, hepatic clearance and drug-drug interaction potential are frequently under-

predicted (Ito and Houston, 2005; Miners et al., 2006). Numerous methodological issues

contribute to the accuracy of predicted disposition parameters (Hallifax and Houston, 2009;

Miners et al., 2010). Notably, non-specific binding (NSB) of a drug to the in vitro enzyme

source, especially HLM, decreases the availability of unbound drug in the incubation medium

(fu(mic)) for metabolism (Obach, 1997; McLure et al., 2000; Venkatakrishnan et al., 2000;

Austin et al., 2002; Margolis and Obach, 2003). Thus, Km and Ki values are over-estimated,

resulting in under-estimation of predicted hepatic clearance and magnitude of an inhibitory

drug-drug interaction. Furthermore, complex effects on reaction kinetics may be observed

when substrate concentrations are in the saturable binding range (McLure et al., 2000).


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

5

Measurement of fu(mic and correction for binding is therefore essential for the calculation of in

vitro kinetic parameters. Equilibrium dialysis and ultra-filtration are the most commonly

employed experimental techniques for the determination of fu(mic) and hence measurement of

this parameter can be onerous, especially early in pre-clinical drug development (Grime and

Riley, 2006). Algorithms based on published or in-house databases have been reported

(Austin et al., 2002; Hallifax and Houston, 2006; Sykes et al., 2006; Gao et al., 2008; Kilford

et al., 2008; Li et al., 2009), although most consider only charge state and lipophilicity as

variables.

Fluorescent probes provide an alternative approach to the measurement of drug binding to

HLM and other membranes. 1-Anilinonaphthalene-8-sulfonate (ANS) binds non-covalently

to the polar head groups of phosphatidylcholine present in membranes with negligible

disruption of bilayer structure (Slavik, 1982). Importantly, ANS fluorescence is inversely

proportional to the negative surface potential of the membrane (Slavik, 1982). Thus, the

binding of cations to phosphate and carboxyl groups, which decreases surface electrostatic

potential, facilitates ANS binding and fluorescence (Slavik, 1982). Consistent with the known

effects of membrane charge on ANS fluorescence, it has been demonstrated that several

drugs that are positively charged at physiological pH enhance the fluorescence of microsome-

bound ANS, whereas negatively charged drugs decrease the fluorescence of ANS in the

presence of microsomes (Diaugustine et al., 1970; Hawkins and Freedman, 1973; Birkett,

1974). These data suggest that drug induced shifts in baseline ANS fluorescence may provide

a measure of the NSB of drugs to HLM.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

6

Drug binding to the membranes appears to be primarily dependent on charge state and

lipophilicity (measured as log P or log D), as might be predicted from passive partitioning

into the phospholipid bilayer (Herbette et al., 1983; Austin et al., 2005; Nussio et al., 2007).

Evidence to date indicates that NSB is greatest for lipophilic organic bases (McLure et al.,

2000; Austin et al., 2002; Sykes et al., 2006), which is thought to arise from favorable

electrostatic interactions with negatively charged phosphate head groups while orienting

lipophilic groups along the hydrocarbon chains of the bilayer. In contrast, neutral and acidic

compounds bind less extensively or not at all. As noted above, most algorithms for the

prediction of fu(inc) therefore consider only log P and charge state at pH 7.4 as variables,

although more recent analyses suggest a role for electrostatic features (Li et al., 2009).

Here, we investigated the relationship between the binding of drugs to HLM as measured by

equilibrium dialysis (fu(mic)) and the magnitude of the change induced in baseline ANS

fluorescence. The results demonstrate that the ANS fluorescence technique provides a simple

and convenient method for estimating fu(mic) and additionally provide further insights into the

physicochemical features of drugs that contribute to membrane binding.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

7

MATERIALS AND METHODS

Materials

ANS, bupropion hydrochloride, chloroquine diphosphate, chlorpromazine hydrochloride,

imipramine hydrochloride, lignocaine hydrochloride monohydrate, propranolol

hydrochloride, and triflupromazine hydrochloride were purchased from Sigma Aldrich

(Sydney, Australia). Other drugs were kindly provided by the following sources; atenolol,

bupivacaine hydrochloride, and ropivacaine hydrochloride from AstraZeneca (Sydney,

Australia); diazepam from Hoffmann La Roche Pharmaceuticals (Basle, Switzerland);

diflunisal from Merck Sharp and Dohme Pty Ltd (Sydney, Australia); flufenamic acid and

meclofenamic acid from Parke, Davis & Co (Sydney, Australia); mianserine hydrochloride

from Organon Pty Ltd (Sydney, Australia); and verapamil hydrochloride from Knoll

Australia Pty Ltd (Sydney, Australia).

Human liver microsomes

NSB experiments were performed with pooled microsomes from two human livers obtained

from the human liver bank of the Department of Clinical Pharmacology of Flinders Medical

Centre. Donors of the respective livers were a 49 year old female and a 22 year old male.

Neither were cigarette smokers, however both received dopamine prior to death. Approval

was obtained from the Flinders Clinical Research Ethics Committee for the use of human

liver tissue in drug disposition studies in vitro. HLM were prepared according to the general

procedure of Bowalgaha et al. (2005).

Equilibrium dialysis

Direct measurement of the binding of drugs to HLM was performed with an equilibrium

dialysis apparatus (Dianorm, Munich, Germany) containing teflon dialysis cells of 1.2 ml


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

8

capacity per side, using a working volume of 1 ml. Spectrapor #4 dialysis membrane

(molecular mass cut off 12,000 –14,000 Da) was purchased from Spectrum Medical

Industries Inc. (Los Angeles, CA, USA) and prepared by soaking overnight in phosphate

buffer (PB; 0.1M, pH 7.4) at 4°C prior to dialysis experiments. Each of the drugs was diluted

1:100 upon addition to the dialysis cell to give the concentration required. PB (0.99 ml; 0.1M,

pH 7.4) and a 0.01 ml aliquot of the test drug were loaded into one side of each dialysis cell.

A suspension of HLM in PB (1 ml: protein concentration 1 mg/ml) was added to the other

side of the dialysis cell. The dialysis cell assembly was immersed in a water bath maintained

at 37°C and rotated at 12 rpm for 3 hr. After dialysis each side of the cell was unloaded by

expelling the contents into a 5 ml glass tube. A negative control and HLM-HLM and PB-PB

controls were employed in each dialysis experiment. The negative control comprised PB (1

ml; 0.1M, pH 7.4) without drug on one side of the dialysis cell and a suspension of HLM in

PB (1 ml; 1 mg/ml protein concentration) on the other side. HLM-HLM controls consisted of

a suspension of HLM (0.99 ml: protein concentration 1 mg/ml) and a 0.01 ml aliquot of the

test drug at a "mid" concentration of the range employed in binding experiments with a

suspension of HLM (1 ml: protein concentration 1 mg/ml) in the other side of the dialysis

cell. PB was substituted for HLM in the PB-PB control samples.

Sample preparation and HPLC conditions

Drug concentrations in the buffer and microsome compartments of the dialysis cell were

measured by HPLC. Prior to chromatography, all samples were treated with acetonitrile or

mixtures containing acetonitrile (plus internal standard, with the exception of imipramine) to

precipitate microsomal protein. After dilution, samples were vortex mixed for 10 sec and then

centrifuged at 2000 g for 10 min at 4°C. An aliquot of the supernatant fraction was injected

onto the HPLC column. The sample preparation procedure for each drug is summarised in

Table 1.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

9

HPLC assays were developed and validated for each drug investigated in dialysis

experiments. The mobile phase, column type, detector wavelength, and retention time of the

analyte and the internal standard are shown in Table 2. The mobile phase flow rate was 1

ml/min for all analytes, except flufenamic acid (1.5 ml/min). Unknown concentrations were

determined by reference of peak height ratios (drug to internal standard) to those of a

standard curve that spanned the concentration range of the drug recovered from the separate

buffer and HLM compartments of the dialysis cell. All standard curves were linear with r2

values ≥ 0.990.

Calculation of fu(mic) and logP

The binding of drugs to HLM is expressed as the unbound fraction of drug in the microsomal

suspension (i.e. fu(mic)), giving a value between 0 and 1. Thus, fu(mic) was calculated as the

proportion of free drug (i.e. concentration in the buffer compartment) to free plus bound drug

(i.e. concentration in the microsome compartment) according to equation 1:

fu(mic) = total

buffer

C

C =

boundfree

free

CC

C

+ (1)

Values of fu(mic) determined in the presence of either 0.25mg/ml or 1mg/ml HLM were inter-

converted according to equation 2 (Austin et al., 2002):

(2)

where C1 and C2 are the two concentrations of HLM, and fu(mic)1 and fu(mic)2 are the unbound

fractions at the two HLM concentrations. LogP values were calculated computationally

using SciFinder Solaris (Advanced Chemistry Software, V4.64).

ANS fluorescence assay


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

10

Aliquots (10 μL) of separate 20 mM stock solutions of each test drug were added to 2 ml of

PB containing ANS (10 μM) and HLM (0.25 mg/ml) in a 4.5 ml glass cuvette to produce

drug concentrations in the range 100 – 1000 μM. Drugs were dissolved in water, except for

diazepam (methanol) and itraconazole (DMSO). Neither methanol nor DMSO, at a

concentration of 0.5% v/v, affected ANS fluorescence. ANS fluorescence, recorded prior to

and after addition of the drug, was measured using a Perkin Elmer 3000 Fluorescence

Spectrometer. The ANS excitation and emission wavelengths were 375 nm and 470 nm,

respectively (Molecular Probes, Leiden, Netherlands). All measurements were performed at

least in duplicate. The occurrence of quenching was excluded for all drugs. Percentage

fluorescence increment or decrement in ANS fluorescence due to added drug was calculated

according to Birkett (1974):

Percent ANS fluorescence increment/decrement

= ]100 x matrix in ANS of cefluorescen initial

matrix in cefluorescen drugmatrix and drug of presence in ANS of cefluorescen [

− - 100 (3)


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

11

RESULTS

Equilibrium dialysis

Table 3 shows logP, pKa and microsomal non-specific binding data, determined using

equilibrium dialysis, for the compounds studied here. Non-specific binding was determined at

3 drug concentrations (100, 200 and 500µM) and results are presented as fu(mic). The

lipophilic acidic drugs diflunisal, flufenamic acid, and meclofenamic acid all bound to HLM,

and fu(mic) values tended to decrease with increasing logP. Atenolol, the most hydrophilic

base investigated, did not bind to HLM. Similarly, neither lignocaine nor the structurally

related amide local anaesthetic agents bupivacaine and ropivacaine, which exhibit logP

values in the range 2.4 to 3.6, bound to HLM. Bupropion was also found not to bind to HLM.

Chloroquine and verapamil bound moderately (fu(mic) ≥ 0.70), while chlorpromazine,

imipramine, mianserine, propranolol, and triflupromazine bound more extensively (fu(mic) ≤

0.60). The neutral drug diazepam bound moderately (fu(mic) 0.76 – 0.77) to HLM.

ANS fluorescence

The percent ANS fluorescence increment/decrement was used as the measure of ANS

fluorescence as this parameter represents the change from baseline (microsomes plus ANS)

due to the effect of added drug. The background fluorescence of drug in the experimental

matrix was accounted for in the calculation of percent ANS fluorescence

increment/decrement. None of the drugs investigated exhibited quenching.

Three basic drugs, imipramine, lignocaine, and propranolol, were selected initially to

characterise effects on ANS fluorescence in the presence of HLM. Imipramine and

propranolol caused concentration dependent increases in ANS fluorescence over the

concentration range tested and the effects on ANS fluorescence were saturable (Table 4 and


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

12

Figure 1). Consistent with the equilibrium binding data, ANS fluorescence was essentially

unaltered by lignocaine (Table 4).

Other drugs characterised for non-specific binding to HLM using equilibrium dialysis were

subsequently investigated using the ANS fluorescence technique at concentrations of 100,

200 and 500 µM. Results are shown in Table 5. The acidic drugs (diflunisal, flufenamic acid,

and meclofenamic acid) and the neutral drug diazepam caused a concentration-dependent

decrease in ANS fluorescence in the presence of HLM. With the exception of the anomalous

result for bupropion, basic compounds were without effect (atenolol, bupivacaine, lignocaine

and ropivacaine) or caused an increase (chloroquine, chlorpromazine, imipramine,

mianserine, propranolol, triflupromazine, and verapamil) in ANS fluorescence in the

presence of HLM.

Equilibrium dialysis experiments were performed at a microsomal protein concentration of 1

mg/ml whereas the maximum HLM protein concentration that could be used in ANS

fluorescence experiments, due to matrix turbidity, was 0.25 mg/ml. Thus, fu(mic) values

determined using equilibrium dialysis were converted to a HLM protein concentration of 0.25

mg/ml using an expression derived and validated by Austin et al. (2002)(see Methods,

Equation 2). Derived fu(mic) values were plotted against the modulus (i.e. absolute value

without regard to sign) of the percent ANS fluorescence increment/decrement (Figure 2).

Data were also plotted as log(1-fu(mic)/ fu(mic)) versus the modulus of percent ANS

fluorescence increment/decrement (Figure 3). Linear and highly significant relationships (p <

0.001: Figure 2 and Figure 3) were observed between the modulus of percent ANS

fluorescence increment/decrement and fu(mic) (y = -149x + 152; r = 0.93) and log{(1-fu(mic))/


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

13

fu(mic)}(y = 47x + 72; r = 0.85). Highly significant relationships (p < 0.001) were also evident

at higher drug concentrations. Expressions for the relationships between the modulus of

percent ANS fluorescence increment/decrement and fu(mic) were y = -258x + 268 (r = 0.96)

and y = -604x + 621 (r = 0.90) at drug concentrations of 200 µM and 500 μM , respectively.

For the relationship between the modulus of percent ANS fluorescence increment/decrement

and log{(1-fu(mic))/ fu(mic)}, expressions were y= 81x + 128 (r = 0.94) at 200 μM, and y= 170

+ 273 (r = 0.85) at 500 μM. Since the correlations shown in Figure 2 are for a HLM

concentration of 0.25 mg/ml, values of fu(mic) are lower than at the usual HLM concentration

employed in NSB studies (viz. 1 mg/ml). As shown in Table 3, the fu(mic) range at an HLM

concentration of 1 mg/ml is 0.08 to 1.

For validation, the ANS fluorescence increment/decrement was measured for 5 compounds

characterised previously using equilibrium dialysis in this laboratory (viz. amitriptyline,

nortriptyline, phenytoin, S-naproxen, and lamotrigine)(McLure et al., 2000; Rowland et al.,

2006), and for itraconazole (see Tables 1 and 2 for equilibrium dialysis sample work-up and

HPLC conditions for itraconazole). ANS fluorescence data for these compounds are shown

Table 6, and predicted and reported fu(mic) values from equilibrium dialysis in Table 7.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

14

DISCUSSION

Consistent with previous reports (Diaugustine et al., 1970; Hawkins and Freedman, 1973;

Birkett, 1974), ANS fluorescence was enhanced by basic drugs that bound to HLM while

acidic drugs and the neutral compound diazepam caused a decrease in ANS fluorescence.

Importantly, there was a highly significant linear relationship between fu(mic) and the modulus

of the fluorescence increment or decrement for the combined dataset at each added drug

concentration. The relationship between the change in ANS fluorescence and log{(1-

fu(mic))/fu(mic)}, which is similar to an equilibrium constant and is considered a more

appropriate expression when dealing with linear free energy relationships (Austin et al.,

2002), was also linear and highly statistically significant. Taken together, these data indicate

that changes in ANS fluorescence may be used to estimate the NSB of drugs to HLM.

The basic drugs shown here to bind to HLM using equilibrium dialysis (chloroquine,

chlorpromazine, imipramine, mianserine, propranolol, triflupromazine, and verapamil) all

increased ANS fluorescence. Consistent with NSB data generated from equilibrium dialysis

(McLure et al., 2000), the increase in ANS fluorescence observed with imipramine and

propranolol was saturable over the concentration range 100 - 1000 µM (Figure 1). In

contrast, no change from baseline fluorescence was observed for those basic drugs (atenolol,

bupivacaine, lignocaine and ropivacaine) shown not to bind to HLM using equilibrium

dialysis. Conversely, the three acidic drugs diflunisal, flufenamic acid and meclofenamic

acid, along with the neutral diazepam, decreased ANS fluorescence compared to baseline.

The only exception to this trend was observed with bupropion. This non-binding basic

compound caused a decrease in ANS fluorescence. The reason for this anomalous result is

unclear, but may be due to the presence of an acidic side-chain carbon atom located between


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

15

electron-withdrawing nitrogen and carbonyl functions, which may cause bupropion to act

more as an acid than a base with respect to NSB.

ANS fluorescence data for amitriptyline, itraconazole, lamotrigine, nortriptyline, phenytoin

and S-naproxen were generated for validation of the procedure. Amitriptyline, itraconazole

and nortriptyline are bases which exhibit moderate to high binding to HLM, while

lamotrigine, phenytoin and S-naproxen are acids that bind negligibly or weakly to HLM. The

NSB of amitriptyline, lamotrigine, nortripyline, phenytoin and S-naproxen to HLM have been

characterised previously in this laboratory by equilibrium dialysis (McLure et al., 2000;

Rowland et al., 2006) and additional data were generated here for itraconazole. ANS

fluorescence was enhanced by amitriptyline, itraconazole and nortriptyline, whereas addition

of lamotrigine, phenytoin or S-naproxen to a suspension of HLM was without effect or

caused a small decrease in ANS fluorescence. Importantly, there was good agreement

between the fu(mic) values predicted for each drug from the ANS fluorescence

increment/decrement and that determined directly by equilibrium dialysis (Table 7).

The NSB binding data presented here are generally consistent with the principle that

extensive NSB is associated with lipophilic organic bases. However, some observations

warrant comment. Notably, the NSB of chloroquine (log P 4.9) was low (fu(mic) ≥ 0.78)

across the concentration range investigated. The binding of mianserine (logP = 3.4) exceeded

that of the more lipophilic bases chloroquine and imipramine (logP = 4.8). By contrast,

bupivacaine and bupropion, which have similar log P values to mianserine (viz. 3.5 and 3.6,

respectively) did not bind to HLM. Two other local anaesthetics structurally related to

bupivacaine, lignocaine (log P 2.4) and ropivacaine (log p 3.1), also exhibited negligible


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

16

NSB. Although most acidic drugs investigated to date bind minimally to HLM (Obach, 1997;

McLure et al., 2000; Austin et al., 2002; Li et al., 2009), 30% to 40% binding was observed

for flufenamic acid and meclofenamic acid. Together, these data indicate that factors other

than charge state and log P influence the degree of NSB. In this regard, recent computational

modeling studies have implicated properties such as polar surface area as determinants of

NSB, although lipophilicity was the most important descriptor (Gao et al., 2008; Li et al.,

2009). Compounds containing a halogen atom(s) (e.g. flufenamic acid, meclofenamic acid,

chlorpromazine, itraconazole, triflupromazine) also appear to be more likely to exhibit

significant NSB. Other factors that potentially affect the extent of NSB include molecular

mass (organic bases with molecular masses > 200 Da appear more likely to exhibit

significant NSB compared to compounds with lower molecular masses), charge distribution,

and the number of hydrogen bond acceptors/donors (JO Miners and JA McLure, unpublished

data). However, exceptions occur. For example, itraconazole, a compound shown here to

bind extensively to HLM, has 14 hetero-atoms (although two of these are chlorines; see

above discussion).

The NSB of chlorpromazine, diazepam, imipramine, propranolol, and verapamil has also

been reported by Austin et al. (2002). Values of fu(mic) determined here and reported by

Austin et al. were similar for chlorpromazine (fu(mic) = 0.08 vs. 0.11), diazepam (fu(mic) = 0.76

vs. 0.66) and propranolol (fu(mic) = 0.48 vs. 0.48). However, there was a 2- and 3-fold

difference in the fu(mic) values for verapamil (fu(mic) = 0.70 vs.0.37) and imipramine (fu(mic) =

0.48 vs. 0.16). The previous study used rat liver microsomes (1 mg/ml) and a drug

concentration of 1 μM, whereas the current study used HLM (1 mg/ml) and drug

concentrations in the range 100 to 500 μM. The lower NSB sometimes observed at high drug

concentrations may contribute to the observed differences.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

17

Limitations of the ANS fluorescence method for the determination of NSB are

acknowledged. As noted previously, the highest concentration of microsomes that can be

used is 0.25 mg/ml due to the effects of matrix turbidity on ANS fluorescence. Changes in

ANS fluorescence are generally not detectable for added drug concentrations below 25 to 50

μM and the method has not been investigated for HLM protein concentrations < 0.25 mg/ml..

Whether sensitivity can be improved with fluorescent probes other than ANS, of which many

are commercially available, is currently under investigation. It is also acknowledged that data

presented here report NSB for microsomes from only two livers. However, experience in this

laboratory demonstrates that fu(mic) values differ to a very minor extent (< 5%) for

microsomes prepared from the livers of multiple donors (JO Miners, unpublished data).

Indeed, lack of an appreciable difference in drug binding to hepatic microsomes across a

range of mammalian species (human, monkey, dog and rat) has recently been reported

(Zhang et al., 2010).

In summary, this study reports ‘proof-of-principle’ evidence that the fluorescent probe ANS

provides an accurate, simple and potentially high throughput method for measuring the NSB

of drugs and other compounds to HLM. In particular, there were highly significant linear

relationships between the modulus of the increment/decrement in ANS fluorescence and both

fu(mic) and log(1-fu(mic)/ fu(mic)), and agreement between measured fu(mic) and that predicted by

ANS fluorescence was very good for a validation set of six compounds. The microsomal

binding data reported here further demonstrate that physicochemical properties other than

charge and lipophilicity (as logP) influence NSB.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

18

AUTHORSHIP CONTRIBUTIONS

Participated in research design: Birkett, Miners, McLure, Williams

Conducted experiments: McLure, Elliot, Rowland

Performed data analysis: McLure, Miners

Wrote or contributed to the writing of the manuscript: Miners, McLure, Birkett, and Williams


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

19

REFERENCES

Austin RP, Barton P, Cockroft SL, Wenlock MC and Riley RJ (2002) The influence of

nonspecific microsomal binding on apparent intrinsic clearance, and its prediction

from physicochemical properties. Drug Metab Dispos 30:1497-1503.

Austin RP, Barton P, Mohmed S and Riley RJ (2005) The binding of drugs to hepatocytes

and its relationship to physicochemical properties. Drug Meta Dispos 33:419 - 425.

Birkett DJ (1974) Interaction of some drugs, metal ions and alcohols with rat liver

microsomes as studied with a fluorescent probe. Clin Exp Pharmacol Physiol 1:415-

427.

Bowalgaha K, Elliot DJ, Mackenzie PI, Knights KM, Swedmark S and Miners JO (2005) S-

Naproxen and desmethylnaproxen glucuronidation by human liver microsomes and

recombinant human UDP-glucuronosyltransferases (UGT): role of UGT2B7 in the

elimination of naproxen. Br J Clin Pharmacol 60:423-433.

Diaugustine RP, Eling TE and Fouts JR (1970) The interaction of a fluorescent probe with rat

hepatic microsomes. Chem-Biol Int 2:17-27.

Gao H, Yao L, Mathieu HW, Zhang Y, Maurer TS, Troutman MD, Scott DO, Ruggeri RB

and Lin J (2008) In Silico Modeling of Nonspecific Binding to Human Liver

Microsomes. Drug Metab Dispos 36:2130-2135.

Grime K and Riley RJ (2006) The impact of in vitro binding on in vitro - in vivo

extrapolations, projections of metabolic clearance and clinical drug-drug interactions.

Curr Drug Metab 7:251-264.

Hallifax D and Houston JB (2006) Binding of drugs to hepatic microsomes: comment and

assessment of current prediction methodology with recommendation for

improvement. Drug Metab Dispos 34:724-726.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

20

Hallifax D and Houston JB (2009) Methodological uncertainty in quantitative prediction of

human hepatic clearance from in vitro experimental systems. Curr Drug Metab

10:307-321.

Hawkins HC and Freedman RB (1973) Fluorescence studies of drug and cation interactions

with microsomal membranes. FEBS Lett 31:301-307.

Herbette L, Katz AM and Sturtevant JM (1983) Comparisons of the interaction of

propranolol and timolol with model and biological systems. Mol Pharmacol 24:259-

269.

Hosea NA, Collard WT, Cole S, Maurer TS, Fang RX, Jones H, Kakar SM, Nakai Y, Smith

BJ, Webster R and Beaumont K (2009) Prediction of human pharmacolinetics from

preclinical information: comparative accuracy of quantitative prediction approaches. J

Clin Pharmacol 49:513-533.

Houston JB (1994) Utility of in vitro drug metabolism data in predicting in vivo metabolic

clearance. Biochem Pharmacol 47:1469-1479.

Ito K, Brown HS and Houston JB (2004) Database analyses for the prediction of in vivo drug-

drug interactions from in vitro data. Br J Clin Pharmacol 57:473-486.

Ito K and Houston JB (2005) Prediction of human drug clearance from in vitro and

preclinical data using physiologically based and empirical approaches. Pham Res

22:103-112.

Ito K, Iwatsubo T, Kanamitsu S, Ueda K, Suzuki H and Sugiyama Y (1998) Prediction of

pharmacokinetic alterations caused by drug-drug interactions: metabolic interaction in

the liver. Pharmacol Rev 50:387-411.

Kilford PJ, Gertz M, Houston JB and Galetin A (2008) Hepatocellular binding of drugs:

correction for unbound fraction in hepatocyte incubations using microsomal binding

or drug lipophilicity data. Drug Metab Disp 36:1194-1197.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

21

Lave T, Chapman K, Goldsmith P and Rowland M (2009) Human clearance prediction:

shifting the paradigm. Exp Opin Drug Metab Toxicol 5:1039-1048.

Li H, Sun J, Sui X, Yan Z, Sun Y, Liu X and Wang YH, Z. (2009) Structure-based prediction

of the nonspecific binding of drugs to hepatic microsomes. AAPS J 11:364-370.

Margolis JM and Obach RS (2003) Impact of nonspecific miicrosomal binding to

microsomes and phospholipid on the inhibition of of cytochrome P4502D6:

Implications for relating in vitro inhibition data to in vivo drug interactions. Drug

Metab Disp 31:606-611.

McLure JA, Miners JO and Birkett DJ (2000) Nonspecific binding of drugs to human liver

microsomes. Br J Clin Pharmacol 49:453-461.

Miners JO, Knights KM, Houston JB and Mackenzie PI (2006) In vitro - in vivo correlation

for drugs and other compounds eliminated by glucuronidation in humans: pitfalls and

promises. Biochem Pharmacol 71:1531-1539.

Miners JO, Mackenzie PI and Knights KM (2010) The prediction of drug-glucuronidation

parameters in humans: UDP-glucuronosyltransferase enzyme-selective substrate and

inhibitor probes for reaction phenotyping and in vitro-in vivo extrapolation of drug

clearance and drug-drug interaction potential. Drug Metab Rev 42:189-201.

Nussio MR, Sykes MJ, Miners JO and Shapter JG (2007) Characterisation of the binding of

cationic amphiphilic drugs to phospholipid bilayers using surface plasmon resonance.

ChemMedChem 2:366-373.

Obach RS (1997) Nonspecific binding to microsomes: impact on scale-up of in vitro intrinsic

clearance to hepatic clearance as assessed through examination of warfarin,

imipramine, and propranolol. Drug Metab Disp 25:1359-1369.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

22

Obach RS (1999) Prediction of human clearance of twenty-nine drugs from hepatic

microsomal intrinsic clearance data: an examination of in vitro half-life approach and

nonspecific binding to microsomes. Drug Metab Dispos 27:1350-1359.

Riley RJ, McGinnity DF and Austin RP (2005) A unified model for predicting human

hepatic, metabolic clearance from in vitro intrinsic clearance data in hepatocytes and

microsomes. Drug Metab Dispos 33:1304 - 1311.

Rowland A, Elliot DJ, Williams JA, Mackenzie PI, Dickinson RG and Miners JO (2006) In

vitro characterisation of lamotrigine N2 glucuronidation and the lamotrigine-valproic

acid interaction. Drug Metab Disp 34:1055-1062.

Slavik J (1982) Anilinonapthalene sulfonate as a probe of membrane composition and

function. Biochim Biophys Acta 694:1-25.

Sykes MJ, Sorich MJ and Miners JO (2006) Molecular modeling approaches for the

prediction of the nonspecific binding of drugs to hepatic microsomes. J Chem Inform

Model 46:2661-2673.

Venkatakrishnan K, von Moltke LL, Obach RS and Greenblatt DJ (2000) Microsomal

binding of amitriptyline: effect on estimation of enzyme kinetic parameters in vitro. J

Pharmacol Exp Ther 293:343-350.

Zhang Y, Yao L, Gao H, Wilson TC and Giragossian C (2010) lack of appreciable species

differences in nonspecific microsomal binding. J Pharm Sci 99:3620-3627.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

23

FOOTNOTE

Project grant support from the National Health and Medical Research Council of Australia

and Pfizer Global Research is gratefully acknowledged.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

24

LEGENDS FOR FIGURES

Figure 1: Relationships between percent ANS fluorescence increment and increasing

imipramine (A), and propranolol (B) concentration in the presence of HLM (0.25 mg/ml)

Figure 2: Relationships between: (A) fu(mic) and the modulus of percent ANS fluorescence

increment/decrement, and (B) log(1-fu(mic)/ fu(mic)) and the modulus of percent ANS

fluorescence increment/decrement. fu(mic) values were normalised to a HLM concentration of

0.25 mg/ml using equation 2 (Materials and Methods).


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

25

Table 1: Sample preparation procedures for equilibrium dialysis samples

Drug Dilutions of dialysis samples and controls

Atenolol 1:10 with 60/40 acetonitrile - water then 1:1 with PB

Bupivacaine 1:10 with mobile phase (Table 2)

Bupropion 2:3 with acetonitrile

Chloroquine 1:10 with 72/28 acetonitrile - water then 1:1 with PB

Chlorpromazine 2:3 with acetonitrile

Diazepam 1:10 with mobile phase (Table 2)

Diflunisal 2:3 with acetonitrile then 1:1 with PB

Flufenamic acid 1:2 with acetonitrile then 1:1 with water

Imipramine 1:10 with mobile phase (Table 2)

Itraconazole 1:2 with acetonitrile

Lignocaine 1:10 with mobile phase (Table 2)

Meclofenamic acid 2:3 with acetonitrile then 1:1 with water

Mianserine 2:3 with acetonitrile

Propranolol As for atenolol

Ropivacaine 1:10 with mobile phase (Table 2)

Triflupromazine 2:3 with acetonitrile

Verapamil 1:10 with mobile phase (Table 2)


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

26

Table 2: HPLC conditions for drug assays

Drug Mobile phase Column Detector

wavelength (nm)

Retention time of analyte (min)

Internal standard / Retention time

(min)

Atenolol 80% water containing 5.2 mM pentanesulfonic acid and 0.3 mM NNN'N' TMED / 20% acetonitrile (adjusted to pH 2.6 with orthophosphoric acid)

C-8 230 7.4 sotalol 10.5

Bupivacaine 85% 100 mM phosphate buffer pH 5.9 / 15% acetonitrile containing 0.1% dimethylamine (adjusted to pH 3.0 with orthophosphoric acid)

C-18 210 13.3 lignocaine 3.5

Bupropion 60% PB (50 mM, pH 7.4) containing 10 mM triethylamine and 7 mM heptanesulfonic acid / 40% acetonitrile

**C-8 214 2.4 verapamil 4.8

Chloroquine 64% water containing 5.2 mM pentanesulfonic acid and 0.3mM NNN'N' TMED / 36% acetonitrile (adjusted to pH 2.6 with orthophosphoric acid)

C-8 220 4.8 ropivacaine 7.5

Chlorpromazine 44% sodium acetate buffer (10 mM, pH 4.3) / 56% acetonitrile

*C-8 254 4.0 thioridazine 3.2

Diazepam 46% water containing 5.2 mM, pentanesulfonic acid and 0.3 mM NNN'N' TMED / 54% acetonitrile (adjusted to pH 2.6 with orthophosphoric acid)

C-18 230 6.1 desmethyl- diazepam

4.3

Diflunisal 70% PB (25mM, pH 7.4) containing 7.2 mM triethylamine / 30% acetontitrile C-18 240 2.7 mefenamic acid 3.2

Flufenamic acid 70% PB (25 mM, pH 7.4) containing 7.2 mM triethylamine / 30% acetonitrile C-18 270 4.1 diclofenac 2.3

Imipramine as for diazepam C-18 259 6.0

Itraconazole 45% water containing 2mM triethylamine (adjusted to pH 2.5 with perchloric acid) / 55% acetnonitrile.

C-18 245 4.6 ketoconazole 2.6

This article has not been copyedited and form

atted. The final version m

ay differ from this version.

DM

D Fast Forw

ard. Published on May 24, 2011 as D

OI: 10.1124/dm

d.111.039354 at ASPET Journals on April 21, 2021 dmd.aspetjournals.org Downloaded from


DMD #39354

27

Drug Mobile phase Column Detector

wavelength (nm)

Retention time of analyte (min)

Internal standard / Retention time

(min)

Lignocaine as for bupivacaine C-18 210 3.5 ropivacaine 6.8

Meclofenamic acid as for diflunisal C-18 240 4.0 diflunisal 3.1

Mianserine 45% PB (25 mM, pH 7.4) containing 7.2 mM triethylamine / 55% acetonitrile **C-8 254 3.6 doxepin 2.1

Propranalol as for chloroquine C-8 220 9.1 pindolol 4.4

Ropivacaine as for lignocaine C-18 210 6.8 lignocaine 3.5

Triflupromazine 47.5% sodium acetate buffer (10 mM, pH 4.3) / 42.5% acetonitrile

**C-8 254 4.1 chlorpromazine 3.2

Verapamil as for diazepam C-8 220 5.9 ropivacaine 3.8

C-18: Waters, Nova Pak, particle size 4 micron, 3.9 (id) x 150 mm C8: Beckman Ultrasphere (Octyl), particle size 5 micron, 4.6 (id) x 250 mm

*C8: Develosil, particle size 5 micron, 4.6 x 150 mm **C-8: Agilent Zorbax Eclipse XDB, particle size 5 micron, 4.6 x 150 mm

This article has not been copyedited and form

atted. The final version m

ay differ from this version.

DM

D Fast Forw

ard. Published on May 24, 2011 as D

OI: 10.1124/dm

d.111.039354 at ASPET Journals on April 21, 2021 dmd.aspetjournals.org Downloaded from


DMD #39354

28

Table 3: Physicochemical characteristics and non-specific binding of the test drugs (100, 200 and 500 µM) to HLM (1 mg/ml) as determined by equilibrium dialysisa

Compound pKa log P fu(mic) ± SD

100 µM 200 µM 500 µM

Acids

Diflunisal 2.9 4.4 0.83 ± 0.02 0.86 ± 0.04 0.85 ± 0.00

Flufenamic acid 3.7 5.6 0.69 ± 0.03 0.68 ± 0.02 0.71 ± 0.02

Meclofenamic acid 3.6 6.7 0.58 ± 0.03 0.59 ± 0.02 0.69 ± 0.08

Bases

Atenolol 9.2 0.1 1.02 ± 0.07 0.99 ± 0.04 1.00± 0.00

Bupivacaine 8.2 3.6 0.98 ± 0.04 1.02 ± 0.02 0.97 ± 0.03

Bupropion 7.2 3.5 1.12 ± 0.16 1.10 ± 0.10 1.15 ± 0.04

Chloroquine 10.5 4.7 0.87 ± 0.03 0.84 ± 0.07 0.78 ± 0.04

Chlorpromazine 9.4 5.2 0.08± 0.02 0.08 ± 0.01 0.14 ± 0.04

Imipramine 9.5 4.8 0.48 ± 0.01 0.57 ± 0.04 0.69 ± 0.04

Lignocaine 8.5 2.4 1.08 ± 0.10 1.01 ± 0.03 1.04 ± 0.06

Mianserine 8.3 3.4 0.29 ± 0.11 0.37 ± 0.01 0.37 ± 0.08

Propranolol 9.1 3.1 0.48 ± 0.05 0.61 ± 0.13 0.59 ± 0.06

Ropivacaine 8.2 3.1 0.94 ± 0.03 0.97 ± 0.04 0.97 ± 0.03

Triflupromazine 9.4 5.7 0.17 ± 0.01 0.27 ± 0.01 0.44 ± 0.03

Verapamil 9.0 3.9 0.70 ± 0.04 0.73 ± 0.05 0.70 ± 0.11

Neutral

Diazepam 3.4 3.0 0.76 ± 0.06 0.77 ± 0.03 0.76 ± 0.03

aEquilibrium dialysis data shown as mean ± SD of 3 measurements.


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

29

Table 4: Effects of imipramine, lignocaine and propranolol (100 – 1000 µM) on ANS fluorescence in the presence of HLM (0.25 mg/ml)

Drug Percent ANS fluorescence increment/decrement at drug concentrations in the range 100 - 1000 μM:

100 200 300 400 500 600 700 800 900 1000

Imipramine 46 77 107 133 154 176 192 213 225 241

Lignocaine 1 1 -1 2 -2 -3 -3 -3 -2 -2

Propranolol 27 47 65 77 86 107 118 131 133 136

Data shown as means of duplicate measurements (< 5% variance)


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

30

Table 5: Effects of test drugs (100, 200 and 500 µM) on ANS fluorescence in the presence of HLM (0.25 mg/ml)

Compound Percent ANS fluorescence increment/decrement

100 µM 200 µM 500 µM

Acids

Diflunisal -8 -18 n/d

Flufenamic acid -35 -43 n/d

Meclofenamic acid -36 -62 -95

Bases

Atenolol -1 1 3

Bupivacaine 0 1 3

Bupropion -11 -15 -24

Chloroquine 1 2 6

Chlorpromazine 107 179 353

Imipramine 46 77 154

Lignocaine 1 1 -2

Mianserine 45 77 158

Propranolol 27 47 86

Ropivacaine 1 1 1

Triflupromazine 97 144 290

Verapamil 8 26 57

Neutral

Diazepam -19 -35 n/d



at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

31

Table 6: Effect of amitriptyline, nortriptyline, itraconazole, phenytoin, S-naproxen and lamotrigine on ANS fluorescence in the presence of HLM (0.25 mg/ml)

Drug Percent ANS fluorescence increment/decrement at:

100 μM 200 μM 500 μM

Itraconazole 138 239 548

Amitriptyline 43 66 139

Nortriptyline 40 64 122

Phenytoin -8 -13 -21

S-Naproxen 3 -3 -8

Lamotrigine 4 6 6



at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


DMD #39354

32

Table 7: Predicted (from ANS fluorescence) versus observed (from equilibrium dialysis) fu(mic) values for the validation seta

Drug

Predicted fu(mic) and observed fu(mic) at:

100 μM 200 μM 500 μM

Predicted Observed Predicted Observed Predicted Observed

Itraconazole 0.02 0.03 0.03 0.02 0.03 0.02

Amitriptyline 0.40 0.44 0.47 -b 0.50 0.55

Nortriptyline 0.43 0.46 0.48 0.40 0.55 0.61

Phenytoin 0.87 0.83 0.96 0.89 0.96 -b

S-Naproxen 0.96 0.99 1.0 0.99 1.0 0.97

Lamotrigine 0.96 0.93 1.0 1.0 1.0 0.96

a Equilibrium dialysis data taken from McLure et al. (2000) and Rowland et al. (2006), except for itraconazole (determined here). fu(mic) values generated from ANS fluorescence normalised to HLM protein concentration of 1 mg/ml using equation 2 (Materials and Methods).

b Not determined by McLure et al. (2000)


at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from



at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from



at ASPE

T Journals on A

pril 21, 2021dm


ownloaded from


Documents

DMD #39354dmd.aspetjournals.org/content/dmd/early/2011/05/24/... · 5/24/2011 · James A McLure, Donald J Birkett, David J Elliot, J Andrew Williams, Andrew Rowland and John O Miners