ATP Serves as an Endogenous Inhibitor of UDP...

DMD #46862

- 1 -

ATP Serves as an Endogenous Inhibitor of

UDP-Glucuronosyltransferase (UGT): A New Insight into

the ‘Latency’ of UGT

Yuji Ishii, Kie An, Yoshio Nishimura, and Hideyuki Yamada

Graduate School of Pharmaceutical Sciences, Kyushu University, Fukuoka,

Japan

DMD Fast Forward. Published on July 30, 2012 as doi:10.1124/dmd.112.046862

Copyright 2012 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 July 30, 2012 as DOI: 10.1124/dmd.112.046862

at ASPE

T Journals on D

ecember 18, 2020

dmd.aspetjournals.org

Dow

nloaded from

http://dmd.aspetjournals.org/

DMD #46862

- 2 -

Running title: Novel Mechanism of UGT ‘latency’

Correspondence author:

Yuji Ishii, Ph.D.

Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1

Maidashi, Higashi-ku, Fukuoka 812-8582, Japan,

Phone: +81-92-642-6586

Fax:+81-92-642-6588

E-mail: [email protected]

Number of text pages is 45.

Number of tables is 6.

Number of figures is 6.

Number of references is 45.

Number of words in abstract is 250 (words).

Number of words in introduction is 750 (words)

Number of words in discussion is 1311 (words)

Number of supplemental materials is 3.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 3 -

Abbreviations: UGT, UDP-glucuronosyltransferase; ER, endoplasmic reticulum;

RLM, rat liver microsomes; HLM, human liver microsomes; GRP,

glucose-regulated protein; 4-MU, 4-metylumbelliferone; 4-MUG,

4-Methylumbelliferyl-β-D-glucuronide; Brij-58, polyoxyethylene cetyl alcohol

ether; PDI, protein disulfide isomerase; EGFR, epidermal growth factor receptor


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 4 -

Abstract:

We have suggested that adenine-related compounds are allosteric inhibitors of

UGT in rat liver microsomes (RLM) treated with detergent. To clarify whether the

same occurs with a pore-forming peptide, alamethicin, the effects of

adenine-related compounds on 4-metylumbelliferone (4-MU) glucuronidation

were examined using RLM and human liver microsomes (HLM). ATP inhibited

4-MU glucuronidation when Brij-58-treated RLM were used (IC50= around 70

µM). However, alamethicin-treated RLM exhibited a lower susceptibility (IC50=

around 460 µM) than Brij-58-treated RLM. A similar phenomenon was observed

when pooled HLM were used. Then, the endogenous ATP content of RLM was

determined in the presence and absence of alamethicin or detergent. While no

ATP remained in the microsomal pellets after Brij-58-treatment, more than half

the microsomal ATP remained even after treatment with alamethicin. Further, the

Vmax in the absence of an adenine-related compound was approximately three

times higher in Brij-58-treated than in alamethicin-treated RLM. The difference in

the inhibitory potency observed would be due to the difference in remaining

endogenous ATP and the accessibility of exogenous ATP to the luminal side of

the endoplasmic reticulum (ER) where the active site of UGT is located. Gefitinib,


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 5 -

a protein tyrosine kinase inhibitor, markedly inhibited human UGT1A9 activity.

Interestingly, AMP antagonized Gefitinib-provoked inhibition of UGT1A9 and ATP

exhibited an additive inhibitory effect at a lower concentration. Therefore,

Gefitinib inhibits UGT1A9 at the common ATP binding site shared with ATP and

AMP. Releasing adenine nucleotide from the ER is suggested to be one of the

mechanisms to explain the ‘latency’ of UGT.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 6 -

Introduction:

Glucuronidation is one of the most important steps in drug metabolism.

Many endo- and xeno-biotics and/or their metabolites are converted to

water-soluble glucuronides (Dutton, 1980; Bock et al, 1987; Ritter, 2000). This

step is catalyzed by UDP-glucuronosyltransferase (UGT) which is expressed

mainly in the liver and gastro-intestinal tract (Tukey and Strassburg, 2000). UGT

is a family of enzymes which transfer the glucuronic acid moiety of the

co-substrate, UDP-glucuronic acid (UDP-GlcUA), to a substrate (Mackenzie et al,

2005). This enzyme is localized in the endoplasmic reticulum (ER) and

experimental evidence has suggested that the most important domain of UGT is

located within the ER (Shepherd et al., 1989; Radominska-Pandya et al, 2005).

A membrane-spanning domain of UGT is located near the carboxy-terminus

which extrudes cytosolic tail consisting of about 20 amino acid residues (Meech

and Mackenzie, 1997). In agreement with the topology of UGT in the ER

membrane, it has been suggested that a UDP-GlcUA transporter is expressed

on the ER membrane, and this protein translocates UDP-GlcUA from the cytosol

to the lumen of the ER (Hauser et al., 1988). Muraoka et al. (2007) have cloned

a UDP-GlcUA transporter, UDP-galactose transporter-related protein 7


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 7 -

(UGTrel7; solute carrier family 35, SLC35D1), which is involved in the above

process.

It has long been known that UGT activity is increased following treatment

of microsomes with detergent (Winsnes, 1969), and this phenomenon is referred

to as the ‘latency’ of UGT. The same is also observed when cells were treated

with a pore-forming peptide, alamethicin (Bánhegyi et al, 1983). The reason why

this latency is observed is thought to be due to the presence of a membrane (ER

membrane) obstructing the free access of substrates to the active site of UGT. It

is generally accepted that the latency of UGT is due to the limited transport of

UDP-GlcUA from cytosol to the lumen of ER, the process of which is mediated

by a UDP-GlcUA transporter(s)(Muraoka et al. 2007). This explanation sounds

reasonable. If this is true, the kinetic Michaellis constant (Km) should be lower

under detergent-treating conditions than under conditions not involving

detergent treatment. However, the Km of the substrate (aglycone) is not often

reduced even after detergent treatment (Puig and Tephly, 1986). It is, therefore,

unlikely that the removal of the membrane obstruction is the sole reason for the

latency of UGT.

We have reported previously that adenine nucleotides, such as ATP,


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 8 -

NADP+ and NAD+, are endogenous inhibitors of UGT (Nishimura et al., 2007).

These adenine nucleotides inhibit UGT in a non-competitive manner, and they

exhibited the effect only when microsomes are disrupted by detergent. On the

basis of a structure-inhibitory effect relationship, while NADPH and NADH

exhibit no or only a minor inhibitory effect on UGT (IC50>1 mM), their oxidized

forms, NADP+ and NAD+, are potent UGT inhibitors. There are

NAD(P)+-dependent dehydrogenases, such as 11β-hydroxysteroid

dehydrogenase and hexose-6-phosphate dehydrogenase, in the lumen of the

ER (Hewitt et al, 2005). Furthermore, molecular chaperones, such as

glucose-regulated protein 78 (GRP78) and GRP94, are expressed within the ER,

and they bind to ATP and hydrolyze it (Dierks et al., 1996; Rosser et al., 2004).

Therefore, significant amounts of ATP, NAD+ and NADP+ should be present in

the lumen of the ER, and they are needed for the oxidoreductase and chaperone

functions. If adenine nucleotides work as endogenous inhibitors of UGT, their

concentration in the ER is considered to be a determinant regulating the action

of certain hormones which are UGT substrates. However, the actual

concentration of adenine nucleotides within the ER has not been accurately

determined.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 9 -

Previously, we observed the nucleotide-dependent inhibition of UGT

using detergent-treated microsomes. To clarify whether the same occurs with a

pore-forming peptide, alamethicin, the effects of adenine-related compounds on

4-metylumbelliferone (4-MU) glucuronidation were examined using liver

microsomes from rats (RLM) and humans (HLM). In addition, we measured the

ATP content in the ER with and without alamethicin/detergent treatment.

Through these experiments, we investigated the hypothesis that a change in the

concentration of adenine nucleotides within the ER is the reason for the latency

of UGT. On the other hand, agents such as Gefitinib inhibiting epidermal growth

factor receptor (EGFR)-protein tyrosine kinase are UGT inhibitors (Liu et al.,

2010; Fujita et al., 2011; Piu et al., 2011). However, the reason why Gefinitib and

related compounds exhibit an inhibitory effect on UGT has not been elucidated.

Since their pharmacological target site is the ATP binding domain of the protein

tyrosine kinase, we also examined whether Gefitinib inhibits UGT through the

ATP binding site on UGT.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 10 -

Materials and Methods

Materials. ATP, NADH, NADPH, 4-MU, D-saccharic acid-1,4-lactone, and

L-α-phosphatidylcholine (egg yolk, type XI-E) were purchased from

Sigma-Aldrich (St. Louis, MO). NAD+, NADP+, and UDP-GlcUA trisodium salt

were obtained from Wako Pure Chemicals, Co. Ltd. (Osaka, Japan).

4-Methylumbelliferyl-β-D-glucuronide (4-MUG) and polyoxyethylene cetyl

alcohol ether (Brij-58) were purchased from Nakalai Tesque (Kyoto, Japan).

ENLITEN® rLuciferase/Luciferin reagent and 5xPassive lysis buffer were

purchased from Promega (Madison, WI). Gefitinib (Iressa®) was purchased

from Tocris Bioscience (Bristol, UK). All other reagents were of the highest grade

commercially available.

Animals and treatment. Animal experiments in this study were conducted

following the approval of the Ethics Committee for Animal Experiments of

Kyushu University. Male Sprague-Dawley rats (5 weeks-old) were purchased

from Kyudo (Kumamoto, Japan) and they were maintained for one week with

free access to water and a standard diet under a 7 am to 7 pm light/dark cycle.

The liver was removed and perfused with ice-cold saline, and then the


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 11 -

microsomes were prepared by differential centrifugation as described previously

(Oguri et al., 1996). The microsomes were thenre-suspended in 0.25 M sucrose.

Protein assay. The protein concentration was determined by the method of

Lowry et al. (1951) using bovine serum albumin as a standard.

Determination of UGT activity. The activity of 4-MU glucuronidation was

determined by the published methods (Hanioka et al., 2001) with slight

modifications (Nishimura et al., 2007). Human UGT1A9 SupersomeTM and

human liver microsomes (pooled microsomes from 150 donors) were purchased

from BDGentest (Woburn, MA).

Determination of ATP by luciferin and luciferase. Rat liver microsomes

were diluted to give a protein concentration of 5 mg/mL and then treated with

Brij-58 (0.25 mg/mg protein) or alamethicin (0.05 mg/mg protein). The mixture

was kept on ice for 1 h, then centrifuged at 105,000 xg (4˚C) for 60 min, and the

supernatant and pellets were separated. Both fractions were mixed/diluted with

20-volume relative to the original microsome amount (protein base) of 1xPassive

Lysis Buffer (Promega, Madison, WI). The mixture was kept at room temperature

for 30 min then a 5 μL aliquot of the diluted sample was mixed with 50 μL

Luciferase/Luciferin reagent. The emission of ATP-dependent


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 12 -

chemiluminescence was measured in a Turner Biosystems Luminometer (Model

TD-20/20)(Promega, Madison, WI). The detection limit was 0.1 nmol ATP/5 μL

sample.

Data analyses. Statistical significance, kinetic paramerers and IC50 were

calculated using the computer software, GraphPadPrism5 (GraphPad, La Jolla,

CA).

Prediction of the three-dimentional structure and ligand binding sites of

UGT1A9. The amino acid sequence of UGT1A9 (GenBank: AAG30418.1) was

subjected to a modeling system on the Web linking to a server installed with

Phyre2 (Protein Homology/analogY Recognition Engine V 2.0)

(http://www.sbg.bio.ic.ac.uk/phyre2/html/; Kelly and Sternberg, 2009). The

following six proteins were selected as the templates based on the Structural

Classification of Proteins and ASTRAL release 1.75A (March 2012): flavonoid

3-O-glucosyltransferases (d2c1xa1, Offen et al, 2006; and c3hbjA, Modolo et al.,

2009), hydroquinone glucosyltransferase (d2vcha1, Brazier-Hicks et al, 2007),

(iso)flavonoid glycosyltransferase (d2pq6a1, Li et al, 2007),

vancosaminyltransferase (Gtf glycosyltransferase) (d1rrva, Mulichak et al, 2004),

and triterpene/flavonoid glycosyltransferase (d2acva1, Shao et al, 2005). This


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 13 -

selection was also based on a heuristics to maximize the confidence,

percentage identity and alignment coverage. In the modeling, the

three-dimentional structure of about 87% of whole UGT1A9 body could be

constructed by overlapping with templates. The confidence of this modeling was

expected to be over 90%. Because the similar way could not be applicable to the

structural simulation of the other region involving the highly variable C-terminal

domain, the structure of this region was tentatively predicted by ab initio

calculation. The construct obtained was then submitted to a 3DLigandSite server

(http://www.sbg.bio.ic.ac.uk/~3dligandsite/; Wass et al, 2010) for the estimation

of ligand binding site.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 14 -

Results

inhibitory effect of adenine-related compounds on glucuronidation by

Brij-58- and alamethicin-treated RLM. The possibility whether

adenine-related compounds inhibit UGT activity both in RLM treated with a

pore-forming peptide, alamethicin, as well as a detergent, Brij-58, was examined

using 4-MU as a substrate. To clarify the specificity of nucleotides and related

substances, the inhibitory potential of 12 substances was assessed. Table 1

shows the IC50 values of the compounds examined. Of the substances tested,

ATP exhibited the strongest inhibitory effect in both Brij-58 and

alamethicin-treated RLM. However, the IC50 was 7-times higher in

alamethicin-treated RLM than in a Brij-50-treated preparation: the IC50 of ATP

was approximately 70 and 460 µM in Brij-58- and alamethicin-treated RLM,

respectively. Although ADP also exhibited an inhibitory effect in both

preparations, its IC50 was 11- and 3-times higher than ATP in Brij-58- and

alamethicin-treated microsomes. However, the inhibitory effect of NADP+ was

stronger than ADP, and NAD+ exhibited a stronger inhibitory effect than ADP only

in Brij-58-treated RLM. CTP had an inhibitory effect close to that of ADP in

Brij-58-treated RLM but not in alamethicin-treated RLM and no inhibitory effect


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 15 -

was observed for adenosine, NADH, NADPH, GTP and UTP in RLM treated with

Brij-58 and alamethicin. Although the relative IC50 (alamethicin/Brij-58) varied

from 1.6 with ADP to 6.9 with ATP, alamethicin-treated microsomes were always

less sensitive to nucleotides than Brij-58-treated preparations. Of the nucleotide

triphosphates tested, only CTP had an inhibitory effect close to that of ADP in

Brij-58-treated RLM. The inhibitory potential of adenine-nucleotides also

depends on the number of phosphate groups [ATP >> ADP > AMP (= almost no

effect)]. Thus, the number of phosphates at the 5’-position of ribose is thought to

be one of the determinants of the inhibitory effect of the adenine nucleotides.

Finally, neither NADH nor NADPH exhibited any inhibitory effect on UGT activity,

while NAD+ and NADP+ did have an inhibitory effect. Therefore, the oxidized

status of the nicotinamide moiety seems to be another factor affecting the

inhibitory effect.

Previously, we have confirmed that the mode of inhibition of 4-MU

glucuronidation catalyzed by Brij-58-treated RLM by adenine-related compounds

was non-competitive. In this study, we examined whether the same is also true

for the glucuronidation catalyzed by alamethicin-treated RLM. For this purpose,

a kinetic experiment was performed by varying the 4-MU concentration (Fig. 1).


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 16 -

Even addition of ATP, or NADP+ to the reaction mixture did not change the

Michaelis-Menten kinetics, and increased the concentrations of nucleotides

producing more pronounced inhibition (Fig. 1A-D). These plots strongly

suggest that 4-MU UGT in both Brij-58- and alamethicin-treated RLM is inhibited

by ATP and NADP+ in a non-competitive manner. This assumption was

supported by re-evaluation of the Eadie-Hofstee plots (Fig. 1E-H); that is,

regression lines with different concentrations of inhibitor run parallel to each

other. Taken together, these results suggest that ATP and NADP+ reduce 4-MU

UGT function in both Brij-58- and alamethicin-treated RLM by the same

mechanism and by interacting with the enzyme(s) at a site distinct from

substrate-binding sites. In agreement with the data shown in Table 1, the

inhibitory constants (Ki) of ATP and NADP+ were smaller in Brij-58-treated than

alamethicin-treated RLM (Table 2).

Effect of AMP on the inhibition caused by ATP and NADP+. As described

before, the 4-MU UGT activity in both Brij-58- and alamethicin-treated RLM is

barely affected by AMP (Table 1). However, we have shown that AMP has an

ability to modify the inhibitory effects of other nucleotides in Brij-58-treated RLM

(Nishimura et al, 2007). To investigate whether the same is also true for


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 17 -

alamethicin-treated RLM, the effect of AMP on the 4-MU glucuronidation activity

was examined. In both Brij-58- and alamethicin-treated RLM, the inhibitory effect

of ATP on 4-MU UGT activity was antagonized by AMP (Fig. 2A-B). Such an

antagonistic effect became stronger along with an increase in AMP

concentration. AMP also antagonized the inhibitory effect of NADP+ (Fig. 2C-D).

These results suggest that AMP competes with adenine-related inhibitors at the

binding site of 4-MU UGT in both Brij-58- and alamethicin-treated RLM, although

AMP itself lacks any inhibitory potential.

Mechanism underlying the UGT latency caused by detergent- and

alamethicin-treatment. Our preliminary study showed that Brij-58-treatment

causes a 17-fold increase in the Vmax of 4-MU glucuronidation in RLM

(Supplemental Table S1). The Michaelis constant, Km, for 4-MU was slightly

increased by Brij-58, while the Km for co-substrate, UDP-GlcUA, tended to be

reduced by the treatment. Therefore, it is likely that an increase in the affinity for

UDP-GlcUA may contribute to the enhancement of Vmax following detergent

treatment. However, an increase in the Km for 4-MU does not agree with the

hypothesis that removal of membrane obstruction is the sole reason for the

latency of UGT. Table 3 shows the kinetic parameters for 4-MU glucuronidation


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 18 -

which were compared for Brij-58- and alamethicin-treated RLM. The Km values

for 4-MU and UDP-GlcUA were slightly higher in alamethicin-treated than

Brij-58-treated RLM. When the kinetics were studied with different 4-MU

concentrations, the Vmax in alamethethin-treated RLM was approximately one

third that of Brij-58-treated RLM. As far as the activating conditions with Brij-58

(0.25 mg/mg protein) and alamethicin (0.05 mg/mg protein) were concerned, the

Vmax obtained after varying the 4-MU/UDP-GlcUA concentrations was higher in

Brij-58-treated than in alamethicin-treated RLM.

Although the ER appears to contain a substantial concentration of adenine

nucleotides for molecular chaperons and dehydrogenases, it is not known

whether the luminal adenine nucleotides are released by treating microsomes

with detergent or alamethicin. If this is true, the inhibitory effect should be

abolished or weakened following detergent- or alamethicin-treatment of

microsomes. To examine this hypothesis, we determined the ATP content in the

ER with and without Brij/alamethicin treatment. Table 4 shows the ATP content in

microsomal pellets and supernatant after treatment with Brij-58 or alamethicin,

the concentrations of which were the same as those used for determining UGT

activity. ATP was released by Brij-58 treatment and no detectable ATP remained


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 19 -

in the detergent-treated microsomes. However, ATP remained in the

microsomes in the control sample. Part of the microsomal ATP was also

released by alamethicin treatment, but almost 60% of ATP remained unreleased

even after alamethicin treatment. The lower Vmax in alamethicin-treated than in

Brij-58-treated RLM (Table 3) seems to correlate with the extent of ATP

remaining in the microsomes. For example, it is suggested that an ATP

concentration sufficient for UGT inhibition still remains in microsomes even after

alamethicin treatment. Conversely, it would be reasonable to believe that the

release of ATP from microsomes after detergent treatment results in the abolition

of its inhibitory effect on UGT. Therefore, it is suggested that the presence of

inhibitory adenine nucleotides is one possible mechanism for the latency of UGT.

The combined effect of adenine-related compounds on UGT activity. Under

normal conditions, the cellular ATP concentration is several mM (Schwiebert,

2000) and this is 5- to 10-fold higher than the ADP level (Taylor et al., 1975). The

proportion of ATP, ADP and AMP concentrations has been assumed to satisfy

the following equation (Hadie et al., 1998; 2001).

[AMP]/[ATP] = ([ADP]/[ATP])2

The concentration of adenine nucleotides in the ER lumen of living cells has


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 20 -

not been accurately determined. However, on the basis of the above equation,

three ratios (ATP:ADP:AMP=1:1:1, 25:5:1 and 100:10:1) at four total

concentrations (ATP+ADP+AMP) from 250 to 1500 µM were examined under

UGT inhibition. In Brij-58-treated RLM, the inhibitory effect on UGT was

increased depending on an increase in the ATP ratio (Fig. 3). Therefore, the

inhibitory effect of ATP can be negatively modulated by its metabolites ADP and

AMP. Although a similar trend was observed in alamethicin-treated RLM, the

effect of combined nucleotides was more marked in Brij-58-treated RLM. Then,

we examined the effect of more complicated mixtures on UGT activity. For this,

all inhibitory substances listed in Table 1 were mixed at their respective IC50 and

half IC50 concentrations and the effects of these cocktails on UGT activity were

examined. In Brij-58-treated RLM, the ‘IC50/2’ cocktail reduced 4-MU

glucuronidation to approximately 40% of the control, while the effect was almost

abolished when alamethicin-treated RLM were used (Fig. 4). It is, therefore,

likely that the access of inhibitory adenine-related compounds to the target site

of UGT is limited in alamethicin-treated RLM.

Inhibition of glucuronidation in HLM and human UGT1A9 by ATP. The

effect of AMP and ADP as well as ATP on UGT activity was examined in Brij-58


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 21 -

and alamethicin-treated HLM. ATP exhibited the strongest inhibitory effect in

both Brij-58 and alamethicin-treated HLM (Table 5). However, the IC50 was

27-times higher in alamethicin-treated HLM. No inhibitory effect was observed

for ADP and AMP in HLM treated with Brij-58 or alamethicin. We then examined

the effect of adenine nucleotides on human UGT1A9 which is one of the major

UGT isoforms involved in 4-MU glucuronidation. In this case, we used

alamethicin-treated UGT1A9 Supersome™ which is used widely for research

purposes. As shown in Table 5, ATP inhibited UGT1A9-catalyzed 4-MU

glucuronidation with an IC50 lower than that obtained in alamethicin-treated HLM.

The IC50 of ADP and AMP was greater than 1000 µM (Table 5).

The mechanism of Gefitinib-induced inhibition of UGT1A9. As shown in Fig.

5, Gefitinib significantly inhibited 4-MU UGT activity mediated by UGT1A9 as

well as Brij-58- and alamethicin-treated HLM. The IC50 of Gefitinib for

UGT1A9-catalyzed 4-MU glucuronidation was about 20 µM (Table 6). To

investigate whether Gefitinib inhibits UGT through the putative ATP binding site

on UGT, the inhibitory effect of Gefitinib on UGT1A9 was examined in the

presence or absence of ATP at its IC50 concentration. Although ATP exhibited an

additive inhibitory effect when the Gefitinib concentration was less than 50 µM,


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 22 -

no such effect was observed in the presence of Gefitinib at a concentration of 75

µM or more (Fig. 6). Furthermore, the Gefitinib-induced inhibition of

UGT1A9-catalyzed 4-MU glucuronidation was antagonized in the presence of

AMP which is a non inhibitory adenine nucleotide but a negative modulator of

ATP-dependent inhibition (Fig. 6). From these observations, Gefitinib and

adenine nucleotides are thought to share the binding site on UGT. Therefore, it is

suggested that Gefitinib inhibits UGT1A9 through the putative ATP binding site

on UGT.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 23 -

Discussion

The inhibitory effect of adenine nucleotides and related compounds on

4-MU UGT was studied in RLM and HLM. ATP exhibited a strong inhibitory effect

in HLM as well as RLM. On comparing the inhibitory effect of ATP on UGT

between microsomes pre-treated with Brij-58 and alamethicin, the inhibitory

effect was more marked in Brij-58-treated RLM than in alamethicin-treated RLM.

As the mode of inhibition of 4-MU UGT was non-competitive in both Brij-58- and

alamethicin-treated RLM, ATP is considered to be an allosteric inhibitor of UGT.

The data suggest the poor ability of adenine-related compounds to gain access

to the target site on UGT in alamethicin-treated microsomes (Fig. 4). Therefore,

the IC50 observed in the detergent-treated microsomes is considered as a

relevant index for the inhibitory potency of the adenine-related compounds.

It is suggested that there is an ATP transporter on the microsomal

membrane which draws ATP into the lumen (Clairmont et al, 1992; Kim et al.,

1996). ATP uptake into microsomes is saturable with a Km of 3 to 4 µM and a

Vmax of 3 to 7 pmol/min/mg protein (Clairmont et al, 1992). When

firefly-luciferase, which contains a signal sequence and a sequence for ER


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 24 -

targeting, was expressed in Chinese hamster ovary (CHO) cells, ATP-dependent

chemiluminescence was observed by microscopy after addition of luciferin

(Dorner and Kaufman, 1994). Thus, it is likely that the luminal side of the ER

contains a substantial level of ATP, although its concentration remains unknown.

In addition, the ER lumen expresses proteins, such as protein disulfide

isomerase (PDI), UGT1A1 and UGT1A7, which are sensitive to phosphorylation

(Quéméneur et al, 1994; Basu et al., 2005). Furthermore, molecular chaperones,

such as PDI, calreticulin, GRP78 and GRP94, capable of binding to ATP are

present within the ER (Guthapfel et al., 1996; Dierks et al., 1996; Rosser et al.,

2004). The dissociation constant of ATP and the Km for ATPase activity of

GRP78 are reported to be 5.4 and 28.6 µM, respectively (Wearsch and Nicchitta,

1997). For PDI, the corresponding values are 9.7 and 7.1 µM, respectively

(Guthapfel et al., 1996). Accordingly, the concentration of ATP needed for the

functions of the above series of proteins is assumed to be around 30 µM, which

is close to the IC50 determined in this study. Since these ATP-binding proteins

are ATPases, it would be possible that they modulate UGT activity by

hydrolyzing inhibitory ATP to less-inhibitory ADP and non-inhibitory AMP in the

ER.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 25 -

As shown in Fig. 3, the inhibitory effect of adenine nucleotides on UGT

activity varied depending on the ratio of ATP:ADP:AMP. If the concentration of

ATP in the ER is kept much higher than its IC50, the UGT activity should be

completely suppressed under such conditions. However, it is reasonable to

assume that the concentration of ATP and antagonistic AMP varies dynamically

in the lumen of the ER to switch the glucuronidation reaction on/off. Our

preliminary study indicated that the ATP level within the hepatic ER is higher in

the evening than in the morning (Supplemental Figure S1). The concentration of

nocturnal ATP was close to 30 µM. Also, the extracellular production of

adenosine from ATP in the retina exhibits a circadian variation (Ribelayga and

Mangel, 2005), and we also found an intra-day variation in the ATP level in the

lumen of the ER.

Although it is assumed that the limited transport of UDP-GlcUA from

cytosol to the ER lumen renders UGT latent, the Vmax of UGT was markedly

increased by Brij-58 treatment, and the Km values for 4-MU and UDP-GlcUA

showed only a minor change (Supplemental Table S1). Thus, although it cannot

be excluded that reducing the Km for UDP-GlcUA may make some contribution

to an apparent increase in UGT function produced by detergent, increasing the


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 26 -

Km for substrates (glucuronic acid acceptors) seems to disagree with the idea

that the removal of interference by the membrane is the absolute mechanism

explaining the latency of UGT.

On the basis of the inhibitory effect of ATP and its level in the lumen of

the hepatic ER determined in this study, we propose a new hypothesis for the

latency of UGT. As mentioned before, in untreated microsomes, the luminal

concentration of ATP is around 30 µM (Supplemental Figure S1), which is

comparable with its IC50 value for UGT inhibition. This observation suggests that

UGT function is partially suppressed by ATP under normal physiological

conditions. However, when microsomes are treated with detergent or

alamethicin, the enclosure of adenine nucleotides in the ER is disrupted, and the

concentration of the nucleotides is markedly reduced. Therefore, the

concentration of adenine nucleotides, especially in Brij-58-treated microsomes,

becomes too low to inhibit UGT. This possibility seems to be consistent with the

fact that the inhibition mode of adenine nucleotides is non-competitive and the

Vmax of UGT is increased by detergent treatment (Supplemental Table S1,

Table 2, Fig. 1). Furthermore, when RLM were treated with Brij-58, ATP did not

remain in the pellets obtained after ultra-centrifugation, while this nucleotide


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 27 -

remained in the microsomal pellets in untreated microsomes. Taking all above

findings into consideration, it is suggested that a detergent/alamethicin-caused

decrease in adenine nucleotides in the ER is one possible mechanism

explaining the latency of UGT.

The inhibitors of EGFR-protein tyrosine kinase have an undesired

inhibitory effect on UGT (Liu et al., 2010; Fujita et al., 2011; Piu et al., 2011). This

may cause a drug-drug interaction in UGT-dependent metabolism (Fujita et al,

2011). However, it has not been discovered why these EGFR-protein tyrosine

kinase inhibitors suppress UGT activity. In this study, we suggest that Gefitinib

inhibits UGT1A9 by binding to the allosteric site shared with ATP and AMP. ATP

at its IC50 concentration additively enhanced the inhibition of UGT catalysis with

Gefinitib at low concentrations, and AMP antagonized the inhibitory effect of

Gefitnib. The inhibitory potency of Gefitinib was far stronger than that of ATP.

Although ATP at IC50 showed an additive effect on the inhibition of UGT1A9 by

Gefitinib at a low concentration, the effect was diminished by increasing the

concentration of Gefitinib. Therefore, Gefitinib and ATP seem to inhibit UGT1A9

through binding to a common site. No additive effect of ATP at the high

concentration of Gefitinib would be owing to the occupation of the binding site


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 28 -

with Gefitinib. On the other hand, the Gefitinib-induced inhibitory effect on

UGT1A9 was partially antagonized by 1 mM AMP, and the suppressive effect

remained even in the presence of high concentration Gefinitib. Although AMP at

a concentration more than 1 mM may completely eject Gefitinib from the

competitive binding site an alternative possibility that AMP exerts its antagonistic

effect against Gefinitib through an allosteric mechanism cannot be excluded. It is

considered that the ATP and AMP contents of cells largely depend on the

mitochondrial function which varies in response to a change in physiological

conditions. However, many more studies will be needed to fully understand how

disease conditions affect the ATP concentration within the ER, hormone

metabolism and drug-drug interaction with EGFR-protein tyrosine kinase

inhibitors. The three-dimensional structure of UGT1A9 has already been

simulated using the crystal structure of TDP-epi-vancosaminyltransferase (PDB

ID: 1PN3) as a template (Fujiwara et al, 2009). In their study, the prediction was

carried out using procedures reported by Ogata and Ueyama (2000). On the

contrary, in Phyre2 system used in this study, multiple templates described in the

Materials and Methods were used to predict the UGT1A9 structure. The

templates used were the enzymes utilizing UDP-sugar as the co-substrate


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 29 -

belonging to UDP-glycosyltransferase/glycogen phosphorylase superfamily and

a flavonoid 3-O-glucosyltransferase (crystal structure of UGT78G1 with binding

UDP as a ligand). Interestingly, we have predicted the binding site of UGT for

ADP in a 3D structural model, using an algorithm “3DLigandSite using similar

structures” and multiple (six) templates substituted for UGT (Kelly and Sternberg,

2009; Wass et al., 2010)(Supplemental Figure S2).

Further studies are needed to confirm which residues are relevant for

the modulation of UGT activity by adenine nucleotides.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 30 -

Authorship Contributions

Participated in research design: Ishii, An, Nishimura, Yamada

Conducted experiments: Ishii, An, Nishimura

Performed data analysis: Ishii, An, Nishimura

Wrote or contributed to the writing of the manuscript: Ishii, An, Yamada


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 31 -

References:

Bánhegyi G, Garzó T, Fulceri R, Benedetti A, and Mandl J (1993) Latency is the

major determinant of UDP-glucuronosyltransferase activity in isolated

hepatocytes. FEBS Lett 328: 149-152.

Basu NK, Kovarova M, Garza A, Kubota S, Saha T, Mitra PS, Banerjee R, Rivera

J and Owens IS (2005) Phosphorylation of a UDP-glucuronosyltransferase

regulates substrate specificity. Proc Natl Acad Sci U S A 102: 6285-6290.

Bock KW, Lilienblum W, Fischer G, Schirmer G, and Bock-Henning BS (1987)

The role of conjugation reactions in detoxication. Arch Toxicol 60: 22-29.

Brazier-Hicks M, Offen WA, Gershater MC, Revett TJ, Lim EK, Bowles DJ,

Davies GJ, and Edwards R (2007) Characterization and engineering of the

bifunctional N- and O-glucosyltransferase involved in xenobiotic metabolism

in plants. Proc Natl Acad Sci U S A 104: 20238-20243.

Clairmont CA, De Maio A and Hirschberg CB (1992) Translocation of ATP into

the lumen of rough endoplasmic reticulum-derived vesicles and its binding to

luminal proteins including BiP (GRP 78) and GRP 94. J Biol Chem 267:

3983-3990.

Dierks T, Volkmer J, Schlenstedt G, Jung C, Sandholzer U, Zachmann K,


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 32 -

Schlotterhose P, Neifer K, Schmidt B, and Zimmermann R (1996) A

microsomal ATP-binding protein involved in efficient protein transport into the

mammalian endoplasmic reticulum. EMBO J 15: 6931-6942.

Dorner AJ and Kaufman RJ (1994) The levels of endoplasmic reticulum proteins

and ATP affect folding and secretion of selective proteins. Biologicals 22:

103-112.

Dutton GJ (1980) Glucuronidation of drugs and other compounds. Boca Raton,

Florida, USA: CRC press.

Fujita K, Sugiyama M, Akiyama Y, Ando Y, and Sasaki Y (2011) The

small-molecular tyrosine kinase inhibitor nilotinib is a potent noncompetitive

inhibitor of the SN 38-glucuronidation by human UGT 1A1. Cancer

Chemother Pharmacol 67: 237-241.

Fujiwara R, Nakajima M, Yamamoto T, Nagao H, and Yokoi T (2009) In silico and

in vitro approaches to elucidate the thermal stability of human

UDP-glucuronosyltransferase (UGT) 1A9. Drug Metab Pharmacokinet 24:

235-244.

Guthapfel R, Guéguen P and Quéméneur E (1996) ATP binding and hydrolysis

by the multifunctional protein disulfide isomerase. J Biol Chem 271:


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 33 -

2663-2666.

Hardie DG, Carling D, and Carlson M (1998) The AMPactivated/SNF1 protein

kinase subfamily: metabolic sensors of the eukaryotic cell? Ann Rev

Biochem 67: 821-855.

Hardie DG and Hawley SA (2001) AMP-activated protein kinase: the energy

charge hypothesis revisited. Bioessays 23: 1112-1119.

Hanioka N, Jinno H, Tanaka-Kagawa T, Nishimura T and Ando M (2001)

Detection of UDP-glucuronosyltransferase UGT1A6 activity in human and rat

liver microsomes by HPLC with UV detection. J Pharm Biomed Anal 25:

65-75.

Hauser SC, Ziurys JC, and Gollan JL (1988) A membrane transporter mediates

access of uridine 5'-diphosphoglucuronic acid from the cytosol into the

endoplasmic reticulum of rat hepatocytes: implications for glucuronidation

reactions. Biochim Biophys Acta 967: 149-157.

Hewitt KN, Walker EA and Stewart PM (2005) Minireview: hexose-6-phosphate

dehydrogenase and redox control of 11β-hydroxysteroid dehydrogenase type

1 activity. Endocrinology 146: 2539-2543.

Kelley LA and Sternberg MJ (2009) Protein structure prediction on the Web: a


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 34 -

case study using the Phyre server. Nature Protoc 4: 363-371.

Kim SH, Shin SJ and Park JS (1996) Identification of the ATP transporter of rat

liver rough endoplasmic reticulum via photoaffinity labeling and partial

purification. Biochemistry 35: 5418-5425.

Li L, Modolo LV, Escamilla-Trevino LL, Achnine L, Dixon RA, and Wang X (2007)

Crystal structure of Medicago truncatula UGT85H2--insights into the

structural basis of a multifunctional (iso)flavonoid glycosyltransferase. J Mol

Biol 370: 951-963.

Liu Y, Ramírez J, House L, and Ratain MJ (2010) Comparison of the drug-drug

interactions potential of erlotinib and gefitinib via inhibition of

UDP-glucuronosyltransferases. Drug Metab Dispos 38: 32-39.

Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ (1951) Protein

measurement with the folin phenol reagent. J Biol Chem 193: 265-275.

Mackenzie PI, Bock KW, Burchell B, Guillemette C, Ikushiro S, Iyanagi T, Miners

JO, Owens IS, and Nebert DW (2005) Nomenclature update for the

mammalian UDP-glycosyltransferase (UGT) gene superfamily.

Pharmacogenet Genomics 15: 677−685.

Meech R and Mackenzie PI (1997) Structure and function of uridine diphosphate


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 35 -

glucuronosyltransferases. Clin Exp Pharmacol Physiol 24: 907-915.

Modolo LV, Li L, Pan H, Blount JW, Dixon RA, and Wang X (2009) Crystal

structures of glycosyltransferase UGT78G1 reveal the molecular basis for

glycosylation and deglycosylation of (iso)flavonoids. J Mol Biol 392:

1292-1302.

Mulichak AM, Lu W, Losey HC, Walsh CT, and Garavito RM (2004) Crystal

structure of vancosaminyltransferase GtfD from the vancomycin biosynthetic

pathway: interactions with acceptor and nucleotide ligands. Biochemistry 43:

5170-5180.

Muraoka M, Miki T, Ishida N, Hara T, and Kawakita M (2007) Variety of

nucleotide sugar transporters with respect to the interaction with nucleoside

mono- and diphosphates. J Biol Chem 282: 24615-24622.

Nishimura Y, Maeda S, Ikushiro S, Mackenzie PI, Ishii Y, and Yamada H (2007)

Inhibitory effects of adenine nucleotides and related substances on

UDP-glucuronosyltransferase: Structure-effect relationships and evidence for

an allosteric mechanism. Biochim Biophys Acta 1770: 1557-1566.

Ogata K and Umeyama H (2000) An automatic homology modeling method

consisting of database searches and simulated annealing. J Mol Graph


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 36 -

Model 18: 258–272, 305–306.

Offen W, Martinez-Fleites C, Yang M, Kiat-Lim E, Davis BG, Tarling CA, Ford CM,

Bowles DJ, and Davies GJ (2006) Structure of a flavonoid

glucosyltransferase reveals the basis for plant natural product modification.

EMBO J 25: 1396-1405.

Oguri K, Kurogi A, Yamabe K, Tanaka M, Yoshiue K, Ishii Y, and Yoshimura H

(1996) Purification of a phenobarbital-inducible UDP-glucuronosyltransferase

isoform from dog liver which catalyzes morphine and testosterone

glucuronidation. Arch Biochem Biophys 325: 159-166.

Piu I, Ramílez J, and Ratain MJ (2011) Inhibition of paracetamol glucuronidation

by tyrosine kinase inhibitors. Br J Clin Pharmacol 76: 917-920.

Puig JF and Tephly TR (1986) Isolation and purification of rat liver morphine

UDP-glucuronosyltransferase. Mol Pharmacol 30: 558-565.

Quéméneur E, Guthapfel R and Gueguen P (1994) A major phosphoprotein of

the endoplasmic reticulum is protein disulfide isomerase. J Biol Chem 269:

5485-5488.

Radominska-Pandya A, Ouzzine M, Fournel-Gigleux S, and Magdalou J (2005)

Structure of UDP-glucuronosyltransferases in membranes. Methods Enzymol


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 37 -

400: 116-147.

Ribelayga C and Mangel SC (2005) A circadian clock and light/dark adaptation

differentially regulate adenosine in the mammalian retina. J Neurosci 25:

215-222.

Ritter JK (2000) Roles of glucuronidation and UDP-glucuronosyltransferases in

xenobiotic bioactivation reactions. Chem-Biol Interact 29: 171-179.

Rosser MF, Trotta BM, Marshall MR, Berwin B and Nicchitta CV (2004)

Adenosine nucleotides and the regulation of GRP94-client protein

interactions. Biochemistry 43: 8835-8845.

Schwiebert EM (2000) Extracellular ATP-mediated propagation of Ca(2+) waves.

Focus on “mechanical strain-induced Ca(2+) waves are propagated via ATP

release and purinergic receptor activation”. Am J Cell Physiol 279: 281-283.

Shao H, He X, Achnine L, Blount JW, Dixon RA, and Wang X (2005) Crystal

structures of a multifunctional triterpene/flavonoid glycosyltransferase from

Medicago truncatula. Plant Cell 17: 3141-3154.

Shepherd SR, Baird SJ, Hallinan T, and Burchell B (1989) An investigation of the

transverse topology of bilirubin UDP-glucuronosyltransferase in rat hepatic

endoplasmic reticulum. Biochem J 259: 617-620.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 38 -

Taylor SI, Mukheriee C, and Jungas RL (1975) Regulation of pyruvate

dehydrogenase in isolated rat liver mitochondria. Effects of octanoate,

oxidation-reduction state, and adenosine triphosphate to adenosine

diphosphate ratio. J Biol Chem 250: 2028-2035.

Tukey RH and Strassburg CP (2000) Human UDP-glucuronosyltransferases:

metabolism, expression, and disease. Annu Rev Pharmacol Toxicol 40:

581-616.

Wass MN, Kelley LA, and Sternberg MJ (2010) 3DLigandSite: predicting

ligand-binding sites using similar structures. Nucleic Acids Res 38:

W469-473.

Winsnes A (1969) Studies on the activation in vitro of glucuronyltranferase.

Biochim Biophys Acta 191: 279-291.

Wearsch PA and Nicchitta CV (1997) Interaction of endoplasmic reticulum

chaperone GRP94 with peptide substrates is adenine

nucleotide-independent. J Biol Chem 272: 5152-5156.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 39 -

Footnotes

This work was supported in part by Grants-in-Aid for Scientific Research (C)

from the Japan Society for the Promotion of Science (JSPS) [Research

No.19590147], [No.21590164].

Address correspondence to: Yuji Ishii, Ph.D., Graduate School of

Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku,

Fukuoka 812-8582, Japan, E-mail: [email protected]


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 40 -

Legends for figures:

Fig. 1 Michaelis-Menten and Eadie-Hofstee plots for the inhibition by ATP

and NADP+ of 4-MU glucuronidation catalyzed by RLM pretreated

with Brij-58 or alamethicin: kinetics by varying 4-MU concentration.

Effect of ATP (A, E) and NADP+ (B, F) on 4-MU glucuronide formation in

Brij-58-treated microsomes was estimated at seven concentrations of

4-MU (10 to 1000 µM). The effect of ATP (C, G) and NADP+ (D, H) in

alamethicin-treated microsomes was also determined. The

UDP-GlcUA concentration was fixed at 2 mM.

Fig. 2. Antagonist effect of AMP on an ATP- and NADP+-evoked reduction

in 4-MU glucuronidation catalyzed by Brij-58- or alamethicin-treated

RLM. Effects of ATP (A, B) and NADP+ (C, D) (10 to 2500 μM) on the

4-MU glucuronide formation were assayed in the presence or absence

of AMP (closed circle, 0 μM; open square, 200 μM; closed triangle, 1000

μM). Microsomes were treated with Brij-58 (0.25 mg/mg protein) (A, C)

or alamethicin (0.05 mg/mg protein) (B, D). 4-MU and UDP-GlcUA

concentrations were fixed at 100 µM and 2 mM, respectively. Each plot


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 41 -

represents the average of a triplicate assay.

Fig. 3. Effect of the ratio of exogeneously added adenine nucleotides on

the UGT activity in detergent- or alamethicin-treated RLM. Effects of

the ratio of adenine nucleotides (ATP:ADP:AMP=100:10:1, 25:5:1, and

1:1:1) on 4-MU glucuronide formation were examined by varying the

total concentration of ATP, ADP and AMP (250, 500, 1000 and 1500 μM).

Microsomes were treated with Brij-58 (0.25 mg/mg protein) or

alamethicin (0.05 mg/mg protein). 4-MU and UDP-GlcUA concentrations

were fixed at 100 µM and 2 mM, respectively. Each bar represents the

mean ± S.E. of a triplicate assay. Significantly different from the activity

in the absence of adenine nucleotides (*, p<0.05; **, p<0.01; ***,

p<0.001).

Fig. 4. Effect of the mixture of inhibitory adenine-related compounds on

the UGT activity in Brij-58- or alamethicin-treated RLM. Effects of a

mixture of inhibitory compounds [ATP, ADP, adenine, NAD+, NADP+ and

CTP (Brij-58); and ATP, ADP, adenine, NAD+ and NADP+ (alamethicin)]


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 42 -

on 4-MU glucuronide formation were examined. The 'IC50' and 'IC50/2'

cocktails contain the inhibitory compounds described above at a IC50

concentration and half the IC50 concentration (see Table 1), respectively.

Microsomes were treated with Brij-58 (0.25 mg/mg protein) or

alamethicin (0.05 mg/mg protein). 4-MU and UDP-GlcUA concentrations

were fixed at 100 µM and 2 mM, respectively. Significantly different from

the activity in the absence of inhibitory compounds (**, p<0.01; ***,

p<0.001).

Fig. 5. Effects of Gefitinib on pooled human liver microsome- and human

UGT1A9 microsome-catalyzed 4-MU glucuronidation. Effects of

Gefitinib on 4-MU glucuronidation were assayed. Microsomal protein

(HLMs, 20 μg; UGT1A9 supersomes, 10 μg) was used in each assay.

The 4-MU concentration was fixed at 100 and 30 µM for HLM and

UGT1A9 supersomes, respectively. Each bar represents the mean ± S.E.

of triplicate assays. Abbreviation: HLMs, human liver microsomes.

Significantly different from the activity in the absence of Gefitinib (*,

p<0.05; **, p<0.01; ***, p<0.001).


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 43 -

Fig. 6. Effects of AMP and ATP on a Gefitinib-evoked reduction in 4-MU

glucuronidation catalyzed by humen UGT1A9. Effect of Gefitinib on

the 4-MU glucuronide formation was assayed in the presence of 500 μM

ATP (A) and 1 mM AMP (B). UGT1A9 supersomes were treated with

alamethicin (0.05 mg/mg protein). The 4-MU concentration was fixed at

30 µM. Each plot represents the average of triplicate assays.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 44 -

Table 1. IC50 value of nucletotides and adenine-related substances toward

4-MU UGT in RLM pretreated with either Brij-58 or alamethicin.

IC50 (μM)

Compounds Brij-58 Alamethicin

ATP 66.8±2.2 463±36 [6.9]

ADP 769±251 1258±54 [1.6]

Adenine <1000 1890±89 [-]

AMP >1000 >2000 [-]

Adenosine >1000 >2000 [-]

NADP+ 385±65 1079±65 [2.8]

NAD+ 334±116 1594±167 [4.8]

NADPH >1000 >2000 [-]

NADH >1000 >2000 [-]

CTP 853±47 >2000 [-]

GTP >1000 >2000 [-]

UTP >1000 >2000 [-]

RLM pretreated with Brij-58 (0.25 mg/mg protein) or alamethicin (0.05 mg/mg

protein) were used as the enzyme sources. 4-MU and UDP-GlcUA

concentrations were fixed at 100 µM and 2 mM, respectively. Each value

represents the estimated IC50 ± S.E. Value in bracket is the IC50 relative to that


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 45 -

determined using Brij-58-treated microsomes (=1.0). Control activities in the

absence of inhibitor was 11.2 ± 0.3 and 4.63 ± 0.16 for Brij-58- and

alamethicin-treated RLM, respectively.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 46 -

Table 2. Ki values of ATP and NADP+ in terms of 4-MU glucuronidation

activity in RLM.

Compounds Ki (μM)

Brij-58 Alamethicin

ATP 204 ± 17 635 ± 42

NADP+ 298 ± 30 607 ± 23

Each value represents the estimated Ki ± S.D. Details are described in Fig. 1


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 47 -

Table 3. Kinetic parameters of 4-MU glucuronidation catalyzed by RLM

pretreated either with Brij-58 or alamethicin: kinetics by varying the

UDP-GlcUA concentration

Activator

Km Vmax Vmax/Km

(μM) (nmol/min/mg protein)

Kinetics by varying 4-MUa

Brij-58 93.6 ± 0.3 26.3 ± 0.3 0.28

alamethicin 163 ± 9 8.32 ± 0.15 0.05

Kinetics by varying UDP-GlcUA b

Brij-58 348 ± 22 19.0 ± 0.3 0.05

Alamethicin 586 ± 49 4.73 ± 0.11 0.01

RLM pretreated with Brij-58 (0.25 mg/mg protein) and alamethicin (0.05 mg/mg

protein) were used as the enzyme source. a4-MU glucuronide formation was

estimated at ten concentrations of 4-MU (10 to 5000 µM). The concentration of

UDP-GlcUA was set at 2 mM in this assay. b4-MU glucuronide formation was

estimated at ten concentrations of UDP-GlcUA (50 to 7500 µM). The

concentration of 4-MU was set at 100 µM in this assay. Each value represents

the optimized value ± S.E.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 48 -

Table 4. Release of ATP from RLM by Brij-58- and alamethicin-treatment.

ATP (pmol)

Pellet Sup. Total [%]

Intact 36.8 ± 1.0 4.83 ± 0.73 41.6 [100]

Brij-58 N. D. 13.4 ± 0.1 13.4 [32.2]

Alamethicin 21.1 ± 0.5 11.5 ± 2.0 32.6 [78.4]

Each value represents the mean ± S.D. of three determinations. Value in bracket

is the percentage of the sum of the ATP recovered in the pellet and sup. relative

to that determined using intact microsomes (=100). Abbreviation: sup.,

supernatant; N.D., not detected.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 49 -

Table 5. IC50 of adenine nucleotides for 4-MU glucuronidation catalyzed

by pooled HLM and UGT1A9.

compounds

HLM UGT1A9

IC50 (µM) Ratio IC50 (µM)

Brij-58 alamethicin (Brij-58=1.0) alamethicin

ATP 33.8±5.2 935±162 27.7 491±22

ADP >1000 >2000 - >1000

AMP >1000 >2000 - >1000

Microsomes (20 μg) were pretreated with Brij-58 (0.25 mg/mg protein) or

alamethicin (0.05 mg/mg protein). Values represent the mean ± S.E. of

triplicate assays. ‘Ratio’ is the IC50 relative to that determined using

Brij-58-treated microsomes (=1.0). For UGT1A9 supersomes, microsomal

protein (10 µg) was pretreated with alamethicin (0.05 mg/mg protein).

Values represent the mean ± S.E. of triplicate assays.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


DMD #46862

- 50 -

Table 6. IC50 of Gefitinib for 4-MU glucuronidation catalyzed by pooled

HLM and human UGT1A9 microsomes.

IC50 (μM)

HLMs (Brij-58-treated) >250

HLMs (alamethicin-treated) >250

UGT1A9 (alamethicin-treated) 19.4±6.6

Human microsomal protein (20 μg) was pretreated with either Brij-58 (0.25

mg/protein) or alamethicin (0.05 mg/mg protein). UGT1A9 microsomes (10 μg

protein) were pretreated with alamethicin (0.05 mg/mg protein). Values

represent the mean ±S.E. of triplicate assays.


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


A ATP (Brij-58)

B NADP+ (Brij-58)

D NADP+ (alamethicin)

C ATP (alamethicin)

E ATP (Brij-58)

F NADP+ (Brij-58)

H NADP+ (alamethicin)

G ATP (alamethicin)

Figure 1


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


ATP (μM)

Rel

ativ

e ac

tivi

ty(%

of

con

tro

l)

0 200 400 600 800 1000

0

20

40

60

80

100

120

1000 μM

250 μM

0 μM

NADP+ (μM)

Rel

ativ

e ac

tivi

ty(%

of

con

tro

l)

0 500 1000 1500 2000

0

20

40

60

80

100

120

1000 μM

250 μM0 μM

NADP+ (μM)

Rel

ativ

e ac

tivi

ty(%

of

con

tro

l)

0 500 1000 1500 2000 2500

0

20

40

60

80

100

120

1000 μM

250 μM

0 μM

ATP (μM)

Rel

ativ

e ac

tivit

y(%

of

con

tro

l)

0 500 1000 1500 2000

0

20

40

60

80

100

120

1000 μM

250 μM

0 μM

A

B

C

D

Figure 2

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 July 30, 2012 as DO

I: 10.1124/dmd.112.046862

at ASPET Journals on December 18, 2020 dmd.aspetjournals.org Downloaded from


Brij-58-treated microsomes

Alamethicin-treated microsomes

Figure 3

*** ***

*** *** *** ***

*** ***

**

**

**

******

***

******

**

******

******


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


14

12

10

8

6

4

2

0control IC50/2 IC50

4-M

UG

fo

rmed

(nm

ol/m

in/m

g p

rote

in)

5

4

3

2

1

0control IC50/2 IC50

4-M

UG

fo

rmed

(nm

ol/m

in/m

g p

rote

in)

Brij-58-trated microsomes

Alamethicin-trated microsomes

Figure 4

***

***

***

**


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


Relative activity

(% of control)

120

100

80

60

40

20

0

0 2.5 25 250

Gefitinib (μM)

*

*

**

120

100

80

60

40

20

0

Relative activity

(% of control)

0 2.5 25 250

Gefitinib (μM)

**

120

100

80

60

40

20

0

Relative activity

(% of control)

0 2.5 25 250

Gefitinib (μM)

*

*

**

Alamethicin-treated HLM

Brij-58-treated HLM

Alamethicin-treated UGT1A9

Figure 5


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


Gefitinib +ATP 500 μM

Gefitinib (μM)

Rel

ativ

e ac

tivi

ty(%

of

con

tro

l)

0 100 200 300

0

20

40

60

80

100

Gefitinib

Gefitinib (μM)

Rel

ativ

e ac

tivi

ty(%

of

con

tro

l)

0 100 200 300

0

20

40

60

80

100

GefitinibGefitinib +AMP 1000 μM

Figure 6


at ASPE

T Journals on D

ecember 18, 2020


Dow

nloaded from


ATP Serves as an Endogenous Inhibitor of UDP...

Documents

Transcript of ATP Serves as an Endogenous Inhibitor of UDP...