Download Human carboxylesterase 1: from drug metabolism to drug discovery

620
Biochemical Society Transactions (2003) Volume 31, part 3
Human carboxylesterase 1: from drug metabolism
to drug discovery
M.R. Redinbo*†1 , S. Bencharit*‡ and P.M. Potter§
*Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, U.S.A., †Department of Biochemistry and Biophysics and the
Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, U.S.A., ‡School of Dentistry, University of North
Carolina at Chapel Hill, Chapel Hill, NC 27599, U.S.A., and §Department of Molecular Pharmacology, St Jude Children’s Research Hospital, Memphis,
TN 38105, U.S.A.
Abstract
Human carboxylesterase 1 (hCE1) is a serine esterase involved in both drug metabolism and activation,
as well as other biological processes. hCE1 catalyses the hydrolysis of heroin and cocaine, and the
transesterification of cocaine in the presence of ethanol to the toxic metabolite cocaethylene. We have
determined the crystal structures of hCE1 in complex with either the cocaine analogue homatropine or
the heroin analogue naloxone. These are the first structures of a human carboxylesterase, and they provide
details about narcotic metabolism in humans. hCE1’s active site contains rigid and flexible pockets, explaining
the enzyme’s ability to act both specifically and promiscuously. hCE1 has also been reported to contain
cholesteryl ester hydrolase, fatty acyl-CoA hydrolase and acyl-CoA:cholesterol acyltransferase activities, and
thus appears to be involved in cholesterol metabolism. Since the enzyme may be useful as a treatment
for cocaine overdose, and may afford protection against chemical weapons like Sarin, Soman and VX gas,
hCE1 could serve as both a drug and a drug target. Selective hCE1 inhibitors targeted to several sites on the
enzyme may also pave the way for novel clinical tools to manage cholesterol homoeostasis in humans.
Introduction
Human carboxylesterase 1 (hCE1) is a broad-spectrum serine
hydrolase involved in drug and xenobiotic metabolism,
as well as several additional biological processes. hCE1
is primarily expressed in the liver, with lesser amounts
in the intestine, kidney, lung, testes, heart, monocytes
and macrophages [1,2]. The enzyme shares 47% sequence
identity with human intestinal carboxylesterase (also called
human carboxylesterase 2; hCE2), which exhibits a distinct
expression profile and a preferred set of substrates [3]. hCE1’s
catalytic triad, composed of Ser-221, Glu-353 and His-467,
employs a two-step hydrolysis mechanism. In the first step,
a covalent acyl-enzyme intermediate is formed with the
catalytic serine residue, and the alcohol product is released.
In the second step, a water molecule attacks the acyl-enzyme
linkage, releasing the acyl product [1]. Compounds other than
water can attack the acyl-enzyme intermediate, leading to
transesterification rather than hydrolysis.
The primary biological role of hCE1 appears to be
xenobiotic metabolism. The enzyme is able to cleave ester,
amide and thioester linkages in a wide variety of chemically
distinct compounds [1]. Several clinical drugs are metabolized
by hCE1, including cocaine, heroin, meperidine, demerol
Key words: cholesterol metabolism, cocaine, heroin, human carboxylesterase 1 inhibitor, X-ray
crystallography.
Abbreviations used: hCE1, human carboxylesterase 1; CEH, cholesteryl ester hydrolase; ACAT,
acyl-CoA:cholesterol acyltransferase; ER, endoplasmic reticulum; CRP, C-reactive protein; rCE,
rabbit carboxylesterase; AcChE, acetylcholinesterase.
1
To whom correspondence should be addressed, at the Department of Chemistry, University
of North Carolina at Chapel Hill (e-mail [email protected]).
C 2003
Biochemical Society
and lidocaine [4–6], and the enzyme processes pro-drugs
like lovastatin to their active forms in vivo [7]. hCE1
cleaves the methyl ester linkage on R-cocaine to generate benzoylecgonine, the primary urinary metabolite of
cocaine detected in humans (Figure 1a). hCE1 is also the
only enzyme known to generate the toxic cocaine metabolite
cocaethylene, formed when cocaine and alcohol are abused
together [8]. hCE1 efficiently cleaves the 3-acetyl linkage,
and, to a lesser extent, the 6-acetyl linkage on heroin to
generate 6-monoacetylmorphine and morphine respectively
[9]. Single-site mutations in the active site of hCE1 also
convert the enzyme into an efficient organophosphate
hydrolase capable of detoxifying potent nerve agents. The
United States military is developing variant forms of hCE1
into a prophylactic treatment for personnel at risk of exposure
to chemical weapons such as Sarin, Tabun and VX gas [10].
The second biological role of hCE1 appears to be the
processing of cholesterol and fatty acids in the liver and
peripheral tissues (Figure 1b). Although there are many
enzymes involved in cholesterol metabolism and trafficking,
hCE1 has been shown to be the bile-salt-independent
cholesteryl ester hydrolase (CEH) found in monocytes,
macrophages and hepatocytes [11], and to also have fatty
acyl-CoA hydrolase activity [12]. In addition, hCE1 has been
reported to act as an acyl-CoA:cholesterol acyl-transferase
(ACAT), capable of transferring a fatty acid from an acylCoA to cholesterol [13]. Thus hCE1 appears to be able to
catalyse both the creation and the elimination of cholesteryl
esters using transesterification and hydrolysis reactions
respectively. Overexpression of hCE1 in vitro induces the
Structural Biology in Drug Metabolism and Drug Discovery
efflux of cholesterol from the cell [14]. These ester transferase
and hydrolase activities are crucial to the transportation of
cholesterol substrates within, and among, cells and tissues
[11,15]. hCE1 and its homologues in other mammals are also
essential for testosterone synthesis [16], retinol metabolism
[15] and the creation of fatty acid ethyl esters (FAEEs), the
toxic metabolites associated with chronic alcohol abuse [12].
The third biological role of hCE1 is the trafficking
and retention of proteins in the endoplasmic reticulum
(ER), where hCE1 is modified by N-linked glycosylation
sites. While in the ER, hCE1 has been shown to bind
directly to the C-reactive protein (CRP) and to hold this
small protein before its release into the plasma [17]. CRP
is secreted in response to tissue injury, and is the most
sensitive marker identified to date for the development
of atherosclerosis [18]. Reduction of hCE1’s affinity for
CRP is accountable for the release of CRP during acute
injury, at levels up to 1000-fold higher than normal [19].
In addition, hCE1 (also called egasyn) binds to the phase II
drug metabolism β-glucuronidase enzymes in the ER. The βglucuronidase enzymes generate glucuronic acid conjugates
of xenobiotic and endobiotic compounds, and mark them
for elimination by efflux pumps [20]. It appears that
organophosphate compounds are capable of inducing the
release of β-glucuronidase from the ER by disrupting the
β-glucuronidase–hCE1 complex [21].
We recently reported the crystal structure of a rabbit carboxylesterase (rCE), the first structure of a mammalian
carboxylesterase [22]. rCE shares 81% sequence identity with
hCE1 [23]. We found that the monomeric rCE enzyme is
an α/β-hydrolase related in structure to the acetylcholinesterases (AcChEs), but with a large and conformable active
Figure 1 For legend, see facing page
(a)
C 2003
Biochemical Society
621
622
Biochemical Society Transactions (2003) Volume 31, part 3
Figure 1 Roles of hCE1 in cocaine and heroin metabolism (a), and roles of hCE1 in cholesterol and fatty acid metabolism (b)
(b)
site. rCE is the most efficient enzyme identified to date in
the conversion of the anticancer drug CPT-11 into its active
metabolite, SN-38, a member of the camptothecin family
of topoisomerase I poisons. A leaving group of CPT-11
activation, 4-piperidino-piperidine, binds in our structure not
at the active site of rCE, but at a surface site separated from
the active one by a thin wall of amino acid side chains. We
termed this the ‘side door’ and proposed that it may shuttle
small substrates or products into or out of the catalytic site of
the enzyme [22]. The side door is related in function to the
‘back door’ in the AcChEs, which was proposed to facilitate the release of small products from the enzyme’s active
site. The side door in rCE is structurally distinct from the
back door in AcChE, however, and is located in a discrete region of the active site. The side door in mammalian
CEs may play an important role in the varied catalytic actions
of these enzymes, as discussed below.
Figure 2 Surface representation of the hCE1 trimer, viewed from
above, relative to the active-site cavities of the enzyme
The substrate-binding gorge, the ‘side door’ and the ‘Z site’ are indicated
in each monomer with white, red and blue arrows respectively.
Heroin and cocaine metabolism by hCE1
We have determined the crystal structure of hCE1 in complex
with the cocaine analogue homatropine to 2.8 Å resolution
(1 Å = 0.1 nm), and with the heroin analogue naloxone to
2.9 Å resolution (Figure 2) [24]. These are the first structures
of a human carboxylesterase or a human enzyme bound to
analogues of cocaine or heroin, and they provide explicit
details about narcotic metabolism in humans. hCE1 shares
C 2003
Biochemical Society
a 1.0 Å root-mean-square deviation over 455 equivalent Cα
positions with rCE, but exhibits several distinct structural
features. hCE1’s oligomerization state is in equilibrium
Structural Biology in Drug Metabolism and Drug Discovery
between a trimer with C3 symmetry and a stacked hexamer
with 32-point group symmetry. The active sites of the enzyme
are buried in the hexamer, although the side door regions are
freely exposed and face out into solvent, whether the enzyme
is a trimer or a hexamer. Whereas two 17-amino-acid loops
proposed to cover the active site in rCE were disordered and
not observed in that structure, these loops are well ordered
in the hCE1 structures and close over the catalytic gorge
of the enzyme. The binding of ligands to a surface Z-site
on hCE1 shifts the trimer–hexamer equilibrium to trimer.
Thus this surface position likely functions as an allosteric
binding site, as has been proposed for a related mammalian
carboxylesterase [25].
The active site of hCE1 contains a small, rigid pocket
and a large, flexible region [24]. The small ester linkages of
cocaine and heroin pack into the small, rigid pocket, whereas
the large, flexible pocket accommodates the structurally
distinct tropine and morphine rings of cocaine and heroin
respectively. These structural results, elucidated from the
crystal structures of hCE1 with the cocaine analogue
homatropine and the heroin analogue naloxone, explain how
hCE1 is able to act both promiscuously and specifically in
narcotic metabolism. The enzyme specifically aligns only the
methyl ester linkage of cocaine into the small, rigid pocket.
The benzoyl ester linkage or the inactive S-stereoisomer of
cocaine would clash sterically with the enzyme, and would
not align productively for catalysis. In the structure of
hCE1 with naloxone, however, the heroin analogue binds
in one orientation in seven of the twelve monomers in the
asymmetric unit, and in a second orientation in the remaining
five. In the first orientation, the site of the 3-acetyl linkage
on heroin is aligned for hydrolysis, whereas in the second
orientation, the 6-acetyl linkage is aligned. This explains
hCE1’s ability to act on both the 3- and 6-positions on heroin.
The large, flexible pocket of hCE1’s active site is lined with
non-polar residues and allows the structurally distinct tropine
ring of homatropine, and the morphine ring of naloxone,
to dock in approximately the same place. Thus hCE1 uses
a combination of a small, rigid pocket and a large, flexible
pocket to achieve both specificity and non-specificity in drug
metabolism [24].
hCE1 and drug discovery
hCE1 may be useful as both a drug and a drug target. hCE1
converts cocaine into benzoylecgonine, the primary urinary metabolite of this narcotic [8,9]. The enzyme’s ability
to form a stable hexamer is likely to enhance its ability to
circulate stably in human serum [24]. Thus hCE1 may be
an effective new treatment for cocaine overdose. hCE1
has several advantages over the bacterial cocaine esterase [26] and catalytic antibodies [27] that were proposed
previously for this purpose. In particular, hCE1 possesses
an efficient catalytic activity, and is unlikely to elicit a
potent immune response. The United States military is also
developing hCE1 as a prophylactic for the potential exposure
to chemical weapons [10]. Single-point mutations in the
active-site region of hCE1 converts the enzyme into an
efficient organophosphate hydrolase capable of detoxifying
Sarin, Soman, Tabun and VX gas. Both civilian and military
personnel at risk of exposure to such compounds could
be injected with purified forms of hCE1, which is likely
to afford protection for periods of days to weeks. In
addition, hCE1 may be useful in the detoxification of
equipment and buildings after exposure to organophosphate
toxins.
As described above, hCE1 appears to play an important
role in cholesterol trafficking and metabolism. Although the
details of hCE1’s contribution to the creation and elimination
of cholesteryl esters in particular tissues are not known,
it is clear that hCE1 joins a small number of enzymes
capable of contributing to these processes. The combined
action of hCE1 in both hydrolysing cholesteryl esters and
in transesterifying cholesterol to create cholesteryl esters
raises a significant structural and mechanistic question. A
similar question arises when one considers hCE1’s ability
to transesterify cocaine with ethanol to create cocaethylene,
a long-lived toxic metabolite detected when cocaine and
alcohol are abused together [8]. When the active site of hCE1
is occupied by either naloxone or homatropine in our crystal
structures, there is little room for compounds larger than
water to enter the active site via the top of the catalytic gorge.
We have proposed that for hCE1 to be able to transesterify
cocaine to cocaethylene, ethanol must gain access to the acylenzyme intermediate by passing through the side door of the
enzyme [24].
A similar mechanism may exist when hCE1 acts as a
CEH, a fatty acyl CoA hydrolase, an ACAT and in FAEE
synthesis. The 16–18 carbon fatty acyl tail of the cholesteryl
ester, for example, may extend out from the active site of
the enzyme through the side door, allowing the ester linkage
of the substrate to align for catalysis. Indeed, it has been
shown that a single-point mutation in a rat homologue of
hCE1 converts this enzyme from a poor into an efficient
CEH [28]. Based on amino acid sequence alignment, this
point mutation impacts one of the four residues that creates
the side door in hCE1. Thus both the active site and the side
door of hCE1 may serve as targets for drug discovery and
design to regulate hCE1’s action in cholesterol trafficking
and metabolism. Indeed, we have recently identified acridinebased compounds that selectively inhibit hCE1, but not other
hCEs (hiCE, hCE2), and appear to act at the active site of the
enzyme [29]. These compounds may pave the way towards
unravelling the biological roles of hCE1 in cholesterol and
xenobiotic metabolism in human tissues, and provide novel
avenues for drug discovery and development.
Supported by National Institutes of Health (NIH) grants CA77340,
DK62229 and CA98468, a Burroughs Wellcome Career Award in the
Biomedical Sciences (M.R.R), and by the NIH, including Core Grant
CA21765 and the American Lebanese Syrian Associated Charities
(P.M.P.).
C 2003
Biochemical Society
623
624
Biochemical Society Transactions (2003) Volume 31, part 3
References
1 Satoh, T. and Hosokawa, M. (1998) Annu. Rev. Pharmacol. Toxicol. 38,
257–288
2 Satoh, T., Taylor, P., Bosron, W.F., Sanghani, S.P., Hosokawa, M. and
La Du, B.N. (2002) Drug Metab. Dispos. 30, 488–493
3 Schwer, H., Langmann, T., Daig, R., Becker, A., Aslanidis, C. and
Schmitz, G. (1997) Biochem. Biophys. Res. Commun. 233, 117–220
4 Kamendulis, L.M., Brzezinski, M.R., Pindel, E.V., Bosron, W.F. and
Dean, R.A. (1996) J. Pharmacol. Exp. Ther. 279, 713–717
5 Zhang, J., Burnell, J.C., Dumaual, N. and Bosron, W.F. (1999) J. Pharmacol.
Exp. Ther. 290, 314–318
6 Alexson, S.E., Diczfalusy, M., Halldin, M. and Swedmark, S. (2002)
Drug Metab. Dispos. 30, 643–647
7 Tang, B.K. and Kalow, W. (1995) Eur. J. Clin. Pharmacol. 47, 449–451
8 Brzezinski, M.R., Abraham, T.R., Stone, C.L., Dean, R.A. and Bosron, W.F.
(1994) Biochem. Pharmacol. 48, 1747–1755
9 Brzezinski, M.R., Spink, B.J., Dean, R.A., Berkman, C.E., Cashman, J.R. and
Bosron, W.F. (1997) Drug Metab. Dispos. 25, 1089–1096
10 Broomfield, C.A. and Kirby, S.D. (2001) J. Appl. Toxicol. (suppl.), 21,
S43–S46
11 Ghosh, S. (2000) Physiol. Genomics 2, 1–8
12 Diczfalusy, M.A., Bjorkhem, I., Einarsson, C., Hillebrant, C.G. and
Alexson, S.E. (2001) J. Lipid Res. 42, 1025–1032
13 Becker, A., Bottcher, A., Lackner, K.J., Fehringer, P., Notka, F., Aslanidis, C.
and Schmitz, G. (1994) Arterioscler. Thromb. 14, 1346–1355
14 Ghosh, S. and Natarajan, R. (2001) Biochem. Biophys. Res. Commun.
284, 1065–1070
15 Harrison, E.H. (1998) Annu. Rev. Nutr. 18, 259–276
C 2003
Biochemical Society
16 Mikhailov, A.T. and Torrado, M. (2000) Front. Biosci. 5, E53–E62
17 Yue, C.C., Muller-Greven, J., Dailey, P., Lozanski, G., Anderson, V. and
Macintyre, S. (1996) J. Biol. Chem. 271, 22245–22250
18 Labarrere, C.A., Lee, J.B., Nelson, D.R., Al-Hassani, M., Miller, S.J. and
Pitts, D.E. (2002) Lancet 360, 1462–1467
19 Macintyre, S., Samols, D. and Dailey, P. (1994) J. Biol. Chem. 269,
24496–24503
20 Islam, M.R., Waheed, A., Shah, G.N., Tomatsu, S. and Sly, W.S. (1999)
Arch. Biochem. Biophys. 372, 53–61
21 Satoh, T. and Hosokawa, M. (2000) Neurotoxicology 21, 223–227
22 Bencharit, S., Morton, C.L., Howard-Williams, E.L., Danks, M.K.,
Potter, P.M. and Redinbo, M.R. (2002) Nat. Struct. Biol. 9, 337–342
23 Danks, M.K., Morton, C.L., Krull, E.J., Cheshire, P.J., Richmond, L.B.,
Naeve, C.W. and Pawlik, C.A. (1999) Clin. Cancer Res. 5, 917–924
24 Bencharit, S., Morton, C.L., Xue, Y., Potter, P.M. and Redinbo, M.R. (2003)
Nat. Struct. Biol., doi 10.1038/nsb919
25 Suzuki-Kurasaki, M., Yoshioka, T. and Uematsu, T. (1997) Biochem. J.
325, 155–161
26 Larsen, N.A., Turner, J.M., Stevens, J., Rosser, S.J., Basran, A., Lerner, R.A.,
Bruce, N.C. and Wilson, I.A. (2002) Nat. Struct. Biol. 9, 17–21
27 Larsen, N.A., Zhou, B., Heine, A., Wirsching, P., Janda, K.D. and Wilson, I.A.
(2001) J. Mol. Biol. 311, 9–15
28 Wallace, T.J., Kodsi, E.M., Langston, T.B., Gergis, M.R. and Grogan, W.M.
(2001) J. Biol. Chem. 276, 33165–33174
29 Bencharit, S., Morton, C.L., Hyatt, J.L., Kuhn, P., Danks, M.K., Potter, P.M.,
and Redinbo, M.R. (2003) Chem. Biol., in the press
Received 23 January 2003