Identification of human UGT2B7 as the major isoform involved in the O-glucuronidation of chloramphenicol

Article abstract of DOI:10.1124/dmd.109.029900

Chloramphenicol (CP), a broad spectrum antibiotic, is eliminated in humans by glucuronidation. The primary UGT enzymes responsible for CP O-glucuronidation remain unidentified. We have previously identified the 3-O-CP (major) and 1-O-CP (minor) glucuronides by β-glucuronidase hydrolysis, liquid chromatography-tandem mass spectrometry, and 1D/2D H NMR. Reaction phenotyping for the glucuronidation of CP with 12 expressed human liver UGT isoforms has identified UGT2B7 as having the highest activity for 3-O- and 1-O-CP glucuronidation with minor contributions from UGT1A6 and UGT1A9. The kinetics of CP 3-O-glucuronidation by pooled human liver microsomes (HLMs) exhibited biphasic Michaelis-Menten kinetics with the apparent high-affinity K _m1 and low-affinity K_m2 values of 46.0 and 1027 μM, whereas expressed UGT2B7 exhibited Michaelis-Menten kinetics with the apparent K_m value of 109.1 μM. The formation of 1-O-CP glucuronide by pooled HLM and expressed UGT2B7 exhibited substrate inhibition kinetics with apparent K_m values of 408.2 and 115.0 μM, respectively. Azidothymidine (AZT) and hyodeoxycholic acid (substrates of UGT2B7) inhibited 3-O- and 1-O-CP glucuronidation in pooled HLMs. In 10 donor HLM preparations, both CP 3-O- and CP 1-O-glucuronidation showed a significant correlation with AZT glucuronidation (UGT2B7) (r_s = 0.85 and r_s = 0.83, respectively) at 30 μMCP, whereas no significant correlation was observed between CP 3-O-glucuronidation and serotonin glucuronidation (UGT1A6) or propofol glucuronidation (UGT1A9) at this CP concentration. These results suggest that UGT2B7 is the primary human hepatic UDP-glucuronosyltransferase isoform catalyzing 3-O- and 1-O-CP glucuronidation with minor contributions from UGT1A6 and UGT1A9. Copyright

Full text of DOI:10.1124/dmd.109.029900

0090-9556/10/3803-368–375$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
DMD 38:368–375, 2010
Vol. 38, No. 3
29900/3567562
Printed in U.S.A.
Identification of Human UGT2B7 as the Major Isoform Involved in
the O-Glucuronidation of Chloramphenicol
Mei Chen, Barbara LeDuc, Stephen Kerr, David Howe, and David A. Williams
Massachusetts College of Pharmacy and Health Sciences, Boston, Massachusetts (M.C., B.L., S.K., D.A.W.); and Vertex
Pharmaceuticals Incorporated, Cambridge, Massachusetts (D.H.)
Received August 25, 2009; accepted December 10, 2009
ABSTRACT:
Chloramphenicol (CP), a broad spectrum antibiotic, is eliminated in 1-O-CP glucuronide by pooled HLM and expressed UGT2B7 exhibited
humans by glucuronidation. The primary UGT enzymes responsible substrate inhibition kinetics with apparent Km values of 408.2 and
for CP O-glucuronidation remain unidentified. We have previously 115.0 ␮M, respectively. Azidothymidine (AZT) and hyodeoxycholic
identified the 3-O-CP (major) and 1-O-CP (minor) glucuronides by acid (substrates of UGT2B7) inhibited 3-O- and 1-O-CP glucuronida-
␤-glucuronidase hydrolysis, liquid chromatography-tandem mass tion in pooled HLMs. In 10 donor HLM preparations, both CP 3-O- and
spectrometry, and 1D/2D H NMR. Reaction phenotyping for the glu- CP 1-O-glucuronidation showed a significant correlation with AZT
curonidation of CP with 12 expressed human liver UGT isoforms has glucuronidation (UGT2B7) (rs ؍ 0.85 and rs ؍ 0.83, respectively) at 30
identified UGT2B7 as having the highest activity for 3-O- and 1-O-CP ␮M CP, whereas no significant correlation was observed between CP
glucuronidation with minor contributions from UGT1A6 and UGT1A9. 3-O-glucuronidation and serotonin glucuronidation (UGT1A6) or
The kinetics of CP 3-O-glucuronidation by pooled human liver micro- propofol glucuronidation (UGT1A9) at this CP concentration. These
somes (HLMs) exhibited biphasic Michaelis-Menten kinetics with the results suggest that UGT2B7 is the primary human hepatic UDP-
apparent high-affinity Km1 and low-affinity Km2 values of 46.0 and glucuronosyltransferase isoform catalyzing 3-O- and 1-O-CP glucu-
1027 ␮M, whereas expressed UGT2B7 exhibited Michaelis-Menten
kinetics with the apparent Km value of 109.1 ␮M. The formation of
ronidation with minor contributions from UGT1A6 and UGT1A9.
Chloramphenicol (CP) (Fig. 1) is a broad-spectrum antibiotic that pharmacokinetic studies in these neonates confirmed an elevation of
has been used to treat serious infections in humans, often when other
antibiotics have failed. However, the use of this antibiotic is limited
because of its potentially life-threatening side effects, such as aplastic
anemia in adults or Gray Baby syndrome in premature and newborn
infants (Suhrland and Weisberger, 1963). Despite these dangers, CP is
still widely used in developing countries due to the increased cost and
decreased availability of newer antibiotics (Brook, 2004).
Although CP has a complex metabolic pattern, glucuronidation has
been recognized as the major pathway for adult human elimination with
up to 85% of CP appearing in the urine as its glucuronide (Glazko et al.,
1950; Glazko, 1966). Nonetheless, the UDP-glucuronosyltransferase
(UGT) isoforms responsible for this glucuronidation remain unidentified.
CP is rapidly absorbed with an oral bioavailability of 80 to 90%, indi-
cating little presystemic glucuronidation, and with 5 to 15% of CP doses
excreted unchanged in the urine. The drug is distributed into all tissues
including the brain and spinal cord. Its elimination half-life of 4 to 6 h
suggests that no enterohepatic cycling occurs.
plasma CP levels, suggesting that the parent drug was responsible for
the toxic effects of the Gray Baby syndrome (Weiss et al., 1960;
Kaufmann et al., 1981). These clinical findings illustrate the impor-
tance of understanding the impact of age-dependent developmental
differences on the activity of drug-metabolizing liver enzymes and
drug clearance, which can markedly alter the pharmacokinetics of a
drug in infants compared with adults (Kaufmann et al., 1981; Kearns
et al., 2003). Because the mechanisms underlying these age-dependent
developmental differences are largely unknown, the identification of
the main human liver UGT isoforms involved in drug clearance is
crucial to understanding the age-related variations in plasma drug
concentration that lead to altered drug efficacy and/or toxicity in
infants, which may necessitate an age-based dose adjustment.
Glucuronidation is inefficient at birth (Kaufmann et al., 1981;
Kearns et al., 2003). As a result of low UGT isoform activity at birth,
the clearance half-life for CP in newborns Ͻ3 months postnatal is
ϳ10 h, at 6 months 4 to 6 h, and at 4 years of age 4 to 6 h (Miles,
1983). The reduced glucuronidation and thus slow elimination of CP
has been cited as being responsible for the delayed systemic clearance
and subsequent toxicity of this drug in newborns (Lambdin et al.,
1960; Weiss et al., 1960; Suhrland et al., 1963).
The incidence of toxic effects of CP in neonates was found to be
strongly related to age, because 9 of 10 babies who showed toxic
effects were less than 9 days old (Kaufmann et al., 1981). Subsequent
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
Despite the widespread clinical use of CP for over five decades, the
role of the liver in human CP O-glucuronidation has not been char-
doi:10.1124/dmd.109.029900.
ABBREVIATIONS: CP, chloramphenicol; UGT, UDP-glucuronosyltransferase; HLM, human liver microsome; UDPGA, UDP-glucuronic acid; AZT,
azidothymidine; HPLC, high-performance liquid chromatography; LC-MS/MS, liquid chromatography-tandem mass spectrometry; MRM, multiple
reaction monitoring; CLint, intrinsic clearance; CLmax, maximum clearance.
368

O-GLUCURONIDATION OF CHLORAMPHENICOL
369
FIG. 1. Chloramphenicol and its major 3-O-glucuronide
and minor 1-O-glucuronides.
Chloramphenicol O-Glucuronidation by Pooled HLMs. CP (300 ␮M)
was incubated with alamethicin-activated pooled HLMs fortified with 20 mM
UDPGA as described previously (Chen et al., 2007). The CP 3-O- and CP
1-O-glucuronide concentrations in the incubation mixture were determined by
liquid chromatography-tandem mass spectrometry (LC-MS/MS) using biosyn-
thesized CP 3-O-glucuronide as the standard (Chen et al., 2007). In brief, the
quantitation of the CP glucuronides was achieved by using a LC-MS/MS
system, which was equipped with an Agilent model 1100 series HPLC system
and a Sciex API 4000 tandem mass spectrometer (Applied Biosystems, Foster
City, CA). Liquid chromatography separation was performed on a Synergi
acterized. Furthermore, the UGT isoforms contributing to CP O-
glucuronide formation have yet to be identified. We have previously
identified the CP 3-O- (major) and CP 1-O-glucuronide (minor)
conjugates (Chen et al., 2007). The determination of the specific
human UGT isoforms responsible for the O-glucuronidation of CP
could provide evidence into the reasons for the newborn toxicity
of CP.
Therefore, in this study we sought to gain insight into the in vivo
hepatic CP clearance by 1) phenotyping the specific UGT isoforms
involved in the in vitro CP 3-O- and 1-O-glucuronidation by ex- Polar-RP column (Phenomenex, Torrance, CA) with a gradient consisting of
mobile phase A water/formic acid (100:0.1, v/v) and mobile phase B aceto-
nitrile/formic acid (100:0.1, v/v) at a flow rate 0.5 ml/min. Negative multiple
reaction monitoring (MRM) mode was used to monitor CP, CP glucuronides,
and warfarin (internal standard) at mass transitions m/z of 321.2/152.1 for CP,
497.2/193.1 for CP glucuronides, and 307.1/250.1 for warfarin (Chen et.al.,
2007). Incubations without UDPGA or with boiled pooled HLMs were per-
formed simultaneously and served as negative controls. Zero-time, zero-
protein, and zero-substrate incubations served as blanks.
Chloramphenicol O-Glucuronidation by Expressed Human UGT Iso-
forms. CP 3-O- and 1-O-glucuronidation were measured in alamethicin-
activated Supersomes expressing human UGT1A1, UGT1A3, UGT1A4,
UGT1A6, UGT1A7, UGT1A8, UGT1A9, UGT1A10, UGT2B4, UGT2B7,
UGT2B15 (all 5 mg of protein/ml), or UGT2B17 (0.5 mg of protein/ml)
Supersomes with 30 or 300 ␮M CP and 20 mM UDPGA. UGT control
Supersomes (0.5 mg of protein/ml) served as the negative control. Incubations
were done in duplicate, and samples were directly analyzed by the described
LC-MS/MS method for pooled HLMs. All of the incubations were performed
at 37°C for 50 min.
Enzyme Kinetic Data Analysis. All data points represent the mean of
triplicate estimations. The kinetic parameters for CP 3-O- and 1-O-glucu-
ronidation were calculated by fitting the untransformed experimental data to
the following enzyme kinetic equations using Prism 5.2 (GraphPad Software
Inc., San Diego, CA), designed for nonlinear regression analysis. The selection
of the “best-fit” kinetic model was based on comparison of the sum-of-squared
residuals, SD of fit, coefficient of determination (r2), and F test (Prism 5.2).
The Michaelis-Menten equation (eq. 1),
pressed human UGTs, 2) determining the kinetics of CP 3-O- and
1-O-glucuronidation in pooled human liver microsomes (HLMs) and
by expressed human UGTs, and 3) confirming the contribution of the
UGTs for CP O-glucuronidation from inhibition studies with UGT-
selective substrates as inhibitors and correlation studies in donor
human HLMs. We now report our results from phenotyping with
expressed UGTs and from inhibition and correlation studies, which
indicate that UGT2B7 is the primary UGT isoform catalyzing CP
3-O- and 1-O-glucuronidation.
Materials and Methods
Chemicals and Reagents. Unless otherwise indicated, reagents including
CP, Tris-HCl, UDP-glucuronic acid (UDPGA), magnesium chloride, ala-
methicin, dimethyl sulfoxide, azidothymidine (AZT), serotonin, propofol,
and hyodeoxycholic acid were purchased from Sigma-Aldrich (St. Louis,
MO). AZT glucuronide sodium salt was purchased from Toronto Research
Chemicals Inc. (North York, ON, Canada). High-performance liquid chro-
matography (HPLC) grade acetonitrile and methanol were obtained from
Mallinckrodt Baker, Inc. (Phillipsburg, NJ). Pooled mixed gender HLM
(n ϭ 20) and expressed UGTs are microsomes from baculovirus-infected
insect cells expressing human UGT1A1, UGT1A3, UGT1A4, UGT1A6,
UGT1A7, UGT1A8, UGT1A9, UGT1A10, UGT2B4, UGT2B7, UGT2B15,
UGT2B17 (Supersomes), and UGT control Supersomes were purchased
from BD Gentest (Woburn, MA). The expressed UGT2A enzymes as well
as UGT1A5, UGT2B10, UGT2B11, and UGT2B28 were not available from
BD Gentest and thus were not tested. The pooled HLMs and expressed
enzymes were kept frozen at Ϫ80°C and used shortly after receiving the
enzymes. The protein content of all Supersomes was 5 mg/ml. Human liver
v ϭ ͑Vmax ϫ ͓S͔͒/͑Km ϩ ͓S͔͒
(1)
where v is the rate of metabolite formation, Vmax is the maximum velocity
microsomes from 10 individual donors were purchased from BD Gentest. (as pmol product/min/mg microsomal or cell lysate protein), Km is the
All other chemicals were of reagent grade or of the highest purity available
commercially.
Michaelis constant (substrate concentration at 0.5 Vmax), and [S] is the
substrate concentration.

370
CHEN ET AL.
The two-enzyme Michaelis-Menten equation (eq. 2),
v ϭ ͕͑Vmax1 ϫ ͓S͔͒/͑Km1 ϩ ͓S͔͖͒ ϩ ͕͑Vmax2 ϫ ͓S͔͒/͑Km2 ϩ ͓S͔͖͒
were analyzed by the previously described LC-MS/MS method for the glucu-
ronidation of CP. The AZT glucuronide was analyzed with a standard curve of
AZT glucuronide (Toronto Research Chemicals Inc.) spiked in the incubation
matrix by LC-MS/MS in positive ionization MRM mode at mass transitions of
m/z 444/268. The concentrations of serotonin, propofol, and their glucuronides
in the incubation mixture were determined using the method as described by
Yamanaka (2005) and analyzed by the described LC-MS/MS method with
serotonin and propofol standard curves. Serotonin and propofol glucuronides
were monitored in negative MRM mode at mass transitions of m/z 351/175 and
m/z 353/177, respectively.
Correlation analyses between these CP-UGT activities and the other UGT
activities in 10 individual donor HLMs were determined by Spearman’s rank
method. When the rs value was greater than or equal to 0.5 and the P value was
less than 0.05, the correlations were considered statistically significant.
(2)
where Km1 and Km2 are the apparent high- and low-affinity components and
Vmax1 and Vmax2 are the apparent maximum velocities for the high- and
low-affinity components, was used when the Eadie-Hofstee plot exhibited a
biphasic shape (Houston and Kenworthy, 2000; Court et al., 2001; Yamanaka
et al., 2005).
The two-site model (eq. 3) (Houston and Kenworthy, 2000; Uchaipichat et
al., 2008),
v/Vmax ϭ ͑S/Ks ϩ ␤S2/␣Ks2͒/͑1 ϩ 2S/Ks ϩ S2/␣Ks2͒
(3)
where Ks represents the binding affinity and ␣ and ␤ are modifying factors that
reflect changes in Ks and Kp (the catalytic rate constant), respectively.
The intrinsic clearance (CLint) was calculated as CLint ϭ Vmax/Km, and
expressed as l/min/mg protein. Enzyme activities were calculated by dividing
metabolite concentration by protein concentration and incubation time and
expressed as pmol/min/mg protein.
Results
Chloramphenicol O-Glucuronidation by Expressed Human UGTs.
UGT1A1, UGT1A3, UGT1A4, UGT1A6, UGT1A7, UGT1A8,
UGT1A9, UGT1A10, UGT2B4, UGT2B7, UGT2B15, and UGT2B17
were screened for CP 3-O- and 1-O-glucuronidation activity at 30 ␮M
CP (Ogilvie et al., 2008), the approximate therapeutic plasma con-
centration (National Toxicology Program, 2002), and at 300 ␮M CP,
the approximate toxic plasma concentration (National Toxicology
Program, 2002) (Fig. 2). CP 3-O-glucuronidation was observed for
UGT1A6, UGT1A9, UGT2B4, and UGT2B7 with activities Ͼ1 pmol/
min/mg at 30 ␮M CP concentration. At 300 ␮M CP concentration, the
additional isoforms UGT1A3 and UGT2B17 were observed to cata-
lyze the 3-O-glucuronidation of CP (Fig. 2A). The isoforms UGT1A1,
UGT1A4, UGT1A7, UGT1A8, UGT1A10, and UGT2B15 did not
form the CP 3-O-glucuronide. Kinetic studies were not performed
with UGT2B4 and UGT2B17 because these enzymes have exhibited
low or negligible activity toward drugs and other xenobiotics (Miners
et al., 2006). Thus, CP 3-O-glucuronidation kinetics were character-
ized for UGT1A6, UGT1A9, and UGT2B7. As shown in Fig. 2, A and
B, UGT2B7 showed the highest CP 3-O- and 1-O-glucuronosyltrans-
ferase activities at 30 and 300 ␮M concentrations of CP. Negligible
activities for the formation of CP 1-O-glucuronide were observed for
all of the other expressed UGT isoforms.
When the Eadie-Hofstee plot exhibited a sigmoidal “U” shaped curve, the
Hill equation (eq. 4) was selected for estimating S50 and Vmax (Weiss, 1997):
v ϭ Vmax ϫ Sn/͑S50n ϩ Sn͒
(4)
where S50 is the substrate concentration resulting in 50% Vmax (analogous to
Km in the Michaelis-Menten equation) and n is the Hill coefficient. The
maximum clearance (CLmax) [CLmax ϭ Vmax ϫ (n Ϫ1)/{S50 ϫ n(n Ϫ 1)1/n}]
provides an estimate of the highest clearance attained, i.e., when the enzyme is
fully activated before saturation occurs (Houston and Kenworthy, 2000).
When the Michaelis-Menten plot showed a hyperbolic shape with no clearly
defined plateau at high substrate concentrations and the Eadie-Hofstee plot
showed a “hook” curve at high substrate concentrations, the uncompetitive
substrate inhibition model (eq. 5) was used (Houston and Kenworthy, 2000;
Venkatakrishnan et al., 2001).
v ϭ Vmax ϫ S/͑Km ϩ S ϫ ͑1 ϩ S/Ksi͒
(5)
where Ksi is the constant describing the substrate inhibition interaction.
Inhibition of Chloramphenicol O-Glucuronidation by AZT, Serotonin,
and Propofol. The 3-O- and 1-O-glucuronidation of CP in pooled HLMs was
studied using known selective UGT isoform probe substrates as inhibitors
(Miners et al., 2004; Miners et al., 2006). Of the expressed UGT isoforms
glucuronidating CP at the 3-hydroxyl position, expressed UGT1A6, UGT1A9,
and UGT2B7 exhibited significant activity and were therefore used for the
inhibition studies. Because only expressed UGT2B7 exhibited activity for
glucuronidating CP at the 1-hydroxyl group, it was used for the inhibition
The Michaelis-Menten model for the formation of the CP 3-O-
glucuronide by expressed UGT2B7 (Fig. 3A) was best fitted to the
experimental data (Fig. 3A, Eadie-Hofstee plot insert), and the derived
kinetic parameters were Km 109.1 Ϯ 5.2 ␮M and Vmax 46.9 Ϯ 0.4
studies. The selective inhibitors were as follows: for UGT1A6, serotonin pmol/min/mg protein and are summarized in Table 1. The apparent
(Krishnaswamy et al., 2003; Court, 2005); for UGT1A9, propofol (Court et al.,
2001; Yamanaka et al., 2005); and for UGT2B7, hyodeoxycholic acid (Pillot
et al. 1993; Barre et al., 2007) and AZT (Court et al., 2003). Serotonin and
AZT were each dissolved in water, hyodeoxycholic acid was dissolved in 1:1
acetonitrile and water, and propofol was dissolved in dimethyl sulfoxide. The
concentration ranges were as follows: for hyodeoxycholic acid or propofol, 0.1
to 1000 ␮M; for serotonin, 100 to 40,000 ␮M; and for AZT, 0.1 to 5000 ␮M.
All the incubations of inhibitors were carried out as described above for CP
O-glucuronidation in pooled HLM. The 3-O-CP- and 1-O-CP-UGT activities
at 100 ␮M CP in pooled HLMs in the absence of any probe substrate as
CLint for the formation of 3-O-CP glucuronide by UGT2B7 is 0.43
l/min/mg protein. The sigmoidal model (Hill equation; Weiss, 1997)
for the UGT1A9 catalyzed CP 3-O-glucuronidation was best fitted to
the experimental data (Fig. 4A, Eadie-Hofstee plot inset), and the
derived kinetic parameters for S50 were 714.4 Ϯ 30.2 ␮M and 41.9 Ϯ
0.6 pmol/min/mg expressed protein, with a positive Hill coefficient of
1.4, suggestive of positive cooperativity kinetics (Houston and Ken-
worthy, 2000), and are summarized in Table 1. The apparent CLmax
for the formation of 3-O-CP glucuronide by UGT1A9 is 0.059
inhibitor were used as controls. The uninhibited glucuronidation activities were l/min/mg protein. Likewise, for UGT1A6-catalyzed CP 3-O-
calculated as a percentage of control activity.
glucuronidation, the Hill equation was best fitted to the experimental
data (Fig. 4B, Eadie-Hofstee plot inset), and the derived kinetic
parameters for S50 were 1557.0 Ϯ 122.0 ␮M, Vmax 12.9 Ϯ 0.4
pmol/min/mg protein, and Hill coefficient of 1.2, with an apparent
CLmax of 0.0053 l/min/mg protein, and summarized in Table 1.
On the other hand, 1-O-CP glucuronidation by UGT2B7 exhibited
atypical kinetics characteristic of substrate inhibition when initially
fitted to the Michaelis-Menten equation (Fig. 3B, Eadie-Hofstee plot
inset) (Houston and Kenworthy, 2000). Thus, the substrate inhibition
Correlation Analysis. The 3-O- and 1-O-glucuronides produced from 30 or
3000 ␮M CP concentrations were measured in a bank of HLMs isolated from
10 individual donors, using the same incubation conditions as previously
described for CP O-glucuronidation by pooled HLM. Other UGT activities
measured included propofol glucuronidation (50 ␮M propofol; UGT1A9 sub-
strate), AZT glucuronidation (2000 ␮M AZT; UGT2B7 substrate), and sero-
tonin glucuronidation (4000 ␮M serotonin; UGT1A6 substrate). The concen-
trations of these probe substrates approximated their Km values in pooled
HLMs, which were 1331 ␮M for AZT, 90 ␮M for propofol, and 4854 ␮M for
serotonin (data not shown). AZT, propofol, serotonin, and their glucuronides equation was best fitted to the experimental data, and the derived

O-GLUCURONIDATION OF CHLORAMPHENICOL
371
FIG. 2. CP-3-O-glucuronidation (A) and CP 1-O-glucu-
ronidation (B) at 30 and 300 ␮M CP by a panel of 12
expressed human UGT isoforms in microsomes from bac-
ulovirus-infected insect cells. Each bar represents the mean
of duplicate determinations.
kinetic parameters by UGT2B7 were for Km 115.0 Ϯ 14.2 ␮M and expressed UGT2B7, Ͼ75% inhibition was observed for CP 3-O- and
Vmax 5.4 Ϯ 0.2 pmol/min/mg expressed protein and are summarized 1-O-glucuronidation (Fig. 6A). Hyodeoxycholic acid at 1000 ␮M
in Table 1. The apparent CLint value was 0.047 l/min/mg protein.
exhibited Ͼ90% inhibition for both CP 3-O-glucuroniude and CP
Kinetics of Chloramphenicol O-Glucuronidation by Pooled Hu- 1-O-glucuronide formation in pooled HLMs (Fig. 6B). For the other
man Liver Microsomes. CP 3-O-glucuronidation by pooled HLM inhibitors, propofol inhibited CP 3-O-glucuronide formation ϳ60%,
exhibited atypical kinetics was characteristic of biphasic kinetics (Fig. 5A, but Ͻ20% inhibition for CP 1-O-glucuronide formation, whereas
Eadie-Hofstee plot inset) (Houston and Kenworthy, 2000). Therefore, serotonin exhibited insignificant inhibition (Ͻ5%) for both glucu-
the glucuronidation data were analyzed using the two-enzyme ronides by pooled HLMs (data not shown). The control activities of
Michaelis-Menten equation (eq. 2), the two-site model equation (eq. the UGTs for the formation of 3-O-CP and 1-O-CP glucuronides at
3), or the Hill equation (eq. 4). The two-enzyme Michaelis-Menten 100 ␮M CP in pooled HLMs were determined to be 25.2 pmol/
equation was best fitted to the experimental data, and the derived min/mg and 4.0 pmol/min/mg protein.
kinetic parameters for the high-affinity components (Km1 and Vmax1
were 46.0 Ϯ 27.0 ␮M and 37.9 Ϯ 13.0 pmol/min/mg protein, respec- experiments were conducted with pooled alamethicin-activated HLM
tively, whereas those for the low-affinity components (Km2 and Vmax2 from 10 human donor livers at 30 and 3000 ␮M CP concentrations. As
)
Correlation Study in Human Liver Microsomes. Correlation
)
were 1027.0 Ϯ 129.8 ␮M and 232.0 Ϯ 11.2 pmol/min/mg protein, shown in Fig. 7, CP 3-O- and 1-O-glucuronidation in the individual
respectively, and are summarized in Table 1. The apparent intrinsic donor HLMs were significantly correlated with AZT glucuronidation
clearance CLint1 for the high-affinity component of CP 3-O-glucu- activity at 30 and 3000 ␮M (Fig. 7, A–D) (rs ϭ 0.85, 0.83 and rs ϭ
ronidation in pooled HLM was 0.82 l/min/mg protein.
0.65, 0.79, respectively) and summarized in Table 2, but they were not
CP 1-O-glucuronidation by pooled HLM also exhibited substrate in- correlated with propofol or serotonin glucuronidation activity at sub-
hibition kinetics (Fig. 5B, Eadie-Hofstee plot inset) (Hutzler and Tracy, strate concentration of 30 ␮M (rs Ͻ 0.5), except for propofol glucu-
2002). The substrate inhibition equation was fitted to the experimental ronidation activity at 3000 ␮M CP (rs ϭ 0.71) (Table 2). These results
data, and the derived kinetic constants for Km were 408.2 Ϯ 41.4 ␮M and indicate that UGT2B7 is the primary source of 3-O- and 1-O-glucu-
Vmax 16.7 Ϯ 0.6 pmol/min/mg expressed protein (Table 1). The apparent ronides with a lesser contribution by UGT1A9 at CP concentrations
CLint value was 0.041 l/min/mg protein.
Ͼ30 ␮M.
Inhibition of Chloramphenicol O-Glucuronidation by AZT,
Hyodeoxycholic Acid, Serotonin, and Propofol. At 5000 ␮M, AZT
inhibited the formation of 3-O-CP-glucuronide by approximately 30%
Discussion
We have previously identified CP 3-O- and 1-O-glucuronides gen-
and 76% of 1-O-CP glucuronide production by pooled HLM at 100 erated by pooled HLMs as the only glucuronide metabolites in hu-
␮M CP (Fig. 6A). When this inhibition study was repeated with mans (Chen et al., 2007). In this study, we have identified the main

372
CHEN ET AL.
FIG. 4. Kinetics of CP-3-O-glucuronidation activity by expressed human UGT1A9
(A) and UGT1A6 (B) in microsomes from baculovirus-infected insect cells repre-
sent fitting the data to the Hill equation, and the Eadie-Hofstee plot is shown as an
inset. The CP-3-O-glucuronidation activity was determined as described under
Materials and Methods. The kinetic parameters are summarized in Table 1. Each
data point represents the mean of triplicate determinations Ϯ S.D.
FIG. 3. Kinetics of CP-3-O-glucuronidation (A) and CP-1-O-glucuronidation (B)
by expressed human UGT2B7 in microsomes from baculovirus-infected insect cells.
For CP-3-O-glucuronidation (A), the solid line represents fitting the data to the
single-enzyme Michaelis-Menten equation and the Eadie-Hofstee plot is shown as
an inset. For 1-O-CP glucuronidation (B), the solid line represents fitting the data to
the substrate inhibition model and the Eadie-Hofstee plot is shown as an inset. The
CP-3-O- and 1-O-glucuronidation activity was determined as described under
Materials and Methods. The kinetic parameters are summarized in Table 1. Each
data point represents the mean of triplicate determinations Ϯ S.D.
trations, UGT1A6, UGT1A9, and UGT2B7 are expressed in the liver
and may potentially contribute to its hepatic clearance. However, the
low activities for UGT1A6, UGT1A9, and UGT2B4 suggest that
involvement of these isoforms is likely to be minor at best. UGT2B7
was the only isoform catalyzing CP 1-O-glucuronidation at either CP
concentration.
UGT isoforms responsible for the CP 3-O- and 1-O-glucuronidations
by HLM. Phenotyping with expressed UGT isozymes showed the
highest activity for UGT2B7 as the primary UGT isoform catalyzing
CP 3-O-glucuronidation with lesser contributions from UGT1A6,
UGT1A9, and other UGT isoforms at 30 and 300 ␮M CP. Of the
The kinetic profiles for CP 3-O-glucuronidation by expressed
isoforms capable of metabolizing CP at physiological plasma concen- UGT2B7 are typical of the single-enzyme Michaelis-Menten
TABLE 1
Best-fit-derived kinetic parameters for CP O-glucuronidation by HLM and baculovirus-expressed human UGT isoforms
Each value represents best-fit values Ϯ S.D.
1-O-CP Glucuronide Kinetics Ϯ S.D. (n ϭ 3)
3-O-CP Glucuronide Kinetics Ϯ S.D. (n ϭ 3)
UGT Isoform
c
c
c
e
b
Km
Vmax
CLint
Km or S50
Vmax
CLint1 or CLmax
Km
Vmax
CLint2
␮M
pmol/min/
mg protein
l/min/mg
protein
␮M
pmol/min/mg
protein
l/min/mg protein
␮M
pmol/min/mg
protein
l/min/mg
protein
Pooled HLMd
2B7
1A6
408.2 Ϯ 41.4
115.0 Ϯ 14.2
16.7 Ϯ 0.6
5.4 Ϯ 0.2
ϽLOQ
0.041
0.047
46.0 Ϯ 27.0d
109.1 Ϯ 5.2
37.9 Ϯ 13.0d
46.9 Ϯ 0.4
12.9 Ϯ 0.3e
41.9 Ϯ 0.6e
0.82d
1027 Ϯ 129.8d
232 Ϯ 11.2d
0.226d
0.43a
1557.0 Ϯ 122.1e
714.4 Ϯ 30.2e
0.008b
0.059b
1A9
ϽLOQ
ϽLOQ, below limit of quantitation.
a CLint calculated from Michaelis-Menten kinetics.
b CLmax calculated from Hill equation fitted kinetics.
c Best-fit kinetic parameters for 1-O-CP glucuronide calculated with substrate inhibition equation Ksi (HLM) was 28500 Ϯ 8200 ␮M and Ksi (UGT2B7) was 8090 Ϯ 1240 ␮M.
d Best-fit Km1 and Km2 and Vmax1 and Vmax2 kinetic parameters calculated for 3-O-CP glucuronide with two-enzyme Michaelis-Menten equation.
e Best-fit kinetic parameters for 3-O-CP glucuronide calculated using Hill equation.

O-GLUCURONIDATION OF CHLORAMPHENICOL
373
FIG. 5. Kinetics of CP-3-O-glucuronidation (A) and CP-1-O-glucuronidation (B) in
pooled HLMs. For CP-3-O-glucuronidation (A), the solid line represents fitting the
data to the two-enzyme Michaelis-Menten equation, and the Eadie-Hofstee plot is
shown as an inset. For CP-1-O-glucuronidation (B), the solid line represents fitting
the data to the substrate inhibition model and the Eadie-Hofstee plot is shown as an
inset. The CP-3-O- and 1-O-glucuronidation activity was determined as described
under Materials and Methods. The kinetic parameters are summarized in Table 1.
Each data point represents the mean of triplicate determinations Ϯ S.D.
FIG. 6. The inhibitory effects of AZT (A) and hyodeoxycholic acid (B) on the CP
3-O- and 1-O-glucuronidation activity were determined in pooled HLMs at 100 ␮M
CP. Each data point represents the mean of duplicate determinations.
model, whereas expressed UGT1A6 and UGT1A9 are characteris-
tic of the sigmoidal (Hill) model. CP 1-O-glucuronidation by
expressed UGT2B7 displayed substrate inhibition, and the kinetics
were derived using the substrate inhibition equation. The CLint for
1-O-CP glucuronide in pooled HLMs was similar to that for
UGT2B7, suggesting that this isoform was the only contributor.
The 10- to 20-fold greater CLint value for the CP 3-O-glucuronide
than that for the 1-O-CP glucuronide by pooled HLM and ex-
pressed UGT2B7 implies that the 3-OH site of CP is glucu-
ronidated more readily than the more sterically hindered benzylic
OH (Fig. 1). The Km values for CP glucuronidation by expressed
UGT1A6 and UGT1A9 were helpful for confirming the minor
involvement of these isoforms in CP O-glucuronidation.
The kinetics for the formation of CP 3-O-glucuronide by pooled
HLM were characteristic for the two-enzyme Michaelis-Menten
model with high- and low-affinity components. The mean Km value
for the high-affinity component was 20-fold lower than the Km for the
low-affinity reaction, and the apparent Clint difference was 4-fold.
Comparisons of the enzyme kinetic parameters between pooled HLMs
and expressed UGT2B7 were made to further determine the relative
contribution of UGT2B7 to CP 3-O-glucuronidation. A substrate that
is glucuronidated solely by a single expressed UGT isoform should
apparent high-affinity Km1 value (ϳ46 ␮M) was similar to the appar-
ent Km for UGT2B7 (ϳ109 ␮M), which suggests that in human liver,
UGT2B7 may be the predominant UGT isoform catalyzing CP 3-O-
glucuronidation. When the mean Km and Vmax values for the high- and
low-affinity components of CP 3-O-glucuronidation by HLM were
substituted into the two-enzyme Michaelis-Menten equation, the high-
affinity component (UGT2B7) is responsible for ϳ90% of CP 3-O-
glucuronidation activity at 30 ␮M and ϳ18% of activity at 3000 ␮M
CP, and as a result it contributes significantly to the higher CLint value
by pooled HLM at low CP concentrations. Hence, CP was 3-O-
glucuronidated in pooled HLM much more rapidly by UGT2B7 than
by either UGT1A6 or UGT1A9, the CLint of which were 10- to
100-fold less, consistent with their minor role in the clearance of CP
at therapeutic plasma concentrations. It is likely that the low-affinity
component of hepatic CP 3-O-glucuronidation, which comprises the
minor UGT isoform activities, are also collectively contributing to CP
3-O-glucuronidation at CP concentrations Ͼ30 ␮M and would there-
fore be expected to make minor contributions (Ͻ10%) to CP clear-
ance in patients receiving this drug. Furthermore, UGT1A6 and
UGT1A9 have S50 values of similar order to the low-affinity compo-
have an apparent Km similar to the apparent Km for pooled HLM nent of HLM CP 3-O-glucuronidation. It should be noted that CP
(Court et al., 2003). Based on the results of the present study, the exhibits minor nonspecific binding to pooled HLMs (M. Chen, un-

374
CHEN ET AL.
FIG. 7. The correlation between 3-O-CP (A and B) and 1-O-CP glucuronosyltransferase (C and D) activities in a bank of HLMs from 10 individual donors at 30 and 3000
␮M CP and AZT glucuronosyltransferase activities at 2000 ␮M AZT. Each data point represents the mean of duplicate determinations.
TABLE 2
Correlation of chloramphenicol glucuronosyltransferase activity with the glucuronosyltransferase activities of marker substrates
3-O-CP UGT Activity
1-O-CP UGT Activity
Marker Substrate Activity
CP ϭ 30 ␮M
CP ϭ 3000 ␮M
CP ϭ 30 ␮M
CP ϭ 3000 ␮M
rs
P
rs
P
rs
P
rs
P
AZT glucuronidation
Propofol glucuronidation
Serotonin glucuronidation
0.85
0.44
Ϫ0.01
0.000a
0.180
0.97
0.65
0.71
0.31
0.038a
0.018a
0.36
0.83
0.16
Ϫ0.10
0.000a
0.63
0.76
0.79
0.31
Ϫ0.079
0.0038a
0.36
0.81
rs, Spearman rank order correlation coefficient.
a The correlations were considered statistically significant when the rs value was greater than or equal to 0.50 and the P value was less than 0.05.
published data), precluding nonspecific microsomal binding as a (Tsoutsikos et al., 2004; Rowland et al., 2007), resulting in a de-
cause of atypical kinetics. CP 1-O-glucuronidation in both pooled creased inhibition of CP 3-O-glucuronidation by the unsaturated fatty
HLM and UGT2B7 exhibited substrate inhibition kinetics (Hutzler acids. Because of the complexity of the 1-O-glucuronidation substrate
and Tracy, 2002), which implies binding of more than one substrate inhibition kinetics of CP by UGT2B7, its inhibition kinetics were not
molecule in the active site.
studied further (Tsoutsikos et al., 2004). The 10- to 20-fold greater
The coincubation of CP with 5000 ␮M AZT in pooled HLM was CLint for UGT2B7-catalyzed O-glucuronidation in pooled HLMs and
observed as 30 and 76% concentration-dependent inhibition for the expressed UGT2B7 of the primary 3-hydroxyl group than the more
3-O- and 1-O-glucuronidation of CP, respectively. We attribute the sterically hindered 1-hydroxyl group may also be contributing factors
partial inhibition observed for CP 3-O-glucuronidation to its inhibi- to the observed differences in the inhibition kinetics between CP 3-O-
tion by the long-chain unsaturated fatty acids (oleic acid, linoleic acid, and 1-O-glucuronidation. When CP is coincubated with hyodeoxy-
and arachidonic acid) present (as phospholipids) in pooled HLMs, cholic acid in pooled HLM, Ͼ90% inhibition was shown at 1000 ␮M
which when released from the microsomal membrane during the hyodeoxycholic acid for both the 3-O- and 1-O-glucuronidation of
course of an incubation, act as potent inhibitors of hepatic UGT2B7 CP. Coincubation of CP with propofol demonstrated moderate inhi-
(Ki 0.2–2 ␮M), competing with AZT (Km 1331 ␮M) for the inhibition bition (ϳ60%) of CP 3-O-glucuronidation and Ͻ20% for CP 1-O-
of hepatic UGT2B7-catalyzed CP 3-O-glucuronidation (Tsoutsikos et glucuronidation in pooled HLM and no inhibition (Ͻ5%) for seroto-
al., 2004; Rowland et al., 2007). However, when CP was coincubated nin. These results help to support UGT2B7 as the primary UGT
with AZT and expressed UGT2B7 under the same incubation condi- isoform catalyzing both CP 3-O- and 1-O-glucuronide formations in
tions as for HLMs, Ͼ75% inhibition was observed for both CP 3-O- pooled HLMs and UGT1A9 as having a minor role in the O-glucu-
and 1-O-glucuronidation, consistent with the lower content of inhib- ronidation of CP.
itory unsaturated long-chain fatty acids in the expressed UGT2B7
To assess the contribution of UGT2B7 to CP 3-O- and 1-O-

O-GLUCURONIDATION OF CHLORAMPHENICOL
375
Postoperative pain in the neonate: age-related differences in morphine requirements and
metabolism. Intensive Care Med 29:2009–2015.
Brook I (2004) Management of anaerobic infection. Expert Rev Anti Infect Ther 2:153–158.
Chen M, Howe D, Leduc B, Kerr S, and Williams DA (2007) Identification and characterization
of two CP glucuronides from the in vitro glucuronidation of CP in human liver microsomes.
Xenobiotica 37:954–971.
Coffman BL, King CD, Rios GR, and Tephly TR (1998) The glucuronidation of opioids, other
xenobiotics, and androgens by human UGT2B7Y(268) and UGT2B7H(268). Drug Metab
Dispos 26:73–77.
Court MH, Duan SX, von Moltke LL, Greenblatt DJ, Patten CJ, Miners JO, and Mackenzie PI
(2001) Interindividual variability in acetaminophen glucuronidation by human liver micro-
somes: identification of relevant acetaminophen UDP-glucuronosyltransferase isoforms.
J Pharmacol Exp Ther 299:998–1006.
Court MH, Krishnaswamy S, Hao Q, Duan SX, Patten CJ, Von Moltke LL, and Greenblatt DJ
(2003) Evaluation of 3Ј-azido-3Ј-deoxythymidine, morphine, and codeine as probe substrates
for UDP-glucuronosyltransferase 2B7 (UGT2B7) in human liver microsomes: specificity and
influence of the UGT2B7*2 polymorphism. Drug Metab Dispos 31:1125–1133.
Court MH (2005) Isoform-selective probe substrates for in vitro studies of human UDP-
glucuronosyltransferases. Methods Enzymol 400:104–116.
Glazko AJ, Dill WA, and Rebstock MC (1950) Biochemical studies on chloramphenical
(chloromycetin). III. Isolation and identification of metabolic products in urine. J Biol Chem
183:679–691.
glucuronidation, a correlation analysis was carried out between CP
3-O-glucuronidation versus AZT (UGT2B7), propofol (UGT1A9), or
serotonin (UGT1A6) glucuronidation by 10 donor HLMs. A signifi-
cant correlation occurred between CP 3-O- and 1-O-glucuronidation
and AZT glucuronidation at 30 and 3000 ␮M CP, but only a signif-
icant correlation for propofol (UGT1A9) occurred at 3000 ␮M CP
concentration. No correlation was observed for serotonin (UGT1A6)
at these CP concentrations. These correlation results help to confirm
that UGT2B7 is the likely isoform catalyzing the hepatic 3-O- and
1-O-glucuronidation of CP at therapeutic concentrations.
When CP was incubated with pooled human kidney microsomes,
the apparent CLint for CP 3-O-glucuronide was an order of magnitude
less than pooled HLMs (data not reported), indicating the contribution
of renal 3-O-glucuronidation to systemic CP clearance in the neonate
is much less than hepatic 3-O-glucuronidation.
Glazko AJ (1966) Identification of CP metabolites and some factors affecting metabolic dispo-
In summary, our data from incubation of CP with human expressed
UGT isoforms, inhibition studies with UGT-selective substrates, and
correlation studies clearly indicate that CP was efficiently O-glucu-
ronidated to its major CP 3-O-glucuronide by pooled HLMs and
expressed UGT2B7, and to a lesser extent by UGT1A9, confirming
that hepatic 3-O-glucuronidation by UGT2B7 plays a major role in the
systemic clearance of CP at therapeutic plasma concentrations. As a
result of low UGT isoform activity in infants Ͻ6 months postnatal,
clinical studies have shown that hepatic glucuronidation undergoes
significant changes during neonatal development requiring age-re-
lated changes in drug therapy and dosages as a result of delayed
systemic clearance of the drug. For example, the clearance of AZT by
UGT2B7 is approximately 50% lower in neonates than found in
children 14 to 99 days of age (Boucher et al., 1993), whereas the
clearance of morphine by UGT2B7 in postoperative infants was
significantly reduced during the neonatal period and in infants and
children under the age of two years, which required an adjustment in
the intravenous morphine dosage (Bouwmeester et al., 2003). The
results from our study indicate that the clearance of CP primarily by
UGT2B7 suggests that the systemic clearance of CP is likely to be
minimal in neonates and infants Ͻ6 months postnatal, which is
consistent with the argument that reduced O-glucuronidation of CP is
responsible for its delayed clearance and subsequent toxicity in new-
borns. Available evidence suggests that UGT2B7 polymorphism has
little effect on drug glucuronidation (Coffman et al., 1998; Bhasker et
al., 2000), but an influence on CP elimination cannot be discounted in
this age group. This study highlights the fact that the identification of
the liver UGT isoform(s) involved in drug clearance is crucial to
understanding the age-dependent variations in plasma drug concen-
tration that lead to altered drug efficacy and/or toxicity in infants,
which might require an age-based dosage adjustment.
sition. Antimicrob Agents Chemother 6:655–665.
Houston JB and Kenworthy KE (2000) In vitro-in vivo scaling of CYP kinetic data not consistent
with the classical Michaelis-Menten model. Drug Metab Dispos 28:246–254.
Hutzler JM and Tracy TS (2002) Atypical kinetic profiles in drug metabolism reactions. Drug
Metab Dispos 30:355–362.
Kauffman RE, Thirumoorthi MC, Buckley JA, Aravind MK, and Dajani AS (1981) Relative
bioavailability of intravenous chloramphenicol succinate and oral chloramphenicol palmitate
in infants and children. J Pediatr 99:963–967.
Kearns GL, Abdel-Rahman SM, Alander SW, Blowey DL, Leeder JS, and Kauffman RE (2003)
Developmental pharmacology–drug disposition, action, and therapy in infants and children.
N Engl J Med 349:1157–1167.
Krishnaswamy S, Duan SX, Von Moltke LL, Greenblatt DJ, Sudmeier JL, Bachovchin WW, and
Court MH (2003) Serotonin (5-hydroxytryptamine) glucuronidation in vitro: assay develop-
ment, human liver microsome activities and species differences. Xenobiotica 33:169–180.
Lambdin MA, Waddell WW Jr, and Birdsong M (1960) CP toxicity in the premature infant.
Pediatrics 25:935–940.
Miles MV (1983) Pediatric pharmacokinetics, in Applied Clinical Pharmacokinetics (Mungall
DR ed), pp 367–388, Raven Press, New York.
Miners JO, Smith PA, Sorich MJ, McKinnon RA, and Mackenzie PI (2004) Predicting human
drug glucuronidation parameters: application of in vitro and in silico modeling approaches.
Annu Rev Pharmacol Toxicol 44:1–25.
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.
National Toxicology Program (2002) Chloramphenicol. Rep Carcinog 10:48–50.
Ogilvie BW, Usuki E, Burton A, Yerino P, and Parkinson A (2008) In vitro approaches for
studying the inhibition of drug-metabolizing enzymes and identifying the drug-metabolizing
enzymes responsible for the metabolism of drugs (reaction phenotyping) with emphasis on
cytochrome P450, in Drug-Drug Interactions. 2nd ed (Rodrigues AD ed) pp 231–358, Marcel
Deckker, Inc., New York.
Pillot T, Ouzzine M, Fournel-Gigleux S, Lafaurie C, Radominska A, Burchell B, Siest G, and
Magdalou J (1993) Glucuronidation of hyodeoxycholic acid in human liver. Evidence for a
selective role of UDP-glucuronosyltransferase 2B4. J Biol Chem 268:25636–25642.
Rowland A, Gaganis P, Elliot DJ, Mackenzie PI, Knights KM, and Miners JO (2007) Binding of
inhibitory fatty acids is responsible for the enhancement of UDP-glucuronosyltransferase 2B7
activity by albumin: implications for in vitro-in vivo extrapolation. J Pharmacol Exp Ther
321:137–147.
Suhrland LG and Weisberger AS (1963) Chloramphenicol toxicity in liver renal disease. Arch
Intern Med 112:747–754.
Tsoutsikos P, Miners JO, Stapleton A, Thomas A, Sallustio BC, and Knights KM (2004)
Evidence that unsaturated fatty acids are potent inhibitors of renal UDP-glucuronosyltrans-
ferases (UGT): kinetic studies using human kidney cortical microsomes and recombinant
UGT1A9 and UGT2B7. Biochem Pharmacol 67:191–199.
Uchaipichat V, Galetin A, Houston JB, Mackenzie PI, Williams JA, and Miners JO (2008)
Kinetic modeling of the interactions between 4-methylumbelliferone, 1-naphthol, and zidovu-
dine glucuronidation by UDP-glucuronosyltransferase 2B7 (UGT2B7) provides evidence for
multiple substrate binding and effector sites. Mol Pharmacol 74:1152–1162.
Venkatakrishnan K, Von Moltke LL, and Greenblatt DJ (2001) Human drug metabolism and the
cytochromes P450: application and relevance of in vitro models. J Clin Pharmacol 41:1149–
1179.
Acknowledgments. We are grateful to Drs. Caroline Decker, Hong
Gao, and Hongying Gao (Vertex Pharmaceuticals) for assistance and
helpful discussions.
Weiss CF, Glazko AJ, and Weston JK (1960) Chloramphenicol in the newborn infant. A
physiologic explanation of its toxicity when given in excessive doses. N Engl J Med 262:
787–794.
References
Weiss JN (1997) The Hill equation revisited: uses and misuses. FASEB J11:835–841.
Yamanaka H, Nakajima M, Katoh M, Kanoh A, Tamura O, Ishibashi H, and Yokoi T (2005)
Trans-3Ј-hydroxycotinine O- and N-glucuronidations in human liver microsomes. Drug Metab
Dispos 33:23–30.
Barre L, Fournel-Gigleux S, Finel M, Netter P, Magdalou J, and Ouzzine M (2007) Substrate
specificity of the human UDP-glucuronosyltransferase UGT2B4 and UGT2B7. Identification
of a critical aromatic amino acid residue at position 33. FEBS J 274:1256–1264.
Bhasker CR, McKinnon W, Stone A, Lo AC, Kubota T, Ishizaki T, and Miners JO (2000)
Genetic polymorphism of UDP-glucuronosyltransferase 2B7 (UGT2B7) at amino acid 268:
ethnic diversity of alleles and potential clinical significance. Pharamacogenetics 10:679–685.
Boucher FD, Modlin JF, Weller S, Ruff A, Mirochnick M, Pelton S, Wilfert C, McKinney R Jr,
Crain MJ, and Elkins MM (1993) Phase I evaluation of zidovudine administered to infants
exposed at birth to the human immunodeficiency virus. J Pediatr 122:137–144.
Address correspondence to: Dr. David A. Williams, Massachusetts College
Pharmacy and Health Sciences, Department of Pharmaceutical Sciences, 179
Longwood Ave., Boston, MA 02115. E-mail: david.williams@mcphs.edu
Bouwmeester NJ, Hop WC, van Dijk M, Anand KJ, van den Anker JN, and Tibboel D (2003)