toxicological sciences 137(1), 12–25 2014
doi:10.1093/toxsci/kft221
Advance Access publication October 1, 2013
Evaluation of Serum Bile Acid Profiles as Biomarkers of Liver Injury
in Rodents
Lina Luo,* Shelli Schomaker,* Christopher Houle,† Jiri Aubrecht,* and Jennifer L. Colangelo*,1 *Biomarkers of Drug Safety Research and Development and †Toxicologic Pathology of Drug Safety Research and Development, Pfizer Inc., Groton,
Connecticut 06340
To whom correspondence should be addressed at Biomarkers of Drug Safety Research and Development, Pfizer Inc., 8274-1429 Eastern Point Road, Groton,
CT 06340. Fax: (860) 715-8045. E-mail: [email protected].
1
Received June 19, 2013; accepted September 9, 2013
Bile acids (BAs) have been studied as potential biomarkers of
drug-induced liver injury. However, the relationship between levels of individual BAs and specific forms of liver injury remains to
be fully understood. Thus, we set out to evaluate cholic acid (CA),
glycocholic acid (GCA), and taurocholic acid (TCA) as potential
biomarkers of liver injury in rodent toxicity studies. We have
developed a sensitive liquid chromatography-tandem mass spectrometry (LC/MS/MS) assay applicable to rat and mouse serum
and evaluated levels of the individual BAs in comparison with the
classical biomarkers of hepatotoxicity (alanine aminotransferase
[ALT], aspartate aminotransferase [AST], glutamate dehydrogenase (GLDH), alkaline phosphatase, total bilirubin, gammaglutamyl transferase, and total BAs) and histopathology findings
in animals treated with model toxicants. The pattern of changes
in the individual BAs varied with different forms of liver injury.
Animals with histopathologic signs of hepatocellular necrosis
showed increases in all 3 BAs tested, as well as increases in ALT,
AST, GLDH, and total BAs. Animals with histopathologic signs
of bile duct hyperplasia (BDH) displayed increases in only conjugated BAs (GCA and TCA), a pattern not observed with the other
toxicants. Because BDH is detectable only via histopathology, our
results indicate the potential diagnostic value of examining individual BAs levels in serum as biomarkers capable of differentiating specific forms of liver injury in rodent toxicity studies.
Key Words: bile acids; drug-induced liver necrosis; bile duct
hyperplasia; LC/MS/MS; biomarkers.
Adverse drug reactions, especially drug-induced liver
injury (DILI), represent a major challenge for drug development. Hepatotoxicity has been considered the most frequent
cause of safety-related drug withdrawals for the past 50 years
(FDA, 2009; Kola and Landis, 2004; Lazarou et al., 1998;
Pirmohamed et al., 2004). Serum enzymatic activity of alanine
aminotransferase (ALT) is considered the gold standard clinical chemistry biomarker of liver injury in preclinical species
and humans (Amacher, 2002; Amacher et al., 1998; Ozer et al.,
2010). However, ALT assessments in preclinical studies may
present a challenge, especially when increases in ALT activity
do not correlate with histopathology findings (Ennulat et al.,
2010). In many cases, these increases can be attributed to induction or extrahepatic injury, such as muscle damage or metabolic
state, and might be addressed by additional biochemical parameters. On the other hand, histopathologic analysis may be the
only indicator for certain types of liver injury, such as bile duct
hyperplasia (BDH). BDH can occur secondary to other abnormalities, such as portal inflammation, cholestasis, and biliary
and/or hepatocellular injury, and can also occur as a primary
lesion. Thus, the development of additional biomarkers capable
of facilitating the interpretation of serum ALT increases and
differentiating between various histopathologic findings in preclinical toxicity studies is important.
Efforts to identify and develop additional biomarkers for
DILI and/or to enhance the current biomarker panels have been
initiated (Amacher et al., 2005; Ozer et al., 2008). Standard
clinical chemistry panels currently include monitoring some
combination of ALT, aspartate aminotransferase (AST), alkaline phosphatase (ALP), and gamma-glutamyl transferase
(GGT) serum activity, as well as serum concentration of total
bilirubin (TBIL). However, these often are of limited use
for detecting BDH. Several research groups have employed
genomics, proteomics, and metabonomics platforms in hopes
of identifying potential biomarkers in this area. Recent attention has been given to bile acids (BAs), which have been identified in several metabonomic studies. These studies indicated
that an altered BA profile was a key characteristic of the toxic
response in the liver (Beckwith-Hall et al., 1998; Davis and
Thompson, 1993; Lin et al., 2009; Yamazaki et al., 2013).
BAs are a class of structurally similar compounds that play
essential roles in cholesterol homeostasis, lipid absorption, and
intestinal signaling (Xiang et al., 2010). Synthesized in the liver
from cholesterol, BAs are excreted into the small intestine via
the bile duct mainly as glycine or taurine conjugates and then
© The Author 2013. Published by Oxford University Press on behalf of the Society of Toxicology. All
rights reserved. For permissions, please email: [email protected].
Evaluation of Serum Bile Acid Profiles
undergo enterohepatic circulation with further metabolism by
bacterial and hepatic enzymes (Martin et al., 2007). Liver and
gastrointestinal diseases often disturb the hepatic synthesis and
clearance of individual BAs giving rise to quantitative changes
in the pattern of serum BAs (Burkard et al., 2005; Meng et al.,
1997; Thompson et al., 1987). In contrast, the measurement
of total BAs may provide a general overall assessment of liver
function, particularly in advanced stages of liver disease, but
provide little insight into specific liver pathology especially
early on in the disease process (Berg et al., 1986; Reyes and
Sjövall, 2000). Because the total BA test measures the sum of
all serum BAs, which is over 20 BAs in most species, and can
be influenced by the presence of other endogenous molecules,
the analysis of individual BAs has been proposed to provide
valuable information regarding the pathogenesis of toxic liver
injury and disease (Alnouti et al., 2008; Bentayeb et al., 2008;
Ducroq et al., 2010; Turley and Dietschy, 1978).
Numerous analytical methods have emerged to determine
BA concentrations in plasma and serum (Gatti et al., 1997; Lee
et al., 1997; Street et al., 1985; Thompson et al., 1987). The
complexity of metabolism, the typically low concentration of
BAs in biological fluids, and the existence of multiple isobaric
structural isomers make BA separation and quantitation challenging (Janzen et al., 2010; Ostrow, 1993). In recent years,
liquid chromatography coupled with mass spectrometry (LC/
MS) has become an ideal option for the analysis of individual BAs due to the high sensitivity and selectivity of the platform (Alnouti et al., 2008; Ando et al., 2006; Bentayeb et al.,
2008; Bobeldijk et al., 2008; Griffiths and Sjövall, 2010; Hagio
et al., 2009; Scherer et al., 2009). Other benefits of utilizing
the LC/MS platform for these analyses include simple sample
preparation, low sample volume requirements, and relatively
low cost for reagents. Of the 2 primary BAs, cholic acid (CA)
and chenodeoxycholic acid (CDCA), CA is more abundant in
rats. Because multiplexing individual BAs into a single LC/
MS assay often impairs accuracy and precision (Suzuki et al.,
2013), we selected the most relevant primary BA to rat, CA,
and its direct conjugates, glycocholic acid (GCA) and taurocholic acid (TCA), for our studies.
Goals of this study were to develop a sensitive LC/MS
method for detection of CA, GCA, and TCA in rodent serum
and to evaluate the potential of BA serum profiles as a biomarker of DILI with the capability of differentiating biliary and
hepatocellular damage in rodents.
Materials and Methods
Chemicals and reagents. CA, GCA, and TCA were purchased from
Sigma-Aldrich (St Louis, Missouri). d4-CA and d4-GCA were purchased
from C/D/N ISOTOPES Inc. (Quebec, Canada). d4-TCA was purchased from
Toronto Research Chemicals Inc. (Ontario, Canada). HPLC grade methanol, acetonitrile, water, and formic acid were purchased from Honeywell
Burdick & Jackson (Muskegon, Michigan). Charcoal-stripped serum was from
Bioreclamation (Westbury, New York).
Galactosamine (GalN), microcystin-LR (MC), α-naphthylisothiocyanate
(ANIT), acetaminophen (APAP), and isoproterenol (ISO) were purchased from
13
Sigma. Drug candidates A, B, C, and D were obtained from Pfizer (Groton,
Connecticut).
Animals. All studies were approved and conducted under the oversight
of the Institutional Animal Care and Use Committee. Studies were conducted
on male Sprague Dawley rats (approximately 6–9 weeks old) or male CD-1
mice (approximately 6–8 weeks old), with the exception of the ISO study that
used Hanover Wistar rats (approximately 6–9 weeks old). Hanover Wistar was
selected for the ISO study because we had conducted full characterization of
the cardiac toxicity in this strain, and no differences in serum concentrations of
individual BAs have been observed between the 2 strains of rats (data not published). Drinking water and a standard commercial laboratory certified rodent
diet were provided ad libitum throughout the studies. All rats were in the fasted
state prior to necropsy.
General study design. Test articles (N = 9) were selected to induce liver
injury or injury to other organs. These compounds included classic hepatotoxicants, a cardiac toxicant, a testicular toxicant, a compound known to elicit
BDH, and a nontoxic comparator compound. Within each study, a vehicle control group was utilized with the same number of animals as treatment groups.
Treatments of test article or vehicle were administrated to rats or mice for a
period of time by oral (PO) gavage or subcutaneous (SC) or intraperitoneal
(IP) injection (Table 1). Blood samples were collected prior to necropsy at the
end of treatment.
Sample collection. Blood samples were collected via the vena cava at
necropsy, unless otherwise noted. Serum was collected in tubes containing no
anticoagulant and was obtained from all rats dosed with vehicle or compound.
Terminal serum samples were collected at the time of necropsy. All serum samples were stored at −80°C until analysis.
Clinical chemistry and histopathology. ALT, AST, ALP, GGT, GLDH,
TBIL, and total bile acids (TBAs) were analyzed by standard clinical chemistry techniques on the Siemens Advia 2400 platform. Cerner HNA Millennium
Laboratory Information System was used to acquire data.
Histopathologic examination of the liver was performed in all cases with
additional evaluation of other major organs when appropriate. Organ samples
were fixed in 10% buffered formalin, embedded in paraffin, sectioned, and
stained with hematoxylin and eosin. Liver pathology was generally graded
using a 4-point scheme as follows: (1) minimal, (2) mild, (3) moderate, and
(4) marked.
Sample preparation for LC/MS/MS analysis. Fifty microliters of each
serum sample was placed into an individual well of a Sirocco protein precipitation plate on the top of a 96-well collection plate. Three hundred microliters of
methanol was added to each well for protein precipitation. Deuterated standards of CA, GCA, and TCA were used as internal standards (ISs). Ten microliters of the 1.0 µg/ml working solution of IS mixture was added, followed by
vortex mixing. After 10 min of centrifugation, the Sirocco plate was removed,
the supernatant was evaporated to dryness under a steady stream of N2 at 37°C.
The samples were reconstituted in 50 µl methanol/water (1/1, vol/vol) and then
3 µl was injected onto the LC/MS system.
Calibration and quality control standard preparation. One milligram per
milliliter of BA stock solutions or IS stock solutions were prepared by dissolving each BA reference in methanol. The individual BA stock solutions were
combined and diluted to achieve the concentration of 20 µg/ml for a working
standard solution or 1 µg/ml for a working IS solution. Nine calibration standard solutions ranging from 5 to 5000 ng/ml were prepared by serially diluting
the working standard solution into charcoal-stripped serum. Quality control
(QC) standards were prepared in the same manner at 20, 500, and 4000 ng/ml in
stripped serum. The calibration standards and QC standards then went through
the sample preparation process described above.
Individual bile acid analysis. Individual bile acid (IBA) analysis was
performed by LC/MS/MS. The mass spectrometer was an ABSciex 5500
QTrap equipped with Turbo Spray ion source, operating in negative mode.
An Acquity UPLC system was interfaced to the front end of the MS system.
All chromatographic separations were performed by gradient elution with an
Negative controls (other organ toxicants)
BDH negative control
BDH
Toxicants with necrosis
Category
D
C
ISO
B
APAP
A
GalN
MC
ANIT
Compound
1000 mg/kg
0.2 mg/kg
30 mg/kg
100 mg/kg
1000 mg/kg
150 mg/kg
300 mg/kg
500 mg/kg
1000 mg/kg
150 mg/kg
300 mg/kg
100 µg/kg
500 µg/kg
4000 µg/kg
100 µg/kg
500 µg/kg
4000 µg/kg
5 mg/kg
50 mg/kg
500 mg/kg
75 mg/kg
150 mg/kg
Dose
Levels
IP
IP
PO
PO
PO
PO
PO
PO
PO
PO
PO
SC
SC
SC
SC
SC
SC
PO
PO
PO
PO
PO
Administration
Vehicle
PBS
PBS
0.5% methylcellulose
0.5% methylcellulose
0.5% methylcellulose
0.5% methylcellulose and 5% PEG400
0.5% methylcellulose and 5% PEG400
0.5% methylcellulose and 5% PEG400
0.5% methylcellulose and 5% PEG400
0.5% methylcellulose and 5% PEG400
0.5% methylcellulose and 5% PEG400
PBS
PBS
PBS
PBS
PBS
PBS
20% PEG400 and 20% hydroxypropyl betacyclodextrin
20% PEG400 and 20% hydroxypropyl betacyclodextrin
20% PEG400 and 20% hydroxypropyl betacyclodextrin
20% hyroxypropyl betacyclodextrin
20% hyroxypropyl betacyclodextrin
Table 1
In Vivo Animal Studies
24 h
24 h
24 h
24 h
24 h
5 days
5 days
5 days
5 days
5 days
5 days
6h
6h
6h
24 h
24 h
24 h
7 days
7 days
7 days
21 days
21 days
Treatment
5 rats
8 rats
10 rats
10 rats
5 rats
5 rats
5 rats
5 rats/6 mice
5 rats/6 mice
5 rats
5 rats
4 rats
4 rats
4 rats
4 rats
4 rats
4 rats
5 rats
5 rats
5 rats
5 rats
5 rats
No. of Animals/Group
14
Luo et al.
Evaluation of Serum Bile Acid Profiles
Acquity Shield RP18 column, 50 × 2.1 mm, 1.7 µm, maintained at 55°C at a
flow rate of 400 µl/min. The gradient program started at 65% mobile phase
A (0.05% formic acid in water) and 35% mobile phase B (5% acetonitrile in
methanol), increased to 90% of B in 7 min, decreased to 35% B in 0.5 min, and
then held at 35% B for 2.5 min. The mass spectrometer was operated with the
source and desolvation temperatures set at 120°C and 350°C, respectively. The
curtain gas was 40 psi; the ion spray voltage was 4500 V; probe temperature
was 600°C; and ion source gas 1 and ion source gas 2 were 40 and 30 psi,
respectively. Deuterated standards d4-TCA, d4-GCA, and d4-CA were used as
ISs for TCA, GCA, and CA, respectively. Analysis in the mass spectrometer
was performed in multiple reaction monitoring (MRM) mode. The MRM transitions (m/z) were 407.3>407.3 for CA, 464.3>74.3 for GCA, 514.3>80.3 for
TCA, 411.3>411.3 for d4-CA, 468.2>73.8 for d4-GCA, and 518.1>79.7 for d4TCA. Peak integration and quantification were performed using Analyst 1.5.1
software. Individual standard curves for each BA were constructed by plotting
the ratio of the BA peak area to its deuterated standard peak area versus concentration. Slope and y-intercept were calculated using a linear curve fit with
1/x2 weighting. The concentrations of BAs in study samples were calculated
relative to the regression line.
Freshly prepared standard curve and QC samples were included in each
analysis run. A run was deemed acceptable if the QC samples were ± 15% of
the nominal concentrations and the coefficient of variance (CV) did not exceed
10%.
Method validation. The method was validated using QC samples at
3 concentration levels (20, 200, and 2000 ng/ml) from the calibration curve.
Four replicates of each QC sample were analyzed in a single run to determine
the intraassay accuracy and precision. This process was repeated 4 times over
4 days in order to determine the interassay accuracy and precision. Accuracy
and precision were calculated from the % relative error (RE) [%(measured − theoretical)/theoretical concentration] and relative standard deviation
[%RSD = % standard deviation/mean], respectively. The assay recovery was
evaluated by comparing the mean detector response of extracted QC samples
at low, medium, and high concentrations (20, 200, and 1000 ng/ml) in 4 replicates to those of postextracted serum blanks spiked at equal concentrations.
The matrix effect was estimated by comparing the extracted serum residue and
the neat solution.
Statistical analysis. Data are presented as individual animals or group
mean ± SD. Statistical analyses were conducted by 2-tailed Student’s t test
to compare drug treatment groups with vehicle control groups. Values significantly different from control are indicated as **p < .01 and *p < .05.
Results
Development of an LC/MS Assay for Detection of CA, GCA,
and TCA
We have developed a sensitive method for the quantification of BAs using LC/MS. Because BAs are endogenous
molecules and already present in rodent serum, we used charcoal-stripped rat serum for the standard curve and QC sample
preparations. To increase the method accuracy and precision,
deuterium-labeled ISs for each corresponding BA were used
for the quantification. The accuracy and precision of TCA have
been reported to be acceptable only when d4-TCA was used
as the IS (Xiang et al., 2010). All BAs gave excellent linear
response over a 103 dynamic range of 5–5000 ng/ml, with coefficients of determination (R2) above 0.999. The lower detection
limit for these BAs was 0.5 ng/ml. Mean intraassay accuracy
was 95%–109% for the 3 IBAs, with a mean CV of 3%–8%;
mean interassay accuracy was 97%–106%, with a mean CV
15
of 4%–10% (Table 2). The mean recovery rates of the extraction procedures were between 103% and 105%. No significant
matrix effect was observed, and the signal difference was less
than 5%. These values were within the acceptable range, and
the method was judged to be suitably accurate and precise for
the analytes.
Currently, no published data can be found on the reference
ranges of individual serum BA levels. The 3 IBA serum concentrations from various vehicle-treated rodents in toxicology
studies were generated here to provide baseline endogenous levels. Endogenous levels of TCA, GCA, and CA in various vehicle-treated rats (n = 46) measured by this LC/MS assay were
0.184 ± 0.153 µg/ml, 0.559 ± 0.333 µg/ml, and 5.455 ± 2.196 µg/
ml, respectively; the serum levels of TCA, GCA, and CA in
various vehicle-treated mice (n = 18) were 0.470 ± 0.464 µg/
ml, 0.004 ± 0.002 µg/ml, and 0.091 ± 0.11 µg/ml, respectively.
Serum BA Profiles After Treatment With Model Liver
Toxicants
As expected, the treatment of rats with model liver toxicants
GalN, MC, ANIT, and APAP caused a wide degree of hepatocellular/hepatobiliary effects detected via histopathology and
serum biochemical analyses (Table 3). As expected, no histopathologic or serum biomarker changes were observed in vehicle-treated animals.
Rats dosed with a single dose of 1000 mg/kg GalN for 24 h
revealed moderate to marked panlobular hepatocellular necrosis that was randomly distributed throughout the liver (Fig. 1A).
Statistically significant, treatment-related increases in serum
ALT, AST, and GLDH activity levels, together with increased
concentrations of total BAs, were also noted. The LC/MS analysis of CA, GCA, and TCA levels from the GaIN-treated rats
showed statistically significant elevations of serum concentrations for all measured BAs. CA, as one of the primary BAs
in rats, remained the most abundant BA in GalN-treated animals (Fig. 2). Interestingly, the BA proportions changed after
treatment with GalN with CA accounting for 51% of the total
3 BAs measured as compared with 93% in the control group.
Conversely, TCA accounted for 41% of the total 3 BAs in the
profile after treatment compared with 2% in controls.
Rats dosed with a single dose of 1000 mg/kg APAP for 24 h
produced characteristic hepatocellular necrosis that was accompanied by significant increases in ALT, AST, ALP, and GLDH
activity and elevation of total BA concentrations (Fig. 1B). No
other clinical chemistry parameters, such as GGT and TBIL,
exhibited significant changes. LC/MS analysis revealed statistically significant changes in serum concentrations of the 3
IBAs when compared with their corresponding control groups
(Table 3). APAP treatment resulted in an IBA profile similar to
GalN treatment with CA remaining the major BA (77% of the
total 3 BAs) as shown in Figure 2.
Treatment of rats with 30 or 100 mg/kg of ANIT for 24 h
caused mild to moderate hepatobiliary portal inflammation and biliary degeneration and necrosis in most animals
16
106
102
102
8
6
6
2.11
2.04
2.04
Abbreviations: %RSD = % standard deviation/mean; %RE = %(measured − theoretical)/theoretical concentration.
97
100
99
0.193
0.200
0.198
6
10
8
98
104
104
4
6
4
109
103
108
8
5
5
2.17
2.06
2.17
97
104
100
0.193
0.208
0.201
Intraday validation (n = 4)
CA
0.0198
GCA
0.0200
TCA
0.0190
Interday validation (n = 4)
CA
0.0196
GCA
0.0209
TCA
0.0207
3
3
3
99
100
95
7
5
5
%RSD
%RE
%RSD
Average (ng/ml)
%RE
%RSD
Average (ng/ml)
IBAs
0.2 μg/ml
0.02 μg/ml
Table 2
Performance of the QC Standards in the LC/MS Quantitative Assay for 3 BAs
Average (ng/ml)
2 μg/ml
%RE
Luo et al.
(Fig. 1C). These changes were slightly more extensive in rats
given 100 mg/kg of ANIT, as compared with rats given 30 mg/
kg. Focal or multifocal, minimal, randomly scattered areas of
hepatocellular necrosis were observed in some rats given each
dose. The biomarker analysis showed statistical increases in
levels of all tested biomarkers (ALT, AST, ALP, GLDH, GGT,
TBIL, and total BAs) at both dose levels. The LC/MS analysis of CA, GCA, and TCA concentrations in ANIT-treated rats
revealed statistically significant elevations. Furthermore, the
proportions of the 3 IBAs were altered markedly. Conjugated
BAs, especially TCA, were substantially elevated in both the 30
and 100 mg/kg groups (682× and 856× fold increases over the
control group) and became the predominate BA in the treated
groups. The unconjugated BA (CA) exhibited a comparatively
small, yet statistically significant increase (Fig. 2). The ratio
of the taurine conjugate over the total 3 BAs was higher in the
100 mg/kg group as compared with the 30 mg/kg group (data
not shown). Paradoxically, ALT, AST, and GLDH had larger
fold changes in the 30 mg/kg dose group (4.1×, 5.7×, and 64×
control, respectively) than in the 100 mg/kg group (1.9×, 3.4×,
and 35× control, respectively). The fold increases of ALP,
GGT, and TBIL, which are generally considered indicators of
hepatobiliary toxicity, were slightly increased in the 100 mg/kg
group as compared with the 30 mg/kg group. This was consistent with TCA levels, the predominant BA here, which showed
dose-dependent increases from 30 to 100 mg/kg (Table 3).
Treatment of rats with a single dose of 0.2 mg/kg MC for
24 h caused moderate to marked centrilobular necrosis with frequent involvement of adjacent portal tracts, which include biliary tracts. These findings were accompanied by significantly
increased activity levels for ALT, AST, GLDH, ALP, TBIL, and
total BAs (Fig. 1D). The LC/MS analysis of CA, GCA, and
TCA levels from the rats treated with MC showed statistically
significant elevations of serum concentrations of all measured
BAs. Again, the BA proportions changed dramatically. TCA
had the largest fold increases of 903× over the control (Table 3)
and became the major BA accounting for 58% of the total 3
BAs compared with 1.4% in control (Fig. 2).
Serum BA Profiles and BDH
To evaluate the BA profile in response to BDH, we treated
rats with a well-characterized drug candidate (compound A)
that was discontinued from development due to prevalent BDH
detected in rat safety studies. To assess specificity of the biomarker response, we used a structurally similar discontinued
drug candidate of the same class (compound B) that did not
cause BDH in rat safety studies. As expected, histopathologic
examination of rats dosed with compound A revealed mild to
moderate BDH in the 3 highest dose groups, 300, 500, and
1000 mg/kg (Fig. 1E). In agreement with previous studies,
the classical biomarkers of liver injury measured in our studies (ALT, AST, ALP, GLDH, GGT, TBIL, and TBAs) failed to
detect BDH. Although total BA levels detected by an enzyme
cycling–based assay was unchanged, the LC/MS analysis of
A
B
BDH negative
control
APAP
ANIT
MC
GalN
Compound
BDH
Toxicants with
necrosis
Category
1000 mg/kg
(24 h)
0.2 mg/kg
(24 h)
30 mg/kg
(24 h)
100 mg/kg
(24 h)
1000 mg/kg
(24 h)
150 mg/kg
(5 days)
300 mg/kg
(5 days)
500 mg/kg
(5 days)
1000 mg/kg
(5 days)
150 mg/kg
(5 days)
300 mg/kg
(5 days)
Dose
No
(2)
No
No
(1) to (2)
No
No
(1)
No
No
No
No
No
No
BD, HC (2) to (3)
HC (3) to (4)
No
No
No
BDH
BD, HC (2) to (3)
HC (3) to (4)
HC (3) to (4)
Necrosis
Liver Histopathologya
0.9
1.0
1.4
1.4
1.4
1.2
12*
1.9**
4.1**
181**
18**
ALT
0.8
1.2
0.8
0.9
0.8
1.1
31*
3.4**
6**
130**
33**
AST
1.0
0.9
0.9
0.9
0.9
1.0
1.4*
1.4*
1.3*
2.2*
NP
ALP
Table 3
Summary of Study Endpoints for Rats
0.8
0.8
1.3
1.4
1.4
2.0
141**
35**
65**
663*
121**
GLDH
1.0
1.0
1.0
1.0
1.0
1.0
1.0
24**
23**
66**
NP
TBIL
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.4**
1.7*
2.0
NP
GGT
Biomarkersb
0.84
0.83
1.3
1.4
1.0
0.7
4.5**
22**
22**
9.3**
3.4**
Total BAs
0.7
1.1
1.5
1.7
1.2
0.8
5.3**
2.8**
3.7**
5.4**
4.4**
CA
1.6
0.6
69**
27*
13*
0.7
26*
856**
682**
903*
82**
TCA
1.6
0.6
10**
18**
6.0*
0.8
5.7*
52**
84**
39**
7.1**
GCA
Evaluation of Serum Bile Acid Profiles
17
De
Cd
ISOc
Compound
100 µg/kg
(6 h)
500 µg/kg
(6 h)
4000 µg/kg
(6 h)
100 µg/kg
(24 h)
500 µg/kg
(24 h)
4000 µg/kg
(24 h)
5 mg/kg
(7 days)
50 mg/kg
(7 days)
500 mg/kg
(7 days)
75 mg/kg
(21 days)
150 mg/kg
(21 days)
Dose
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
BDH
No
Necrosis
2.1**
1.7**
3.2**
1.8**
1.3*
1.0
1.0
1.0
1.2
1.2
1.2
ALT
1.2
1.1
3.5**
0.9
1.1
1.5
1.3
1.0
3.2**
1.9*
1.0
AST
1.3*
1.3*
1.1
1.1
0.9
1.0
0.9
1.0
0.9
0.8
0.8
ALP
NP
NP
NP
NP
NP
1.4
1.2
1.1
1.2
1.1
2.3
GLDH
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
TBIL
1.0
1.0
1.1
1.2
1.2
1.0
1.0
1.0
1.0
1.0
1.0
GGT
Biomarkersb
NP
NP
NP
NP
NP
1.6
1.4
2.1
1.0
2.8*
1.2
Total BAs
0.8
1.2
1.1
0.7
0.6
1.3
1.2
1.7
1.3
2.0
1.2
CA
0.8
1.2
0.8
0.8
0.5
0.8
1.1
1.2
0.7
1.7
1.3
TCA
0.2**
1.2
0.6
0.4
1.1
0.6
1.8
1.3
0.1
0.5
0.7
GCA
b
a
HC, hepatocyte; BD, bile duct or hepatobillary; the number in the parentheses stands for the severity of liver histopath findings: (1) minimal; (2) mild; (3) moderate; (4) marked.
Values represent fold changes compared with their corresponding controls; NP, not performed.
c
At the 6-h time point changes in the heart consisted predominantly of minimal to mild subepicardial hemorrhage/edema at all dose levels; by 24 h, this progressed to also include minimal to mild
myocardial degeneration/necrosis and inflammation at all dose levels.
d
Minimal mesenteric lipid depletion in the 500 mg/kg group; moderate to marked zymogen decrease in pancreas and mesenteric lipid depletion in the 500 mg/kg group.
e
Minimal epidydimal spermatic granulomas in the 75 mg/kg group with mild epidydimal interstitial inflammation and skeletal muscle inflammation in the 150 mg/kg group.
Statistically significant changes are indicated by *p < .05, **p < .01.
Negative
controls
(other organ
toxicants)
Category
Liver Histopathologya
Table 3—Continued
18
Luo et al.
Evaluation of Serum Bile Acid Profiles
19
Fig. 1. Histopathology analysis. Representative hematoxylin and eosin stained sections of liver from rats treated with 1000 mg/kg of galactosamine (A);
1000 mg/kg of acetaminophen (B); 100 mg/kg of α-naphthylisothiocyanate (C); 0.2 mg/kg of microcystin (D); 1000 mg/kg of compound A (E); or 4000 μg/kg
of isoproterenol (F). The isoproterenol image is representative of all those without any liver findings (compound B, compound C, and compound D). All images
taken using ×400 magnification.
CA, GCA, and TCA revealed statistically significant increases
of the conjugated BAs, GCA, and TCA, correlating with the
BDH findings (Fig. 2). No statistically significant changes
were observed for CA concentrations in any dose group,
regardless of the presence or absence of BDH (Table 3). None
of IBAs significantly changed in the groups that had no histopathological findings. As expected, the treatment of rats with
compound B did not affect any biomarkers measured in our
studies, and no histopathology findings were observed in the
liver (Supplementary Data). Due to the limited tolerability of
compound B in rats, animals could not be dosed above 300 mg/
kg (Table 3).
A similar study was conducted in mice to investigate the translation of biliary hyperplasia from rats to mice with compound A. Moderate to marked portal tract inflammation with associated
bile duct degeneration and hyperplasia was observed in the 2
groups treated with compound A. Some mice exhibited poor tolerance at the high dose of 1000 mg/kg, so those samples were not
analyzed. TCA and GCA again were significantly increased in
the compound A–treated groups (Fig. 3). No significant changes
were observed for CA in the treated groups compared with the
concurrent control group. Even though mice and rats exhibit different profiles for the 3 BAs (TCA is the predominate BA in
mice instead of CA in rats.), these results in mice mimicked
those in the rat studies in that increases in conjugated BAs were
observed in animals with BDH and no changes were observed
in other parameters, providing confidence in the potential use of
individual BAs as biomarkers of BDH.
20
Luo et al.
Fig. 2. Effects of various compounds on levels of serum IBAs in rats measured by LC/MS/MS. Each bar represents the mean ± SD for 4–10 rats as described
in Table 1. Statistical analysis was performed by 2-tailed Student’s t test. Abbreviations: ANIT, α-naphthylisothiocyanate; CA, cholic acid; GalN, galactosamine;
GCA, glycocholic acid; IBA, individual bile acid; ISO, isoproterenol; MC, microcystin, TCA, taurocholic acid.
Serum BA Profile After Treatment With Model Nonhepatic
Toxicants
To evaluate whether or not the changes in individual BAs are
specific to liver injury, we analyzed serum samples from studies with model nonhepatic toxicants (Table 1). ISO is known
to induce cardiac damage at certain dosages (York et al., 2007;
Zhang et al., 2008). In the ISO study, male Hanover Wistar rats
received a single dose of 100, 500, or 4000 µg/kg of ISO SC. As
expected, the ISO-treated rats exhibited a minimal to mild subepicardial hemorrhage/edema at all dose levels, accompanied
by increases in AST activities in the 500 and 4000 µg/kg dose
groups at the 6-h time point after treatment. At the later 24-h
time point after ISO treatment, the rats showed cardiac toxicity characterized by myocardial inflammation and necrosis,
without significant changes in clinical chemistry biomarkers
(Table 3). The livers of ISO-treated rats had no histopathological findings at both the 6- and 24-h time points (Fig. 1F)
and no statistically significant changes in serum levels of CA,
GCA, or TCA (Fig. 2). A statistically significant increase of
total BAs was noted in animals treated with 500 µg/ml of ISO
at 6 h postdose, but was not dose or time dependent (Table 3).
The fact that the levels of individual BAs in vehicle-treated rats
Evaluation of Serum Bile Acid Profiles
Fig. 3. Effects of compound A on levels of serum IBAs in mice measured
by LC/MS/MS. Each bar represents the mean ± SD for 6 mice after 5-day
treatments. Statistical analysis was performed by 2-tailed Student’s t test.
Abbreviations: CA, cholic acid; GCA, glycocholic acid; IBA, individual bile
acid; TCA, taurocholic acid.
in the ISO study did not differ from those of vehicle-treated
Sprague Dawley rats in the other studies (Fig. 2), indicates the
absence of strain-specific differences for serum concentrations
of individual BAs between the Hanover Wistar and Sprague
Dawley rats.
Compounds C and D are discontinued drug candidates
with well-characterized, nonhepatotoxic findings. Compound
C caused decreases in body weight and reductions in food
consumption, as well as effects associated with the expected
pharmacology of this compound, including lipid depletion of
mesenteric fat that ranged from minimal at 50 mg/kg to moderate or marked at 500 mg/kg, and decreased pancreatic zymogen
granules ranging from minimal to mild at 50 and 500 mg/kg.
Compound D caused moderate to marked epididymal toxicity. With the exception of decreases in GCA for animals in the
high-dose group of compound D, none of the 3 individual BAs
exhibited statistically significant increases in their serum levels for animals treated with compound C or D (Table 3). The
biological meaning of the decreases observed in GCA with
the high dose of compound D is unknown at this time, and
further experimentation is needed to understand this observation. Because of sample size limitations, we could not analyze
GLDH and total BAs. Although we observed small increases
in ALT and AST in compound C–treated animals, and in ALT
and ALP in compound D–treated animals (Table 3), their livers
were microscopically normal (Supplementary Data).
Discussion
DILI is the single greatest cause for the termination of drug
development candidates and for the withdrawal of approved
drugs from the market (Kaplowitz, 2005; Nathwani et al.,
2005). A well-characterized panel of standard biomarkers
that includes ALT, AST, GLDH, ALP, TBIL, and GGT is used
for monitoring liver injury in preclinical species and clinical
21
practice (Antoine et al., 2009; EMEA, 2010). Nevertheless,
more specific biomarkers capable of distinguishing various
components of liver injury, such as hepatocellular and hepatobiliary damage, are needed for assessing the risk of liver injury
to humans during drug development (Bailey et al., 2012).
Recently, individual BAs, such as CA, GCA, and TCA, and/
or BA profiling have been proposed as potential biomarkers
of DILI in preclinical and clinical studies (Tribe et al., 2010;
Trottier et al., 2011; Yamazaki et al., 2013). In this study, we
set out to evaluate the performance of CA, GCA, and TCA as
potential biomarkers of DILI in rodents by comparing them
with standard biomarkers of liver injury in rats and mice treated
with model hepatotoxicants that cause hepatocellular and/or
hepatobiliary damage.
We have developed a rapid, quantitative LC/MS assay for
the detection of BAs. This assay provided high sensitivity and
specificity to simultaneously measure 3 serum BAs in a short
run time using a single-step sample preparation. This method
was validated to ensure high accuracy and precision, and therefore the reliability of its results as applied to toxicology studies
in rodents. The predominant BA in rats was CA, followed by
GCA at 10-fold lower level than CA. TCA was observed to be
3-fold lower than GCA. On the other hand, the BA levels in
mice were 10- to 100-fold lower than in rats with TCA being
predominant followed by CA and GCA. In humans, the levels
of glycine conjugates have been reported as being higher than
the unconjugated forms, and taurine conjugates reported as
approximately 3-fold lower than the free forms (Trottier et al.,
2011). These species-specific differences are in line with published reports (Alnouti et al., 2008; Ando et al., 2006; Shimada
et al., 2010) and likely due to species-specific variation in BA
biosynthesis and metabolism.
Treatment of rats with model hepatotoxicants GaIN, MC,
ANIT, and APAP resulted in characteristic histopathologic
signs of liver injury. A wide degree of hepatocellular and/
or hepatobiliary effects were detected via histopathology.
The liver injury correlated with significant increases in
serum levels of standard biomarkers of liver injury, including total BAs measured using the total BA enzyme assay
(Advia 2400). As expected, the serum levels of CA, GCA,
and TCA were also significantly increased. The biomarker
response showed a high degree of specificity because none
of the rats treated with vehicle or nonhepatic toxicants (ISO,
compounds C and D) showed treatment-related increases in
biomarker response. These results are in accord with published findings (Wang and Stacey, 1990; Want et al., 2010;
Yamazaki et al., 2013). For example, Yamazaki et al. (2013)
observed increases in plasma levels of TCA and GCA, as
well as increases in urine CA levels, upon dosing rats with
APAP and ANIT. Levels of TCA were found to remain
unchanged or decreased in APAP-treated animals at a later
time point, yet these animals did not exhibit the expected
APAP-induced necrosis. These results further confirm our
observations.
22
Luo et al.
In contrast to model hepatotoxicants, compound A induced
BDH in the absence of hepatocellular damage, and the severity of injury correlated with increasing dose. Treatment of rats
with compound A did not change levels of the standard biomarkers of liver injury, including total BAs. However, a dosedependent increase of conjugated BAs GCA and TCA was
observed. Interestingly, the levels of unconjugated CA remained
unchanged. These changes in both GCA and TCA were not
observed with a structurally related compound, compound B,
that was devoid of any liver pathology and did not exhibit any
changes in any of the other endpoints measured. Compounds
that induce primary BDH are rare, yet the concern is how this
form of injury translates into the clinic, which is currently
unknown. In addition, BDH preclinical findings can be accompanied by other forms of liver injury or other toxicities, making it difficult to discriminate any response in the serum. The
fact that similar increase of conjugated BAs GCA and TCA in
the absence of changes in unconjugated CA was found in mice
treated with compound A (Fig. 3) suggested applicability of the
BA profile as a biomarker of BDH in rodents. Furthermore, our
data indicated the superior specificity of CA, GCA, and TCA
as biomarkers of liver injury when compared with standard biomarkers, such as ALT. In contrast, the total BA assay was not
useful for detection of BDH indicating the need for assessment
of individual BAs and BA profiles. These results are significant
because they are the first to indicate that serum TCA and GCA
are useful tools for detecting BDH in rodents, in the absence of
changes of any other biomarkers.
Close examination of BA profiles, in particular the evaluation of relative proportion of conjugated and unconjugated BAs,
across the compound set revealed that hepatocellular damage
was associated with predominant elevation of unconjugated CA,
whereas hepatobiliary hyperplasia was associated with predominant increase of conjugated BAs GCA and TCA (Fig. 2). For
instance, APAP treatment that produced hepatocellular damage
increased mainly CA, whereas compound A that produced BDH
increased levels of GCA and TCA exclusively. This suggested
that a BA profile characterized by prominent increases in CA may
be a potential marker for hepatocyte damage, whereas a profile
marked by increases in TCA and GCA may serve as an indicator
of biliary injury. Because in many instances, both hepatocellular
and hepatobiliary damage occur together, one could see a continuum of changes in our data. Such as, in the MC study, histopathology revealed extensive hepatocellular necrosis that spilled
over into and involved biliary regions as well, and the LC/MS/
MS analysis demonstrated a significantly altered serum BA profile with TCA being the predominant BA. In the case of ANIT,
massive fold increases of the conjugated BAs indicated severe
biliary injury, whereas the mild fold increases of CA also indicated less intensive hepatocellular injury. The different intensities of fold increases in conjugated and unconjugated BAs may
indicate a different site or severity of hepatic injury. As summarized in Table 4, the increased CA fold changes correlated
well with hepatocellular injury/necrosis across the compound
set. Our findings supported the publication from InnoMed
PredTox Consortium regarding the pathogenesis of liver injury
(Ellinger-Ziegelbauer et al., 2011), which proposed that hepatocyte damage may have interfered with BA metabolism and/or
conjugation, leading to accumulation of toxic unconjugated BAs
in hepatocytes and subsequent leakage into serum, whereas the
biliary toxicity may have arisen from impaired transport of conjugated BAs across the canalicular membrane of hepatocytes or
leakage as a result of bile duct injury. Our observations were also
in accord with the publications from Trottier et al. (2011) and
Meng et al. (1997). Taken together with our findings, these studies indicate that the 3 BA proportions/profiles themselves are
significant, and these profiles are likely to be specific to the site
of damage and provide insight into the mechanisms involved in
different types of liver toxicity. Our data suggested both serum
BA levels and serum BA patterns can be sensitive indicators of
hepatic damage, and patterns in IBAs may be established for
differentiating various types of liver injury and findings, such
as BDH, cholestasis, necrosis, steatosis, and adaptive responses.
For example, in the publication by Yamazaki et al. (2013), the
compounds that induced steatosis had significant increases in
the tauro-conjugated BAs, without increases in any other BAs
measured, suggesting a different profile for steatosis than we
observed for necrosis or BDH.
Currently, modest ALT elevations may necessitate placing a hold upon clinical development of a compound. If these
elevations were better understood through newly developed
biomarker tools, the drug development process could progress
Table 4
Serum CA Fold Increases Corresponded to Hepatocellular Injury
Compound
Compound A
ANIT
MC
GalN
APAP
Hepatocellular Injury
Fold Changes of CA to Control
−
++
++++
++++
++++
1.5
2.8**
5.4**
4.4**
5.3**
−, no finding; ++, mild of severity degree; ++++, marked of severity degree.
**Signifies p < .01 for t-test analysis.
23
Evaluation of Serum Bile Acid Profiles
faster with more confidence in safety to bring needed medicines
to market. At this point, none of the new biomarkers have the
potential to be the universal safety parameter for liver in isolation. Instead, a combination of conventional endpoints and newly
developed markers may provide a panel showing adequate sensitivity and specificity for most preclinical and clinical contexts
of relevance. Biomarkers, such as troponin, myosin light chain,
and fatty acid–binding protein, provide data regarding skeletal
muscle and cardiac injury, which can also induce elevations in
ALT. Individual BAs and markers such as these could be run
concurrently with ALT for better interpretation of ALT elevation
in preclinical toxicology studies. Based on this fast and accurate
LC/MS/MS method, the assessment of these 3 individual BAs
could add information to the assessment of liver toxicity during
compound development. Although more data are needed, our
study indicated hepatotoxicants differentially affect circulating
BAs, resulting in different BA profiles. Particularly, the specific
correlation of CA to hepatocellular necrosis and the correlation of GCA and TCA without changes in CA to BDH could
provide mechanistic insight for a specific toxicity. Additional
compounds need to be tested to verify if this pattern holds
true, and additional BAs, such as CDCA, could be evaluated to
determine if they could add specificity to these findings. In the
clinic, Burkard et al. (2005) reported that conjugated BAs were
elevated in serum from patients with liver disease by differentiated quantification of human individual BAs in serum samples.
Alternations in the BA profile and shifts in BA synthesis have
been reported in various human hepatic disease states, such as
primary biliary cirrhosis, chronic hepatitis C, and nonalcoholic
fatty liver diseases (Crosignani et al., 2007; Lake et al., 2013).
Therefore, the individual BAs as potential biomarkers might be
translatable to clinical practice and may further be used as personalized medicine biomarkers by monitoring an individual’s
serum BA profiles (Shimada et al., 2010).
In conclusion, our data suggest the potential for relationships to be established between IBAs and specific liver toxicity findings. Incorporation of serum individual BA analysis
may enhance conventional safety assessment in preclinical
drug development studies, and BA profiles could be particularly
valuable in situations where traditional clinical chemistry measurements are ambiguous, such as BDH. Further studies in preclinical models, including nonhuman primates, will allow better
understanding of the relationship between IBAs and DILI.
Supplementary Data
Supplementary data are available online at http://toxsci.
oxfordjournals.org/.
Acknowledgments
The work included in this paper would not have been possible without the original work of Carol Fritz and the input of
Dr Michael Aleo. We thank Cindy Drupa, Kim Navetta, David
Wolford, and many others within the Drug Safety Department
at Pfizer. The authors declare no conflict of interest.
References
Alnouti, Y., Csanaky, I. L., and Klaassen, C. D. (2008). Quantitative-profiling
of bile acids and their conjugates in mouse liver, bile, plasma, and urine
using LC-MS/MS. J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci. 873,
209–217.
Amacher, D. E. (2002). A toxicologist’s guide to biomarkers of hepatic response.
Hum. Exp. Toxicol. 21, 253–262. doi:10.1191/0960327102ht247oa.
Amacher, D. E., Adler, R., Herath, A., and Townsend, R. R. (2005). Use of
proteomic methods to identify serum biomarkers associated with rat
liver toxicity or hypertrophy. Clin. Chem. 51, 1796–1803. doi:10.1373/
clinchem.2005.049908.
Amacher, D. E., Schomaker, S. J., and Burkhardt, J. E. (1998). The relationship
among microsomal enzyme induction, liver weight and histological change
in rat toxicology studies. Food Chem. Toxicol. 36, 831–839.
Ando, M., Kaneko, T., Watanabe, R., Kikuchi, S., Goto, T., Iida, T., Hishinuma,
T., Mano, N., and Goto, J. (2006). High sensitive analysis of rat serum bile
acids by liquid chromatography/electrospray ionization tandem mass spectrometry. J. Pharm. Biomed. Anal. 40, 1179–1186.
Antoine, D. J., Mercer, A. E., Williams, D. P., and Park, B. K. (2009).
Mechanism-based bioanalysis and biomarkers for hepatic chemical stress.
Xenobiotica 39, 565–577. doi:10.1080/00498250903046993.
Bailey, W. J., Holder, D., Patel, H., Devlin, P., Gonzalez, R. J., Hamilton, V.,
Muniappa, N., Hamlin, D. M., Thomas, C. E., Sistare, F. D., et al. (2012). A
performance evaluation of three drug-induced liver injury biomarkers in the
rat: Alpha-glutathione S-transferase, arginase 1, and 4-hydroxyphenyl-pyruvate dioxygenase. Toxicol. Sci. 130, 229–244. doi:10.1093/toxsci/kfs243.
Beckwith-Hall, B. M., Nicholson, J. K., Nicholls, A. W., Foxall, P. J., Lindon,
J. C., Connor, S. C., Abdi, M., Connelly, J., and Holmes, E. (1998). Nuclear
magnetic resonance spectroscopic and principal components analysis investigations into biochemical effects of three model hepatotoxins. Chem. Res.
Toxicol. 11, 260–272.
Bentayeb, K., Batlle, R., Sánchez, C., Nerín, C., and Domeño, C. (2008).
Determination of bile acids in human serum by on-line restricted access
material-ultra high-performance liquid chromatography-mass spectrometry.
J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci. 869, 1–8.
Berg, B., Helm, G., Petersohn, L., and Tryding, N. (1986). Cholestasis of pregnancy. Clinical and laboratory studies. Acta Obstet. Gynecol. Scand. 65,
107–113. doi:10.3109/00016348609158363.
Bobeldijk, I., Hekman, M., de Vries-van der Weij, J., Coulier, L., Ramaker, R.,
Kleemann, R., Kooistra, T., Rubingh, C., Freidig, A., and Verheij, E. (2008).
Quantitative profiling of bile acids in biofluids and tissues based on accurate
mass high resolution LC-FT-MS: Compound class targeting in a metabolomics workflow. J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci. 871,
306–313.
Burkard, I., von Eckardstein, A., and Rentsch, K. M. (2005). Differentiated
quantification of human bile acids in serum by high-performance liquid
chromatography-tandem mass spectrometry. J. Chromatogr. B. Analyt.
Technol. Biomed. Life Sci. 826, 147–159.
Crosignani, A., Del Puppo, M., Longo, M., De Fabiani, E., Caruso, D., Zuin,
M., Podda, M., Javitt, N. B., and Kienle, M. G. (2007). Changes in classic
and alternative pathways of bile acid synthesis in chronic liver disease. Clin.
Chim. Acta 382, 82–88. doi:10.1016/j.cca.2007.03.025.
Davis, D. G., and Thompson, M. B. (1993). Nuclear magnetic resonance
identification of the taurine conjugate of 3 alpha,6 beta,7 beta-trihydroxy-5
beta,22-cholen-24-oic acid (tauro-delta 22-beta-muricholate) in the serum
24
Luo et al.
of female rats treated with alpha-naphthylisothiocyanate. J. Lipid Res. 34,
651–661.
Ducroq, D. H., Morton, M. S., Shadi, N., Fraser, H. L., Strevens, C., Morris, J.,
and Thomas, M. A. (2010). Analysis of serum bile acids by isotope dilutionmass spectrometry to assess the performance of routine total bile acid methods. Ann. Clin. Biochem. 47(Pt 6), 535–540. doi:10.1258/acb.2010.010154.
Ellinger-Ziegelbauer, H., Adler, M., Amberg, A., Brandenburg, A., Callanan,
J. J., Connor, S., Fountoulakis, M., Gmuender, H., Gruhler, A., Hewitt, P.,
et al. (2011). The enhanced value of combining conventional and “omics”
analyses in early assessment of drug-induced hepatobiliary injury. Toxicol.
Appl. Pharmacol. 252, 97–111.
EMEA. (2010). Reflection paper on non-clinical evaluation of drug-induced
liver injury (DILI). Available at: http://www.ema.europa.eu/docs/en_GB/
document_library/Scientific_guideline/2010/07/WC500094591.pdf.
Accessed September 03, 2013.
Ennulat, D., Magid-Slav, M., Rehm, S., and Tatsuoka, K. S. (2010). Diagnostic
performance of traditional hepatobiliary biomarkers of drug-induced liver
injury in the rat. Toxicol. Sci. 116, 397–412. doi:10.1093/toxsci/kfq144.
FDA. (2009). Guidance for Industry. Drug-Induced Liver Injury: Premarketing
Clinical Evaluation. Available at: http://www.fda.gov/downloads/Drugs/
GuidanceComplianceRegulatoryInformation/Guidances/UCM174090.pdf.
Accessed September 03, 2013.
Gatti, R., Roda, A., Cerre, C., Bonazzi, D., and Cavrini, V. (1997). HPLCfluorescence determination of individual free and conjugated bile acids in
human serum. Biomed. Chromatogr. 11, 11–15.
Griffiths, W. J., and Sjövall, J. (2010). Bile acids: Analysis in biological fluids
and tissues. J. Lipid Res. 51, 23–41. doi:10.1194/jlr.R001941.
Hagio, M., Matsumoto, M., Fukushima, M., Hara, H., and Ishizuka, S. (2009).
Improved analysis of bile acids in tissues and intestinal contents of rats using
LC/ESI-MS. J. Lipid Res. 50, 173–180. doi:10.1194/jlr.D800041-JLR200.
Janzen, N., Sander, S., Terhardt, M., Das, A. M., Sass, J. O., Kraetzner, R.,
Rosewich, H., Peter, M., and Sander, J. (2010). Rapid quantification of conjugated and unconjugated bile acids and C27 precursors in dried blood spots
and small volumes of serum. J. Lipid Res. doi:10.1194/jlr.D003814.
Kaplowitz, N. (2005). Idiosyncratic drug hepatotoxicity. Nat. Rev. Drug
Discov. 4, 489–499.
Kola, I., and Landis, J. (2004). Can the pharmaceutical industry reduce attrition
rates? Nat. Rev. Drug Discov. 3, 711–715.
Lake, A. D., Novak, P., Shipkova, P., Aranibar, N., Robertson, D., Reily, M.
D., Lu, Z., Lehman-McKeeman, L. D., and Cherrington, N. J. (2013).
Decreased hepatotoxic bile acid composition and altered synthesis in progressive human nonalcoholic fatty liver disease. Toxicol. Appl. Pharmacol.
268, 132–140. doi:10.1016/j.taap.2013.01.022.
Lazarou, J., Pomeranz, B. H., and Corey, P. N. (1998). Incidence of adverse
drug reactions in hospitalized patients: A meta-analysis of prospective studies. JAMA 279, 1200–1205. doi:10.1001/jama.279.15.1200.
Lee, B. L., New, A. L., and Ong, C. N. (1997). Comparative analysis of conjugated bile acids in human serum using high-performance liquid chromatography and capillary electrophoresis. J. Chromatogr. B. Biomed. Sci. Appl.
704, 35–42.
Lin, Y., Si, D., Zhang, Z., and Liu, C. (2009). An integrated metabonomic
method for profiling of metabolic changes in carbon tetrachloride induced
rat urine. Toxicology 256, 191–200.
Martin, F. P., Dumas, M. E., Wang, Y., Legido-Quigley, C., Yap, I. K., Tang, H.,
Zirah, S., Murphy, G. M., Cloarec, O., Lindon, J. C., et al. (2007). A topdown systems biology view of microbiome-mammalian metabolic interactions in a mouse model. Mol. Syst. Biol. 3, 112.
Meng, L.-J., Reyes, H., Palma, J., Hernandez, I., Ribalta, J., and Sjövall,
J. (1997). Effects of ursodeoxycholic acid on conjugated bile acids
and progesterone metabolites in serum and urine of patients with
intrahepatic cholestasis of pregnancy. J. Hepatol. 27, 1029–1040.
doi:10.1016/S0168-8278(97)80147-X.
Nathwani, R. A., Pais, S., Reynolds, T. B., and Kaplowitz, N. (2005). Serum
alanine aminotransferase in skeletal muscle diseases. Hepatology 41,
380–382.
Ostrow, J. D. (1993). Metabolism of bile salts in cholestasis in humans. In
Hepatic Transport and Bile Secretion: Physiology and Pathophysiology
(Tavoloni, N. and Berk, P. D, eds.), pp. 673–712, Raven Press, New York.
Ozer, J. S., Chetty, R., Kenna, G., Palandra, J., Zhang, Y., Lanevschi, A.,
Koppiker, N., Souberbielle, B. E., and Ramaiah, S. K. (2010). Enhancing
the utility of alanine aminotransferase as a reference standard biomarker for
drug-induced liver injury. Regul. Toxicol. Pharmacol. 56, 237–246.
Ozer, J., Ratner, M., Shaw, M., Bailey, W., and Schomaker, S. (2008). The
current state of serum biomarkers of hepatotoxicity. Toxicology 245,
194–205.
Pirmohamed, M., James, S., Meakin, S., Green, C., Scott, A. K., Walley, T.
J., Farrar, K., Park, B. K., and Breckenridge, A. M. (2004). Adverse drug
reactions as cause of admission to hospital: Prospective analysis of 18 820
patients. BMJ 329, 15–19. doi:10.1136/bmj.329.7456.15.
Reyes, H., and Sjövall, J. (2000). Bile acids and progesterone metabolites in intrahepatic cholestasis of pregnancy. Ann. Med. 32, 94–106.
doi:10.3109/07853890009011758.
Scherer, M., Gnewuch, C., Schmitz, G., and Liebisch, G. (2009). Rapid quantification of bile acids and their conjugates in serum by liquid chromatography-tandem mass spectrometry. J. Chromatogr. B. Analyt. Technol. Biomed.
Life Sci. 877, 3920–3925.
Shimada, T., Nakanishi, T., Toyama, A., Yamauchi, S., Kanzaki, A., Fujiwake,
H., Sato, T. A., and Ikegawa, M. (2010). Potential implications for monitoring serum bile acid profiles in circulation with serum proteome for carbon
tetrachloride-induced liver injury/regeneration model in mice. J. Proteome
Res. 9, 4490–4500. doi:10.1021/pr1002388.
Street, J. M., Trafford, D. J., and Makin, H. L. (1985). Extraction and fractionation of bile acids and their conjugates using pre-packed microparticulate silica cartridges (Sep-Pak Sil and Bond-Elut C18). J. Chromatogr. B. Biomed.
Sci. Appl. 343, 259–270.
Suzuki, Y., Kaneko, R., Nomura, M., Naito, H., Kitamori, K., Nakajima, T., Ogawa,
T., Hattori, H., Seno, H., and Ishii, A. (2013). Simple and rapid quantitation of
21 bile acids in rat serum and liver by UPLC-MS-MS: Effect of high fat diet on
glycine conjugates of rat bile acids. Nagoya J. Med. Sci. 75, 57–71.
Thompson, M. B., Blair, P. C., Morris, R. W., Neptun, D. A., Deyo, D. F., and
Popp, J. A. (1987). Validation and application of a liquid-chromatographic/
enzymatic assay for individual bile acids in the serum of rats. Clin. Chem.
33, 1856–1862.
Tribe, R. M., Dann, A. T., Kenyon, A. P., Seed, P., Shennan, A. H., and Mallet,
A. (2010). Longitudinal profiles of 15 serum bile acids in patients with intrahepatic cholestasis of pregnancy. Am. J. Gastroenterol. 105, 585–595.
Trottier, J., Białek, A., Caron, P., Straka, R. J., Milkiewicz, P., and Barbier, O.
(2011). Profiling circulating and urinary bile acids in patients with biliary
obstruction before and after biliary stenting. PLoS One 6, e22094.
Turley, S. D., and Dietschy, J. M. (1978). Re-evaluation of the 3 alpha-hydroxysteroid dehydrogenase assay for total bile acids in bile. J. Lipid Res. 19,
924–928.
Wang, G., and Stacey, N. H. (1990). Elevation of individual serum bile acids
on exposure to trichloroethylene or α-naphthylisothiocyanate. Toxicol. Appl.
Pharmacol. 105, 209–215. doi:10.1016/0041-008X(90)90182-T.
Want, E. J., Coen, M., Masson, P., Keun, H. C., Pearce, J. T., Reily, M. D.,
Robertson, D. G., Rohde, C. M., Holmes, E., Lindon, J. C., et al. (2010).
Ultra performance liquid chromatography-mass spectrometry profiling of
bile acid metabolites in biofluids: Application to experimental toxicology
studies. Anal. Chem. 82, 5282–5289.
Evaluation of Serum Bile Acid Profiles
Xiang, X., Han, Y., Neuvonen, M., Laitila, J., Neuvonen, P. J., and Niemi, M.
(2010). High performance liquid chromatography-tandem mass spectrometry for the determination of bile acid concentrations in human plasma. J.
Chromatogr. B. Analyt. Technol. Biomed. Life Sci. 878, 51–60.
Yamazaki, M., Miyake, M., Sato, H., Masutomi, N., Tsutsui, N., Adam, K.-P.,
Alexander, D. C., Lawton, K. A., Milburn, M. V., Ryals, J. A., et al. (2013).
Perturbation of bile acid homeostasis is an early pathogenesis event of drug
induced liver injury in rats. Toxicology and Applied Pharmacology 268,
79–89. doi:10.1016/j.taap.2013.01.018.
25
York, M., Scudamore, C., Brady, S., Chen, C., Wilson, S., Curtis, M., Evans,
G., Griffiths, W., Whayman, M., Williams, T., et al. (2007). Characterization
of troponin responses in isoproterenol-induced cardiac injury in the Hanover
Wistar rat. Toxicol. Pathol. 35, 606–617. doi:10.1080/01926230701389316.
Zhang, J., Knapton, A., Lipshultz, S. E., Weaver, J. L., and Herman, E. H.
(2008). Isoproterenol-induced cardiotoxicity in sprague-dawley rats:
Correlation of reversible and irreversible myocardial injury with release of
cardiac troponin T and roles of iNOS in myocardial injury. Toxicol. Pathol.
36, 277–278. doi:10.1177/0192623307313010.