![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 288, Issue 1, 326-334, January 1999
Department of Pediatrics, Division of Basic Sciences, National Jewish Medical and Research Center, Denver, Colorado (P.W., J.A.H., R.C.M.); and Department of Gastroenterology-Hepatology, University of Colorado Health Sciences Center, Denver, Colorado (B.B., D.G.)
 |
Abstract |
|---|
Top
Abstract
Introduction
Procedures
Results
Discussion
References
|
|---|
Leukotriene B4 (LTB4) is a potent lipid
mediator of the inflammatory response whose biological half-life is
believed to be mediated principally by metabolism to inactive forms
either in the tissue of origin or in the liver. Pathways of metabolic
degradation of LTB4 along with structural identification of
metabolites have been elucidated previously in isolated rat liver
cells, human keratinocytes, human polymorphonuclear leukocytes, and
cultured HepG2 cells. Research advances in human liver transplantation and preservation have made isolated human hepatocytes available for
studying the metabolism of LTB4 in vitro.
LTB4 was added to plated human hepatocytes from
three different subjects for 24-h periods whereupon the substrate was
analyzed by high-performance liquid chromatography coupled with
scintillation counting, UV spectroscopy, and negative ion electrospray
ionization tandem mass spectrometry. Each set of hepatocytes yielded a
different distribution of metabolites, but several metabolites appeared in all three sets of cells. These central metabolites included the
previously identified 20-carboxy-LTB4 and
18-carboxy-LTB4, implicating the presence in the liver of
specific P-450-mediated 
-oxidation as well as the enzymes involved
in 
-oxidation from the -terminus. Each set of hepatocytes
produced the metabolite 10,11-dihydro-20-COOH-LTB4, a
product of the 12-hydroxyeicosanoid dehydrogenase/
10
reductase pathway. Glucuronides of LTB4 and several
metabolites were found, which represents the first description of
glucuronidation as a pathway of LTB4 metabolism. Finally, a
series of novel metabolites were observed corresponding to
-oxidation from the carboxyl terminus of LTB4.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Procedures
Results
Discussion
References
|
|---|
Leukotriene
B4 (LTB4) is thought to
play an important role as a lipid mediator of the inflammatory response
and to be a potent chemotactic factor stimulating the human
polymorphonuclear leukocyte through a G protein-linked receptor
(Yokomizo et al., 1997
). LTB4 is formed from
arachidonic acid (released from cell membrane phospholipids by
phospholipase A2) through a series of enzymatic
transformations. The direct product of 5-lipoxygenase metabolism of
arachidonic acid is leukotriene A4
(LTA4) (Ford-Hutchinson et al., 1994; Drazen et
al., 1995), which is enzymatically hydrated by
LTA4 hydrolase (EC 1.1.1.) (Mueller et al.,
1995). It is now recognized that many cells export
LTA4, such as the human polymorphonuclear
leukocyte, and that other cells in the immediate vicinity, which
contain LTA4 hydrolase, generate the biologically
active LTB4. The biological half-life of
LTB4 is mediated by metabolic degradation either in the tissue of origin or through hepatic metabolism if
LTB4 escapes into the circulatory system. The rat
liver is known to extract LTB4 from the blood
with a high-affinity uptake system (Keppler, 1992) and then metabolize
LTB4 into a variety of products (Shirley and
Murphy, 1990). Considerably less is known about the metabolism of
LTB4 in the human subject or from cells derived from human liver.
The metabolism of LTB4 has been studied in
several human cell types including the human polymorphonuclear
leukocyte (Shak and Goldstein, 1984; Soberman et al., 1988),
keratinocyte (Wheelan et al., 1993), and cultured HepG2 (Wheelan and
Murphy, 1995b) among others. Several pathways of
LTB4 metabolism now have been recognized. The
first pathway to be characterized was the -oxidation of
LTB4 into 20-hydroxy-LTB4
by cytochrome P-450. The human neutrophil was shown to express a
specific NADPH-dependent microsomal P-450 (Kikuta et al., 1993), which
now has been termed CYP4F3 and found on human chromosome 19 (Kikuta et
al., 1998). This enzyme is also thought to catalyze the formation of
20-carboxy-LTB4 by further oxidation of the 
carbon atom. Interestingly, this specific cytochrome CYP4F3 is not
expressed in rat or human hepatocytes (Romano et al., 1987), but rather
a closely related cytochrome P-450 of the CYP4F superfamily is in these
hepatocytes (Kikuta et al., 1994 and Kawashima et al., 1997).
LTB4 is metabolized by P-450-mediated -oxidation in the rat hepatocyte to
20-hydroxy-LTB4, but is then oxidized further by
alcohol dehydrogenase- and aldehyde dehydrogenase-dependent reactions
into 20-carboxy-LTB4 (Baumert et al., 1989;
Shirley et al., 1992). Detailed studies of the further metabolism of
20-carboxy-LTB4 in the isolated rat hepatocyte
revealed that a second major metabolic pathway, -oxidation,
proceeded after -oxidation and after formation of the CoA ester at
the carboxyl group at carbon-20, ultimately leading to a
chain-shortened and reduced 16-carboxy-LTB3
(Shirley and Murphy, 1990).
Human cells also have been found to express a third pathway of
LTB4 metabolism termed the 12-hydroxyeicosanoid
dehydrogenase/11 reductase pathway (Wainwright
and Powell, 1991; Yokomizo et al., 1993), which leads to the initial
oxidation of a hydroxyl group at carbon-12 to the 12-oxo derivative,
which then is reduced at the adjacent double bond, yielding a series of
10,11 dihydro-LTB4
metabolites (Powell and Gravelle, 1989). This pathway appears to be in
competition with the cytochrome P-450 pathway for the
LTB4 substrate. In cells that do not express a
specific cytochrome P-450 isozyme capable of acting on
LTB4, the 12-hydroxyeicosanoid dehydrogenase
pathway can account for the majority of metabolic products.
Recent developments in human hepatic cell isolation and tissue culture from livers obtained through organ donation have made possible studies of in vitro metabolic transformations under controlled conditions. Such human hepatocyte cultured cells were used to investigate the biochemical conversion of LTB4 and provide some insight into the potential competent metabolic pathways in human liver cells integrating enzymatic systems located in cytosol, endoplasmic reticulum, mitochondria, and peroxisomes.
| |
Experimental Procedures |
|---|
|
Top
Abstract
Introduction
Procedures
Results
Discussion
References
|
|---|
Materials. LTB4 and [6,7,14,15-2H4]LTB4 (d4 LTB4) were purchased from Biomol Research Laboratories (Plymouth, PA). [5,6,8,9,11,12,14,15-3H8]LTB4 (195 Ci/mol) was purchased from Dupont/New England Nuclear (Boston, MA). All solvents were high-performance liquid chromatography (HPLC) grade and obtained from Fisher Scientific (Fair Lawn, NJ).
Isolation and Culture of Human Hepatocytes. Human livers were obtained through a collaboration with the Organ Bank International Institute for the Advancement of Medicine (Exton, PA). Human livers used for hepatocyte isolation were procured under the United Network for organ-sharing guidelines but determined unsuitable for orthotopic liver transplantation because of physical trauma to a portion of the organ, anoxy, or a high interstitial fat content. On occasion, lobes also were made available for isolation when, because of size considerations, the entire liver was not given to the recipient.
Hepatocytes were isolated by modification of Seglen's perfusion technique (Seglen, 1976). Briefly, human liver pieces were cannulated
and perfused with Hanks' balanced salt solution (HBSS) without calcium
or magnesium ions. This HBSS solution contained 4.2 mM sodium
bicarbonate, 0.5 mM EGTA, and 9.0 mM HEPES buffer (Sigma Chemicals, St.
Louis, MO). Perfusion was continued with L-15 (Leibovitz) medium (Sigma
Chemical, St. Louis, MO) containing 0.05% collagenase. After
collagenase digestion, the softened liver was gently teased apart and
the suspension was filtered through a 100-µm nylon mesh and washed
three times in ice-cold L-15. Hepatocytes were plated at densities of 3 to 5 × 104 cells/cm2
on modified culture plates with a minimal essential medium/Waymouth's medium, 3:1 tyrosine-free (GIBCO, Gaithersburg, MD) containing 10%
fetal bovine serum (HyClone Laboratories, Inc., Logan, UT). These
modified culture plates consisted of surface-modified polystyrene containing a high density and variety of nitrogen and oxygen containing functional groups (Primeria; Becton Dickinson, Franklin Lakes, NJ).
These functional groups on the plate surface are believed to enhance
cell adhesion onto the plate (Chilkoti et al., 1995). After an 18- to
24-h incubation, cells were placed in a serum-free, defined medium to
maintain the stable differentiation of normal human hepatocytes (Ledley
et al., 1993). Modified culture dishes and tyrosine-free media were
used to select against the proliferation of fibroblasts and
nonparenchymal cells. Hepatocytes were incubated in a 92.5% air/7.5%
CO2 humidified environment at 37°C during the
course of all experiments. Experiments with LTB4
were conducted within 24 to 48 h after placing the cells in a
serum-free, hormonally defined medium. Cells were 70 to 100% confluent
at the time of experiments and still maintained a somewhat flattened
but round morphology.
Incubation of Human Hepatocytes with LTB4.
LTB4 and
[3H8]LTB4 were evaporated to
dryness under nitrogen and reconstituted in HBSS to a final
concentration of 12 µM LTB4 (0.2 µCi/ml). Culture media
were removed from the plated hepatocytes and replaced with HBSS
containing LTB4 media (3 ml for hepatocytes in wells and 10 ml for hepatocytes in 75-cm2 culture flasks) and then
incubated at 37°C for 24 h. The HBSS incubation buffer had been
used in previous rat hepatocyte studies (Shirley and Murphy, 1990;
Wheelan and Murphy, 1995a) to reduce binding of LTB4
to extracellular proteins in serum-containing buffers. Light
microscopic examination of cells after the 24-h LTB4 light
incubation did not reveal any morphologic alterations. Supernatants
were decanted and the hepatocytes were washed with ice-cold ethanol
(3-5 ml) several times, with the ethanol washes added to supernatants.
The supernatant/ethanol solution was brought to 80% ethanol and kept
at 0°C for at least 3 h to allow precipitation of protein.
Samples were centrifuged and supernatants were decanted and evaporated
to near dryness by rotary evaporation. Samples were reconstituted in
reversed-phase (RP)-HPLC solvents at proportions of initial gradient,
filtered, and kept at 0°C until analysis.
Metabolite Purification. Samples were analyzed by RP-HPLC using an Ultremex C18 column (4.6 × 250 mm; Phenomenex, Torrance, CA). The initial mobile phase consisted of methanol/aqueous 0.05% acetic acid (8.7 mM), with pH adjusted to pH 5.0 with ammonium hydroxide (10/90) at a flow rate of 1 ml/min. A linear gradient was started immediately to 70% methanol at 60 min followed by a second linear gradient to 95% methanol at 75 min. Column effluent was monitored using UV detection and on-line radioactivity monitoring. For analysis of samples by electrospray mass spectrometry, an Ultremex C18 column (1.0 × 150 mm) was used with the above mobile phase conditions at a flow rate of 50 µl/min, and the effluent was introduced into the mass spectrometer by a 0.5-m, 50-µm fused silica capillary.
Mass Spectrometry.
Mass spectrometry was performed on a
Sciex API III+ triple quadrupole mass spectrometer
(Perkin-Elmer Sciex, Ontario, Canada) operating in the negative ion
mode with a spray voltage of 
2600 to 3400 V and an orifice voltage
of 60 V. Highly purified air was used as nebulizer gas to reduce the
glow discharge at these negative voltages. Collisionally induced
decomposition (CID) spectra were obtained using an offset potential of
20 eV and argon as the collision gas at a thickness equivalent to
200 × 1013 molecules/cm2. Mass spectra
were obtained by scanning from m/z 100 to
1000 in 3 s as previously described (Wheelan and Murphy, 1995b).
| |
Results |
|---|
|
Top
Abstract
Introduction
Procedures
Results
Discussion
References
|
|---|
Incubation of LTB4 (12 µM in 3 ml of media) for 24 h with isolated human hepatocytes (1 × 106) resulted in recovery of 60.7% of LTB4 or LTB4 metabolites in the supernatant as determined by recovery of radioactivity. RP-HPLC analysis with radioactivity monitoring revealed numerous LTB4 metabolites (Fig. 1A). In one series of experiments with a single human hepatocyte preparation, eight separate flasks were prepared with identical aliquots of cells plated into the serum-free, defined media. After an initial 24-h incubation, one flask was taken for an LTB4 incubation and metabolite profile analysis. At 24-h sequential intervals for 1 week, an additional flask was taken for LTB4 metabolite profiling. There was no difference observed in the metabolic profile for any of the time points, suggesting no major cellular differentiation altered LTB4 metabolism while culturing cells in the defined media.
|
Several metabolites displayed a UV absorption spectrum indicative of a conjugated triene structure with an absorption maximum at 270 nm (Fig. 1B), whereas others contained a conjugated diene structure as evidenced by an absorption maximum at 230 nm (Fig. 1C). Negative ion electrospray mass spectral analysis of metabolites resulted in identification of several previously known metabolites as well as elucidation of several novel structures.
Metabolite A, 18-COOH-LTB4.
The RP-HPLC
retention time, triene chromophore, and molecular anion at
m/z 337 for metabolite A suggested the
previously identified LTB4 metabolite,
18-COOH-LTB4. The CID mass spectrum of the molecular anion
generated by electrospray ionization has not been reported previously
and contained many of the fragment ions observed in the CID mass
spectrum of LTB4 (Wheelan et al., 1996a), including abundant ions at m/z 195, m/z 71, and
m/z 59 and a lesser abundant ion at
m/z 129 (Fig.
2A). A fragment ion observed at
m/z 141, analogous to
m/z 169 observed in the CID mass spectrum
of 20-COOH-LTB4, was most likely a result of fragmentation
of the C(11)-C(12) bond with formation of an aldehyde and charge
retention on the -terminus. Fragmentation of the same bond with
charge retention on the carboxyl terminus results in formation of the
ion at m/z 195. An abundant ion at
m/z 206 was most likely formed by an
oxy-Cope rearrangement with loss of H2O as observed
previously in the formation of odd electron fragment ions containing
the 3-OH 1,5-diene moiety (Fig. 3)
(Wheelan et al., 1996b).
|
|
Metabolite B, 10,11-Dihydro-18-COOH-LTB4.
A
relatively minor yet novel metabolite was observed eluting with
slightly longer HPLC retention time than 18-COOH-LTB4 (Fig. 1A). This metabolite retained a diene chromophore and a molecular anion
at m/z 339, two mass units higher than
that for 18-COOH-LTB4. This suggested saturation of one
double bond present in the triene unit of 18-COOH-LTB4. In
addition to ions at m/z 321 (loss of H2O), m/z 303 (loss of
2H2O), and m/z 277 (loss of
H2O and CO2), the CID mass spectrum of
metabolite B contained abundant ions at
m/z 115 and
m/z 141 (Fig. 2B). The ion at
m/z 141 likely was identical with that
formed after collisional activation of the 18-COOH-LTB4
anion whereas the presence of an abundant ion at m/z 115 had been recognized previously as
indicative of 10,11-dihydro LTB4 metabolites (Wheelan et
al., 1996a). This ion was reported to arise from the cleavage of the
C(5)-C(6) bond with transfer of the hydroxyl proton, formation of a C-5
aldehyde, and charge retention at C(1). These data suggested that
metabolite B differed from 18-COOH-LTB4 by saturation of
the 10,11 double bond and that metabolite B was
10,11-dihydro18-carboxy-LTB4.
Metabolite C, Glucuronide Conjugate of
20-COOH-LTB4.
With an observed molecular anion at
m/z 541 and the presence of a conjugated
triene chromophore, the structure of metabolite C was consistent with a
novel glucuronide conjugate of 20-COOH-LTB4. The CID mass
spectrum at a collision energy of 20 eV showed a low abundant fragment
ion at m/z 523 (loss of H2O)
and a second low abundant ion at m/z 365, consistent with the molecular anion of 20-COOH-LTB4. The
facile loss of the glucuronide moiety (loss of 176 Da) during
collisional activation did not allow determination of the position of
glucuronidation, but this neutral ion loss is quite indicative of the
presence of a glucuronic acid conjugate (Straub et al., 1987).
Metabolite D, 12-Oxo-10,11-Dihydro-20-COOH-LTB4.
Metabolite D eluted by RP-HPLC approximately 3 min earlier than
standard 20-COOH-LTB4 yet the observed molecular anion,
m/z 365, was identical with that of
20-COOH-LTB4. The presence of a conjugated diene
chromophore for this metabolite instead of a triene chromophore
suggested a 10,11-dihydro metabolism and, in combination with the
molecular weight, also suggested the presence of the C(12) oxo moiety.
Also, a 12-oxo-10,11-dihydro modification of 20-COOH-LTB4
would be expected at an earlier RP-HPLC retention time when only
methanol was used as the organic phase (Wheelan et al., 1993). The CID
mass spectrum displayed fragment ions at m/z 347 (loss of H2O) and
m/z 303 (loss of H2O and
CO2) (Fig. 4A). A prominent
fragment ion at m/z 115 was consistent
with a 10,11-dihydro modification resulting in fragmentation of
C(5)-C(6) with charge retention at C(1), and the fragment ion at
m/z 249 was likely a result of the same
fragmentation but with charge retention at C(20).
|
Metabolite E, 20-COOH-LTB4.
The RP-HPLC retention
time, UV triene chromophore, and molecular anion at
m/z 365 for the radioactive metabolite
eluting at 39 min were identical with authentic
20-COOH-LTB4. The CID mass spectrum showed characteristic
ions at m/z 347 (loss of
H2O), m/z 329 (loss of
2H2O), and m/z 303 (loss of
H2O and CO2) as well as significant ions
resulting from C(11)-C(12) bond fragmentation with charge retention on
the carboxyl terminus (m/z 195) or on the
-carboxyl terminus (m/z 169). Abundant
ions also were observed at m/z 141 [C(12)-C(13) bond fragmentation, -carboxyl terminus charge] and at
m/z 129, which also is observed in the
CID mass spectrum of LTB4 (Wheelan et al., 1996a).
Metabolite F, 10,11-Dihydro-20-COOH-LTB4.
Metabolite F, at a slightly longer RP-HPLC retention time than
20-COOH-LTB4, displayed a diene chromophore and a molecular anion at m/z 367, 2 Da higher than
20-COOH-LTB4. This suggested the saturation of one double
bond in the conjugated triene of 20-COOH-LTB4. The CID mass
spectrum revealed a prominent ion at m/z
115 [C(5)-C(6) bond fragmentation], which is characteristic of the
10,11-dihydro-LTB4 structure (Fig. 4B). The fragment ion at
m/z 251 also was likely due to the same
bond fragmentation but with charge localized on the -terminus.
Fragment ions at m/z 141 and
m/z 169 also were observed in the CID
mass spectrum of 20-COOH-LTB4 and result from charge
localization at the -carboxyl terminus (Wheelan and Murphy, 1995b).
These data were consistent with a structure of metabolite F as
10,11-dihydro-20-carboxy-LTB4.
Metabolite G, 18-COOH-10-Oxo-4,6,12-Octadecatrienoic
Acid.
The molecular anion for metabolite G at
m/z 321, 44 atomic mass units (amu) less
than for metabolite D, suggested a chain-shortened metabolite resulting
from -oxidation at the carboxyl terminus with loss of the C(5)
hydroxyl substituent. The diene chromophore for this metabolite also
was consistent with a chain-shortened analog of
12-oxo-10,11-dihydro-20-COOH-LTB4. The most prominent ion
in the CID mass spectrum, m/z 277, most
likely resulted from loss of CO2, and a second loss of
CO2 was suggested by the ion at
m/z 233 (Fig.
5A). The facile loss of 44 amu for the
chain-shortened metabolites, observed earlier in the CID spectra of
10-hydroxy-4,6,8,12-octadecatetraenoic acid (10-HOTE) and
10-hydroxy-4,6,12-octadecatrienoic acid (10-HOTrE) (Wheelan and Murphy,
1995b), also was apparent in the low-mass ions at
m/z 135 and 123. The ion at
m/z 135 was likely due to fragmentation
of the C(10)-C(11) bond of the C(1) decarboxylated anion whereas the
ion at m/z 123 may have arisen by
fragmentation of the C(9)-C(10) bond of the -terminus decarboxylated
anion. This metabolite, identified as
18-carboxy-10-oxo-4,6,12-octadecatrienoic acid, has not been observed
in previous in vitro metabolic studies of LTB4.
|
Metabolite H, 18-COOH-10-hydroxy-4,6,12-octadecatrienoic Acid.
Even though it is a minor component, the molecular anion of metabolite
H at m/z 323 was readily observed,
suggesting -oxidation from the carboxyl terminus and a structure
similar to metabolite G. The two additional protons in this molecule
suggested reduction of the ketone moiety in metabolite G and the
presence of a C(10) hydroxyl substituent. Product ions observed in the
CID mass spectrum were consistent with fragmentation of the C(10)-C(11)
bond and formation of a terminal alkene and an -terminal carboxylate
anion (m/z 141). Cleavage of the
C(9)-C(10) bond would lead to formation of a terminal aldehyde and the
-terminal carboxylate anion (m/z 169)
(Fig. 5B). The UV data (
max = 230 nm) and electrospray
tandem mass spectrometry were consistent with identification of this previously unidentified metabolite as
18-carboxy-10-hydroxy-4,6,12-octadecatrienoic acid.
Metabolites I and K, Glucuronide Conjugates of LTB4. Metabolites I and K both displayed triene chromophores and a molecular anion at m/z 511. Decomposition of the molecular anion revealed loss of 176 Da with formation of a fragment ion at m/z 335. This neutral ion loss, vide supra, UV spectra, and molecular weight were consistent with isomeric glucuronide conjugates of LTB4. The exact position of the glucuronide attachment could not be ascertained in this experiment.
Metabolite J, Glucuronide Conjugate of 10,11-Dihydro-LTB4. Metabolite J displayed a molecular anion during electrospray ionization at m/z 513, which, upon collisional activation, fragmented to m/z 337. This was consistent with the glucuronide conjugate of the previously identified metabolite, 10,11-dihydro-LTB4. No further structure information was possible for this metabolite because of the inability to yield structurally relevant ions indicative of the position of glucuronide conjugation.
Human hepatocytes in two additional hepatocyte preparations were incubated with [3H8]LTB4, unlabeled LTB4, and the stable isotopically labeled [6,7,14,15-2H4]LTB4 (d4-LTB4). Recovery of LTB4 in the cell supernatant (based on radioactivity) was 90.8%. Analysis of cell supernatants by RP-HPLC with radioactivity monitoring showed more extensive metabolism than observed in the first experiment, with most of the radioactivity eluting within the first 10 min. Many of the metabolites identified in the first experiment were also present, and the ratio of d0:d4 of the molecular anions was used to verify these metabolites as derived from LTB4. The ratio of d0:d4 LTB4 used in the incubation was approximately 100:75 as assessed from the ratio of molecular anions at m/z 335 and 339 after mass spectral analysis of the initial incubation media. An identical ratio was observed for m/z 337:341 at the appropriate RP-HPLC retention time for metabolite A, 18-COOH-LTB4. Likewise, 10,11-dihydro-18-COOH-LTB4 (m/z 339:343), the glucuronide conjugate of 20-COOH-LTB4 (m/z 541:545) and two peaks at m/z 365:369 corresponding to metabolites D and E, and ions at m/z 367:371 for 10,11-dihydro-20-COOH-LTB4 also showed the expected ratio of d0:d4 for LTB4-derived metabolites. Two peaks also were observed for the glucuronide conjugates of LTB4 and displayed the same ratio of ions at m/z 511:515 as the starting d0:d4 LTB4. Collisional activation of both of these ions resulted in the expected loss of 176 amu with formation of ions at m/z 335 and 339, respectively, corresponding to the molecular anions of LTB4 and d4-LTB4 (Fig. 6).
|
Metabolites L and M,
6-Cysteinyl-5,12-Dihydroxy-7,9,14-Eicosatrienoic Acid,
e-LTB3.
Two additional metabolites, not identified in
the first experiment, were identified by the ratio of
d0:d4 molecular
anions at RP-HPLC retention times slightly longer than the two
glucuronide conjugates of LTB4. Both peaks contained the
molecular anion at m/z 456, with
m/z 460 for the deuterated analog,
suggesting the addition of cysteine to LTB4. The presence
of two such metabolites separated approximately 2 min in the RP-HPLC
analysis (Table 1) was consistent with
stereoisomeric compounds, and both fragmented after collisional
activation to identical mass spectra. The CID mass spectra revealed
prominent ions resulting from loss of 87 amu, fragmentation of the
cysteine moiety, from the molecular anions resulting in ions at
m/z 369 and
m/z 373 for the unlabeled and
d4-labeled metabolites, respectively (Fig.
7. Loss of H2O from these two
ions was also observed, resulting in ions at
m/z 351 and
m/z 355. Loss of the cysteine moiety
likely was responsible for the observed ion at
m/z 335 with additional losses of
H2O (m/z 317) and
CO2 (m/z 273) as observed in
the CID mass spectrum of LTB4. The collisional mass
spectrum of the d4 metabolite revealed an
ion at m/z 339, consistent with the
expected molecular anion of d4
LTB4, and a loss of 19 amu from this ion at
m/z 320, suggesting the loss of
H2O also involved one of the deuterated positions. Loss of
the cysteine moiety with charge retained on the amino acid rather than
on the LTB4 backbone also was apparent as evidenced by the
ion at m/z 120. A fragment ion at
m/z 229 also was observed in the CID mass
spectrum of a previously identified dipeptide conjugate of
LTB4, d-LTB3 (Wheelan et al.,
1993). Formation of this ion likely was a result of fragmentation of
the C(11)-C(12) bond of the initially formed
m/z 369 ion. This ion was shifted by 2 amu to m/z 231 in the CID mass spectrum
of the deuterated analog, reflecting the presence of two deuterium
atoms, at positions C(6) and C(7), and loss of two deuterium atoms, at
positions C(14) and C(15). The ion at m/z
195 also was present in the CID mass spectrum of LTB4 and
likely resulted from initial loss of the cysteine moiety followed by
fragmentation of the C(11)-C(12) bond as for LTB4 (Wheelan
et al., 1996a). This ion was shifted to
m/z 197 in the CID mass spectrum of the
deuterated analog, consistent with the C(1)-C(11) backbone structure.
|
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Procedures
Results
Discussion
References
|
|---|
A number of specific metabolic pathways have been described in
animal as well as human cells that can participate in the
metabolism of LTB4. These include
pathways mediated by specific cytochrome P-450 isozymes, alcohol
dehydrogenase, 12-hydroxyeicosanoid dehydrogenase, as well as
-oxidation from both the C-1 carboxyl and methyl terminus of
LTB4. The combination of these numerous pathways
results in a complex profile of metabolites observed during in vitro
incubations of LTB4 with isolated cells (Murphy
and Wheelan, 1997; Wheelan and Murphy, 1998). A central role for
hepatic metabolism of LTB4 has emerged with
studies of a high-affinity uptake system for LTB4
(Keppler, 1992). Initial studies of the metabolic transformation of
LTB4 in isolated rat hepatocytes suggested a
predominant pathway for cytochrome P-450-dependent metabolism because
all metabolites were derived from initial -oxidation with no
evidence of an operational 12-hydroxy dehydrogenase eicosanoid pathway.
The majority of metabolites in rat hepatocytes were the result of a
combination of cytochrome P-450-dependent oxidation followed by
-oxidation from the -terminus. Detailed metabolic studies with
human cells such as human keratinocytes, macrophages, and lung-derived
cells have suggested that the 12-hydroxyeicosanoid dehydrogenase
pathway may play a more important role in nonhepatic human cells.
Investigation of the metabolic disposition of
LTB4 in human subjects or even within human
hepatic tissue has been somewhat difficult to pursue. Recent
developments in isolation and maintenance of isolated human hepatocytes
(Seglen, 1976; Ledley et al., 1993) have enabled studies of
LTB4 metabolism in vitro with human hepatocytes. The results of these studies suggest that some variability exists in
ultimate metabolism of LTB4 between each
hepatocyte preparation. Significant interindividual variation in
biotransformation of xenobiotics has been reported in primary human
hepatocyte cultures (Straub et al., 1987). Such differences had been
attributed to natural polymorphisms in enzymatic pathways and specific
enzymes that arise in outbred populations. Such variations have been
suggested to be responsible for individual differences in responses to
toxicological and environmental exposures including particular adverse
reactions. Nonetheless, it is clear that these hepatocytes express many
of the enzymes responsible for LTB4 metabolism
observed in other animal as well as human cell types. Therefore, some
qualitative insight into the potential for LTB4
metabolism within human hepatocytes was obtained after detailed
structural analysis of the metabolites formed in these experiments. For
this purpose, electrospray tandem mass spectrometry proved to be of
great value in providing critical information to suggest structural
assignment of these metabolites. This information in combination with
UV absorption and UV retention time was used to identify more than 20 different metabolites of LTB4 formed by cultured
human hepatocytes after in vitro incubation (Table 1).
One of the features of LTB4 metabolism in
cultured human hepatocytes was that overall metabolism was relatively
slower than it was in freshly isolated hepatocytes. This might
suggest the loss of enzymatic activity in the human hepatocyte
preparations and uncontrolled variables that could contribute to this
included cell density in plating, cell viability, and preparation
delays before suspension experiments. The LTB4
metabolite profile in the rat was largely determined by cytochrome
P-450-dependent -oxidation. There are numerous examples of the loss
of specific cytochrome P-450 isozymes from cells in culture (Glatt et
al., 1987; Utesch and Oesch, 1992), and possibly a cytochrome P-450
isozyme specific for LTB4 was lost in these
preparations of human hepatocytes. Nevertheless, some of the most
abundant metabolites after a 24-h incubation of
LTB4 with human hepatocytes were -oxidation
products. The presence of the 12-hydroxy eicosanoid
dehydrogenase/10,11 reductase pathway in human
hepatocytes was clearly evident by the identification of a series of
10,11 reduced metabolites that retained the conjugated diene structure
and characteristic UV absorption maximum at 235 nm. This pathway of
LTB4 metabolism was identified previously in
several human cell types, including keratinocytes (Wheelan et al.,
1993), human monocytes (Schonfeld et al., 1988), as well as cells in
the lung parenchyma (Kumlin and Dahlén, 1990). This has been
found to be a general pathway of eicosanoid metabolism involving
initial formation of an 
,-unsaturated ketone from a secondary
carbonyl group adjacent to a trans double bond. Such a metabolic
transformation occurs for prostaglandins in the formation of the
15-oxo-prostaglandin intermediates catalyzed by 15-hydroxyprostaglandin
dehydrogenase (Matsuo et al., 1997).
Several glucuronide metabolites of LTB4 were
observed in these studies, and these represent the first identification
of LTB4 glucuronide conjugates observed during in
vitro incubations. Although the exact position of glucuronidation could
not be ascertained, several isomers were observed, suggesting that each
hydroxyl group could be conjugated as well as the possibility of
formation of acyl glucuronides. These results suggest that glucuronide
conjugation may play a more important role in the elimination of
LTB4 and its metabolites into the urine of
humans. Interestingly, primary cultures of hepatocytes have been found
to retain both glucuronide and sulfate conjugation pathways of
xenobiotics and possibly provide a faithful picture of phase II
reactions that occur in vivo (Mertes et al., 1985; Mennes et al.,
1994).
Another pathway for LTB4 metabolism is
-oxidation from the carboxyl terminus. Previous studies of
LTB4 metabolism with HepG2 cells (Wheelan and
Murphy, 1995b) revealed this route of -oxidation as the most
prevalent with formation of 10-HOTE. The metabolic events leading to
this metabolite are poorly understood and involve loss of the oxygen
substituent at C-5. As summarized in Fig.
8, some of the more abundant metabolites
(Table 1, 18-COOH-10-oxo-OTrE and 18-COOH-10-HOTrE) must arise from
involvement of -oxidation, the 12-hydroxyeicosanoid dehydrogenase
pathway, as well as -oxidation from the carboxyl terminus with loss
of the C-5 oxygen atom. Thus, human hepatocytes express multiple
enzymatic pathways to metabolically transform
LTB4. Although a quantitative picture of
LTB4 metabolism in the human liver in vivo cannot
be ascertained from these studies, it is clear that in the human
hepatocyte, LTB4 is processed sequentially by
multiple oxidative pathways as well as by glucuronidation.
|
| |
Footnotes |
|---|
Accepted for publication August 26, 1998.
Received for publication May 18, 1998.
1 This work was supported in part by grants from the National Institutes of Health (HL25785 and AG00619).
2 Current address: Glaxo/Wellcome, 5 Moore Drive, Research Triangle Park, NC 27709.
Send reprint requests to: Dr. Robert C. Murphy, National Jewish Medical and Research Center, 1400 Jackson Street, Denver, CO 80206. E-mail: murphyr{at}njc.org
| |
Abbreviations |
|---|
LTB4, leukotriene B4; HPLC, high-performance liquid chromatography; RP, reversed-phase; amu, atomic mass unit; HBSS, Hanks' balanced salt solution; CID, collisionally induced decomposition; 10-HOTE, 10-hydroxy-4,6,8,12-octadecatetraenoic acid; 10-HOTrE, 10-hydroxy-4,6,12-octadecatrienoic acid.
| |
References |
|---|
|
Top
Abstract
Introduction
Procedures
Results
Discussion
References
|
|---|
-hydroxylase from human liver.
FEBS Lett
348:
70-74[Medline].6-reductase metabolic pathway.
J Biol Chem
270:
19845-19852
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
