| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Article |
From the John Cochran Veteran’s Administration Medical
Center,*
St. Louis; the Departments of Internal
Medicine
and
Pathology,
Saint Louis University School of
Medicine, St. Louis; and the Department of
Medicine,
Washington University School of
Medicine, St. Louis, Missouri
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In vivo and in vitro studies of myeloid cells
indicate that myeloperoxidase is synthesized at the promyelocytic stage
of differentiation.4,7
The enzyme represents 
5% of
neutrophil and 1% of monocyte protein but has long been believed to
be absent from macrophages.8
Indeed, myeloperoxidase
protein and mRNA rapidly disappear from freshly isolated human
monocytes as they differentiate into macrophages in
vitro.9
In contrast, recent studies suggest that
myeloperoxidase is present in tissue macrophages of human
atherosclerotic lesions. Enzymatically active myeloperoxidase has been
extracted from this macrophage-rich atherosclerotic tissue and
immunohistochemical studies have shown co-localization of
myeloperoxidase and macrophages in atherosclerotic
lesions.10,11
Moreover, elevated levels of
3-chlorotyrosine, a specific product of myeloperoxidase at plasma
concentrations of halide, have been detected in atherosclerotic tissue
and in low-density lipoprotein isolated from these vascular
lesions.12
Immunohistochemical studies have also
demonstrated HOCl-modified proteins in these lesions.13
Collectively, these data strongly support the hypothesis that
myeloperoxidase is present in atherosclerotic lesions and is
catalytically active in vivo. These observations raise the
intriguing possibility that macrophages express myeloperoxidase under
certain conditions in vivo and that oxidants generated by
macrophage-associated myeloperoxidase may participate in the
pathogenesis of inflammatory diseases.
Myeloperoxidase may contribute to tissue injury by several mechanisms.
HOCl is a potent oxidant that attacks nucleophilic amino groups,
generating reactive aldehydes and chloramines.14,15
It
also reacts with unsaturated lipids to form
chlorohydrins.16
Lipid peroxidation and protein
cross-linking can be catalyzed by tyrosyl radical, which results from
the oxidation of tyrosine by myeloperoxidase.17,18
In
addition, HOCl has been shown to inactivate the protease inhibitor
-1-antitrypsin19
and to activate latent neutrophil
collagenase.20
Conversely, myeloperoxidase has inhibitory
effects on lymphocyte function21
and can suppress
inflammation by inactivating soluble chemotactic factors such as
C5a.22
Thus, in addition to being a direct cause of tissue
injury, myeloperoxidase may modulate aspects of the inflammatory
response.
Kupffer cells, the resident macrophages of the liver, comprise 80 to 90% of the body’s fixed tissue macrophage population.23 Because these cells seem to be involved in the pathogenesis of a variety of liver diseases and because myeloperoxidase has been detected in the macrophage foam cells of atheromatous lesions, we hypothesized that Kupffer cells might be a source of myeloperoxidase. We report here that two monospecific rabbit polyclonal antibodies to myeloperoxidase recognized a protein in detergent extracts of human liver tissue that co-migrated with myeloperoxidase on Western blotting. Myeloperoxidase was also detected immunohistochemically in the Kupffer cells of both nondiseased and diseased human livers. These observations indicate that myeloperoxidase is present in human Kupffer cells and raise the possibility that the enzyme may be an important source of oxidative damage during liver injury.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Liver Tissues
Fresh human liver was obtained at the time of liver transplantation for end-stage or fulminant disease or resection of mass lesions and was immediately stored at -70°C. Archived blocks of normal and diseased human tissues were obtained from the Department of Pathology, Saint Louis University Health Sciences Center. This study was approved by the Institutional Review Board of Saint Louis University.
Fresh mouse liver was obtained from myeloperoxidase-deficient and wild-type animals.25 The animals were perfused with 10 ml of ice-cold phosphate-buffered saline and the livers rapidly removed, fixed in formalin, and embedded in paraffin. The Animal Studies Committee of Washington University School of Medicine approved all animal studies.
Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis and Immunoblotting
Liver tissue was homogenized in 20 volumes of Tris buffer (50 mmol/L, pH 6.8) containing 5% (v/v) ß-mercaptoethanol and 1% (w/v) sodium dodecyl sulfate. A 1-ml aliquot of the homogenate was boiled for 5 minutes, sonicated three times for 10 seconds, and spun at 10,000 x g for 10 minutes. The protein content of the supernatant was quantitated using a modification of the Lowry assay and the indicated quantities loaded into the wells. Proteins were subjected to electrophoresis on 10 to 20% (w/v) gradient polyacrylamide gels (Mini Protean II system; BioRad Laboratories, Richmond, CA). The resolved proteins were electrophorectrically transferred to polyvinylidene difluoride membranes (MSI, Westboro, MA). Nonspecific binding sites were blocked with 20% nonfat dried milk/0.5% Tween-Tris-buffered saline. After thorough washing, the membrane was first incubated with rabbit anti-human myeloperoxidase antibody diluted 1:10,000 then with peroxidase-conjugated goat anti-rabbit IgG diluted 1:10,000 (Sigma, St. Louis, MO). After extensive washing, bound antibody was detected using chemiluminescence according to the manufacturer’s instructions (ECL; Amersham International, Little Chalfont, Bucks, UK).
Immunoaffinity Purification of Myeloperoxidase Antibody
Electrophoresis was performed as above using genuine myeloperoxidase (500 ng/well). The myeloperoxidase used was apparently pure as determined by spectroscopy (R/Z = 0.81) and by electrophoresis under denaturing and nondenaturing conditions followed by Coomassie blue staining and assessment of peroxidase activity, respectively.26 After transfer to the polyvinylidene difluoride membrane and incubation with the DAKO anti-myeloperoxidase antibody, a strip was cut from one side of the membrane and processed as indicated above to localize the antigen-antibody complex. This strip was aligned with the remaining membrane, the area containing the antigen and primary antibody excised, and antibody bound to myeloperoxidase eluted in 100 mmol/L of glycine buffer (pH 2.5) for 10 minutes and neutralized with 1 mol/L of Tris (pH 8.0).27
Immunohistochemistry
Tissue sections (4- to 5-µm thick) were cut from formalin-fixed, paraffin-embedded livers. Sections were deparaffinized and rehydrated by passage through a graded series of ethanol and distilled water. For myeloperoxidase immunohistochemistry, the antigen was retrieved by heating the slides in a pressure cooker in Tris-buffered saline with 0.075% Tween-20 (pH 7.6) for 10 minutes.28 Endogenous peroxidase activity was quenched by incubation in 0.3% v/v H2O2 in methanol for 20 minutes at room temperature.29 Sections were incubated at room temperature for 30 minutes with polyclonal rabbit anti-human myeloperoxidase antibody diluted 1:1500. Staining of murine liver sections was performed identically, with the exception that the polyclonal rabbit anti-human myeloperoxidase antibody was diluted 1:150. Sections for CD68 and HOP-1 were handled in a similar manner except for the omission of the antigen-retrieval step. The CD68 antibody was used at a dilution of 1:250 and HOP-1 antibody at 1:500 with incubation times of 1 and 2 hours, respectively. Biotinylated secondary antibodies were used at a 1:200 dilution. Immunostaining was performed using an avidin-biotin-horseradish peroxidase system (Vector Laboratories, Burlingame, CA) with 3-amino-9-ethylcarbazole as the chromogen for myeloperoxidase, diaminobenzidine for CD68, and True Blue (Kirkegaard and Perry Laboratories, Gaithersburg, MD) for HOP-1. Primary antibodies were omitted on sections serving as negative controls. Sections were counterstained and those stained with 3-amino-9-ethylcarbazole were mounted in aqueous mounting medium before application of coverslips. For immunofluorescence, liver sections were deparaffinized, rehydrated, and antigen-retrieved as above before incubation with a mixture of the myeloperoxidase and CD68 antibodies for 2 hours. The sections were subsequently incubated with fluorescein isothiocyanate-labeled anti-mouse IgG and Texas Red-labeled anti-rabbit IgG antibodies (Vector Laboratories) at a dilution of 1:50 and examined using confocal microscopy.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
60 kd) and two light subunits (15 kd).30
To assess
the specificity of the two different rabbit polyclonal myeloperoxidase
antibodies (DAKO and Biodesign) used for our studies, purified
myeloperoxidase was subjected to sodium dodecyl sulfate-polyacrylamide
gel electrophoresis and Western blotting. A major band and a minor band
with apparent molecular masses of 60 kd and 15 kd, respectively, were
recognized by both antibodies (Figure 1
,
lane 1). Minor bands were also observed at 40 kd and 20 kd, consistent
with autocatalytic products of myeloperoxidase previously
reported.31
In extracts of liver obtained from patients
undergoing transplantation or liver resection for a variety of liver
diseases, both antibodies recognized a single 60-kd protein that
co-migrated with authentic myeloperoxidase (Figure 1
, lanes 2 to 6).
Immunoaffinity-purified myeloperoxidase antibody also reacted with a
protein in liver tissue that co-migrated with the heavy chain of
myeloperoxidase (data not shown). A 15-kd band that co-migrated with
the light chain of myeloperoxidase was observed in liver homogenates on
immunoblots in which large concentrations of protein were
loaded and higher concentrations of primary antibody were used (data
not shown). Thus, the antibody was monospecific for myeloperoxidase in
these tissues. All five liver samples examined by Western blotting
contained immunoreactive myeloperoxidase (Figure 1)
.
|
. In contrast, hepatocytes
showed negligible immunoreactivity. Similar results were obtained with
two different antibodies to myeloperoxidase (data not shown).
Immunoreactivity was absent in sections exposed to primary antibody
that had been preincubated with myeloperoxidase (Figure 2B)
.
Furthermore, no immunoreactivity was observed in the livers of mice
deficient in myeloperoxidase, whereas livers from wild-type mice
demonstrated sinusoidal lining cell immunoreactivity similar to that
seen in human livers (Figure 3)
.
Collectively, these observations indicate that the antibodies used for
the immunohistochemical studies are monospecific for myeloperoxidase
and that sinusoidal lining cells of both human and murine livers
reacted with the antibodies.
|
|
.
Morphologically, these cells resembled quiescent and activated Kupffer
cells, respectively. A similar pattern of sinusoidal lining cell
staining with the myeloperoxidase antibody was observed in sections
from livers with a variety of acute and chronic liver diseases.
Although the number of immunoreactive sinusoidal lining cells varied
among the samples, there was no obvious difference between nondiseased
and diseased liver nor was there an apparent lobular distribution
gradient. Occasional cells within sinusoidal lumina were reactive with
the antibody (data not shown) but their multilobed nuclei identified
them as polymorphonuclear leukocytes and distinguished these cells from
the immunoreactive cells lining the sinusoids.
To further characterize the sinusoidal lining cells that were reactive
with the myeloperoxidase antibody, an antibody that recognizes CD68, an
epitope specific for monocytes and macrophages,32
was
applied to serial sections of the samples. Abundant CD68-immunoreactive
cells were observed lining the sinusoids of both nondiseased and
diseased livers (Figure 2D)
. CD68-positive cells were morphologically
identical to the sinusoidal lining cells that expressed
myeloperoxidase, but were more numerous in all specimens examined.
Based on the morphology and spatial correspondence with CD68-positive
cells, the myeloperoxidase-expressing sinusoidal lining cells seemed to
represent a distinct subpopulation of Kupffer cells.
Double-immunofluorescence confocal microscopy with the rabbit antibody
to myeloperoxidase and the murine antibody to CD68 confirmed
co-localization of the enzyme and CD68 in cells lining the hepatic
sinusoids (Figure 4)
.
|
. Spindle cell
immunoreactivity was frequently accentuated in areas of bile duct
proliferation. The myeloperoxidase-positive spindle cells in the septa
also expressed CD68 (Figure 5B)
. These observations suggest that
hepatic fibrosis is associated with increased numbers of
myeloperoxidase-expressing macrophages in connective tissue septa.
|
. The
presence of HOCl-modified proteins in these hepatocytes was associated
with the presence of myeloperoxidase-expressing cells in the septa
(Figure 5, A and C)
. These observations indicate that HOCl-modified
proteins may be present in both hepatocytes and Kupffer cells of
diseased human liver. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
60-kd protein reactive with two different rabbit polyclonal
antibody monospecific for myeloperoxidase, was present in detergent
extracts of human liver. The apparent molecular mass of this protein
was indistinguishable from that of the heavy chain of genuine
myeloperoxidase. We also detected an immunoreactive protein that
migrated with the light chain of myeloperoxidase when large amounts of
liver protein were studied. Second, both monospecific antibodies
recognized hepatic sinusoidal lining cells that seemed to be Kupffer
cells by morphological criteria. Immunoreactivity was blocked when the
antibodies were preincubated with myeloperoxidase and was absent in the
livers of myeloperoxidase-deficient mice. The human cells also reacted
with a monoclonal antibody to CD68, a specific marker for monocytes and
macrophages. Third, double-immunofluorescence confocal microscopy
demonstrated co-localization of the myeloperoxidase antibody and the
CD68 antibody in cells lining the hepatic sinusoid. Collectively, these
observations provide strong evidence that myeloperoxidase is present in
Kupffer cells, the body’s largest population of fixed tissue
macrophages.
Previous investigations have suggested that Kupffer cells lack
myeloperoxidase. For example, Wang and colleagues35
failed
to detect HOCl after stimulation of rat Kupffer cells with phorbol
ester, a treatment that stimulated rat neutrophils to produce
substantial amounts of HOCl. In preliminary experiments, we observed
far fewer myeloperoxidase-positive sinusoidal lining cells in rat liver
than in human liver (KEB and EMB, unpublished observation). Given this
difference, the inability to detect HOCl production by rat Kupffer
cells by biochemical means is not surprising. In contrast, we observed
significant numbers of myeloperoxidase-expressing sinusoidal lining
cells in mouse and gerbil livers, suggesting significant differences
among species (KEB and EMB, unpublished observation). Similar
observations have been made for monocytes. For example, myeloperoxidase
represents 1% of human monocyte protein but is absent from
circulating rabbit monocytes.4,36
It is interesting to
speculate that the presence of myeloperoxidase in the Kupffer cells of
some species but not others may contribute to the differential
susceptibility of various species to liver damage.
An important question raised by our observations is the cellular source of myeloperoxidase. Intracellular myeloperoxidase might arise from enzyme present in storage granules or in phagolysosomes or from secreted myeloperoxidase that has bound to mannose receptors and been internalized.37 Uptake of myeloperoxidase by macrophage mannose receptors is supported by the observation that levels of myeloperoxidase are lower in peripheral blood than in portal blood, suggesting clearance of circulating myeloperoxidase by the liver.38 However, it is also possible that cytokines or growth factors induce myeloperoxidase gene expression in Kupffer cells or cause monocytes to continue to express the enzyme as they differentiate into macrophages. Indeed, we have recently shown that granulocyte-macrophage colony-stimulating factor causes cultured human monocytes to continue to express functional myeloperoxidase as they differentiate into macrophages.11 The presence of myeloperoxidase in liver macrophages may increase their capacity for cytotoxicity, because internalization of myeloperoxidase has been shown to enhance the ability of macrophages to kill microorganisms.39,40
The ability of liver macrophages to use myeloperoxidase is demonstrated by the finding of HOCl-modified proteins in acute and chronic liver pathology. Although myeloperoxidase is present in Kupffer cells in both nondiseased and diseased livers, reactivity with HOP-1, the monoclonal antibody that specifically recognizes HOCl-modified proteins, was observed in close association with macrophages only in diseased liver, suggesting that myeloperoxidase activity arising from macrophages is enhanced during liver injury. It is important to point out, however, that because the liver sections we examined were derived primarily from explants, the patterns of immunostaining for HOCl-modified proteins that we observed may not be characteristic of less severe forms of liver injury.
In neutrophils and monocytes, myeloperoxidase is stored in the azurophilic granules.4,6 Depending on the agonist for phagocyte activation, it can be secreted into intracellular phagolysosomes or discharged into the extracellular milieu.41 The two distinct patterns of immunostaining for HOCl-modified proteins that we observed seem to indicate that hepatic macrophages also mobilize myeloperoxidase by both intra- and extracellular routes. In cases of acute, severe liver injury, HOP-1 reactivity was most prominent in necrotic areas infiltrated by macrophages. This pattern is consistent with HOCl production in the phagolysosomes of macrophages that have ingested necrotic debris. The pattern of HOP-1 reactivity was strikingly different in cases of chronic liver disease with cirrhosis in which HOCl-modified proteins were localized to hepatocytes adjacent to fibrous septa. HOP-1 staining of cells and extracellular matrix near myeloperoxidase-containing cells has previously been observed in atherosclerotic lesions and in diseased glomeruli and is presumed to result from diffusion of HOCl from its site of generation.13,24 The present observation suggests that myeloperoxidase contained in septal macrophages had been enzymatically active in the extracellular environment.
Myeloperoxidase could contribute to liver disease in a number of ways. The HOCl it generates is a potent oxidant that directly chlorinates and oxygenates a wide range of nucleophilic moieties in vitro.4,12,14-18 The enzyme also inactivates protease inhibitors and converts zymogen precursors of proteases into their active forms19,20 and has been shown to exert damaging effects on components of the extracellular matrix.42,43 Myeloperoxidase may therefore promote degradation of the normal hepatic extracellular matrix. Alterations in the extracellular matrix have been linked to the activation of hepatic stellate cells, which is a critical step in the development of hepatic fibrosis.44 Furthermore, reactive species generated by myeloperoxidase may in turn peroxidize lipids, thereby enhancing collagen production by activated stellate cells.45
The progression of chronic liver disease to fibrosis and ultimately to cirrhosis has been linked to oxidative damage to lipids and proteins.46 In most cases, however, the source of oxidants has not been clearly identified. In acute inflammatory processes, the generation of reactive oxygen species during the respiratory burst of neutrophils, monocytes, and macrophages is one potential source. It is noteworthy, however, that with the exceptions of alcoholic hepatitis and nonalcoholic steatohepatitis, neutrophilic infiltrates are not characteristic of chronic liver disease. Although Kupffer cells have been implicated in the pathogenesis of liver injury from a variety of causes, these macrophages have been shown to produce less superoxide on stimulation than neutrophils or other macrophages.47 Moreover, superoxide is chemically a reducing agent and is not highly reactive with biomolecules.3,5 Its dismutation product, H2O2, is a relatively weak oxidant that is much less bactericidal than HOCl.3,5 It is therefore unclear whether reactive oxygen species derived from the respiratory burst of Kupffer cells adequately explain the oxidative damage observed in chronic liver disease. Our finding that human Kupffer cells express myeloperoxidase suggests a more likely oxidant because these cells may have the capacity to convert H2O2 to HOCl. Thus, Kupffer cell myeloperoxidase may be an important source of oxidative damage during chronic liver injury.
In summary, our study has shown by Western blot and by immunohistochemical studies that human Kupffer cells contain myeloperoxidase, and that this enzyme’s characteristic oxidation products are detectable in acute and chronic liver disease. These observations point to a novel oxidative pathway by which Kupffer cells could participate in liver injury. These findings may have therapeutic implications because biochemical defenses against the products of myeloperoxidase differ from those that prevent damage by other reactive species.48
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by the Medical Research Service of the Veterans Administration (to K. E. B.) and the National Institutes of Health (AG15012, HL64344, DK56341 to J. W. H.), and the Washington University-Pharmacia-Monsanto-Searle Research Program.
Accepted for publication August 17, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, by macrophages enhances killing activity toward micro-organisms. DNA Cell Biol 1996, 15:617-624[Medline]
1(I) gene expression in human liver fat storing cells. Biochem Biophys Res Commun 1993, 194:1044-1050[Medline]
1(I) gene expression is associated with lipid peroxidation in hepatocellular injury: a link to tissue fibrosis? Hepatology 1994, 19:1262-1271[Medline]
This article has been cited by other articles:
 |
|
 |
|
  W. Lu and J. P. Uetrecht Peroxidase-Mediated Bioactivation of Hydroxylated Metabolites of Carbamazepine and Phenytoin Drug Metab. Dispos., August 1, 2008; 36(8): 1624 - 1636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. van Leeuwen, M. J.J. Gijbels, A. Duijvestijn, M. Smook, M. J. van de Gaar, P. Heeringa, M. P.J. de Winther, and J. W. C. Tervaert Accumulation of Myeloperoxidase-Positive Neutrophils in Atherosclerotic Lesions in LDLR / Mice Arterioscler. Thromb. Vasc. Biol., January 1, 2008; 28(1): 84 - 89. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Odobasic, A. R. Kitching, T. J. Semple, and S. R. Holdsworth Endogenous Myeloperoxidase Promotes Neutrophil-Mediated Renal Injury, but Attenuates T Cell Immunity Inducing Crescentic Glomerulonephritis J. Am. Soc. Nephrol., March 1, 2007; 18(3): 760 - 770. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hirsch, K. C. Hansen, S. Choi, J. Noh, R. Hirose, J. P. Roberts, M. A. Matthay, A. L. Burlingame, J. J. Maher, and C. U. Niemann Warm Ischemia-induced Alterations in Oxidative and Inflammatory Proteins in Hepatic Kupffer Cells in Rats Mol. Cell. Proteomics, June 1, 2006; 5(6): 979 - 986. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Esme, H. Fidan, T. Koken, and O. Solak Effect of lung ischemia-reperfusion on oxidative stress parameters of remote tissues Eur. J. Cardiothorac. Surg., March 1, 2006; 29(3): 294 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. M. Hanumegowda, B. L. Copple, M. Shibuya, E. Malle, P. E. Ganey, and R. A. Roth Basement Membrane and Matrix Metalloproteinases in Monocrotaline-Induced Liver Injury Toxicol. Sci., November 1, 2003; 76(1): 237 - 246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Marsche, A. Hammer, O. Oskolkova, K. F. Kozarsky, W. Sattler, and E. Malle Hypochlorite-modified High Density Lipoprotein, a High Affinity Ligand to Scavenger Receptor Class B, Type I, Impairs High Density Lipoprotein-dependent Selective Lipid Uptake and Reverse Cholesterol Transport J. Biol. Chem., August 23, 2002; 277(35): 32172 - 32179. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. H. Gregory and E. J. Wing Neutrophil-Kupffer cell interaction: a critical component of host defenses to systemic bacterial infections J. Leukoc. Biol., August 1, 2002; 72(2): 239 - 248. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
