Journal of Hepatology
Volume 32, Issue 4 , Pages 689-702, April 2000

Mallory bodies revisited

Department of Pathology, University of Graz School of Medicine, Graz, Austria

Article Outline

 

Since their first description in 1911 by F.B. Mallory 1., 2., 3., 4., 5., 6., Mallory bodies (MBs) or alcoholic hyalin have stimulated the interest of hepatopathologists for several reasons:

1.(i) MBs are characteristic cytoplasmic hyaline inclusions in hepatocytes and thus resemble a peculiar manifestation of liver cell injury.

2.(ii) MBs are morphologic hallmarks particularly of alcoholic and non-alcoholic steatohepatitis, liver disorders with poorly understood pathogenesis but considerable clinical importance.

3.(iii) MBs share morphologic and physicochemical features with inclusion bodies associated with chronic degenerative disorders of the central nervous system, such as Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis, diseases with etiology and pathogenesis still undetermined.

Thus, elucidation of the mechanisms involved in MB formation and itsconsequences for the affected cell may throw light not only on a special type of liver cell injury but also on degenerative disorders affecting other organs and tissues.

Back to Article Outline

Diseases Associated with MBs 

MBs are characteristically associated with alcoholic and non-alcoholic steatohepatitis (NASH), but are also present in benign and malignant hepatocellular neoplasms, and a diversity of metabolic, toxic and chronic cholestatic liver disorders (for review, see 6., 7., 8., 9., 10., 11., 12.). Their appearance is related to alterations of the cytokeratin (CK) intermediate filament (IF) cytoskeleton of hepatocytes 5., 13., 14..

The spectrum of alcoholic liver disease comprises steatosis, alcoholic hepatitis (alcoholic steatohepatitis), fibrosis and finally cirrhosis 6., 7., 8., 11., 12.. Steatosis, a common and reversible consequence of alcohol intoxication, appears within a relatively short time of abuse. More serious and life-threatening consequences, namely alcoholic hepatitis and cirrhosis, develop only in 20 to 40% of heavy and chronic drinkers 3., 6., which suggests a multifactorial pathogenesis, including environmental and genetic factors in addition to toxic effects of ethanol and its metabolites 3., 8., 15., 16., 17.. Several risk factors, alone or in combination, may determine individual susceptibility.

Alcoholic hepatitis is morphologically characterized by liver cell injury with macro- and/or microvesicular steatosis of various degrees, ballooning of hepatocytes, necrosis, cholestasis, occurrence of MBs, mostly neutrophil granulocytic inflammation and pericellular and perivenular fibrosis. A diversity of non-alcoholic liver disorders, collectively termed NASH, shares some or even all morphologic features with alcoholic hepatitis 9., 10., 11., 12., 18.. Particularly morbid obesity, weight-reducing surgery and type II diabetes mellitus may closely mimick in their liver pathology alcoholic liver disease, including the tendency to progress to cirrhosis. Drugs and toxins (like amiodarone, perhexiline maleate, 4,4-diethylaminoethoxyhexestrol) are able to produce similar lesions in human liver, and so do griseofulvin (GF) and 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC) in mice 19., 20.. In contrast to earlier views, MBs do not seem to be signs of adverse prognosis and their presence does not significantly influence mortality rates (for review, see 3., 4.). Therefore, on the basis of clinico-pathologic experience MBs seem to be indicators of a special type of liver cell injury rather than active players in the necroinflammatory process; nevertheless, the analysis of their composition and pathogenesis is essential for our understanding of the pathogenetic principles involved in chronic alcoholic liver disease and, at least morphologically, related disorders.

Back to Article Outline

Morphology of MBs 

MBs are cytoplasmic inclusions in hepatocytes, ranging in size from small granules to large irregular masses (Fig. 1; see ref. 2., 6., for further information). They display a predominantly filamentous ultrastructure consisting of 10–20-nm-thick filaments coated by fuzzy more electron-dense material2,13,21,Fig. 2). According to their ultrastructural appearance, MBs can be classified as type I (bundles of filaments in parallel arrays), type II (randomly oriented filaments) and type III (granular and amorphous material). Types II and III prevail and are often seen in combination of type II at the periphery and type III in the center (21). MBs are usually present in enlarged, ballooned hepatocytes, most of them looking vital with large nuclei and nucleoli. Most MB-containing hepatocytes lack large fat vacuoles but may display microvesicular steatosis. Although MBs are not obligatory for the diagnosis of alcoholic hepatitis in hematoxylin-eosin stained sections, more sensitive immunohistochemical stainings using CK or ubiquitin antibodies disclose MBs in most cases of otherwise morphologically characteristic alcoholic hepatitis. Neutrophils predominantly, but not constantly, cluster around MB-bearing hepatocytes. Pericellular fibrosis seems to be more pronounced around MB-containing cells 8., 11..

  • View full-size image.
  • Fig. 1. 

    Liver biopsy with alcoholic hepatitis. Note enlarged (hydropic) hepatocytes containing MBs. In addition, there is pronounced pericellular fibrosis as revealed by connective tissue stain (blue; chromotrope aniline blue stain). The inset shows an enlarged (hydropic) hepatocyte with an MB. Note the large nucleus containing a large nucleolus (H-E). Bars, 100 μm.

  • View full-size image.
  • Fig. 2. 

    Electron microscopy of MBs in human alcoholic hepatitis (a; Hu - AH), 2 months DDC (b; 2 m DDC-ML)- and 4 months GF (c; 4 m GF-ML)-fed mouse livers. The figures show the filamentous ultrastructure of MBs with filaments in irregular arrangement (type II). Note identical morphology of human and murine MBs. Bar, 500 nm.

Back to Article Outline

MB Formation in Animal Models 

Animal models are indispensable tools for elucidation of pathogenetic principles involved in human disease, although usually not all features of human disease can be reproduced in a single model. Several animal models for human alcoholic hepatitis and NASH have been designed 2., 19., 20., 22., 23.. In the alcohol-feeding models fatty liver, centrilobular necrosis, inflammation and fibrosis were markedly exaggerated if ethanol was administered together with a diet rich in polyunsaturated fat 22., 23., 24., 25., 26., 27.. MBs and associated liver cell changes, such as ballooning and pericellular accumulation of polymorphonuclear leukocytes (“satellitosis”), are, however, not evoked in animal livers under these conditons. On the other hand, chronic GF and DDC feeding of mice leads to the appearance of MBs mostly in enlarged hepatocytes, but also to liver cell necrosis, steatosis, fibrosis and ductular reaction, in addition to protoporphyrin accumulation. Satellitosis is occasionally seen but less pronounced than in human alcoholic hepatitis (19., 20., 28.; see also 2, for review). Thus, although the latter models are obviously not models for alcohol toxicity in the strict sense, they provide important morphologic features of alcoholic hepatitis and NASH and allow studies of specific aspects of their pathogenesis. The common pathogenetic pathways of alcoholic hepatitis and NASH remain to be uncovered. However, radical injury, believed to be of central importance in alcoholic liver disease, also seems to be involved in NASH and related models 29., 30..

Back to Article Outline

Mallory Bodies have a Complex Chemical Composition and are Related to the CK-IF Cytoskeleton and Components of the Protein-Folding and Degradation Machinery 

MB composition 

The first evidence of a relationship between MBs and the CK cytoskeleton was provided by observations that MBs react with CK antibodies and contain polypeptides closely resembling CKs of hepatocytes Fig. 3, Fig. 4. An additional non-CK protein component with a molecular weight between 62–65 kDa has been found in MBs but not in CKs of normal hepatocytesFig. 4,2,13). The predominantly filamentous ultrastructure of MBs together with diminution of hepatocellular cytoplasmic CK-IF immunostaining was originally (and erroneously) interpreted to indicate that MBs result from collapse of the CK-IF network. However, poly- and monoclonal CK antibodies, including those which recognize conformation-dependent epitopes on CKs, revealed that the MB-associated CK polypeptides are not organized as in normal IFs 2., 13., 31., 32., 33., 34., 35., 36., 37.. This is in line with data obtained by biochemical and immunochemical analyses of isolated and purified MBs which showed that in MBs the equimolar ratio of type I and type II CKs (i.e. CK18 and CK8), which is essential for normal IF assembly, is not maintained and that CK8 prevails over CK18 35., 36.. Interestingly, in DDC-fed mice already in early stages of intoxication a shift of the CK8 to CK18 ratio towards CK8 occurred when tested by immunoblots in liver homogenates (Fig. 5).

  • View full-size image.
  • Fig. 3. 

    Immunofluorescence microscopy of MBs in human liver with alcoholic hepatitis (Hu - AH; a, d, g, j, m) and in 2 months DDC (2 m DDC-ML; b, e, h, k, n)- and 4 months GF (4 m GF-ML; c, f, i, l, o)-fed mouse livers with antibodies against CKs8 and 18 (CK8/18), antibody 5B3 against phosphorylated CK8 (pCK8), antibodies against ubiquitin, and the MM120-1 and the SMI 31 antibodies. Note the pronounced immunostaining of MBs in enlarged hepatocytes with all these antibodies. With CK8/18 in addition to MBs a cytoplasmic CK filament network in hepatocytes is decorated which is clearly diminished in most enlarged MB-containing hepatocytes. With antibodies against phosphorylated CK8 (pCK8) in addition to MBs of different sizes (small granular to large irregular inclusions) the cytoskeleton with cell-peripheral accentuation of some hepatocytes is immunostained. Ubiquitin antibodies and the MM120-1 and SMI 31 antibodies exclusively react with MBs. Bar, 20 μm.

  • View full-size image.
  • Fig. 4. 

    One-dimensional SDS-polyacrylamide gel electrophoresis of MBs isolated and semipurified from the livers of 2 months DDC-fed mice. Note the presence of three major polypeptide bands (I, II, III). Bands II and III react in immunoblotting with CK antibodies and thus are of CK nature. Band II corresponds to CK8 and band III to CK18. Band I is a non-CK component. In addition, poorly soluble high molecular material remains at the interphase between stacking and resolving gels.

  • View full-size image.
  • Fig. 5. 

    The upper panel shows relative amounts of CKs 8 (K8) and 18 (K18) protein as revealed in mouse liver homogenates by immunoblotting with a polyclonal antibody directed to CKs 8 and 18. Note that in mice fed a normal diet CKs 8 and 18 are present in a 1:1 ratio. Already after 1 week of DDC feeding (1 w DDC) CK8 clearly prevails over CK18 and this situation is maintained after a 2.5-month (2.5 m DDC) DDC feeding period. The lower CK protein levels after prolonged DDC intoxication (2.5 m DDC) may reflect the higher number of hepatocytes with diminished intermediate filament cytoskeleton. The increased amount of CK8 resembles unassembled polypeptide. The lower panel shows mRNA levels in mouse liver homogenates. DDC intoxication (for 1 week and 2.5 months) leads to a conspicuous increase of CKs 8 (K8) and 18 (K18) mRNA levels. A peak level is already reached after 1 week (1 w DDC) of DDC feeding.

In addition to CKs a variety of non-CK components were identified in MBs. They include the MM120-1-, SMI 31- and MPM-2-reactive antigens, ubiquitin and high and low molecular weight heat shock proteins as revealed by immunohistochemistry and immunoblot analysesFig. 3,38,39,40),41)). The MM120-1 antigen is associated with a still uncharacterized high molecular weight protein that in vivo is an exclusive MB component (40). In vitro studies revealed that MM120-1-protein can be induced in tissue culture cells by stress treatments (e.g., Ca-ionophore, sodium arsenite, or heat shock) but also by transfection with a human CK18 gene construct (Fig. 6), suggesting a relationship to stress proteins (42). The antibody SMI 31, which is directed against a phosphorylated epitope present on neurofilaments and on abnormal tau protein associated with Alzheimer's neurofibrillary tangles, recognizes the non-CK MB component with an apparent molecular mass of 62 to 65 kDa and very acidic isoelectric pH around pH 4.5 (41). Although the nature of the SMI 31-reactive MB component is as yet unclear, the apparently common phospho-epitope in MBs and neurofibrillary tangles implies that similar protein kinases (i.e. proline-directed kinases) are responsible for hyperphosphorylation in both instances (43., 44.; for review, see 45). The similarity in the phosphorylation state of MB and neurofibrillary tangle proteins is further underlined by the observation that the 62–65-kDa MB protein reacted in immunoblots with MPM-2 antibodies. The MPM-2 antibody also reacts with neurofibrillary tangles and is directed to hyperphosphorylated epitopes generated on diverse proteins by mitotic kinases in the M-phase of the cell cycle. Therefore, M-phase specific kinases, like p34 cdc2- and MAP-kinase, seem to create this epitope 46., 47., 48..

  • View full-size image.
  • Fig. 6. 

    Chinese hamster ovary cells (CHOK 1) (a) and embryonal mouse hepatocytes (TIB 73) (b) were transfected with human CK18 gene constructs (using the receptor-mediated adenovirus-augmented gene delivery system). Double immunofluorescence was performed 24 h post-transfection with MM120-1 (green) and polyclonal CK8/18 (CK8+18) antibodies (red). Confocal images of both stainings were superimposed. Note that in transfected cells the foreign CK18 accumulates together with the MM120-1 antigen as small aggregates. Bar, 10 μm.

This SMI 31/MPM-2-reactive MB component closely resembles in its electrophoretic coordinates p62 which has recently been identified by our group as the major protein constituent of intracellular hyaline bodies (not identical with MBs!) present in hepatocellular carcinoma cells (49). The staining of MBs in human and mouse livers with these antibodies strongly suggests that p62, or at least a closely related protein, is indeed an integral component of MBs (Stumptner et al., manuscript in preparation). P62 has recently been discovered as phosphotyrosine-independent ligand of the SH2 domain of p56lck, a member of the c-src family of cytoplasmic kinases. It is a cytoplasmic non-proteasomal ubiquitin chain-binding protein which rapidly increases upon a variety of extracellular signals, including those related to oxidative stress 50., 51., 52., 53.. The p62 gene is activated without de novo protein synthesis and, therefore, behaves like immediate early response genes. In tissue culture cells activation of p62 gene occurs after stimulation of cells with growth factors or upon various stress treatments (54). P62 is apparently involved in cell regulatory processes by modifying the fate of ubiquitinated proteins. It may, therefore, on the one hand modulate the activation of transcription factors, such as NF-κB, and protein kinases involved in signal transduction, and on the other lead to the accumulation of ubiquitinated proteins. It is interesting in this context that treatment of tissue culture cells with proteasomal inhibitors induced granular accumulation of p62 together with multiubiquitin-protein conjugates as“sequestosomes” (52). Therefore, MBs may actually resemble sequestosome-like structures. Since p62 (as well as the MM120-1 antigen) is present in the earliest stages of MBs, an intimate relationship to MB development is plausible. However, its exact role in the cells and the consequences of its aggregation and segregation remain to be elucidated.

Ubiquitin is another non-CK component that MBs have in common with neurofibrillary tangles and other cytoplasmic protein aggregates, such as Lewy bodies inParkinson's disease, Rosenthal fibers in astrocytomas and inclusion bodies in motor neuron disease 38., 39., 55., 56., 57.. Ubiquitin plays a major role in targeting proteins to proteasomal degradation 58., 59.. Besides ubiquitin, other stress proteins, such as αB crystallin and HSP 70, associate with MBs or are induced under conditions leading to MB formation (60., 61.; Stumptner et al., unpublished observation). HSP70 belongs to a family of ubiquitous chaperone proteins that assist in protein-folding processes in order to prevent aggregation or promote refolding of conformationally altered proteins 62., 63.. These chaperones recognize, and bind to, exposed hydrophobic surfaces of proteins, that in the native state are buried and form noncovalent interactions, and thus prevent irreversible multimeric aggregation. Release of the polypeptide is then driven by an ATP-dependent conformational change of the chaperone. The expression of stress proteins is triggered by the intracellular accumulation of altered proteins resulting from a variety of stresses, including oxygen radicals, heat shock, toxins and heavy metals. The capacity of the cell to degrade damaged and mis- or unfolded proteins is a major factor opposing the expression of stress proteins. Therefore, if degradation is blocked, e.g. by inhibition of proteasome action, heat shock and other proteins accumulate. Thus the presence of HSP70 and ubiquitin in association with MBs implies impairment of proteasome function. Indeed, chronic ethanol administration to rats as well as oxidative stress inhibit proteasome activity in hepatocytes (64).

It is not yet clear whether the different MB components are directly involved in the initiation of MBs or are consequences of abnormal protein aggregation. Since ubiquitination is a late event in MB formation, at least as revealed by immunofluorescence microscopy in GF- or DDC-intoxicated mice (Stumptner et al., manuscript in preparation), ubiquitin binding seems to be a consequence rather than a cause of MB formation.

Posttranslational modification of CKs and other components in the course of MB formation 

The association of CKs with high molecular weight MB material suggests that CK polypeptides present in MBs are modified by covalent cross-linking 36., 65.. It has, indeed, been shown that hepatocellular CKs are substrates of transglutaminase, a Ca++-dependent cross-linking enzyme, and that MBs contain high amounts of ε-(γ-glutamyl) lysine cross-links resulting from transglutaminase action. Thus, transglutaminase contributes to the stability of these inclusions (66). Other mechanisms that could be responsible for crosslinking and stabilization of MB proteins are the formation of aldehyde adducts and a conformational transition to β-sheet which favor aggregation 67., 68., 69., 70.. Analysis of MBs by infrared spectroscopy demonstrated an increase in β-sheet conformation as compared to normal IF, which are predominantly α-helical (67). Poor solubility, β-sheet conformation, filamentous ultrastructure as well as the presence of glycosylated proteins are common features of diverse types of amyloid 71., 72.. Although MBs do not show the apple-green birefringence after Congo red staining (Denk et al., unpublished observation), which is characteristic of classical amyloid, MBs still fulfill most of the criteria for classification as a special form of intracellular amyloid.

Phosphorylation (in addition to proteolysis) is one of the most important regulatory principles of protein function, particularly in response to external stimuli. Hyperphosphorylation of CKs is involved in the cellular stress response and seems to play a role in the protection from hepatotoxic injury 73., 74.. In human alcoholic hepatitis phosphorylation of CKs at multiple sites and accumulation of phosphorylated CKs in MBs was revealed by antibodies that selectively recognize phosphorylated epitopes of CKs 8 or 18 (75). Two-dimensional gel electrophoresis of CK preparations derived from GF-treated animals showed an increase of the more acidic isoelectric CK variants reminiscent of phosphorylation (76). Increased phosphorylation of CKs in response to GF intoxication was furthermore confirmed in vitro by 32P-incorporation (77). In vivo, hyperphosphorylation occurred rapidly already after 1 day of DDC intoxication and preceded architectural changes of the cytoskeleton. In chronically DDC-intoxicated mice with MB-containing livers phosphorylated CKs were preferentially associated with MBs but not with the residual CK network adjacent to MBs (Fig. 7), suggesting that hyperphosphorylation of CKs contributes to aggregation and MB formation (75).

  • View full-size image.
  • Fig. 7. 

    Double-label immunofluorescence microscopy of livers of 2-month DDC-intoxicated (2 m DDC-ML) mice using monoclonal antibodies (a. 5B3; b. LJ 4; see 75) to different phosphoepitopes of CK8 (green) and polyclonal CK antibodies (CK8+18; red). Note liver cells without MBs containing phosphorylated CK8 associated with the CK cytoplasmic network (asterisks). MB-containing hepatocytes lack a phosphorylated CK network and phosphorylated CK8 is only associated with MBs (arrow heads). This suggests that phosphorylated CK8 preferentially aggregates in MBs. Bar, 10 μm.

Sequence of events in the assembly of MB components 

In patients recovered from alcoholic hepatitis MBs rapidly reappear if drinking is resumed (78). Like chronic alcoholic liver disease in man, experimental induction of MBs requires long-term intoxication with GF or DDC, indicating that MBs result from longlasting metabolic alterations of hepatocytes 19., 20.. In mice allowed to recover from GF or DDC intoxication, most MBs disappeared within 4 weeks and only remnants persisted for a longer period at the cell periphery, mostly in association with desmosomes 2., 28., 37., 79., 80.. If mice were reintoxicated after recovery, newly-formed MBs rapidly (within 2–3 days) reappeared in many hepatocytes, comparable to the situation in patients, mimicking an immunologic memory response (81). This proves that the response of chronically intoxicated and recovered livers (“primed livers”) is different from the response of“naive” livers to toxin exposure. The “primed” state, responsible for increased sensitivity with respect to MB formation, persists in the mouse for at least 2 months after recovery from chronic continuous intoxication. Since, at least in the mouse model, no evidence exists that immune mechanisms are responsible, this phenomenon reflects a “toxic memory” response (82).

The rapid reinduction of MBs in “primed” livers allowed the assembly of MB components to be followed in more detail. We found that CKs and the MM120-1 antigen were the earliest detectable MB components, whereas the SMI 31 and MPM-2 epitopes, ubiquitin and heat shock proteins appeared later (Stumptner et al., manuscript in preparation). Thus, during MB formation an “initiation” phase (primary accumulation of CKs and MM120-1 antigen) and a “maturation” phase (immunoreactivity with SMI 31, MPM-2 antibodies, and ubiquitination) can be distinguished.

The phenomenon of“priming” and “triggering” has recently been studied in detail by French and coworkers, showing that a variety of treatments such as ethanol feeding, inhibition of protein phosphatases, and heat shock treatment of primed recovered mice were able to reinduce MB formation 61., 83., 84., 85., 86.. The mechanisms leading to rapid MB re-formation in drug-primed mouse livers are still unresolved. In our original reports we stated that reinduction of MBs was not only achieved by readministration of the original agent, i.e. GF, but also by colchicine, suggesting an antimicrotubular effect to be involved in MB reinduction 28., 81.. This is supported by the observation that lumicolchicine, which lacks antimicrotubular properties, is inactive in DDC- as well as GF-primed livers (Stumptner et al., manuscript in preparation). French and coworkers 61., 85., 86.. emphasized that necrotizing processes and consecutive regeneration play an essential role in MB reinduction, although no direct correlation on the cellular level between regeneration (as assessed by PCNA immunostaining of nuclei) and MB formation could be found. Oxidative stress may also be a good candidate as a trigger of MBs for several reasons:

(i) Oxidative stress leads to NF-κB activation which has been found in the context of MB reinduction (83).

(ii) Endotoxin serum levels are elevated in both alcoholics and related animal models and cause the activation of NF-κB as well as the release of various cytokines from Kupffer cells and other cells contributing to inflammation and cell injury. In addition, free radicals generated by activated Kupffer cells and inflammatory cells damage DNA, protein and phospholipid 87., 88., 89., 90., 91., 92., 93., 94., 95., 96., 97., 98., 99., 100., 101., 102..

(iii) Several inducers of MBs, such as GF, DDC, ethanol and okadaic acid, are inducers of free radicals and thus initiate oxidative stress 29., 30., 83., 95., 97., 98., 100., 101., 103., 104., 105., 106..

(iv) Oxidative stress is a major responsible factor for abnormal protein folding 107., 108..

MB formation requires metabolically active cells with stimulated protein synthesis, and is not, as originally assumed, a simple passive process with collapse and aggregation of pre-existing components, particularly CKs. MB formation correlated with a significant increase of CK mRNA and protein concentrations 85., 86., 109., 110.. In refed but also in naive intoxicated mice CK mRNA levels, as revealed by quantitative RT-PCR, rapidly rose and so did CK protein levels (Fig. 5). However, overproduction of CK proteins alone is not sufficient for MB formation, as shown with shortterm intoxicated naive and primed mice. Despite a rapid increase in CK mRNA and protein levels, as well as hyperphosphorylation, MBs developed only in primed mice. Only in the latter situation did the MM120-1 antigen appear, suggesting that the trigger of MB formation is more closely related to occurrence of the MM120-1 antigen than to increased CK levels and hyperphosphorylation (Stumptner et al., manuscript in preparation).

Under our experimental conditions, MB reinduction occurred only in hepatocytes with residual IF cytoskeleton, but not in enlarged hydropic and“empty” cells completely lacking a demonstrable IF cytoplasmic cytoskeleton, as revealed by in situ hybridization combined with immunohistochemistry (Stumptner et al., manuscript in preparation). This again emphasizes that MB formation requires a vital, metabolically active, yet modified, cell still able to synthesize CKs (and other cellular components) upon toxic stimulation. The residual CK IF cytoskeleton may act as a scaffold for MB assembly, as suggested by the observation that young, tiny MB granules appear in association with intersections of CK IF bundles in hepatocytes. The apparent enlargement of pre-existing MBs, particularly in GF and colchicine reinduction (Stumptner et al., manuscript in preparation) but also during ethanol feeding of drug-primed mice (85), could be due to direct incorporation of newly synthesized CK proteins into pre-existing MBs, as observed by Kachi et al. (110) in primary cultures of GF-treated hepatocytes. According to the in vitro studies, incorporation of newly synthesized CK monomers into MBs is even more pronounced than into pre-existing IFs 77., 110., 111. which indicates that MBs are dynamic structures with a high turnover of MB proteins. Most ballooned hepatocytes without immunohistochemically demonstrable IF cytoskeleton and very low keratin mRNA expression seem to be unable to regenerate their cytoskeleton and produce MBs (Stumptner et al., manuscript in preparation). They may subsequently disappear by necrosis or apoptosis. This, however, does not mean that every hepatocyte with a deranged or lost CK IF cytoskeleton is“burned-out” since Yuan et al. (86) reported that predominantly “empty” hepatocytes are “sensitized” for MB formation by prior drug treatment.

Back to Article Outline

Clues from CK Gene Knockout Mice 

New insights into the role of CKs in MB formation and its functional consequences were recently provided by mice in which the gene for CK8 had been disrupted 112., 113., 114.. Since CK8 (type II CK) and CK18 (type I CK) are the only CKs expressed in normal hepatocytes, no CK IF can be formed in the absence of CK8 115., 116. and the remaining type I CK18 is rapidly degraded. Despite lack of a CK8/18 cytoskeleton homozygous CK8 knockout (−/−) knockout FVB/N mice develop normally without liver defects (113). DDC feeding, however, revealed a markedly higher toxicity in these animals than in wild-type mice with development of severe porphyria and significantly increased mortality (114; Zatloukal et al., manuscript submitted). As expected, analysis of livers of DDC-fed CK8−/−mice showed the absence of MBs (Fig. 8). Since also no other non-CK MB component accumulated in these livers, CK8 has to be the core protein of MBs, and the other MB components either bind to or coassemble with CK. Surprisingly, DDC-intoxicated mice with only one inactivated CK8 allele (CK8+/−), which were able to form a regular hepatocytic CK network, also lacked MBs despite the presence of all other signs of DDC intoxication, particularly porphyria. In the absence of the second CK8 allele, the DDC-induced overexpression of CKs resulted in a two-fold dominance of CK18 over CK8 mRNA, suggesting that, because of the excess of CK18, CK8 is prevented from interacting with other MB components to initiate MB formation.

  • View full-size image.
  • Fig. 8. 

    Immunofluorescence microscopy of DDC-intoxicated (2 months of DDC-feeding) wild-type (a; CK8/18 - WT), CK18 knockout (b; CK18−/−) and CK8 knockout (c; CK8−/−) mouse livers, using polyclonal CK8/18 antibodies. In livers of wild-type mice CK-positive MBs of different sizes are immunostained with the CK antibodies (a). The MB-containing hepatocytes are enlarged and show a reduced CK network mostly present only as a peripheral rim of filament bundles. In CK18−/− and CK8−/− mice no CK network is visible. Whereas CK8−/− animals do not develop MBs (c) CK18−/− mice (b) clearly develop small MBs (also positive with MM120-1) mostly situated at the cell periphery upon DDC intoxication. Bar, 10 μm.

Further important information on the role of CKs in MB formation came from experiments with CK18−/− mice (117). CK18−/− mice were also devoid of a hepatocytic CK IF cytoskeleton. However, they responded differently to DDC intoxication, in that in contrast to CK8−/− animals they were not abnormally sensitive to DDC (Magin et al., manuscript submitted). This is further proof that the higher mortality rate seen in DDC-treated CK8−/− mice is not due to impaired mechanical stability of hepatocytes caused by the loss of the CK cytoskeleton but related to the alteration of non-assembled CK polypeptides. This suggests that CKs are involved in cellular metabolic processes in an as yet unknown way and that CK8 and CK18 differ in this respect. Another difference between CK8−/− and CK18−/− mice was that DDC intoxication of CK18−/− mice led to the formation of classical MBs consisting of CK8 and non-CK components (Fig. 8; Magin et al., manuscript submitted). Thus, under certain conditions CK8 is stabilized and aggregates in cells even without its corresponding type I partner. Moreover, the fact that mice which formed MBs had fewer signs of toxicity than those which were unable to generate MBs, implies that the MB itself is not detrimental to the hepatocyte but is rather a product of a cellular defense mechanism involving CK8. A role of CKs in the cellular response to toxins has recently also been shown in in vitro experiments where overexpression of CKs by transfection of cells resulted in up to 450-fold increased resistance to a variety of toxins, including colcemid (118).

Further clues to the roles of CKs in normal liver and liver disease came from studies of transgenic mice expressing a mutated CK18 (R89C) (119). This mutation prevented CK IF formation and led to CK aggregates in the hepatocytes. This situation was associated with liver cell necroses and inflammation, indicating that abnormal accumulation of CK material and/or the disturbance of CK function can damage hepatocytes. In addition to these spontaneous lesions, a markedly increased sensitivity of the transgenic mice to liver toxins was found.

The relevance of CK alterations in human liver disease has recently been underlined by the detection of a mutation in the CK18 gene (H127L) in one patient with cryptogenic liver cirrhosis (120; for review on phenotypes in transgenic mice and CK mutation in human liver disease see 74).

Back to Article Outline

The MB: a Cellular Defense Mechanism to Stress-Induced Protein Damage 

As revealed by immunohistochemical, electron microscopic and biochemical analyses MBs resemble filamentous aggregates of phosphorylated, partially proteolytically degraded, ubiquitinated and cross-linked CK and non-CK proteins (including proteins with stress protein characteristics) with CK8 as the essential nucleating factor. It has now amply been documented that aggregation and deposition as abnormal inclusion body is the fate of unfolded or misfolded proteins, and one of the ways the cell eliminates abnormal, functionally impaired and thus potentially harmful proteins 62., 121., 122., 123.. Under normal circumstances, the nascent polypeptide chain is converted to a unique threedimensional functional structure. Altered amino acid composition, resulting from mutation, erroneous transcription, faulty mRNA processing or translation, interference by environmental factors, metabolic or viral gene products, and other adverse effects may disturb proper protein folding. Moreover, unbalanced synthesis of subunits of heterooligomeric protein complexes also favors the appearance of unassembled and aggregation-prone polypeptide chains. In misfolded or unassembled proteins hydrophobic sequences may be exposed, which are normally buried in the folds of the mature protein or covered by interaction with a partner in the case of heterooligomeric protein complexes. Inappropriate exposure of hydrophobic sequences favors adoption of abnormal protein conformation and thus fosters aggregation. The result is the sequestration of potentially important cellular components as inactive inclusion body. Mechanisms aimed at preventing misfolding and subsequent aggregation are, therefore, important survival strategies of a cell. Molecular chaperones play an essential role in this respect. They bind to proteins in their non-native configuration, mostly by recognizing hydrophobic sequences, and assist them in an ATP-dependent manner to either adopt a functional three-dimensional conformation, or to correctly refold in case of misfolding. However, if this rescue effort fails, e.g., because of lack of functional chaperones or ATP deficiency, misfolded proteins have to be discarded by proteolytic degradation.

Three major systems execute protein breakdown in eukaryotic cells: the lysosome apparatus, the calpains, and the ATP-dependent proteasome. The latter system is responsible for degradation of the bulk of cytosolic proteins, and most of its substrates are marked for degradation by covalent linkage to multiple molecules of ubiquitin. Hence, protein aggregation is again favoured if proteolytic degradation is impaired either by decreased proteasome activity or by increased substrate concentration exceeding the degradative capacity. Once aggregated, proteins are notoriously difficult to unfold or degrade 62., 122., 123..

Applying this general knowledge to MBs, the following scenario ensues: experimental models but also human liver diseases with MB formation, e.g. alcoholic hepatitis, are associated not only with an increase of CKs but also with a disturbance of the 1:1 type II to type I CK ratio essential for spontaneous assembly of IFs. Thus, the CK8 (type II) subunit prevails, favoring aggregation with other proteins, like p62 and stress proteins. Moreover, (hyper) phosphorylation as well as partial proteolysis alter the physiochemical properties of CKs and other proteins and may thus contribute to misfolding and aggregation. Ubiquitination then follows in a futile attempt to initiate degradation and disposal by the proteasomal pathway.

However, at least under the experimental conditions of DDC or GF intoxication, neither increased synthesis of CKs, unequal synthesis of IF subunits, i.e. shift to CK8, nor hyperphosphorylation are by themselves sufficient to produce MBs. Consequently, it is most likely that the final trigger of MB formation is the association of CKs with non-CK proteins and impairment of degradation. In vitro studies with CK-transfected cells support this concept by showing that CK-containing“aggresomes” only develop after proteasome inhibition (Stumptner et al., manuscript in preparation). Moreover, according to Rouach and coworkers 64., 106., proteasome and other protein-degrading pathways are inhibited in liver of rats fed alcohol by intragastric tube. This suggests that inclusion bodies, like MBs, apparently consisting of altered and misfolded proteins, only occur when the capacity of proteasomal degradation is exhausted or overstressed by high concentrations of aggregation-prone substrates. This may also be true in old CK18−/− livers which spontaneously develop MBs (117). It is well known that with age the adverse influence of reactive oxygen species on proteins, DNA and lipids exceeds the cellular defense mechanisms, leading to accumulation of damaged proteins, which may then overwhelm the available rescue capacity of the cell and accumulate as amyloid-like deposits (107).

In conclusion, protein aggregation seems to be an important mechanism of the cell in disposing of, as inert inclusions, irreversibly damaged proteins which can neither be refolded to an active functional configuration nor proteolytically degraded. Several examples of human diseases fit this concept. Inclusion body formation can be observed in chronic degenerative and aging-related disorders, including amyloidoses, Alzheimer's and Parkinson's diseases and motor neuron disease 55., 56., 57.. On that basis, MBs and their highly reproducible animal models may tell us a story of interest and application beyond alcoholic liver disease.

Back to Article Outline

References 

  1. Mallory FB. Cirrhosis of the liver: five different lesions from which it may arise. Bull Johns Hopkins Hosp. 1911;22:69–75
  2. Denk H, Franke WW, Kerjaschki D, Eckersdorfer R. Mallory bodies in experimental animals and man. Int Rev Exp Pathol. 1979;20:77–121
  3. Jensen K, Gluud C. The Mallory body: morphological, clinical and experimental studies. (part 1 of a literature survey) Hepatology. 1994;20:1061–1077
  4. Jensen K, Gluud C. The Mallory body: theories on development and pathological significance. (part 2 of a literature survey) Hepatology. 1994;20:1330–1342
  5. Denk H, Lackinger E. Cytoskeleton in liver diseases. Semin Liver Dis. 1986;6:199–211
  6. French SW, Nash J, Shitabata P, Kachi K, Hara C, Chedid A, et al.  Pathology of alcoholic liver disease. Semin Liver Dis. 1993;13:154–169
  7. Hall P. Pathological spectrum of alcoholic liver disease. In:  Hall P editors. Alcoholic Liver Disease. Pathology and Pathogenesis. London, Boston, Melbourne, Auckland: Edward Arnold; 1995;p. 41–68
  8. Ishak KG, Zimmerman HJ, Ray MB. In: Alcoholic liver disease: pathologic, pathogenetic and clinical aspects. 15:Alcoholism: Clin Exp Res; 1991;p. 45–66
  9. Zimmerman HJ, Ishak KG. Non-alcoholic steatohepatitis and other forms of pseudo-alcoholic liver disease. In:  Hall P editors. Alcoholic Liver Disease. Pathology and Pathogenesis. London, Boston, Melbourne, Auckland: Edward Arnold; 1995;p. 175–198
  10. James OFW, Day CP. Non-alcoholic steatohepatitis (NASH): a disease of emerging identity and importance. J Hepatol. 1998;20:495–501
  11. Baptista A, Bianchi L, DeGroote J, Desmet VJ, Gedigk P, Korb G, et al.  Alcoholic liver disease: morphological manifestations. Review by an international group. Lancet. 1981;i:707–711
  12. Burt AD, Mutton A, Day CP. Diagnosis and interpretation of steatosis and steatohepatitis. Semin Diagnostic Pathol. 1998;15:246–258
  13. Franke WW, Denk H, Schmid E, Osborn M, Weber K. Ultrastructural, biochemical and immunologic characterization of Mallory bodies in livers of griseofulvin-treated mice. Fimbriated rods of filaments containing prekeratin-like polypeptides. Lab Invest. 1979;40:207–220
  14. Katsuma Y, Swierenga SHH, Khettry U, Marceau N, French SW. Changes in the cytokeratin intermediate filament cytoskeleton associated with Mallory body formation in mouse and human liver. Hepatology. 1987;7:1215–1223
  15. Bosron WF, Ehrig T, Li TK. Genetic factors in alcohol metabolism and alcoholism. Semin Liver Dis. 1993;13:126–135
  16. Grove J, Daly AK, Bassendine MF, Day CP. Association of a tumor necrosis factor promoter polymorphism with susceptibility to alcoholic steatohepatitis. Hepatology. 1997;26:143–146
  17. Lindros KO. Alcoholic liver disease: pathobiological aspects. J Hepatol. 1995;23(Suppl):7–15
  18. Ludwig J, Viggiano TR, McGill DB, Ott BJ. Non alcoholic steatohepatitis. Mayo Clinic experiences with a hitherto unnamed disease. Mayo Clin Proc. 1980;55:434–438
  19. Denk H, Gschnait F, Wolff K. Hepatocellular hyalin (Mallory bodies) in long term griseofulvin-treated mice: a new experimental model for the study of hyalin formation. Lab Invest. 1975;32:773–776
  20. Tsunoo C, Harwood TR, Arak S, Yokoo H. Cytoskeletal alterations leading to Mallory body formation in livers of mice fed 3,5-diethoxycarbonyl-1,4-dihydrocollidine. J Hepatol. 1987;5:85–97
  21. Yokoo H, Minick OT, Batti F, Kent G. Morphologic variants of alcoholic hyalin. Am J Pathol. 1972;69:25–40
  22. Lieber CS, DeCarli LM, Sorrell MF. Experimental methods of ethanol administration. Hepatology. 1989;10:501–510
  23. French SW, Morimoto M, Tsukamoto H. Animal models of alcohol-associated liver injury. In:  Hall P editors. Alcoholic Liver Disease. Pathology and pathogenesis. London, Boston, Melbourne, Auckland: Edward Arnold; 1995;p. 279–296
  24. Tsukamoto H, French SW, Benson N, Delgado G, Rao GA, Larkin EC, et al.  Severe and progressive steatosis and focal necrosis in rat liver induced by continuous intragastric infusion of ethanol and low fat diet. Hepatology. 1985;5:224–232
  25. Tsukamoto H, Gad K, French SW. Insights into the pathogenesis of alcoholic liver necrosis and fibrosis: status report. Hepatology. 1990;12:599–608
  26. Tsukamoto H, Horne W, Kamimura S, Niemelä O, Parkkila S, Yiä-Herttuala S, et al.  Experimental liver cirrhosis induced by alcohol and iron. J Clin Invest. 1995;96:620–630
  27. Tsukamoto H, Cheng S, Blander WS. Effects of dietary polyunsaturated fat on ethanol-induced Ito cell activation. Am J Physiol. 1996;270(Gastrointest Liver Physiol 33):G581–G586
  28. Denk H, Eckerstorfer R, Gschnait F, Konrad K, Wolff . K Experimental induction of hepatocellular hyalin (Mallory bodies) in mice by griseofulvin treatment. I. Light microscopic observations. Lab Invest. 1976;35:377–382
  29. Ortiz de Montellano PR, Beilan HS, Kunze KL. N-alkylprotoporphyrin IX formation in 3,5-dicarbethoxy-1,4-dihydrocollidine-treated rats. Transfer of the alkyl group from the substrate to the porphyrin. J Biol Chem. 1981;256:6708–6713
  30. Knasmüller S, Parzefall W, Helma C, Kassie F, Ecker S, Schulte-Hermann R. Toxic effects of griseofulvin: disease models, mechanisms, and risk assessment. Crit Rev Toxicol. 1997;27:495–537
  31. Denk H, Franke WW, Eckerstorfer R, Schmid E, Kerjaschki D. Formation and involution of Mallory bodies (alcoholic hyalin) in murine and human liver revealed by immunofluorescence microscopy with antibodies to prekeratin. Proc Natl Acad Sci USA. 1979;76:4112–4116
  32. Denk H, Franke WW, Dragosics B, Zeiler I. Pathology of cytoskeleton of liver cells: demonstration of Mallory bodies (alcoholic hyalin) in murine and human hepatocytes by immunofluorescence microscopy using antibodies to cytokeratin polypeptides from hepatocytes. Hepatology. 1981;1:9–20
  33. Denk H, Krepler R, Lackinger E, Artlieb U, Franke WW. Immunological and biochemical characterization of the keratinrelated component of Mallory bodies: a pathological pattern of hepatocytic cytokeratins. Liver. 1982;2:165–175
  34. Okanoue T, Ohta M, Ou O, Kachi K, Kagawa K, Yuki T, et al.  Relationship of Mallory bodies to intermediate filaments in hepatocytes. A scanning electron microscopy study. Lab Invest. 1985;53:534–540
  35. Hazan R, Denk H, Franke WW, Lackinger E, Schiller DL. Change of cytokeratin organization during development of Mallory bodies as revealed by a monoclonal antibody. Lab Invest. 1986;54:543–553
  36. Zatloukal K, Boeck G, Rainer I, Denk H, Weber K. High molecular weight components are main constituents of Mallory bodies isolated with a fluorescence activated cell sorter. Lab Invest. 1991;64:200–206
  37. Zatloukal K, Spurej G, Rainer I, Lackinger E, Denk H. Fate of Mallory body-containing hepatocytes: disappearance of Mallory bodies and restoration of the hepatocytic intermediate filament cytoskeleton after drug withdrawal in the griseofulvintreated mouse. Hepatology. 1990;11:652–661
  38. Lowe J, Blanchard A, Morell K, Lennox G, Reynolds L, Billet M, et al.  Ubiquitin is a common factor in intermediate filament inclusion bodies of diverse type in man, including those of Parkinson's disease, Pick's disease, and Alzheimer's disease, as well as Rosenthal fibers in cerebellar astrocytomas, cytoplasmic bodies in muscle, and Mallory bodies in alcoholic liver disease. J Pathol. 1988;155:9–15
  39. Ohta M, Marceau N, Perry G, Manetto V, Gambetti P, Autilio-Gambetti L, et al.  Ubiquitin is present on the cytokeratin intermediate filaments and Mallory bodies of hepatocytes. Lab Invest. 1988;59:848–856
  40. Zatloukal K, Denk H, Spurej G, Lackinger E, Preisegger KH, Franke WW. High molecular weight component of Mallory bodies detected by a monoclonal antibody. Lab Invest. 1990;62:427–434
  41. Preisegger KH, Zatloukal K, Spurej G, Riegelnegg D, Denk H. Common epitopes of human and murine Mallory bodies and Lewy bodies as revealed by a neurofilament antibody. Lab Invest. 1992;66:193–199
  42. Stumptner C, Fuchsbichler A, Denk H, Zatloukal K. Overexpression of cytokeratin induces accumulation of the Mallory body-specific component MM120-1 in vitro. Hepatology. 1996;24:129A
  43. Nukina N, Kosik KS, Selkoe DJ. Recognition of Alzheimer paired helical filaments by monoclonal neurofilament antibodies is due to crossreaction with tau protein. Proc Natl Acad Sci USA. 1987;84:3415–3419
  44. Lichtenberg-Kraag B, Mandelkow EM, Biernat J, Steiner B, Schröter C, Gustke N, et al.  Phosphorylation - dependent epitopes of neurofilament antibodies on tau protein and relationship with Alzheimer tau. Proc Natl Acad Sci USA. 1992;89:5384–5388
  45. Mandelkow EM, Mandelkow E. Tau as a marker for Alzheimer's disease. Trends Biol Sci. 1993;18:480–483
  46. Westendorf JM, Rao PN, Gerace L. Cloning of cDNAs for Mphase phosphoproteins recognized by the MPM2 monoclonal antibody and determination of the phosphorylated epitope. Proc Natl Acad Sci USA. 1994;91:714–718
  47. Vincent I, Rosado M, Davies P. Mitotic mechanisms in Alzheimer's disease. J Cell Biol. 1996;132:413–425
  48. Kondratick CM, Vandre DD. Alzheimer's disease neurofibrillary tangles contain mitosis-specific phosphoepitopes. J Neurochem. 1996;67:2405–2416
  49. Stumptner C, Heid H, Fuchsbichler A, Hauser H, Mischinger HJ, Zatloukal K, et al.  Analysis of intracytoplasmic hyaline bodies in a hepatocellular carcinoma. Demonstration of p62 as major constituent. Am J Pathol. 1999;154:1701–1710
  50. Vadlamudi RK, Joung I, Strominger JL, Shin J. p62, a phosphotyrosine-independent ligand of the SH2 domain of p56lck belongs to a new class of ubiquitin-binding proteins. J Biol Chem. 1996;271:20235–20237
  51. Joung I, Strominger JL, Shin J. Molecular cloning of a phosphotyrosine-independent ligand of the p56lck SH2 domain. Proc Natl Acad Sci USA. 1996;93:5991–5995
  52. Shin J. P62 and the sequestosome, a novel mechanism for protein metabolism. Arch Pharm. 1998;21:629–633
  53. Vadlamudi RK, Shin J. Genomic structure and promoter analysis of the p62 gene encoding a non-proteasomal multi-ubiquitin chain binding protein. FEBS Lett. 1998;435:138–142
  54. Lee YH, Ko J, Joung I, Kim JH, Shin J. Immediate early response of the p62 gene encoding a non-proteasomal multiubiquitin chain binding protein. FEBS Lett. 1998;438:297–300
  55. Lowe J, Mayer RJ, Landon M. Ubiquitin in neurodegenerative diseases. Brain Pathol. 1993;3:55–65
  56. Bergmann M. Motor neuron disease/amyotrophic lateral sclerosis - lessons from ubiquitin. Path Res Pract. 1993;189:902–912
  57. Julien JP. Neurofilaments and motor neuron disease. Trends Cell Biol. 1997;7:243–248
  58. Ciechanover A. The ubiquitin-proteasome proteolytic pathway. Cell. 1994;79:13–21
  59. Weissman AM. Regulating protein degradation by ubiquitination. Immunol Today. 1997;18:189–198
  60. Lowe J, McDermott H, Pike I, Spendlove I, Landon M, Mayer RJ. αB crystallin expression in non-lenticular tissues and selective presence in ubiquitinated inclusion bodies in human disease. J Pathol. 1992;166:61–68
  61. Yuan QX, Marceau N, French BA, Fu P, French SW. Heat shock in vivo induces Mallory body formation in drug primed mouse liver. Exp Mol Pathol. 1995;63:63–76
  62. Hartl FU. Molecular chaperones in cellular protein folding. Nature. 1996;381:571–580
  63. Rassow J, Ahsen O, Boiner U, Pfanner N. Molecular chaperones: towards a characterization of the heat-shock protein 70 family. Trend Cell Biol. 1997;7:129–133
  64. Fataccioli V, Andrand E, Gentil M, French SW, Rouach H. Effects of chronic ethanol administration on rat liver proteasome activities: relationship with oxidative stress. Hepatology. 1999;29:14–20
  65. Cadrin M, Marceau N, French SW. Cytokeratin of apparent high molecular weight in livers from griseofulvin-fed mice. J Hepatol. 1992;14:226–231
  66. Zatloukal K, Fesus L, Denk H, Tarcsa E, Spurej G, Böck G. High amount of ε-(γ-glutamyl) lysine cross-links in Mallory bodies. Lab Invest. 1992;66:774–777
  67. Cadrin M, French SW, Wong TT. Alteration in molecular structure of cytoskeleton proteins in griseofulvin-treated mouse liver: a pressure tuning infrared spectroscopy study. Exp Mol Pathol. 1991;55:170–179
  68. Li CJ, Nanji AA, Siakotos AN, Lin RC. Acetaldehyde-modified and 4-hydroxynonenal-modified proteins in the livers of rats with alcoholic liver disease. Hepatology. 1997;26:650–657
  69. Lieber CS. Biochemical factors in alcoholic liver disease. Semin Liver Dis. 1993;13:136–153
  70. Tuma DJ, Sorrell MF. The role of acetaldehyde adducts in liver injury. In:  Hall P editors. Alcoholic Liver Disease. Pathology and Pathogenesis, London, Boston, Melbourne, Auckland: Edward Arnold; 1995;p. 89–99
  71. Breathnach SM. The cutaneous amyloidoses. Pathogenesis and therapy. Arch Dermatol. 1985;121:470–475
  72. Röcken C, Schwotzer EB, Linke RP, Saeger W. The classification of amyloid deposits in clinicopathological practice. Histopathology. 1996;29:325–335
  73. Ku NO, Liao J, Chon CF, Omary MB. Implications of intermediate filament protein phosphorylation. Cancer Metastasis Rev. 1996;15:429–444
  74. Omary MB, Ku NO. Intermediate filament proteins of the liver: emerging disease association and function. Hepatology. 1997;25:1043–1048
  75. Stumptner C, Omary MB, Fickert P, Denk H, Zatloukal K. Hepatocyte cytokeratins are hyperphosphorylated at multiple sites in human alcoholic hepatitis and in a Mallory body mouse model. Am J Pathol, in press.
  76. Salmhofer H, Rainer I, Zatloukal K, Denk H. Posttranslational events involved in griseofulvin-induced keratin cytoskeleton alterations. Hepatology. 1994;20:731–740
  77. Cadrin M, McFarlane , Anderson N, Aasheim LH, Kawahara H, Franks DJ, et al.  Modifications in cytokeratin and actin in cultured liver cells derived from griseofulvin-fed mice. Lab Invest. 1995;72:453–460
  78. Klassen LW, Tuma D, Sorrell MF. Immune mechanisms in alcohol-induced liver disease. Hepatology. 1995;22:355–357
  79. Denk H, Lackinger E, Cowin P, Franke WW. Maintenance of desmosomes in mouse hepatocytes after drug-induced rearrangement of cytokeratin filament material. Exp Cell Res. 1985;161:161–171
  80. Denk H, Franke WW. Rearrangement of the hepatocyte cytoskeleton after toxic damage: involution, dispersal and peripheral accumulation of Mallory body material after drug withdrawal. Eur J Cell Biol. 1981;23:241–249
  81. Denk H, Eckerstorfer R. Colchicine-induced Mallory body formation in the mouse. Lab Invest. 1977;36:563–565
  82. Zatloukal K, Stumptner C, Fuchsbichler A, Lehner M, Denk H. Induction of Mallory bodies reflects a toxic memory response and is linked to overexpression of cytokeratin mRNA in hepatocytes. Hepatology. 1996;24:601A
  83. Yuan QX, Nagao Y, Gaal K, Hu B, French SW. Mechanism of Mallory body formation induced by okadaic acid in drugprimed mice. Exp Mol Pathol. 1998;65:87–103
  84. Yuan Q, French BA, French SW. Tautomycin induces extensive Mallory body formation in drug primed mouse livers. Hepatology. 1998;28:256A
  85. Zhang-Gouillon ZQ, Yuan QX, Hu B, Marceau B, French BA, Gaal K, et al.  Mallory body formation by ethanol feeding in drug-primed mice. Hepatology. 1998;27:116–122
  86. Yuan QX, Marceau N, French BA, Fu P, French SW. Mallory body induction in drug-primed mouse liver. Hepatology. 1996;24:603–612
  87. Adachi Y, Bradford BU, Gao W, Bojes HK, Thurman RG. Inactivation of Kupffer cells prevents early alcohol-induced liver injury. Hepatology. 1994;20:453–460
  88. Bjarnson I, Peters TJ, Wise RJ. The leaky gut of alcoholism: possible route of entry for toxic compounds. Lancet. 1984;i:179–182
  89. Bode C, Kugler V, Bode JC. Endotoxemia in patients with alcoholic and non-alcoholic cirrhosis and in subjects with no evidence of chronic liver disease following acute ethanol excess. J Hepatol. 1987;4:8–14
  90. Caraceni P, Ryu HS, Subbotin V, De Maria N, Colantoni A, Roberts L, et al.  Rat hepatocytes isolated from alcohol-induced fatty liver have an increased sensitivity to anoxic injury. Hepatology. 1997;25:934–949
  91. Deaciuc IV. Alcohol and cytokine networks. Alcohol. 1997;14:421–430
  92. Dilger K, Metzler J, Bode JC, Klotz V. CYP 2E1 activity in patients with alcoholic liver disease. J Hepatol. 1997;27:1009–1014
  93. Ganey PE, Bailie MB, VanCise S, Colligan ME, Madhukar BV, Robinson JP, et al.  Activated neutrophils from rat injured isolated hepatocytes. Lab Invest. 1994;70:53–60
  94. Iimuro Y, Gallucci RM, Luster MI, Kono H, Thurman RG. Antibodies to tumor necrosis factor alfa attenuate hepatic necrosis and inflammation caused by chronic exposure to ethanol in the rat. Hepatology. 1997;26:1530–1537
  95. Ingelman-Sundberg M, Johansson I. Mechanisms of hydroxyl radical formation and ethanol oxidation by ethanol-inducible and other forms of rabbit liver microsomal cytochrome P-450. J Biol Chem. 1984;259:6447–6458
  96. Jaeschke H, Farhood A. Neutrophil and Kupffer cell-induced oxidant stress and ischemia-reperfusion injury in rat liver. Am J Physiol. 1991;260:G355–G362
  97. Kukielka E, Dicker E, Cederbaum AI. Increased production of reactive oxygen species by rat liver mitochondria after chronic ethanol treatment. Arch Biochem Biophys. 1994;309:377–386
  98. Lauterburg BH, DeQuay B. Radicals and oxidants in ethanol induced liver injury. In:  Moslen MT,  Smith CV editor. Free radical mechanisms of tissue injury. London, Tokyo, Boca Raton, Ann Arbor: CRC Press; 1992;p. 33–43
  99. Nanji AA, Khettry U, Sadrzadeh SMH, Yamanaka T. Severity of liver injury in experimental alcoholic liver disease. Correlation with plasma endotoxin, prostaglandin E2, leukotriene B4, and thromboxane B2. Am J Pathol. 1993;142:167–173
  100. Rouach H, Fataccioli V, Gentil M, French SW, Morimoto M, Nordmann R. Effect of chronic ethanol feeding on lipid peroxidation and protein oxidation in relation to liver pathology. Hepatology. 1997;25:351–355
  101. Wieland P, Lauterburg BA. Oxidation of mitochondrial proteins and DNA following administration of ethanol. Biochem Biophys Res Comm. 1995;213:815–819
  102. Hoek JB. Endotoxin and alcoholic liver disease: tolerance andsusceptibility. (editorial) Hepatology. 1999;29:1602–1604
  103. Bailey SM, Cunningham CC. Acute and chronic ethanol increase reactive oxygen species generation and decrease viability in fresh, isolated rat hepatocytes. Hepatology. 1998;28:1318–1326
  104. Castillo T, Koop DR, Kamimura S, Triadafilopoulos G, Tsukamoto H. Role of cytochrome P-450 2E1 in ethanol-, carbon tetrachloride- and iron-dependent microsomal lipid peroxidation. Hepatology. 1992;16:992–996
  105. Weltman MD, Farrell GC, Hall P, Ingelman-Sundberg M, Liddle C. Hepatic cytochrome P450 2E1 is increased in patients with non alcoholic steatohepatitis. Hepatology. 1998;27:128–133
  106. Rouach H, Fataccioli V, Gentil M, French SW, Marimoto M, Nordmann R. Effect of chronic ethanol feeding on lipid peroxidation and protein oxidation in relation to liver pathology. Hepatology. 1997;25:351–355
  107. Johnson FB, Sinclair DA, Guarente L. Molecular biology of aging. Cell. 1999;96:291–302
  108. Ando Y, Suhr O, El-Salhy M. Oxidative stress and amyloidosis. Histol Histopathol. 1998;13:845–850
  109. Hutter H, Zatloukal K, Winter G, Stumptner C, Denk H. Disturbance of keratin homeostasis in griseofulvin-intoxicated mouse liver. Lab Invest. 1993;69:576–582
  110. Kachi K, Cadrin M, French SW. Synthesis of Mallory body, intermediate filament, and microfilament proteins in liver cell primary cultures. Lab Invest. 1993;68:71–81
  111. Skalli O, Chou YH, Goldman RD. Intermediate filaments: not so tough after all. Trends Cell Biol. 1992;2:308–312
  112. Baribault H, Price H, Miyai K, Oshima RG. Mid-gestational lethality in mice lacking keratin 8. Genes Dev. 1993;7:1191–1202
  113. Baribault H, Penner J, Iozzo RV, Wilson-Heiner M. Colorectal hyperplasia and inflammation in keratin 8-deficient FVB/N mice. Genes Dev. 1994;8:2964–2974
  114. Zatloukal K, Stumptner C, Lechner M, Fuchsbichler A, Eshkind LG, Baribault H, et al.  Cytokeratin protects hepatocytes from intoxication: a possible new function of intermediate filaments revelaed by cytokeratin 8 knock-out mice. Hepatology. 1997;26:182A
  115. Franke WW, Denk H, Kalt R, Schmid E. Biochemical and immunological identification of cytokeratin proteins present in hepatocytes of mammalian liver tissue. Exp Cell Res. 1981;131:299–318
  116. Moll R, Franke WW, Schiller D, Geiger B, Krepler R. The catalog of human cytokeratins: Patterns of expression in normal epithelia, tumors and cultured cells. Cell. 1982;31:11–24
  117. Magin TM, Schröder R, Leitgeb S, Wanninger F, Zatloukal K, Grund C, et al.  Lessons from keratin 18 knockout mice; formation of novel keratin filaments, secondary loss of keratin 7 and accumulation of liver specific keratin 8-positive aggregates. J Cell Biol. 1998;140:1441–1451
  118. Baumann PA, Dalton WS, Anderson JM, Cress AE. Expression of cytokeratin confers multiple drug resistance. Proc Natl Acad Sci USA. 1994;91:5311–5314
  119. Ku NO, Michie SA, Soetikno RM, Resurreccion EZ, Broome RL, Oshima RG, et al.  Susceptibility to hepatotoxicity in transgenic mice that express a dominant-negative human keratin 18 mutant. J Clin Invest. 1996;98:1034–1046
  120. Ku NO, Wright TL, Terrault NA, Gish R, Omary MB. Mutation of human keratin 18 in association with cryptogenic cirrhosis. J Clin Invest. 1997;99:19–23
  121. Taubes G. Misfolding the way to disease. Science. 1996;271:1493–1496
  122. Johnston JA, Ward CL, Kapito RR. Aggresomes: a cellular response to misfolded proteins. J Cell Biol. 1998;143:1883–1898
  123. Lee DH, Goldberg AL. Proteasome inhibitors: valuable new tools for cell biologists. Trends Cell Biol. 1998;8:397–403

PII: S0168-8278(00)80233-0

Journal of Hepatology
Volume 32, Issue 4 , Pages 689-702, April 2000

Article Tools

* Image Usage & Resolution