carcinogen-inducedabnormalitiesin rat livercells …...incubation of methanol-fixed hyaline...

[CANCER RESEARCH 39, 800-807, March 1979]0008-5472/79/0039-0000$02.00

Carcinogen-induced Abnormalities in Rat Liver Cells and TheirModification by Chemical Agents1

E. Borenfreund,2 P. J. Higgins, M. Steinglass, and A. Bendich

Laboratory of Cell Biochemistry, Memorial Sloan-Kettering Cancer Center, New York, New York 10021

ABSTRACT

Epithelial cells derived from livers of rats beamingdiethylnitrosamine-induced hepatocancinomas were propagatedin vitro and cloned. These cells exhibited acentnic nucleiand j uxtanuclear cytoplasmic lesions indistinguishablefrom the Mallory hyaline bodies found in the cirrhotic liversof alcoholics. The hyaline region was found to be cornpnised of a filamentous aggregate in the cytoplasm whichcontained components that bound concanavalin A, an intemaction that was blocked by a-methylmannopyranoside.This hepatocellular aggregate also bound immunoglobulinG from the semaof normal rabbits and from patients withadvanced alcoholic cirrhosis. The immunoglobulin G fraction of semafrom normal goats, rats, or mice did not showsuch an affinity. The morphological appearance of cellscontaining this Mallory body-like lesion was altered, in areversible reaction, to one resembling normal rat liverepithelial cells after incubation with dimethyl sulfoxide orsodium butyrate. Such treatment also caused the cells tolose their acquired ability to localize concanavalin A andimmunoglobulin G in the juxtanucleam region. The presenceof the filamentous aggregate may be responsible for theacentnic displacement of the nucleus and might be comelated with the ability of the cells to express certain in vitroproperties associated with the transformed state. The cytoplasmic lesion may reflect an impairment of microtubulamfunction and polymerization.

INTRODUCTION

Long-term in vitro propagation of transformed epithelioidliver cells isolated from rats after p.o. administration ofdiethylnitrosamine has been reported (6, 7). These unusualcells possessed acentnic nuclei and were characterized bythe appearance of a juxtanuclear aggregation of filamentsresembling the Mallory bodies seen in the livers of patientswith advanced alcoholic cirrhosis and other degenerativeliver diseases (Ref. 34; for review, see Ref. 19). Cellscontaining Mallory body-like abnormalities were reisolatedfrom carcinomas formed following inoculation of the cultuned liver cells into newborn rats or nude mice (6). Thefilamentous aggregate was not modified when the cultureswere exposed to cytochalasin B or Colcemid (6), agentsknown to interfere with the polymerization of rnicmofilaments or microtubules, respectively. The cells and their

I This work was supported in part by USPHS Grant CA08748 fnom the

National Cancer Institute, National Cancer Institute Grant 1 RO 1 CA 20019-01 ET, and American Cancer Society Grant PDT-68.

2 To whom requests for reprints should be addressed, at Memorial Sloan

Kettering Cancer Center, 1275 York Avenue, New York, N. Y. 10021.Received August 17, 1978: accepted November21, 1978.

prominent morphological marker had not been previouslydescribed in vitro . They represent a unique model systemfor the investigation of the relationship between cytoskeletal structure and neoplastic growth characteristics. We nowreport studies concerning the nature of these carcinogeninduced abnormalities and the ability of certain chemicalagents to modify them.

MATERIALSAND METHODS

Cells and Cell Cultures. Cell lines containing Mallorybody-like lesions were derived, as described (6, 7), fromcollagenase-dissociated livers of Wistar rats sacrificed 10 to14 weeks after daily p.o. administration of diethylnitmosamine (50 ppm) in the drinking water and were established inculture. One of several cloned cell lines (72/22), all of whichproduced tumors when injected into rats or nude mice, wasselected for further investigation. Tumor-producing cellsderived from well-demarcated neoplastic liver nodes whichdid not contain these cytoplasmic lesions were studied forcomparison. Long-term cultures of matliver cells (9) whichmaintain a normal karyotype, do not grow in semisolid agarmedium, and do not produce tumors upon inoculation intonewborn rats or nude mice served as control.

Cells were cultured in Ham's F-12 medium supplementedwith 10.6 M dexamethasone, penicillin, streptomycin, Fungizone (2.5 j.@g/rnl),and 10% fetal calf serum in a humid 5%CO2 atmosphere at 37Â°.To examine the effect of chemicalagents on cell morphology, Me2SO3(Fisher Scientific Co.,N. V.) at final concentrations ranging from 1 to 3% orsodium butyrate (Pfaltz and Bauer, Stamford, Conn.), 2 to 4mM, was added to the growth medium. Covemslips wereremoved at various times; the cells were fixed with methanol and stained with Giemsa for viewing by light microscopy. Con A and the polysacchanides a-methylmannopymanoside and a-methylglucopyranoside (all from SigmaChemical Co. , St. Louis, Mo.) were used for binding studies. Rabbit sera were obtained from New Zealand Whiterabbits; human semawere supplied by the serum bank ofour Institute (courtesy of Dr. V. Hirshaut).

For tests of anchorage-independent growth, soft-agancultureoverlayscontaining1 x 10@to1 x [email protected] 0.33% Difco-Bacto agamIn F-12 medium supplementedwith 0.3% Tryptose phosphate and 10% fetal bovine serumwere prepared and placed on a base of 5.0 ml of 0.5% agarmedium in 60-mm plastic Petni dishes. Cultures were incubated for 2 weeks at 37Â°,at which time colonies werecounted . Several agamcolonies were isolated , disrupted by

3 The abbreviations used are: Me2SO, dimethyl sulfoxide; Con A, concan

avalin A; PBS, phosphate-buffered saline (137 mM NaCl:O.146 m,,i KH2PO4:8.1 m@ Na2HPO4:2.68 m@ KCI); FITC, fluorescein isothiocyanate: lgG,immunoglobulin G..

800 CANCER RESEARCH VOL. 39

on March 19, 2020. © 1979 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Carcinogen-induced Rat Liver Cell Abnormalities

trypsin, and further propagated in liquid medium.Immunofluorescence. For the indirect immunofluomes

cence assay (36), cells on coverslips were either fixed for 30mm with 3.5% formaldehyde in PBS (pH 7.2); postfixed withacetone:H2O (1:1) for 2 mm, with acetone (100%) for 5 mm,and with acetone:H,O (1:1) for 2 mm; and washed with PBSor else simply fixed with methanol for 30 mm. Then the cellswere incubated for 1 hr in moist chambers with normalrabbit, goat, mat,or mouse sera; with IgG purified therefrom(21); or with sera from normal individuals or from patientswith advanced alcoholic cirrhosis. The covemslipswere thenwashed with 2 changes of PBS for 30 mm, followed byincubation with the appropriate FITC-conjugated secondaryantibody to the primary IgG source. After 1 hn, the covenslips were washed with PBS and mounted in 50% glycerolin PBS. Preparations were viewed with a Zeitz photomicroscope equipped with UV optics.

Lectin Binding. The binding of Con A to the surface ofviable cells was studied using lectin which had been directlycoupled with fluorescein as described (22) according to themethod of Ash and Singer (2). Cells were incubated inmedium containing FITC-Con A (50 pg/mI) for 5 or 20 mmand then were fixed with 2% formaldehyde. In other expemiments, cells were first fixed with methanol, followed byincubation for 1 hr with FITC-Con A (80 @g/rnI),on else theywere incubated with unlabeled Con A (80 @g/ml)prior toexposure to FITC-Con A. In still other experiments, a mixtune of FITC-Con A with a-methylglucopyranoside or a-methylmannopyranoside was used to stain methanol-fixedcell monolayers. Coverslips were mounted in 50% glycerolin PBS and viewed with UV optics.

Electron Microscopy. For transmission electron microscopy studies, cells were pelleted, fixed in 2% glutaraldehyde, postfixed in 1% osmium tetroxide, and embedded inepoxy resin. Thin sections were stained with uranyl acetateand lead citrate and examined with a Siemens electronmicroscope.

Autoradiography. The effect on DNA synthesisof chemical agents added to the culture medium was examined withthe aid of [3H]thymidine. Cells were grown on covenslips inculture dishes in the presence of n-butyrate or Me2SOfor 3or 5 days, respectively. [3H]Thymidine [final concentration,0.25 @Ci/rnl(6 Ci/mmol)] was then added to freshly meplaced culture medium containing the agents and in parallelsets of dishes to medium from which the agent had beenomitted; the cells were incubated for an additional 18 hr.Unlabeled thymidine (10 j.@g/ml)was then added, followedafter 10 mm by a medium wash and fixation with methanol.Covemslip preparations were extracted with cold 5% trichloroacetic acid for 5 mm to remove unincorporated label,rinsed, and then coated with Kodak NTB@emulsion, diluted1:1. A total of 300 to 500 cells were scored for labeled nucleiin randomly selected fields.

RESULTS

Incubation of methanol-fixed hyaline body-containing ratliver cells (Fig. la) with normal rabbit sera or lgG purifiedtherefrom, followed by FITC-conjugated goat anti-rabbitIgG, resulted in the appearance of a brightly fluorescingjuxtanuclear area coincident with the Mallory body region

c..

Fig. 1. a, clonal subline (72/22) of an established culture of liver cellsisolated from a rat 3.5 months after daily p.o. administration of diethylnitnosamine and propagated in vitro for 7 months. Note darkly stained hyalinebodies adjacent to acentnic nuclei. Cason's Mallory stain; partial Nomanskioptics, x 510. b, indirect immunofluorescence. Cloned liver cells (a) weremethanol fixed, incubated with nonimmune rabbit serum, and stained withFITC-conjugated goat anti-rabbit IgG as described in â€œMaterialsand Methods.â€•Note fluorescence in the juxtanuclear area coincident with the Mallorybody region. uv optics, x 330.

when viewed in UV light (Fig. ib). No fluorescence wasobserved after incubation with lgG from goats, mats, ormice . Neither neoplastic carcinogen-transformed rat livercells devoid of Mallory body-like lesions nor normal rat livencells established in vitro (9) and examined in parallel experiments gave the fluorescence reaction after incubation withnonimmune rabbit lgG.

To investigate whether the cytoplasmic lesion could bemodified by agents known to interact with cell surfacereceptors (2), cultures of the juxtanuclear aggregate-contaming liver cells were grown in the presence of Con A (4 to20 @g/ml;2 to 10 x 10@ M). This carbohydrate-bindinglectin combines with sites most complementary to a-methyl-D-mannopymanosyl residues of polysacchamides orglycoproteins (24). No alterations in cellular morphology orgrowth characteristics could be observed when Con A waspresent in the medium. However, when methanol-fixed cellswere incubated with 4 x 10@ M FITC-Con A a brightfluorescence in the hyaline region was observed (Fig. 2).This binding was partially inhibited by preincubating thecells with 4 x 10@ M unlabeled Con A and completelyinhibited with 5 x 10@M unlabeled Con A. No fluorescencewas observed when methanol-fixed cells were incubated

801MARCH 1979

.â€œ4@ .T@



E. Borenfreund et a!.

with a 4 x 10â€”iM solution of FITC-Con A which hadpreviously been mixed with a-methylmannopyranoside (5x 102 M) 30 mm prior to overlaying on the cells. Theaddition of a-methylglucopyranoside (5 X 102 M) to a FITCCon A solution (4 x 106 M) only partially prevented Con Abinding to sites in the juxtanuclear region. This is inkeeping with the finding that Con A binding to the a-Dglucans is less specific than for the a-D-mannans (24). Itwas of interest to determine whether viable filamentousbody-containing liver cells differed from normal liver cellsin the pattern of Con A binding to surface membranereceptors. FITC-Con A was, therefore, added to the incubation medium. Whereas normal rat liver epithelial cellsshowed a uniformly distributed surface fluorescence after a20-mm incubation with FITC-Con A, hyaline body-containing cells exhibited a patchy clustering of fluorescence(Figs. 3 and 4) at the surface. Similar fluorescent clustersinduced by Con A have been observed in virus-transformedrat kidney fibroblasts (3, 18) and in rat kidney or chickfibroblasts and mouse lymphocytes after incubation withcolchicine, an agent known to interfere with microtubuleassembly (2, 18, 54).

Fluorescence was not observed in either the lgG or Con Aincubation experiments if cells had been fixed with formaldehyde rather than with methanol. Methanol-fixed normalor transformed rat liver cells that did not contain Mallorytype hyaline bodies did not fluoresce after incubation withrabbitlgG.Fluomescein-coupledphytohemagglutininat aconcentration of 200 @g/mldid not stain the Mallory-likebodies.

Theme have been reports that patients with alcoholiccirrhosis might produce liver-directed autoantibodies (55,57). Sera from such patients have produced bright immunofluorescence reactions in the Mallory bodies (alcoholichyalin) of cells in cryostat sections of cirrhotic livers ofalcoholics (38, 56). Examination of our cultured liver cellswith sera obtained from patients with advanced alcoholiccirrhosis revealed that these cells similarly showed a brightfluorescence in the hyaline body area with 2 of 5 of thecirrhotic human sera tested (Fig. 5). Normal rat liver cells orneoplastic liver cells which did not contain such cytoplasmic abnormalities did not react with these sera. Semafrom4 normal individuals were fluorescence negative with allcell lines tested.

Reversal Studies. A series of chemical agents found toinduce differentiation in various cell systems (1, 10, 16, 20,27, 32, 39, 45) were tested for their effects on the hyalinebody-containing liver cells and on their growth characteristics. Whereas bmomodeoxyunidine, ouabain , hypoxanthineor dibutyryl cyclic adenosine 3':5'-monophosphate, andtestosterone were without significant effect, inclusion of2% Me,SO in the culture medium for 2 weeks resulted in amarked flattening and a gradual disappearance of thecytoplasrnic aggregate in about 60% of the cells (Fig. 6).Although the population doubling time was increased overthat of cultures growing in the absence of Me2SO, cellscontinued their propagation in the presence of this agentfor more than 2 months. However, when Me@SOwas memoved from the medium, cells reverted to their originalmorphology within 2 to 3 days. Cells in which the cytoplasmappeared normal after growth in the presence of Me,SO no

longer bound nonimmune IgG or FITC-Con A in the juxtanuclear area; those cells which had retained the filamentous bodies still showed fluorescence. Reappearance of theMallory-like hyaline body and subsequent lgG- and Con Abinding was again demonstrated 3 days after return of thecultures to normal growth medium.

Sodium butymatewas found to be even more effective inaltering the characteristic cytoplasmic lesion. The agentwas tested over a concentration mangeof 2 to 4 mM at pH7.0 to 7.4. Whereas 4 mM butyrate was toxic to the cellswhen present in the medium for more than 1 week, 2 mMbutymatewas not deleterious, and 3 mM solutions were welltolerated. A concentration of 3 mM, which slowed but didnot inhibit proliferation, produced marked cell flatteningand the disappearance of the cytoplasmic aggregate in 70%of the cells after 4 to 6 days of incubation (Fig. 7). Incubation with sodium butyrate, inhibited the binding of rabbitpreimmune lgG and FITC-Con A to the Mallory body regionof methanol-fixed cells. Removal of the butyrate from themedium led to the reappearance of the Mallory-like bodyafter 2 to 3 days, and with it the ability to bind lgG and ConA returned. Half of the cells that had lost their Mallory-likehyaline bodies after incubation with 3 mM sodium butynateregained this cytoplasmic lesion after treatment with 106 MColcemid for 3 hr. One might infer that the disappearanceof the hyaline body in the butyrate-treated cell was causedby a â€œnormalization'â€of the cytoskeletal structure, whichwas again disrupted by Colcemid.

Changes in the cytoplasmic organization were followedby transmission electron microscopy. Whereas cells grownin normal medium contained a well-demarcated juxtanuclear fiiamentous cytoplasmic area surrounded by a ring ofrnitochondnia (6, 7), incubation with Me2SO or butyrateresulted in a more homogeneous distribution of cytoplasmic organelles and in an apparent dispersal of the filamentous aggregate (Figs. 8 and 9).

Examination of the capacity of the Mallory body-containing liver cells and that of their Me@SO-or butyrate-treatedcompanion cultures for anchorage-independent growth insemisolid agar medium revealed that cells with the cytoplasmic lesion could grow in soft agar. Cultures previouslyincubated with Me2SO or n-butymate did not grow if theagents were also present in the agammedium; they did formcolonies, however, when the agents were omitted. Thus,the transformed cells (72/22) not only assumed a morphologically more normal appearance during incubation withMe@SOor butynate, but they were also anchorage dependent for growth, as is characteristic of normal cells. However,other transformed cell lines such as HeLa, Meth A, andseveral of our hepatoma lines, which usually grow in softagar, also did not grow when Me2SO or butyrate wasincluded in the agar medium, although they were able toreplicate (but more slowly) in liquid medium which contamed these agents.

Autoradiography. The autoradiographic technique wasused in order to assess the ability of the Mallory bodycontaining cells grown in the presence of 3 mM butyrate for3 to 6 days to synthesize DNA; the results indicate that only10 to 13% of the cellsso treatedhad incorporated[3H]thymidine. However, when normal medium containing[3H]thyrnidine was added to cells which had previously been





incubated for 3 or 5 days with 3 mM butyrate, the labelingindex increased to 64 or 80Â°/o,respectively (Table 1).Repression of DNA synthesis was dependent upon theconcentration of butynate. In the presence of 2 mM butyrate,27 to 30% of the nuclei were labeled, and with 4 mMbutyrate6 to 9% were labeled.Increasesinthe labelingindex upon change into normal medium were in the 60 to80% range for all concentrations tested. The results indicatethat the cells were still viable despite the suppression ofDNA synthesis induced by butyrate. The DNA labeling indexof control, untreated cells containing Mallory body-likelesions when propagated in normal medium was 95%.

DISCUSSION

We recently reported a carcinogen-induced juxtanuclearcytoplasmic abnormality in rat liver cells (6, 7). The abnormal cytoplasmic region of these neoplastic cells was characterized by light and electron microscopy as a membraneless but well-demarcated homogenously staining area consisting of an aggregate of 100- to iso-A filaments summounded by mitochondnia and elements of the Golgi apparatus. An interacting cytoskeletal network of actin filaments, microtubules, and ioo-A filaments (intermediatefilaments) is believed to be necessary for the regulation ofcellular function. Antisera that are specific for these cytoskeletal components have served to identify them and helpreveal their mole,distribution, and interrelationship in normal and virus-transformed cells (28). An altered cytoskeleton, observed in rat kidney or chick fibroblasts infected invitro with tumor viruses, has been attributed to abnommalities of microfilamentous or microtubular structure or to amarked decrease of their components (3, 11, 18, 42, 49,50). This suggests that cytoplasmic filaments may play amole in the process of transformation and the loss ofcontact-inhibited growth. The appearance of patchy clustenings of cell surface membrane receptors to Con A ontransformed but not on normal fibroblasts has been attributed to a transmembmane linkage of the receptors withfilamentous elements inside the cell (3, 17, 18, 35). Suchvirus-induced changes resemble the cytoskeletal abnormalities (Mallory bodies) we observed in epithelial liver cellsafter administration of the carcinogen diethylnitrosamine torats. A cap-like juxtanucleam formation of filaments, similarto those in our system, has also been seen after treatmentof cultured fibroblasts with Colcemid (4, 23, 41, 47) and inthe livers of mice either after long-term in vivo administration of griseofulvin or after short-term feeding of gniseofuiyin followed later by injections of colchicine (14, 15). SinceColcemid, colchicine, and gniseofulvin are antimitoticagents known to interfere with the polymerization of tubulin(43, 51-53), it has been suggested that these filaments wereformed at the expense of microtubules (13).

Whereas the abnormalities induced by the antimitoticagents in other cell systems (4, 23, 41, 47) are reversibleafter the removal of the agents from the culture media, thechanges in our rat liver cell system after the administrationof diethylnitrosamine are permanent and consequently appear to be under genetic control.

The reversion of the liver cells from an abnormal to amomphologically more normal-appearing cytoplasm me

Table 1

Effect of butyrate on incorporation off3HJthymidine into DNAHyaline body-containing rat liven cells were incubated in the

presence of butyrate for the designated time interval; [3H]thymidinewas added, and incubation was continued for 18 hr in the presenceof butyrate on in normal medium. Nuclei of 300 to 500 cells were

@@Â±nrandomly selected areas. _@@ - - - -

Nuclei labeled (%)Concentration of bu- Incubation In presence of In normal

@@ time (days) butyrate medium2 3 30 702 5 27 733 3 13 643 5 10 804 3 6 584 5 9 77

None - â€”---- -@- _ -- - --

quired the continuous presence of Me2SOor butynate; theiromission from the medium led to the reappearance of thelesion. It is not clear whether these agents affected theassembly of microtubules or whether the observed resultswere due to an indirect effect on membrane receptors orother cellular constituents. Me2SO has recently been foundto induce the polymerization of tubulin in vitro in theabsence of tubulin-associated proteins (26). It is intriguingto speculate that tubulin polymerization is enhanced bythese agents; indeed, we observed a strikingly rapid reappearance of the hyaline body when cells which had revertedto a more normal type of morphology because of butyratetreatment were incubated for 2.5 hr with 106 M Colcemid,an agent known to interfere with microtubule formation(53).

The reasons for the fluorescence seen after incubation ofthe cells with lgG from nonimmune rabbit sera or with serafrom patients with advanced alcoholic cirrhosis are currently under investigation. Normal rabbits are known toproduce autoantibodies to actin (48) and tubulin (29). However, nonimmune lgG absorbed with actin or tubulin wasstill reactive in our system (8); these cytoskeletal components thus do not seem to be the active constituentsinvolved. Furthermore, we found that extracts of preparations of the cytoplasmic bodies did not show actin ortubulin bands by polyacrylamide gel electmophoretic analysis (4, 8, 46). Human sera containing smooth muscle autoantibodies, obtained from cancer patients, reacted withvinblastine-induced perinuclear intermediate filaments inhuman embryonic skin fibroblasts. This reaction, observedby indirect immunofluorescence, was unaffected by punfied actin; it was inhibited by a preparation of isolatedintermediate filaments (30).

If the components involved in the Mallory body formation

are indeed constituents of intermediate filaments, theirabnormal, juxtanuclear aggregation might have an adverseeffect on cell physiology and growth patterns. Intermediatefilaments are believed to play a role in the anchorage ofnuclei (33). They appear to attach to nuclear pores and tothe plasma membrane and could thus serve as a link in thetransmission of information between the nucleus and thecell surface (37). The retraction of these filaments into ajuxtanuclear cap, with a concomitant acentnic displacement

MARCH 1979 803




of the nucleus, may lead to a disruption in the orderly flowof information.

Since the continuous presence of Me2SO or butymate inthe culture medium was required for the maintenance of anormal-appearing phenotype, it was not feasible to test thetumonigenicity of these cells in vivo. Examination of theirgrowth potential in soft agam, on the other hand, waspossible by including the respective chemicals in the agammedium. Under those experimental conditions, Me2SO- orbutymate-treated cells did not grow in soft agam mediumand, therefore, behaved like normal cells, but they prolifemated and formed colonies as expected of transformed cellswhen the agents were not included in the agar. A suppression of the growth of neoplastic Syrian hamster cells, grownin butymate-containing agam medium, has recently beenshown (31).

It has been reported thatn-butyrate shuts off DNA synthesis reversibly in chick embryonic fibmoblasts and HeLa cellsand that cytosol from butymate-tmeatedcells will turn offDNA synthesis in control nuclei (25); butymatealso has beenfound to inhibit histone deacetylase (12, 44). Although therewas a marked decrease of DNA synthesis in our cell systemafter butyrate exposure, utilization of [3H]thymidine was notcompletely inhibited in all cells even after 6 days of exposure to butymate. The mechanism of the effect of butyrateon the cytoplasmic organization is not clear at this time.

The hyaline body-containing matliver cell lines establishedin our laboratory (6) thus provide a system to explore theaction and reversibility of â€˜â€˜differentiation'â€˜-inducingagentsand the interaction between cytoskeletal components andtheir relationship to cellular function and oncogenic tmansformation. Cellular abnormalities in the Chediak-Higashisyndrome (5), an inherited autosornal recessive disorder ofchildren, also involve cytoskeletal (microtubular) and othermorphological alterations (40) similar to those discussed inthe present study. The possible relationship between animpaired cytoskeletal structure and neoplastic transfommation deserves further exploration.

ACKNOWLEDGMENTS

We thank C. Wand for competent technical assistance and L. Clements forvaluable assistance with the electron microscopy.

REFERENCES

1. Altenburg, B. C., Via, D. P., and Steiner, S. H. Modification of thephenotype of murine sarcoma virus-transformed cells by sodium butyrate. Exp. Cell Res., 102: 223-231 , 1976.

2. Ash, J. F., and Singer, J. Concanavalin A-induced tnansmembranelinkage of concanavalin A surface receptors to intracellular myosincontaining filaments. Proc. Nati. Acad. Sd. U. S. A., 73: 4575â€”4579,1976.

3. Ash, J. F., Vogt, P. K., and Singer, S. J. Reversion from transformed tonormal phenotype by inhibition of protein synthesis in rat kidney cellsinfected with a temperature-sensitive mutant of Rous sarcoma virus.Proc. Natl. Acad. Sci. U. S. A., 73: 3603-3607, 1976.

4. Blose, S. H., Shelanski, M. L., and Chacko, S. Localization of bovinebrain filament antibody on intermediate (100 A) filaments in guinea pigvascular endothelial cells and chick cardiac muscle cells. Proc. Nati.Acad. Sci. U. S. A., 74: 662-665, 1977.

5. Blume, R. S., and Wolff, S. M. The Chediak-Higashi syndrome: studiesin four patients and a review of the literature. Medicine, 51; 247-280,1972.

6. Borenfreund, E., and Bendich, A. In vitro studies of hepatocellularchanges induced in rats exposed to diethylnitrosamine. Lab. Invest. , 3:295-303, 1977.

7. Bonenfreund, E., Higgins, P. J., and Bendich, A. In vivo initiated rat livercarcinogenesis studied in vitro ; formation of alcoholic hyaline-typebodies. Cancer Left., 3: 145-150, 1977.

8. Borenfneund, E. , Higgins, P. J., and Peterson, E. Carcinogen-inducedMallory bodies in liver cells studied in vitro. Proc. Am. Assoc. CancerRes., 19: 19, 1978.

9. Borenfreund, E., Higgins, P. J., Steinglass, M., and Bendich, A. Properties and malignant transformation of established rat liver parenchymalcells in culture. J. Nati. Cancer Inst., 55: 375â€”384,1975.

10. Bonenfneund, E., Steinglass, M., Korngold, G., and Bendich, A. Effect ofdimethylsulfoxide and dimethylformamide on the growth and morphology oftumor cells. Ann. N. Y. Acad. Sci.,243: 164â€”171,1975.

11. Brinkley, B. A., Fuller, G. M., and Highfield, D. P. Cytoplasmic microtubules in normal and transformed cells in culture: analysis by tubulinantibody immunofluorescence. Proc. NatI. Acad. Sci. U. S. A., 72: 4981-4985, 1975.

12. Candido, E. P. M., Reeves, A., and Davie, J. A. Sodium butyrate inhibitshistone deacetylation in cultured cells. Cell, 14: 105â€”113,1978.

13. De Brabander, M., Aerts,F.,Van de Veire,A.,and Borges, M. Evidenceagainst interconversion of microtubules and filaments. Nature, 253: 119â€”120, 1975.

14. Denk, H., and Eckerstorfer, A. Colchicine-induced Mallory body formation in the mouse. Lab. Invest. , 36: 563-565, 1977.

15. Denk, H., Eckerstorfer, A., Gschnait, F., Konrad, K., and Wolff, K.Experimental induction of hepatocellular hyalin (Mallory bodies) in miceby griseofulvin treatment. Lab. Invest., 35: 377-382, 1976.

16. Deutsch, S. I., Silvers, 0. N., Cox, A. P., Griffin, M. J., and Ghosh, N. K.Ultrastructunal and enzymic modulation of HeLa cells induced by sodiumbutyrate and the effects of cytochalasin B and Colcemid. J. Cell. Sd.,21,391-406,1976.

17. Edelman, G. M. Surface modulation in cell recognition and cell growth.Science, 192: 218-226, 1976.

18. Edelman, G. M., and Yahara, I. Temperature-sensitive changes in surface modulation assemblies of fibroblasts transformed by mutants ofRous sarcoma virus. Proc. NatI. Acad. Sci. U. S. A., 73: 2047-2051 , 1976.

19. French, S. w., and Davies, P. L. The Mallory body in the pathogenesis ofalcoholic liver disease. In: J. M. Khanna, Y. Israel, and H. Kalant (eds.),Alcoholic Liver Pathology, pp. 113-143. Toronto: House of Lind, 1975.

20. Friend, C., Scher, W., Holland, J. G., and Sato, T. Hemoglobin synthesisin munine virus-induced leukemic cells in vitro. Stimulation of erythroiddifferentiation by dimethylsulfoxide. Proc. NatI. Acad. Sd. U. S. A., 68:378-382,1971.

21. Garvey, J. S., Cremer, N. E., and Sussdorf, D. H. Isolation of rabbitantibodies and their subunits. In: Methods in Immunology, Ed. 3, pp.183-234. New York: W. A. Benjamin, Inc., 1977.

22. Goldman, A. D. Labeling agents and procedures for conjugation. In:Fluorescent antibody methods, Chap. 7, pp. 97-117. New York: AcademicPress,Inc.,1968.

23. Goldman, A. D. The role of three cytoplasmic fibers in BHK-21 cellmotility.J.CellBiol.,51:752-762,1971.

24. Goldstein, I. J., Reichert, C. M., and Misaki, A. Interaction of concanavalin A with model substrates. Ann. N. V. Acad. Sci., 234: 283-295, 1974.

25. Hagopian, H. K., Riggs, M. G., Swartz, L. A., and Ingram, M. V. Effect ofn-butyrate on DNA synthesis in chick fibroblasts and HeLa cells. Cell,12:855-860,1977.

26. Himes, R. H., and Burton, P. The DMSO-induced assembly of tubulin inthe absence of associated proteins. J. Cell Biol., 75: 269a, 1977.

27. Hsie, A. W., and Puck, T. T. Morphological transformation of Chinesehamster ovary cells by dibutyryl adenosine 3' ,5'-monophosphate andtestosterone. Proc. Natl. Acad. Sci. U. S. A., 68: 358â€”361, 1971.

28. Hynes, A. 0. , and Destree, A T. 10 nm filaments in normal andtransformed cells. Cell, 13: 151-163, 1978.

29. Karsenti, E., Guilbert, B., Bonneus, M., and Avrameas, S. Antibodies totubulin in normal non-immunized animals. Proc. NatI. Acad. Sci. U. S.A.,74:3997-4001,1977.

30. Kurki, P., Linder, E. , Virtanen, I. , and Stenman, S. Human smoothmuscle autoantibodies reacting with intermediate (100 A) filaments.Nature, 268: 240-241 , 1977.

31. Leavitt, J., Barrett, J. C., Crawford, B. 0., and Ts'o, P. Butynic acidsuppression of the in vitro neoplastic state of Syrian hamster cells.Nature,271: 262-265, 1978.

32. Leder, A., and Leder, P. Butyric acid, a potential inducer of erythroiddifferentiation in cultured erythroleukemic cells. Cell, 5: 319â€”322,1976.

33. Lehto, V. P., Virtanen, I. , and Kurki, P. Intermediate filaments anchorthe nuclei in nuclear monolayers of cultured human fibroblasts. Nature,272: 175-177, 1978.

34. Mallory, F. B. Cirrhosis of the liven. Five different types of lesions fromwhich it may arise. Bull. Johns Hopkins Hosp. , 22: 69â€”75,1911.

35. McClain, 0. A., D'Eustachio, P., and Edelman, G. M. Role of surfacemodulating assemblies in growth control of normal and transformedfibroblasts. Proc. NatI. Acad. Sci. U. S. A., 74: 666â€”670,1977.

36. Mellors, A. C. The application of labeled antibody technics in studyingcell antigens. Cancer Res., 28: 1372-1381 , 1968.





37. Metuzals, J., and Mushynski, E. Electron microscope and experimentalinvestigations of the neurofilamentous network in Deiters' neurons. J.CellBiol.,61:701-721,1974.

38. Nenci, K. Identification of actin-Ilke proteins in alcoholic hyaline byimmunofluorescence. Lab. Invest., 32: 257-260, 1975.

39. Nudel, U., Salmon, J. E., Terada, M., Bank, A., Aifkind, A. A., andMarks, P. A. Differential effects of chemical inducers on expression of 13globin genes in murine erythroleukemia cells. Proc. NatI. Acad. Sci. U.S. A., 74: 1100-1104, 1977.

40. Oliver, J. M. Impaired microtubule function correctable by cyclic GMPand cholinergic agonists in the Chediak-Higashi syndrome. Am. J.Pathol., 85: 395-417, 1976.

41. Osborn, M., Franke, W. W., and Weber, K. Visualization of a system offilaments 7-10 nm thick in cultured cells of an epithelloid line (Pt K2) byimmunofluorescence microscopy. Proc. NatI. Acad. Sci. U. 5. A. , 74:2490-2494, 1977.

42. Pollack, R., Osborn, M., and Weber, K. Patterns of organization of actinand myosin in normal and transformed cells. Proc. NatI. Acad. Sci. U. S.A., 72: 994-998, 1975.

43. Roobol, A., Gull, K., and Pogson, C. I. Evidence that griseofulvin bindsto a microtubule associated protein. FEBS Left., 75: 149â€”153,1977.

44. Sealy, L. , and Chalkley, A. The effect of sodium butyrate on histonemodification. Cell, 14: 115-121, 1978.

45. Silagi, S., and Bruce, S. A. Suppression of malignancy and differentiation in melanotic melanoma cells. Proc. NatI. Acad. Sci. U. S. A., 66: 72-78, 1970.

46. Sim, J. S., Franks, K. E., French, S. W., and Caldwell, M. G. Mallorybodies compared with microfilament hyperplasia. Arch. Pathol. Lab.Med.,101:401-404,1977.

47. Starger, J. M., and Goldman, A. D. Isolation and preliminary characterization of 10 nm filaments from baby hamster kidney (BHK-21) cells.

Proc. NatI. Acad. Sci. U. 5. A., 74: 2422-2426, 1977.48. Trenchev, P., and Holborow, E. J. The specificity of anti-actin serum.

Immunology, 31: 509-517, 1976.49. Wang, E., and Goldberg, A. A. Changes in microfilament organization

and surface topography upon transformation of chick embryo fibroblastsby Rous sarcom$iitrus. Proc. NatI. Acad. Sci. U. S. A., 73: 4065-4069,1976.

50. Weber, K., Lazanides, E., Goldman, A. D., Vogel, A., and Pollack, R.Localization and distribution of actin fibers in normal, transformed andrevertant cells. Cold Spring Harbor Symp. Quant. Biol., 39: 363-369,1974.

51. Weber, K., Wehland, J., Herzog. W. Gniseofulvin interacts with microtubules both in vivo and in vitro. J. Mol. Biol., 102: 817-829. 1976.

52. Wehland, J., Herz@g,W., and Weber, K. Interaction of griseofulvin withmicrotubules, mlctotubule protein and tubulin. J. Mol. Biol., 111: 329â€”342, 1977.

53. Weisenberg, A., Borisy, G. G., and Taylor, E. W. The colchicine-bindingprotein of mammalian brain and its relation to microtubules. Biochemistry, 7: 4466-4479, 1968.

54. Yahara, I. , and Edelman, G. M. Electron microscopic analysis of themodulation of lymphocyte receptor mobility. Exp. Cell Aes., 91: 125-142, 1975.

55. Zetterman, A., Chen, T., and Leevy, C. M. Autoimmunity and alcoholicliver disease. In: C. M. Leevy (ed), Diseases of the Liver and BiliaryTract, pp. 27-31 . Basal: S. Karger, 1975.

56. Zinneman, H. H. Autoimmune phenomena in alcoholic cirrhosis. Am. J.Dig. Dis., 20: 337-345, 1975.

57. Zinneman, H. H., and Levi, D. F. Separate autoimmune mechanisms for75 and 195 globulins in Laennec's cirrhosis. Arch. Intern. Med., 124:153-159, 1969.

805MARCH 1979




- \@%@â€¢

..@

. -@@

:.@@ ..

@ ,

!4@

@â€˜@L @1@I@ â€œ@

. .. . ..,b@..,!_- â€˜@

@ a

Fig. 2. Liver cells (72/22) were methanol fixed and then incubated with FITC-Con A (80 @tg/ml)for 1 hr as described. Note fluorescence concentrated inthe juxtanuclear area. UV optics. x 430.

Fig. 3. Liven cells (72/22) incubated in the living state with FITC-Con A (50 j@g/ml) for 20 mm and then fixed with 2% formaldehyde. Note the patchydistribution of Con A-binding sites. UV optics. x 330.

Fig. 4. Normal rat liven cells (9) incubated in the living state with FITC-Con A as in Fig. 3. Note uniform distribution of Con A-binding sites. UV optics, x330.

Fig. 5. Indirect immunofluorescence. Liver cells (72/22) were methanol fixed and then incubated with serum from a patient with advanced alcoholiccirrhosis and stained with FITC-conjugated goat anti-human lgG as described. UV optics. x 410.

Fig. 6. Liver cells (72/22) grown in the presence of 2Â°/oMe@SOfor 42 days. Note the absence of aggregate in the cytoplasm. Methanol fixation, Giemsastain.x 330.

Fig. 7. Liven cells (72/22) grown in the presence of 3 mM sodium butyrate for 12 days. Note the absence of juxtanuclear aggregate in the cytoplasm.Methanol fixation. Giemsa stain. x 330.

. . a,

â€˜.::. . . @,. p.

@@lLi

I a

. .â€˜ :@ .@

0. - S

@...

- , 3'â€˜a

- â€˜ -; a

â€¢@â€¢_.#. .,-,@-.

. S

,@

- @.

\ 3, -@@ â€˜

.@.:,

:@

JoV

â€¢*;@ â€¢â€¢ â€¢.@ +

--:@ .

... .-

-...,. 6

CANCER RESEARCH VOL. 39806

Os â€¢ ,1@@)â€¢@4'.@

@Ã³..@_ .@. @. &

- a@

@ 0

a -

,, , ,- .bi@$@â€”,@

@ .@â€¢$s:',

dI@;@ lIp@

@1

h@;:T'â€˜@ @a â€¢

-@ â€¢@@l@sF/@

@ 0@.




I ,v:.@'@'@â€¢

1'@@@

-.4. ,@ ,.@@

5'. . â€¢.

,-â€˜,, , â€”

. â€˜.. t @.-. . :â€˜;

@ . :.â€˜1#' :. r'@ â€¢@@ A

. . -- , )â€˜ -. - . v.-. @.â€˜

@@ @; 9

â€˜@ â€˜@ a

@ p@ __@ @; - .5@Fig. 8. Transmission electron micrograph of a liver cell (72/22) cloned from a culture derived from a rat after daily administration of diethylnitrosamine for

3.5 months. Note the Mallory-type body (MB) next to the nucleus (N) and the ring of mitochondria. Glutaraldehyde fixed. Uranyl acetate and lead citrate, x6500.

Fig. 9. Transmission electron micrograph of a liver cell (72/22) after incubation for 6 days in the presence of 4 mM sodium butyrate in the culture medium.Note the absence of filamentous aggregate and the random distribution of mitochondnia. N, nucleus. x 6500.

807MARCH 1979



1979;39:800-807. Cancer Res E. Borenfreund, P. J. Higgins, M. Steinglass, et al. Modification by Chemical AgentsCarcinogen-induced Abnormalities in Rat Liver Cells and Their

Updated version

http://cancerres.aacrjournals.org/content/39/3/800

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/39/3/800To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



carcinogen-inducedabnormalitiesin rat livercells …...incubation of methanol-fixed hyaline...

Documents