[CANCER RESEARCH 36, 4491-4499, December 1976]

lions in the plasma membrane accompanying an oncogenicconversion of cultured cells will undoubtedly be present.Certainly, hepatocarcinoma cells differ from normal hepatocytas in plasma membrane enzyme characteristics (2, 7).Histochamical studies of the distribution of phosphataseshave indicated that, while nucleosida tniphosphatase achyity is localized only on the plasma membranes facing bile

canaliculi in normal mathapatocytas, an intense activity isdetectable over the entire surface of neoplastic cells in theliver of rats fed carcinogens (17, 18). This altered distnibulion of surface nucleoside phosphatasa activity has beenpreviously reported for transplantable well-differentiatedhapatomas (9, 24). It is uncertain, though, whether thesechanges occur during continuous transplantation in viva orwith the initial transformation. A significant increase inATPase activity along the entire surface membrane duringpassages of rat liver apithelial cells in vitro can be demonstratad(18,19).

This cell surface alteration appears to begin coincidantally with the acquisition of tumonigenicity (18, 19). Takingnote of this, an analysis of tumomiganicity of cultured livercalls and the histochemical localization of surface ATPaseactivity was extended to several independently isolated calllines from normal rat livers as well as some transformedliver call lines. All of these cell lines were chosen from thosethat have shown evidence of specific functions on components characteristic of hepalocytes. In view of the problemsinherent in the histochamical technique (8, 10, 12, 16, 22,23), it was desirable to add a biochemical procedure for thedemonstration of an ATPase activity with intact cells inwhich viability was unimpaired during the enzyme assay.Agran at a!. (1) have demonstrated that incubating culturedcalls with [y-32PJATPreleases 32p1into the medium in relalion to membrane-bound ATPase. Applying this technique,the characteristics of membrane-bound ATPase of tumonigenic liver epithalial calls can be analyzed and comparedwith results of histochamical techniques. Finally, the possibla functional significance of the phosphalasa and its usafulness as a parameter for epithelial oncogenasis in vitro isconsidered.

MATERIALS AND METHODS

, This work was supported by grants from the National Cancer Institute ofCanada, le Ministère des Affaires Sociales du Québec,La Fondation J. H.Biermans, Les Fondations J. Rheaume, and by USPHS Grant CA-13376.

a Research Associate of the National Cancer Institute of Caneda. To whomrequests for reprints should be addressed, at Institut du Cancer de Montréal,centre Hospitalier Notre-Dame, Montréal, Canada H2L 4M1.

Received May 4, 1976; accepted September 1, 1976.

Cell Lines. Table 1 describes the origin of the call linesestablished for this investigation. Normal diploid call linesware isolated from the primary cultures of dissociated hapalocytes of Wistar rats (25). Other lines have been previouslyreported on (11, 18, 19, 26-28, 33, 35). Cell lines studied

DECEMBER1976 4491

Surface Membrane Nucleoside Triphosphatase Activity andTumorigenicity of Cultured Liver Epithelial Cells1

Shuichi Karasaki2 and Tohru Okigaki

Institut du Cancerde Montréal,CentreHospitalierNotre-Dame,Montréal,H2L4M1and DOpartementd'Anatomie,Universitdde Montréal,Canada(S. K.J, and Pasadena Foundation of Medical Research, Pasadena, California 91 101 fT. 0.)

SUMMARY

A call surface-located nucleosida tniphosphataseactivitycan be assayed in liver apithalial cultures in situ with theincubation of intact calls in a medium containing [y32Pjadanosina tniphosphata and correlated with the tumonigenicity of these calls in neonatal Wistar rats. The actoanzyme activity of normal diploid cell lines is minimal,whereas a considerably high activity has been found in alltumorigenic cell lines tested . The optimum condition for theadenosinatniphosphatase activity is physiological with megard to osmolanity, ionic composition, pH, and substrateconcentration in the medium. The enzyme is significantlystimulated by Ca2@,and its activation is controlled by Mg2@.Histochemical examinations indicate that glutamaldahydafixed cells of tumoriganic lines have Ca2@-stimuIatedadenosinetriphosphatasa activity on the external surface. The isotopic assay of adanosina tniphosphata hydrolysis by intactcells may provide a rapid method for screening oncoganasis in vitro of liver apithelial calls.

INTRODUCTION

Epithalial cell cultures derived from rat liver can provide auseful system for in vitro analysis of chemical carcinogenasis (4, 6, 15, 20, 35, 38, 39). However, many studies havebeen frustrated by difficult reproduction of differentiatedliver cells (13, 15, 26), inadequate morphological panamatens for epithalial oncogenasis in vitro (6, 27, 28, 31, 35, 38,39), spontaneous neoplastic transformation (3, 18—20,28,32), and viral expressions of oncomna type (19, 35). Theproduction of invasive carcinomas on implantation into appropniate hosts has been a criterion for oncogenesis (18-20,27, 28, 32, 38). However, negative results of such lasts arenot conclusive and may in fact be due to immunologicalsuppressions (3, 4, 6, 35). Consequently, Weinstein at a!.(35) proposed that the capacity of growth in soft agar suspension is a more maliableindication of neoplastic transfonmation. While this may be useful, a simpler, rapid, andaccurate assay technique is still elusive.

Regardless of what form that assay will develop, altema

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

Cell linesOriginHepatocyticmar

kems―NormalRL-1Wistar

rat adult liver, explanted in May 1972 (25)A,BRL-33Wistarrat 27 days liver, explanted in June 1974(25)A, B,CRL-34Wistarrat 7 days liver, explanted in June 1974 (25)A, B,CRL-36Wistarrat 12 days liver, explanted in June 1974(25)A,BTransformed―RL34-ESARL-34,

treatedfor 7 dayswith benzo(a)pyrene(1 @.&g/ml)in Oct 1974and grown insoft agar in December1975(27).A,

B,CR134-HIIRL-34,

treatedton2 hr with 4-nitroquinoline N-oxide, (5 x 1O@M) in October1974(27).A,

B,CRLCWistar

adult rat liver, explanted in May 1969 (11).A, B,CRLCC-iRLC, cloned in February1972(26).A, B,DRLC-TVIRLC

C-i , explantedafter growth in vivo in a Wistar rat in September1973(26).A, B,DHNTWistaradult rat liver, explanled in May 1973(19).B,DHTCMorrishepatoma7288Cof a Buffalo rat, explanted in October 1964(33).A

S. Karasaki and T. Okigaki

Table1Liver epithelial cell lines

a A, hydrocortisone-inducible tyrosine aminotransferase; B, peroxisomes; C, glycogen granules; D, canaliculi.b Cells positive for growth in soft agar.

have bean in continuous culture for years (Table 2). Asshown in Table 1, these cells may retain hepatocytic funclions on components such as hydmocortisone-inducibla tymosine Inansaminidasa activity (11, 33), paroxisomes (19),glycogan granules (11), and canaliculi (19).

Cell Culture. All calls ware well grown in sealed flasks(Falcon Plastics, Oxnard, Calif.) at 37°in Ham's F12 medium(Grand Island Biological Co., Grand Island, N. V.) with 10%fatal bovine serum (Grand Island Biological Co.) supplemented with streptomycin (100 @g/ml)and penicillin (100units/mi). Cultures ware refed twice weakly and kept at pH6.8 to 7.1 . For subcultuning, weekly monolayemswere dissociatad with 0.25% trypsin in GIBCO Solution A (GrandIsland Biological Co.). For growth in agamsuspension, cellswane suspended in F12 medium containing 0.4% agam(Grand Island Biological Co.) and 10% fetal bovine serum,following the method of MacPherson and Montagnier (21).

Tumorlgenicity Tests. Monolayer cultures ware hamvested by trypsinizahion, and 106 to 10@calls ware suspended in 0.1 ml of Dulbecco's phosphate-buffered saline.These ware injected s.c. or i.p. into each newborn Wistammatwithin 18 hr of birth . Cell lines were scored as tumonigenic ifa palpable nodule appeared at the site of injection andincreased in size thereafter (19).

Biochemical Assays. Cells (10kto 106)ware seeded in 10ml of the culture medium in the 25-sq cm flask. Unlessotherwise stated , biochemical assays were performed 1 to 3days after seeding when they were all in a logarithmicgrowing phase. The calls were washed in situ 2 or 3 limeswith Tyrode's balanced saline solution and ovenlayed with 2ml of an isotonic (300 ±10 mosM) incubation medium. Thestandard medium contained: 130 mM NaCI, 10 mM KCI, 2mM CaCI2, 1 mM MgCI@, 30 mM Tnis-acatate buffer (pH 7.2),and 10 @MATP (disodium salt; Sigma Chemical Co., St.Louis, Mo.). Details on the conditions of modified assaysystems are given in the legends to the charts and tables.For enzyme inhibition experiments, cultured cells werepreincubated for appropriate periods in the culture mediumcontaining inhibitors.

For the isotopic ATPasa assays, the reaction was started

at 37°by adding [‘y-32PJATP(tetraammonium salts; NewEngland Nuclear, Boston, Mass.) and was usually terminated at 10 mm. Following incubation, a 100-pi aliquot ofthe incubation medium and 2 ml of 1.5% molybdic acid in0.5 M H2S04 were vigorously emulsified with 2 ml of banzane:isobutyl alcohol (1:1). Aliquots (100 pi) of the organicphase containing 32p1and the aqueous phase ([y-32P]ATP)were transferred to separate scintillation vials containing 10ml of Bray's solution (New England Nuclear), and the radioactivity was determined in a Nuclear Chicago Mark II scintillationsystem.

Cell number was determined with a Coulter electroniccounter following trypsinization of the cultures. The viabilityafter ATPasa assay was tested by trypan blue exclusion andwas greater than 95%. Trypsinizad cells were also assayedin suspension by the isotopic procedure.

For the colonimatnic assay of ATPasa activity, calls wanesuspended in 2.0 ml of the incubation medium. The reactionwas carried out by adding 2 @moIesof ATP at 37°and wasterminated by chilling followed by the addition of 200 @lof20% HCIO4.After a rapid cantnifugation, a 2.0-mI aliquot ofthe supemnatant was assayed for the presence of P1by amodified Fiske and SubbaRow procedure (22, 30). Substrata specificity of the enzymatic hydrolysis of ATP wasdetermined by substituting GTP, CTP, UTP, ADP, AMP, pnitrophenyl phosphate or /3-glycerophosphate (all sodiumsalts, Sigma) for the ATP. The colonimatnic method of analysis, however, required excessive amounts of ATP (morethan 0.5 mM) and cells (more than 5 x 10@)and was not usedroutinely.

Histochemistry and Electron Microscopy. Calls warefixed in situ for 5 to 60 mm in 2% glutaraldehyde (0.1 Msodium cacodylate-HCI buffer, pH 7.2, containing 4 mMCaCI@,)and washed in ice-cold 0.25 M sucrose for 3 hr. Theywere treated for 30 mm by a modification of the method ofWachstein and Meisel (34) for histochemical demonstrationof ATPase activity. The modified reaction medium contamed 100 mM NaCI, 5 m@KCI, 4 mM CaCl@,1 mM MgC@,40mM Tns-maleate buffer (pH 7.2), 1 mM ATP (disodium salt;Sigma), and 0.3 mM Pb(NO3)@.Following incubation, the

4492 CANCERRESEARCHVOL. 36

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

Call Membrane Activity and Tumorigenicity

Table 2Tumorigenicityof cultured liver cells in newborn Wistarrats and

cytochemical reaction of surface-located ATPase activity

were nontumoniganic, when 10@to 10@calls were injectedinto newborn Wistar rats either s.c. or i.p. A positive resultof tumoniganicity was obtained with s.c. transplantations of3 x 106 cells of RL34-Hll. Injections s.c. of 10@cells fromRLC C-i , RLC-TVI, HNT, and HTC produced palpable tumomsat the sites of injection within 10 days. The humorsgnaw continuously and, in many cases, revealed histological features resembling those of hepatocarcinomas (cf. Ref.

N@@Dc 19).

Cytogenatic analyses of normal liven call lines revealed— that a majority of cells had a diploid number of chromo

— somas (Table 3). RL-i demonstrated an epithelial appear

— anca in early cultures but had shifted toward fusiform or

ND bipolar appearance (cf. Refs. 11, 13, and 26) at the time of— cytochamical studies and tumoriganicity lasts. These sari

ally subculturad calls ware frozen and stored before enter+ ing a nondividing stage in culture. RL-33, RL-34, and RL-36

— were established as continuously growing lines with an

+ epithalial appearance. Although the plating efficiency was

often less than 1%, these epihhelial cells revealed a relativelyND high growth rate during exponential phases (Table 3). Fig. 1

+ represents a phase-contrast micrograph of apithalial calls

ND of the RL-34 line. Normal cells were extremely flat and+ adhesive to each other, as wall as to the plastic substratum.

+ As reported in other liver cell lines (3, 13, 15, 35, 37), they

+ had little or no locomotion and continued to divide even: whentheywereinclosecontactwitheachother.Their+ confluent cultures consisted of a monolayer of tightly

packed but nonovanlapping calls. All transformed cell lines+ studied here consisted of aneuploid cells and had charac

tanistics of apithelial- rather than fusiform-lype calls. The:: platingefficiencywasmorethan10%inRL34-ESAand+ RL34-Hll, while it reached 40% in RLC C-i , RLC-TVI, HTC,

+ . and HNT. Sparse cultures revealed irregularities in size and

+ shape of calls. However, they often grew into a monolayer

+ of lightly packed cells. There was a tendency for polygonal

+ calls to pile up in multiple layers, but it was not a consistent

+ feature of these transformed call lines. No definite comrela

lion could be established between the tumoniganicity andthe cell generation time or saturation density of these calls(Table 3).

Cytochemical Studies. Fig. 2 represents an electron micrograph of apithalial cells of the RL-34 line. In this andother normal call lines, the plasma membranes were oftenclosely apposad with intercellular adhesion zones. Microvilli were present at the apical surface of apithalial cells ofnormal cell lines but in fewer numbers when compared tothose of the tumoniganic lines. It was particularly significantin normal call lines that microfilamants ware frequentlyorganized into bundles or networks. Except for the distnibution of microfilaments, the cytoplasmic architectures ofnormal liver apithelial calls appeared to be similar to thoseof tumonigenic cell lines (cf. Refs. 11, 15, 19, 35, and 37).

When glutaraldahyda-fixad samples of RL-34 cells waneincubated in the medium containing ATP and CaCI2 withPb(N03)2, no staining reaction was seen anywhere in saclion with the electron microscope (Fig. 2). With the samecondition, RLC C-i cells (Fig. 3) showed a positive ATPasacytochamical reaction along intercellular boundaries andfree call surfaces with the light microscope. Electron micrographs of RLC C-i cells (Figs. 4 and 5) revealed that lead

Surface-bcated ATP

ass―Time of investiga

lionOct. 1972

Nov. 1974Feb. 1976

Feb. 1975Dec. 1975Feb. 1976

Nov. 1974Feb. 1976

Feb. 1976

Mar. 1975Mar. 1976

Oct. 1971Apr.1974

Apr. 1972Sept.1972Jan. 1973Mar. 1973Feb.1974Jan. 1975Mar. 1976

Jan. 1976

Feb. 1974Mar. 1974May 1974Jan. 1975Nov. 1975Mar. 1976

Feb. 1974Mar. 1976

Tumor mcidence―

0/10

0/110/4

0/380/50/15

0/il0/14

0/36

0/122/6

0/i 52/7

0/123/54/59/1012/149/i 2

il/li

4/5

0/52/75/83/42/3

10/12

19/223/4

Celllines

RL-1

RL-33

RL-34

RL-36

RL34-ESA

RL34-Hll

RLC

RLCC-i

RLC-TVI

HNT

HTC

a Number of animals with carcinoma/number of newborns inoculated.

b Cytochemical reaction: +, positive; —, negative.

C ND, not done.

samples were postfixed for 1 hr in s-collidina-HCI-bufferad0504, dehydrated through graded ethanol, and embeddedin Epon for sectioning. Thick (1 @m)sections ware treatedwith ammonium sulfide and examined with a phase-contrash microscope. Thin sections were either unstained ordouble stained with aqueous uranyl acetate and lead citrateand examined in a JEM-7A electron microscope.

Growth Properties of Cultured Liver Cells. Those livercell lines tested were defined as normal or transformed,according to their ability to grow in 0.4% agar suspension(Table 1). None of the normal liver apithelial cells grew insoft agar (31, 34, 39).

The results of tumorigenicity tests are listed in Table 2. All4 normal cell lines and a transformed call line, RL34-ESA,

DECEMBER1976 4493

RESULTS

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

Comparison of surface-located ATPase activity and growth properties in normal andtransformed liver celllinesCell

Saturagener-lionCell

typeChromo some no.ation

densi- ATPase activity (units'9time― ty―(x(hr) l0@) PreconfluenteConfluent―NormalRL-334222

79 0.6 ±0.2 0.2 ±0.1RL-34422854 <0.1<0.1RL-36423260 0.3 ±0.1<0.1TransformedeRL34-ESA65-7024

58 7.1 ±2.1 4.8 ±1.5RL34-Hll60-702360 12.1 ±4.3 5.8 ±1.1RLC

C-i58-6319 62 40.9 ±14.9 7.8 ±1.2RLC-TVI58-632372 14.6 ±3.5 5.1 ±1.9HNT57-611878 60.8 ±14.3 13.5 ±4.0HTC60-701965 42.6 ±17.0 5.6 ±2.0

S. Karasaki and T. Okigaki

Table 3

5 ml F12-1O% fetal calf serum were cultured in 25 sq-cm T flasks.b Cells were incubated in situ in the standard salt solution with 100 @MATP; nmoles 32P@

releasedfrom [y-32P]ATPper i0@cells per 10 mm ±S.D.C@ cells or less per flasks.

d 3 X i0@ or more per flasks.

C,Cells positive for growth in soft agar.

reaction products deposited on the extracellular side ofentire plasma membranes. In Fig. 4, the free surfaces ofepithalial calls showed numerous microvilli and were associated with the distinct reaction product. The intercellularboundaries of adjacent calls often ran undulating courses,and the patterns of lead precipitates followed exactly thecourses of folded lateral membranes. In Fig. 5, the basalsurface membrane along the substratum was also associated with distinct reaction products. In any one section, theamount of reaction product varied from cell to cell or regionto region , ranging from a near-complete absence hoa denseband of lead precipitates on the plasma membranes.

The cytochemical results given in Table 2 indicate that acorrelation exists between membrane-bound nucleosida Inphosphatasa activity of cultured liver calls and their tumoniganicity in newborn mats.

Biochemical Assays. Monolayer cultures of both normaland transformed call lines were incubated in a physiologically balanced salt solution containing [y-32PJATPin situ orfollowing trypsinization. The viability of calls was not impaired during a 20-mm period of incubation. The isotopicassay of ATP-splihting activity was found to be linear for atleast 10 mm and oven the range of i0@to 3 x 10@calls parassay vassal. Results of these experiments are presented inTable 3. The cells of transformed lines revealed a considerably high mateof ATPase activity: more than 40% of addedATP was hydrolyzed by 10@cells for 10 mm. Exponentiallygrowing calls exhibited an activity about 2 to 6 times that ofconfluent stationary cells. When dissociated by a trypsinlmaatment, the activity was unchanged in preconfluent cultunes, whereas it was increased in confluent culturesroughly to the same degree as the untreated growing cells.Both growing and stationary monolayer cultures of the dipbid call lines exhibited a very low degree of ATPasa activity,and trypsinization of those cultures did not increase it.

The release of 32@from ATP correlated with an increase inP1in the reaction medium as measured by the cobonimetmic

method. Virtually all (99.9% or more) of the 32p added wasrecovered in the reaction medium after incubation withtransformed or normal calls. Solubilization of the enzymeinto the medium was negligible during the incubation panod.

Some of the biochemical properties of surface-locatedATPase ware explored with the transformed RLC C-i lineand the normal call line, RL-34, as a comparison. Substrataspecificity was demonstrated utilizing the cobonimetnic procaduras. Other nucleosida phosphates and organic phosphate compounds wane substituted for ATP. The matesofhydrolysis of GTP, CTP, UTP, and ADP ware, respectively,80, 60, 30, and 5% of the rate of ATP hydrolysis. Neither RLCC-i nor RL-34 hydrolyzed AMP, /3-glycarophosphata, or pnitrophanyl phosphate at pH 7.2. Chart 1 shows the dependancy of the ATPasa activity of RLC C-i cells on cationiccompositions. With Na@and K@in the isotonic incubationmedium, the activity was very low. It required divalant cations, since the presence of 1 mM EDTA reduced the activityfurther by 50%. The addition of 1 mM Ca2@to the monovalent cahions increased the activity 5-fold. Mg2@did not appraciably stimulate ATP hydrolysis and was inhibitory to theCa2@activation at the concentration of 1 mM. Thus, Ca2@was the most effective cation for stimulating the hydrolysisof ATP. ATP hydrolysis was maximal at about 1 mM CaCI2,when NaCI and KCI were also present at physiological concentrations (Chart 2). One mM MgCI@suppressed enzymeactivation at Ca2@concentrations greater hhan 1 mM. Theeffect of increasing concentrations of CaCI2was dependenton NaCI and KCI, since ATP hydrolysis was significantlyreduced when monovalent cations were replaced by isoIonic sucrose (Chart 2). The pH optimum for the enzymeactivity extended over a range of pH 6 to 7.5, with 2 maximalpeaks at pH 6.0 and 7.2 (Chart 3).

At the optimal conditions of ionic composition and pH,ATP hydrolysis by 10@calls was maximal at about 1 mM ATP.The Lineweaver-Burk plots ware obtained with RLC C-i and

4494 CANCER RESEARCH VOL. 36

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

Ca!! Membrane Activity and Tumorigenicity

@: @Na,k',Ca@'.@ 4 /“ Na', K, Ca2M12'

@3 / .@.

@ 2 /• 7@.Na'.K .M@2'

°E' _

Incubation time minutes

Chart 1. Cation-dependent ATP hydrolysis by dissociated cells as a function of time. Cells (10@)of the RLC C-i line were incubated at 37°for varyingtimes in 1 ml of an isotonic salt solution with Ca2@(C), Ca2@and Mg2@(U),Mg'@ (V), or without divalent cations (0). The basic incubation mediumcontained10nmoles[y-@PJATP,30 mMTile-aceticacid buffer (pH7.2), 130mM NaCl, and 10 mM KCI. The concentration of divalent cations, whenpresent, was 1 mM. P1release into incubation medium was expressed asnmoles P1liberated per 10 cells. RL-34 cells were incubated with Ca2@andMg2@(0).

a‘5

:@ r-@ Na.k'.4 1:2 Y@ Na'. K'. Mg2'

:@3 @/ __________@ ,,,/‘d'@ Sucrose

@ ,—@-----—@ @—@-----@ ,

1 2 3 4

Chart 2. The effect of Ca2@on ATP hydrolysis. Trypsinized RLC C-i cells(10w)were incubated with varying concentrations of CaCI, (0 to 4 mM) in 1 mlof an isotonic medium. The medium contained 10 nmoles [y-32P]ATPand 30mM Tns-acetic acid buffer (pH 7.2) with 130 mM NaCI, 10 mM KCI, and 1 mp.iMgCI, (0); 130 mM NaCI and 10 m@.iKCI (•);or 250 mu sucrose (V). RL-34cells were similarly incubated in the 1st medium (0). Incubation was at 37°for 10 mm.

RLC-TVI when the reciprocals of the velocity of Ca2@-stimu

lated ATP hydrolysis ware plotted against reciprocal substrata concentrations (Chart 4). The K,,@calculated for ATPass was 0.29 mM for RLC C-i and 0.33 mM for RLC-TVI. Thecorresponding maximal velocities were 125 and 36 nmolesof P1liberated per 10 mm per i0@calls, respectively.

The direct effects of different agents on the ATPase achyity were tested with RLC C-i calls. La3', which was knownto be a specific antagonist of Ca2@in a number of biologicalsystems (14), was affective in suppressing 80% of ATP hydrolysis at the concentration of 0.2 mM. Ouabain, a cardiacglycosida which is a potent inhibitor of Na@-K@-ATPasa(8,22), did not inhibit it at the concentration of iO j.@M.There islittle reduction of enzyme activity in the incubation mediumcontaining 10 @MOncovin, 10 @Mcolchicine or 5 @Mcytochalasin B. The addition of dithiolhraitol up to 1 @Mdid notstimulate the activity.

RLC C-i cells fixed for 5 mm in 0.5% glutaraldahydebefore assay showed 40% of the ATPase activity of unfixedcells. After routine glutaraldehyda fixation for alechnon microscopy (2%, iS mm to 1 hr), 30% of the activity was

4 5 S 7 1 9

detachable. The Ca2@-shimulatedATP hydrolysis of Ihe fixedcells was linear for a 20-mm period and gradually reached amaximum of 30% hydrolysis of added ATP, within 1 hr(Chart 5). The reaction required divalent cations and wassuppressed by La3@but not by ouabain . The activity of fixedRL-34 cells was also low (Chart 5).

4C

E

.a

CE

pHChart 3. The effect of pH on Ca2@-dependentATP hydrolysis by RLC C-i

(C) and RL-34 (0) cells. The reaction mixture of 1 ml contained 130 mu NaCI,10 mu KCI, 1 mu CaCI,, and 10 nmoles [y-@P]ATP at different pH of 30 muTne-acetic acid buffer. Incubation was at 37°for 10 mm.

E

a

‘S

‘-.2

a.5

aSC

—@@ ,

1/K., 5 10 15 201 /( mM ATP)

Chart 4. Lineweaver-Burk plot of the rate of P@release from ATP by intactcells of RLC C-i (0) and RLC-TVI (V). Reaction mixtures contained 30 muTns-acetic acid buffer, pH 7.2, 130 mu NaCI, 10 mu KCI, 2 mu CaCI,, and 1mu MgCl, in a final volume of 2 ml. Absciasa, reciprocal of ATP concentration expressed as mu; ordinate, reciprocal by velocity expressed as nmolesPCliberated per 10@cells per 10 mm.

24

1:C@@ 1

10 20 40 60 80Incubatientime minutes

Chart5. Glutaraldehyde-resistantATPaseactivityof RLCC-i (•)and RL34 cells (0). The cultured cells were dissociated by trypsinization and fixedfor 30 mm in cacodylate-buffered 2% glutaraldehyde. Afterwashing, i0@cellswere incubated in suspension at 37°in 1 ml of the standard incubationmedium with 10 nmoles ATP.

mM CaCI2added

DECEMBER1976 4495

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

S. Karasaki and T. Okigaki

DISCUSSION

The primary purpose of this work is to investigate thesurface alterations of cultured liven apihhalial calls with refemance to their tumomigenic potential. The morphologicalcriteria often used for assessing neoplastic growth of fibroblast cultures such as piling up or cniss-crossing of cellsdue ho the loss of contact inhibition (4, 6, 31) cannot bereliably applied to epihhelial cultures (20, 27, 31, 35, 38, 39).At the subcallulam level, the increase of microvilli at the callsurface (19, 35, 39) and the loss of cytoplasmic microfilament bundles (15) have been observed in association withoncoganic conversion of liver epithalial calls, althoughthese are hardly detectable as constant features. The prasanh histochamical survey of an ATPasa reaction of culturedliver calls has revealed a surface membrane alterationunique to tumonigenic cell lines and absanh from diploid celllines that ware derived from normal rat livers. The nature ofthe nucleosida tniphosphatase localized to plasma membranes has been characterized by a biochemical method.Since a close correlation exists between the activity of asurface-located ATPasa and the tumoniganicity of cullumadcalls, the isotopic assay of [‘y-32P]ATPhydrolysis by intactcalls may provide a rapid and quantitative method forscreening oncogenic conversion in vitro of liver apithelialcalls.

The hydrolysis of added ATP by inlact calls has beandescribed for other normal and neoplashic cell types including human glia-like astrocytes in culture (1), Ehnlich ascitascells (29), human granubocytes (5), and the nucleated erythrocytes of vartebraha species except mammals (36). In contrash ho Ihe present results with liver epithalial call lines,Agran at a!. (1) have reported that human glia cell linesexhibit a decreased rate of surface-membrane ATPasa activity following neoplastic transformation in viva (gliomalines), as wall as in vitro (glia cells exposed to SV4O).Only alow or moderate activity of surface-membrane ATPasa hasbean detected either in normal or transformed fibroblasts(1). The biochemical analysaswith tumonigenic liver epihhe

hal calls have characterized a Ca2@-,Mg2@-dependenl ATPase on the basis of substrahe specificihies, activating cations, pH optima, sensitivities to inhibitors, and effects offixation. The specific characteristic of the ATPase, aspecially with the pattern of Mg2@inhibition of Ca2@activationand sensitivity ho La3@,is somewhat different from the onedescribed by Ronquist and Agmen (29) for the surface-bcated ATPasa on intact Ehnich ascites calls. The ectoenzyme activity demonstrated in transformed liver apithalialcalls appears to be unique for tissue hype or call form ratherIhan for a neoplastic slate par Se. In fact, morphology,behavior, and growth properties of cultured liver epithalialcalls(4,il, 13, 15, 19, 26, 35, 37-39) are quitediffenentfromthose of fibroblasts (1, 4, 6), astrocytes (1), and ascitas calls(9). Thus, the appearance of Iha ectoanzyma is a uniquephenomenon of cytodiffarentiation following a long-termcultivation of liver epithalial cells. The lumomiganic call populations amacleanly subject to selection processes in vitro.Calls expressing surface ATPasa activity may have a growthadvantage in vitro over other cells with Iowan activity thatware suppressed and eventually eliminated from the cubtune. Whether a correlation of Iha enzyme activity to mabig

nancy is applicable ho other tumor calls that arose in vivo isnot yet established. It should be noted, however, that thetopographic pattern of surface enzyme reaction in hhe cultuned liver calls is similar ho Ihosa described in primary andtransplantable hapatomas (9, 17, 18, 24).

It is well eslablished that ATP cannot penetrate theplasma membrane barmierof intact cells (1, 5, 10, 22). Manyphysiological and biochemical studies with human emythrocytes and other call types suggest that, except for Ihe spacializad surface regions of some differentiated cells (e.g.,glia cells), the enzyme sites active in ATP hydrolysis aremostly located at the inner leaflet of the plasma membrane(2, 14, 22). Thus, the negative ATPasa reaction with inlactcalls of normal liven cell lines reported in this study shouldnot be interpreted as indicating an absence of this enzymefrom the plasma membrane. Our cytochamical studies offrozen-thawed and formaldehyde-fixed samples of cultureddiploid liver cell lines have bean in agreement with those ofWilliams et a!. (37), who also showed positive reactions ofATPase activity at the plasma membranes as well as at thecytoplasmic sites. Since these preparative procedurescould preserve cytoplasmic ATPasa activity and permiltedpenetration of ATP into Ihe calls (8, 10), the intracellularsites revealed enzymatic reactions.

As assayed biochemically, growing calls of all tumonigenic lines tested show a rapid hydrolysis of ATP added to aphysiologically balanced salt solution. The ATP-splihting activity is due neither to leakage of enzyme from the cells intothe medium nor to penetration of added substrala into thecalls (1). The enzymatic activity of tumoniganic call lines isprobably different from that of the Na@-K@-dapandantATPass, since the latter was known to be completely suppressed by glutaraldehyde fixahion (8, 16, 22). Only a Ca2@-Mg2@-dapandenh ATPase could survive in glutaraldahyde tmaalment with a long fixation time (16). The presenthistochemical study has demonstrated that anzymic raaclion products of glularaldahyda-fixed cells are distributedrandomly on the external side of entire plasma membranes.The reaction pattern revealed by electron microscopic histochamishny is presumably related either ho the polarity ofmembrane enzyme constituents or the vectorial flow ofanzymic and-products in the plasma membrane (2, 12, 22).The ATP-splitting activity of Iransformad liver cells could beascribed to an exposure of enzyme proteins on the externalsurface that are cryptic in normal cells. Trypsinization ofnormal liver calls, however, was not enough to expose thecryptic sites. Some architectural changes of plasma mambranas associalad with naoplastic transformation may beresponsible for the altered enzyme topography. The avidance that the actoenzyme activity of growing cells is always many limes higher than in stationary calls indicatesthat there may be a mechanism for generating hhe alteredmembrane conformation, which appears to be inherent inthe process of membrane assembling or biogenesis following aberrant call growth.

ACKNOWLEDGMENTS

We thank Luciano Borsato for valuable technical help and Bernard Szirthfor assistance in photography. We also thank Shirley G. Quan and MarshaDavis who helped with preparations of cell cultures during the course of thiswork.

CANCERRESEARCHVOL. 364496

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

Cell Membrane Activity and Tumoriganicity

of Malignant Conversion in Vitro of Epithelial Cells Derived from NormalAdult Rat Liver. European J. Cancer, 12: 527-534, 1976.

20. Katsuta, H., and Takaoka, T. Carcinogenesis in Tissue Culture. XW.Malignant Transformation of Rat Liver Parenchymal Cells Treated with 4-Nitroquinoline 1-Oxide in Tissue Culture. J. NatI. Cancer Inst., 49: 1563-1576,1972.

21. MacPherson, I., and Montagnier, L. Agar Suspension Culture for theSelective Assay of Cells Transformed by Polyoma Virus. Virology, 23:291-294,1964.

22. Marchesi, V. T., and Palade, G. E. The Localization of Mg-Na-K-activatedAdenosineTnphosphataseon RedCellGhostMembranes.J. Cell Biol.,35: 385-404, 1967.

23. Moses, H. L., and Rosenthal, A. S. Pitfalls in the Use of Lead Ions forHistochemical Localization of Nucleoside Phosphatases. J. Histochem.Cytochem., 16: 530-539, 1968.

24. Novikoff, A. B., and Biempica, L. Cytochemical and Electron Microscopic Examination of Morris 5123 and Reuber 1435 Hepatomas afterSeveralYearsof Transplantation.GannMonograph,1: 65-87, 1966.

25. Okigaki,T. A ReproducibleTechniquefor RatLiverCulture.In: T.C.A.Manual. Rockville, Md.: Tissue Culture Association, Inc., 1976.

26. Okigaki, T., and Quan, S. G. Morphological Inconsistency in Liver Cellsin Cufture. Proc. Japan Aced., 50: 517-720, 1974.

27. Okigaki, T., and Tokes, Z. Characterization of Various Stages PrecedingRat Liver Epithelial Cell Transformation. In Vitro, 12: 296, 1976.

28. Oshiro,Y., Gerschenson,L.E.,andDiPaolo,J.A.CarcinomasfromRatLiver Cells Transformed Spontaneously in Culture. Cancer Res.,32: 877-897, 1972.

29. Ronquist, G., and Agren, G. K. A Mg2@-and Ca2@-stimulatedAdenosineTriphosphatase at the Outer Surface of Ehrich Ascites Tumor Cells.Cancer Res., 35: 1402-1406, 1975.

30. Rosenthal,A.S., Kregenow,F.M.,andMoses,H.L.SomeCharacteristics of a Ca2@-DependentATPase Activity Associated with a Group ofErythrocyte Membrane Proteins Which Form Fibrils. Biophys. Biochem.Acta, 196: 254-262, 1970.

31 . Sanford, K. K. Biologic Manifestation of Oncogenesis in Vitro: A Critique. J. NatI. Cancer Inst., 53: 1481-1485, 1974.

32. Sato,J., Namba,M.,Usui,K.,andNagano,D.CarcinogenesisinTissueCulture. VIII. Spontaneous Malignant Transformation of Rat Liver Cellsin Long-TermCulture.Japan.J. Exptl. Med.,38: 105-118,1968.

33. Thompson,E.B.,Tomkins,G.M.,andCurran,J. F.InductionofTyrosine-a-keto Glutarate Transaminase by Steroid Hormones in a NewlyEstablished Tissue Culture Line. Proc. NatI. Aced. Sci. U. S., 56: 296-303, 1966.

34. Wachstein,M.,andMeisel,E.Histochemistryof HepaticPhosphatasesat a Physiologic pH. With Special Reference to the Demonstration of BileCanaliculi. Am. J. Clin. Pathol., 27: 13-23, 1957.

35. Weinstein,I.B.,Orenstein,J.M.,Gebert,R.,Kaighn,M.E.,andStadles,U. C. Growth and Structural Properties of Epithelial Cell Culture Established from Normal Rat Liver and Chemically Induced Hepatomas. Cancer Res., 35: 253-263, 1975.

36. Wenkstern,T. W., andEngelhardt,W. A. Diean der OberfláchederKernhaltiger roten BlutkOrperchen lokalisierte Adenosmnpolyphosphatase. Folia Haematol., 76: 422-431 , 1959.

37. Williams,G.M.,Stromberg,K.,andKroes,R.CytochemicalandUltrastructural Alterations Associated with Confluent Growth in Cell Culturesof Epithelial-like Cells from Rat Liver. Lab. Invest., 29: 293-303, 1973.

38. Williams,G. M., Elliott,J. M.,andWeisburger,J. H. CarcinomaafterMalignant Conversion in Vitro of Epithelial-like Cells from Rat Liverfollowing Exposure to Chemical Carcinogens. Cancer Res. ,33: 606-612,1970.

39. Yamaguchi, N. , and Weinstein, I. B. Temperature-Sensitive Mutants ofChemicallyTransformed Epithelial Cells. Proc. NatI. Acad. Sci. U. S., 72:214—218,1975.

Fig. 1. Phase-contrast micrograph of a confluent monolayer of the normal rat liver cell line, RL-34 cells, indicating a fiat and regular appearance of cellularmorphology. x 400.

Fig. 2. Electron micrograph of RL-34 cells treated for the histochemical demonstration of ATPase activity. Thin section is parallel to the monolayer plane.No reactionIs detectable.Microfilaments(F)andmicrotubules(1)areabundantin the cytoplasmjust beneaththe plasmamembraneattachedto thesubstrate. Organized bundles of microfilaments appear toextend from the cell surface toward the nuclei. The plasma membranes of adjacent cells are closelyapposed with intercellular adhesion zones (A). x 10,000.

Fig. 3. Phase-contrast micrograph of RLC C-i cells, treated for the cytochemical demonstration of ATPase activity. RLC C-i cells are in a confluent regionof culture. Arrows, positive reactions of ATPase activity along the intracellular boundaries and free surfaces. x 1000.

Fig. 4. Electron micrograph of a thin section of RLC C-i cells, adjacent to that of Fig. 3. The entire surfaces of plasma membranes deposited opaquereaction products of ATPase activity (arrow). Many microvilli (MV) are protruding at the free surfaces. x 5,500.

Fig. 5. A thin section of RLC C-i cells which is perpendicular to the plastic substratum (S). Opaque reaction products of ATPase activity are unevenlydeposited on the external surface of plasma membranes as well as in the narrow intercellular space of lateral membranes (arrows). x 20,000.

DECEMBER1976 4497

REFERENCES

1. Agren, G., Pontén,J., Ronquist, G., and Westermak, B. Demonstrationof an ATPase at the Cell Surface of Intact Normal and Neoplastic HumanCellsin Culture.J. CellularPhysiol.,78:171-176,1971.

2. Benedetti, E. L., and Emmelot, P. Structure and Function of PlasmaMembranes Isolated from Liver. In: A. J. Dalton and F. Haguenau (eds.),The Membranes, pp. 33-120. New York: Academic Press, Inc., 1968.

3. Borek, C. Neoplastic Transformation in Vitro of a Clone of Adult LiverEpithelial Cells into Differentiated Hepatoma-like Cells under Conditionof Nutritional Stress. Proc. NatI. Aced. Sci. U. S., 69: 956-959, 1972.

4. Borek, C., Grob, M., and Burger, M. Surface Alterations in TransformedEpithelial and Fibroblastic Cells in Culture: A Disturbance of MembraneDegradation Versus Biosynthesis? Exptl. Cell Res., 77: 207-215, 1973.

5. DePierre, J. W., and Karnovsky, M. L. Edo-Enzyme of Granulocytes: 5'-Nucleotidase.Science,183:1096-1098,1974.

6. DiPaolo, J. A. Susceptibility of Mammalian Cells in Vitro to NeoplasticTransformation by Chemical Carcinogens. In: L. E. Gerschenson and E.B. Thompson (eds.), Gene Expression and Carcinogenesis in CulturedLiver, pp. 402-411 . New York: Academic Press, Inc., 1975.

7. Emmelot, P., and Bos, C. J. Studieson Plasma Membranes. IX. A Surveyof Enzyme Activities Displayed by Plasma Membranes Isolated fromNormal and Preneoplastic Livers and Primary and Transplanted Hepatomas of the Rat. Intern. J. Cancer, 4: 705-722, 1969.

8. Ernst, S. A., and Philpott, C. W. Preservation of Na-K-activated and Mgactivated Adenosine Triphosphatase Activities of Avian Salt Gland andTeleost Gill with Formaldehyde as Fixative. J. Histochem. Cytochem., 18:251-263, 1970.

9. Essner, E., and Reuber, M. D. Localization of Acid Phosphatase Activityin Well-Differentiated Hepatocellular Carcinoma 146. J. NatI. CancerInst., 47: 25-33,1971.

10. Essner, E., Rogh, J., and Fabrizio, P. Localization of MitochondrialAdenosine Triphosphatase Activity in Cultured Human Cells. J. Histochem. Cytochem., 13: 647-656, 1975.

11. Gerschenson, L. E., Okigaki, T., Andersson, M., Molson, J., and Davidson, M. B. Fine Structural and Growth Characteristics of Cultured RatLiver Cells. Insulin Effects. Exptl. Cell Res., 71: 49-58, 1972.

12. Goldfischer, S., Essner, E., and Novikoff, A. B. The Localization ofPhosphatase Activities at the Level of Ultrastructure. J. Histochem. Cytochem., 12: 72-95, 1964.

13. Grisham, J. W., Thai, S. B., and Nagel, A. Cellular Derivation of Continuously Cuftured Epithelial Cells from Normal Rat Liver. In: L. E. Gerschenson and E. B. Thompson (eds.), Gene Expression and CarcmnogenesisinCulturedLiver,pp. 1-23. NewYork: AcademicPress,Inc., 1975.

14. Haksar, A., Maudsley, D. v., Péron,F. G., and Bedigian, E. Lanthanum:Inhibition of ACTH-stimulated Cyclic AMP and Corticosterone Synthesisin Isolated Rat Adrenocortical Cells. J. Cell Biol., 68: 142-153, 1976.

15. lype, P. T., Allen, T. D., and Pillinger, D. J. Certain Aspects of ChemicalCarcmnogenesisin Vitro Using Adult Rat Liver Cells. In: L. E. Gerschenson and E. B. Thompson (eds.), Gene Expression and Carcinogenesis inCultured Liver, pp. 425-440. New York: Academic Press, Inc., 1975.

16. Jacobsen, N. 0., and Jorgensen, P. L. A Quantitative Biochemical andHistochemical Study of the Leed Method for Localization of AdenosineTriphosphate-Hydrolyzing Enzymes. J. Histochem. Cytochem., 17: 443-453,1969.

17. Karasaki, S. Subcellular Localization of Surface Adenosine Tnphosphatase Activity In Preneoplastic Liver Parenchyma. Cancer Res., 32: 1703-1712, 1972.

18. Karasaki, S. Ultrastructural and Cytochemical Studies on Hyperbasophilic Foci with Special Reference to the Demonstration of Cell SurfaceAlterations in Hepatocarcinogenesis. Cancer Res., 36: 2567-2572, 1976.

19. Karasaki, S., Simard, A., and de Lamirande, G. Ultrastructural Analysis

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

S. Karasaki and T. Okigaki

:::@.@@ , -

@ 1@c_@@@ :@.@ :?4@-@L@@

,,.? ,@

@ @. , - .)_.@ ‘.

-@@@ . .@@

@@ , @.@ . ‘ .:@

J .(, ,@.,@ -

;..:@.‘,.@ ::@. @-

@ .. .@

@ @. .@ ._@..*_t. .@-@@ -. ; £‘

•.@

@@@ ,.,;@

;@ @, ‘ ,/ .:@..‘t

@; ),, \L_,:

4

F

, —@ ,*@‘@ &:.@@@ .-@ .@ @•:@ @‘;@ ..@,‘@

.1 :@@ :@4@ :@@.,.

, @:‘;@@@ c@I@F)CANCERRESEARCHVOL. 364498

@ @.-

@ ‘.,,.,.@ &. S.

a ‘ r,'@ •@@ .

@ :@“@ ‘@

.@...@@

‘,a@@i.,itW@ ... ..;;.i,@ .

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

,.@. .@ -@ ;@ ._,s4.@ ., .@ 44@.,* :-@ .@@ @‘@ I@@

@@ . @[@

..@@ ,@

.@@@@ ,@..@@ _@ .-. ..@-A

@ 4-@@@i . ,@@(2Y ‘@

-,@ -@ ‘@

. -@

@.

. a.. -‘

.4

*@. .@

@;

@.

.,@••.@.‘.. - ,.—,@

4499

@,.@

@@ . .. a , •‘•@@•@•

.@@ . . .- S.@@@@@@

.@1.@ @1

@ . ,,@..

.@ â€-̃4., :@,I, • @.

I_@°@

S.,.

S

DECEMBER1976

—@@ : .@@@ 4@ ‘:@•.@@@@@ @‘.@@@@@@ ,@.@?@k@@ •@-@;_@@@@ I@

... ,@. i@@

,@

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from

1976;36:4491-4499. Cancer Res   Shuichi Karasaki and Tohru Okigaki  Tumorigenicity of Cultured Liver Epithelial CellsSurface Membrane Nucleoside Triphosphatase Activity and

  Updated version

  http://cancerres.aacrjournals.org/content/36/12/4491

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

  .pubs@aacr.orgDepartment 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/36/12/4491To request permission to re-use all or part of this article, use this link

Research. on August 11, 2019. © 1976 American Association for Cancercancerres.aacrjournals.org Downloaded from