[CANCER RESEARCH 42. 563-568, February 1982]0008-5472/82/0042-OOOOS02.00

Induction of Alkaline Phosphatase Activity in Cultured Human

Intracranial Tumor Cells

Nobuhiko Takahara,1 Fritz Herz,2 Robert M. Singer, Asao Mirano, and Leopold G. Koss

Department of Pathology and Division of Neuropathology, Montefiore Hospital and Medical Center, Albert Einstein College of Medicine, Bronx, New York 10467[N. T., F. H., A. H., L. G. K.], and Department of Anatomy, Fair/eigh Dickenson University School of Dentistry, Hackensack, New Jersey 07601 [R. M. SJ

ABSTRACT

Alkaline phosphatase activity in several cultured primaryhuman intracranial tumor cells varied over a relatively widerange, and there was no correlation between specific activityand the type of tumor from which the cultures were derived.The enzyme was thermolabile, and its activity was stronglyinhibited by l-bromotetramisole, levamisole, and L-homoargi-nine but not by L-phenylalanine and L-phenylalanylglycylgly-cine. These are the characteristics of the liver-bone-kidney

form of alkaline phosphatase. Prednisolone induced increasedlevels of enzyme activity in most cultures, and sodium butyrateacted as an inducer in cultures of pituitary adenoma andhemangioblastoma cells. The increase was most pronouncedwhen responsive cells were exposed to both stimuli simultaneously. The induced alkaline phosphatase had the sameproperties as the enzyme of cells grown in the absence ofinducers. Increased alkaline phosphatase activity was not induced by osmolality changes of the culture medium; this feature appears to be characteristic of cells producing the liver-bone-kidney enzyme form.

INTRODUCTION

In humans, at least 3 different forms of alkaline phosphatase[orthophosphoric monoester phosphohydrolase (alkaline optimum) (EC 3.1.3.1 )] have been recognized. They are the term-placental, intestinal, and liver-bone-kidney forms (32, 34, 38).They can be distinguished from each other by several parameters (13). It has been known for some time that the heat-stable, term-placental type is produced ectopically by a varietyof human tumors and is usually referred to as "Regan" isoen-

zyme (11 ). The production of heat-labile, non-Regan isoen-

zymes by human tumors has been described recently (7, 18).In cultured cancer cells, the 3 forms of alkaline phosphatasehave been demonstrated (3, 24, 43). However, studies onenzyme regulation have been restricted mainly to cells producing the term-placental form (4, 5, 15, 22, 23, 35, 36).

In this report, we present evidence that, irrespective of theirhistogenetic derivation, cells of primary intracranial tumorsproduce a thermolabile enzyme which has the characteristicsof the liver-bone-kidney form of alkaline phosphatase. In addition, we will show that some of these cells respond withincreased levels of enzyme activity to inducers of term-placen

tal alkaline phosphatase such as glucocorticoids (5, 22, 25)and sodium butyrate (4, 17). By contrast, activity is not influ-

1 Present address: Department of Neurosurgery, Kansai Medical University.

Fumizono-cho, Moriguchi City, Osaka 570, Japan.2 To whom requests for reprints should be addressed, at Montefiore Hospital

and Medical Center, 111 East 210th Street, Bronx, N. Y. 10467.3 The abbreviations used are: CMS, central nervous system; prednisolone,

11ß,17,21 -trihydroxypregna-1,4-diene-3,20-dione.

Received May 26, 1981 ; accepted October 27, 1981.

enced by osmolality changes of the culture medium whichcontrol the term-placental and intestinal enzymes (22, 25, 36).This is the first report documenting the modulation of the liver-bone-kidney form of alkaline phosphatase in cultured human

brain tumor cells.

MATERIALS AND METHODS

As described previously (26), aliquots of surgically removed primarybenign and malignant tumors of the CMS3 were used to initiate the

cultures. The benign tumors were 7 meningiomas, 2 pituitary adenomas, and one hemangioblastoma. The malignant tumors were 5 glio-

blastomas, 3 astrocytomas of low grade, 2 astrocytomas of high grade,one ependymoma and one pineal region tumor. There were also 5metastatic tumors to the brain. Of these, one was from breast, one frommelanoma, one from bone, and 2 from lung. The tumor tissues wereplaced in Earle's balanced salt solution and immediately taken to the

laboratory. Following removal of blood and the electrocoagulated portion, the tumor tissue was minced with small curved scissors and stirredat 25°with 0.1 % trypsin in a Ca2+- and Mg2+-free Earle's salt solution

until most of the material was dissociated into single cells and smallfragments which were decanted and washed with complete culturemedium (26). Following centrifugation at 800 x g, the cells wereresuspended in medium and inoculated into 25- and/or 75-sq cm

plastic culture flasks. Minimum essential medium supplemented with10% fetal bovine serum, penicillin (100 units/ml), streptomycin (100/ig/ml), and amphotericin B (0.25 fig/ml) was used throughout. Thecultures were incubated at 37°in a humidified atmosphere of 5% CO2

in air. The medium was changed 3 times a week. None of the culturesdisplayed the morphological arrangements typical of normal fibro-

blasts. Cells were subcultured as they reached confluency (generallyafter 7 to 10 days) using trypsin (0.05%)-EDTA (0.02%).

Where indicated, the osmolality of the medium was increased from284 to 384 mosmol/kg by the addition of 50 mw NaCI (from anautoclaved 3 M stock solution) 24 hr after cell transfer (22). A stocksolution of 100 ¡igof prednisolone per ml was prepared in ethanol, anda final concentration of 0.5 /ig per ml (1.4 JIM)was added to the cultures24 hr after cell transfer; ethanol (0.5%) was added to the respectivecontrols. To parallel cultures, a final 2 mw concentration of sodiumbutyrate was added also 24 hr after cell transfer. In other experiments,the cultures were supplemented with various combinations of NaCI,prednisolone, and butyrate. With primary cultures, the agent(s) wasadded 1 day after culture initiation. Cells growing in regular mediumserved as controls. To study the effect of the removal of prednisolone,after 24 hr and every 24 hr thereafter, the steroid-containing medium

was removed and the cultures were washed twice with and thencovered with regular medium. Other cultures growing in steroid-sup

plemented medium were similarly treated, except that they werewashed with and refed prednisolone-containing medium. Cells growing

in regular medium throughout the experiment were also included. Cellsfrom experimental and control cultures were harvested at 24-hr inter

vals.Cultures for enzyme assays were washed 3 times with 15 to 20 ml

of cold 0.15 M NaCI and lysed with 0.5 ml of 0.25% sodium deoxycho-late (17). To optimize the recovery of lysates, the 75-sq cm flasks were

centrifugea for 2 min at 50 x g in an IEC Model UV centrifuge with a

FEBRUARY 1982 563

on April 9, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

N. Takahara et al.

240 rotor modified with string loops so as to accommodate the cultureflasks. If not immediately assayed, the lysates were stored in liquid

nitrogen.Preparation of Cell-free Extracts from Original Tumor Tissues.

Aliquots of the original tissue were washed with and minced in a buffersolution containing 130 rriM NaCI, 30 mw Tris-HCI, and 1 mw MgCI,.(pH 7.4), followed by disruption for 5 to 10 min with an ice water-cooled Raytheon 10-kc ultrasonic oscillator (24).

Enzyme Assays. Alkaline phosphatase activity was determined at37° by the hydrolysis of p-nitrophenyl phosphate (24) in 0.4 M 2-

amino-2-methyl-1 -propanol-HCI buffer (pH 10.6). Acid phosphatase

activity was measured with the same substrate at pH 4.8 using 0.1 Macetate buffer (21 ). Enzyme reactions were stopped with 0.25 N NaOH.The variation of duplicate assays was less than 10%. Activity wasexpressed in units, 1 unit being defined as the amount of enzyme thathydrolyzes 1 fimol of substrate per min. Specific activity was expressedin units per mg of protein, the latter determined according to themethod of Lowry ef al. (31 ) using crystalline bovine serum albumin asstandard.

Partial Purification of Alkaline Phosphatase. Cell lysates werediluted with 0.25 volume of NaCI (130 mM)-Tris (30 mM)-MgCI2 (1 rnw)buffer solution (pH 7.4) and extracted for 15 min with 1 volume of 1-

butanol (24). After centrifugation at 1500 x g for 10 min, the butanollayer was discarded and the aqueous layer was used either immediatelyor frozen in liquid nitrogen. Sonicates of tumor tissues were similarlyextracted. Butanol extraction yielded a 10- to 15-fold purification of

alkaline phosphatase activity (24).Thermal Inactivation. Thermostability of alkaline phosphatase was

investigated by incubating duplicate aliquots of 0.05 ml of cell lysatesor of butanol-extracted preparations at 56°with 0.1 ml of 1 M 2-amino-

2-methyl-1 -propanol-HCI buffer (pH 10.6). After incubation for variouslengths of time, tubes were transferred to 4°and the remaining activitywas subsequently measured at 37°by the addition of 0.1 ml of 0.016

M p-nitrophenyl phosphate containing 2 mw MgCI2 (24). Enzyme re

actions were stopped with 0.25 N NaOH. Thermal inactivation was alsocarried out at pH 7.4 using the buffer solution indicated above. The pHof the buffers was monitored during and after preincubation, and nosignificant changes were noted. The percentage of residual activitywas computed from controls kept in the respective buffers at 4°.

Inhibition Studies. For these experiments, butanol-extracted en

zyme preparations were used. Extracts from cultured cancer cells ofcervical (C4I) and colonie (HT-29) origin, which produce, respectively,the term-placental (23) and the intestinal (25) alkaline phosphatase

forms, were included for comparison. The following specific inhibitorswere tested (ranges of final concentrations in parentheses): L-phenyl-alanine (1 to 20 mw), t-leucylglycylglycine (1 to 20 mw); L-phenylala-nylglycylglycine (1 to 20 mw); L-homoarginine (1 to 20 mw); levamisole(0.02 to 2 mw); l-bromotetramisole (0.005 to 1 mw); and Na3VO4 (0.05

to 5 mw). Where necessary, enzyme preparations were diluted so thatless than 3% of the substrate was hydrolyzed during the test. Totriplicate 0.05 ml of enzyme samples, 0.1 ml of inhibitor and 0.2 ml ofsubstrate-buffer-MgCb mixture were added. 0.1 ml of noninhibitory Disomer or inactive ion was added to controls. After incubation at 37°,

the reaction was stopped with 0.25 N NaOH and the percentage of

residual activity was calculated from controls. The results were expressed as the concentration of inhibitor required to produce 50%inhibition, (I50), obtained from graphs depicting the reciprocal of thepercentage of residual activity versus the concentrations of inhibitor(13). With all inhibitors tested, there was a linear relationship betweenthe reciprocal of the reaction velocity and inhibitor concentration.

Electrophoretic and Immunological Characterization. In order tocharacterize the enzyme further, polyacrylamide gel electrophoresiswas done by the method of Fishman (10), using 7% gels prepared with0.5% Triton X-100. For the immunological characterization, antiserumto purified human term-piacental alkaline phosphatase (gift from Dr. G.J. Doellgast, Bowman Gray School of Medicine, Winston-Salem, N. C.)

was added to the specimens prior to electrophoresis. The retardation

of migration or the diminution of activity in an isoenzyme band wasused as a criterion for positive cross-reactivity (39). Following electro

phoresis, enzyme activity was visualized by staining the gels accordingto Angelus ef al. (1) with o-naphthyl phosphate as substrate and 4-

aminodiphenylamine diazonium salt (Dajac Laboratories, Philadelphia,Pa.) as azo dye in 1 M propanediol buffer (pH 9.7). For inhibitionstudies, 20 rriM i.-phenylalanine was incorporated into the reactionmixture, using o-phenylalanine in the controls (39). In order to verifythat the antibody was active, specimens of term-placental alkaline

phosphatase of similar activity were run concurrently; retardation ofmigration was seen in each instance.

RESULTS

Alkaline phosphatase activity of cultured intracranial tumorcells varied over a relatively wide range, and there was nocorrelation with the type of tumor from which the cultures werederived. The specific activity of cultures derived from 7 differentmeningiomas was 4.2 x 10~4, 6.1 x 1CT4, 2.0 x 10~3,3.1 x 10~3, 5.4 x 10~3, 1.1 x 10"2, and 7.8 x 1CT2, andthat of cultures derived from 5 glioblastomas was 9.4 x 10"3,1.0 X 1Q-2, 2.4 x 1Q-2, 5.5 x 10~2, and 7.8 X 10'2. The

specific activity of the cultures was similar to that of the originaltumor specimens tested. For example, the activity of a glio-blastoma was 3.1 x 10~2 and that of its primary culture 5.5x 10~2. After the third transfer, the activity of the cultured cellswas 3.6 x 10~2. Alkaline phosphatase activity in an astrocy-toma was 2.0 x 10~3 and that of its fourth and eighth subcultures 1.0x10"3 and 1.4x10~3, respectively. Repeatedly

subcultured cells, obtained from the same patient, had essentially the same specific activity. The results with cultures initiated from the metastatic tumors were as variable as thoseobtained with cultures derived from primary tumors. Specificactivity of the cultured cells of metastatic tumors was alsosimilar to that of the original specimens (Table 1). The relativelylow enzyme activity of some cultures cannot be ascribed to adifferential effect of the harvesting procedure used, sinceequivalent results were seen when the cell monolayers weredispersed with a rubber policeman. The possibility of alkalinephosphatase leakage into the culture medium was considered,but upon testing used media, none was detected. Acid phosphatase measurements were included in this study because itsspecific activity does not vary significantly from one type of

Table 1

Alkaline phosphatase activity in cultured cells derived from tumors of the brainand from tumors metastasizing to the brain

Preparation of tumor extracts and initiation of cultures are as indicated in thetext. Cells were grown for 7 days after each subculture.

Specific activity (units/mgprotein)NonculturedT2T3T4T5T6T7T8a

Identificationcode."T, transfer number.78-1

78a (ma

lignantastro-cytomaMx10-3)2.01.51.01.20.91.21.479-30(glio-blastoma)(xIO'2)3.15.54.43.65.079-54(lungcancer

metastasis)(x10~3)7.17.05.5

564 CANCER RESEARCH VOL. 42

on April 9, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Alkaline Phosphatase in Cultured Brain Tumor Cells

cultured cell to another (22) and it is not influenced by stimuliaffecting alkaline phosphatase (36). Acid phosphatase activityof all cultures was within the same range, varying between 1.5x 10~2 and 3.7x10~2. These values were similar to those of

other cultured human tumor cells (22).Increased levels of alkaline phosphatase activity are induced

in certain human tumor cell lines by increasing the osmolalityof the culture medium (22, 36) or by exposure to glucocorti-

coids (5, 35) or sodium butyrate (4,17). Induction experimentswere conducted on cultures derived from the primary intracra-

nial tumors. Primary cultures and cells transferred up to 20times were used. Each experiment with cultured cells wasrepeated at least twice. The experimental conditions used werethose found to be optimal for enzyme induction (4, 22, 35). Aresponse was considered positive only when the specific activity in experimental cultures was 2.5 times higher than in controls. Of the stimuli tested, hyperosmolality (384 mosmol/kg)did not increase alkaline phosphatase activity in any of thecultures examined. By contrast, prednisolone (1.4 ¡IM)elicitedinduction in cultures derived from pituitary adenomas, glioblas-

tomas, and malignant astrocytomas. Enzyme induction wasalso seen in primary cultures of glioblastomas (Table 2). Sodium butyrate (2 mw) induced increased activity only in culturesof pituitary adenoma and hemangioblastoma cells. Inducibilityby either agent was independent of the base level activity.When prednisolone and sodium butyrate were added simultaneously, the levels of activity were higher than those obtainedwith each agent individually, and in some instances the effectwas synergistic (Table 2). The alkaline phosphatase of cellswhich did not respond to each agent alone was also refractoryto their combination. With respect to the cultures derived frommétastases to the brain, only the enzyme of the metastaticbreast and lung tumors was inducible by prednisolone; sodiumbutyrate had no effect on any of these cultures. The addition ofthe stimuli to intact cells attached to their growth surface andcovered with balanced salt solution (which does not supportcell growth) or to lysates did not affect enzyme activity. Thepossibility that prednisolone or sodium butyrate promoted theformation of enzyme-activating factors was ruled out by mixingexperiments of control and experimental cell lysates. The effectof the stimuli was discernible within 48 hr and reached itsmaximum by 72 hr. Removal of inducer from the medium ofgrowing cells resulted in lower specific activity (Chart 1). Hy

perosmolality, prednisolone, sodium butyrate, and their combinations did not affect acid phosphatase activity (results notshown).

Thermostability is a very sensitive parameter for distinguishing isoenzymes of alkaline phosphatase (13, 21, 24). Thus,preincubation of the term-placental enzyme at 56°in the pres

ence of 0.67 M 2-amino-2-methyl-1 -propanol-HCI buffer (pH

10.6) caused a relatively small loss of activity. By contrast, theenzyme of brain tumors and of the cultures derived therefromwas rapidly inactivated; more than 98% of the activity was lost

26

N Op

bXî18>

I 14o

I}0w

24 48 72 96 120 144

Hours of Qrowth(after last cell transfer)

Chart 1. Effect of prednisolone and its removal on alkaline phosphataseactivity. Glioblastoma (79-115) cells, at the fifth passage, were dispersed asindicated in the text and inoculated into regular medium. At 24 hr (upward arrow)the cells were rinsed and covered with medium containing 0.5 /aj prednisoloneper ml. At 48, 72, and 96 hr, cultures in prednisolone-containing medium wereeither rinsed with and refed regular medium (downward arrows) or washed withand refed steroid-containing medium. Duplicate cultures of each set were harvested at 24-hr intervals with 0.25% sodium deoxycholate (17). Cell lysates werestored in liquid nitrogen until the end of the experiment. Alkaline phosphataseactivity was determined as indicated. •¿�,activity of cells growing in regularmedium at time of harvest; O, activity of cells growing in prednisolone-containingmedium at time of harvest. Specific activity is expressed in units/mg protein.

Table 2

Effects of va

Cells were inoculated into regular medium. NaCI, prednisolone, and sodium butyrate were added 24 hr later. Cells were grown for 7 days.77-73" (T9)" (pituitary

adenoma)Additions

tomediumNone

NaCI (50 (TIM)Prednisolone (0.5 /ig/ml)Sodium butyrate (2 rriM)NaCI + prednisoloneNaCI + sodium butyratePrednisolone + sodium butyrateNaCI + prednisolone + sodium

butyrateSpecific

activity(x10-')5.3

2.8108.340.538.936.3

389.8197.1IR0.5

20.47.67.36.8

73.537.278-1

60 (T4) (hemangioblastoma)Specific

activity (x

10-4)4.9

5.420.1

103.719.6

162.1319.8330.7IR1.14.1

21.24.0

33.165.367.579-1

15 (T0)c (glioblas- 77-131 (T10) (malig-

toma) nantastrocytoma)Specific

activity (x10^2)6.1

10.782.314.275.914.258.462.8IR1.8

13.52.3

12.42.39.6

10.3Specific

activity (x10"4)3.7

4.014.47.1

11.68.0

22.336.3IR1.1

3.91.93.12.26.09.879-55

(T2) (metastaticlungcancer)Specific

activity (x10~3>3.1

3.127.2

6.222.9

6.349.349.8IR1.0

8.72.07.42.0

15.816.0

Identification code.T, transfer number; IR, activity ratio of experimental to control cultures.Primary culture.

FEBRUARY 1982 565

on April 9, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

N. Takahara et al.

after 2 min. At pH 7.4, the inactivation was more gradual, yetafter 5 min only 7% of the activity remained (Table 3). Theresults were the same for the enzyme of control cultures andof inducer-treated cultures with increased specific activity. Thealkaline phosphatase of the cultured cells derived from métastases to the brain was equally thermolabile.

The thermolabile alkaline phosphatase was further characterized with specific inhibitors that readily distinguish the various enzyme forms (3, 13, 37). For control purpose, preparations of term-placental and intestinal enzymes were included inthe tests. An example of the plots (13) used to determine theinhibitor concentration required to produce 50% inhibition isshown in Chart 2. The enzyme of the original CMS tumorspecimens tested and of the cultured cells grown with orwithout inducer(s) was strongly inhibited by l-bromotetramisole,levamisole, and L-homoarginine. It was unaffected by L-phe-nylalanine, u-phenylalanylglycylglycine, and L-leucylglycylgly-

cine (Table 4). By comparing these inhibition patterns withthose of the various human alkaline phosphatase forms (3, 13,37), it is evident that the CNS tumor cells produced the liver-bone-kidney enzyme. Na3VO4, an alkaline phosphatase inhibitor (30) not heretofore used for this purpose, also distinguishedthis enzyme form from the thermolabile, intestinal type. The

Table 3

Thermostability of alkaline phosphatase

Enzyme preparations from cultured human brain tumor cells grown with orwithout inducer were preincubated at 56°at pH 10.6 and pH 7.4 as indicated inthe text. Specific activity varied between 2.8 x 10"" and 8.2 x 10~'. Term-

placental alkaline phosphatase was included for comparison. The percentage ofresidual activity was computed from controls kept at 4°.Data at pH 7.4 represent

the average of 15 determinations.

% of residual enzyme activity

Time at 56°

(min)123457.510Term-placentalpH10.6NO"98.8NDND89.6ND77.5pH7.4100100100100100100100Brain

tumorcellspH

10.6ND<2<2ND<2<2<2pH7.451321910732

enzyme produced by the cultured cells derived from brainmétastaseswas also of the liver-bone-kidney type.

Some cancer cell lines which normally produce a heat-labile

alkaline phosphatase respond to glucocorticoids by synthesizing the term-placental isoenzyme (20, 40). In order to rule out

the possibility that brain tumor cells exposed to prednisoloneproduced otherwise undetectably small amounts of this enzymeform, preparations from control and from steroid-treated cultures were pretreated with antiserum to human term-placental

alkaline phosphatase and then subjected to polyacrylamide gelelectrophoresis (39). As shown in Fig. 1, the enzymes fromboth sources had identical isoenzyme profiles and their migration was not retarded by the antiserum. These findings demonstrated the absence of term-placental enzyme and provided

additional evidence for the identity of the base level and the¡nducer-mediated increased alkaline phosphatase activity in

cultured primary intracranial tumor cells. It is of interest to notethe presence in the preparations of a weak, slow-moving, heat-stable isoenzyme (Fig. 1). This enzyme which did not cross-

Table 4

Concentrations of inhibitors producing 50% inhibition of alkaline phosphataseactivity

Assays were carried out in triplicate using p-nitrophenyl phosphate as substrate in 2-amino-2-methyl-1-propanol-HCI buffer (pH 10.6). At least 5 concentrations of each inhibitor were used. Three or more determinations were made oneach specimen at each inhibitor concentration, and the average values (% ofactivity remaining) were used in subsequent calculations.

Iso" (mM)

Inhibitor

Cultured Term-pla-brain tumor cental en- Intestinal

cells zyme enzyme

L-PhenylalanineL-PhenylalanylglycylglycineL-LeucylglycylglycineL-HomoarginineLevamisolel-BromotetramisoleNa3VO4»50.0»50.0»50.03.70.090.010.224.50.56.0»40.02.00.70.143.014.025.0»40.011.03.00.8

'' Iso. concentration of inhibitor producing 50% inhibition of alkaline phospha

tase activity, obtained from graphs depicting 100/the percentage of activityremaining versus inhibitor concentrations (13).

3 ND, not done.

12

: 10

- 8

0.1 0.2 1.00.5L-Bromotetr»mi to I •¿�mM )

Chart 2. Inhibition of alkaline phosphatase by l-bromotetramisole. The effectof l-bromotetramisole on enzyme activity was determined as indicated in the text.Abscissa, final concentrations of inhibitor. O, alkaline phosphatase of brain tumorcells; •¿�,term-placental enzyme; Q. intestinal enzyme.

1234

B5 6

Fig. 1. Isoenzyme profile of alkaline phosphatase of cultured brain tumor cellsgrown with and without prednisolone. Cells derived from a glioblastoma (79-115)were grown for 7 days in regular medium (A) and in prednisolone-containingmedium (B). Specific activities were 3.0 x 10~2 and 22 x 10~2, respectively.

Before electrophoresis, aliquots of each specimen were heated for 5 min at 65°

(pH 7.0) and mixed with antiserum to placental alkaline phosphatase. Nonheatedaliquots were similarly mixed. Conditions of electrophoresis were as described(39), and visualization of activity was as indicated in the text. Gel 1, untreatedsamples; Gel 2, heated samples; Gel 3, p-phenylalanine (20 mM) in the reactionmixture; Gel 4, L-phenylalanine (20 mM) in the reaction mixture; Gel 5, unheatedsamples treated with antiserum to term-placental alkaline phosphatase; Gel 6,heated samples treated with antiserum.

566 CANCER RESEARCH VOL. 42

on April 9, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Alkaline Phosphatase in Cultured Brain Tumor Cells

react with the antiserum to term-placental alkaline phosphataseand was not sensitive to L-phenylalanine inhibition may repre

sent an as yet uncharacterized isoenzyme.

DISCUSSION

Alkaline phosphatase activity has heretofore not been investigated in cultured primary intracranial tumor cells. As theforegoing results indicate, there were quantitative differencesamong cultures initiated from aliquota of the same tumor typesobtained from different individuals. These differences wereobserved in cultures of benign and malignant tumors, and theywere in keeping with the results of earlier histochemical andbiochemical investigations (8, 9, 12, 41, 42) on a variety ofCMS tumors. It is of interest to note that in one histochemicalstudy no alkaline phosphatase activity was detected in glio-

blastomas and astrocytomas (12). In our studies, there wereno significant enzyme activity differences between the originaltumor and the cultured cells derived therefrom and activity wasindependent of the number of transfers; cells derived from thesame patient had essentially the same specific activity uponrepeated subculture. These findings differ from the changes inalkaline phosphatase activity observed in other systems inwhich there was no correlation between the specific activity ofthe original tissue and of the cultured cells initiated from thesame (45). It should be noted that the results obtained withcultures of tumor métastasesto the brain were similar to thoseof cultured primary CNS tumor cells. It would be of considerableinterest to ascertain whether brain métastasesof primary tumors producing term-placental alkaline phosphatase and cul

tured cells derived therefrom would preserve the expression ofthis enzyme form.

The observation that the enzyme of primary intracranialtumor tissue and of tumor métastasesto the brain as well as ofthe cultures derived therefrom was thermolabile and not inhibited by L-phenylalanine is in keeping with previous findings

with meningiomas (9, 42) and craniopharyngiomas (41). Byapplying a variety of specific inhibitors, we have now established that the enzyme represents the liver-bone-kidney form

of alkaline phosphatase. The presence of this enzyme form inhuman brain has been described recently (14).

The liver-bone-kidney type of alkaline phosphatase has beenfound in some cultured cells of human origin, including fibro-blasts (44), Chang liver cells (19), and KMK-2, a continuous

line derived from a gastric carcinoma (43). However, in contrastto the enzyme of responsive CNS tumor cells, the liver-bone-

kidney activity of these cells is not inducible (35, 36, 43). Ofthe inducers tested, prednisolone elicited increased activitylevels in most cultures obtained from benign and malignanttumors of the CNS. The effect of the steroid was also observedin primary cultures, and the responsiveness of the cells persisted upon repeated subculture. This is of special interestsince most studies on the control of alkaline phosphataseactivity have been conducted on continuous cell lines that hadbeen transferred many times (4, 5, 16, 22, 23, 36). Sodiumbutyrate, a substance known to affect a variety of other cellularprocesses (4, 29), was uniformly effective only with culturesfrom pituitary adenomas and hemangioblastomas. The reasonfor this selectivity is not clear. The lack of response by culturedintracranial tumor cells to hyperosmolality was in keeping withour previous findings that only term-placental (21, 22, 36) and

intestinal (25) alkaline phosphatases are inducible by this factor.

The mechanisms responsible for the control of alkaline phosphatase in cultured human tumor cells have not been elucidated. Studies with HeLa cells suggest that the hormonallyinduced increase in activity is due to the synthesis of a modifiermolecule that interacts with the enzyme to produce an alkalinephosphatase with enhanced catalytic efficiency (2). The modeof action of sodium butyrate appears to depend on the type ofcells used. Thus, DNA synthesis does not seem to be necessaryfor enzyme induction in HeLa cells, whereas in choriocarci-

noma cells it seems to require new synthesis of DNA, RNA,and protein (4). However, because alkaline phosphatase is acell membrane-associated enzyme (28), it is possible that the

increase in specific activity does not necessarily represent aneffect on the enzyme and its synthesis per se but that thestimuli may elicit alterations in the enzyme microenvironmentleading to increased activity. The finding that dimethyl-DL-2,3-distearoyloxypropyl-2'-hydroxyethylammonium ace

tate, a phospholipase A? inhibitor, causes alkaline phosphatasestimulation in HeLa S3G (33) and the suggestion that 5-iodo-2'-deoxyuridine, another inducer, acts by affecting the cell

membrane (16) would support such an interpretation (22).The functional significance of alkaline phosphatase activity

in brain and CNS tumors is not known. It has been suggestedthat in brain it may have a role in the activation and inactivationof enzymes through dephosphorylation (14). Alkaline phosphatase may be involved in the metabolism of pyridoxal phosphate,a cofactor of glutamate decarboylase and glutamate transami-

nase (14). Pyridoxal phosphate is a substrate for the purifiedsheep brain enzyme (6). The membrane location of alkalinephosphatase could be of significance in the design of chemo-

therapeutic agents. For example, it has been suggested thatphosphorylated derivatives of cytotoxic agents might be usefulin the treatment of 6-thiopurine-resistant acute leukemia (27).

ACKNOWLEDGMENTS

We wish to thank Alexander Schermar and Elissa M. Leahy for expert technicalassistance and Pearl Parsowith for her help in preparation of the manuscript

REFERENCES

1. Angelus, D.. Inglis, N. R., and Fishman. W H. Isoelectric focusing of alkalinephosphatase isoenzymes in polyacrilamide gels. Am. J. Clin. Pathol., 66.929-934. 1976.

2. Bazzell, K. L.. Price, G., Tu, S.. Griffin, M., Cox, R., and Ghosh. N Cortisolmodification of HeLa 65 alkaline phosphatase Decreased phosphate content of the induced enzyme. Eur. J. Biochem.. 6f 493-499, 1976.

3. Benham, F. J., and Harris, H. Human cell lines expressing intestinal alkalinephosphatase. Proc. Nati. Acad. Sei. U. S. A., 76. 4016-4019, 1979.

4. Chou, J. Y.: Regulation of the induction of alkaline phosphatase in chorio-carcinoma cells by sodium butyrate. In Vitro (Rockville). 75 789-795, 1979.

5. Cox, R. P.. and MacLeod. C M. Hormonal induction of alkaline phosphatasein human cells in tissue culture. Nature (Lond.). /90 85-87. 1961.

6. Dorai, D. T., and Bachhawat, B. K. Purification and properties of brainalkaline phosphatase. J. Neurochem., 29 503-511. 1977.

7. Ehrmeyer, S. L., Joiner. B. L., Kahan, L.. Larson, F. C., and Metzenberg, R.L. A cancer-associated, fast, homoarginine-sensitive electrophoretic form ofserum alkaline phosphatase. Cancer Res.. 36. 599-601, 1978.

8. Feigin. I., and Wolf, A. The alkaline phosphomonoesterase activities of braintumors. Arch. Pathol., 67. 670-678. 1959.

9. Fischer, W.. and Müller,E. Untersuchungen überdie alkalische Phosphatasein Meningiomen. Enzymol. Biol. Clin.. 1Ì:450-458, 1970.

10. Fishman, L. Acrylamide disc gel electrophoresis of alkaline phosphatase ofhuman tissues, serum and ascites fluid using Triton X-100 in the sample andgel matrix. Biochem. Med., 9. 309-315. 1974.

11. Fishman. W. H., Inglis. N. R., Stolbach, L., and Krant, M. J. A serum alkaline

FEBRUARY 1982 567

on April 9, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

N. Takahara et al.

phosphatase ¡soenzyme of human neoplastia cell origin. Cancer Res.. 28:150-154, 1968.

12. Gluszcz, A. A histochemical study of some hydrolytic enzymes in tumours ofthe nervous system. Acta Neuropathol., 3. 184-201, 1963.

13. Goldstein, D. J., and Harris, H. Human placental alkaline phosphatase differsfrom that of other species. Nature (Lond.), 280. 602-605. 1979.

14. Goldstein. D. J., and Harris, H. Mammalian brain alkaline phosphatase:expression of liver/bone/kidney locus. Comparison of fetal and adult activities. J. Neurochem., 36. 53-57, 1981.

15. Goz. B.: The induction of alkaline phosphatase in HeLa cells by 5-iodo-2'-deoxyuridine. Cancer Res., 34: 2393-2398, 1974.

16. Goz, B., Orr, C., and Wharton, W. Inhibition by deoxycytidine, cytidine, and/j-cytosine arabinoside of the induction of alkaline phosphatase activity inHeLa cells. J. Nati. Cancer Inst., 64: 1355-1361, 1980.

17. Griffin. M. J.. Price, G. H.. Bazzell, K. L., Cox, R. P., and Ghosh, N. K. Astudy of adenosine 3':5'-cyclic monophosphate, sodium butyrate and corti-

sol as inducers of HeLa alkaline phosphatase. Arch. Biochem. Biophys.,r 64. 612-623. 1974.

18. Hada, T.. Higashino, K., Yamamoto, H., Okochi, T., and Yamamura, Y.Phenotypic changes of alkaline phosphatase in association with the novely-glutamyl transpeptidase in renal cell carcinomas. In: F-G. Lehmann (ed.),Carcino-Embryonic Proteins, Chemistry, Biology, Clinical Applications, Vol.2, pp. 689-692. Amsterdam: Elsevier/North Holland, 1979.

19. Hamilton. T. A., Tin, A. W., and Sussman, H. H. Alkaline phosphataseisoenzyme expression in Chang liver cells. Exp. Cell Res., 722. 31-38,

1979.20. Hanford, W., and Fishman, W. H. Induction of term placental alkaline

phosphatase in cultured human cells by prednisolone or butyrate. Proc. Am.Assoc. Cancer Res., 21: 41, 1980.

21. Herz, F. Alkaline phosphatase in KB cells: influence of hyperosmolality andprednisolone on enzyme activity and thermostability. Arch. Biochem. Biophys., 758. 225-235, 1973.

22. Herz, F., and Koss. L. G. Alkaline phosphatase activity in cultured urinarybladder cancer cells. Arch. Biochem. Biophys., 794. 30-36, 1979.

23. Herz, F., Miller, O. J., Miller, D. A.. Auersperg, N., and Koss, L. G. Chromosome analysis and alkaline phosphatase of C4I, a cell line of humancervical origin distinct from HeLa. Cancer Res., 37: 3209-3213, 1977.

24. Herz, F., and Nitowsky, H. M. Alkaline phosphatase activity of human cellcultures: kinetic and physical-chemical properties. Arch. Biochem. Biophys.,96:506-515. 1962.

25. Herz, F., Schermer, A., Halwer. M., and Bogart, L. H. Alkaline phosphatasein HT-29, a human colon cancer cell line: influence of sodium butyrate andhyperosmolality. Arch. Biochem. Biophys., 270: 581-591, 1981.

26. Kajikawa. H.. Kawamoto, K., Herz, F., Wolley, R. C., Hirano, A., and Koss,L. G. Flow-through cytometry of meningiomas and cultured meningiomacells. Acta Neuropathol., 44: 183-187, 1978.

27. Lee. S. H., Shansky, C. W.. and Sartorelli, A. C. Evidence for the externallocation of alkaline phosphatase activity on the surface of Sarcoma 180cells resistant to 6-thioguanine. Biochem. Pharmacol., 29: 1859-1861,1980.

28. Lin, C-W.. Sasaki, M., Orcutt, M. L., Miyayama, H., and Singer, R. M. Plasma

membrane localization of alkaline phosphatase in HeLa cells. J. Histochem.Cytochem., 24: 659-667, 1976.

29. Littlefield, B. A., Cidlowski, N. B., and Cidlowski. J. A. Modulation ofglucocorticoid effects and steroid receptor binding in butyrate-treated HeLaS3 cells. Arch. Biochem. Biophys., 207. 174-184, 1980.

30. Lopez, V., Stevens, T., and Lindquist, R. N. Vanadium ion inhibition ofalkaline phosphatase-catalyzed phosphate ester hydrolysis. Arch. Biochem.Biophys., 775. 31-38. 1976.

31. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall. R. J. Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem.. 793. 265-275,1951.

32. McKenna, M. J., Hamilton, T. A., and Sussman, H. H. Comparison of humanalkaline phosphatase isoenzymes: structural evidence for three proteinclasses. Biochem. J., 787: 67-73, 1979.

33. Melnykovych, G., and Lopez, I. C. Rapid cell surface-related stimulation ofalkaline phosphatase in HeLa cells by dimethyl-DL-2,3-distearoylpropyl-2'-hydroxyethylammonium acetate (Rosenthal's inhibitor). J. Cell. Physiol., 92.

91-96, 1977.34. Mulivor, R. A., Hannig, V. L., and Harris, H. Developmental change in human

intestinal alkaline phosphatase. Proc. Nati. Acad. Sei. U. S. A.. 75: 3909-3912, 1978.

35. Nitowsky, H. M., and Herz, F. Hormonal regulation of alkaline phosphatasein dispersed cell cultures. Biochem. Biophys. Res. Commun., 77. 261-266,1963.

36. Nitowsky, H. M., Herz, F., and Geller, S. Induction of alkaline phosphatasein dispersed cell cultures by changes in osmolarity. Biochem. Biophys. Res.Commun., 72: 293-299, 1963.

37. Sakiyama, T., Robinson, J. C., and Chou, J. Y. Characterization of alkalinephosphatases from human first trimester placentas. J. Biol. Chem., 254:935-938, 1979.

38. Seargent, L. E., and Stinson, R. A. Evidence that three structural genescode for human alkaline phosphatase. Nature (Lond.), 287:152-154, 1979.

39. Singer, R. M., and Herz, F. Multiple effects of glucocorticoids and hyperosmolality on isoenzyme expression in cultured KB cells. J. Nati. CancerInst., 60: 573-577, 1978.

40. Singh, I., Tsiangh, K. Y., and Blakemore, W. S. Placenta-like alkalinephosphatases from human osteosarcoma cells. Cancer Res.. 38: 193-198,1978.

41. Timperley, W. R., Turner, P., and Davies, S. Alkaline phosphatase in cran-iopharyngiomas. J. Pathol., 703: 257-262, 1971.

42. Timperley, W. R., and Warnes, T. W. Alkaline phosphatase in meningiomas.Cancer (Phila.), 26: 100-103, 1970.

43. Tokumitsu, S-i., Tokumitsu, K., Kohnoe, K., and Takeuchi, T. Characterization of liver-type alkaline phosphatase from human gastric carcinoma cells(KMK-2) in vitro. Cancer Res., 39: 4732-4738. 1979.

44. Vanneuville. F. J., Van Eisen, A. F., and Leroy, J. G. Alkaline phosphatase:presence and behaviour in human diploid fibroblasts. Arch. Int. Physiol.Biochim., 83. 1017-1018, 1975.

45. Wilson, P. D. Reversible and irreversible effects of tissue culture on enzymepatterns of spontaneous mouse tumors and mouse and human embryotissues. Cancer Res., 33. 375-379, 1973.

568 CANCER RESEARCH VOL. 42

on April 9, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1982;42:563-568. Cancer Res   Nobuhiko Takahara, Fritz Herz, Robert M. Singer, et al.   Intracranial Tumor CellsInduction of Alkaline Phosphatase Activity in Cultured Human

  Updated version

  http://cancerres.aacrjournals.org/content/42/2/563

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/42/2/563To request permission to re-use all or part of this article, use this link

on April 9, 2021. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from