alterations in the activity and isozymic profile of …...foundation, la jolla, ca. heparinized...

[CANCER RESEARCH 45, 2993-3001, July 1985]

Alterations in the Activity and Isozymic Profile of Human Phosphofructokinaseduring Malignant Transformation in Vivo and in Vitro: Transformation- andProgression-linked Discriminants of Malignancy1

Shobhana Vora,2 James P. Halper, and Daniel M. Knowles3

Departments of Pediatrics [S. V.], and Internal Medicine [J. P. H.], and Pathology [D. M. K.], Columbia University, College of Physicians and Surgeons, New York,New York 10032, and Department ot Basic and Clinical Research, Scripps Clinic and Research Foundation, La Jolla, California [S. V.]

ABSTRACT

6-Phosphofructokinase (PFK) plays a central role in the regu

lation of glycolysis in both normal and neoplastia cells. SincePFK also mediates the Pasteur effect, it coordinates the twomodes of energy production in most cell systems, i.e., glycolysisand respiration. The energy production in the cancer cell ischaracterized by a predominance of aerobic glycolysis (the Warburg effect) and a diminution or lack of the Pasteur effect.

Previous studies from this laboratory have demonstrated thatPFK in humans and in the rat exists in multiple tetramericisozymic forms consisting of three unique subunits under separate genetic controls, M, L, and P types. These isozymes aredistinguishable from one another by ion-exchange chromatog-raphy and subunit-specific antibodies. Various organs exhibit

unique isozyme distribution patterns which essentially reflect thepreferred mode of carbohydrate metabolism utilized, i.e., glycolysis or gluconeogenesis or both.

In order to investigate whether the high aerobic glycolysis ofthe cancer cell can be explained on the basis of a lack of theregulatory function of PFK due to an altered isozyme distributionpattern, we compared the activity and isozymic profile of theenzyme from malignant cells of human leukemias, lymphomas,virus-transformed cell lines, and established malignant cell linesof lymphoid, myeloid, erythroid, and fibroblastic origin and theirnormal counterparts. The myeloid and erythroid cell lines werealso investigated after in vitro differentiation induced by dimethylsulfoxide, sodium butyrate, hemin, etc. Our results show that,as is the case with hexokinase and pyruvate kinase, the othertwo rate-limiting enzymes of glycolysis, PFK shows both quan

titative increases and isozymic alterations secondary to alteredgene expression during neoplastic transformation, both in vivoand in vitro. In contradistinction to the isozymic alteration inhexokinase and pyruvate kinase, where highly regulated liver-

type isozymes decrease or disappear and are replaced by thenonregulated ones, in the case of PFK, the highly regulated liver-

type isozyme not only persists but actually increases, followedby an increase in the platelet-type isozyme. These isozymic

alterations closely parallel the quantitative increases in total PFKactivity, which in turn is closely related to the rate of replicationof cancer cells and hence an increase in metabolism. Thus,

1This work was supported in part by Grants AM 33445 and EY 03357 from the

NIH and by grants from the Cancer Research Institute and the New York StateAIDS Institute. This is Publication 3760-BCR from the Research Institute of Scripps

Clinic and Research Foundation.2 Recipient of a Research Career Development Award (K04 AM01260) from the

NIH. To whom requests for reprints should be addressed, at Scripps Clinic andResearch Foundation, 10666 North Torrey Pines Road, BCR-7, La Jolla, CA 92037.

'Present address: New York University School of Medicine, Department of

Pathology, 550 First Avenue, New York, NY 10016.Received 12/3/84; revised 4/9/85; accepted 4/11/85.

human PFK is both a transformation- and a progression-linked

discriminant of malignancy (For definitions of these terms, seeWeber ef al., N. Engl. J. Med., 296: 486-493, 1977.). Thesedata, taken together with those reported in the literature, suggestthat altered regulatory properties of tumor PFKs secondary toisozymic alteration may be partly responsible for the Warburgeffect and a weakening of the Pasteur effect in the cancer cell.

INTRODUCTION

In mammals, PFK4 (ATP:D-fructose-6-phosphate-1-phospho-

transferase, EC 2.7.1.11), one of the key regulatory enzymes ofglycolysis, is a tetrameric protein, consisting of 4 identical ornonidentical subunits. The enzyme is regulated by several metabolites, hormones, and nutritional and developmental stateswhich alter the rates of glycolysis in accord with cellular need forenergy or glycolytic intermediates. Thus, PFK plays a critical rolein the energy metabolism of cells/organs largely or entirely dependent on glycolysis, i.e., mature erythrocytes, exercising muscle, brain under most conditions, ischemie heart, and neoplasticcells. Previous studies from this laboratory have shown that thehuman enzyme is under the control of 3 structural loci, whichcode of M-, L-, and P-type subunits which are differentially

expressed by various organs (1, 2). These subunits undergorandom tetramerization to produce various homo- and hetero-

tetrameric isozymes which are distinguishable from one anotherby ion-exchange chromatography and subunit-specific antibod

ies (1,3).The carbohydrate metabolism of the cancer cell is character

ized by an increase in the rate of glycolysis and a decrease inthe rate of gluconeogenesis (for review, see Refs. 4 to 7). Thismetabolic state is associated with and therefore presumablyachieved by the reciprocal alterations in the key regulatoryenzymes of these 2 opposing catabolic and synthetic pathways(5, 6). The increase in glycolysis is associated with quantitativeand qualitative alterations in the key regulatory enzymes, i.e.,PFK, HK (EC 2.7.1.1 ), and PK (EC 2.7.1.40) (8).

Previous studies of the experimental rodent tumors haveshown that isozymic alterations in PFK differ from those exhibitedby HK and PK. Although the non-muscle and non-liver PFKisozyme(s) which are poorly regulated appear anew, the liver-

type PFK, a highly regulated isozyme, not only persists butactually increases significantly (for review, see Refs. 6 and 9 to11). In human neoplasia, PK and HK exhibit alterations analogous

4The abbreviations used are: PFK, phosphofructokinase; M, P, and L, muscle,

platelet, and liver type, respectively; PK, pyruvate kinase; HK, hexokinase; PHA,phytohemagglutinin; FCS, fetal calf serum; DMSO, dimethyl sulfoxide; TPA, tetra-decanoylphorbol-13-acetate; PBS, phosphate-buffered saline, pH 7.4; Slg, surfaceimmunoglobulin.

CANCER RESEARCH VOL. 45 JULY 1985

2993

Research. on December 14, 2020. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

PFK ISOZYMES IN HUMAN CANCERS AND MALIGNANT CELL LINES

to those present in rat tumors (12-18). In contrast, very limited

information is available with regard to alterations in PFK activityand isozymes in human cancers (19-22).

In order to define such changes for human PFK, we investigated malignant cells from various types of leukemia, lymphoma,and virus-transformed and established malignant cell lines for

their PFK contents as well as their isozymic profiles. Our resultsdemonstrate that PFK exhibits both quantitative increases andisozymic shifts secondary to altered gene expression in neoplasia, i.e., retention of the liver-type isozyme and increase in theexpression of both liver and platelet-type isozymes. The isozymic

alterations closely parallel the quantitative increases in total PFKactivity, which in turn is closely related to the rate of replicationand therefore metabolism of cancer cells. Thus, in accord withthe prediction of the "molecular correlation concept," human

PFK(s)/is/both/a /transformation- /and/progression-linkeddiscriminant5 of malignancy (5,6). The identification of the P- and

L-isozymes of PFK as essential to the survival of cancer cells

may provide sensitive targets for the design of selective cancerchemotherapy.

MATERIALS AND METHODS

Chemicals and Reagents. Adenme nucleotides, NADH, fructose 6-phosphate, fructose 1,6-diphosphate, and dithiothreitol were purchasedfrom Sigma Chemical Co., St. Louis, MO. Aldolase, Â«-glycerol-phosphate

dehydrogenase, triosephosphate isomerase, and other auxiliary enzymeswere purchased from Boehringer Mannheim Biochemicals, Indianapolis,IN. DEAE-Sephadex A-25 came from Pharmacia, Piscataway, NJ, andNonidet was from Particle Data, Inc., Chicago, IL. RPMI 1640, Eagle's

minimal essential medium, PCS, and hemin were products of GrandIsland Biological Co., Grand Island, NY. Staphylococci bearing protein A(IgGSorb) were obtained from The Enzyme Center, Inc., Boston, MA.Isolymph, a Ficoll-Hypaque density gradient medium, was obtained fromGallard-Schlesinger Chemical Co., Carle Place, NY. Sodium butyrate,

TPA, and DMSO came from Sigma. All other chemicals were of reagentgrade.

Normal and Malignant Human Cells and Established Cell Lines.Cases with leukemia and lymphoma were randomly selected for PFKstudies from the patient populations attending the Hematology Servicesof the Departments of Pediatrics and Internal Medicine, Columbia University, College of Physicians and Surgeons, New York, NY, and Department of Basic and Clinical Research, Scripps Clinic and ResearchFoundation, La Jolla, CA. Heparinized fresh bone marrow aspirates orperipheral blood samples (10 to 100 ml) were obtained from the newlydiagnosed patients with acute lymphoblastic or myelogenous leukemiaafter informed consent. Peripheral blood specimens were also obtainedfrom patients suffering from chronic lymphocytic or myelogenous leukemia who were not on any chemotherapy for at least 1 month prior toblood sampling. Fresh biopsy specimens of lymphoma were obtained atthe initial diagnostic work up of newly diagnosed patients.

Heparinized and defibrinated peripheral blood samples (50 to 100 ml)were freshly obtained from healthy normal volunteers to obtain normalgranulocyte and lymphocyte suspensions to serve as controls. Normalthymic and tonsiller tissue samples were obtained at surgery. The human

5The qualifying terms "transformation-linked" and "progression-linked" are used

as defined by Weber et al. (5) when an enzymic alteration is linked to malignanttransformation per se, i.e., without any relation to the degree of malignancy,differentiation, or growth rate, and thus discriminates between normal and neo-plastic cells, it is termed "transformation-linked." Such an alteration will be present

in all cancers of all grades of malignancy. In contrast, when an enzymic alterationshows a positive or negative correlation with the degree of malignancy of a giventype of cancer, it is termed "progression-linked," i.e., the higher the degree of

malignancy, the more profound will be the alteration.

malignant cell lines, HeLa, KB, K-562, HL-60, and KG-1, were obtained

through the courtesy of various investigators at Columbia University,College of Physicians and Surgeons. Epstein-Barr virus-transformed

lymphoblastoid cell lines (GM 605, GM 558, and GM 3201) and culturedT-cell leukemia cell lines (GM 3639 and MOLT-4) were obtained from the

Human Mutant Genetic Cell Repository, Camden, NJ. At least 3 or moreseparate donors were investigated where possible. At least 3 observations were made on each of the established malignant cell lines.

Cell Cultures. The HeLa and KB cells were grown in large tissueculture flasks (Falcon; 75 sq cm) containing 20 ml of Eagle's minimalessential medium with Earle's salts supplemented with 1% each of L-

glutamine, penicillin, streptomycin, and nonessential amino acids and10% FCS. All other cell lines were grown in plastic flasks containingRPMI 1640, supplemented with 1% each of L-glutamine, penicillin andstreptomycin and with 15% FCS. The flasks were incubated at 37Â°Cin

5% CO2 and 95% air. The cells were transferred when cultures becameconfluent or the cell count exceeded 1 x 106/ml density by dilution with

fresh culture medium. The K-562 cell line was induced with hemin (100

JIM) and sodium butyrate (1 mw) as described previously (23), whereasHL-60 and KG-1 cell lines were induced with DMSO (1.25%) and TPA

(150 nM), respectively. The cell suspensions were induced at a densityof 3 to 5 x 104/ml, and the exposure of various inducers was maintained

for 7 and 11 days without any change of medium. At the time of harvest,the cultures were compared with concurrent noninduced control cultures.Generally, the induced cultures showed a mild decrease in cell replicationand viability as compared to the control cultures. For suspension cultures,the viable cells were counted by the trypan blue dye exclusion technique.

Preparation of Pure Cell Suspensions and Cell Extracts. Mononu-clear cells were isolated from freshly obtained heparinized blood, fromaspirated bone marrow, and from biopsy tissues (after teasing the latterin RPMI 1640) by Ficoll-Hypaque density gradient centrifugaton usingIsolymph according to the manufacturer's instructions. If necessary,

mononuclear cell suspensions were subjected to repeated centrifugationsteps at 600 x g to remove platelets, which do not sediment at thiscentrifugal force. Mononuclear cells from normal individuals were furtherseparated into pure populations of T- and B-lymphocytes as follows. T-cells were separated over Ficoll-Hypaque according to their capacity toform rosettes with neuraminidase-treated sheep erythrocytes and were

quantified by resetting of these erythrocytes as described by Hoffmanand Kunkel (24). Monocytes were removed from the mononuclear cellsuspensions by allowing them to adhere to plastic flasks for 1 to 2 h inRPM11640 plus 10% FCS at 37Â°Cor by a carbonyl iron (Superfine; GAF

Corp, New York, NY) method as described previously (25). B-cellsbearing surface IgM were isolated from T-cell and monocyte-depleted

peripheral blood mononuclear cell suspensions according to their capacity to form rosettes with bovine erythrocytes coated with purified anti-

IgM antibodies covalently linked by the CrCI3 method (26).Immature myeloid elements (myeloblasts) were separated from either

peripheral blood or bone marrow using Isolymph. Mature polymorpho-nuclear leukocytes were separated from defibrinated blood according tothe method described by Clark and Kimball (27). Total leukocytes wereprepared from defibrinated blood from patients with chronic myelogenousleukemia by dextran sedimentation. The contaminating platelets wereremoved by a density gradient centrifugation step using 20.6% Stractanaccording to Corash ef al. (28).

Harvested cultured cells were washed 3 times with PBS, and final cellsuspensions were counted using the standard manual method; viabilitywas determined using the trypan blue dye exclusion technique. Thedegree of contamination of final cell suspensions with other cell typeswas monitored by morphological examination of the smears.

The cell extracts were prepared just prior to PFK assays and otherstudies as described previously (1).

PFK Assays. PFK assays were performed using a Gilford Model 260spectrophotometer at 26Â°C in a final volume of 1 ml as described

previously (2). One unit of enzyme is defined as that amount of enzymethat converts 1 ^mol of fru 6-phosphate to fructose 1,6-diphosphate in


2994




1 min in the above system. The enzyme activity is expressed as units/10'Â°cells.

Production of Subunit-specific Antibodies. Subunit-specific monoclonal antibodies against M- and L-subunits of human PFK were produced by the hybridoma technique. Antibody-rich ascites fluids were

raised by injecting hybridomas into the peritoneal cavities of BALB/cxDBA/2 Ft mice. The details of the immunization protocol, techniquesof hybridization, screening of the secretory hybrids, and characterizationof the secreted antibodies are described elsewhere (3). The productionand characterization of a human P-subunit-specific mouse antiserum has

been described previously (29).Enzyme Immunoprecipitation Studies. Cell extracts were diluted to

a final concentration of 0.06 unit/ml using the extraction buffer. Fifty-/jlaliquots of a given extract were mixed in duplicate with 50 Â¿tlof the anti-M or anti-L monoclonal antibody-rich diluted ascites fluids (1:200 in PBS)or anti-P antiserum (1:64 in PBS). After incubation at 37Â°Cfor 30 min,

100 ni of a 10% suspension of freshly washed staphylococci bearingprotein A (IgGSorb) were added (30). The mixtures were incubated at37Â°C for 30 min and centrifuged, and the supernatants were assayed

for residual enzyme activity; only tubes showing more than 7% precipitation in comparison with their controls were considered to be positive(31). Each cell suspension was tested in duplicate using 3 types ofantibodies with concurrent duplicate controls.

Isozymic Complements of Various Cell Types. Chromatographieseparation of PFK isozymes was carried out at 4Â°C using DEAE-

Sephadex A-25 and a concave elution gradient of salt as described

previously (2, 32).

RESULTS

Purity of Isolated Cell Suspensions. Mononuclear Å“il suspensions consisting of lymphoblasts or myeloblasts were generally >95% pure; the occasional contaminants were neutrophilsand erythrocytes. Normal T-lymphocytes isolated from peripheralblood were usually very pure, i.e., >90% with 50% yield. Thethymic specimens yielded >95% pure T-cells. The compositionof non-T-cells varied with the source of the lymphoid material.The non-T-cell fractions obtained from peripheral blood contained

20 to 60% monocytes; the remaining cells were a mixture ofSlg-positive cells (B-cells) and a heterogeneous population of laantigen-positive cells, most of which are in the B-cell series.Typically, >90% pure Slg-positive B-cells are obtained with

approximately 50% yield. Tonsils provided an excellent sourceof B-cells; they often contained >50% non-T cells and <5%monocytes. The tonsillar non-T-cell fraction often contained ahigher proportion of Slg-positive B-cells (>70%) than did periph

eral blood. The final polymorphonuclear cell preparations showed95 to 98% purity. However, a high degree of purity was oftendifficult to obtain with peripheral blood specimens from patientswith chronic myelogenous leukemia; the major contaminantswere lymphocytes and platelets. The use of defibrinated bloodresulted in negligible contamination with platelets; however, allpreparations studied contained 10 to 20% lymphocytes.

Characterization of the Anti-PFK Antibodies. As reported

earlier, each monoclonal antibody exhibits strict subunit specificity, i.e., anti-M or anti-L. Since a given antibody reacts with therespective subunit, whether contained within a homo- or a het-erotetramer, it precipitates all M- or L-bearing heterotetramers.For instance, anti-M precipitates not only M4, but also M3L andMP2L isozymes (3). The dilutions of the ascites used (1:200)provided an antibody excess.

Similarly, the mouse anti-P antibody also shows strict reactivity

with the P-subunit only, and it precipitates not only P4 but alsoP-containing heterotetramers. As reported earlier, the anti-Pantibody is not a truly "monospecific" antibody; however, it is

functionally rendered monospecific in the conditions of the activeenzyme precipitation assay (29). A 1:64 dilution of the antibodyretains full precipitating capacity, if used in conjunction withIgGSorb (29); therefore, this dilution was used in the studiesreported in this article.

Total PFK Contents of Various Cell Types. Table 1 lists thePFK activity levels of different types of malignant cells, expressedboth as units/1010 cells and as a percentage of the respective

normal cell type. There was no significant difference betweenthe total PFK contents of B- and T-lymphocytes and between

normal lymphocytes and normal granulocytes, i.e., 10 to 15units/1010 cells. The malignant cells isolated from all leukemias

and lymphomas exhibited none to minimal increases in total PFKactivity on a per cell basis, i.e., 102 to 196% of normal. Incontrast, very high activity levels were present in all establishedmalignant cell lines of lymphoid, myeloid, erythroid, and fibro-

blastic origin. The highest activity levels were present in rapidlyreplicating and therefore heavily glycolyzing cells, i.e., HeLa andKB. PHA-stimulated T-cell blasts showed moderately high activ

ity levels on the third and fourth days following stimulation.Chromatographie Studies. Charts 1 to 4 illustrate the repre

sentative isozymic profiles of PFK from native and cultured celllines of lymphoid (Chart 1), myeloid (Chart 2), erythroid (Chart3), and fibroblastic (Chart 4) origins. As shown in Chart 1,malignant cells from the cases of acute and chronic lymphocyticleukemia did not show any significant deviation in their isozymic

TabtelPFK activity levels of various normal and malignant cell types and cultured

cell lines

CelltypeLymphoidNormal

lymphocytes(6)"Acute

lymphoblastic leukemia(8)Chroniclymphocytic leukemia(27)B-cell

lymphomas(9)Epstein-Barrvirus-transformed B-cell lines(8)dT-cell

leukemia cell lines(5)dPHA-stimulated

T-cell blasts(10)MyeloidNormal

polymorphonuclear leukocytes(5)Acutemyeloblastic leukemia(11)Chronic

myelogenous leukemia(6)HL-60cell line(3)HL-60cell line, induced(3fKG-1

cell line(3)KG-1cell line, induced(3)"ErythroidRBC

(6)K-562cell line(4)K-562cell line, induced(3)"FibroblasticFibroblast

cell lines(6)HeLacell line(3)KB

cell line (4)PFK

activity(units/1010cells)11

Â±1.7Â°15.3

Â±4.117.6Â±1116.9

Â±7.498Â±12.761

.2Â±5.654.8Â±9.512.4

Â±2.524.3Â±11.212.7Â±1.959.1Â±10.457.2Â±0.375.4Â±20.650.1Â±92.1

Â±0.185.6Â±8.175.1Â±5.4244.4

Â±73.3264Â±19.5255.3

Â±30.7%

of normal8100149160154893558532100196102477462608405

a Percentage of the activity of the normal counterpart, i.e., lymphocytes or

neutrophils.b Numbers in parentheses, number of separate donors or observations on each

cell line(s) investigated.c Mean Â±SD." Number of observations on 2 separate T-cell and 5 separate B-cell lines.8 Induced with DMSO, hemin, sodium butyrate, or TPA (see text).


2995




500

70

FractionChart 1. Representative isozymic profiles of PFKs from normal and malignant

cellsof iymphoidorigin.A: normal lymphocytes.8:0, acute lymphoblasticleukemia;â€¢,chronic lymphocytic leukemia.C: PHA-stimulatedT-cell blasts. D: Epstein-Barrvirus-transformed lymphoblastoid cell line. E: established T-cell leukemia cell line,MOLT-4. F: B-cell lymphoma.

profiles (Chart 10) compared with normal B- or T-lymphocytes(Chart 1A). Malignant lymphoma cells exhibit predominance ofhybrid forms (Chart 1F) as compared with L, in normal andleukemic lymphocytes. In contrast, PHA-stimulated T-cell blasts(Chart 1C), Epstein-Barr virus-transformed B-cell lines (Chart10), and T-cell leukemia cell-lines (Chart 1E) exhibit dramaticalterations in their isozyme profiles. In each case, there is apredominance of P4- and P-containing isozymes instead of thatof U- and L-containing isozymes seen in normal lymphocytes.In particular, MOLT-4 exhibits P4 as its predominant species(Chart 1Â£).In B-cell lines and PHA-stimulated T-cell blasts, Uand L-containing isozymes (and occasionally M4 peak) are distinctly present (Chart 1, C and D) as compared with T-cell lines,

where these species are absent.The isozymic profile of malignant cells from the cases of acute

myelogenous leukemia (Chart 2B, closed circles) is not significantly different from that of normal polymorphonuclear leukocytes (Chart 2A). However, that of cells from chronic myelogenous leukemia consistently demonstrated a distinct P4 peak,which could not be accounted for by the degree of plateletcontamination present (Chart 28, open circles). In cultured mye-loid cell lines, KG-1 (Chart 2C) and HL-60 (Chart 2D) multipleisozymes composed of all 3 subunits were present with L-

containing isozymes predominating. Both cell lines showed disappearance of their early-eluting hybrid isozymes following in

duction with TPA and DMSO, respectively (Chart 2, E and F);

x

S 5

CD 01E g

tae

Â«.p.

B

30 50 70 30

Fraction

50 70

Chart 2. Representative isozymic profiles of PFKs from normal and malignantcells of myeloid origin. A: normal polymorphonuclear leukocytes. B: O, chronicmyelogenous leukemia;â€¢,acute myelogenous leukemia.C: KG-1 cell line. D: HL-60 cell line. E: KG-1 cell line induced with TPA. F: HL-60 cell line induced withDMSO.

their isozymic profiles now appeared to approach that of normalpolymorphonuclear leukocytes (Chart 2A).

As shown in Chart 3, in contrast to normal erythrocytes, whichexpress 5-membered set composed of the M and L subunits,

i.e., M4, M3L, M2L2, MLÂ«,and L, (Chart 3A), or 8 to 10 isozymesdue to polymorphism of the M subunit6 (composed of M, M',

and L subunits (Chart 3B), the K-562 cell line exhibits multiple

species composed of all 3 subunits (Chart 3C). Hemin inductionof K-562 did not result in any significant alteration in isozymicprofile (Chart 3D), although >90% of cells exhibited hemoglobinsynthesis as indicated by benzidine positivity on Days 7 and 9following induction. Although normal diploid fibroblasts expressa profile similar to that of the cultured Iymphoid line (P4- and P-containing isozymes predominating) (Chart 4A), a single fibro-blast cell line (ICT-272) was found to express an additional major

peak in the position of M4 isozyme. HeLa (Chart 4C) and KB(Chart 40) cell lines exhibited predominantly P4species with onlya few hybrids of P+L and M+L subunits.

Immunoprecipitation Studies. Our interpretation of the Chromatographie data were essentially confirmed by the immunopre-

' S. Vora. Genetic polymorphismof human muscle-type phosphofructokinaseinvarious racial groups: implication in the phenotypic expression of the erythrocytephosphofructokinaseisozymes, manuscript in preparation.


2996




M.tÃ

O

l 5

Â«.p. u B M, P,* t

ut

30 50 70 30

Fraction

50 70

Chart 3. Representative Â¡sozymicprofiles of PFKs from normal and malignantcells of erythroid origin. A, normal erythrocytes (typical profile);B, normal erythro-cytes (atypical profile); C, K-562 cell line; D, K-562 cell line induced with hemin orsodium butyrate.

xM

8CD

il

"g.

CO

B M. P,t *

30 50 70 30

Fraction

50 70

Chart 4. Representative isozymic profiles of normal and malignant cells offibroWastic origin. A. normal diploid fibroblasts (typical profile); B, normal diptoidfibroblasts (atypical profile);C, HeLa cell line;D, KB cell line, a derivative of HeLa.

cipitation studies. As shown in Table 2, the subunit compositionsof PFKs from leukemic cells do not differ significantly from thatof normal lymphocytes, except perhaps an increase in the Lsubunit in leukemic cells. The M subunit contributes relativelylittle to the PFKs from all the lymphoid cells except the Epstein-Barr virus-transformed B-cell lines. Note that the M subunit isalmost absent in T-cell blasts and that the L subunit contributesin a minor way to PFK of T-cell leukemia cell lines. Similarly, cells

Table 2Immunoprecipitationvaluesof PFKfrom various cell types using subunit-specific

antibodies% of precipitation8

Cell type Anti-M Anti-L Anti-P

LymphoidNormal lymphocytes (6)" 18.4Â±4.6C 62.5 Â±6.2 68.6 Â±6.7Acute lymphoblastic teuke- 15.4 Â±2.7 92 Â±0.8 50.6 Â±9.1

mia (4)Chronic lymphocytic teuke- 11.4Â±7.1 82.4Â±12.4 68.7 Â±9.7

mia (5)B-cell lymphomas (3) 2 Â±2.9 89.3 Â±8.1 59.7 Â±10.1EB virus-transformed B-cell 38.6 Â±13.4 68.9 Â±9.5 76.2 Â±5.5

lines (6)"T-cell leukemiacell lines (5)" 2.5 Â±1.7 25.7 Â±7.5 85.7 Â±8.5PHA-stimulatedT-cell blasts 0 71.5 Â±7.2 63.8 Â±10.9

(9)

MyeloidNormalpc4ymorphonuclearleukocytes

(5)Acutemyeloblastic leuke

mia(3)Chronicmyelogenousleu

kemia(3)HL-60cell line(3)HL-60cell line, induced(3fKG-1

cell line(4)KG-1cell line, induced(3)"ErythrokjRBC

(6)K-562cell line(4)K-562cell line, induced(3)FibroblasticFibroblast

cell lines(6)HeLacell line(3)KB

cell line (4)18.5

Â±49.5

Â±3.811.1

Â±7.17.3

Â±2.302.9

Â±4.21.4Â±2.374.9

Â±6.143.4Â±10.144.8Â±12.924.3

Â±10.450.3Â±3.325.5Â±2.079.7

Â±7.086.6

Â±3.487.7

Â±2.291.1

Â±1.492.2Â±3.381

.8Â±6.591.8+9.988

Â±4.180.7Â±5.875.1Â±9.577

Â±4.458.5Â±1.748.5

Â±7.156.1

Â±5.676.6

Â±3.564.8

Â±12.653.8

Â±1037.2Â±7.155Â±17.259.1Â±12.11

Â±2.157.2Â±7.664.8

Â±18.488.2

Â±11.985.3Â±3.583.3Â±8.0

a Percentage of the activity precipitated as compared with that of the controlunder the conditions described in "Materials and Methods"; values less than 7%

precipitation are considerednot significant.6 Numbers in parentheses,number of separate donors or observations on each

cell lme(s)investigated.0 Mean Â±SO.a Number of observationson 2 separate T-cell and 5 separate B-cell lines.8 Induced with DMSO, hemin, sodium butyrate, or TPA (see text).

of myeloid origin express none or very few M subunits; theirPFKs consist predominantly of the P+L subunits, the subunitspresent in the mature neutrophils. Although the differentiatedKG-1 and HL-60 cell lines exhibit profiles resembling that of

mature neutrophils, no alterations are seen in their precipitationvalues because of the nature of the assay (see "Materials andMethods"). In contrast to the near absence of the M subunit

from the cells of myeloid origin, the K-562 cell line exhibits a

substantial expression of the M subunit. However, induction withconcomitant differentiation in vitro did not result in any significantalteration in the isozymes of this cell line. As reported previously,PFKs from cultured fibroblasts predominantly consist of P+Lsubunits with the M subunit contributing only in a minor way. Incontrast, PFKs from HeLa and KB cell lines consist predominantly of P4 and of hybrid isozymes consisting of all 3 types ofsubunits.

DISCUSSION

Since the initial description of the Warburg effect, i.e., the highaerobic glycolysis of malignant tumors by Warburg (33) in 1926,the enzymatic and/or metabolic mechanism(s) responsible forthis phenomenon have remained an enigma (for review, see


2997




Refs. 4 and 7). Since PFK is now known to be the primary rate-

limiting reaction of the glycolytic pathway and to mediate thePasteur effect (34-36), we reasoned that the alterations in theactivity and isozymes of PFK may be invoked to explain theobserved decrease or lack of Pasteur effect and high aerobicglycolysis of cancer cells.

The total activities of PFK, PK and HK were first found to beelevated in experimental rat hepatomas of differing growth rateand differentiation (37-40). The PFK activity showed a 2- to 3-

fold increase in poorly differentiated (rapidly growing) tumors,while it was normal or even diminished in well-differentiated

(slow-growing) tumors, indicating a direct correlation of the enzyme activity with the degree of differentiation (39-43). Weber

and Lea (40) also demonstrated the close correlation betweenthe PFK activity (and also PK and HK activities) and the growthrate of the tumor on the one hand and lactate production by thetumor on the other.

Subsequent studies of rodent hepatomas and other tumors,including Yoshida sarcomas and ascites hepatomas and Ehrlichascites tumor cells, revealed that HK and PK also exhibit iso-zymic shifts which appear to facilitate glycolysis. The liver-typehigh-Km (allosteric) isozymes, which are responsive to hormonaland nutritional regulation, disappeared or decreased, whereasthe non-liver type, i.e., muscle- or kidney-type isozymes (nonal-

losteric), which are much less regulated, if at all, either appearsanew or increased (38, 44-46). In contrast, isozymic alterationsexhibited by PFK in rat neoplasia differed fundamentally fromthose in HK and PK. Although the non-muscle and non-liver PFKisozymes (most likely platelet or brain-type) appeared anew, theliver-type PFK, a highly regulated isozyme, not only persistedbut actually increased significantly in all the rodent tumors studied (10, 41, 42, 47-50). Only a single study reported on the totalloss of the U isozyme and appearance of P4 and hybrid speciesof P+L in Yoshida ascites hepatomas and sarcomas (51). Ourrecent studies of the well-differentiated and anaplastic medullarythyroid carcinomas of the rat dramatically reinforce the preferential expression of the liver-type isozyme by the cancer cell(10). The rapidly growing anaplastic tumors showed an increasein the liver-type isozyme as compared to the slow-growing

differentiated tumors. This increase in the L subunit is remarkable, since PFKs from both types of tumors predominantly consistof P4 and hybrids of P+L subunits (10).

In contrast to the extensive and elegant studies of PFK inrodent tumors, only a very few studies have attempted to investigate the enzyme in human neoplasia (19-22, 52-54). This is

probably due to the recent recognition of the PFK isozymes inhumans and to the difficulties encountered in resolving andidentifying them. As a matter of fact, some of the previousobservations appear to be inaccurate in light of our present dataobtained using Chromatographie and immunochemical techniques discussed below. Meienhofer ef al. (21) first reported thatPFK from leukemic myeloid cells showed altered immunochemical and kinetic properties consistent with an increased expression of the M subunit as compared to normal leukocytes. Subsequent immunochemical studies by this group showed thatthere were no differences in the isozymic complements of malignant cells from acute and chronic lymphocytic leukemia as compared with normal lymphocytes (22). Bennett eÃal. (19) comparedthe isozymic profiles of various types of brain tumors with thoseof adult and fetal brain tissue using polyacrylamide gel electro-

phoresis. Although the technique used did not resolve the multiple isozymes now known to exist in adult brain (32), it clearlydemonstrated the general trend of isozyme alterations. Theisozyme pattern of the slow-growing, well-differentiated tumorsresembled that of adult brain, whereas the fast-growing undiffer-entiated tumors exhibited liver-type isozyme, in addition to iso

zymes normally present in adult brain. Recent immunochemicalstudies by Cottreau ef a/. (20) have confirmed these results andhave suggested further that the isozymic complements are correlated with the degree of differentiation of the tumors.

We found that PFK activity is normal or only mildly elevated(1- to 2-fold increment) in the neoplastic cells from various acute

and chronic leukemias and lymphomas, as compared with theirnormal counterparts (Table 1). In contrast, all types of culturedcell lines, irrespective of their cell types of origin, i.e., lymphoid,myeloid, erythroid, or fibroblastic, possess very high PFK activitylevels (4- to 9-fold increment). The highest increases in the PFKactivity, /.e., 10-fold, are shown by the Epstein-Barr virus-transformed lymphoblastoid cell lines derived from B-lymphocytes.Similarly, PHA-stimulated T-cell blasts or established T-cell leukemia cell lines show a 5- to 6-fold increase in PFK activity.Interestingly, the established human malignant cell lines, i.e., K-562, HL-60, KG-1, HeLa, and KB, exhibit relatively high PFK

activities, and the highest activities are shown by rapidly replicating and heavily glycolyzing HeLa and KB cell lines (200 to300 units/1010 cells).

Our Chromatographie (Charts 1 to 4) and immunological data(Table 2) illustrate the fact that in these cell types the isozymicalterations of PFK parallel the quantitative increases in the totalenzyme activity. Various types of leukemia and lymphoma cellsexhibit no to mild increases in the activity (Table 1) and no tominimal alterations in isozymic profiles (Charts 1 and 2). Themalignant cells from acute and chronic lymphoid leukemias andB-cell lymphomas exhibit no alterations in isozymic profiles (Chart

1, B and F) compared with normal lymphocytes (Chart 1X\). Incontrast, those from acute and chronic myeloid leukemias exhibitminimal increases in hybrid species of P+L and P4, respectively(Chart 20) as compared to normal neutrophils (Chart 2A). T-cellblasts irrespective of their origin, i.e., PHA-induced (Chart 1C) orT-cell leukemia cell lines (Chart 1Â£),exhibit striking increases in

PFK activity (Table 1) which are associated with drastic alterations in their isozyme profiles. Note that U- and L-containing

isozymes predominate in normal lymphocytes (Chart 1A),whereas P4- and P-containing Â¡sozymespredominate in T-cellblasts (Chart 1, C and E). Although P4- and P-containing isozymes also predominate in Epstein-Barr virus-transformed B-cell

lines (chart 1D), these cell lines consistently exhibit a significantexpression of the M subunit (approximately 40 to 50% M-

containing species) (Table 2). It is noteworthy that neither leukemic nor PHA-induced T-cell blasts express the M subunit insignificant amounts (approximately 0 to 5% M-containing spe

cies) (Table 2). In fact, within hours of induction with PHA, thenormally expressed M subunit in the lymphocyte (approximately26% precipitation) becomes undetectable (0 to 5% precipitation),indicating a repression of the M-gene expression.7

In order to investigate whether PFK gene(s) expression iscorrelated with the state of differentiation of these cell lines andwhether some of the observed isozymic alterations are related

7S. Vora, unpublished data.


2998




to the culture conditions rather than differentiation, we investigated several malignant erythroid and myeloid cell lines whichcan be induced to differentiate in vitro. HL-60 and KG-1, the

human promyelocytic and granulocytic leukemia cell lines, canbe induced to terminally differentiate into morphologically andfunctionally mature granulocytes by a wide variety of compoundsincluding DMSO and TPA. Prior to induction, KG-1 and HL-60

cell lines express all 3 PFK subunits with P+L subunits predominating (Chart 2, C and D, respectively; Table 2); again, the Msubunit is almost absent. Concomitant with differentiation, however, the total PFK activity does not show any decrement, thehybrid isozymes of P+L decrease or disappear in both cell lines,and the isozymic profile now approaches that of mature neutro-

phils; the minimal expression of the M subunit is also lost.In contrast, K-562, a myeloid leukemia cell line (55) possesses

a number of phenotypic properties of embryonic erythroid progenitor cells including embryonic-fetal hemoglobin synthesis

when induced with hemin and sodium butyrate (23, 56). Theuninduced K-562 cell line showed a simultaneous expression of

all 3 subunits with the P and L subunits predominating (Chart3C). However, unlike the myeloid cell lines, the K-562 cell line

expresses the M subunit in significant amounts (approximately40 to 50% precipitation; Table 2). However, despite differentiation with hemin and DMSO as indicated by hemoglobinization,this cell line did not alter its total PFK activity (Table 1) or isozymicprofile (Chart 3D). Chart 3 compares the isozyme complementsof both uninduced and induced K-562 cell line with those of

mature erythrocytes. The normal erythrocyte PFK consists of 5isozymes composed of the M and L subunits, M4, M3L, M2L2,ML3, and U (2). Among certain ethnic groups, a "fibroblastic"

pattern (Chart 3B) is prevalent which most probably results froma genetic polymorphism of the M subunit.6

The observed lack of isozymic alterations in PFK despitedifferentiation of K-562 is consistent with those reported for

other glycolytic enzymes, lactate dehydrogenase (23), PK (57,58), and HK (57). Neither the total activities (57) nor the isozymicprofiles of these enzymes showed any significant alterations.Although these data have been interpreted to indicate a non-coordinate regulation of mature erythrocyte-specific proteins, it

is generally forgotten that fetal and not adult hemoglobins aresynthesized by these cells. It is conceivable that the erythrocyte-

specific glycolytic isozyme expression may be manifested bythese cells, if these cells were to express adult genetic programas indicated by the expression of adult hemoglobins. However,it is noteworthy that mature erythrocyte-specific isozymes of

HK, PFK, PK, and lactic acid dehydrogenase are expressedalthough in variable amounts by both uninduced and induced K-

562 cells.As reported previously, fibroblast cell lines were found to

exhibit very high PFK activity levels and a simultaneous expression of all 3 PFK subunits (1). The latter should theoreticallyresult in 15 distinct isozymic species. However, due to theunavoidable overlap of some heterotetramers during gradientelution, only 8 to 10 species are chromatographically distinguishable (Chart 4, A and ÃŸ).Most fibroblast cell lines exhibit anisozyme pattern similar to that illustrated in Chart 4A, where P4isozyme predominates with hybrids of P+L and M+L constitutingthe less abundant species. However, in an occasional cell line,M4 isozyme predominates as determined both chromatographically (Chart 10) and immunologically (approximately 50% precip

itation). As reported previously, the expression of the M subunitis variable in cultured human fibroblasts and the reasons for thisvariability are not entirely clear (31). Two permanent malignantcell lines, HeLa and KB, which are known to glycolyze veryactively also exhibit PFK activity levels comparable to culturedfibroblasts (Table 1). However, Chromatographie results indicatethat their PFKs consist primarily of P4 and hybrids of P+L andM+L subunits (Chart 4, C and D). Immunological data confirmthe predominance of the P subunit (approximately 85% precipitation) over the L (50 to 60% precipitation) and M subunits (25to 50% precipitation) (Table 2).

Our results taken together with those in the literature indicatethat human PFK undergoes quantitative increases and isozymicshifts during neoplastic transformation both in vivo and in vitro.These changes are apparently not correlated with the type ofcell from which cancer originated but rather the rate of growthand concomitant increase in metabolism. The isozymic alterations seem to parallel the quantitative increases in total enzymeactivity; the higher the increment, the more dramatic are theisozymic changes. Considering the expression of PFK in humancancers, in virus-transformed and permanent malignant cell lines,and in PHA-induced blasts, a unifying feature emerges; i.e.,

whenever there is an extra demand for glycolysis, the relativeexpression of the L subunit increases first, followed by that inthe P subunit, whereas the expression of the M subunit isvariable; the latter may decrease, increase, or remain unchanged.

The preferred expression of the L and P subunits may beexplained in terms of the biochemical and metabolic advantagesthey may confer upon the cancer cell because of their uniquekinetic and regulatory properties. Tanaka er al. (50) found thatthe P4 isozyme isolated from the rat hepatomas was the leastallosteric among the 3 isozymes. Sumi and Ui (48) also reportedon the novel regulatory properties (K+ inhibition and pH inde

pendence of ATP inhibition) of the Ehrlich ascites tumor cell PFKwhich appears to be of U type (50). Rat hepatoma PFKs (mainlyliver type) are also reported to be relatively insensitive to inhibitionby ATP and citrate (42, 59). Our recent studies also demonstratethe desensitization of the enzyme from undifferentiated thyroidtumors to potent inhibitors (ATP and citrate) and increasedsensitivity to the activator (fructose 2,6-diphosphate) (10), whichmost probably reflect the increase of the liver-type subunit (60).

The available metabolic studies also point to the dysfunction oftumor PFK with the loss of its regulatory activity on tumorglycolysis which again most probably reflects its altered isozymiccomposition, i.e., an increased expression of the L subunit (61-

63).It is now generally accepted that the Pasteur effect, i.e.,

inhibition of glycolysis during active respiration, is mediated viaan inhibition of PFK by ATP, citrate, and other inhibitory metabolites. Considering the known kinetic and regulatory propertiesof liver PFK, it is conceivable that an increase in this subunit mayresult in the weakening of the Pasteur effect, since ATP andcitrate are unable to exercise a negative feedback inhibition ofthe enzyme.

It is of interest to note that a minor increase in the L subunitmay suffice to alter the regulatory properties of the total tumorPFK since the former can tetramerize with the non-L subunits to

alter the overall isozymic composition. It is well established thatthe kinetic properties of a given hybrid isozyme, e.g., M2L2 orM3L, are not an arithmetic average of those of M4 and U


2999




Â¡sozymes,but unique to the hybrid per se (64). It is conceivablethat a reduction in the rate-limiting activity of PFK in concert with

similar changes for PK (44, 50) and HK (45) account for theobserved high rate of aerobic glycolysis of tumors as suggestedpreviously (50).

In summary, these studies show that PFK is both a transformation- and progression-linked discriminant of malignancy and

that the reprogramming of PFK genes occurring during malignanttransformation proceeds in an orderly fashion. The available dataon the regulatory properties of the P and L isozymes implicatethem in the mediation of the Warburg effect exhibited by thecancer cell. Detailed analyses of the contribution of different PFKisozymes to the regulation of glycolysis and respiration manifested as the Pasteur effect and the Crabtree effect, respectively,may assist toward the elucidation of the mechanism(s) of theWarburg effect. Moreover, if L and P isozymes of PFK areessential to the survival of cancer cells, they may provide sensitive targets for the design of selective cancer chemotherapy.

ACKNOWLEDGMENTS

The authors wish to thank Lisa Dodson and Rita Raab for excellent technicalassistance and Elizabeth Goddard for her help in the preparation of the typescript.

REFERENCES

1. Vora, S. Isozymes of human phosphofructokinase in blood cells and culturedcell-lines: molecular and genetic evidence for a trigenic system. Blood, 57:724-732, 1981.

2. Vora, S., Seaman, C., Durham, S., and Piomelli, S. Isozymes of humanphosphofructokinase: identification and subunit structural characterization ofa new system. Proc. Nati. Acad. Sci. USA, 77: 62-66,1980.

3. Vora, S.. Wims, L. A., Durham, S., and Morrison, S. L. Production of monoclonal antibodies to the subunits of human phosphofructokinase: new tools forthe immunochemical and genetic analyses of isozymes. Blood, 58: 823-829,1981.

4. Racker, E. Resolution and reconstitution of biological pathways from 1919 to1984. Fed. Proc., 42: 2899-2909, 1983.

5. Weber, G. Enzymology of cancer cells. N. Engl. J. Med., 296: 486-493,1977.6. Weber, G. Enzymology of cancer cells. N. Engl. J. Med., 296: 541-551,1977.7. Weinhouse, S. Isozyme alterations, gene regulation and the neoplastic trans

formation. Adv. Enzyme Regul., 27: 369-386,1983.8. Weber, G., Singhai, R. L., Stamm, N. B., Lea, M. A., and Fisher, E. A.

Synchronous behavior pattern of key glycolytic enzymes: glucokinase, phosphofructokinase and pyruvate kinase. Adv. Enzyme Regul., 4: 59-81, 1966.

9. Dunaway, G. A. A review of animal phosphofructokinase isozymes with anemphasis on their physiological role. Mol. Cell. Biochem., 52: 75-91, 1983.

10. Oskam, R., Vora, S., Rijksen, G., and Staal, G. E. J. Isozymic composition andregulatory properties of phosphofructokinase from well-differentiated and an-aplastic medullary thyroid carcinomas of the rat. Cancer Res., 45: 135-142,1985.

11. Schapira. F. Resurgence of fetal isozymes in cancer: study of aldolase,pyruvate kinase, lactic dehydrogenase and /3-hexosaminidase. In: M. C. Rat-tazzi, J. G. Scandalios, and G. S. Whitt (eds.), Isozymes: Current Topics inBiological and Medical Research, Vol. 5, pp. 27-75. New York: Alan R. Uss,1982.

12. Beemer, F. A., Vlug, A. M. C., Rijksen, G., Hamburg, A., and Staal, G. E. J.Characterization of some glycolytic enzymes from human retina and retino-blastoma. Cancer Res., 42: 4228-4232,1982.

13. Bennett, M. J., Timperley, W. R., Taylor, C. B., and Hill, A. S. Fetal forms ofpyruvate kinase isozymes in tumors of the human nervous system. Neuro-pathol. Appi. Neurobiol., 7: 347-356, 1975.

14. Bennett, M. J., Timperley, W. R., Taylor, C. B., and Hill, A. S. Isozymes ofhexokinase in the developing normal and neoplastic human brain. Eur. J.Cancer, 74:189-193,1978.

15. Hammond, K. D . and Balinsky, D. Isozyme studies of several enzymes ofcarbohydrate metabolism in human adult and fetal tissues, tumor tissues, andcell cultures. Cancer Res., 38: 1323-1328,1978.

16. Kamel, R., and Schwarzfischer, F. Hexokinase isozymes in human neoplasticand fetal tissue: the existence of HK II in malignant tumors and in placenta.Hum. Genet., 30:181-185,1975.

17. Sato, S., Kikuchi, Y., Takakura, K., Chien, T. C., and Sugimura, T. Diagnosticvalue of aldolase and hexokinase isozymes of human brain and uterine

tumours. Gann Monogr. Cancer Res., 73: 279,1972.18. Van Veelen, C. W. M., Verbiest, H., Vlug, A. M. C., Rijksen, G., and Staal, G.

E. J. Isozymes of pyruvate kinase from human brain, meningiomas andmalignant gliomas. Cancer Res., 38: 4681-4687,1978.

19. Bennett, M. J., Taylor, C. B., and Timperley, W. R. Etectrophoretic study ofphosphofructokinase isoenzymes from human tissues and tumours of thenervous system. Eur. J. Cancer, 77: 359-363,1975.

20. Cottreau, B., Rousseau-Merck, M. F., Nezelof, C., and Kahn, A. Pyruvatekinase and phosphofructokinase isozymes in childhood cancer. Pediatr. Res.,76:199-202,1982.

21. Meienhofer, M-C., Carayon, P., and Gacon, G. Les isozymes de la phosphofructokinase humaine. Biochimie (Paris), 54:1399-1403,1972.

22. Meienhofer, M-C., Lagrange, J-C., Cottreau, D., Lenoir, G., Dreyfus, J-C., andKahn, A. Phosphofructokinase in human blood cells. Blood, 53: 389-4001979.

23. Benz, E. J., Jr., Mumane, M. J., Tonkonow, B. L., Berman, B. W., Mazur, E.M., Cavallesco, C., Jenko, T., Snyder, E. L., Forget, B. G., and Hoffman, R.Embryonic-fetal erythroid characteristics of a human leukemic cell line ProcNati. Acad. Sci. USA, 77: 3509-3513,1980.

24. Hoffman, T., and Kunkel, H. G. The E rosette test. In: B. R. Bloom and J. R.David (eds.), In Vitro Methods in Cell-mediated and Tumor Immunity, pp. 71-81. New York: Academic Press, Inc., 1980.

25. Halper, J. P., Knowles, D. M., II, and Wong, C. Y. la antigen expression byhuman malignant lymphoma: correlation with conventional lymphoid markers.Blood, 55: 373-382, 1980.

26. Winchester, R. J., Wang, C. Y., Halper, J., and Hoffman, T. Studies with B cellallo- and heteroantisera: parallel reactivity and special properties. Scand. J.Immunol., 5:745-757, 1976.

27. Clark, R. A., and Kimball, H. R. Defective granulocyte chemotaxis in theChediak-Higashi syndrome. J. Clin. Invest., 50: 2645-2652, 1971.

28. Corash, L., Shafer, B., and Periow, M. Heterogeneity of human whole bloodplatelet subpopulations. II. Use of sub-human primate model to analyze therelationship between density and platelet age. Blood, 52: 726-734,1978.

29. Vora, S., Miranda, A., Hernandez, E., and Francke, U. Assignment of thehuman gene for platelet-type phosphofructokinase (PFKP) to chromosome10p: Novel use of polyspecific rodent antisera in mapping human enzymegenes. Hum. Genet., 63: 374-379, 1983.

30. Kessler, S. W. Cell membrane antigen isolation with the staphylococcal proteinA-antibody adsorbent. J. Immunol., 777:1482-1490,1976.

31. Vora, S., Durham, S., de Martinville, B., George, D. L., and Francke, U.Assignment of the human gene for muscle-type phosphofructokinase (PFKM)to chromosome 1 (cen -Â»q32) using somatic cell hybrids and monoclonal anti-M antibody. Somatic Cell Genet., 8:95-104,1982.

32. Davidson, M., Collins, M., Byrne, J., and Vora, S. Alterations in phosphofructokinase isozymes during early human development. Establishment of adultorgan-specific patterns. Biochem. J., 274: 703-710,1983.

33. Warburg, 0. Ãœberdie Wirkung von Blausaureathylester (Athylbarbylamin) aufdie Pasteursche Reaktion. Biochem. Z., 772: 432-441, 1926.

34. Newsholme, A. E., and Gevers, W. Control of glycolysis and gluconeogenesisin liver and kidney cortex. Vitamins Hormones, 25: 1-79,1967.

35. Ramaiah, A. Pasteur effect and phosphofructokinase. Curr. Top. Cell RegulS: 297-345, 1974.

36. Scrutton, M. C., and Utter, M. F. The regulation of glycolysis and gluconeogenesis in animal tissues. Annu. Rev. Biochem., 37: 249-302,1968.

37. Olson, R. E. Oxidation of C"-labeled carbohydrate intermediates in tumor and

normal tissue. Cancer Res., 77: 571-584,1951.38. Shatton, J. B., Morris, H. P., and Weinhouse, S. Kinetic, electrophoretic and

Chromatographie studies on glucose-ATP phosphotransferases in rat hepato-mas. Cancer Res., 29:1161-1172,1969.

39. Shonk, C. E., Morris, H. P., and Boxer, G. E. Patterns of glycolytic enzymesin rat liver and hepatoma. Cancer Res., 25: 671-676,1965.

40. Weber, G., and Lea, M. A. The molecular correlation concept of neoplasia.Adv. Enzyme Regul., 4: 115-145,1966.

41. Dunaway, G. A., Leung, L-Y., Thrasher, J. R., Naqui, D., and Morris, H. P.Regulation of rat liver phosphofructokinase levels in Morris hepatomas.Biochem. Biophys. Res. Commun., 88: 882-886,1979.

42. Dunaway, G. A., Morris, H. P., and Weber, G. A comparative study of rat liver,muscle, and hepatoma 3924A phosphofructokinase isozymes. Cancer Res34:2209-2216,1974.

43. Kirby, W., Morris, H. P., and Taylor, C. B. A comparison of the phosphofructokinase isoenzyme profiles of transplantable hepatomas and normal liver(letter). Eur. J. Cancer, 70: 629-631, 1974.

44. Imamura, K., Taniuchi, K., and Tanaka, T. Multimolecular forms of pyruvatekinase. II. Purification of M2-type pyruvate kinase from Yoshida ascites hepatoma 130 cells and comparative studies on the enzymological and immunolog-ical properties of the three types of pyruvate kinases, L, M1 and M2. J.Biochem. (Tokyo) 72:1001-1015, 1972.

45. Sato, S., Matsushima, T., and Sugimura, T. Hexokinase isozyme patterns ofexperimental hepatomas of rats. Cancer Res., 29:1437-1446,1969.

46. Taylor, C. B., Morris, H. P., and Weber, G. A comparison of the properties ofpyruvate kinase from hepatoma 3924-A, normal liver and muscle. Life Sci fi-635-644,1969.


3000




47. Dunaway, G. A., Naqui, D., Kruep, D., Thrasher, J. R., and Morris, H. P.Phosphofructokinase isozymes of Morris hepatomas. Arch. Biochim. Biophys.,2/2:1-8, 1981.

48. Sumi, T., and Ui, M. Purification and some properties of phosphofructokinasefrom Ehrlich ascites tumor cells. Biochim. Biophys. Acta, 268:354-363,1972.

49. Tanaka, T., An, T., and Sakaue, Y. Studies on multimolecular forms ofphosphofructokinase in rat tissues. J. Biochem. 69: 609-612,1971.

50. Tanaka, T., Imamura, K., Ann, T., and Tanteuhi, K. Multimolecular forms ofpyruvate kinase and phosphofructokinase in normal and cancer tissues. GannMonogr. Cancer Res., 13: 219-234,1972.

51. Kurata, N., Matsushima, T., and Sugimura, T. Multiple forms of phosphofructokinase in rat tissues and rat tumors. Biochem. Biophys. Res. Commun., 48:473-479,1972.

52. Vora, S. Isozymes of phosphofructokinase: biochemical and genetic aspects.In: M. C. Rattazzi, J. G., Scandaltos, and G. S. Whitt (eds.), Isozymes: CurrentTopics in Biological and Medical Research, Vol. 11, pp. 3-23. New York: Alan

R. Uss, 1983.53. Vora, S., Durham, S., and Piomelli, S. Phosphofructokinase (PFK) isozymes:

biochemical discriminants of malignancy (abstract). Pediatr. Res., 14: 542,1980.

54. Vora, S., Halper, J., Knowles, 0., Ill, and Durham, S. Changes in the activityand isozymic pattern of phosphofructokinase (PFK) during blastogenesis andneoplastia transformation. Am. J. Hum. Genet., 32: 58A, 1980.

55. Lozzio, C. B., and Lozzio, B. B. Human chronic myelogenous leukemia cell-line with positive Philadelphia chromosome. Blood, 45: 321-334,1975.

56. Hoffman, R., Mumane, M. J., Burger, D., Tonkonow, B., Ibrahim, N., Elder, J.T., Mazur, E., Scarpa, A., Diamond, A., Bruno, E., Levere, R. D., Forget, B.G., and Benz, E. J., Jr. Characterization of a human leukemia cell line. In: G.Stamatoyannopoulos and A. W. Nienhuis (eds.), Hemoglobins in Developmentand Differentiation, pp. 487-506. New York: Alan R. Uss, Inc., 1981.

57. Jansen, G., Rijksen, G., de Gast, G. C., and Staal, G. E. J. Glycolytic enzymesof an erythroleukemic cell line, K562, before and after hemoglobin induction.Exp. Hematol., 11: 626-638,1983.

58. Max-Audit, I., Testa, U, Kechemir, D., Titeux, M., Vainchenker, W., and Rosa,R. Pattern of pyruvate kinase isozymes in erythroleukemia cell lines and innormal human erythroblasts. Blood, 64:930-936,1984.

59. Kirby, W., and Taylor, C. B. Differences in the properties of phosphofructokinase isoenzymes of rat tissues. Biochem. Soc. Trans., 2: 491-494,1974.

60. Vora, S., Oskam, R., and Staal, G. E. J. Isoenzymes of phosphofructokinasein the rat: demonstration of the three non-identical subunits by biochemical,

immunochemical and kinetic studies. Biochem. J., in press, 1985.61. Meldotesi, M. F., and Lacetti, P. Phosphofructokinase regulation in some

transplantabte thyroid tumors. Cancer Res., 39: 2796-2801,1979.

62. Meldotesi, M. F., Macchia, V., and Laccetti, P. Differences in phosphofructokinase regulation in normal and tumor thyroid cells. J. Biol. Chem . 257: 6244-

6251,1976.63. Meldotesi, M. F., and Maccia, V. Regulation of the Embden-Meyerhof pathway

in a transplantaWe rat thyroid tumor. Cancer Res., 32: 2793-2798,1972.64. Gonzalez, F., and Kemp, R. G. The A2B2 hybrid isozyme of phosphofructoki

nase. J. Bid. Chem., 253:1493-1497,1978.


3001



1985;45:2993-3001. Cancer Res Shobhana Vora, James P. Halper and Daniel M. Knowles Discriminants of Malignancy

: Transformation- and Progression-linkedin Vitroand in VivoPhosphofructokinase during Malignant Transformation

Alterations in the Activity and Isozymic Profile of Human

Updated version

http://cancerres.aacrjournals.org/content/45/7/2993

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/45/7/2993To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



alterations in the activity and isozymic profile of …...foundation, la jolla, ca. heparinized...

Documents