[CANCER RESEARCH 52, 3361-3366, June 15, 1992]

Induction of Interleukin 2 Production but not Methionine AdenosyltransferaseActivity or S-Adenosylmethionine Turnover in Jurkat T-Cells1

James De La Rosa, Arthur M. Geller, H. Leighton LeGros, Jr., and Malak Kotb2

Veterans Administration Medical Center, Research Service. Memphis, Tennessee 38104 [J. D. L. R., H L. L., M. K.]; and the Departments of Surgery, Microbiology,and Immunology ¡J.D. L. R., M. K.] and Biochemistry [A. M. G.J, University of Tennessee, Memphis, Tennessee 38163

ABSTRACT

We have recently reported that methionine adenosyltransferase(MAT) in resting human peripheral blood I -colls is primarily present inthe form of a precursor which we named X.This protein decreases uponcell stimulation, as both MAT activity and the amount of the catalytic «/a' subunits of the enzyme increase. When resting cells are activated byphytohemagglutinin, the decrease in X and increase in a/a' occurs after

interleukin 2 (IL-2) production and before DNA synthesis. The humanT-leukemia cell line, Jurkat, is unique in its ability to produce IL-2 inresponse to exogenous stimuli such as T-cell mitogens and thereforeprovides a convenient model for studying biochemical reactions involvedin T-cell activation. In this study the regulation of MAT activity and .V-adenosylmethionine (AdoMet) in resting and activated Jurkat cells wasinvestigated. Here we report that MAT activity in unstimulated Jurkatcells is about 10- and 3-fold higher than the activity in resting andactivated peripheral blood mononuclear cells, respectively. Activation ofJurkat cells with phytohemagglutinin resulted in increased IL-2 production, but not an increase in MAT activity. Identical results were obtainedusing freshly isolated cells from acute lymphoblastic leukemia patients.AdoMet utilization and pool size were approximately 3- and 10-foldhigher, respectively, in Jurkat cells compared to peripheral blood mono-nuclear cells, and both parameters were unaffected by phytohemagglutinin stimulation. Jurkat MAT was determined to be structurally indistinguishable from enzyme from T- or B-leukemia cells but was differentfrom resting, normal I -rolls in that it lacked the X form. Furthermore,unlike MAT in resting T-cells, the relative amounts of the a, a', and ß

subunits of the enzyme did not change throughout the course of IL-2induction. We conclude that AdoMet metabolism and MAT activity inJurkat cells are constitutive!}' high and that induction of IL-2 synthesis

in these cells is independent of changes in AdoMet synthesis or turnover.The lack of the X form and the difference in MAT regulation betweenleukemic T-cells and peripheral blood mononuclear cells may be exploitedin the design of specific chemotherapeutic agents.

INTRODUCTION

AdoMet3 donates methyl groups to proteins, nucleic acids,

and various lipids and donates propylamine groups for thebiosynthesis of polyamines. The importance of AdoMet inproliferating cells is underscored by the well established rolepolyamines play in proliferation, the recently discovered role ofprotein methylation in the processing of ras proteins (1,2), and

Received 11/20/91; accepted 4/7/92.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported in part by grants from the United States VeteransAdministration (M. K.), and from the NIH, GM-38530 (M. K.). The ALL-2 cellswere obtained from The Leukemia Cell Bank (LCB) at St. Jude's Children's

Research Hospital with the assistance of Dr. Fredrick Behm. The LCB issupported by the National Cancer Institute and by funds from the AmericanLebanese and Syrian Associated Charities.

2To whom requests for reprints should be addressed, at the V. A. Medical

Center (151), 1030 Jefferson Ave., Memphis, TN 38104.3The abbreviations used are: AdoMet, S-adenosy Imethionine: PBMC, periph

eral blood mononuclear cells; MAT, methionine adenosyltransferase (ATP:L-methionineS-adenosyltransferase. EC 2.5.1.6); PHA, phytohemagglutinin; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; PBS, phosphate buffered saline ( 140 mM NaCl, 2.7 mivi KCI, 1.5 mM KH2PO4, and 8.1 miviNa¡HPCX,pH 7.4); HPLC, high performance liquid chromatography; LDH,láclatedehydrogenase; IL-2, interleukin 2; PMA, phorbol myristic acetate.

the reports that DNA methylation is important in the regulationof different lymphocyte genes upon cell activation and differentiation (4-6).

Considering the importance of AdoMet in lymphocytes, it issurprising that there is very little information regarding its rateof utilization during the course of cell activation. One exceptionis the report by German et al. (7) which documented that afterstimulation of PBMC with PHA, the AdoMet pool size andrate of AdoMet utilization increased (7). Subsequent studies byKotb et al. (8) and De La Rosa et al. (9, 10) showed that uponactivation of PBMC, MAT activity also increased. More recently, we discovered that in resting human T-cells, MAT ispresent in a precursor form called X (10). Upon lymphocyteactivation X, which has low MAT activity, disappears and thea/a' subunits, which have high MAT activity, increase concom-itantly. The replacement of Aby a/a' subunits occurs after IL-

2 synthesis and before DNA synthesis (10).The human leukemic T-cell line, Jurkat, provided cell biolo

gists with an excellent tool to study events of T-cell activationin a homogeneous population of cells because of its uniqueability to produce IL-2 in response to mitogens and certainsuperantigens (11). We have used this T-cell line to study theregulation of MAT activity and AdoMet metabolism in lymphocytes. We report here that in Jurkat cells, unlike restingnormal human T-cells, activation is not accompanied by eithera change in the relative amounts of M AT subunits or an increasein MAT activity, AdoMet levels, or AdoMet turnover. AdoMetlevels and MAT activity are constitutively higher in Jurkat cellscompared to either resting or activated T-cells. The differentialregulation of MAT in normal and leukemic T-cells may in partbe due to the fact that Jurkat cells lack the X form of theenzyme.

MATERIALS AND METHODS

Materials. Aprotinin, antipain, chymotrypsin, leupeptin, pepstatinA, soy bean trypsin inhibitor, benzamidine. o-phenanthroline, phenyl-methylsulfonyl fluoride, PHA, bicinchoninic acid, Ficoll-Hypaque, andthe iodide salt of AdoMet were from Sigma Chemical Co. (St. Louis,MO). The tissue culture media and fetal bovine serum were from GibcoBRL (Gaithersburg, MD). L-["S]Methionine was purchased from DuPont New England Nuclear (Wilmington, DE), and the chemilumines-cence detection system (ECL) was from Amersham (Arlington Heights,IL). Nitrocellulose membranes and horseradish peroxidase-conjugatedgoat anti-rabbit polyclonal antibody were from Bio-Rad (Richmond,CA) and XAR-5 film was from Kodak (Rochester, NY).

Leukemic T-Cells. Freshly isolated ALL-2 cells (T-cells, midthymo-cyte) from pediatrie patients were provided by Dr. F. Behm at St. JudeChildren's Research Hospital. The Jurkat cells were from the American

Type Tissue Culture Collection (Rockville, MD). Cells were culturedin RPM1 supplemented with 10% fetal bovine serum, 2 mM L-gluta-mine, 50 ng/ml of streptomycin, and 50 units/ml of penicillin (referredto as RPMI complete). All cells were cultured at 37°Cin humidified

air containing 5% CO?.Preparation of Cell Extracts. Cell extracts were prepared for analysis

of AdoMet and L-methionine by lysing cells, precipitating protein with2 N perchloric acid, and then neutralizing the extract with KOH and

3361

Research. on August 16, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

MAT ACTIVITY AND AdoMel METABOLISM IN JURKAT CELLS

KHCOj. as described previously (7-9).MAT activity was assayed from extracts prepared by three cycles of

freeze-thawing. The enzyme extraction buffer contained 50 mM potassium yV-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid, pH7.4; 50 mM KCI; 15 mM MgCl2; 0.3 mM EDTA; and the followingproteolytic inhibitors: 0.5 mM phenylmethylsulfonyl fluoride, 1 mM o-phenanthroline; 10 mM benzamidine; 300 Mg/ml soy bean trypsininhibitor; 50 Mg/ml aprotinin; 1 Mg/ml pepstatin A; 25 Mg/ml antipain;25 Mg/ml chymostatin; and 25 Mg/ml leupeptin. If the extract was to beanalyzed by SDS-PAGE, the proteolytic inhibitors were dissolved in50 mM Tris, pH 7.4-50 mM NaCl-5 mM MgCl2.

L-Methionine and AdoMet Turnover Studies. The isotope kineticmodel described by German et al. (7) was used to determine thefractional turnover of intracellular L-methionine (k\) and AdoMet (k2).The model showed that if the specific radioactivity of extracellularmethionine, denoted by MI,is known, then the time-dependent changein the specific radioactivity of intracellular methionine, denoted by M2.in a culture of cells can be described by Equation A. Similarly, the time-dependent change in the specific radioactivity of AdoMet, denoted byMS,can be described by Equation B.

I. *2g-*"

MJ = Ml M - . _ .k2-k.

(A)

(B)

In these experiments MI,m, and MJwere determined empirically overtime so that the turnover constants of k¡and A2could then be determined graphically by iteration. In most experiments, the specific radioactivity of intracellular L-methionine (MZ)and AdoMet (MJ)approachedasymptotes which ranged from 80 to 99% of the value for extracellularmethionine (MI).A similar observation in PBMC was made previously(7) and suggests the presence of metabolically inaccessible pools ofAdoMet and methionine. It was these asymptote values which wereused in modeling to obtain A, and Ac2.

The specific activity of intracellular and extracellular L-methioninewas determined after adding 2.5 to 3.0 MCiof L-["S]methionine to snaptop vials containing a culture of IO7cells in 1 ml of RPMI complete.The vials were kept in a 37°Cwater bath and at specific times, within

90 s, 0.8 ml of culture was removed and diluted into 50 ml of ice coldPBS. Both the vial and the conical centrifuge tube containing thediluted cells were then centrifuged for 1 min at 1800 x g. The mediumin the vial was collected and used to determine the specific radioactivityof extracellular L-methionine. The cells in the conical tube were washedbriefly with ice cold PBS, frozen in liquid nitrogen, and stored at -70°C

until extracted (as described above) for intracellular methionine. Endogenous AdoMet was removed from the cell extract by adsorption toacid washed charcoal, which was added to the extract at 65 Mg/106cells. Both extracellular and intracellular L-methionine in the extractswere converted enzymatically to AdoMet. To convert methionine toAdoMet, extract was incubated with 100 mM Tris (pH 8.3), 100 mMKCI, 20 mM MgCl2, 2.5 mM ATP, 2 mM dithiothreitol, 150 Mg/ml ofbovine serum albumin, and 10 Mg/ml of Escherichia coli MAT for l hat 37°C.Following this conversion, a cell extract was made as described

above. AdoMet was then separated from other radiolabeled compoundsby HPLC using a Whatman SCX strong cation exchange column (0.46x 25 cm) described by German et al. (7), with the modification that themobile phase was 0.2 M ammonium formate, pH 4, containing 10%acetonitrile. (-)AdoMet was purified chromatographically (12) andused as the HPLC standard. The radioactivity in the AdoMet-peakfrom each HPLC run was measured by liquid scintillation counting.The specific radioactivity of AdoMet was then calculated as the ratioof cpm to nmol of AdoMet.

The equilibration of radioactivity into AdoMet produced by the cellswas determined similarly. Jurkat cells, at a cell density of 2 x IO6cells/

ml, were incubated with radiolabeled methionine at 3 nC\/m\ for up to2 h. At specific times 3 x IO6 cells were harvested, centrifuged for 1

min at 1800 x g, washed with PBS, and frozen as a cell pellet in liquidnitrogen until extracted. Cell extracts were made as described above,

and the specific radioactivity of the AdoMet was determined by HPLCand liquid scintillation counting.

Other Methods, Activity Units, and Kinetics. MAT and LDH wereassayed by the methods of Kotb and Kredich (12) and Bergmeyer andBern t (13), respectively. Protein was determined spectrophotometri-cally by the bicinchoninic acid method (14), and IL-2 was determinedby the CTLL bioassay ( 15). Ten units of IL-2 is defined as that amountwhich will stimulate a proliferative response of the CTLL cell line to50% of the maximum obtained with a standard preparation of IL-2.Proliferation was determined by the incorporation of pHjthymidineinto precipitable radioactivity.

One unit (U) of MAT activity is that amount of enzyme which willcatalyze the formation of 1 nmol of AdoMet in 1 h, whereas 1 unit ofLDH is that amount of enzyme which will catalyze the formation of 1Mmol of lactate in 1 min.

SDS-PAGE and Western Blotting. Samples were dissolved in loadingbuffer (60 mM Tris-Cl, pH 6.8; 2% SDS; 5% 2-mercaptoethanol; and5% glycerol) and heated in a boiling water bath for 3-4 min. The SDS-PAGE resolving gel consisted of 10% acrylamide and 0.27% bisacry-lamide. Electrophoresis was done at 15-20 V/cm with cooling. Thecharacterization of antibodies produced to human lymphocyte MAThas been published recently (16). Secondary antibody was horseradishperoxidase-conjugated goat anti-rabbit polyclonal antibody. The transfer for Western blotting was done for 1 h at 25 to 30 V/cm with cooling.The blots were blocked overnight with 6% nonfat milk in 50 mM Tris,pH 7.5-150 mM NaCI. After incubating the blot with primary andsecondary antibody, immunoreactive proteins were detected with theluminol-chemiluminescence system (ECL) from Amersham. The processed blot was sandwiched between two overhead transparencies andexposed to XA-R film for 10 to 60 s. The limit of detection under theseconditions was about 2 ng for each of the human lymphocyte MATsubunits.

RESULTS

The Jurkat human T-cell line was studied for two reasons:(a) T-cells respond to PHA and are responsible for the changesobserved in AdoMet metabolism and MAT activity in PHA-activated PBMC; (b) although Jurkat is a transformed cell linewhich proliferates continuously, it is considered a relativelyresting T-cell line because, unlike other T-cell lines, it can beinduced to secrete IL-2. Therefore, it was reasoned that becauseJurkat cells can be activated, they might serve as a model tostudy the role of AdoMet metabolism in the induction of IL-2synthesis.

Induction of IL-2 Synthesis in Jurkat Cells Is Not Accompanied by an Increase in MAT Activity. Jurkat cells were incubated with different concentrations of PHA to determine theoptimal concentration for cell activation, as indicated by IL-2production, and for the induction of MAT activity. This doseresponse experiment demonstrated that IL-2 production by theJurkat cells increased concomitantly with increasing PHA concentration, reaching a maximum at 2.5-5 Mg/ml (Fig. 1/4).However, MAT activity did not increase upon Jurkat activation,as was previously found for PBMC, and in fact appeared todecrease (Fig. IB). LDH, which is a noninducible enzyme, wasused as a negative control, and its activity remained relativelyconstant upon activation (Fig. 1C).

Freshly Isolated Leukemic T-Cells Behave Like Jurkat Cells.Next we asked whether the lack of induction of MAT activityrepresented an idiosyncratic behavior of the Jurkat clone or ageneral characteristic of leukemic T-cells. Freshly isolatedPBMC from patients with ALL-2 were incubated for varioustimes with optimal mitogenic concentrations of PMA plusionomycin and then tested for IL-2 production and MATactivity. PMA and ionomycin were used instead of PHA be-

3362

Research. on August 16, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

MAT ACTIVITY AND AdoMet METABOLISM IN JURKAT CELLS

20

50e

S 40

I 30

<

0.0 1.0 2.0 3.0 4.0 5.0

PHA Concentration, ng/ml

0.0 I 0 2.0 3.0 4.0 5.0

PI IA Concentration, ng/ml

1.0 2.0 3.0 4.0

PHA Concentration, (Jg/ml

5.0

Fig. 1. PHA dose response of IL-2 production (A). MAT activity (B). andLDH activity (C). Jurkat cells at a density of 4.0 x 10' cells/ml were incubatedwith different concentrations of PHA and the cells were harvested after either 10h (O) or 25 h (•).The media were assayed for IL-2 and the cell extracts wereassayed for both MAT and LDH activity.

Table 1 Induction of IL-2 synthesis but not MAT activity in freshly isolated ALL-2 cells

ALL-2 cells were cultured for 24 h at 2 x 106/rnl ¡nthe presence or absence

of 50 ng/ml phorbol myristate acetate and 0.5 ^M ionomycin. These concentrations were determined in a separate experiment to be optimal for IL-2 production.After 24 h the cells »ereseparated from the culture media by centrifugation. Theculture media were assayed for IL-2 activity, and the cell pellet was extracted andassayed for MAT activity and protein concentration as described in "Materialsand Methods."

Incubation time(h)Zero

time24(-)24(+)IL-2

activity(units/ml)3.6

118MAT

activity(units/mgprotein)11.3

10.312.6

MAT Subunit Composition in Resting and Activated JurkatCells. Recent data from our laboratory revealed the presence ofa precursor of MAT in resting T-cells. This M, 68,000 precursor(X) which has low MAT activity disappears when T-cells arestimulated with PHA while the «and «'subunits of the enzyme

which have high MAT activity increase. The amount of the ßsubunit, on the other hand, remained constant throughout thecourse of stimulation. These changes in MAT subunits alwaysoccurred after IL-2 production and before DNA synthesis. Itwas of interest, therefore, to determine whether the inductionof IL-2 production by Jurkat cells is also accompanied by achange in the relative amounts of MAT subunits. Cell extractswere prepared from resting and 24-h PHA-stimulated Jurkatcells and analyzed in Western blots for MAT subunits. Incontrast to normal T-cells, Jurkat cells as well as ALL-2 cellslacked the Xform and only had the a, a', and ßsubunits (data

not shown). Furthermore, when the cell extracts from the 24-htime point of the dose response experiment in Fig. 1 wereanalyzed by Western blotting, there was no change in therelative amounts of the different MAT «,a', and ßsubunits

(Fig. 2). There was a slight but insignificant increase in the ßsubunit.

AdoMet Utilization in Resting and Activated Jurkat Cells.Although the above experiments showed that MAT activity didnot increase upon activation of Jurkat cells, this did not preclude the possibility that AdoMet utilization or pool size couldchange. To estimate the absolute rate of AdoMet utilization instimulated and unstimulated Jurkat cells, the equilibration ofextracellular radiolabeled methionine with intracellular methi-onine and AdoMet was determined. The time course of thisequilibration is shown in Fig. 3. In both stimulated and unstimulated Jurkat cells methionine specific radioactivity reachedhalf-maximum in about 15 s (Fig. 3/1). The equilibration ofradiolabel with AdoMet reached half-maximum within 24 min.Both the specific radioactivity of methionine and AdoMet approached an asymptote between 80 and 99% of the specificactivity of extracellular methionine.

The fractional turnover of intracellular methionine (k,) andAdoMet (&:) was determined graphically from these equilibration curves using Equations A and B under "Materials andMethods," and these values are listed in Table 2. Included in

Table 2 are the values from stimulated and unstimulated PBMCreported previously (7). The fractional turnover of L-methioninewas similar in both stimulated and unstimulated Jurkat cellsranging from 2.5 to 4.0 min"1, which is similar to that forPBMC (2.5 min"1; Table 2). Upon cell activation, the AdoMet

a,a1

-ß

cause a number of these cells showed aberrant CD3 expressionwhich can interfere with PHA responsiveness. As shown inTable 1, when freshly isolated T-leukemic cells were stimulatedwith PMA and ionomycin, the IL-2 production increased fromless than 4 units/ml to greater than 100 units/ml after a 24-hincubation. However, as the case with Jurkat cells, MAT activity remained essentially the same.

0 0.25 1.0 2.5 5.0 std

PHA concentration, jig/mlFig. 2. Western blot analysis of Jurkat cell extracts stimulated with different

concentrations of PHA. The first five lanes correspond to the 25-h samples fromFig. 1. and the last lane is a MAT standard (std) purified from chronic lymphocyticleukemia cells. The blot was probed with anti-holoenzyme antisera, and processedas described in "Materials and Methods."

3363

Research. on August 16, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

MAT ACTIVITY AND AdoMel METABOLISM IN JURKAT CELLS

1.0

0.8

0.6

0.4

0.2

0.00 0.25 0.50 0.75 1.00 1.25 1.50Time, min

1.0

0.8

0.6

0.4

0.2

0.050 75 100Time, min

Fig. 3. Time course of equilibration of extracellular i.-["S]mcthionine withintracellular i.-methionine and AdoMet in unstimulated and stimulated Jurkat T-cells. After the addition of radiolabeled methionine, cells were harvested at theindicated times and the specific radioactivities of extracellular methionine (n¡),intracellular methionine (n¡).and AdoMet (n,) was determined as described in"Materials and Methods." Equilibration as defined here is the MI^I ratio for i.-methionine and M.VMIfor AdoMet. A, equilibration of radiolabel into I.-methioninein unstimulated (D) and stimulated cells (•).B, equilibration of radiolabel intoAdoMet in unstimulated (O) and stimulated cells (•).The open and closed symbolsrepresent actual data points, while the curves represent theoretical cunes modeledto fit the data using Equations A and B in "Materials and Methods." In I the

dashed line is to data from unstimulated cells modeled to 85% full equilibrationresulting in A, = 3.0 min "', and the solid line is to data from stimulated cellsmodeled to 99% full equilibration resulting in A, = 2.5 min ~*. In B the dashedline for data from unstimulated cells using k, = 3 min "', modeled to 95% fullequilibration resulting in A2= 0.035 min "'; and the solid line is to data fromstimulated cells using A. = 2. modeled to 95% full equilibration and resulting inAj = 0.025 min ~'.

Table 2 Comparison of AdoMel turnover, pool si:e, utilisation, and MAT activityin PBMC

Unstimulated (-PHA) and stimulated (+PHA) PBMC and Jurkat cells wereincubated with radiolabeled methionine to determine the fractional turnover ofu-methionine (A,) and AdoMet (Aj). The AdoMet content of the cells wasmeasured by HPLC. and MAT activity was determined by a radiochemical assay.One unit of MAT activity catalyzes the formation of 1 nmol of AdoMet/h.

PBMCA,

(min"')A2(min~')AdoMet

pool sizepmol/ 10' cells

jimol/min*AdoMet

utilizationpmol/min/10' cells

¿imol/minMAT

activity (units/mg protein)-PHA2.5

0.029-0.06°1.4-3.4

3.3-8.10.05-0.17

0.12-0.42-5f+PHA2.5

0.04-0.09117-23

25-330.97-1.08

1.4-1.58-15Jurkal-PHA2.7-4.0

0.025-0.036140

1843.5-5.0

4.6-6.630-40cells+PHA2.5-3.5

0.022-0.03140

1843.1-4.2

4.1-5.525-35

°Turnover estimates for PBMC arc from report of German et al. (7).* AdoMet pool size (nmol/ml cell volume) and AdoMet utilization (nmol/

min/ml cell volume) were calculated based on cell volume for stimulated andunstimulated PBMC of 0.42 ml/10' cells and 0.69 ml/10' cells estimate by

German et al. (7). The cell volume for (he clone of Jurkat cells used in thesestudies was calculated to be 0.76 ml/106 cells.

' MAT activities for PBMC are from De La Rosa et ai (9).

fractional turnover increased in PBMC by 38-200%; however,in Jurkat cells, AdoMet turnover was not significantly affected(Table 2). After activation of PBMC, the pool size of AdoMetalso increased 4-10 fold compared to the unactivated cells(Table 2). On the other hand, the pool size of AdoMet in Jurkatcells remained constant at 140 pmol/106 cells, regardless of

whether the cells were activated or not (Table 2). Although thefractional turnover of AdoMet in Jurkat cells was slightly lessthan that of PBMC, the fact thai the AdoMet pool size inJurkat cells ( 184 ^M) was at least 23 times that of unstimulatedPBMC (3-8 MM)indicated that AdoMet utilization in Jurkatcells was higher than in PBMC. In fact, AdoMet utilization inJurkat cells was 5-7 ^M/min compared to a value of 0.12-0.4and 1.5 /¿M/minin unstimulated and stimulated PBMC, respectively (Table 2).

Correlation between Predicted and in Situ MAT Activity inResting PBMC versus Stimulated PBMC and Jurkat Cells.Using the steady state kinetic model of Kotb and Kredich (17)describing the MAT catalyzed reaction, we predicted the MATactivity in both PBMC and in Jurkat cells (Table 3). Enzymeactivity can be predicted as a fraction of the Vmaxif the concentrations of the substrates and products of the reaction areknown. The Vmaxitself was estimated empirically by assaying

Table 3 Predicted MAT reaction velocity and predicted MAT activity in PBMCand Jurkat T-cells

The MAT activity of unstimulated (-PHA) and stimulated (+PHA) PBMCand Jurkat cells was predicted using a previously published steady state kineticmodel describing the enzyme activity (16).

PBMC

-PHA +PHA

Jurkat cells

-PHA +PHA

Predicted reaction velocityexpressed as a fraction ofVm„°

Predicted MAT activity*(pmol/min/10'cells)

In situ MAT activity'(pmol/min/10'cells)

0.49 0.28 0.09 0.09

0.36-0.49 0.56-0.76 2.1-2.7 1.7-2.4

0.05-0.17 0.97-1.08 3.5-5.0 3.1-4.2

a The predicted reaction velocity for PBMC was estimated using the steady

state kinetic model of Kotb and Kredich (16) shown below, where K, is Vm„,v isthe initial velocity. A is the ATP concentration. B is the i -methionine concentration, P is the AdoMet concentration, Q is the pyrophosphate concentration, andR is the phosphate concentration.

y,AB(\ + dQ + gQR)

A'„Ä+ AB + C„P+ C„AP(—T-H\ A,»/

„¿�/l+ WA',0A»

+ c„

C,rQRI + Kw\d + C„,PQR

The values obtained from the report of De La Rosa et al. (9) of 5 UMpyrophosphate. 5 HIMphosphate. 1.5 HIMATP, 20 /*Mmethionine, and 5 or 30 nM AdoMet»ereused in solving for v. The same values were used for predicting i- Jurkatcells, except that methionine was 100 >iM(concentration in RPMI), and AdoMetwas calculated to be 184 ^M, based on a measured cell volume of 0.76 ml/10'cells.

* This value is the product of the predicted reaction velocity (fraction of V„„)and the enzyme activity assayed under conditions of near-substrate saturation,after it was corrected to 100% activity under total substrate saturation. Unstimulated and 24-h-stimulated PBMC have activities of 0.73-1.0. and 2.0-2.7 pmol/min/10' cells, respectively. Unstimulated and stimulated Jurkat cells have activities of 23-30 and 19-26 pmol/min/10* cells, respectively.

' The in situ MAT activity is the rate of AdoMet utilization (listed in Table 2),

since under steady state conditions the rate of AdoMet utilization should equalits rate of svnthesis.

3364

Research. on August 16, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

MAT ACTIVITY AND AdoMel METABOLISM IN JURKAT CELLS

the enzyme under conditions of near-substrate saturation. Thus,

the predicted MAT activity in situ is simply a product of thefraction of Vnux and the assayed MAT activity (which wascorrected for 100% substrate saturation). Under steady stateconditions it is expected that the rate of AdoMet utilizationwould equal the rate of input, i.e., the in situ MAT activity. Inunstimulated PBMC, the in situ MAT activity was 7-3 foldlower that the predicted value. In contrast, in situ MAT activityin stimulated PBMC, as well as in Jurkat cells, was 2-foldhigher than the predicted MAT activity (Table 3).

DISCUSSION

Abnormalities in methionine metabolism have been reportedin a variety of cancer cells (18-21). Whether these changes area consequence or cause of malignant transformation is notclear. However, inhibitors of AdoMet synthesis or inhibitors ofDNA methylation have been shown to induce the differentiationof several transformed cell lines (22, 23). As a result of thesereports, several laboratories have dedicated their efforts tostudying the metabolism of AdoMet and designing specificinhibitors of its metabolism for use as chemotherapeutic agents(24-28). We are particularly interested in the role of AdoMetin the activation and differentiation of human T-lymphocytesand therefore have engaged in studies of the regulation of MATactivity in normal resting, activated, and transformed T-cells.Although several studies have compared MAT activity andAdoMet turnover in normal and malignant cell lines (21, 29,30) the data are rather confusing. In some cases there was nodifference in AdoMet metabolism, while in others, MAT specific activity was much higher in the transformed cells. We havefound that both immortal T- and B-cell lines have much higherMAT specific activity than activated normal T- or B-lympho-cytes or freshly isolated leukemic cells. However, MAT specificactivity varied depending on the lineage of the leukemic cells.4

The Jurkat cell line has been used in many studies of T-cellactivation because unlike other transformed T-cell lines, thisline is able to respond to mitogenic stimuli by producing highlevels of IL-2. These Jurkat cells, as shown in this study had aresting level of MAT activity of 30-40 units/mg protein andAdoMet utilization rate was 20-55 times that of unstimulatedPBMC, and 3-5 times that of stimulated PBMC. Activation ofJurkat cells by PHA had no effect on either MAT activity orAdoMet turnover. This was not unique to this cell line becausefreshly isolated ALL-2 cells when stimulated to produce IL-2behaved the same way. These results are similar to those obtained by Chiba et al. (31 ) with the promyelocytic cell line, HL-60 cells. These investigators showed that in these transformedcells MAT activity as well as the synthesis and pool size ofAdoMet and S-adenosylhomocysteine were cell cycle independent. The data with Jurkat cells and HL-60 cells are quitedifferent from that observed with resting PBMC, where activation triggers marked increase in both MAT activity, AdoMetpool size, and AdoMet turnover (7-9).

In PBMC, the change in AdoMet turnover is a relativelyearly event, occurring within 3-5 h of stimulation (7). Thissuggests that the increase in AdoMet level and turnover is notinvolved in early signal transduction but may be important inthe initial stages of programming of cell activation. The increasein MAT activity in resting T-cells appear to follow the transformation of the less active \ precursor form to the more active

4J. De La Rosa, H. L. LeGros, and M. Kolb. manuscript in preparation.

a/a' form (10). Although this process is detected by 24 h of

cell activation, it reaches a maximum at 48 h following IL-2production and preceding DNA synthesis. It was not clear fromour study whether the change in these subunits is a result orprerequisite of the cell cycle Gìto S transition. As shown inthis study Jurkat and ALL-2 cells lack the X form, and theenzyme is constitutively present in the highly catalytic form a/«'.This may explain the lack of induction of enzyme activity

despite induction of high levels of IL-2. Again the issue ofwhether this constitutively activated AdoMet metabolism is acause or result of the process of malignant transformationrequires further experimentation.

Clearly intact AdoMet metabolism is essential for normal T-cell proliferation. Previously, we demonstrated that inhibitorsof AdoMet synthesis block T-cell blastogenesis in response tomitogenic and superantigenic stimulation (8). This is consistentwith clinical and experimental reports that lymphocytic cellsdepend on AdoMet mediated reactions for immune function(32-35), which is not surprising since these cells undergo rapiddifferentiation. Indeed, the role of DNA methylation and theregulation of the expression of certain genes in lymphocyteshas been shown in several studies (4-6). A comparison betweenthe predicted and in situ MAT activity in resting and stimulatedT-cells showed some interesting correlations. In unstimulatedcells the in situ MAT activity was lower than the predictedactivity, but in stimulated T-cells and Jurkat cells the situationwas reversed. These observations can be interpreted in severalways: (a) Resting T-cells may have a less active form of MATcompared to stimulated cells; (b) an inhibitor of MAT may bepresent in resting but not in stimulated T-cells; (c) activationof T-cells may be accompanied by induction of a MAT activa-

tor(s). None of these possibilities are mutually exclusive, andfurther studies should reveal the mechanism of MAT regulation. However, the finding in this study that the differences inAdoMet metabolism are at least in part related to the presenceof different forms of MAT in quiescent and actively dividingT-cells raises the possibility that this may be a feature that canbe exploited for chemotherapy and immunomodulation.

Previous reports by Liau et al. (36) showed that tumor cellsmay have a different form of MAT. Such differences betweennormal and transformed cells provide the impetus for a morerigorous search for selective inhibitors of the various forms ofMAT that can be of potential benefit in chemotherapy. Ofparticular interest in this regard is the report by Sufrin andLombardini (37) which showed, based on studies with severalsubstrates and inhibitors of MAT, that there are differences inthe active-site of tumor versus normal isozymes of MAT. Itmay be possible, therefore, to synthesize inhibitors specific forthe a/«' but not the X form of MAT which can be used to

target actively dividing T-cells without affecting normal T-cells.

REFERENCES

1. Clarke, S., Vogel. J. P.. Deschenes. R. J.. and Stock. J. Posttranslationalmodification of the Ha-roi oncogene protein: evidence for a third class ofprotein carboxyl methyltransferases. Proc. Nati. Acad. Sci. USA, 85: 4643-4647. 1988.

2. Gutierrez, L., Magee, A. I., Marshall. C. J., and Hancock, J. F. Posttranslational processing of p21 ras is two-step and involves carboxyl-methyl-ation and carboxy-terminal proteolysis. EMBO J., 8: 1093-1098. 1989.

3. Hancock. J. F.. Cadwalladen. K., and Marshall. C. J. Methylation andproteolysis are essential for efficient membrane binding of prenylated p21K""". EMBOJ.. 10: 641- 646. 1991.

4. Scyfert. V. L., McMahon. S. B.. Glen. W. D.. Yellen, A. J.. Sukhatme, V.P., Cao, X.. and Monroe. J. G. Methylation of an immediate-early induciblegene as a mechanism for B cell tolerance induction. Science (WashingtonDC), 250: 797-800. 1990.

3365

Research. on August 16, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

MAT ACTIVITY AND AdoMct METABOLISM IN JURKAT CELLS

5. Sakamoto, S., Ortaldo, J. R., and Young, H. A. Analysis of the methylationstate of the T cell receptor ßchain gene in T cells and large granularlymphocytes. J. Biol. Chem.. 264: 251-258. 1989.

6. Carbone, A. M., Marrack. P.. and Kappler, J. W. Demethylated CDS genein CD4* T cells suggests that CD4* cells develop from CD8* precursors.Science (Washington DC), 242: 1174-1176, 1988.

7. German, D. C., Bloch. C. A., and and Kredich. N. M. Measurements of S-adenosylmethionine and L-homocysteine metabolism in cultured humanlymphoid cells. J. Biol. Chem.. 258: 10997-11003, 1983.

8. Kotb, M., Dale, J. B., and Beachey, E. H. Stimulation of S-adenosylmethion-ine synthetase in human lymphocytes by streptococcal M protein. J Immu-no!., 139: 202-206, 1987.

9. De La Rosa, J., Kotb, M., and Kredich, N. M. Regulation of S-adenosylme-thionine synthetase activity in cultured human lymphocytes. Biochim. Bio-phys. Acta, 1077: 225-232, 1991.

10. De La Rosa. J.. LeGros, H.. Geller, A. M., and Kotb, M. Changes in therelative amount of subunits of methionine adenosyltransferase in humanlymphocytes upon stimulation with a polyclonal T cell mitogen. J. Biol.Chem., 267: 10699-10704. 1992.

11. Kotb, M., Majumdar, G., Tomai, M., and Beachey, E. H. Accessory cell-independent stimulation of human T cells with superantigens. J. Immun»]..145: 1332-1336, 1990.

12. Kotb, M., and Kredich, N. M.S-Adenosylmethionine synthetase from humanlymphocytes. Purification and characterization. J. Biol. Chem., 260: 3923-3930, 1985.

13. Bergmeyer, H. U., and Bernt, E. Láclate dehydrogenase U-V assay withpyruvate and NADH. In: H. U. Bergmeyer (ed.). Methods of EnzymaticAnalysis, Vol. 2, pp. 574-579. New York: Academic Press, 1974.

14. Smith, P. K., Krohn, I., Hermanson, G. T., Mallia, A. K.. Gartner, F. H.,Provenzano. M. D., Fujimoto, E. K., Goeke, N. M., Olson. B. J., and Klenk,D. C. Measurement of protein using bicinchoninic acid. Anal. Biochem.. 750:76-85, 1985.

15. Gillis, S., Perm. M. M.. Ou, W., and Smith, K. A. T-Cell growth factor:parameters of production and a quantitative microassay for activity. J.Immunol., 120: 2027-2032. 1978.

16. Kotb, M., Geller, A. M., Markham, G. D.. Kredich, N. M., De La Rosa, J.,and Beachey. E. H. Antigentic conservation of primary structural regions ofS-adenosylmethionine synthetase. Biochim. Biophys. Acta, I040: 137-144,1990.

17. Kotb, M., and Kredich, N. M. Regulation of of human lymphocyte 5-adenosylmethionine synthetase by product inhibition. Biochim. Biophys.Acta, I039: 253-260, 1990.

18. Jacobsen, S. J., Hoffman. R. M.. and Erbe, R. W. Regulation of methionineadenosyltransferase in normal diploid and simian virus 40-transformed human fibroblasts. J. Nati. Cancer Inst., 65: 1237-1244, 1980.

19. Coalson, D. W., Mecham. J. O., Stern, P. H., and Hoffman, R. M. Reducedavailability of endogenously synthesized methionine for S-adenosylmethion-ine formation in methionine-dependent cancer cells. Proc. Nati. Acad. Sci.USA, 79:4248-4251, 1982.

20. Stern, P. H., Mecham. J. O.. Wallace. C. D.. and Hoffman. R. M. Reducedfree-methionine in methionine-dependent SV40-transformed human fibroblasts synthesizing apparently normal amounts of methionine. J. Cell. Phys-iol., 7/7:9-14, 1983.

21. Hoffman, R. M. Altered methionine metabolism, DNA methylation andoncogene expression in carcinogcnesis. Biochim. Biophys. Acta, 783:49-87.1984.

22. Mendelson. N.. Michi, J.. Gilbert!, H. S., Acs, G., and Christman, J. K. L-Ethionine as an inducer of differentiation in human promyelocytic leukemiacells (HL-60). Cancer Res., 40: 3306-3310, 1980.

23. Creusot, F., Acs, G., and Christman, J. K. Inhibition of DNA methyltrans-ferase and induction of friend erythroleukemia cell differentiation by 5-azacytidine and 5-aza-2'-deoxycytidine. J. Biol. Chem., 257: 2041-2046,

1982.24. Kappler, F., Hai, T. T.. and Hampton, A. Isoenzyme-specific enzyme inhib

itors. 10. Adenosine S'-triphosphate derivatives as substrates or inhibitors of

methionine adenolsyltransferase of rat normal and hepatoma tissues. J. Med.Chem., 29:318-322. 1986.

25. Kappler. F.. Vrudhula. V. M., and Hampton. A. Toward the synthesis ofisozyme-specific enzyme inhibitors. Potent inhibitors of rat methionine ad-enosyltransferases. Effect of one-atom elongation of the ribose-P a bridge intwo covalent adducts of L-methionine and /i.-y-imido-ATP. J. Med. Chem.,37:384-389, 1988.

26. Kramer, D. L., Sufrin, J. R., and Porter, C. W. Relative effects of 5-adenosylmethionine depletion on nucleic acid methylation and polyaminebiosynthesis. Biochem. J.. 247: 259-265. 1987.

27. Sufrin, J. R., Lombardini, J. B.. Kramer, D. L., Alks, V., Bernaki, R. J., andPorter, C. W. Methionine analog inhibitors of S-adenosylmethionine biosynthesis as potential antitumor agents.. In: R. T. Borchardt, C. R. Creveling,and P. M. Ueland (eds.). Biological Methylation and Drug Design, pp. 373-384. New Jersey: Humana Press. 1986.

28. Sufrin, J. R., Lombardini, J. B., and Keith, D. D. A potent inhibitor of theenzymatic synthesis of S-adenosylmethionine. Biochem. Biophys. Res. Commun., 706:251-255. 1982.

29. Oden, K. L., Carson. K., Mecham, J. O., Hoffman, R. M., and Clarke, S. 5-Adenosylmethionine synthetase in cultured normal and oncogenically-trans-formed human and rat cells. Biochim. Biophys. Acta. 760: 270-277, 1983.

30. Liau, M. C., Lin, G. W., and Huribert, R. B. Partial purification andcharacterization of tumor and liver .S-adenosylmethionine synthetases. Cancer Res., 37: 427-435. 1977.

31. Chiba, P., Wallner, C., and Kaiser, E. S-Adenosylmethionine metabolism inHL-60 cells: effect of cell cycle and differentiaion. Biochim. Biophys. Acta,977:38-45, 1988.

32. Pike, M. C., Kredich, N. M., and Snyderman, R. Requirement of S-adenosyl-L-methionine-mediated methylation for human monocyte chemotaxis. Proc.Nati. Acad. Sci. USA, 75: 3928-3932, 1978.

33. Johnson, J., and Kredich, N. M. Inhibition of methylation by adenosine inadenosine deaminase-inhibited, phytohemagglutinin-stimulated human lymphocytes. J. Immunol., 123: 97-103, 1979.

34. Kredich. N. M.. and Hershfield. M. S. 5-Adenosylhomocysteine toxicity innormal and adenosine kinase-deficient lymphoblasts of human origin. Proc.Nati. Acad. Sci. USA, 76: 2450-2454, 1979.

35. Kredich, N. M.. and Hershfield. M. S. Immunodeficiency diseases caused byadenosine deaminase deficiency and purine nucleoside phosphorylase deficiency. In:: C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle (eds.), TheMetabolic Basis of Inherited Disease, Vol. 1. pp. 1045-1076. New York:McGraw-Hill, 1989.

36. Liau. M. C.. Chang. C. F.. and Becker, F. F. Alteration of .S-adenosylmethionine synthetases during chemical hepatocarcinogenesis and in resulting carcinomas. Cancer Res., 39: 2113-2119, 1979.

37. Sufrin, J. R., and Lombardini. J. B. Differences in the active-site region oftumor versus normal isozymes of mammalian ATP:t-methionine 5-adeno-syltransferase. Mol. Pharmacol., 22: 752-759, 1982.

3366

Research. on August 16, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from

1992;52:3361-3366. Cancer Res   James De La Rosa, Arthur M. Geller, H. Leighton LeGros, Jr., et al.   Turnover in Jurkat T-Cells

-AdenosylmethionineSAdenosyltransferase Activity or Induction of Interleukin 2 Production but not Methionine

  Updated version

  http://cancerres.aacrjournals.org/content/52/12/3361

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/52/12/3361To request permission to re-use all or part of this article, use this link

Research. on August 16, 2021. © 1992 American Association for Cancercancerres.aacrjournals.org Downloaded from