[CANCER RESEARCH 49, 158-163, January 1, 1989]
Inhibition of Glycinamide Ribonucleotide Formyltransferase and Other Folate
Enzymes by Homofolate Polyglutamates in Human Lymphoma and Murine
Leukemia Cell Extracts1
J. Thorndike, Y. Gaumont, R. L. Kisliuk,2 F. M. Sirotnak, B. R. Murthy, M. G. Nair, and J. R. Piper
Department of Biochemistry, Tufts University Health Science Campus, Boston, Massachusetts 02111 [J. T., Y. G., R. L. K.]; Memorial Sloan-Kettering Cancer Center,
New York, New York 10021 [F. M. S.]; Department of Biochemistry, University of South Alabama, Mobile, Alabama 36688 [B. R. M., M. G. NJ; and Drug Synthesis
Section, Southern Research Institute, Birmingham, Alabama 35255 [J. R. P.]
ABSTRACT
In order to determine the biochemical basis for the cytotoxicity of
homofolates, poly--y-glutamyl derivatives of homofolate (HPteGlu) and
tetrahydrohomofolate (H4HPteGlu) were synthesized and tested as in
hibitors of glycinamide ribonucleotide formyltransferase (GARFT), aniinoimidazolecarboxamide ribonucleotide formyltransferase (AICARFT),
ih\mid)l:iu- synthase, and serine hydroxymethyltransferase (SHMT) in
extracts of Manca human lymphoma and 1.12111murine leukemia cells.
The most striking inhibitions are that of GARFT by (6R,S)-H4HI'tc
(.lu., ,, with ICso values from 1.3 to 0.3 MM.Both diastereomers, (6R)H4HPteGlu«and (6S>H4HPteGlu«,inhibit GARFT activity. In Manca
cell extracts, the (6S) form is more potent than the (6R) form whereas
in the murine system the reverse is true. The (6K,S)-II4IIlVGIu polyglutamates are weak inhibitors of human AICARFT (ICso, 6-10 AIM).
Polyglutamates of HPteGlu, however, are more inhibitory to AICARFT,
with HPteGliu-6 having ICso values close to 2 MM.Polyglutamates of
HPteGlu and of H4HPteGlu are weaker inhibitors of thymidylate syn
thase (ICso, 8 MM for HPteGlus-6 and >20 MM for H4HPteGlu,.5).
Polyglutamates of HPteGlu and of H<HPteGlu are poor inhibitors of
SHMT (ICso, >20 MM).Manca cell growth is inhibited 50% by HPteGlu
and (6R,S)-5-methyl-H4HPteGlu at 6 and 8 MM,respectively. Both of
these effects are reversed by 0.1 HIMinosine. Trimetrexate at a subinhibitory concentration, 10 IIM,antagonizes growth inhibition by HPteGlu,
raising the ICso from 6 to 64 MM,but enhances inhibition by (6R,S)-5methyl-H4HPteGlu, lowering the ICsofrom 8 to S MM.Our results support
the view that homofolates become toxic after conversion to II4IIPtcGlu
Polyglutamates which block GARFT, a step in purine biosynthesis.
INTRODUCTION
Homofolate differs from folate only in having an additional
méthylène
group between the C-9 and N-10 positions (Fig. 1).
Homofolates increase the survival time of LI 210 leukemic mice
(1,2). The ani itumor activity of homofolates and methotrexate
differ in that (a) homofolates do not inhibit DHF3 reducÃ-ase
(1, 3) and (b) homofolates show enhanced antitumor activity
against a strain of I 1210 that is resistant to the DHF reducÃ-ase
inhibitor methotrexate (1, 2). Homofolates inhibit growth of
mammalian tumor cell lines in culture (1, 4-7), an effect that
is antagonized by added purines (4, 6, 7). The synthesis of
Received 6/24/88; revised 9/20/88; accepted 10/5/88.
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 in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by NCI Grants CA 10914 (R. L. K.), CA 22764,
CA 08748, and CA 18856 (F. M. S.), CA 32687 (M. G. N.), and CA 25236 (J.
R. P.) from the National Cancer Institute, USPHS/Department of Health and
Human Services.
2 To whom requests for reprints should be addressed, at Department of
Biochemistry, Tufts University Health Science Campus, Boston MA 02111.
3The abbreviations used are: DHF, dihydrofolate; HPteGlu, homofolate;
H<HPteGlu, tetrahydrohomofolate: TMTX, trimetrexate; GAR, glycinamide ri
bonucleotide; GARFT, glycinamide ribonucleotide formyltransferase; AICARFT,
aminoimidazolecarboxamide ribonucleotide formyltransferase; TS, thymidylate
synthase; SHMT, serine hydroxymethyltransferase; MTT, 3-(4.5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; IC50,the inhibitor concentration caus
ing 50% inhibition; mu, milliunits.
formyl GAR on the purine synthetic pathway is blocked by
homofolates (6, 7). GARFT, which catalyzes the transfer of the
formyl group from 10-formyltetrahydrofolate to GAR, is inhib
ited by the 5,11-methenyl derivative of H4HPteGlu in Sarcoma
180 extracts (8). We now show that H4HPteGlu polyglutamates
are potent, selective inhibitors of GARFT in extracts of human
and murine lymphoma cells. Our results indicate that both
reduction and polyglutamation are essential for homofolate
toxicity.
MATERIALS AND METHODS
Cell Cultures. Human B-cell lymphoma (Manca, SKD-HLA) (9) and
murine leukemia (LI210) cells were grown in RPMI 1640 medium
(folate concentration: 2.2 MM)supplemented with 10% horse serum, 2
HIMglutamine, 0.01 M sodium pyruvate, and O.OSM 2-mercaptoethanol
in a humidified 5% CO2:95% air atmosphere at 37°C(10). A strain of
Manca cells was made 180-fold resistant to the TS inhibitor 5,8dideaza-10-propargylfolate (II) by adding the inhibitor at increasing
concentrations up to 0.1 HIMto the culture medium.
For studies of growth inhibition, the original Manca cell strain was
grown in RPMI 1640 medium with 2 mM L-glutamine and 10% fetal
calf serum (doubling time, 15-17 h). Exponentially growing cells, 7000
in 0.14 ml medium, were transferred to 96-well microplates and incu
bated for 2 h. Antifolates, dissolved in Hanks' balanced salt solution
(GIBCO Laboratories, Grand Island, NY) were added to a total vol of
0.15 ml. After 72 h (the number of cells per well had reached 180,000),
0.015 ml of 15 mg/ml MTT (Sigma M 2128) was added. Incubation
was continued for 2 h and the MTT formazan produced was solubili/ed
by adding 0.17 ml of 0.01 N HC1 in isopropanol and quantitated
spectrophotometrically (12). To calculate the inhibition of growth to
be expected from two agents added simultaneously and acting inde
pendently, the fractional inhibition method of Harvey (13) was used.
When expressed as the fraction inhibited, the expected effect is the sum
of the two separate inhibitory effects minus their product. If the
observed inhibition by the combination is greater than expected, the
combination is synergistic; if smaller, it is antagonistic. In reversal
experiments, inosine (Sigma) was added to 0.1 mM with the inhibitors.
Control cultures produced 158 ±8 nmol MTT formazan (nine deter
minations) in 2 h.
Cell Extracts. A modification of the method of Caperelli ( 14) was
used for the preparation of cell extracts. Frozen cell pellets were thawed
and mixed with two volumes of phosphate buffered saline. The suspen
sion was treated with an equal volume of 10 mM potassium phosphate,
pH 7.5, containing (v/v) 25% ethylene glycol, 10% dimethylsulfoxide,
0.25 M sucrose, 10 mM 2-mercaptoethanol, 1 mM EDTA, and the
following protease inhibitors (Sigma): 25 Mg/ml a-1-antitrypsin, 0.12
units/ml aprotinin and 1 mM phenylmetnylsulfonyl fluoride. After 20
min at 4"( '. the cells were disrupted in a glass homogenizer and the
solution centrifuged at 11,000 x g for 10 min. The supernatants used
for enzyme assays contained 6-9 mg protein/ml as determined by the
Lowry method, with bovine serum albumin as standard (15).
HPteGlu Derivatives. Homofolic acid, homopteroic acid, and (6R,S)5-methylH4HPteGlu were provided by the Drug Development Branch
of the National Cancer Institute through the courtesy of Dr. J. A. R.
Mead. Poly--y-glutamyl derivatives of HPteGlu with 1-5 additional
glutamates were prepared from homopteroic acid as described (16).
158
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
ENZYME INHIBITION BY HOMOFOLATE POLYGLUTAMATES
=v
CH2CH2N
O
n H COOH
CNCH
CH2
CH2
COOH
HOMOFOLICACID
O
/=.
n H COOH
kCH2CH2N¿
u v />CNCH
'/
r*uo
H
M
UM¿
CH2
COOH
H
(6S)-5,6,7,8-TETRAHYDROHOMOFOLICACID
O
H
H
il HCOOH
CNCH
CH2
CH2
COOH
(6R)-5,6,7,8-TETRAHYDROHOMOFOLICACID
Fig. l. Structures of homofolate derivatives.
Tetrahydro derivatives, (6R,S)-H4HPteGlu and its polyglutamates,
were prepared from the corresponding homofolates by catalytic reduc
tion (17). The 6S diastereomer of H4HPteGlu6 was prepared by enzy
matic reduction of HPteGlu6 as described for the preparation of (6S)tetrahydrofolate from folie acid (18). The product, (6S)-H4HPteGlu6,
was purified by chromatography on a DEAE-cellulose column, using a
linear 0.2-1.0 M NaCI gradient as described (18), except that the pH
of the buffer was 7.5 and the temperature, 25°C.Five-mi fractions were
collected and (6S)-H4HPteGlu6 (1.5 //mol) was recovered from a peak
eluted at 220-250 ml. To obtain H2HPteGlu6 and (6R)-H4HPteGlu6,
(6R,S)-H4HPteGlu6 (1.5 /¿mol)was incubated with 220 milliunits of
Lactobacillus casei TS in the standard TS assay mixture with increased
dUMP (30 Mmol) in 30 ml for 5 h at 37°C(19). Sodium chloride (9
mmol) was included to minimize TS inhibition by polyglutamated
antifolates (19). The sample was diluted with 30 ml water and chromatographed on DEAE-cellulose. H2HPteGlu6 and the unreacted 6R
form of H4HPteGlu6 were eluted from the peak extending from 280
through 350 ml. Pure H2HPteGlu6 was recovered at 325-350 ml. A
mixture of H2HPteGlu6 and (6R)-H4HPteGlu6 was recovered at 280320 ml. To resolve the mixture, fractions were pooled, diluted with an
equal volume of NaCl-free elution buffer, and rechromatographed to
isolate (6R)-H4HPteGlu6 by the above procedure, except that a 0.2-0.6
M NaCI gradient was used. Pure (6R)-H4HPteGlu6 was recovered at
600 ml. All homofolate derivatives were identified and quantitated
spectrophotometrically. The molar absorbance coefficients used were
19,400 at 281 nm (pH 13) for HPteGlu, 19,700 at 282 nm (pH 7) for
H2HPteGlu and 20,500 at 295 nm (pH 7) for H4HPteGlu (17).
Other Materials. To prepare the natural (6R) diastereomer of 10formyl-tetrahydrofolate, (6S)-5-formyl-tetrahydrofolate was incubated
in 0.1 N HC1 for 1 h, followed by neutralization (20). The 5-formyl
compound was formed from (6S)-tetrahydrofolate (18) and formic acid
(21). TMTX was supplied by Warner-Lambert Co, Ann Arbor, MI.
Enzyme Assays. Extracts of Manca and L1210 cells were used for
enzyme assays except for TS. Extracts of 5,8-dideaza-10-propargylfolate-resistant cells containing 10 times the TS activity of extracts of the
original strain were used as the source of TS. Reaction rates were linear
with time and enzyme concentration. All assays were begun by addition
of enzyme unless otherwise specified. Blanks lacking enzyme or one of
the substrates gave negligible activity. The specific activity is expressed
in milliunits. or nanomoles of product formed/min per mg protein ±
SD for four extracts. GARFT (EC 2.1.2.2) was assayed by a modifica
tion of a continuous spectrophotometric method (22). Cell extracts (0.1
mg protein) were incubated with 45 /¿molof Tris-HCl, pH 7.5. 90 >imol
of 2-mercaptoethanol, 54 nmol of (6R)-10-formyltetrahydrofolate and
0.22 Mmolof «,/3-GARin 0.9 ml for 5 min at 30°C.Incubation mixtures
were flushed with argon before initiating the reaction. The formation
of tetrahydrofolate was measured by following the increase in absorb
ance at 298 nm using a molar absorbance of 19,700. Activities in the
controls lacking inhibitors were 2.9 ±0.7and 7.9 ±1.2 mu/mg protein
for Manca and LI210, respectively. Since the inhibitors (6S)H4HPteGlu6, (6R)-H4HPteGlu6, and H2HPteGlu6 were purified on
NaCI gradients, it was necessary to test the effect of NaCI on GARFT
assays. In Manca extracts, NaCI at 0.01-0.04 M did not interfere with
the inhibition assay but at 0.05 M it antagonized the inhibitory effect
of H4HPteGlu6. In L1210 extracts, NaCI began to interfere with inhi
bition at 0.03 M. The preparations of (6R)-H4HPteGlu6 and
H2HPteGlu«could not be tested at concentrations above 1.5 MM.The
(6S)-H4HPteGlu6 preparation was less dilute than the other two sam
ples and could be used at concentrations through 5 MMwithout inter
ference by NaCI.
AICARFT (EC 2.1.2.3) was assayed in Manca extracts as described
for GARFT except that 22 Mmol of KC1 were included, GAR was
replaced by 50 nmol of aminoimidazolecarboxamide ribonucleotide,
and incubation was at 37°C(23). The activity of control extracts was
1.5 ±0.1 mu/mg protein. To assay for TS (EC 2.1.1.45), the cell extract
(0.1 mg protein) was incubated with 8 /¿molof Tris-HCl, pH 7.4, 4
Mmol MgCl2, 2.5 Mmol HCHO, 0.16 Mmol EDTA, 16 Mmol 2-mercaptoethanol, 0.13 ninni (6R,S)-tetrahydrofolate,
and 16 nmol
[5-3H]dUMP, 10 MCi/Mmol,in 0.2 ml for 10 min at 30°C.The tritium
released was determined in the charcoal supernatant as described (24).
The activity in the control (high TS Manca strain) was 2.6 ±0.6 mu/
mg protein. SHMT (EC 2.1.2.1) was measured by following the con
version of [3-14C]serine to H[14C]HO. Extracts (0.08 mg protein) were
preincubated 5 min at 37°Cwith 50 Mmol potassium phosphate, pH
7.5, 70 nmol (6R,S)-tetrahydrofolate, 50 nmol pyridoxal phosphate
and 5 Mmol 2-mercaptoethanol in 0.19 ml. The reaction was begun by
the addition of 0.01 ml of 0.25 M [3-'4C]L-serine (0.012 MCi/Mmol).
After incubation for 15 min at 37°C,H[14C]HO was determined as the
dimedon derivative (25, 26). Activities were 5.9 ±0.4 and 20.0 ±5.7
mu/mg protein in Manca and LI210 extracts, respectively.
To determine if any degradation of HPteGlu polyglutamate chains
took place during enzyme assays, HPteGlu6 (18 nmol) was incubated
with a Manca cell extract (0.1 mg protein), 45 Mmol of Tris-HCl, pH
7.5, and 90 Mmol 2-mercaptoethanol in 0.9 ml for 5 min at 30°C.
Protein was removed by precipitation with 0.09 ml of 2 N perchloric
acid at 4°Cand samples were neutralized with 0.02 ml of 10 N KOH.
Zero-time samples were used as controls. Aliquots of the resulting
supernatant (2.5 nmol in 0.05 ml) were analyzed by reversed-phase
high-performance liquid chromatography (27). HPteGlu6 was eluted as
a single peak before and after incubation with the cell extract. No
species corresponding to HPteGlui-5 was detected.
RESULTS
Growth Inhibition
Homofolates were tested as inhibitors of the growth of cul
tured Manca cells. HPteGlu and 5-methyl-(6R,S)-H4HPteGlu
inhibit growth 50% at 6 and 8 /¿M,respectively, as shown in
Fig. 2 (each value represents the mean of three determinations,
159
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
ENZYME INHIBITION BY HOMOFOLATE POLYGLUTAMATES
g
20
40
5-Me-H4HPteGlu(uM)
120-,
60
2
45
HOMOFOLATE
Fig. 2. Effect of combinations of homofolate (HPteGlu) or (6R,S)-5methylH4HPteGlu with TMTX on growth of cultured Manca cells. A, cultures
were grown 72 h in 96-well microtiter plates at 7000 cells per well and growth
was estimated by MTT reduction. Cultures were treated with varying concentra
tions of HPteGlu: added alone (O), with 10 nM TMTX (•),with 0.1 IHMinosine
(A), and with 10 nM TMTX and 0.1 min inosine (A). Dashed line, additive effects
expected for each combination of agents, calculated from the results of TMTX
and HPteGlu when added singly (13). B, cultures were grown as described for/4,
except that HPteGlu was replaced by (6R,S)-5-methylH4HPteGlu.
(uM)
Fig. 3. Effect of polyglutamate chain length on the inhibition of GARFT by
homofolate derivatives in cell extracts. A, Manca; B, L1210. The total number of
glutamate residues is indicated on each curve. HPteGlu (A); HjHPteGlu (A);
(6R,S)-H4HPteGlu (•).
SD < 5%). Inhibition by these agents is reversed >90% by the
simultaneous addition of inosine, indicating a block in purine
biosynthesis. The effects of HPteGlu and 5-methylH4HPteGlu
are also completely reversed by (6S)-5-formyltetrahydrofolate
at 10 nM (not shown). To determine if the inhibitory action of
HPteGlu requires a reduction step mediated by DHF reductase,
the effect of the DHF reductase inhibitor TMTX was tested. In
the presence of 10 nM TMTX, inhibition by HPteGlu is dimin
ished 10-fold (Fig. 2A). When added alone at 10 nM, TMTX
inhibits cell growth less than 5%. With 5-methyl-H4HPteGlu,
the reduction step is by-passed and no antagonism by TMTX
would be expected. Fig. IB shows that inhibition by 5-methylH4HPteGlu is not antagonized but enhanced by TMTX.
Enzyme Inhibition
GARFT. The influence of the following factors on the inhi
bition of GARFT by HPteGlu derivatives was studied: (a)
polyglutamate chain length; (b) the extent of reduction of the
pyrazine ring; (c) stereochemical configuration at C6; and (d)
enzyme source. The inhibition curves for GARFT are shown in
Fig. 3. Based on a minimum of three trials, the standard
deviations of the plotted values were within 15%. The IC50
values derived from the inhibition curves are shown in Table 1.
The IC50values used for comparisons of inhibitory potency are
valid for these experiments because none of the inhibitors bind
tightly, that is, with K¡values in the 5-60 pM range (28). The
ICso values obtained for the Manca extracts can be compared
with those of LI210 because assay conditions are identical and
both extracts have similar Kmvalues for (6R)-formyltetrahydrofolate (25 and 35 nM, respectively) and for GAR (29 and 40
MM).
With H4HPteGlui_6, increasing polyglutamate chain length
enhances inhibition in both Manca and LI 210 cell extracts. In
Manca extracts, (6R,S)-H4HPteGlu2 is 12 times more inhibi-
tory than Glui. As more Glu residues are added to (6R,S)H4HPteGlu, little further change occurs until Glu5, which is 50
times more inhibitory than Glui. The inhibitory pattern is
similar in LI210 extracts, except that Glu2 is 1/6 as active as
in Manca cell extracts. The potency of (6R,S)-H4HPteGlu6 is
similar in both extracts. In Manca extracts, inhibition by
(6R,S)H4PteGlu6 is competitive with (6R)-10-formyltetrahydrofolate, with a A/ value (29) of 0.3 nM (data not shown).
Polyglutamates of HPteGlu are weak inhibitors of GARFT and
the potency of H2HPteGlu6 is between that of HPteGlu6 and
(6R,S)-H4PteGlu6
(Fig. 3). Higher
concentrations
of
H2HPteGlu6 were not tested because NaCl in this preparation
interfered with inhibition assays.
The effect of the stereochemical configuration at carbon 6 of
H4HPteGlu6 on GARFT inhibition is shown in Fig. 4 and Table
1. In Manca extracts the natural (6S) form is about four times
more inhibitory than the unnatural (6R) form. The results differ
in LI210 extracts, the (6R) form being slightly more inhibitory
than the (6S) form. The (6R,S) mixture gives inhibition curves
similar to those calculated for the equimolar mixture by the
fractional inhibition method (13) in both Manca and LI210
extracts. The individual diastereomers of H4HPteGlu6 were
chosen for further studies because Glu6 is the most potent
(6R,S)-H4HPteGlu derivative against L1210 GARFT.
We had observed that (6R)-H4HPteGlu6 was eluted from
DEAE-cellulose more slowly than the (6S) form. During prep
aration of (6R)-H4HPteGlu6, the (6R,S) mixture is incubated
with formaldehyde in the TS assay to form 5,11-méthylène
(6R,S)-H4HPteGlu6.
It is therefore possible that (6R)H4HPteGlu6 is recovered as its 5,11-méthylène
derivative after
chromatography on DEAE-cellulose. To exclude the possibility
that formation of 5,1 l-methylene-H4HPteGlu6 alters inhibitory
potency, (6R,S)-H4HPteGlu6 (1.6 HIM)was incubated with 12
itiM HCHO in 0.05 M Tris-HCl and 0.2 M 2-mercaptoethanol,
pH 7.5, for 5 min at 37°Cbefore adding to the GARFT assay
mixture. This treatment had no effect on the potency of (6R,S)H4HPteGlu6.
160
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
ENZYME INHIBITION BY HOMOFOLATE POLYGLUTAMATES
O 120 n
oc
80-
CC
<
3
10
2.0
1.0
H„HPteGlu6(|jM)
HOMOFOLATE
o 12°i
(uM)
K
Z 100-
o
u
u.
o
*
'
SO 60 H
u<
a
S
1.0
H4HPteGlu6(uM)
2.0
Fig. 4. Effect of diastereomers of H4HPteGlu6 on GARFT activity in cell
extracts. A, Manca: B. LI210. S, natural diastereomer (•);A',unnatural diastereomer (A); R,S, equimolar mixture of diastereomers (A). In I and /'. the dashed
line represents the additive values expected for an equimolar R,S mixture, calcu
lated from lines R and 5 by the fractional inhibition method (13).
10
20
H4HPteGlu6(pM)
Fig. 5. Effect of polyglutamate chain length on inhibition of AICARFT by
homofolates in extracts of Manea cells. The total number of glutamate residues
is indicated on each curve. A, HPteGlu; B, H4HPteGlu.
Table 1 Inhibition ofeniyme activities by homofolates
Concentration causing 50% inhibition of enzyme activity. IC50 values inter
polated from curves in Figs. 3-6.
GARFTCompoundHPteGlu,HPteGlu;HPteGlu,HPteGlu.HPteGlu,HPteGlu«(6R,S)-H4HPteGlu,(6R,S)-H4HPteGlu2(6R.S)-H4HPteGluj(6R,S)-H4HPteGlu4(6R,S)-H4HPteGlu,(6R,S)-H4HPt
10
HOMOFOLATE
Fig. 6. Effect of polyglutamate chain length on inhibition of TS by HPteGlu
in extracts of Manca cells. The total number of glutamate residues is indicated
on each curve.
AICARFT. The effect of HPteGlu polyglutamate chain
length and reduction of the pyrazine ring on the inhibition of
AICARFT was studied in extracts of Manca cells. With
HPteGlu, inhibitory potency increases with polyglutamate
chain length, a sharp increase occurring between Gluj and Glu4,
with little change thereafter (Table 1, Fig. 5/Õ).Conversion to
the corresponding (6R,S)-H4HPteGlu forms weakens inhibition
(Table l, Fig. 5Ä);the Glu4_6derivatives of (6R,S)-H4HPteGlu
are 3-5-fold less inhibitory than the corresponding HPteGlu
derivatives. The inhibition curve for (6R,S)-H4HPteGlu2 (not
shown) was nearly identical to that for Glu3. As with the
unreduced series, a marked increase in the potency of (6R,S)H4HPteGlu occurs between Gluj and Glu4 with no further
increase with Glus, and the effect of Glu6 is weaker than that
of Glu4.
TS. Inhibitory' potency of HPteGlu increases with the length
of the polyglutamate chain, but effects are weaker for TS than
for AICARFT. The Glu5 and Glu6 polyglutamates of HPteGlu
are slightly inhibitory (Table 1 and Fig. 6). The curve for
HPteGlus (not shown) is nearly identical to that for Glu6. If the
concentration of (6R,S)-tetrahydrofolate is decreased from 650
to 180 tiM and that of dUMP from 0.08 to 0.04 MM,the IC50 of
HPteGlu6 is lowered from 9 pM to the previously reported
value, 2 /IM (30). Conversion of HPteGlu derivatives to their
tetrahydro derivatives reduces their inhibitory potency against
TS (Table 1).
SHMT. In extracts of Manca and L12IO cells, SHMT activ
ity was inhibited less than 50% by HPteGlu^
and (6R,S)H4HPteGlu,_6 at 20 ^M.
DISCUSSION
The results show that HPteGlu inhibits growth of cultured
Manca cells, and that this effect is antagonized by the DHF
reducÃ-aseinhibitor, TMTX. In contrast, inhibition of Manca
cell growth by a tetrahydro derivative of HPteGlu, 5-methylH4HPteGlu, is not antagonized by TMTX. This supports the
hypothesis that the mechanism for growth inhibition by
HPteGlu involves reduction to H4HPteGlu. Our findings are
consistent with results of earlier studies with LI210 leukemic
mice: (a) the antitumor effects of homofolates are increased if
161
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
ENZYME INHIBITION BY HOMOFOLATE
dTMP <-*-< dUMP
AICAR cx=> Purines
HPie(poly)Glu
HPleGlu f—
ÃŽ
H4HPleGlu <=i
HPteGlu
5-CH3-H4HPteGlu
5-CH3-H4HPleGlu
l
H4HPte(poly)Glu
GAR
Purines
Fig. 7. Potential metabolites and inhibitory sites of homofolates.
animals are inoculated with a methotrexate-resistant strain of
LI210 with elevated levels of DHF reducÃ-ase(1, 2), and (b)
they are diminished when animals are pre-treated with methotrexate (1, 31). The reduction of H2HPteGlu to HJHPteGlu
occurs in tissues of normal and leukemic mice in vivo and in
cell-free extracts (3, 32). The in vivo reduction reaction is
blocked in tissues of animals pretreated with methotrexate (32).
Reversal of the inhibitory effects of HPteGlu on Manca cell
growth by inosine indicates that a step in purine biosynthesis is
involved. Previous studies have shown that synthesis of
formylGAR is inhibited by homofolates in sarcoma-180 (7).
We have now demonstrated that polyglutamates of H4HPteGlu
are potent inhibitors of GARFT activity in cell-free extracts of
Manca and LI 210 tumor cells. Polyglutamates of HPteGlu are
considerably less inhibitory. This supports the hypothesis that
a reduction step is required. Since GARFT is inhibited by the
natural diastereomer of H4HPteGlu6, growth inhibition by
HPteGlu could be explained by its conversion to a polyglutamate of (6S)-H4HPteGlu. Polyglutamation of homofolates is
likely to occur in vivo, since homofolates are substrates for
mammalian folylpolyglutamate synthetase (33, 34). Other antifolates have been shown to be converted into polyglutamated
forms with increased potency as inhibitors of DHF reductase,
TS, and AICARFT (10, 35-39).
Polyglutamate chain length also regulates the activity of
folates as GARFT substrates and inhibitors; studies of the avian
enzyme show that the ratio of VmaK
to the Kmfor polyglutamates
of the 10-formyltetrahydrofolate
precursor, 5,10-methenyltetrahydrofolate, increases with chain length from Glui to Glu6
(40). Folie acid polyglutamates inhibit avian GARFT and their
potency increases from Glui to Glu? (40).
Inhibition of AICARFT by homofolates may also be involved
in the inhibition of cell growth, since polyglutamates of unre
duced HPteGlu inhibit AICARFT. These effects, however, are
more modest than the inhibition of GARFT by H4HPteGlu
derivatives. TS is an unlikely target enzyme since inhibition of
cell growth by homofolates is completely reversed by inosine in
the absence of thymidine.
The results show a lack of stereospecificity for the inhibition
of GARFT in mammalian systems. Both the 6S and 6R diastereomers of H4HPteGlu6 show inhibitory effects of the same
magnitude against GARFT. This is consistent with the obser
vation that both (6S) and (6R)-H4HPteGlu have antitumor
activity in LI210 leukemic mice (41). Similarly, both diastereomers at C-6 of the GARFT inhibitor 5,10-dideazatetrahydro-
POLYGLUTAMATES
folate are active against LI210 cell growth (42). With L. casei,
by contrast, (6R)-H4HPteGlu inhibits growth (IC5o, 0.018 /¿M)
while the 6S form acts as a growth factor (43). The (6R) form
of H4HPteGlu6 is more potent (IC50,0.1 MM)than the (6S) form
(ICso, 5 MM)as an inhibitor of GARFT activity in extracts of L.
casei.4 It is of interest that avian GARFT is inhibited by the
unnatural diastereomer of its substrate, 10-formyltetrahydro
folate (22).
The target enzyme for HPteGlu in vivo would depend on
relative rates of reduction and polyglutamation of this agent.
As shown in Fig. 7, the balance among these reactions could
determine whether GARFT, AICARFT, or TS is the principal
site of inhibition. Our results suggest that HPteGlu is converted
to H4HPteGlu polyglutamates. Since GARFT is the only re
action found to be sensitive to H4HPteGlu in cell extracts, it is
the most likely target in vivo. With 5-methyl-H4HPteGlu, no
reduction step is required. It is possible that 5-methylH4HPteGlu is demethylated, since it is a substrate for N-5methyltetrahydrofolate-homocysteine
methyltransferase (44).
Conversion of HPteGlu to folate coenzyme analogues such as
5-methyl or ll-formyl-H4HPteGlu
may also occur, since
H4HPteGlu is a substrate for 10-formyltetrahydrofolate synthe
tase and other folate-dependent enzymes (5, 44). The potency
of polyglutamates of these coenzyme analogues as enzyme
inhibitors is as yet undetermined. The intracellular distribution
of polyglutamated folate substrates may also play a role since
the degree of polyglutamation can regulate enzyme reaction
rates and sensitivities to antifolates (23, 35-40). Conversely,
these folate pools may be altered by homofolate coenzyme
analogues competing for the same enzyme sites.
The ability of TMTX to enhance the potency of 5-methylH4HPteGlu as a growth inhibitor is similar to effects of com
bining an inhibitor of DHF reductase with 5,10-dideazatetrahydrofolate (45, 46). It is unlikely that subtoxic levels of
TMTX inhibit GARFT by causing depletion of the GARFT
substrate 10-formyltetrahydrofolate (47). TMTX may block the
AICARFT reaction indirectly instead by causing the accumu
lation of dihydrofolate polyglutamates, since it is known that
these compounds inhibit AICARFT (36, 47). It is also possible
that polyglutamates of both HPteGlu and H4HPteGlu are
formed in the presence of low concentrations of TMTX, block
ing both AICARFT and GARFT. It has been suggested that
homofolates become inhibitors of purine biosynthesis after
conversion to polyglutamate forms (8, 48). Our studies provide
direct evidence for this view and point to GARFT as the most
likely target enzyme.
ACKNOWLEDGMENTS
The authors wish to thank Dr. Jacob Selhub of the U. S. Department
of Agriculture Human Nutrition Research Center on Aging, Tufts
University, Boston, MA, for performing high-performance liquid chromatography analyses of homofolate polyglutamates.
REFERENCES
1. Risiilik, R. L. Homofolates and other 2-amino-4-oxy antifolates. In: T. J.
Bardos and T. I. Kaiman (eds.), New Approaches to the Design of Antineoplastie Agents, pp. 201-214, New York: Elsevier Science Publishing Co, Inc.,
1982.
2. Mead, J. A. R., Goldin, A., Kisliuk, R. L., Friedkin, M., Plante, L., Crawford,
E. J., and Kwok, G. Pharmacological aspects of homofolate derivatives in
relation to amethopterin-resistant murine leukemia. Cancer Res., 26: 23742379, 1966.
3. Plante, L. T., Crawford, E. J., and Friedkin, M. Enzyme studies with new
4 J. Thorndike, R. L. Kisliuk, and Y. Gaumont, unpublished results.
162
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
ENZYME INHIBITION BY HOMOFOLATE POLYGLUTAMATES
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
analogues of folie acid and homofoUc acid. J. Biol. Chem., 242: 1466-1476.
1967.
Hakala, M. T. Homofolate and tetrahydrohomofolate. inhibitors of purine
synthesis. Cancer Res., 31: 813-816, 1971.
Scanion, K. J., Cashmore, A. R., Moroson, B. A. Dreyer, R. N., and Berlino,
J. R. Inhibition of serine metabolism by tetrahydrohomofolate in L1210
mouse leukemia cells. Mol. Pharmacol., 19: 481-490, 1981.
Daubner, S. C., and Benkovic, S. J. Characterization of mammalian films
phoribosylglycineamide formyltransferase from transformed cells. Cancer
Res., 45:4990-4997. 1985.
Divekar, A. Y., and Hakala, M. T. Inhibition of the biosynthesis of 5'phosphoribosyl-A'-formylglycinamide in sarcoma 180 cells by homofolate.
Mol. Pharmacol., //: 319-325, 1975.
Caperelli, C. A., Domanico, P., and Benkovic, S. J. Synthesis of 5,11methenyltetrahydrohomofolate
and its antifolate activity in vitro. J. Med.
Chem., 24: 1086-1088, 1981.
Nishikori, M., Hansen, H., Jhanwar, S., Fried, J., Sordillo. P., Koziner, B.,
Lloyd, K., and Clarkson, B. Establishment of a near-petraloid B-cell lymphoma line with duplication of the 8:14 translocation. Cancer Genet. Cytogenet., 72:39-50, 1984.
Samuels, L. L., Moccio, D. M., and Sirotnak, F. M. Similar differential for
total polyglutamation and cytotoxicity among various folate analogues in
human and murine tumor cells in vitro. Cancer Res., 45: 1488-1495, 1985.
Jones, T. R., Calvert, A. H., Jackman, A. L., Brown, S. J., Jones, M., and
I larrup. K. R. A potent antitumor quinazoline inhibitor of thymidylate
synthetase: synthesis, biological properties and therapeutic results in mice.
Eur. J. Cancer., 17: 11-19, 1981.
Mossmann, T. Rapid colorimetrie assay for cellular growth and survival:
application to proliferation and cytotoxicity assays. J. Immunol. Methods,
65:55-63, 1983.
Harvey, R. J. Synergism in the folate pathway. Rev. Infect. Dis., 4:255-260,
1982.
Caperelli, C. A. Mammalian glycinamide ribonucleotide transformylase:
purification and some properties. Biochemistry, 24: 1316-1320, 1985.
Lowry, O. H., Rosebrough, N. J., Fair, A. L., and Randall, R. J. Protein
measurements with the Polin phenol reagent. J. Biol. Chem., 193: 265-275,
1951.
Nair, M. G., Nanavati, N. T., Kumar, P., Gaumont, Y., and Kisliuk, R. L.
Synthesis and biological evaluation of poh -, -niniaim 1 metabolites of 10deazaaminopterin and 10-ethyl-lO-deazaaminopterin. J. Med. Chem., 31:
181-185, 1988.
Kisliuk, R. L. Synthesis and assay of tetrahydrohomofolic acid and tetrahydrohomopteroic acid. Methods Enzymol., 18B: 663-670, 1971.
Kisliuk, R. L., Gaumont, Y., Lafer, E., Baugh, C. M., and Montgomery, J.
A. Polyglutamyl derivatives of tetrahydrofolates as substrates for Lactobacillus casei thymidylate synthase. Biochemistry, 20:929-934. 1981.
Kisliuk, R. L., Gaumont, Y., and Baugh, C. M. Polyglutamyl derivatives of
folate as substrates and inhibitors of thymidylate synthase. J. Biol. Chem.,
249:4100-4103, 1974.
Rabinowitz, J. C. Preparation and properties of 5,10-methenyltetrahydrofolic
acid and 10-formyltetrahydrofolic acid. Methods Enzymol., 4: 814-815,
1963.
Sirotnak, F. M., Chello, P. L., Moccio, D. M., Kjsliuk, R. L., Combepine,
G., Gaumont, Y., and Montgomery, J. A. Stereospecificity at carbon 6 of
formyltetrahydrofolate as a competitive inhibitor of transport and cytotox
icity of methotrexate in vitro. Biochem. Pharmacol., 28: 2993-2997, 1979.
Smith, G. K., Benkovic, P. A., and Benkovic, S. J. L-(-)-10-Formyltetrahydrofolate is the cofactor for glycinamide ribonucleotide transformylase from
chicken liver. Biochemistry, 20: 4034-4036, 1981.
Baggott, J. E., and Krumdieck, C. L. Folypoly-y-glutamates as cosubstrates
of 10-formyltetrahydrofolate:
S'-phosphoribosyl-S-amino^-imidazolecarboxamide formyltransferase. Biochemistry, 18: 1036-1041, 1979.
Herzfeld, A., and Raper, S. M. Relative activities of thymidylate synthetase
and thymidine kinase in rat tissues. Cancer Res., 40: 744-750, 1980.
Taylor, R. T., and Weissbach, H. Radioactive assay for serine transhydroxymethylase. Anal. Biochem., 13: 80-84, 1965.
Thorndike, J., Pelliniemi, T. P., and Beck, W. S. Serine hydroxymethyltransferase and serine incorporation in leukocytes. Cancer Res., 39: 3435-3440,
1979.
Kamen, B. A., and Winick, N. Analysis of methotrexate polyglutamate
derivatives in vivo. Methods Enzymol., 122: 339-346, 1986.
Baccanari, D. P., and Tansik, K. L. Kinetics of methotrexate binding to
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
dihydrofolate reducÃ-asefrom Neisseria gonorrhoea. Biochem. Pharmacol..
33: 2101-2107. 1984.
Dixon, M. The determination of enzyme inhibitor constants. Biochem. J.,
55: 170-171, 1953.
Thorndike, J., Gaumont, Y., Kisliuk, R. L., Sirotnak. F. M.. Murthy, B. R.,
and Nair, M. G. Interaction of homofolate polyglutamates with folate en
zymes from a human B-cell lymphoma. Proc. Am. Assoc. Cancer Res., 29:
285, 1988.
Mishra, L. C., and Mead, J. A. R. Further evaluation of the antitumor activity
of homofolate and its reduced dérivâtes
against methotrexate-insensitive
tumors. Chemotherapy, 17: 238-292. 1972.
Mishra, L. C., Parmar, A. S., and Mead, J. A. R. Effect of pretreatment with
methotrexate on the reduction of dihydrohomofolic acid in mice. Biochem.
Pharmacol.. 20: 2871-2878, 1971.
McGuire, J. J., Hsieh, P., Coward, J. K., and Benino, J. R. Enzymatic
synthesis of folylpolyglutamates: characterization of the reaction and its
products. J. Biol. Chem., 255: 5776-5788. 1980.
Cook, J. D., Cichowicz, D. J., Sabu, G., Lawler. A., and Shane, B. Mam
malian folyl-7-glutamate synthetase. 4. In vitro and in vivo metabolism of
folates and analogues and regulation of folate homeostasis. Biochemistry,
20:530-539, 1987.
Allegra, C. J., Drake, J. C., Jacques, J., and Chabner, B. A. Inhibition of
folate-dependent enzymes by methotrexate polyglutamates. In: I. D. Gold
man (ed.). Proceedings of the Second Workshop on Folyl and Antifolyl
Polyglutamates, pp. 348-356. New York: Praeger, 1985.
Allegra, C. J.. Drake, J. C., Jolivet. J., and Chabner, B. A. Inhibition of
phosphoribosylaminoimidazolecarboxamide
transformylase by methotrexate
and dihydrofolic acid polyglutamates. Proc. Nati. Acad. Sci. USA, 82:48814885, 1985.
Sikora, E., Jackman, A. L., Newell, D. R., Harrap, K. R., Calvert, A. H.,
Jones, T. R., Pawelczak, K., and Rzeszotarska, B. A"°-Propargyl-5,8-dideazafolic acid polyglutamates as inhibitors of thymidylate synthase and their
intracellular formation. In: B. A. Cooper and V. M. Whitehead (eds.).
Chemistry and Biology of Pteridines, pp. 675-679. New York: de Gruyter,
1986.
Kumar, P., Kisliuk, R. L., Gaumont, Y., Nair, M. G., Baugh, C. M., and
Kaufman, B. T. Interaction of polyglutamyl derivatives of methotrexate, 10deazaaminopterin and dihydrofolate with dihydrofolate reducÃ-ase.Cancer
Res., 46: 5020-5023, 1986.
Kisliuk, R. L. Pteroylpolyglutamates. Mol. Cell. Biochem., 39: 331-345,
1981.
Chan. V. T., and Baggott, J. E. Polyglutamyl folate coenzymes and inhibitors
of chicken liver glycinamide ribotide transformylase. Biochim. Biophys. Acta,
702:99-104, 1982.
Kisliuk, R. L., Gaumont, Y., Thorndike, J., Schmid, F. A., Sirotnak, F. M.,
Piper, J. R., Murthy, B. R., and Nair, M. G. Antitumor, cytotoxic and
enzyme inhibitory properties of homofolates. Pharmacol. Ther.. in press,
1989.
Moran, R. G., Taylor, E. C., Shih, C., and Beardsley, G. P. The two
diastereomers of 5,10-dideazatetrahydrofolate
(DDATHF) are equiactive.
Proc. Am. Assoc. Cancer Res., 28: 274, 1987.
Kisliuk, R. L., and Gaumont, Y. Action of diastereoisomers of tetrahydro
homofolate on the growth of Lactobaciltus casei. Ann. NY Acad. Sci., 186:
438-440, 1971.
Taylor, R. T., and Hanna, M. L. 5-Methyltetrahydrohomofolate: a substrate
for cobalamin methyltransferases and an inhibitor of cell growth. Arch.
Biochem. Biophys., 163: 122-132, 1974.
Thorndike. J., Gaumont, Y., Powers, J., Kisliuk, R. L., and Piper, J. R.
Synergistic growth inhibition of human lymphoma cells by combination of
trimetrexate with 5,10-dideazatetrahydrofolate.
Proc. Am. Assoc. Cancer
Res.. 29: 285, 1988.
Galivan, J., Nimec, Z., Rhee, M., Boschelli, D., Oronsky, A. L., and Kerwar,
S. S. Antifolate drug interations: enhancement of growth inhibition due to
the antipurine 5,10-dideazatetrahydrofolic acid by the lipophilic dihydrofo
late reducÃ-aseinhibitors metoprine and trimetrexate. Cancer Res., 48:24212425, 1988.
Allegra, C. J., Hoang, K., Yeh, G. C., Drake, J. C., and Baram, J. Evidence
for direct inhibition of de novo purine synthesis in human MCF-7 breast cells
as a principal mode of metabolic inhibition by methotrexate. J. Biol. Chem.,
262: 13,520-13,526, 1987.
Slicker, L. J., and Benkovic, S. J. Inhibition of HKSV28 cell growth by 5,11methenyltetrahydrohomofolate. Mol. Pharmacol., 25: 294-302, 1984.
163
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
Inhibition of Glycinamide Ribonucleotide Formyltransferase and
Other Folate Enzymes by Homofolate Polyglutamates in Human
Lymphoma and Murine Leukemia Cell Extracts
J. Thorndike, Y. Gaumont, R. L. Kisliuk, et al.
Cancer Res 1989;49:158-163.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/49/1/158
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.