![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 282, Issue 3, 1305-1311, 1997
Department of Pharmacology, University of Bergen, N-5021 Bergen, Norway
 |
Abstract |
|---|
 Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
We compared the effects of methotrexate (MTX) and nitrous oxide on the methionine (Met) synthase system in two variants of a human glioma cell line. The cells were protected from cytotoxic effect of MTX by adding thymidine and hypoxanthine to the cell culture medium. MTX (0-1 µM) was associated with a dose- and time-dependent reduction in 5-methyltetrahydrofolate (5-methyl-THF) in both cell lines. Already after 3 hr of exposure, 5-methyl-THF was reduced by 50% and after additional 48 hr, the level was undetectable. In addition to reduction in folate level, homocysteine (Hcy) remethylation in intact cells was markedly inhibited as judged by an increased export of Hcy from the cells, and Met synthase activity in cell extracts and level of cellular methylcobalamin (CH3Cbl) declined. MTX reduced Hcy remethylation and CH3Cbl level more efficiently than nitrous oxide. In both cell variants, the inactivation of Met synthase by nitrous oxide was almost completely prevented in cells pre-exposed to MTX. This indicates that there is no catalytic turnover in cells exposed to MTX, and emphasizes the importance of the sequence of administration for synergistic effect of this drug combination. In conclusion, our data show that MTX through depletion of 5-methyl-THF reduces both the Met synthase activity and the cellular CH3Cbl level. Moreover, the effect of MTX on the Hcy remethylation is more pronounced than the inhibition caused by nitrous oxide. These observations should be taken into account in studies on MTX pharmacodynamics.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
MTX
is an antifolate drug that inhibits dihydrofolate reductase, the enzyme
responsible for the regeneration of tetrahydrofolate from
dihydrofolate. This effect is associated with depletion of cellular
reduced folates, and thereby inhibition of numerous folate-dependent processes. Impeding of the thymidylate synthase and enzymes involved in
purine biosynthesis impairs DNA synthesis, which probably explains the
cytotoxicity of MTX (Bertino, 1993
).
Met synthase (5-methyltetrahydrofolate-homocysteine methyltransferase,
EC2.1.1.13) catalyzes a folate-dependent reaction, in which
5-methyltetrahydrofolate (5-methyl-THF) functions as methyldonor,
thereby converting Hcy to Met. Cobalamin (Cbl) serves as cofactor in
this reaction. Because Met synthase operates at the point of
convergence of folate, Cbl and sulfur amino acids (Finkelstein, 1990),
impaired function of this enzyme may have secondary effects on many
cellular processes, including S-adenosylmethionine-dependent methylation reactions and polyamine synthesis.
The anesthetic agent nitrous oxide is the only drug which has been
reported to directly inhibit Met synthase, and it probably oxidizes
enzyme-bound cob(I)alamin formed during the catalytic cycle (Banerjee
and Matthews, 1990). Notably, this effect of nitrous oxide may account
for its diverse biological effects, including the megaloblastic changes
in human bone marrow (Nunn, 1987; Amess et al., 1978),
antileukemic effect reported in patients (Ikeda et al.,
1989; Eastwood et al., 1963) and experimental animals (Ermens et al., 1989b; Abels et al., 1990) and
altered metabolism and plasma level of Hcy and related sulfur compounds
(Nunn, 1987; Ermens et al., 1991a; Christensen et
al., 1994).
The side effects reported for nitrous oxide (Nunn, 1987) point to
significant biological consequences of Met synthase inhibition. Depletion of 5-methyl-THF by MTX (Bunni 1988, Ermens 1991b, Baram 1987)
suggests that some effect of this antifolate drug may be related to
inhibition of Met synthase. Furthermore, because the action of both
nitrous oxide and MTX converges on 5-methyl-THF metabolism, one may
expect a dynamic interaction between these two drugs (Ueland et
al., 1986b; Goldhirsch et al., 1987; Ermens et
al., 1991b).
The aim of our work was to investigate changes of the Met synthase system secondary to the folate depletion induced by MTX, and to compare these changes to those caused by nitrous oxide. First we confirmed that MTX depleted cellular 5-methyl-THF, and then we studied the relation between folate depletion and alteration in Met synthase, including the susceptibility of the enzyme to inactivation by nitrous oxide, the influence on intact cell Hcy remethylation and the content of Cbl cofactor. The study was performed with two genetically related human glioma cell lines, characterized by diverse functional state of the enzyme. The cells were cultured in the presence of Thd and Hx, to protect against MTX cytotoxicity.
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Chemicals.
The sources of various chemicals used in the cell
culture experiments and in the assays have been reported previously
(Fiskerstrand et al., 1994).
L-[14C]Hcy thiolactone (56 mCi/mmol), [57Co]Cbl (0.3 Ci/µmol) and
(±)-L-5-[methyl-14C]methyl-THF (50 mCi/mmol, barium salt) were purchased from Amersham International
(Buckinghamshire, UK).
L-5-[3
,5-7(N)-3H]formyl-THF (25 Ci/mmol) was
obtained from Moravek Chemicals. Inc. (Brea, CA). A custom- made
powdered DMEM, identical to a standard DMEM but without folic acid and
Met, was from GIBCO-BRL (Paisley, Scotland). Hog kidney hydrolase (10 mg protein/ml), prepared as described by McMartin et al.
(1981), was a gift from Dr. Ermens, Erasmus University, Rotterdam.
Cells and cell culture conditions.
The two variants of the
human glioma cell line GaMg used in these experiments, have been
described (Fiskerstrand et al., 1994, 1997). The parent cell
line was isolated in 1984 from a human glioblastoma multiforme tumor in
a 42-yr-old woman (Akslen et al., 1988). Both variants are
passage number 60 of GaMg. The Met-dependent variant, P60, was
developed by repeated passages in Met-containing medium (Fiskerstrand
et al., 1994), and the other variant, P60H, retained its Met
independence by being passed in Hcy containing medium (Fiskerstrand
et al., 1997).
Experimental designs.
For all experiments, cells were seeded
in Met medium at a density of 1000 cells/cm3 in
6-cm dishes (Nunc, Denmark) (most experiments), or in 10-cm dishes (Cbl
determinations). When reaching log growth phase, the cells received
fresh medium with 0 to 3 µM MTX, and the medium volume was adjusted
to 40,000 cells/ml (Fiskerstrand et al., 1997). After 3 hr
of incubation, the dishes were placed in modular incubator chambers
(Billups-Rothenberg, Del Mar, CA) and exposed to air (75%
N2, 20% O2 and 5%
CO2) or nitrous oxide (50%
N2O, 25% N2, 20%
O2 and 5% CO2). The gas
was moistened and tempered by passing through sterile
H2O at 50°C, and then delivered to the chambers at a flow rate of 5 liters/min for 10 min.
80°C until analysis. Medium was kept at 20°C
until Hcy determination. Cells were counted using a Coulter counter
(Coulter Electronics Ltd., Luton, UK).
Cells for Met synthase activity and Cbl measurements were harvested 48 hr after start of gas exposure, and for Hcy export after 0, 6, 12, 24, 36, 48 and 72 hr. Samples for folate (and simultaneous enzyme activity)
determinations were first depleted of folates by growing the cells for
10 days in medium without folate, but containing Hx/Thd. The cells were
then seeded at the density given above, but medium was replaced after
24 hr with fresh medium containing either 42 nM
5-[3H]formyl-THF (for determination of folate
distribution) or unlabeled 5-formyl-THF (for measurements of total
folate or Met synthase activity). After incubation for 24 hr, medium
was changed to standard DMEM, to allow saturation and equilibration of
the folate pools. Medium was changed 36 hr later, and MTX was added.
Dishes were incubated for 3 hr with this medium before gas exposure
(air or nitrous oxide).
Met synthase assay.
The enzyme activity was determined
according to a slight modification (Christensen et al.,
1992) of the assay described by Weissbach et al.(1963).
Hcy export rates.
The concentration of Hcy in the medium was
determined by an automated HPLC method (Refsum et al., 1989;
Fiskerstrand et al., 1993), and the logarithm of the
cell number and the concentration of Hcy in the medium was plotted
against time of incubation. The equations obtained by curve fit to
polynomial functions were used to calculate Hcy export rate, which
describes change in Hcy concentration per hr per
106 cells. Export rates were plotted against
time. The mathematic processing of data has been described in detail
(Refsum et al., 1991; Christensen et al., 1991).
Determination of folates.
This was based on a method
developed by Ermens et al. (1991b), which was modified by us
(Fiskerstrand et al., 1994). Cells were trypsinized and then
carefully washed twice with medium (without supplements). The cell
pellet was dissolved in 600 µl 2% ascorbic acid and 2%
mercaptoethanol containing unlabeled folates as internal standard.
Further processing included heat denaturation of proteins, enzymatic
cleavage of polyglutamates by hog kidney hydrolase, heat inactivation
of the hydrolase and centrifugation. Samples were injected on a 3-µm
octadecylsilane Hypersil column (10 × 0.5 cm) equilibrated with
20 mM ammonium sulfate, 10 mM tetrabutylammoniumhydroxide at pH 6.5, and eluted with a nonlinear methanol gradient (12.5-27.5% in 35 min)
at a flow of 1.5 ml/min. The retention times were 25 and 36 min for
5-formyl-THF and 5-methyl-THF, respectively. Fractions were collected
by Foxy model 200 fraction collector, and radioactivity was counted.
).
Determination of Cbl.
Extraction of Cbl was performed with
slight modification (Fiskerstrand et al., 1994) of the
method described by van Kapel et al. (1983). After
trypsination, cells were washed four times with PBS containing 0.1 mg/ml albumin, and 300 µl cell suspension was stored at 80°C
until analysis. After thawing, 8 µl N-ethylmaleimide and 8 µl
acetic acid were added, and the volume was adjusted to 800 µl with
distilled water. Then, Cbl was extracted by heating at 80°C for 30 min.
;
Fiskerstrand et al., 1994). A sample volume of 400 µl was injected into a 3-µm octadecylsilane Hypersil column (10 × 0.4 cm) equilibrated with 50 mM sodium phosphate buffer, pH 3.0. The Cbl
forms were eluted at a flow of 1.3 ml/min with a linear acetonitrile gradient (5-30% in 13.8 min). The retention times were 10.5 min (OHCbl), 12 min (CNCbl), 14 min (adenosylCbl) and 16 min
(CH3Cbl).
A radioisotope dilution assay described by van Kapel et al.
(1983) was used to determine both total Cbl and the different Cbl forms
in the HPLC fractions. Salivary R-binder was used as Cbl binding
protein.
Statistical analyses. Results are given as mean and S.D. We used analyses of variance for comparison between different treatment groups. The significant levels were expressed as two-tailed.
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Cells and culture conditions.
The doubling time of P60H and
P60 cultured in Met medium (with Thd/Hx) were 23 and 17 hr,
respectively. Only the P60 cell line showed growth arrest when Hcy
replaced Met (data not shown), which confirmed the Met-dependent
phenotype of P60 cells (Fiskerstrand et al., 1994), also in
Thd/Hx-containing medium.
Cellular folate content. Total folate content of cells grown in a medium with Thd/Hx was about 35 pmol/106 cells in both cell lines. MTX induced a progressive depletion of total folate, and the folate content was reduced to 60% (P60H cells) and 75% (P60 cells) after 3 hr exposure to 1 µM, and to 25% (P60H) and 8% (P60) after 48 hr (tables 1 and 2).
|
|
|
Assessment of cellular remethylation by Hcy export rates.
Inhibition of Hcy remethylation is reflected by an increased
concentration of Hcy extracellularly both in vivo an
in vitro (Refsum et al., 1997). Because the
amount of Hcy released to the medium conceivably depends on both the
number of cells and the duration of the experiment, we studied the
changes in export of Hcy by determining the export rate,
i.e., the amount of Hcy exported from one million cell per
unit of time (Christensen et al., 1991; Refsum et
al., 1991). We compared the Hcy export rate in the presence and
absence of drug and assume that an increased difference in export rate
reflected a more pronounced inhibition of Hcy remethylation.
|
|
Met synthase activity.
We first investigated the separate
effect of MTX or nitrous oxide on Met synthase in P60H and P60 cells.
MTX caused a dose-dependent reduction in enzyme activity in both cell
lines with a half maximal effect at 0.2 µM. After exposure to MTX (3 µM) for 48 hr, the activity was reduced to 30 to 50% as compared to
control cells (fig. 2A). Nitrous oxide
inactivated the enzyme at an initial rate of 0.07 hr
1 in
P60 H and 0.12 hr1 in P60 cells (Fiskerstrand et
al., 1994), and after 48 hr, the remaining activity was ~ 20% of activity in cells not exposed to nitrous oxide (fig. 2B).
|
Cellular Cbl content.
We confirmed (Fiskerstrand et
al., 1994, 1997) that total Cbl content was essentially the same
in the two cell lines, and the level of CH3Cbl
was higher in P60H (83 fmol/106 cells) than in
P60 (30 fmol/106 cells) (table
4). Nitrous oxide markedly reduced the
level of CH3Cbl (to 20-40%) in both cell lines.
Notably, MTX reduced CH3Cbl (to 25%) equally or
even more efficiently than nitrous oxide. In MTX-exposed cells, nitrous
oxide caused essentially no further reduction of
CH3Cbl (table 4).
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Aim and study design.
We have studied the effect of MTX and
nitrous oxide in two cell lines characterized by different ability to
use Hcy instead of Met for growth. To prevent growth arrest and
toxicity induced by MTX, we supplied the medium with Thd/Hx.
Interestingly, these culture conditions increased the level of
5-methyl-THF and CH3Cbl in both cell lines
compared to medium with no (Fiskerstrand et al., 1994) or
lower Thd/Hx concentration (Fiskerstrand et al., 1997).
Despite this folate sparing effect of Thd/Hx, P60 cells remained Met
dependent, and MTX still caused an extensive and rapid depletion of
5-methyl-THF.
Folate depletion and inhibition of Hcy remethylation.
MTX
induced a marked time- and concentration-dependent reduction of
cellular 5-methyl-THF (tables 1 and 2; fig. 3). Such an MTX effect on
folate distribution has also been reported by others both in
vitro (Bunni et al., 1988; Baram et al.,
1987; Ermens et al., 1991b) and in vivo (Bunni
et al., 1994; Selhub et al., 1991), and it has
been suggested that the depletion of 5-methyl-THF may be partly
explained by an inhibition of methylene-THF reductase (Chabner et
al., 1985).
; Refsum et al., 1986;
Ermens et al., 1991a; Christensen et al., 1992;
Refsum et al., 1997). We showed that MTX is an equal (P60
cells) or even stronger (P60H cells) inhibitor of Hcy remethylation
than nitrous oxide (fig. 1; table 3). The increased sensitivity to
nitrous oxide in Met-dependent P60 cells compared to P60H cells is
probably related to low CH3Cbl content in P60
cells (Fiskerstrand et al., 1997).
Effects of MTX and nitrous oxide on Cbl metabolism.
Cbl
deficiency eventually leads to deprivation of the intracellular folate
pool, in accordance with the folate trap hypothesis (Shane and
Stokstad, 1985). The data on the influence of folates on Cbl metabolism
are however sparse (Quadros et al., 1976; Quadros and
Jacobsen, 1995). We found that MTX reduced the
CH3Cbl content to the same extent (P60 cells), or
more (P60H) than nitrous oxide (table 4). This effect of MTX is
probably secondary to insufficient supply of 5-methyl-THF for Hcy
remethylation, and adds strong support to the notion (Fiskerstrand
et al., 1997) that the amount of cellular
CH3Cbl reflects the catalytic activity of Met
synthase. Our data may therefore provide some insight into the
mechanisms behind low serum or erythrocyte Cbl after MTX treatment or
during folate deficiency. Leeb et al. (1995) demonstrated
normal serum Cbl and low Cbl in erythrocytes from patients receiving
low dose MTX therapy. A related phenomenon is low serum Cbl in patients with folate deficiency and no overt Cbl deficiency, which was reported
by Mollin and Ross in 1957 (Mollin et al., 1962) and confirmed in later studies (Van der Weyden et al., 1972). We
recently observed that hyperhomocysteinemic subjects with the C677T
mutation in the methylene-THF reductase gene and low plasma folate
(i.e., low 5-methyl-THF) often had a low serum Cbl level
(Guttormsen et al., 1996).
MTX effects on Met synthase activity. We could distinguish two effects of cellular exposure to MTX on enzyme activity in cell free extract.
Conceivably, the decreased level of Met synthase in the presence of MTX (figs. 2 and 3) is induced either by folate depletion (tables 1 and 2; fig. 3) or accumulation of MTX polyglutamates. The former possibility agrees with published reports (Christensen et al., 1992;
Quadros and Jacobsen, 1995) demonstrating effects on Met synthase
activity from medium folate.
MTX antagonized the enzyme inactivation induced by nitrous oxide. Our
data (fig. 3; table 1) suggest that this effect is related to a
substantial reduction of 5-methyl-THF. This agrees with a previous
observation that inactivation of intracellular Met synthase by nitrous
oxide is prevented when the cells are cultured in low folate medium
(Christensen et al., 1992). According to the prevailing
model (Banerjee and Matthews, 1990), catalytic turnover is a major
determinant of the Met synthase inactivation rate, and MTX probably
acts by decreasing the availability of 5-methyl-THF and thereby
decreases the catalytic turnover of the enzyme.
Interaction between MTX and nitrous oxide.
Synergism between
MTX and nitrous oxide has been reported, including their effect on
folate distribution (Kano et al., 1981; Ermens et
al., 1991b), on proliferation of human bone marrow cells (Kano
et al., 1981) and on the antileukemic effect in a rat model (Kroes et al., 1986; Abels et al., 1990).
However, the synergistic effect of this combination is probably
critically dependent on sequence of administration (Ermens et
al., 1989a). This may be explained by our data which showed that
MTX administered before N2O protected the enzyme from inactivation by
nitrous oxide, and in line with this, the Hcy export rate did not
increase further by the combination of the two agents, than by MTX
given alone.
Conclusions and perspectives. The implications of our findings are 3-fold.
MTX inhibits the Met synthase reaction and may therefore have remote effects related to Hcy accumulation as well as Met depletion. Low Met leads to reduced S-adenosylmethionine formation. MTX may in this way inhibit the transmethylation reactions, which may be partly responsible for the antiinflammatory effect of this drug (Cronstein, 1992; Nesher
et al., 1991).
MTX severely decreases cellular CH3Cbl level, and
this may mirror an early event in the development of Cbl deficiency
reported in patients with folate deficiency (Mollin et al.,
1962) or patients receiving MTX therapy (Van der Weyden et
al., 1972). Thus, impaired Cbl function may represent an
additional component of MTX pharmacodynamics.
Pretreatment with MTX prevents irreversible inactivation of methionine
synthase by nitrous oxide. This may explain published data on the
toxicity of the drug combination being dependent on the sequence of the
administration (Ermens et al., 1989a), and should guide
strategies to obtain synergy or prevent side-effects.
| |
Acknowledgments |
|---|
The authors thank Mrs. Eli Gundersen, Elfrid Blomdal, Wenche Breyholtz, Anne Halhjem and Gry Kvalheim and Mr. Halvard Bergesen for expert technical assistance.
| |
Footnotes |
|---|
Accepted for publication May 27, 1997.
Received for publication October 4, 1996.
1 This work was supported by grants from the Norwegian Cancer Society and the Norwegian Research Council.
Send reprint requests to: Dr. Torunn Fiskerstrand Department of Pharmacology, Armauer Hansens hus, University of Bergen, N-5021 Bergen, Norway.
| |
Abbreviations |
|---|
MTX, methotrexate; Met, methionine; THF, tetrahydrofolate; Hcy, homocysteine; Cbl, cobalamin; Thd, thymidine; Hx, hypoxanthine; DMEM, Dulbecco's modified Eagle's medium; PBS, phosphate-buffered saline; HPLC, high-performance liquid chromatography.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
40 µmol/liter). The Hordaland Homocysteine study.
J. Clin. Invest.
98: 2174-2183, 1996[Medline].This article has been cited by other articles:
|
|
 |
|
  N. Di Simone, P. Riccardi, N. Maggiano, A. Piacentani, M. D'Asta, A. Capelli, and A. Caruso Effect of folic acid on homocysteine-induced trophoblast apoptosis Mol. Hum. Reprod., September 1, 2004; 10(9): 665 - 669. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kishi, J. Griener, C. Cheng, S. Das, E. H. Cook, D. Pei, M. Hudson, J. Rubnitz, J. T. Sandlund, C.-H. Pui, et al. Homocysteine, Pharmacogenetics, and Neurotoxicity in Children With Leukemia J. Clin. Oncol., August 15, 2003; 21(16): 3084 - 3091. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. Selzer, D. S. Rosenblatt, R. Laxova, and K. Hogan Adverse Effect of Nitrous Oxide in a Child with 5,10-Methylenetetrahydrofolate Reductase Deficiency N. Engl. J. Med., July 3, 2003; 349(1): 45 - 50. [Full Text] [PDF]
|
|
|
|
 |
|
 A. B. Guttormsen, P. M. Ueland, P. E. Lonning, O. Mella, and H. Refsum Kinetics of Plasma Total Homocysteine in Patients Receiving High-Dose Methotrexate Therapy Clin. Chem., September 1, 1998; 44(9): 1987 - 1989. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
P < .001) as a function of MTX concentrations as revealed by
analysis of variance.