Download Enzymes of nucleotide biosynthesis: differences between intact and

The Enzymes of Nucleotide
Biosynthesis
Molecular Enzymology Group Colloquium Organized and Edited by E. A. Carrey (Department of Biochemistry,
University of Dundee) and H. A. Simmonds (Purine Laboratory, UMDS, London) and Sponsored by Cancer
Research Campaign, Roche Products Ltd and Waters Chromatography. 655th Meeting held at the University of
Manchester, 18-2 I July, I995
Enzymes of nucleotide biosynthesis: differences between intact and lysed cells as well
as between species and tissues can be important
H. A. Simmonds
Purine Research Laboratory, UMDS Guy’s Hospital, London SE I 9RT, U.K.
action between the synthetic and salvage pathways in normal humans comes from studies in
patients with genetic defects of enzymes catalysing individual steps in these pathways. From
these patients we have learnt that salvage of
purines is essential for the control of de novo
synthesis and for as yet unexplained reasons for
the brain, but not for the immune system
(reviewed in [2]). However, removal of the toxic
purine waste from DNA turnover is vital to the
immune system, and indeed to all rapidly dividing cells, to ensure a balanced supply of precursors for the cell-cycle-specific enzyme
ribonucleotide reductase. T h e pyrimidine disorders confirm the importance of de nouo synthesis
and salvage to the immune and haematopoietic
systems, as well as to the central nervous system.
T h e fact that hereditary oroticaciduria does
respond to life-long uridine indicates that pyrimidine salvage can compensate for the genetic
lack of de novo synthesis, due to blockage at the
level of UMP synthase, even in rapidly dividing
cells such as lymphocytes.
T h e anucleate erythrocyte is an interesting
example of cell co-operativity [1,2]. It does not
salvage uridine, but readily picks up orotic acid,
which it recycles to deliver as uridine to other
tissues. T h e reverse is true of nucleated cells,
which salvage uridine but not orotic acid. T h e
erythrocyte also picks up adenosine, which, in
the absence of de novo synthesis, it uses to sustain its ATP [2]. T h e malarial parasite, as described in the paper by Christopherson and
co-workers, can only synthesize pyrimidines de
novo, and not purines; it is dependent on salvage
for purines, but not pyrimidines. Again these are
Purine and pyrimidine nucleotides play a crucial
role in virtually all biological processes. As described in the following presentations, these
nucleotides are derived exclusively from endogenous sources in humans: a combination of de
nouo synthesis, i.e. multistep synthetic pathways
utilizing
simple
precursors
(ammonia,
bicarbonate, amino acids, etc.), and ‘salvage’, i.e.
the recycling of preformed purine bases or pyrimidine nucleosides, with salvaged nucleotides
normally exerting feedback control on de nooo
synthesis [ 11.
This colloquium focuses on structural and
functional studies of enzymes controlling individual steps in the pathways of nucleotide biosynthesis, and the application of this information to
the development of more effective anti-proliferative, immunosuppressive and anti-parasitic
drugs. This is a worthy goal, for to understand
the abnormal requires a clear picture of the
normal. What is not always appreciated is that
there are considerable differences between prokaryotes and eukaryotes, as well as between different mammalian species, and in particular
tissue-specific differences in enzyme expression
within a given species [2]. For instance, microorganisms salvage pyrimidines at the base level
and purines at the nucleoside level, the converse
of the situation in humans, representing an
important focus for devising therapy that is
selectively lethal to the invader but not to the
host.
Another problem is that information gleaned
from disrupted cells, or from studies in uitro, is
not necessarily applicable to the in oivo situation
[ 13. Much of our knowledge relating to the inter-
877
I995
Biochemical Society Transactions
878
important differences which are being exploited
therapeutically [3].
It is a commonly held belief that nucleated
cells have a requirement for a balanced supply of
purine and pyrimidine nucleotides, but is this
true for all cells? Our own recent studies [4]
accord with those of others [5] and indicate that
this does not apply to phytohaemagglutininstimulated T-lymphocytes, where expansion of
the cytosolic pyrimidine nucleotide pool is up to
4-fold greater than for purines. T h e importance
of de novo synthesis for this is highlighted by (a)
the arrest in the expansion of all pools from 24 h
by the glutamine antagonist azaserine, indicating
complete blockage of pyrimidine, NAD’, and
purine synthesis; and (b) the G T P depletion and
cell cycle arrest at 48 h induced by ribavirin, with
pyrimidine nucleotide depletion evident at 72 h.
This confirms that T-lymphocytes need to activate their synthetic pathways to provide the additional ribonucleotide precursors necessary not
only for RNA and DNA synthesis, but also for
the other processes associated with cell division,
e.g. the purine and pyrimidine sugars essential
for membrane lipid synthesis and protein glycosylation, and the pyrimidine nucleotides required
for strand-break repair.
T h e human lymphocyte has been of particular interest because of its role in the immune
response, and it may be an example of a situation
where the allosteric controls on de novo synthesis are quite different from those in other cells.
In contrast to most cells where phosphoribosylpyrophosphate synthase (EC 2.7.6.1) is normally
subject to feedback inhibition by both adenosine
and guanosine nucleotides, this enzyme in
human lymphocytes is seemingly inhibited by
adenosine nucleotides, but stimulated by guanosine nucleotides [6]. Consequently, inhibitors of
G T P synthesis have been the focus of strategies
to induce immunosuppression in organ transplantation, and several novel drugs
- are now in
clinical trial [6,7]. Numerous studies also point
to important differences in controls between
malignant and normal lymphocytes, such as the
rapid incorporation of salvaged uridine into cytidine nucleotides by malignant but not blasting
lymphocytes L8]* It is generally assumed
from studies in cultured cells that a decrease in
purine synthesis de novo results in a corresponding stimulation of pyrimidine synthesis [ 11. T h e
antagonists azaserine and ribavirin, which inhibit
de novo purine synthesis in malignant lymphocytes, have been reported to induce a comple-
Volume 23
mentary stimulation of de novo pyrimidine
biosynthesis [9,10]. This contrasts with the inhibition of both pyrimidine and purine synthesis
found by us in blasting T-lymphocytes [4] and
suggests that allosteric regulation of pyrimidine
metabolism may also differ in normal lymphocytes [6].
Another interesting question is: do viruses,
as well as bacterial and parasitic infections, alter
the purine and pyrimidine metabolism of the
host cell in a manner that is detrimental to
normal function? Our recent studies in HIV-1’
lymphocytes suggests that they do. A striking
abnormality already detectable in cells from
asymptomatic patients was the impaired ability of
pyrimidine pools to expand in response to mitogen stimulation, although purine responses were
normal, while in cells from symptomatic patients
de novo synthesis of both purines and pyrimidines was seriously impaired [ l l ] . These findings underline the particular importance of
pyrimidine ribonucleotide availability to proliferating T-lymphocytes.
In summary, when studying the enzymes of
nucleotide biosynthesis it is important to remember that: (a) activity in disrupted cells may not
reflect the in vivo situation; (b) differences
between humans, micro-organisms and other
mammalian species can be considerable; (c)
there are remarkable tissue-specific variations in
enzyme expression within a given species; (d)
the requirements of cells undergoing rapid division may vary from those of normal cells, and
hence the regulatory controls might be quite different; and (e) as we shall see in the different
presentations in this colloquium, important differences may exist between normal, parasiteinfected and malignant cells which can be
exploited to the benefit of the healthy cell, while
avoiding toxic side-effects.
1 Henderson, J. F., Lowe, J. K. and Barankiewicz, J.
(1977) Ciba Found. Symp. 48, 3-23
2 Simmonds, H. A. (1994) in the Inherited Metabolic
Diseases (Holton, J. B., ed.), pp. 297-350,
Churchill Livingstone, Edinburgh
3 Gero, A. M. and O’Sullivan, W. J. (1990) Blood
Cells 16, 467-484
4 Fairbanks, L. D., Bofill, M., Ruckemann, K. and
Simmonds, H. A. (1995) J. Biol. Chem., in the
press
5 Marijnen, Y. M. T., de Korte, D., Haverkort, W. A.,
den Breejen, E. J. S., van Gennip, A. H. and Roos,
D. (1989) Biochim. Biophys. Acta 1012, 148-155
6 Allison, A. C. and Eugui, E. M. (1993) Transplant.
The Enzymes of Nucleotide Biosynthesis
Proc. 25, 8-18
7 Dayton, J. S., Turka, L. A., Thompson, C. B. and
Mitchell, B. S. (1992) Mol. Pharmacol. 41, 671-676
8 van den Berg, A. A., van Lenthe, H., Busch, S., de
Korte, D., van Kuilenburg, A. B. P. and van
Gennip, A. H. (1994) Leukaemia 8, 1375-1378
9 Lyons, S. D., Sant, M. E. and Christopherson, R. I.
(1990) J. Biol. Chem. 265, 11377-11381
10 Zimmerman, T. P. and Deeprose, R. D. (1978)
Biochem. Pharmacol. 27, 709-716
11 Bofill, M., Fairbanks, L. D., Ruckemann, K.,
Lipman, M. and Simmonds, H. A. (1995) J. Biol.
Chem., in the press
Received 21 July 1995
The carbamoyl-phosphate synthase family and carbamate kinase: structure-function
studies
V. Rubio and J. Cervera
lnstituto de lnvestigaciones Citobgicas (FIB) and CSIC, Amadeo de Saboya, 4, 460 I 0-Valencia, Spain
-
~
Enzymatic synthesis of carbamoyl
phosphate
Carbamoyl phosphate is synthesized irreversibly by carbamoyl-phosphate synthase (CPS;
EC6.3.4.16 and EC 6.3.5.5) in the first step of
the routes of biosynthesis of pyrimidines, arginine and urea. In most bacteria, including
Escherichia cofi, the same CPS is used for arginine and pyrimidine synthesis, whereas in higher
organisms different CPS enzymes are used for
making pyrimidines and argininelurea.
Carbamoyl phosphate is also synthesized by
carbamate kinase, an enzyme found in some
bacteria that grow anaerobically using arginine as
the major nutrient. Carbamate kinase uses ATP
to phosphorylate carbamate reversibly, but its
physiological role is to generate ATP from ADP
and carbamoyl phosphate. T h e carbamoyl phosphate used in this reaction is generated from
citrulline and phosphate by catabolic ornithine
transcarbamylase.
T h e reactions catalysed by carbamate kinase
( 1 ) and CPS (2) are compared below:
ATP+NH,COO-t-,
-
ADP +NH,COOPO,H2ATP+HOCOO-+NH,
(1)
2ADP+NH,COOPO3H-+P, (2)
Carbamate is in chemical equilibrium with
bicarbonate and ammonia, but the equilibrium
does not favour its synthesis [ 11. Compared with
reaction ( l ) , reaction (2) hydrolyses an extra
ATP molecule which is used to displace the
equilibrium in the direction of carbamate syntheAbbreviation
synthase.
used:
CPS,
carbamoyl-phosphate
sis. Carbamate synthesis by CPS occurs in two
steps: (1) phosphorylation of bicarbonate by ATP
and
(ATP HCO; +ADP HOCOOPO,H-);
(2) attack of the resulting carboxyphosphate
by ammonia (HOCOOP03H- NH3+P, NH2COO-). T h e carbamate formed in this way is
then phosphorylated in a final step which is identical to reaction (1). In addition to these steps, all
CPSs except the enzyme from ureotelic organisms hydrolyse glutamine as the source of the
NH3 used in the reaction.
+
+
+
+
General structure of CPS enzymes
In correspondence with the multiplicity of steps
catalysed, CPS has a modular design, with different domains being involved in the various
reaction steps. All well characterized CPSs share
a basic structural plan and exhibit considerable
sequence identity [2], irrespective of their origin
or functional role. Bacterial CPSs, of which the
E. coli enzyme is the best studied [3], are composed of two subunits of about 40 and 120 kDa,
associated by non-covalent bonds (Figure 1).
T h e small subunit is responsible for hydrolysing
glutamine, and the large subunit catalyses the
entire reaction (2) and binds the effectors that
modulate activity. In the ureotelic enzyme the
two subunits are fused into a single polypeptide
of about 160 kDa [4]. T h e pyrimidine-specific
enzyme of eukaryotes is the largest enzymic
component of a 243 kDa trifunctional polypeptide
known as CAD, which also incorporates aspartate transcarbamylase, dihydro-orotase, and a
relatively long linker which is essential for
channelling carbamoyl phosphate to the transcarbamylase [S] . In yeast the product of the Uru2
gene is similar to CAD, although the dihydroorotase component is inactive [6].
I995
879