516
14 Glycolysis
diffusion pathways and interactions between the
enzymes [10]. Most glycolytic enzymes are found
exclusively in the glycosomes; only PGK and
GAPDH have additional cytosomal isoenzymes.
In addition to the glycolytic enzymes, the glycosomes contain further specific proteins [124]. All
glycosomal proteins are formed on free ribosomes
in the cytoplasm and are then introduced into the
glycosomes. This has led to the question of the
topogenic signals that direct this intracellular
transport. Answers have been sought by comparing the glycosomal enzymes with the homologous
cytoplasmic enzymes of mammals and, where
appropriate, with those of the Kinetoplastida.
Only two of the glycosomal enzymes of T. brucei
show any differences to those of other organisms.
The hexokinase of T. brucei is a hexamer of 51kDa subunits, whereas the mammalian hexokinase is a monomer of 80-90 kDa. PFK is tetrameric as in the mammals, but the subunits have a
mass of only 51 instead of 96-100 kDa. The other
glycolytic enzymes of T. brucei have the same quaternary structure as those from mammalian cells
but, with the exception of triose phosphate isomerase, have consistently 1-5-kDa larger subunits.
All the enzymes of the glycosomes (again with the
exception of triose phosphate isomerase) are
strongly basic proteins with pI values between 8.8
and 10.2, i.e. 1-4 pH units higher than the corresponding mammalian enzymes [2, 25,93, 162]. In
contrast to the chloroplast, mitochondrial and secretory proteins, the glycosomal enzymes are not
subjected to proteolytic processing when they
reach the target organelle; the topogenic signals
are thus located in the internal structure of the
chain. According to the hypotheses of the Weirenga group, the signal consists of one or two
closely situated, positively charged amino acid
residues ("hot spots"). Recent results, especially
with Crithidia fasciculata, have cast some doubt
on this hot-spot theory [117, 181]. In addition to
the glycosomal phosphoglycerate kinase (gPGK),
T. brucei and C. fasciculata have a cytoplasmic
PGK (cPGK) which is encoded by its own gene.
The gPGK of T. brucei has a C-terminal extension
of 20 predominantly basic amino acids but agrees
93 % with the cPGK. The C-terminal extension of
the gPGK C. fasciculata has a length of 38 amino
acids and consists mostly of hydrophobic residues
and hydroxyamino acids; the N-terminal sequence
has only two amino acid substitutions compared
with the N-terminal sequence of cPGK. The aldolase and triose phosphate isomerase of T. brucei
have no C-terminal extensions. Thus, there is
apparently no universal topogenic signal for
import into the glycosomes [181]. A polycistronic
transcript is produced from the T. brucei PGK
locus and this is then processed to give the
mRNAs of the individual enzymes. In mammalian
blood, the mRNA and protein of gPGK predominate, whereas in culture (which corresponds to
the forms from the tsetse fly vector) the cPGK
forms predominate [77].
A ubiquitous and intensively investigated
enzyme system such as that of glycolysis provides
an excellent opportunity for posing the following
basic questions about enzyme evolution: Are
enzymes which catalyse the consecutive steps of a
reaction chain homologous, perhaps by derivation from a common gene? Can one assume high
sequence agreement between enzymes of similar
reaction specificity or which have similar ligands?
The amino acid sequences and spatial structures
are known for almost all glycolysis enzymes, and
the rates of evolution of the enzymes appear to be
relatively low (see Table 4.12, p.161). Of the
seven glycolytic enzymes that bind ATP or
NAD+, GAPDH, lactate dehydrogenase (LDH)
and alcohol dehydrogenase (ADH) have similar
spatial structures with an NAD-binding and a
catalytic domain on either side of the active
centre. The coenzyme-binding domains all consist
of two mononucleotide-binding folds with very
similar conformations (Rossmann folds). PGK
also includes two such folds, whereas pyruvate
kinase (PK) has only one, and in hexokinase and
PFK there are none. Despite the similarities of
spatial structure, the mononucleotide-binding
folds of the different enzymes show no sequence
similarity. In the glycolysis chain, there are several groups of enzymes with similar reaction specificities: the kinases, mutases, isomerases and
dehydrogenases. However, there is no significant
sequence similarity within such functional types;
in fact, even enzymes with the same specificity in
different organisms are not homologous: the
ADH of Drosophila is not homologous to the
enzymes from yeast or mammals; among all organisms, there are two types of aldolase; and in the
animal kingdom there are two types of phosphoglycerate mutase (PGM) with very different properties. Thus, the enzyme system of glycolysis
apparently arose early in evolution by the combination of independently derived enzymes; the
present structural similarities are examples of
convergence [59].
The Ubiquitous reaction pathways of primary
cellular metabolism are adapted to the different
metabolic requirements of different animal
groups and cell types. This occurs on the one
14 Glycolysis
diffusion pathways and interactions between the
enzymes [10]. Most glycolytic enzymes are found
exclusively in the glycosomes; only PGK and
GAPDH have additional cytosomal isoenzymes.
In addition to the glycolytic enzymes, the glycosomes contain further specific proteins [124]. All
glycosomal proteins are formed on free ribosomes
in the cytoplasm and are then introduced into the
glycosomes. This has led to the question of the
topogenic signals that direct this intracellular
transport. Answers have been sought by comparing the glycosomal enzymes with the homologous
cytoplasmic enzymes of mammals and, where
appropriate, with those of the Kinetoplastida.
Only two of the glycosomal enzymes of T. brucei
show any differences to those of other organisms.
The hexokinase of T. brucei is a hexamer of 51kDa subunits, whereas the mammalian hexokinase is a monomer of 80-90 kDa. PFK is tetrameric as in the mammals, but the subunits have a
mass of only 51 instead of 96-100 kDa. The other
glycolytic enzymes of T. brucei have the same quaternary structure as those from mammalian cells
but, with the exception of triose phosphate isomerase, have consistently 1-5-kDa larger subunits.
All the enzymes of the glycosomes (again with the
exception of triose phosphate isomerase) are
strongly basic proteins with pI values between 8.8
and 10.2, i.e. 1-4 pH units higher than the corresponding mammalian enzymes [2, 25,93, 162]. In
contrast to the chloroplast, mitochondrial and secretory proteins, the glycosomal enzymes are not
subjected to proteolytic processing when they
reach the target organelle; the topogenic signals
are thus located in the internal structure of the
chain. According to the hypotheses of the Weirenga group, the signal consists of one or two
closely situated, positively charged amino acid
residues ("hot spots"). Recent results, especially
with Crithidia fasciculata, have cast some doubt
on this hot-spot theory [117, 181]. In addition to
the glycosomal phosphoglycerate kinase (gPGK),
T. brucei and C. fasciculata have a cytoplasmic
PGK (cPGK) which is encoded by its own gene.
The gPGK of T. brucei has a C-terminal extension
of 20 predominantly basic amino acids but agrees
93 % with the cPGK. The C-terminal extension of
the gPGK C. fasciculata has a length of 38 amino
acids and consists mostly of hydrophobic residues
and hydroxyamino acids; the N-terminal sequence
has only two amino acid substitutions compared
with the N-terminal sequence of cPGK. The aldolase and triose phosphate isomerase of T. brucei
have no C-terminal extensions. Thus, there is
apparently no universal topogenic signal for
import into the glycosomes [181]. A polycistronic
transcript is produced from the T. brucei PGK
locus and this is then processed to give the
mRNAs of the individual enzymes. In mammalian
blood, the mRNA and protein of gPGK predominate, whereas in culture (which corresponds to
the forms from the tsetse fly vector) the cPGK
forms predominate [77].
A ubiquitous and intensively investigated
enzyme system such as that of glycolysis provides
an excellent opportunity for posing the following
basic questions about enzyme evolution: Are
enzymes which catalyse the consecutive steps of a
reaction chain homologous, perhaps by derivation from a common gene? Can one assume high
sequence agreement between enzymes of similar
reaction specificity or which have similar ligands?
The amino acid sequences and spatial structures
are known for almost all glycolysis enzymes, and
the rates of evolution of the enzymes appear to be
relatively low (see Table 4.12, p.161). Of the
seven glycolytic enzymes that bind ATP or
NAD+, GAPDH, lactate dehydrogenase (LDH)
and alcohol dehydrogenase (ADH) have similar
spatial structures with an NAD-binding and a
catalytic domain on either side of the active
centre. The coenzyme-binding domains all consist
of two mononucleotide-binding folds with very
similar conformations (Rossmann folds). PGK
also includes two such folds, whereas pyruvate
kinase (PK) has only one, and in hexokinase and
PFK there are none. Despite the similarities of
spatial structure, the mononucleotide-binding
folds of the different enzymes show no sequence
similarity. In the glycolysis chain, there are several groups of enzymes with similar reaction specificities: the kinases, mutases, isomerases and
dehydrogenases. However, there is no significant
sequence similarity within such functional types;
in fact, even enzymes with the same specificity in
different organisms are not homologous: the
ADH of Drosophila is not homologous to the
enzymes from yeast or mammals; among all organisms, there are two types of aldolase; and in the
animal kingdom there are two types of phosphoglycerate mutase (PGM) with very different properties. Thus, the enzyme system of glycolysis
apparently arose early in evolution by the combination of independently derived enzymes; the
present structural similarities are examples of
convergence [59].
The Ubiquitous reaction pathways of primary
cellular metabolism are adapted to the different
metabolic requirements of different animal
groups and cell types. This occurs on the one