should soon be possible to record the total spectrum of cellular proteins. It is already possible to
separate in 2-D electrophoregrams more than
1000 proteins from cell extracts and to compare
them by computer analysis [14, 61]. The cleavage
peptides obtained from the enzymic hydrolysis of
proteins can also be separated two-dimensionally
to give a "peptide map" or a "fingerprint". Comparisons with the peptide map of a protein of
known sequence can be used to detect and localize single amino acid differences. This method is
mostly used for the comparison of closely related
proteins.
3.1.4 Folding of the Polypeptide Chain
During the biosynthesis of proteins, the 1-D nucleotide sequence of the mRNA is translated into a
1-D amino acid sequence; however, the biological
function of the protein depends finally on a particular 3-D structure (chain conformation) that
arises by a complicated folding process during the
synthesis of the polypeptide chain. Destruction of
the chain conformation (denaturation) leads to
loss of biological activity [205]. The ageing of
enzyme proteins, which can lead to reductions in
specific activity of 40-60 %, also seems to be the
result of slight alterations in chain conformation.
All the information required for the spatial
structure of a protein is contained in the amino
acid sequence. Evidence for this was already
obtained in the 1930s with the observation that
pancreas RNase denatured by urea spontaneously
renatures in urea-free solution. It was initially
thought that during the folding process many different conformations were tested until, at last,
the most thermodynamically stable one was
achieved. It was then realized that this would
require too much time; even a small protein like
the RNase would, on average, require a century
to assume its specific conformation, whereas, in
reality, the j3-galactosidase of E. coli, for example, is synthesized and correctly folded within 4
minutes, despite its large size of 500 kDa. Selection apparently favours those sequences that
allow the rapid assumption of a unique, functional conformation. Folding is kinetically determined and begins on the partially completed
chain; spontaneous renaturation is actually
observed only for a very few proteins. Initially,
very small regions are apparently folded within
microseconds into specific secondary structures,
like a-helices or j3-sheets; interactions between
such regions bring about larger, stable areas
3.1.4 Folding ofthe Polypeptide Chain
77
(domains) in which the a and 13 structures take up
particular positions. The final chain conformation
comes about by interactions between certain
binding sites on the surface of domains, or by the
formation of disulphide bridges [43, 119, 142].
With multimeric proteins, there follows the
association of the polypeptide chains to form one
of the possible quaternary structures of the protein [83,205]. About 40 % of the proteins of
mammals and bacteria are monomers; most of
the rest are even-numbered oligomers, mainly dior tetramers. Less than 5 % are trimers, for
example, ornithine transcarbamylase, arginase
and carboxylesterase of mammalian liver, the
complement components C4 and P, and the
haemerythrin of the sipunculans, or pentamers
like the C-reactive protein and the serum amyloid
protein of vertebrates. Modification of the quaternary structure is an important mechanism for
the regulation of enzyme activity. This can involve
dissociation and reaggregation of the enzyme subunits, and also reversible binding to other enzymes or structural proteins [253].
The basic unit of protein structure is the
domain, an autonomous subregion of the polypeptide chain that possesses all the characteristics
of a globular protein and can often be isolated
proteolytically without loss of its properties
[98, 245]. It seems reasonable to relate protein
domains to the exon as the structural unit of the
gene. The recombination of exons (exon shuffling) can then only be imagined as an effective
mechanism for the creation of new proteins when
structurally and functionally autonomous protein
parts are linked together in this way. In many
cases, exon and domain boundaries do actually
coincide, e.g. in the immunoglobulins, j3-globin,
lysozyme, pyruvate kinase and many membrane
proteins. There are, however, numerous examples where this is not the case, e.g. carboxypeptidase, prothrombin and other serine proteases, aamylase and carboanhydrase [5]. In such cases,
the exon boundaries may in fact correspond to
points on the surface of the globular molecule, so
that shifts, caused by mutations in splicing signals, in the exon-intron boundaries can lead to
the introduction or removal of single amino acids
or short peptides without alteration of the spatial
structure of the protein. [42].
The most important secondary structures are
the right-handed a-helix and the j3-sheet; occasionally, other helical structures occur such as the
left-handed helix or the polyproline conformation. These structures can be deformed within
certain limits: an a-helix of more than 15 amino
separate in 2-D electrophoregrams more than
1000 proteins from cell extracts and to compare
them by computer analysis [14, 61]. The cleavage
peptides obtained from the enzymic hydrolysis of
proteins can also be separated two-dimensionally
to give a "peptide map" or a "fingerprint". Comparisons with the peptide map of a protein of
known sequence can be used to detect and localize single amino acid differences. This method is
mostly used for the comparison of closely related
proteins.
3.1.4 Folding of the Polypeptide Chain
During the biosynthesis of proteins, the 1-D nucleotide sequence of the mRNA is translated into a
1-D amino acid sequence; however, the biological
function of the protein depends finally on a particular 3-D structure (chain conformation) that
arises by a complicated folding process during the
synthesis of the polypeptide chain. Destruction of
the chain conformation (denaturation) leads to
loss of biological activity [205]. The ageing of
enzyme proteins, which can lead to reductions in
specific activity of 40-60 %, also seems to be the
result of slight alterations in chain conformation.
All the information required for the spatial
structure of a protein is contained in the amino
acid sequence. Evidence for this was already
obtained in the 1930s with the observation that
pancreas RNase denatured by urea spontaneously
renatures in urea-free solution. It was initially
thought that during the folding process many different conformations were tested until, at last,
the most thermodynamically stable one was
achieved. It was then realized that this would
require too much time; even a small protein like
the RNase would, on average, require a century
to assume its specific conformation, whereas, in
reality, the j3-galactosidase of E. coli, for example, is synthesized and correctly folded within 4
minutes, despite its large size of 500 kDa. Selection apparently favours those sequences that
allow the rapid assumption of a unique, functional conformation. Folding is kinetically determined and begins on the partially completed
chain; spontaneous renaturation is actually
observed only for a very few proteins. Initially,
very small regions are apparently folded within
microseconds into specific secondary structures,
like a-helices or j3-sheets; interactions between
such regions bring about larger, stable areas
3.1.4 Folding ofthe Polypeptide Chain
77
(domains) in which the a and 13 structures take up
particular positions. The final chain conformation
comes about by interactions between certain
binding sites on the surface of domains, or by the
formation of disulphide bridges [43, 119, 142].
With multimeric proteins, there follows the
association of the polypeptide chains to form one
of the possible quaternary structures of the protein [83,205]. About 40 % of the proteins of
mammals and bacteria are monomers; most of
the rest are even-numbered oligomers, mainly dior tetramers. Less than 5 % are trimers, for
example, ornithine transcarbamylase, arginase
and carboxylesterase of mammalian liver, the
complement components C4 and P, and the
haemerythrin of the sipunculans, or pentamers
like the C-reactive protein and the serum amyloid
protein of vertebrates. Modification of the quaternary structure is an important mechanism for
the regulation of enzyme activity. This can involve
dissociation and reaggregation of the enzyme subunits, and also reversible binding to other enzymes or structural proteins [253].
The basic unit of protein structure is the
domain, an autonomous subregion of the polypeptide chain that possesses all the characteristics
of a globular protein and can often be isolated
proteolytically without loss of its properties
[98, 245]. It seems reasonable to relate protein
domains to the exon as the structural unit of the
gene. The recombination of exons (exon shuffling) can then only be imagined as an effective
mechanism for the creation of new proteins when
structurally and functionally autonomous protein
parts are linked together in this way. In many
cases, exon and domain boundaries do actually
coincide, e.g. in the immunoglobulins, j3-globin,
lysozyme, pyruvate kinase and many membrane
proteins. There are, however, numerous examples where this is not the case, e.g. carboxypeptidase, prothrombin and other serine proteases, aamylase and carboanhydrase [5]. In such cases,
the exon boundaries may in fact correspond to
points on the surface of the globular molecule, so
that shifts, caused by mutations in splicing signals, in the exon-intron boundaries can lead to
the introduction or removal of single amino acids
or short peptides without alteration of the spatial
structure of the protein. [42].
The most important secondary structures are
the right-handed a-helix and the j3-sheet; occasionally, other helical structures occur such as the
left-handed helix or the polyproline conformation. These structures can be deformed within
certain limits: an a-helix of more than 15 amino