Structure of Transferrins

Determination of the amino-acid sequence of human serum transferrin (MacGillivray et al., 1983) and of human lactoferrin (Metz-Boutique et al., 1984) revealed an internal two-fold sequence repeat. The amino-terminal half has approximately 40 % sequence identity with the carboxyl-terminal half. Similar results have subsequently been found for a number of other transferrins (Baldwin, 1993), suggesting that the modern transferrins have evolved by a gene duplication event from an ancestral gene coding for a protein of molecular weight around 40 kD and with a single metal-binding site. Subsequent duplications of this gene would have given rise to the transferrins, lactoferrins, ovotransferrins and melanotransferrins that we find today. Crystallographic studies, on human lactoferrin (Anderson et al., 1987, 1989), human serotransferrin (Bailey et al., 1988) and ovotransferrin (Kurokawa et al., 1995) have shown that the polypeptide chain is folded into two structurally homologous lobes referred to as the N- and C-lobes. The same folding pattern, found in a number of other transferrins as well as

Lactoferrin Structure

Figure 5.1 Schematic diagram of the lactoferrin molecule. The positions of carbohydrate attachment are marked with a star. O, ovotransferrin; T, human serotransferrin; L, human lactoferrin; R, rabbit serotransferrin; M, melanotransferrin; A, the connecting helix; B, the C-terminal helix. The disulfide bridges are indicated by heavy bars, and the iron and carbonate binding sites by filled or open circles, respectively. Reprinted from Baker et al., 1987. Copyright (1987), with permission from Elsevier Science.

Figure 5.1 Schematic diagram of the lactoferrin molecule. The positions of carbohydrate attachment are marked with a star. O, ovotransferrin; T, human serotransferrin; L, human lactoferrin; R, rabbit serotransferrin; M, melanotransferrin; A, the connecting helix; B, the C-terminal helix. The disulfide bridges are indicated by heavy bars, and the iron and carbonate binding sites by filled or open circles, respectively. Reprinted from Baker et al., 1987. Copyright (1987), with permission from Elsevier Science.

various fragments, mutants, and metal- and anion-substituted proteins, is shown in schematic form for human lactoferrin in Figure 5.1. The organization of the human serotransferrin and hen ovotransferrin genes are consistent with this evolutionary history (Park et al., 1985; Schaeffer et al., 1987). Each gene contains 17 exons separated by 16 introns. Exon 1 codes for a signal peptide, while the remaining 16 exons code for the mature transferrin molecule. Fourteen of these exons constitute seven homologous pairs, coding for corresponding regions in each of the lobes, while the remaining two are unique to the C-terminal lobe. The N-terminal lobe in Figure 5.1 is at the top and the C-terminal lobe at the bottom. They are both made up of some 330 amino acids, joined by a short connecting peptide which, in lactoferrin, forms a three-turn a-helix. Each lobe contains an iron binding site located deep in the cleft between two dissimilar a/b-globular domains, 1 and 2, each of about 160 residues, that follow the folding pattern shown in Figure 5.2. All four domains have a similar super-secondary structure consisting of a central core of five or six irregularly twisted b-sheets of similar topology, with a-helices packed on either side of them. The first 90 or so residues form part of domain 1. The polypeptide chain then crosses to domain 2 via a long b-strand, e. The next 160 residues form the whole of domain 2, and the chain then returns via a long b-strand, j, to complete the folding of domain 1, and finally an a- helix runs back across the domain interface

Helix Strand
Figure 5.2 Schematic representation of the folding pattern for the N-lobe (left) and C-lobe (right) of human lactoferrin. From Anderson et al., 1989. Reproduced by permission of Academic Press.

to domain 2. The pair of antiparallel b-strands e and j, which run behind the iron binding site, forms a flexible hinge between the two domains. Disulfide bridges, six in the N-lobe and ten in the C-lobe are represented as heavy bars. Many of these disulfide bridges are conserved in other transferrins, but none of them crosses from one lobe to the other, which explains how single half-molecules can be isolated from many species of transferrin by proteolytic cleavage in the helix A which joins the N- and C-lobes.

The determination of the structure of the iron transporter, ferric-binding, protein (hFBP)i from Haemophilus influenzae (Bruns et al., 1997) at 0.16 nm resolution shows that it is a member of the transferrin superfamily, which includes both the transfer-rins and a number of periplasmic binding proteins (PBP). The PBPs transport a wide variety of nutrients, including sugars, amino acids and ions, across the periplasm from the outer to the inner (plasma) membrane in bacteria (see Chapter 3). Iron binding by transferrins (see below) requires concomitant binding of a carbonate anion, which is located at the N-terminus of a helix. This corresponds to the site at which the anions are specifically bound in the bacterial periplasmic sulfate- and i Gram-negative pathogenic bacteria such as Haemophilus, Neisseria, Serratia and Yersinia acquire free iron directly from the transferrin (Tf) or lactoferrin (Lf) of their host. Fe3+ is extracted from Tf or Lf at the outer membrane by receptor proteins which are specific for Tf (Tbpl and Tbp2) or Lf (Lbpl and Lbp2), and transported to an inner membrane permease by a periplasmic ferric binding protein (Fbp).

Gene duplication and fusion event

Common ancestor

Gene duplication and fusion event

Common ancestor

Transferrin Structure

Transferrin

Ovotransferrin

Spermidine-binding protein

Sulphate-binding protein

Maltodextrin-binding protein

Fe+3-binding protein (FBP)

Lactoferrin (LF)

Figure 5.3 The deduced evolutionary tree for selected members of the transferrin superfamily, based on comparisons of structures and sequences. The tree combines the transferrins with a number of prokaryotic periplasmic transport proteins. From Bruns et al., 1997. Reproduced by permission of Nature Publishing Group.

Transferrin

Ovotransferrin

Spermidine-binding protein

Sulphate-binding protein

Maltodextrin-binding protein

Fe+3-binding protein (FBP)

Lactoferrin (LF)

Phosphate-binding protein

Figure 5.3 The deduced evolutionary tree for selected members of the transferrin superfamily, based on comparisons of structures and sequences. The tree combines the transferrins with a number of prokaryotic periplasmic transport proteins. From Bruns et al., 1997. Reproduced by permission of Nature Publishing Group.

phosphate-binding proteins (Pflugrath and Quiocho, 1985; Luecke and Quiocho, 1990). This led to the suggestion that the transferrins could have arisen by divergent evolution and subsequent duplication from an anion-binding precursor common to the transferrins and the periplasmic binding proteins (Baker et al., 1987).

HFBP is a member of the class II periplasmic binding proteins, which include those specific for phosphate (Pbp), sulfate and maltodextrin, and Figure 5.3 presents the deduced evolutionary tree for selected members of the transferrin superfamily, based on comparison of structures and sequences. The polypeptide topology of hFBP and the N-lobe of human lactoferrin are compared in Figure 5.4 (Plate 5). The apparent homology between the transferrins and the periplasmic binding proteins implies evolution from a common ancestor that existed prior to the divergence of prokaryotes and eukaryotes, about 1500 million years ago. Despite their structural similarities the periplasmic binding proteins have less than 20 % sequence identity with one another and less than 10 % identity with the transferrins.

Was this article helpful?

0 0

Post a comment