The cDNA sequence and polymorphism of horse transferrin : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Veterinary Clinical Science at Massey University, Palmerston North, New Zealand
Transferrin, the serum protein responsible for iron transport, is highly polymorphic in many species, including the horse. In this study, the cDNA sequence of horse transferrin was determined and used to identify sequence polymorphisms which distinguish some of the many variants of horse transferrin. A horse liver cDNA library was constructed, and was screened using the human transferrin cDNA as a probe. The clones isolated gave 1800 bp of sequence. The remainder of the cDNA was obtained using PCR. The 2305 bp horse transferrin cDNA sequence included part of the 5' untranslated region and extended to the poly(A) tail. It had 80% sequence identity with the human transferrin cDNA, and encoded a protein of 706 residues, including a signal sequence of 19 amino acids. The amino acid sequence was compared to those of related proteins, i.e. the serum transferrins and lactoferrins of several species and human melanotransferrin. The horse transferrin sequence had the duplicated structure and conserved iron binding and cysteine residues which are characteristic of the transferrin family. This is consistent with the structural and functional similarities within the family. Horse transferrin has 73% amino acid sequence identity to human transferrin, 62% identity with human lactoferrin, and 53% identity with chicken transferrin. Sequence polymorphisms distinguishing the variants of horse transferrin which are found in thoroughbreds (D, F1, F2, H2, O, R and ) were identified. First, comparison of the horse transferrin cDNA and protein sequences with two reported amino acid substitutions distinguishing the D and R variants indicated that exons 12 and 15 were likely to be polymorphic. Therefore, these two regions were analysed by Southern blotting and by sequencing PCR products. The D and R variants differed by 10 nucleotide substitutions which encoded 6 amino acid substitutions. The F1, F2, H2 and variants were identical to D, and the O variant was very similar to R, in the regions studied. The data indicated that the horse transferrin variants comprise two distinct groups. Secondly, the positions of differences between the D and F1 alleles were established by producing single stranded conformation polymorphisms. Sequencing then revealed 3 nucleotide substitutions that encode 2 amino acid substitutions. All 8 of the amino acid substitutions which were observed occurred at positions which are variable between members of the transferrin family. Location of the polymorphic residues on the 3-dimensional structure of human lactoferrin revealed that all were clustered at one end of the C-lobe.