Development of an effective vaccine against

Mammaglobin, a promising diagnostic marker for breast cancer, forms a covalent complex with lipophilin B. mRNA levels for each component of the complex were determined for a number of breast tumors and normal tissues, and correlation of message expression was highly significant between mammaglobin and lipophilin B (p < 0.0001). The complex was purified by both standard biochemical techniques and immunoaffinity chromatography. N-Terminal sequencing revealed that mammaglobin and lipophilin B are processed as predicted by cleavage of their signal sequence after amino acids 19 and 21, respectively. Three molecular masses-representing the fully glycosylated form, the complex without one of the carbohydrate chains, and the deglycosylated proteins-are detected by ProteinChip array SELDI-TOF mass spectrometry after partial enzymatic deglycosylation. This is consistent with the two predicted N-linked glycosylation sites in the primary sequence of mammaglobin and each site having an attached sugar of approximately 3500 Da. Reducing agents release lipophilin B from mammaglobin, and the free peptides are seen at their predicted molecular masses in the deglycosylated complex. Molecular modeling, secondary structure prediction, and circular dichroism indicate that the complex is a small alpha-helical globule that has three disulfide bridges and a carbohydrate chain at each pole. LC-ESI-MS shows that mammaglobin and lipophilin B are bonded in a head to tail orientation. This work describes the biochemistry of the mammaglobin/lipophilin B complex and lays the framework for use of this complex as a novel protein-based serological marker for breast cancer.

No abstract

Fibres from human hair and wool are characterised by two main types of proteins: intermediate filament proteins (IFPs) and keratin associated proteins (KAPs). The IFPs, comprising over 50% of the fibre, tend to dominate 2-D electrophoretic maps, hindering identification of the less-abundant KAPs. This has been compounded in wool fibres by the relatively limited amount of sequence information available, with approximately 35 distinct protein sequences from ten KAP families being available, in contrast to human hair, where the sequences from well over 80 proteins from 26 KAP families are known. Additional complications include the high degree of homology within these families, ranging from 70 to 95%, and the dominance of cysteine residues in a number of KAP families with their high propensity to form cross-links. The lack of sequence information for wool KAPs has been partly overcome through the recent acquisition of new sequences. Fractionation of the proteins on the basis of their solubility with pH, urea and DTT concentration has resulted in protein extracts in which the IFP concentration has been considerably reduced. These improvements have enabled the identification of low-abundance proteins in 2-D electrophoretic maps and represent a significant advance in our knowledge of the wool proteome.