Characterisation of hG3, a cDNA isolated from a first trimester human placental library [microform] : a thesis presented in partial fulfilment of the requirements of the degree of Doctor of Philosophy in Molecular Genetics at Massey University, Palmerston North, New Zealand

Abstract

The aim of this study has been to determine if the cDNA hG3, isolated from a placental cDNA library, represents a protein expressed in the placenta. This cDNA had been isolated previously from a first trimester human placental cDNA library. The cDNA appeared to have been derived from a developmentally regulated mRNA which comprises 5 to 10% of the total placental mRNA in the first trimester but was undetectable in the tissue at partuition. The sequence of the cDNA had revealed a main open reading frame that could encode a protein of 108 amino acids. Immunoaffinity techniques were used to identify the likely hG3 encoded protein in first trimester placenta. The hG3 cDNA was subcloned into an E.coli expression vector and the protein product was used to raise antisera in rabbits. Antisera was also raised against synthetic peptides deduced from the sequence of the hG3 clone. Analysis of placental tissue at different stages of development failed to identify the expected hG3 protein. Furthermore, the proteins expressed by the E. coli expression vectors were not of the anticipated size, rather they were truncated. The accumulated evidence suggested that the cDNA sequence maybe in error. Upon resequencing the hG3 clone, a stop codon was identified in the middle of the main open reading frame. This fact, along with computer analyses and a detailed literature review, indicated that the cDNA was unlikely to encode a protein and may represent an unprocessed transcript. To investigate this hypothesis, the first trimester human placental cDNA library was rescreened with the hG3 cDNA clone (plus an oligonucleotide that represented part of hG3) in order to isolate clones which had significant homology to hG3. No such clone was isolated which suggested that the hG3 cDNA does not represent a transcript which is translated in the placenta. Sequence comparison with the July 1992 Genbank data base revealed that 90 nucleotides at the beginning of the hG3 cDNA sequence are identical to part of the 3' untranslated sequence of the human chaperonin-like protein mRNA, HTR3. Thus, the hG3 cDNA may have resulted from a rearrangement of more than one transcript during the construction of the cDNA library.