Search

The multigene family which codes for the mouse major urinary proteins (MUPs) consists of approximately 35 genes. Most of these are members of two different groups, Group 1 and Group 2, which can be distinguished by nucleic acid hybridisation. Here we describe the structure of a Group 1 gene and show that two size classes of MUP mRNA which are found in mouse liver result from different splicing events in the 3'-non-coding region and contain different polyadenylation sites. Short mRNA is approximately 750 nucleotides long, contains six exons, and is the main product of the Group 2 genes. Long mRNA is approximately 880 nucleotides long, contains seven exons and is the main product of the Group 1 genes. Five exons and part of the sixth are common to long and short mRNA and contain the coding region. This codes for an acidic protein of 180 amino acids containing an 18 residue signal peptide. A comparison of the mouse sequence with a homologous rat alpha 2u-globulin sequence shows that the rate of evolutionary divergence of the two proteins has been high. Silent sites have diverged four times more rapidly than replacement sites, showing that there has been selection against change in the protein sequence.

Fourteen different major urinary protein (MUP) genomic clones from BALB/c mice were isolated. By restriction site mapping, six of these form two sets of three overlapping clones. By the criterion of cross-hybridization, the 10 different genes fall into two groups of four (Group 1) and three (Group 2) genes, while three genes fall into neither group. Southern blot analysis of genomic DNA with Group 1 and Group 2 plasmid subclones shows that the haploid mouse (BALB/c) genome contains approximately 15 Group 1 genes, 12 Group 2 genes and at least seven MUP genes that belong to neither group. An analysis of mouse-Chinese hamster hybrid cell lines shows that most, if not all, Group 1 and Group 2 genes are located on mouse chromosome 4.

Here we compare the exonic sequences of four Group 1 mouse major urinary protein (MUP) genes and four Group 1 cDNA sequences. These define seven different nucleotide sequences which differ from each other by 0.35% of bases on average, and which would code for seven different MUP proteins that could probably be resolved physically into at least five classes. The sequences differ at 13 nucleotide positions and at six codons, and although they are closely related their descent cannot be described by a simple series of duplications. We also describe the sequence of another liver cDNA (pMUP15) which has diverged from the Group 1 consensus sequence in 14.6% of bases. The divergence is much greater over exons 1-3 than over exons 4-6, suggesting that an ancestral gene conversion event has occurred. pMUP15 also differs from the Group 1 genes in having a longer signal peptide sequence and a different splice configuration between exons 6 and 7. Unlike the Group 1 sequences, pMUP15 contains a potential N-linked glycosylation site. Other published work has shown that a shorter cDNA clone which is identical over their common sequence to pMUP15 codes for MUP proteins that are unusually large in size and acidic in pI. We show here that mouse urine does indeed contain a glycosylated MUP protein with those properties, presumably the product of the gene that corresponds to pMUP15.