Gene duplications and sequence polymorphism of bovine class II DQB genes

The genetic diversity of bovine class II DQB genes was investigated by polymerase chain reaction amplification and DNA sequencing. The first domain exon was amplified from genomic DNA samples representing 14 class II haplotypes, defined by restriction fragment length polymorphism (RFLP) analysis. The presence of a polymorphism in the copy number of DQB genes was confirmed since two DQB sequences were isolated from certain haplotypes. Four subtypes of bovine DQB genes were found. DQB1 is the major type and was found in almost all haplotypes. DQB2 is very similar to DQB1 but was found only in the duplicated haplotypes DQ9 to 12. DQB3 and DQB4 are two quite divergent genes only present in certain duplicated haplotypes. The bovine DQB complexity thus resembles that in the human DRB region. Bovine DQB genes were found to be highly polymorphic as ten DQB1 alleles and four DQB2 alleles were identified. The observed sequence polymorphism correlated well with previously defined DQB RFLPs. Bovine and human DQB alleles show striking similarities at the amino acid level. In contrast, the frequency of silent substitutions is much higher in comparisons of DQB alleles between species than within species ruling out the possibility that any of the contemporary DQB alleles have been maintained since the divergence of humans and cattle. The frequency of silent substitutions between DQB alleles was markedly lower in cattle than in humans, in agreement with a previous comparison of human and bovine DRB alleles.     

Gene duplications and sequence polymorphism of bovine class II DQB genes

The genetic diversity of bovine class II DQB genes was investigated by polymerase chain reaction amplification and DNA sequencing. The first domain exon was amplified from genomic DNA samples representing 14 class II haplotypes, defined by restriction fragment length polymorphism (RFLP) analysis. The presence of a polymorphism in the copy number of DQB genes was confirmed since two DQB sequences were isolated from certain haplotypes. Four subtypes of bovine DQB genes were found. DQB1 is the major type and was found in almost all haplotypes. DQB2 is very similar DQB1 but was found only in the duplicated haplotypes DQ9 to 12. DQB3 and DQB4 are two quite divergent genes only present in certain duplicated haplotypes. The bovine DQB complexity thus resembles that in the human DRB region. Bovine DQB genes were found to be highly polymorphic as ten DQB1 alleles and four DQB2 alleles were identified. The observed sequence polymorphism correlated well with previously defined DQB RFLPs. Bovine and human DQB alleles show striking similarities at the amino acid level. In contrast, the frequency of silent substitutions is much higher in comparisons of DQB alleles between species than within species ruling out the possibility that any of the contemporary DQB alleles have been maintained since the divergence of humans and cattle. The frequency of silent substitutions between DQB alleles was markedly lower in cattle than in humans, in agreement with a previous comparison of human and bovine DRB alleles.     

Sequence and PCR-RFLP analysis of 14 novel BoLA-DRB3 alleles

The genetic diversity of the bovine class IIDRB3 locus was investigated by polymerase chain reaction (PCR) amplification and DNA sequencing of the first domain exon. Studying 34 animals of various cattle breeds, 14 previously unrecognized DRB3 alleles were identified. In three alleles, amino acid substitutions were observed that had not been previously found in bovine DRB3, but occurred at the same position in bovine DQB and in the DRB alleles of other mammals. For all newly identified alleles, the restriction fragment length polymorphism (RFLP) patterns of PCR products obtained with the enzymes Rsa I, Bst YI, and Hae III were compared with patterns of 38 previously described alleles. Altogether, eleven novel PCR-RFLP types were defined. Twelve out of the 42 PCR-RFLP types identified so far were not found to be fully informative because they corresponded to more than one allelic sequence. PCR-RFLP may therefore be a rapid and useful method for DRB3 typing in cattle families, but for studies on outbred populations, sequencing and hybridization techniques are required.     

Polymorphism and gene organization of water buffalo MHC-DQB genes show homology to the BoLA DQB region

In cattle (Bos taurus), there is evidence of more than 50 alleles of BoLA-DQB (bovine lymphocyte antigen DQB) that are distributed across at least five DQB loci, making this region one of the most complex in the BoLA gene family. In this study, DQB alleles were analysed for the water buffalo (Bubalus bubalis), another economically important bovine species. Twelve alleles for Bubu-DQB (Bubalis bubalis DQB) were determined by nucleotide sequence analysis. A phylogenetic analysis revealed numerous trans-species polymorphisms, with alleles from water buffalo assigned to at least three different loci (BoLA-DQB1, BoLA-DQB3 and BoLA-DQB4) that are also found in cattle. These presumptive loci were analysed for patterns of synonymous (d(S)) and non-synonymous (d(N)) substitution. Like BoLA-DQB1, Bubu-DQB1 was observed to be under strong positive selection for polymorphism. We conclude that water buffalo and cattle share the current arrangement of their DQB region because of their common ancestry.     

Identification of a bovine β-mannosidosis mutation and detection of two β-mannosidase pseudogenes

β-Mannosidase deficiency results in β-mannosidosis, a severe neurodegenerative lysosomal storage disease identified in cattle, goats, and humans. To more fully understand the molecular pathology of this disease, the mutation associated with bovine β-mannosidosis was identified by sequence analysis of cDNA from an affected calf. A transition mutation of G to A at position 2574 of the cDNA coding sequence creates a premature stop codon near the 3′ end of the protein coding region. To aid commercial breeders of Salers cattle, a PCR-based test was developed to detect the mutation for β-mannosidosis carrier screening. Application of this test also revealed the presence of two β-mannosidase pseudogenes. Portions of the pseudogenes were amplified with allele-specific primers and then sequenced. One pseudogene was highly homologous (>99% sequence identity) to the expressed cDNA sequence over the 1292 bp that were sequenced, while the other showed more divergence (83% sequence identity) in the 477 bp that were sequenced. Both are processed pseudogenes that are not expressed. The severity of the bovine β-mannosidosis phenotype suggests that the 22 C-terminal amino acids of β-mannosidase play an important role in the function of this enzyme. Received: 18 June 1999 / Accepted: 13 August 1999     

A deletion of the paracellin-1 gene is responsible for renal tubular dysplasia in cattle

Various hereditary diseases analogous to particular human heritable diseases have been identified in cattle. Investigation of these cattle diseases will provide useful information regarding the pathogenesis of the corresponding human diseases. Renal tubular dysplasia is an autosomal recessive disease of Japanese black cattle characterized by renal failure and growth retardation. We have previously mapped the locus responsible for the disease within a region on bovine chromosome 1. In the present study, we further typed additional markers in this region and found that a genomic segment of bovine chromosome 1 including the microsatellite marker BMS4009 was deleted in the affected animals. Construction of a physical map covering this region with BAC clones and comparison of the nucleotide sequences of this region between normal and affected animals revealed that a region of 37 kb including exons 1 to 4 of the bovine paracellin-1 gene was deleted in the affected animals. The paracellin-1 gene, which is the causative gene for human renal hypomagnesemia with hypercaciuria and nephrocalcinosis, encodes a tight junction protein of renal epithelial cells. Therefore, we concluded that deletion of the paracellin-1 gene is responsible for renal tubular dysplasia of cattle, and the cattle disease could be a good model for the human disease.     

In vitro binding of cattle PstI SINE with a 33-kDa nuclear protein.

A PstI family of SINEs (short interspersed elements) has been identified in some of the members of the family Bovidae, for example, cattle, buffalo and goat. In vitro DNA-protein interactions were studied to provide a better understanding of the function of these SINEs in the genome. Use of one such cattle PstI interspersed repeat sequence, as a probe in gel retardation assays, has lead to the identification of a repeat DNA-binding factor PIRBP (PstI interspersed repeat binding protein) from cattle liver nuclear extract. Southwestern analysis with liver nuclear extracts from cattle, goat, and buffalo revealed the presence of a PIRBP-like nuclear factor in all three species belonging to the family Bovidae. Deletion analysis localized the PIRBP binding site to an 80-bp (337-417 bp) region within the cattle PstI sequence. UV crosslinking and Southwestern analyses clearly indicated that PIRBP is a singular, small polypeptide of 33-kDa molecular mass. Homology search of the nucleic acids database revealed that the cattle PstI sequence was associated with many different genes of the family Bovidae, either in the 5' flanking region, 5' locus activating region, 3' UTR or in intervening sequences. The binding of the cattle PstI SINE by PIRBP and its association with the regulatory regions of the genes suggests that it plays an important role in the bovine genome.     

Sequence variation in the bovine and ovine PRNP genes

A resequencing approach was adopted to identify sequence variants in the PRNP gene that may affect susceptibility or resistance to bovine spongiform encephalopathy. The entire PRNP gene (>21 kb) was sequenced from 26 chromosomes from a group of Holstein-Friesian cows, as well as exon 3 of PRNP (>4 kb) from a further 24 chromosomes from six diverse breeds. We identified 51 variant sequences of which 42 were single nucleotide polymorphisms and nine were insertion/deletion (indel) events. The study was extended to exon 3 of the sheep PRNP gene where 23 sequence variants were observed, four of which were indels. The level of nucleotide diversity in the complete bovine PRNP gene was pi = 0.00079, which is similar to that found at the bovine T-cell receptor alpha delta joining region (pi = 0.00077), but somewhat less than that observed for the bovine leptin (pi = 0.00265). Sequence variation within exon 3 of PRNP in both cattle (pi = 0.00102) and sheep (pi = 0.00171) was greater than that for the complete PRNP gene, with sheep showing greater sequence variation in exon 3 than cattle. The level of sequence variation reported here is greater than previously thought for the bovine PRNP gene in cattle. This study highlights the contribution that recombination plays in increasing allelic diversity in this species.     

Assessment of the nucleotide sequence variability in the bovine T-cell receptor alpha delta joining gene region

The sequence of 2,193 nucleotides from the bovine T-cell receptor alpha/delta joining gene region (TCRADJ) was determined and compared with the corresponding human and murine sequences. The identity was 75.3% for the comparison of the Bos taurus vs. the Homo sapiens sequence and 63.8% for the Bos taurus vs. the Mus musculus sequence. This comparison permitted the identification of the putatively functional elements within the bovine sequence. Direct sequencing of 2,110 nucleotides in nine animals revealed 12 variable sites. Estimates, based on direct sequencing in three Holstein Friesian animals, for the two measures of sequence variability, nucleotide polymorphism (u) and nucleotide diversity (p), were 0.00050 (60.00036) and 0.00077 (60.00056), respectively. The test statistic, Tajima's D, for the comparison of the two measures indicates that the difference between u and p is close to significance (P < 0.05), suggesting the possibility of selective forces acting on the studied genomic region. Allelic variation at 5 of the 12 variable sites was analysed in 359 animals (48 Anatolian Black, 56 Braunvieh, 115 Fleckvieh, 47 Holstein Friesian, 50 Simmental and 43 Pinzgauer) using the oligonucleotide ligation assay (OLA) in combination with the enzyme linked immunoabsorbant assay (ELISA). Nine unambiguous haplotypes could be derived based on animals with a maximum of one heterozygous site. Four to seven haplotypes were present in the different breeds. When taking into account the frequencies of the haplotypes in the different breeds, especially in Anatolian Black, an ancestral cattle population, we could establish the likely phylogenetic relationships of the haplotypes. Such haplotype trees are the basis for cladistic candidate gene analysis. Our study demonstrates that the systematic search of single nucleotide polymorphisms (SNPs) is useful for analysing all aspects of variability of a given genomic region.     

Determination of the sequence of the arylacetyl acyl-CoA:amino acid N-acyltransferase from bovine liver mitochondria and its homology to the aralkyl acyl-CoA:amino acid N-acyltransferase

The arylacetyl acyl-CoA:amino acid N-acyltransferase was previously purified to homogeneity from bovine liver mitochondria, and partial sequences were obtained for peptides generated by cyanogen bromide cleavage of the enzyme. One of these sequences was used to design an oligonucleotide probe that was utilized to screen a bovine liver cDNA library. Several clones were isolated and sequenced, and the sequence is given. The cDNA contains 346 bases of 5′-untranslated region and 439 bases of 3′ untranslated region. The cDNA codes for an enzyme containing 295 amino acid residues. The sequence gives a molecular weight for the enzyme of 38,937, which is larger than that previously estimated for the functional enzyme, which suggests the existence of ca. 5 kDA of signal peptide. The molecular weight of the enzyme was slightly lower than that of the aralkyltransferase, which was previously determined to be 39,229. Comparison of this sequence with that which we previously obtained for the aralkyltransferase indicated that the coding regions were of identical length and that the sequences were 78% homologous. However, the 5′ and 3′ untranslated regions had less than 29% homology. The derived amino acid sequences were 71% homologous. This high homology indicates a common origin for the two enzymes. There are, however, significant differences in amino acid compositions, and these are discussed. © 1998 John Wiley & Sons, Inc. J Biochem Toxicol 12: 275–279, 1998     

Molecular characterization of expressed DQA and DQB genes in the California sea lion ( Zalophus californianus)

To date, there are no published MHC sequences from the California sea lion (Zalophus californianus), a thriving species that, by feeding high on the marine food web, could be a sentinel for disturbances in marine and coastal ecosystems. In this study, degenerate primers and RACE technology were used to amplify near-full-length (MhcZaca- DQB) and full-length (MhcZaca- DQA) expressed class II MHC gene products from the peripheral blood mononuclear cells of two California sea lions in rehabilitation. Five unique Zaca- DQA sequences and eight unique Zaca- DQB sequences, all encoding functional proteins, were identified in the two animals, indicating the presence of multiple DQ- loci in this species. An additional three Zaca- DQB sequences containing features compatible with pseudogenes or null alleles were also identified. Despite the identification of multiple DQA and DQB sequences, the degree of heterogeneity between them was extremely low. To confirm the limited degree of Zaca-DQ nucleotide variation between individuals, we used denaturing gradient gel electrophoresis to examine putative peptide binding region sequences from the peripheral blood leukocyte-derived RNAs of 19 wild-caught California sea lions from physically distinct populations. The pattern of Zaca-DQ sequence migration was identical between individuals and independent of geographical region. This apparent Zaca-DQ sequence identity between sea lions was confirmed by direct sequencing of individual bands. In combination, these findings raise important questions regarding immunogenetic diversity within this thriving species, and should prompt further research into the existence of a highly polymorphic sea lion class II MHC molecule with sequence features that support traditional peptide binding functions.     

Cloning of the bovine major histocompatibility complex class II genes

Class II genes of the bovine major histocompatibility complex (MHC) have been cloned from a genomic library. The library was constructed in the bacteriophage lambda vector EMBL3 and comprises approximately 10 times the equivalent of the haploid genome. Half the library was screened with the human DQA, DQB, DRA and DRB cDNA probes. Of the 100 positively hybridizing phage clones, 37 were eventually fully characterized and mapped by means of Southern blot analysis. The exons encoding the first, second and transmembrane domain of all different A and B genes were subcloned and mapped in more detail. These analyses showed that these 37 clones were derived from five different A and 10 different B genes. The hybridization studies indicate that we have cloned and mapped two DQA genes, one DRA gene, two other A genes, four DQB genes, three DRB genes and three other B genes. Since the library was made from a heterozygous animal, this would suggest that there are at least one DQA, one DRA one other undefined A, two DQB, two DRB and one or two other undefined B genes in the haploid genome of Holstein Friesian cattle.     

Nucleotide sequence of bovine prolactin messenger RNA. Evidence for sequence polymorphism

Hybrid molecules containing DNA sequences complementary to bovine pituitary mRNA were constructed in the Pst I site of pBR322 by the dC . dG tailing technique. Recombinant plasmids containing bovine prolactin (bPRL) sequences were amplified in bacteria and identified by hybridization to purified [32P]bPRL cDNA sequences. Nucleotide sequence analysis was performed on the inserts from two of the positive clones. One clone, pBPRL72, contained a 982-base pair insert that included 67 nucleotides of the 5'-untranslated region, the complete coding region of the preprolactin protein (690 nucleotides), and the entire 3'-untranslated region (150 nucleotides) of bPRL mRNA. The nucleotide sequence analysis of clone pBPRL72 predicted the sequence of a 30-amino acid signal peptide and confirmed the published amino acid sequence of the protein with one exception. A comparison of the pBPRL72 cDNA sequence with a second bPRL clone, pBPRL4, revealed four silent nucleotide differences. Three of the base changes occurred in the third position of amino acid codons, and one occurred in the 3'-noncoding region. The sequence polymorphism suggests the existence of alleles or multiple loci for bPRL that do not alter the protein structure.