A proposal for standardization in forensic bovine DNA typing: allele nomenclature of 16 cattle-specific short tandem repeat loci

Introduction

The capability of PCR amplification of a multiplex of species-specific, polymorphic, short tandem repeat (STR) loci has led to the development of a robust set of 16 bovine STR markers (BM1818, BM1824, BM2113, CSRM60, CSSM66, ETH3, ETH10, ETH225, HAUT27, ILSTS006, INRA023, SPS115, TGLA53, TGLA122, TGLA126 and TGLA227). Although recently a large population study in cattle (Bos taurus) was presented for this robust set of 16 bovine-specific STR markers (Van de Goor L.H.P., Koskinen M.T. & Van Haerignen W.A., unpublished data), no details have yet been published about allele structure or DNA sequence variation in the alleles. This set of 16 markers, which includes the loci recommended by the International Society for Animal Genetics (ISAG) for bovine parentage testing, is included in commonly used commercial kits (Applied Biosystems and Finnzymes Diagnostics) and routinely employed by bovine genotyping providers. For the purpose of more effectively sharing data and for better use in legal casework, a universal nomenclature system should be developed based on the principles of the human repeat-based nomenclature advocated by the International Society of Forensic Genetics (ISFG) recommendations (Gill et al. 1994, 1997; Bär et al. 1997; Schneider et al. 1998). The commonly used current nomenclature system in cattle for parentage verification cases is based on the measured length of PCR amplicons and has evolved in an informal manner from discussions between laboratories during conferences of the ISAG. This allele nomenclature in cattle is based solely on amplicon length and is internationally regulated within ISAG. However, it has major limitations because of the ability to effectively share data generated using different electrophoretic platforms and the fact that allele designation for new STR loci or alleles will not be automatically available. Allele designations based on number of repeats have been advocated for human DNA typing, but it has yet to be universally accepted for animal genetic typing. Although a proposal for allele nomenclature based on number of repeats and an operational comparison with an allelic reference ladder was published for dogs (Canis familiaris) (Eichmann et al. 2004; Hellmann et al. 2006) and partially for cats (Felis catus) (Lipinski et al. 2007), no effort has been made to adapt the same allele nomenclature to a number of widely used important animal species such as cattle, horses and pigs. To develop a repeat-based nomenclature in cattle, a selection of alleles from a dataset containing 22 cattle breeds (N = 9.738) (Van de Goor et al. unpublished data) was sequenced for the 16 polymorphic dinucleotide STR loci. In this study, a proposal is presented for a repeat-based allele nomenclature based on sequenced alleles.

Materials and methods

Results

Previously, all STR loci included in this study were described as dinucleotide repeat sequences (Table 1 shows locus name, chromosomal location, repeat structure and sequence, original reference, primer sequences and size range of the amplicon length). All STR loci revealed a simple or compound variable repeat structure. The proposed nomenclature for the 16 bovine STR loci investigated is based on the number of repeat units and is adopted from the recommendations of the ISFG for the nomenclature of human STRs.

In the following section (1-3 and S1), the investigated STR loci are presented. They are described by (i) the UniSTS identification number (ii) the GenBank accession number, (iii) the general sequence structure including the flanking regions and (iv) the average allele frequency distribution as observed in the dataset (Van de Goor et al. unpublished data). Sequenced alleles are indicated by an asterisk (*), whereas alleles which were not sequenced have been extrapolated based on the allele mobilities from the raw data. Although deduced allele designation will only be operationally correct, additional sequencing of alleles may not be required for routine analyses as these are based on the application of control samples. If this is confirmed, the STR is described according to previous published studies.

Discussion

Dimeric STRs are used for bovine genetic identity testing. Because dimeric STRs are more susceptible than trimeric or tetrameric STRs to slippage during PCR amplification resulting in the generation of stutter bands, the human forensic community chose larger repeat motif STRs, predominately tetrameric STRs. In contrast, for parentage verifications and identity testing in bovines, historically dimeric STRs have been selected because tetranucleotide loci were not available initially. There is less need to change tousing tetrametric STRs in cattle, because of (i) the extensive, global databases with cattle genotypes based on dimeric STRs; (ii) the optimization of amplification conditions that reduce stutter production for these loci; (iii) efforts to implement Single Nucleotide Polymorphisms in cattle genotyping; and (iv) the use of the Phusion DNA-polymerase in the bovine STR panels provides proofreading activity that reduces the tendency observed with DNA-polymerases for non-template adenylation of PCR products (+A). Therefore, the current bovine STR loci are likely to be used in the foreseeable future for identity testing.

The preferred method for naming STR alleles in the human DNA forensic science community is based on the number of repeats (Gill et al. 1994, 1997; Bär et al. 1997; Schneider et al. 1998) and this is applied operationally by comparison with an allelic ladder. Previously, the motif structure had not been described for the dinucleotide STR loci used in bovine genetic identity testing. This is necessary to develop a consistent nomenclature and for generating a community-wide allelic ladder. The repeat structure of the bovine STRs has been described herein by sequence analysis of a sampling of the alleles at each locus.

Homozygous profiles of alleles at each STR locus were directly sequenced from the purified PCR product; no alleles from heterozygote profiles were used. From the 16 investigated loci, 11 showed a simple repeat structure, whereas compound repeat structures were present in the five loci ETH3, ETH225, SPS115, TGLA53 and TGLA122 (2, 3 & S1). None of the 16 STRs showed complex repeat sequences. There were a few inconsistencies in sequence with those in GenBank, but these were predominately in the flanking region and did not affect the repeat structure. The inconsistencies are primarily caused by the original sequence methods, which were used approximately 15 years ago. Allele designation was straightforward in almost all cases, as only one intermediate allele (a single nucleotide variant –SPS115 allele 27.1) was observed in our dataset of 9738 individuals. Considering the low number of intermediate alleles found in the large dataset (>9000 animals covering 22 breeds), the number of other intermediate alleles not yet identified should remain low.

In Table 3, a conversion is provided for the amplicon length-based allele nomenclature in cattle, which is internationally recommended within ISAG. One allele of STR ETH225 has been extensively discussed within the ISAG nomenclature, either being called allele 158 or 160 depending on the analytical conditions used. This allele was identified by our sequencing results to have a repeat-based nomenclature of 28.

In the human forensic community, the recommendations by ISFG (Gill et al. 1994, 1997; Bär et al. 1997; Schneider et al. 1998) include the use of standardized allelic ladders to enable designation of alleles in samples. At present, unfortunately no allelic ladders for the 16 STRs are commercially available. Thus, laboratories performing STR typing will have to base their allele calls only on an internal size standard and one control sample. However, the data from these allele sequences will be useful for future efforts to develop an allelic ladder for all 16 loci.