Abstract: Traditionally, parentage verification in cattle has been performed using blood typing. Parentage verification is very important, since it is the basis for breeding schemes and becomes even more critical with the widespread use of artificial insemination and embryo transfer techniques. Recently, a new methodology employing DNA markers is being used for parentage control. A panel of 9 DNA microsatellite markers was chosen internationally for evaluation in the cattle population of different countries. In order to achieve this goal, information regarding the amount of genetic variation of the microsatellites must be obtained. In the present study, the number of alleles and their frequencies were determined in 8 microsatellites from the panel (BM 1824, BM 2113, SPS 115, ETH 3, ETH 10, ETH 225, TGLA 122 and TGLA 227) using PCR and genotyping. This data was then used to calculate heterocigozity, polymorphism information content and exclusion probabilities of the microsatellites, both individually and as a group. A total of 287 animals belonging to 6 cattle breeds (Holstein, Simmental, Brown Swiss, Beefmaster, Brahman and Brangus) and localized in northeast Mexico were studied. In the Simmental breed, 3 of the 8 microsatellites were not analyzed (BM 1824, ETH 3 and TGLA 122). Considering the 8 microsatellites, exclusion probabilities were as follows: Holstein, 0.9988; Simmental, 0.9924; Brown Swiss, 0.9998; Beefmaster, 0.9999; Brahman, 0.9930; Brangus, 0.9990. All of the eight genetic markers in the 6 cattle breeds analyzed, except BM 1824 and ETH 3 in Brahman (exclusion probabilities of 0.3800 and 0.2200, respectively), as well as SPS 115 in Holstein (exclusion probability of 0.1213) were highly informative and their exclusion probabilities as a group demonstrate their usefulness for parentage verification in cattle from northeast Mexico.

INTRODUCTION

Since, 1980, parentage testing in cattle has been done by blood typing, but it was replaced in the mid-90’s for the so called DNA test or genotyping. The simultaneous use of several genetic markers, named DNA microsatellites, allows the achievement of a higher than 99.9% reliability for genealogy verification. It has been observed than with the use of 22 microsatellites, an exclusion probability of 0.9999 can be obtained, which means that the probability of excluding as progenitor an individual that is not the actual progenitor is of 99,999%. Laboratory materials and protocols for genotyping are easily available, results are comparable among different laboratories and different biological samples can be used for analysis (Glowatzki-Mullis et al., 1995; Heyen et al., 1997; Vankan and Faddy, 1999).

To determine the degree of usefulness of microsatellites for parentage testing in a certain population, the number and frequency of their alleles must be evaluated. With this information it is possible to calculate their heterozygosis and degree of polymorphism, which in turn allows the determination of the exclusion probability of each individual microsatellite, as well as of a panel of microsatellites as a group. Exclusion probability indicates the possibility of excluding as progenitor an individual assigned as such incorrectly. Normally it is proven that it is impossible to exclude a particular individual as sire or dam of a calf and the animal is said that qualifies as progenitor and can be certified as such in the pedigree papers. In other occasions, an assumed progenitor can be excluded as such with a probability of up to 0.9999 (Caldwell, 1996).

Recent studies in cattle have used microsatellite analysis to determine genetic diversity among breeds (Hansen et al., 2002; Maudet et al., 2002), to assign paternity in legal disputes (Liron et al., 2004) and to determine the degree of mistakes existing in genealogy records and its impact in animal breeding programs; the error rate found was of about 10% (Banos et al., 2001; Israel and Seller, 2000; Visscher et al., 2002). This confirms the importance of paternity tests for accurate pedigree data. Also, new statistical methods have been developed to calculate exclusion probabilities based in microsatellite analysis in cases with more than one candidate sire, as happens with multiple sire programs (Neff et al., 2000, 2001; Sherman et al., 2004).

In the present study, it was determined the number of alleles and their frequencies of 8 microsatellites (BM 1824, BM 2113, SPS 115, ETH 3, ETH 10, ETH 225, TGLA 122 and TGLA 227). Also, their heterozygosis, Polymorphism Information Content (PIC) and Exclusion Probabilities (EP), both individually and as a group were calculated. Six different cattle breeds were included (Holstein, Simmental, Brown Swiss, Beefmaster, Brahman and Brangus). The microsatellites analyzed belong to the panel chosen by the International Society of Animal Genetics (ISAG) (Caldwell, 1996).

MATERIALS AND METHODS

Blood samples were obtained and DNA was extracted by the salting-out method from 287 bovines: 39 Holstein, 38 Simmental, 54 Brown Swiss, 59 Beefmaster, 47 Brahman and 50 Brangus, located in ranches at northeast Mexico (states of Nuevo Leon, Coahuila and Tamaulipas). The fluorescent marker, concentration of reagents, PCR primer sequence and grouping of reactions for multiplex PCR are shown in Table 1. Eight microsatellites were analyzed, except in the Simmental breed, in which only 5 were included due to lack of good PCR amplification. These microsatellites have been analyzed in other cattle population and were selected for their high heterozygosis, chromosomal location, wide range of sizes and easiness to be amplified in multiplex PCR reactions (Caldwell, 1996). The forward primer of each microsatellite was labeled with fluorescence according to Table 1.

Multiplex PCR reactions were performed in a reaction mix of 25 μL containing 40 ng genomic DNA, 1 x PCR buffer (including MgCl₂), 200 μM each dNTP and 1.25 units of Taq polymerase. Primer concentrations are shown in Table 1. Guidelines for PCR amplification improvement of Henegariu et al. (1997).

Termocycler (M. J. Research 100) program was as follows: initial denaturalization at 94°C, 4 min; 30 cycles of denaturalization at 94°C, 4 min, hybridization at 56°C, 1 min and extension at 72°C, 3 min; a final extension at 72°C, 7 min, with the exception of multiplex 2, in which hybridization temperatures were 60°C for ETH 10 and ETH 225 and of 66°C for ETH 3. PCR amplification was confirmed by 2% agarose gel electrophoresis at 75 V by 1 h, including a molecular size marker and negative control. The length (in base pairs or b.p.) of PCR products were determined with a DNA sequencer (ABI Prism 373A, Applied Biosystems) using the Genotyper 1.1 program, which automatically determines the fragment sizes according with the internal size marker (TAMRA).

The allele number of each microsatellite was determined by direct count. Genic frequency for each microsatellite was calculated by dividing the number of times each allele was present among the number of times all alleles were present.

Heterozygosis was calculated by the following formulae:

were, Pi is the frequency of allele I (Heyen et al., 1997; Usha et al., 1995).

Table 1:	PCR multiplex conditions

PIC value was calculated with the next formulae:

were, Pi is the frequency of i-esim allele (Usha et al., 1995).

Exclusion probability, which indicates probability that an alleged parent be excluded as potential progenitor, was calculated with following formulae, which assumes it is an individual taken at random:

RESULTS AND DISCUSSION

Three multiplexes and one single PCR reactions were evaluated for parentage verification purposes. Table 2 shows the observed number for each allele. From this data were calculated allele number and frequency, size, heterozygosis, PIC and exclusion probabilities. Allele number had a considerable variation, from 4 for ETH 3 in Brahman to 22 for BM 2113 in Beefmaster. Markers with a low allele number were ETH 3 in Brahman (4), ETH 10 in Simmental and Brahman (6 and 5, respectively), ETH 225 in Simmental (5) and TGLA 227 in Brahman (6). The markers with the highest allele number were BM 2113 in Beefmaster and Brangus (22 and 18), ETH 3 in Holstein (19), as well as ETH 10 (18) and ETH 225 (21) in Brangus.

With regard to heterozygosis, markers with the lowest values were SPS 115 in Holstein, ETH 3 in Brahman and ETH 225 in Beefmaster (0.4132, 0.5179 and 0.5887, respectively), whereas the highest values were BM 2113 in Simmental, Beffmaster and Brangus, ETH 225 in Brangus and TGLA 227 in Simmental (0.8739, 0.8811, 0,8855, 0.8799 and 0.9301, respectively) (Table 3).

PIC values were very high in all markers and breed; of the 46 PIC values obtained, 12 had the maximum value of 1. Six of the obtained exclusion probabilities had values <0.5: BM 1824 in Brahman (0.38), SPS 115 (0.1213) and ETH 3 (0.4524) in Holstein, ETH 3 in Brahman (0.22), ETH 19 in Brangus (0.45), as well as ETH 225 in Simmental (0.4155). These values, when considered independently, could indicate that they may not be useful for paternity testing; however, considering all markers for each breed in order to obtain the combined exclusion probability, we found that only in Simmental and Brahman was low (0.9924 and 0.9930). In the Simmental breed this is due to the fact that only five markers were analyzed, whereas in Brahman is because to the low exclusion probability of some markers (BM 1824 and ETH 3). The other markers in this breed had not so high exclusion probability values. Therefore, it could be necessary to substitute BM 1824 and ETH 3 for other markers in the Brahman breed. Total PE for each breed was: Holstein: 0.9988; Simmental: 0.9924; Brown Swiss: 0.9998; Beefmaster: 0.9999; Brahman: 0.9930; Brangus: 0.9990 (Table 3).

The present study shows, evidence that parentage testing using DNA markers in cattle from northeast Mexico gives an acceptable degree of reliability, such as has been demonstrated in other cattle populations. The DNA markers chosen by the ISAG are therefore convenient for use with the local cattle population. Results obtained demonstrate that the described methodology constitutes a useful, fast and trustworthy way to perform parentage tests in Mexico. The 8 DNA microsatellites used allow the exclusion in the analyzed breeds, on average, of 99.8% of individuals that are not progenitors of a particular animal. This can be applied when only the genotype of the offspring and of the alleged parent is known (Usha et al., 1995).

For the Brahman breed, which had the lowest exclusion probability with 8 markers, it may be convenient to analyze a set of microsatellites different from the ones used in this study. This will also, be necessary for the Simmental breed, in which only 5 markers were analyzed.

Some not previously reported alleles were found, since the size rank of markers ETH 3, ETH 10, ETH 225, TGLA 122 and TGLA 227 in this study is different to the ones reported. For example, TGLA 122 has been reported with a size rank of 135-181 bp, whereas we found its size rank varied from 128-175 bp. Also, the smaller size informed for TGLA 227 was 78 bp, but in this study we report an allele of 66 bp (Heyen et al., 1997).

DNA microsatellite polymorphisms are formed by mutations, which originate new alleles, differing from each other in the number of reported sequences. Therefore, care must be taken when a progenitor is to be disqualified as a parent based only in one marker, since the difference in the number of repetitions could be due to a mutation and not to the fact that it is not the real progenitor (Glowatzki-Mullis et al., 1995; Heyen et al., 1997).

Data in allele number, heterozygosis, PIC and EP values found in this research are in general similar to those reported by Glowatzki-Mullis et al. (1995), Heyen et al. (1997) and Usha et al. (1995).

Previous studies have shown that pedigree errors, that can be as high as 10% (Visscher et al., 2002), can lead to losses in genetic gain expected from breeding programs (Israel and Seller, 2000).

Table 2:	Characterization of the 8 microsatellites in the 6 bovine breeds
N = Number of observations for the estimations of allelic frequencies. PIC = Polimorphism Information Content. P.E. = Probability of Exclusion for each microsatellite

Table 3:	Probabilities of exclusion of the 8 combined microsatellites in the 6 analyzed breeds

These losses can be translated into a diminution of the bull`s breeding values (Banos et al., 2001). Some studies suggest that genetic marker information can replace that of traditional pedigree records to support animal breeding schemes (Doods et al., 2005).

CONCLUSION

The results obtained in this study lead us to conclude that the use of parentage verification based in a DNA test, in order to prove the accuracy of genealogical records, is justified. In the other hand, this type of analysis will also be useful in cases of legal disputes concerning the ownership of a particular animal (Liron et al., 2004). Furthermore, DNA microsatellite analysis is being used in paternity verification in other species, such as human (Dodds et al., 2005), horse (Zupanik et al., 2001), dogs (Tozaki et al., 2001; Ichikawa et al., 2001). Although, some studies have been conducted to compare the usefulness of microsatellite markers with that of Single Nucleotide Polymorphisms (SNPs), most authors concluded that there is no evidence of higher exclusion probabilities when SNPs are used instead of microsatellites. This confirms the validity for this kind of tests.

ACKNOWLEDGEMENTS

This research was conducted with the support of the National Council of Science and Technology, Mexico (CONACYT), under project number 26502-B. We thank the collaboration of Dr. James Derr from the DNA Technologies Laboratory, College of Veterinary Medicine, Texas A and M University in the genotyping of the PCR products.