Testing the breeding strategy of Hungarian Bronze turkey strains for maintaining genetic diversity with microsatellites

The aim of the study was to provide information on the genetic variability of the Hungarian Bronze turkey gene reserve population and its difference from the Broad-breasted turkey, and offer guidance and proposals for its future conservation strategies. Altogether, 239 Hungarian Bronze turkeys from 10 strains and 13 Broad-breasted turkeys as a control population were genotyped for 15 microsatellites. All loci were polymorphic with the average number of alleles per locus 3.20±1.146 in the Hungarian Bronze turkey. The mean expected (Hexp) and observed heterozygosity (Hobs) were not different (0.392 and 0.376, respectively) in the overall population, and similar values were obtained for hens and bucks and among hen strains. Inbreeding coefficient (FIS) and Shannon index (I) indicated that there was low inbreeding within hens and bucks. Our results confirm that the genetic diversity in the Hungarian Bronze turkey population has been preserved by the rotational mating system. Differences between the Hungarian Bronze turkey and the Broad-breasted turkey populations were determined. Nei’s unbiased values clearly indicated that the two populations are highly genetically differentiated.


Introduction
The turkey (Meleagris gallopavo) is the second most important poultry species agriculturally after the chicken.Most literature about poultry concerns genetic background and structure of the chicken genome; however, as the importance of the turkey grows, so biological information about this species must increase.According to the American Poultry Association, turkey variants are considered a single breed (Kamara et al. 2007).Knowledge of the genetic diversity of turkey populations, strains and breeds is important for the diversity of turkey varieties.This essential genetic resource could help breeders improve their birds' health and vigour or respond to changing environmental conditions, production systems and consumer needs (Christman & Hawes 1999).
The Meleagris genus originates in America; however, the turkey (Meleagris gallopavo) has been bred in Hungary since the end of 16th century (Mihók 2000).One native Hungarian turkey breed is the Hungarian Bronze turkey.The Black turkey had previously been common, but is now only kept on small farms.Its population has decreased because of crossing with Bronze and other breeds at the beginning of the previous century.The Bronze turkey has adapted to local conditions to become a native breed.There is another native turkey type -the Hungarian Copper turkey -that is kept in the south of the country (Szalay et al. 2009).Performance and meat production are similar for all three types; therefore, it remains a question whether they are different breeds or only colour varieties.Some meat quality parameters of the Hungarian Bronze turkey, such as water-holding capacity, consistency and tenderness, are better and production level is usually less than that of the intensive breeds (Mihók et al. 1999).The adult weight of a male is 10-12 kg and 5-7 kg for females.The main advantage of this breed is its excellent feed-seeking ability and vitality.In the 1960s, the number of stocks decreased as intensive production systems became widespread (Mihók 2000).The Broad-breasted Bronze turkey is an improved type of Bronze turkey bred in the USA.Adult males weigh 12-13 kg and females 7-9 kg with a broad breast and good carcass performance (Sütő 2006).
Considering that few data are available about native turkey types and their genetic backgrounds, the purpose of this work was to determine the genetic diversity of a Hungarian Bronze turkey gene reserve strains, evaluate the effectiveness of the present mating system to preserve genetic diversity and finally compare this native population to the Broad-breasted turkey breed using modern molecular genetics tools.

Animals
The examined Hungarian Bronze turkey has been owned by the University of Debrecen, Hungary for more than 15 years in order to maintain a gene reserve population.In 1999, 12 unrelated strains were generated containing 14-16 hens and 3-4 tons per strain.To keep the genetic diversity the following breeding program has been used since 2000.Hens in a given strain provide the next generation in the same strain, and the bucks will be the next generation of bucks of the neighbouring strain.In the reproduction cycle, animals mate at random within each strain.Hens lay 19 eggs on average and are kept for two years.Unfortunately, the number of animals in two strains decreased so much that these strains were divided out among the other strains over the past six years.
In this study, 239 Hungarian Bronze turkeys from 10 strains and 13 Broad-breasted turkeys as a control population were examined (Table 5).Blood samples were collected from both strains in 2007.

DNA extraction and microsatellite analyses
Genomic DNA from total blood was extracted using the Zsolnai & Orbán (1999) procedure.Microsatellites were chosen based on their chromosomal location (Table 1).
Polymerase chain reactions (PCRs) were prepared in a total volume of 10 µl containing 30 mM MgCl 2 , 2 mM of each dNTP, 2.5 µM of each primer, 5 U Taq polymerase and 100-200 ng genomic DNA.Amplifications were performed using a Perkin Elmer Gene Amp PCR System 9700 Thermocycler (Applied Biosystems, Foster City, CA, USA) and DNA Engine Peltier Thermal Cycle (Bio-Rad, USA).Running conditions consisted of 10 min at 95 °C followed by 35 cycles of 15 s at 95 °C, 30 s at 51-62 °C (depending on the microsatellites, Table 1), 30 s at 73 °C and a final extension step of 20 min at 73 °C.PCR products were run in two multiplexes and analysed using an ABI Prism 310 Genetic Analyser (Applied Biosystems, Foster City, CA, USA).The size of each fragment was determined relative to the LIZ 500 size standards (Applied Biosystems).GeneMapper 3.7 software (ABI; Perkin Elmer, Foster City, CA, USA) was used for genotype scoring.

Statistical analysis
Popgene version 1.31 software package was used for statistical analysis (Yeh & Yong 1999).Population structure was analysed using Wright's fixation indexes (Weir & Cockerham 1984).Expected (H exp ) and observed heterozygosity (H obs ) were computed using Levene (1949).The effective number of alleles (N e ) was calculated by the Kimura & Crow (1964) formula.Shannon index (I) was calculated to determine genetic diversity (Lewontin 1972).Based on microsatellite data, Nei's (1987) unbiased genetic distance was used for dendrogram construction according to the neighbour-joining algorithm.

Genetic structure of the Hungarian Bronze turkey strains
Altogether, 48 alleles were detected on the 15 microsatellite loci.All loci were polymorphic with the number of alleles per locus (N a ) varying from two (ADL0149, ADL0266, ADL0293, MCW0080) to six (MNT162).The average number of alleles per locus was 3.20±1.146.The effective number of alleles per locus ranged from 1.062 (ADL0293) to 3.615 (MNT162) (Table 2).Rare alleles (with a frequency of less than 0.05) were found for 10 loci (Table 3).The rest of the allele frequencies are available from the authors on request.
The mean of Hobs was 0.376±0.224among loci, ranging between 0.000 (ADL0293) and 0.722 (MNT162).Higher H exp than H obs were obtained in case of nine microsatellites.A significant deviation from the Hardy-Weinberg equilibrium was observed for only three microsatellites at P<0.01 (Table 2).The inbreeding coefficient (F IS ) among loci varied from 1.000 (ADL0292) to -0.186 (MNT199) (Table 2).

Genetic differentiation between Hungarian Bronze turkey hens and bucks
The genetic structure of hens (n=194) and bucks (n=45) were analysed separately.From the examined 15 loci, 47 alleles were detected in hens and 43 in bucks.The same number of alleles was found in 11 loci in hens and bucks.Two more alleles were detected on locus MNT162 and MNT214: one (MNT327) when comparing hens to bucks and one (MNT387) when comparing bucks to hens.Each of those alleles, which were found only in hens or in bucks, had less than a 0.05 frequency in the overall population (Table 3).Mean of N a was higher in hens than in bucks, however, the mean of N e was slightly higher in bucks than in hens (Table 4).Mean H exp was higher than mean H obs in both hens and bucks.F IS and Shannon index indicated that there was a low inbreeding within hens and bucks (Table 4).Multilocus fixation index (F ST ) indicated that around 0.4 % of the total genetic variation could be explained by a sex difference; the remaining corresponded to differences among individuals.Nei's unbiased genetic distance value was low between hens and bucks (0.003).

Genetic differentiation among the Hungarian Bronze turkey hen strains
The genetic structure of the 10 strains was investigated separately and differentiation was determined among them.Three loci (ADL0292, ADL0293 and MNT332) were found to be not polymorphic in at least one strain.All hens were homozygous for one of these loci in strains 1, 3 and 5-8, two loci in strains 2, 9 and 10 and three loci in strain 4. The number of alleles per locus varied between one and five in the strains.The mean number of alleles per locus per strain ranged from 2.333±0.724(strain 9) to 2.600±0.828(strain 3).The examined hen strains had similar levels of allelic variations.N e ranged from 1.598±0.655(strain 1) to 1.934±0.753(strain 8) (Table 5).Different alleles were found with a low frequency (F<0.05) in the different strains.Rare alleles were not detected in strain 8 (Table 6).H exp and H obs per strains ranged from 0.306±0.234(strain 1) to 0.434±0.238(strain 8) and from 0.311±0.210(strain 5) to 0.458±0.252(strain 8), respectively.Genetic diversity was highest in strain 8 (Table 5).F IS values were almost negative in all strains and ranged from 0.035 (strain 5) to -0.116 (strain 9).Nei's unbiased genetic distance values were calculated (Table 7).The greatest distance values were found between the following pairs: strain 5/strain 3, strain 6/strain 1, strain 9/strain 1, strain 9/strain 2, strain 9/strain 4, strain 9/strain 5, strain 9/strain 6, strain 10/strain 1 and strain 7/all other nine strains.Figure 1 shows the neighbour-joining dendrogram obtained using Nei's unbiased genetic distance values.The genetic relationship among strains is shown clearly.

Genetic distance between the Hungarian Bronze turkey and Broad-breasted turkey strains
The mean number of observed alleles was lower (2.769±1.013)and the mean number of effective alleles was higher (1.952±0.738) in the Broad-breasted turkey than in the Hungarian Bronze turkey.Mean H exp was higher than mean Hobs (0.441±0.219 and 0.419±0.232,respectively) in the Broad-breasted turkey.
The result of F ST showed that 15.9 % of the total genetic variation was explained by differences between the two strains, and the remaining 84.1 % was accounted for differences among individuals.Nei's unbiased genetic distance values were calculated to determine the distance between the Hungarian Bronze and Broad-breasted turkey strains (Table 8).5/strain 3, strain 6/strain 1, strain 9/strain 1, strain 9/strain 2, strain 9/s strain 9/strain 5, strain 9/strain 6, strain 10/strain 1 and strain 7/all othe strains.Figure 1 shows the neighbour-joining dendrogram obtained us unbiased genetic distance values.The genetic relationship among stra shown clearly.

Genetic distance between the Hungarian Bronze turkey and Broad-bre turkey strains
The mean number of observed alleles was lower (2.769±1.013)and th number of effective alleles was higher (1.952±0.738) in the Broad-brea turkey than in the Hungarian Bronze turkey.Mean H exp was higher tha Hobs (0.441±0.219 and 0.419±0.232,respectively) in the Broad-breas turkey.
The result of F ST showed that 15.9% of the total genetic variation was explained by differences between the two strains, and the remaining 8 was accounted for differences among individuals.Nei's unbiased gene distance values were calculated to determine the distance between th Hungarian Bronze and Broad-breasted turkey strains (Table 8).
Table 8: Genetic identity (above diagonal) and genetic distances (belo diagonal) between the Hungarian Bronze turkey strains and Broad-bre turkey

Discussion
Microsatellite markers are now used efficiently for determining the genetic diversity of poultry species (Leberg et al. 1994, Rhodes et al. 1995, Ye et al. 1998, Kaiser et al. 2000, Romanov & Weigend 2001, Kong et al. 2006, Tu et al. 2006, Shahbazi et al. 2007, Yan et al. 2008).In our study, 15 microsatellite markers were used to determine the genetic structure of a Hungarian Bronze turkey gene reserve population and its differences from the Broad-breasted turkey.High numbers of alleles (n=48) were found on the examined 15 loci in the Hungarian Bronze turkey.The use of a mixture of highly variable and less variable markers should reduce the danger of overestimating genetic variability, which might occur if only highly variable loci are used (Wimmers et al. 2000).In our study, all loci were polymorphic.The number of observed alleles per locus ranged from two to six, with a mean number of alleles of 3.2±1.1,which demonstrates the utility of the chosen microsatellites as informative molecular markers in the examined breed.However, the effective number of allele was much lower than the observed alleles in all loci.Larger differences were found in cases of those microsatellites where rare alleles occurred because they are less likely to take part in mating.Altogether, 14 rare alleles were found on 10 loci.Szöke et al. (2004) previously studied this Hungarian Bronze turkey five generations earlier using seven microsatellites, from which five (ADL0292, ADL0293, ADL0149, ADL0266 and MCW0080) were the same as we have studied.They found one more allele on every locus except MCW0080, which means that one allele from each of these loci, was lost over the past five years.The frequency of these alleles was lower than 0.01 in their study.They assumed that gender greatly influences the probability of losing a rare allele using the present mating system, which was borne out by their simulation results.They reported that if only one hen had a rare allele from 144 turkeys (allele frequency less than 0.01), it was lost within five generations in 85.5 % of the cases; when the male had the rare allele the value was only 21.5 %.We suppose that those lost alleles were carried by hens in the examined population.In the present study, 6 of the 14 rare alleles were carried only by hens, eight were found in both genders and one was observed only in bucks.The occurrence of the rare alleles among strains was different.Each allele was mainly observed only in one strain (Table 2).We should pay particular attention to the animals carrying the rare alleles because preservation requires more offspring from these animals during selection.However, this is difficult to put it into practice because turkeys have a short generational period and the cost of genotyping is expensive compared to their value.
Concerning heterozygosity, the mean H exp and H obs were not different (0.392 and 0.376, respectively) in the overall population, with similar values obtained for both hens and bucks.Szöke et al. (2004) reported a mean H exp of 0.165 and mean H obs of 0.164 for seven loci in this strain.Differences in heterozygosity might be the result of different sample sizes and studied microsatellites.The F IS and Shannon index indicated a low inbreeding within hens and bucks (Table 4).Multilocus F ST indicated that around 0.4 % of the total genetic variation was explained by a sex difference; the remaining amount corresponded to differences among individuals.The H exp and H obs varied marginally among hen strains, and H obs was higher than H exp in all strains except strain 5.The highest heterozygosity was obtained in strain 8, which could be the reason for the absence of rare alleles.Kalinowski (2004) reported that rare alleles affect allelic richness.F IS and I also indicate similar high genetic diversities within all examined hen strains.All these results confirm that the genetic diversity in the Hungarian Bronze turkey has been preserved by the rotational mating system.Differences between the Hungarian Bronze turkey and Broad-breasted turkey were determined.Nei's unbiased values clearly indicated that the examined two populations were highly genetically differentiated because of their different breeding strategies.
We conclude that the present mating system is suitable for preserving the genetic diversity of the Hungarian Bronze turkey gene reserve population.However, genetic diversity could rapidly decrease by reducing the population size.Bottlenecks cause a rapid loss of rare alleles and also result in the loss of genetic variability because of the effects of genetic drift (Allendorf & Luikart 2007).The Hungarian Bronze turkey population has a similar genetic diversity to other turkeys, suggesting that the studied population has not experienced serious genetic loss from the effects of bottlenecks.Inbreeding or breeding for a long time without fresh blood, especially in small populations, could reduce the level of genetic variability and produce a low heterozygosity value.Besides using the present breeding strategy, bringing fresh blood from other Bronze turkey populations is recommended.The results of this study might be useful as a guide for preserving the Hungarian Bronze turkey.

Figure 1
Figure 1 Neighbour-joining dendrogram among the Hungarian Bronze turkey hen strains

Table 1
Main characteristics of used primers

Table 2
Number of alleles, heterozygosity per locus and inbreeding coefficient in the Hungarian Bronze turkey strains *significant deviation from Hardy-Weinberg equilibrium (P<0.001),N a : observed number of alleles, N e : effective number of alleles, H ob : observed heterozygosity, H exp : expected heterozygosity, F IS : inbreeding coefficient

Table 3
Frequency of rare alleles in the Hungarian Bronze turkey strains

Table 4
Number of alleles, heterozygosity, inbreeding coefficient and Shannon index per hen and buck : observed number of alleles, N e : effective number of alleles, H ob : observed heterozygosity, H exp : expected heterozygosity, F IS : inbreeding coefficient, I: Shannon index This might be the result of the lower number of bucks than hens.Significant deviation from the Hardy-Weinberg equilibrium was observed for three loci in bucks (MNT327 [P<0.05],ADL0293 and MNT332 [P<0.01]), and for six loci in hens (ADL0149, MNT197 and MNT199 [P<0.05] and ADL0293, MNT106 and MNT387 [P<0.001]). a

Table 6
Frequency of rare alleles in the 10 Hungarian Bronze turkey hen strains

Table 8
Genetic identity (above diagonal) and genetic distances (below diagonal) between the Hungarian Bronze turkey strains and Broad-breasted turkey