Evaluation of diversity between different Spanish chicken breeds, a tester line, and a White Leghorn population based on microsatellite markers

Evaluation of diversity between different Spanish chicken breeds, a tester line, and a White Leghorn population based on microsatellite markers S. G. Dávila, 1 M. G. Gil, P. Resino-Talaván, and J. L. Campo Departamento de Mejora Genética Animal, Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Apartado 8111, 28080 Madrid, Spain ABSTRACT The present study was conducted to evaluate the genetic variability and the genetic divergence of 13 Spanish chicken breeds, a tester line, and a White Leghorn population, using 24 microsatellite markers. A total of 150 alleles were detected across all population. The number of alleles by locus ranged from 2 to 13, with the mean value being 6.25. The mean polymorphic information content was 0.591, ranging from 0.847 to 0.172. The combined parentage exclusion probability of excluding 1 parent or 2 parents was 99 and 100%, respectively. The observed heterozygosity was lower than the expected heterozygosity for all loci, the mean values being 0.461 and 0.637. The observed and expected heterozygosity ranged from 0.003 to 0.735 and 0.181 to 0.863, respectively. Mean deficit of heterozygotes within populations (F IS ) was 0.056 and mean fixation index of each population (F ST ) was 0.244. The mean global deficit of heterozygotes across populations (F IT ) was 0.286. A total of 15 private alleles in 10 microsatellites were observed, and in some populations, fixed alleles were found for 7 microsatellites. A total of 300 birds (83%) were properly assigned to the source population. The average observed heterozygosity for each population was 0.461, ranging from 0.328 (Quail Castellana) to 0.538 (Red Villafranquina), and the average expected heterozygosity was 0.488, ranging from 0.320 (Quail Castellana) to 0.550 (White-Faced Spanish). All of the Spanish breeds except the Quail Castellana were more polymorphic than the White Leghorn population. The mean value of the deviation of heterozygote number was 0.052. Nei s genetic distance showed a range from 0.109 (between White-Faced Spanish and Black Menorca) to 0.437 (between Buff Prat and White Leghorn). A phylogenetic tree constructed by the neighbor-joining method, based on Nei s genetic distance, showed a clear separation between the White Leghorn and the remaining breeds. The results indicate that the panel of microsatellite markers was useful in studying the genetic diversity of chicken breeds. Key words: Spanish breed, chicken, microsatellite marker, genetic diversity, phylogenetic tree INTRODUCTION Because of centuries of domestication and breeding, a wide range of chicken breeds exists today. However, an increasing number of local chicken breeds are under threat of extinction, and valuable genotypes and traits are at risk of being lost (Blackburn, 2006). The local breeds represent both a heritage and a reservoir of variability that deserves to be explored and properly managed. With industrial aviculture development in the 1950s and the 1960s, and the creation of hybrids to produce eggs and meat, the number of traditional hens fell dramatically to the point of extinction. Fortunately, due to conservation programs and a few traditional hen breeders, a large number of these breeds were saved 2009 Poultry Science Association Inc. Received July 14, 2009. Accepted September 11, 2009. 1 Corresponding author: sgdavila@inia.es 2009 Poultry Science 88 :2518 2525 doi: 10.3382/ps.2009-00347 from extinction. The evaluation of local breeds as genetic resources is of interest in efforts to maintain genetic variation and includes records of phenotypes and breeding history as well as determination of genetic variability (Hammond, 1994). Quantitative assessment of genetic diversity within and between populations is an important tool for decision making in genetic conservation plans. Spain has a rich genetic diversity of native chickens. A total of 13 Spanish chicken breeds have been reported, based on morphological characteristics (Campo, 1998). Among these native breeds, 6 are white egg layers, 2 are tinted egg layers, 1 is a dark brown egg layer, 1 is a dual-purpose breed (meat and brown egg layer), 2 are feather producers, and 1 is a synthetic egg layer. These native breeds are used for quality chicken production in free-range systems, and they are maintained in a conservation program of genetic resources that was started in 1975 (Campo and Orozco, 1982). The Span- 2518

ish population includes one of the oldest Mediterranean breeds (its ancestor having been introduced into Spain by the Arabians in the 8th century), the only breed in the world with white face, and the original breed carrying the blue gene. The Spanish breeds carry 6 different alleles in the main polyallelic locus determining the plumage color in chickens (extended black, birchen, dominant wheaten, wild type, brown, and buttercup), and some of them can be used for chick sexing of egg lines (Campo, 1991) or have low cholesterol content (Campo, 1995). Recent advances in molecular technology have opened up new horizons for estimating genetic relatedness between and within animal populations, and molecular markers may serve as an important initial guide to develop conservation strategies. Studies in molecular technology have found that local breeds have more genetic diversity when compared with recent commercial breeds (Siegel et al., 1992; Plotsky et al., 1995; Ponsuksili et al., 1996; Zhou and Lamont, 1999; Lee et al., 2000; Okumura et al., 2006). The microsallite markers are extensively used for estimating genetic structure, diversity, and relationships because of many advantages: they are numerous and ubiquitous throughout the genome, show a higher degree of polymorphisms, and have a codominant inheritance (Tautz, 1989). Information in literature has revealed that microsatellite markers are the most accurate and efficient method for estimating genetic diversity and relationships among populations (Takezaki and Nei, 1996). Several studies have been conducted to characterize the diversity in traditional chicken breeds based on microsatellite markers. Many of them have been made in Asian breeds (Takahashi et al., 1998; Zhang et al., 2002; Osman et al., 2006; Ya-Bo et al., 2006; Tadano et al., 2007). Additionally, Wimmers et al. (2000) evaluated genetic distinctness of African, Asian, and South American chickens, and Romanov and Weigend (2001) analyzed genetic relationships between various chicken populations in Europe and jungle fowl. Genetic relationships between a wide range of chicken types from Europe, Asia, and Africa have been analyzed in the AVIANDIV project (Hillel et al., 2003), and Berthouly et al. (2008) studied genetic diversity of French and Asian breeds, comparing them with those of the AVI- ANDIV project. Finally, Twito et al. (2007) compared the biodiversity of 20 chicken breeds with SNP and microsatellite markers, noting that analysis based on microsatellites resulted in significantly higher clustering success due to their multiallelic nature. The objective of this study was to evaluate the genetic variability and the genetic divergence of 13 Spanish chicken breeds, a tester line, and a White Leghorn population using microsatellite markers. Up to now, only 1 study with Spanish breeds has been conducted to evaluate genetic diversity based on DNA polymorphism (the AVIANDIV project included 2 Spanish chicken breeds). SPANISH CHICKEN GENETIC DIVERSITY 2519 Chickens MATERIALS AND METHODS A total of 360 hens, randomly selected, from the 15 populations were used (24 hens by population). All of them came from the experimental station of El Encín (Madrid, Spain), where the conservation program of Spanish genetic resources is located. The number of hens and cocks in each population ranged from 128 to 395 and from 25 to 80, respectively. The effective population size ranged from 66 to 310 (Campo et al., 2002), higher than the minimum size (50) required to avoid inbreeding depression in the short-term. Black Castellana, Black Menorca, White-Faced Spanish, Black- Barred Andaluza, Blue Andaluza, and Black-Breasted Red Andaluza are white egg layers, whereas Buff Prat and White Prat are tinted egg layers, and Red-Barred Vasca and Red Villafranquina lay brown and dark brown eggs, respectively. Birchen Leonesa and Blue Leonesa are used to produce hackles and saddles for fishermen. In the study are included a synthetic Spanish breed (Quail Castellana or Melanotic Prat) originated from an F 2 cross between Black Castellana and Buff Prat (Campo and Orozco, 1986), a tester line carrying the recessive wheaten allele (Smyth, 1976), and a White Leghorn population (Campo and Jurado, 1982). DNA Isolation and Amplification Blood samples were collected by brachial venipuncture aseptically into tubes using EDTA as anticoagulant. Blood samples were stored at 20 C. Deoxyribonucleic acid was extracted from 40 μl of blood using resuspension buffer (0.1 M Tris-HCl, 0.01 M NaCl, 0.1 M EDTA, ph 8), lysis buffer (0.1 M Tris-HCl, 0.1 M EDTA, 0.01 NaCl, 1% SDS, ph 8), 12.5 μl of proteinase K (20 mg/ml), 2 ml of 5 M NaCl, 6 ml of isopropanol, and resuspension in Tris-EDTA. The DNA was quantified spectrophotometrically and the concentration was adjusted to 15 ng/μl. Twenty-four microsatellite markers were chosen based on their genomic location and their degree of polymorphism. The PCR products were obtained in a total volume of 10 µl of reaction, with 1 µl of DNA (15 ng/µl), 0.5 µm of forward and reverse primers, 1.5 mm MgCl 2, 200 µm deoxynucleoside triphosphate, and 0.02 U/µL of Taq polymerase. The amplification involved first denaturation at 94 C for 3 min, 30 cycles of denaturation at 94 C for 1 min, annealing at primerspecific temperature (from 45 to 60 C) for 1 min, and extension at 72 C for 1 min, followed by final extension at 72 C for 10 min. Fluorescent end-labeled (carboxyfluorescein) PCR primers were used and size characterization of PCR product was performed by an ABI 370 DNA Genetic Analyzer (Applied Biosystems, Foster City, CA).

2520 Statistical Analysis Number of alleles, effective allele number, polymorphism information content, observed and expected heterozygosity, Wright s F-statistics, heterozygote deficiency or excess, and test of the Hardy-Weinberg equilibrium using Bonferroni correction were estimated using POPGENE (Yeh et al., 1999), FSTAT (Goudet, 2001), and CERVUS (Kalinowski et al., 2007) computer packages. Neighbor-joining method (Saitou and Nei, 1987) with arithmetic mean based on Nei s genetic distance (Nei et al., 1983) was used to construct the phylogenetic trees. The robustness of the phylogenetic trees was evaluated by resampling bootstrap of the loci with a total of 1,000 replications. All calculations were made using the DISPAN package (Ota, 1993). The GENECLASS program (Cornuet et al., 1999) was used for assigning individuals to populations using Nei s genetic distance. The probability that each individual was assigned or not to a population was calculated using a direct estimation of frequencies, with 10,000 simulated individuals per population and applying a rejection threshold of 0.01. Parentage exclusion probabilities (p i ) of first parent and parents pair (Jamieson and Taylor, 1997) were calculated with CERVUS. RESULTS Polymorphism of Markers Dávila et al. A total of 150 alleles were detected across all population for the 24 microsatellites examined (Table 1), the Spanish breeds having 72 alleles which were missing in the White Leghorn population. The number of alleles by locus ranged from 2 to 13, with the mean value being 6.250 (SD = 2.770; residual sum of squares = 2.009), whereas the effective number ranged from 1.220 to 7.219, with the mean value being 3.339 (SD = 1.506). The mean polymorphic information content was 0.591 (SD = 0.178), ranging between 0.847 and 0.172. The parentage exclusion probabilities of excluding first parent and parents pair ranged from 0.016 to 0.567 and from 0.166 to 0.889, respectively (Table 1). Combined parentage exclusion probabilities [1 Π(1 p i )] across all loci were 99.999 and 100% (first parent and parents pair, respectively). The observed heterozygosity ranged from 0.003 to 0.735 (Table 2). The minimum observed heterozygosity was at the MCW0294 locus, and it was practically zero. The lowest and greatest expected heterozygosity was 0.181 and 0.863, respectively, with only 4 loci having the expected heterozygosity lower than 0.5. The observed heterozygosity was lower than the expected heterozygosity for all loci, the mean values being 0.461 (SD = 0.191) and 0.637 (SD = 0.175), respectively. The F-statistics (F IS, F ST, and F IT ) for each locus are also indicated in Table 2. Mean deficit of heterozygotes within populations (F IS ) was 0.056 and mean fixation index of each population (F ST ) was 0.244, ranging from 0.099 to 0.467. The global deficit of heterozygotes across populations (F IT ) ranged from 0.097 to 0.996, the mean value being 0.286. Six of the Spanish populations (Blue Andaluza, Red- Barred Vasca, White-Faced Spanish, Black Castellana, Birchen Leonesa, and Red Villafranquina) and the White Leghorn had a total of 15 private alleles at 10 Table 1. Microsatellite markers, allele size range, number of alleles (Na), effective number of alleles (Ne), polymorphism information content (PIC), and parentage exclusion probabilities (p i ) for all loci across breeds Microsatellite Chromosome Allele size (bp) Na Ne PIC p i 1 ADL0034 20 124 to 166 13 7.219 0.847 0.567 0.889 ABR0341 28 90 to 124 12 5.863 0.809 0.497 0.846 MCW0034 2 247 to 261 11 5.781 0.805 0.489 0.838 ADL0114 2 173 to 189 6 4.771 0.760 0.414 0.776 ADL0019 1 161 to 181 8 4.469 0.751 0.407 0.782 ABR0345 8 172 to 182 8 4.527 0.746 0.394 0.757 ADL0267 2 99 to 117 7 4.025 0.717 0.358 0.727 ADL0102 10 221 to 249 7 3.777 0.698 0.334 0.704 ADL0181 2 152 to 158 6 3.752 0.693 0.330 0.696 MCW0294 Z 219 to 223 5 3.249 0.645 0.274 0.632 ADL0124 1 223 to 229 5 2.795 0.623 0.262 0.632 LEI0194 1 86 to 102 7 3.067 0.617 0.260 0.601 MCW0081 5 165 to 180 6 3.075 0.616 0.252 0.586 ADL0023 5 111 to 133 5 2.930 0.589 0.221 0.527 MCW0295 4 111 to 127 7 2.717 0.578 0.217 0.553 ADL0278 8 109 to 125 4 2.816 0.570 0.210 0.505 MCW0037 3 93 to 113 4 2.453 0.527 0.177 0.477 MCW0330 17 158 to 176 7 2.296 0.514 0.168 0.491 MCW0069 26 254 to 286 6 2.300 0.510 0.172 0.490 ADL0101 1 102 to 114 3 2.307 0.470 0.160 0.401 ADL0118 14 108 to 140 5 1.817 0.392 0.103 0.355 MCW0150 3 156 to 170 2 1.510 0.281 0.057 0.224 ADL0032 18 220 to 248 3 1.401 0.259 0.041 0.236 MCW0222 3 300 to 316 3 1.220 0.172 0.016 0.166 p i 2 1 First parent. 2 Parents pair.

SPANISH CHICKEN GENETIC DIVERSITY 2521 Table 2. Observed and expected heterozygosity (Ho and He, respectively) and F-statistics (F IS, F IT, and F ST 1 ) for all loci across breeds Microsatellite Ho He F IS F IT F ST ADL0034 0.735 0.863 0.011 0.156 0.146 ABR0341 0.700 0.831 0.071 0.170 0.225 ADL0114 0.667 0.791 0.019 0.166 0.149 ADL0102 0.629 0.736 0.223 0.158 0.311 MCW0034 0.625 0.828 0.042 0.257 0.224 ADL0181 0.616 0.735 0.008 0.172 0.178 ADL0023 0.603 0.660 0.095 0.097 0.176 ADL0267 0.581 0.753 0.010 0.238 0.244 ADL0019 0.542 0.778 0.054 0.315 0.275 ABR0345 0.531 0.780 0.109 0.331 0.249 MCW0295 0.528 0.633 0.020 0.178 0.193 ADL0278 0.494 0.646 0.022 0.248 0.264 MCW0069 0.492 0.566 0.029 0.140 0.164 LEI0194 0.476 0.675 0.099 0.307 0.231 MCW0081 0.464 0.676 0.062 0.326 0.281 MCW0037 0.431 0.593 0.042 0.287 0.256 MCW0330 0.408 0.565 0.068 0.290 0.237 ADL0118 0.393 0.450 0.037 0.132 0.099 ADL0101 0.353 0.567 0.198 0.388 0.239 MCW0150 0.253 0.338 0.031 0.288 0.311 ADL0124 0.247 0.665 0.334 0.628 0.444 ADL0032 0.211 0.286 0.043 0.305 0.326 MCW0222 0.081 0.181 0.407 0.576 0.296 MCW0294 0.003 0.692 0.992 0.996 0.467 1 F IS = mean deficit of heterozygotes within populations; F IT = mean global deficit of heterozygotes across populations; F ST = mean fixation index of each population. microsatellites. The Black Castellana, Birchen Leonesa, and Red Villafranquina breeds had 1 private allele at 3 different loci, whereas the White-Faced Spanish had 2 private alleles at 1 locus. The Red-Barred Vasca had 1 private allele at 2 loci, and the Blue Andaluza had 1 private allele at 1 locus and 2 private alleles at another locus. Finally, the White Leghorn had 1 private allele at 2 loci and 3 private alleles at another locus. A total Table 3. Number of birds assigned to the population (n 1 ), to the population and to another population (n 2 ), to another population (n 3 ), and number assigned to the population or to another population (n 4 ) Population 1 n 1 n 2 n 3 n 4 AA 22 1 (MN) 1 AF 20 1 (PA) 3 AP 22 1 (e y ) 1 B 22 1 1 CB 15 1 (MN, AA) 4 (MN) CC 19 1 (CN, MN) 3 1 (CN) CN 21 3 e y 18 2 (AP) 2 2 IN 19 5 MN 16 6 (CB) 2 LEG 20 4 P 21 2 (PW) 1 PA 20 4 PW 22 2 (CB) VF 23 1 1 AA = Blue Andaluza; AF = Black-Barred Andaluza; AP = Black- Breasted Red Andaluza; B = Red-Barred Vasca; CB = White-Faced Spanish; CC = Quail Castellana; CN = Black Castellana; e y = tester line; IN = Blue Leonesa; MN = Black Menorca; LEG = White Leghorn; P = Buff Prat; PA = Birchen Leonesa; PW = White Prat; VF = Red Villafranquina. of 9 private alleles had frequencies higher than 10%. Fixed alleles were observed in 7 loci. The assignment of the individual to the different populations is presented in Table 3. A total of 300 birds (83%) were properly assigned to the source population, some additional birds being assigned to the source population and to another population (20 birds) or to 2 different populations (2 birds). Thirty-five birds were neither assigned to the source population nor to another population, and only 3 birds were assigned to another population. Diversity of Population Observed and expected heterozygosities for each population are indicated in Table 4. Mean expected heterozygosity was higher than mean observed heterozygosity in most populations except in the Red Villafranquina, White Prat, Blue Andaluza, and Quail Castellana breeds. The average observed heterozygosity was 0.461 (SD = 0.041), ranging from 0.328 to 0.538, whereas the average expected heterozygosity was 0.488 (SD = 0.060), ranging from 0.320 to 0.550. All of the Spanish breeds except the Quail Castellana had greater observed heterozygosity than the White Leghorn population. The number of alleles per locus and breed ranged from 3 to 4, the average mean value being 3.405 ± 0.393. The 4 breeds indicated above showed deficit of heterozygotes, whereas the 11 remaining breeds showed excess of heterozygotes, the average mean value being 0.052. The number of loci that deviated significantly (P < 0.05) from the Hardy-Weinberg equilibrium ranged from

2522 Table 4. Observed and expected heterozygosity (Ho and He, respectively), observed number of alleles per locus (Na), number of loci that deviated from Hardy-Weinberg equilibrium (dhwe), and deviation of heterozygote number (F IS ) Population 1 Ho He Na dhwe F IS VF 0.538 0.528 3.250 2 0.020 CB 0.517 0.550 3.958 2 0.062 AF 0.495 0.518 3.500 2 0.046 CN 0.493 0.540 3.958 3 0.088 MN 0.478 0.535 3.708 3 0.109 AP 0.474 0.502 3.292 4 0.057 B 0.471 0.529 3.708 4 0.113 PW 0.464 0.454 3.042 2 0.023 e y 0.462 0.481 3.583 3 0.040 PA 0.460 0.535 3.791 6 0.142 AA 0.460 0.448 3.458 4 0.024 IN 0.434 0.469 3.208 2 0.076 P 0.425 0.427 2.875 1 0.001 LEG 0.416 0.486 3.125 1 0.146 CC 0.328 0.320 2.625 1 0.026 1 AA = Blue Andaluza; AF = Black-Barred Andaluza; AP = Black- Breasted Red Andaluza; B = Red-Barred Vasca; CB = White-Faced Spanish; CC = Quail Castellana; CN = Black Castellana; e y = tester line; IN = Blue Leonesa; MN = Black Menorca; LEG = White Leghorn; P = Buff Prat; PA = Birchen Leonesa; PW = White Prat; VF = Red Villafranquina. 1 to 6 (Table 4), the average value being 2.666. A total of 330 Hardy-Weinberg equilibrium tests were made (360 theoretical tests without taking in account the sexlinked locus and those corresponding to fixed alleles). A total of 40 (12%) significant tests were observed, 11 of them (28%) having an excess of heterozygotes and 29 of them (72%) having a deficit of heterozygotes. Relationships Between Populations Ne s genetic distance (Table 5) showed a range from 0.109 (between White-Faced Spanish and Black Menorca) to 0.437 (between Buff Prat and White Leghorn). The genetic distance between the Spanish breeds laying white eggs and the White Leghorn ranged from 0.246 (Black Menorca) and 0.352 (Black-Red Andaluza), whereas the genetic distance between all of the Spanish Dávila et al. breeds and the White Leghorn ranged from 0.246 to 0.437. The 2 Spanish breeds included in the AVIANDIV project (Black Castellana and Red Villafranquina) had a genetic distance between them of 0.252, being 0.349 and 0.359 with the White Leghorn. A phylogenetic tree showed a clear separation between the White Leghorn and the remaining breeds (Figure 1). The most evident clusters were those found with the Blue Andaluza and Blue Leonesa, the 3 breeds with black plumage (Black Castellana, Black Menorca, and White-Faced Spanish), and the 3 varieties of the Prat breed (Buff, White, and Quail). DISCUSSION Most of the microsatellite markers used in this study showed a high degree of polymorphism. Barker (1994) suggested that the average number of alleles per locus in studies of genetic distances must be greater than 4 to reduce the SE in the estimation of genetic distances. In the present study, 4 microsatellites markers had a lower value, whereas the mean number of alleles per locus was bigger than 4. Results in the current study were similar to those found by Takahashi et al. (1998), Wimmers et al. (2000), Ya-Bo et al. (2006), Tadano et al. (2007), and Berthouly et al. (2008). On the contrary, Romanov and Weigend (2001), Zhang et al. (2002), Hillel et al. (2003), and Osman et al. (2006) observed higher values of the average number of alleles. The expected heterozygosity was generally higher than 0.5 and was especially high in the markers ADL0034, ABR0341, and MCW0034, suggesting the usefulness of these markers for this type of study. The panel of microsatellites showed a high power of parental exclusion, higher than 99.999%, showing that this group of markers has a great capacity to discriminate paternities. Mean fixation index (F ST ) was 0.244, the global deficit of heterozygotes across populations being 0.286, suggesting a high degree of population differentiation. Typically, a fixation index of about 0.15 is considered to be an indication of significant differentiation among Table 5. Genetic distances between Spanish chicken breeds, a tester line (e y ), and White Leghorn population Population 1 AA AF AP B CN CB CC e y IN LEG MN P PA PW AF 0.286 AP 0.253 0.215 B 0.294 0.233 0.225 CN 0.193 0.220 0.202 0.296 CB 0.227 0.256 0.223 0.302 0.151 CC 0.256 0.262 0.274 0.322 0.232 0.260 e y 0.254 0.244 0.125 0.222 0.234 0.209 0.260 IN 0.163 0.306 0.301 0.319 0.247 0.218 0.272 0.265 LEG 0.321 0.352 0.352 0.382 0.349 0.309 0.423 0.358 0.357 MN 0.189 0.286 0.219 0.290 0.158 0.109 0.229 0.186 0.165 0.246 P 0.324 0.291 0.286 0.294 0.251 0.242 0.249 0.296 0.386 0.437 0.274 PA 0.221 0.231 0.283 0.227 0.240 0.215 0.310 0.275 0.263 0.307 0.216 0.236 PW 0.272 0.257 0.262 0.282 0.234 0.172 0.218 0.233 0.278 0.395 0.233 0.148 0.226 VF 0.320 0.263 0.252 0.235 0.252 0.282 0.279 0.199 0.347 0.359 0.240 0.282 0.235 0.267 1 AA = Blue Andaluza; AF = Black-Barred Andaluza; AP = Black-Breasted Red Andaluza; B = Red-Barred Vasca; CB = White-Faced Spanish; CC = Quail Castellana; CN = Black Castellana; e y = tester line; IN = Blue Leonesa; MN = Black Menorca; LEG = White Leghorn; P = Buff Prat; PA = Birchen Leonesa; PW = White Prat; VF = Red Villafranquina.

SPANISH CHICKEN GENETIC DIVERSITY 2523 populations (Frankham et al., 2002); this value was found for all loci except for the ADL0118 locus. The mean fixation index was similar to those indicated previously by Tadano et al. (2007) in 12 Japanese breeds and 2 commercial lines, and Berthouly et al. (2008) in 14 French breeds and 6 different Japanese and Chinese breeds (0.303 and 0.240, respectively). As an indirect way to measure quantitative genetic diversity, a fixation index of about 0.25 means that 40% of total genetic variance could be explained by the among-breed genetic variance [2F ST /(1 + F ST )], in agreement with the range of values (30 to 50%) indicated in the literature for this parameter (Frankham et al., 2002). The markers used in the current study may be of great interest for the genetic identification of animals because 15 private alleles were observed in the populations. A high percentage of birds were properly assigned to the source population, the birds assigned to the source population and to another population being explained most of the time by a common origin. For example, White-Faced Spanish, Black Menorca, and Blue Andaluza come from the Black Castellana breed, and Buff Prat and White Prat are 2 varieties of the same breed. All of the Spanish breeds except the synthetic Quail Castellana were more polymorphic than the White Leghorn population. The highest value of heterozygosity was observed in the Red Villafranquina breed (0.54); although this breed has been selected for a very dark brown shell color for years, it shows a great variability of shell color ranging from light to very dark brown. Hillel et al. (2003) indicated a value for this breed of 0.46. The lowest heterozygosity was observed in the Quail Castellana, which originated from an F 2 cross between the Black Castellana and Buff Prat breeds followed by 4 generations of selection to uniform the chick down color until a completely black with brown face type (Campo, 1991), the low values of heterozygosity reflecting this selection. Similarly, the White Leghorn population also showed a low value of heterozygosity (0.42) and originated from the crossing of 3 commercial stocks selected for egg production and egg weight (Babcock, Creighton, and Mount Hope; Campo and Jurado, 1982). Hillel et al. (2003) indicated a value for the white egg layer line of 0.34. The mean observed heterozygosity in the 13 Spanish breeds was 0.46, in agreement with the values reported by Hillel et al. (2003) for 23 standardized breeds selected for morphological traits (0.46) and Wimmers et al. (2000), Tadano et al. (2007), and Berthouly et al. (2008) for other native breeds (0.58, 0.40, and 0.49, respectively). All populations showed significant deviations from the Hardy-Weinberg equilibrium, suggesting that some Spanish chicken breeds have been selected for years for morphological traits such as plumage, shank and egg colors, and comb and earlobe sizes, although the presence of null alleles or genotyping error could also be the reason. Figure 1. Neighbor-joining dendrogram based on Nei s genetic distance. AA = Blue Andaluza; AF = Black-Barred Andaluza; AP = Black- Breasted Red Andaluza; B = Red-Barred Vasca; CB = White-Faced Spanish; CC = Quail Castellana; CN = Black Castellana; e y = tester line; IN = Blue Leonesa; MN = Black Menorca; LEG = White Leghorn; P = Buff Prat; PA = Birchen Leonesa; PW = White Prat; VF = Red Villafranquina.

2524 Dávila et al. In the current study, the 2 Spanish breeds (Black Castellana and Red Villafranquina) included in the AVIANDIV project had a Nei s genetic distance between them and the White Leghorn population lower than those indicated by Hillel et al. (2003), which were 0.426, 0.644, and 0.454, respectively. This fact suggests that the values of genetic distances depend on the group of populations studied (52 in the AVIANDIV project); the white egg layer line used in the AVIANDIV project had been selected for more generations, and the white egg layer lines that were used in both studies had a different origin. Although Hillel et al. (2003) also calculated the Cavalli-Sforza and Reynolds genetic distances, Takezaki and Nei (1996) suggested the use of Nei s genetic distance in the analysis with microsatellite markers, when the main objective of the study is focused on the correct distribution of the topology, rather than to studies of evolutionary times. A phylogenetic tree showed a clear separation between the White Leghorn and the Spanish breeds, suggesting that the latter do not come from the former, although most of them are included in the Mediterranean group of breeds. Within the Spanish breeds, one of the most evident clusters was found for the 2 breeds carrying the blue gene, which produces a dilution of the eumelanins coming from the extended black (Blue Andaluza) or the birchen (Blue Leonesa) alleles. Another evident cluster was observed for the 3 breeds with black plumage carrying the extended black allele (Black Castellana, Black Menorca, and White-Faced Spanish), with the Black Castellana separated from the other 2 breeds. This fact is in agreement with the origin of the Black Menorca and White-Faced Spanish from the Black Castellana, due to the selection for a large size of the earlobes (Black Menorca), which are so greatly enlarged that they fall below the wattles covering all of the face (White-Faced Spanish). The last evident cluster was found for the 2 varieties of the Prat breed (Buff and White), and the synthetic breed originated from crossing this breed and the Black Castellana. This fact suggests that the synthetic breed is more similar to the Buff Prat than to the Black Castellana, in agreement with its melanotic columbian genetic background: (e Wh /e Wh Co/Co Ml/Ml), the Buff Prat being columbian (Campo, 1991). In conclusion, the panel of microsatellite markers was of great usefulness in studying the genetic diversity of chicken breeds, showing a high degree of polymorphism, a great capacity to discriminate paternities, and a high degree of population differentiation. The Spanish breeds had a high number of private alleles and were more polymorphic than the White Leghorn population, suggesting their potential to be selected for use in alternative production systems. Although Spanish chickens have been traditionally selected for morphological traits, they have not been selected yet for productive traits. A clear separation between the White Leghorn and the Spanish breeds was found, indicating the importance of including these populations in conservation programs. REFERENCES Barker, J. S. F. 1994. A global protocol for determining genetic distance among domestic livestock breeds. Pages 501 508 in Proc. 5th World Congr. Genet. Appl. Livest. Prod., Guelph, Ontario, Canada. Univ. Guelph, Guelph, Ontario, Canada. Berthouly, C., B. Bed Hom, M. Tixier-Boichard, C. F. Chen, Y. P. Lee, D. Laloë, H. Legros, E. Verrier, and X. Rognon. 2008. Using molecular markers and multivariate methods to study the genetic diversity of local European and Asian chicken breeds. Anim. Genet. 39:121 129. Blackburn, H. D. 2006. The National Animal Germplasm Program: Challenges and opportunities for poultry genetic resources. Poult. Sci. 85:210 215. Campo, J. L. 1991. Use of the sex-linked barring (B) gene for chick sexing on an eumelanotic columbian background. Poult. Sci. 70:1469 1473. Campo, J. L. 1995. Comparative yolk cholesterol content in four Spanish breeds of hens, an F 2 cross, and a White Leghorn population. Poult. Sci. 74:1061 1066. Campo, J. L. 1998. Conservation and genetical study of Spanish chicken breeds. Pages 155 158 in Proc. 6th World Congr. Genet. Appl. Livest. Prod., Armidale, Australia. Univ. New England, Armidale, New South Wales, Australia. Campo, J. L., M. G. Gil, and S. G. Dávila. 2002. El programa de conservación de las razas españolas de gallinas. Pages 183 191 in Proc. V Congr. Soc. Esp. Recur. Genét. Anim., Madrid, Spain. Editorial Punto, Madrid, Spain. Campo, J. L., and J. J. Jurado. 1982. Evaluation of multiple trait selection in strains of layers. Pages 869 874 in Proc. 2nd World Congr. Genet. Appl. Livest. Prod., Madrid, Spain. Editorial Garsi, Madrid, Spain. Campo, J. L., and F. Orozco. 1982. Conservation and genetical study of Spanish chicken breeds. Pages 83 93 in Proc. 2nd World Congr. Genet. Appl. Livest. Prod., Madrid, Spain. Editorial Garsi, Madrid, Spain. Campo, J. L., and F. Orozco. 1986. Genetic basis of the Melanotic Prat phenotype. Br. Poult. Sci. 27:361 367. Cornuet, J. M., S. Piry, G. Luikart, A. Estoup, and M. Solignac. 1999. New methods employing multilocus genotypes to select or exclude populations as origins of individuals. Genetics 153:1989 2000. Frankham, R., J. D. Ballou, and D. A. Briscoe. 2002. Introduction to Conservation Genetics. Cambridge Univ. Press, Cambridge, UK. Goudet, J. 2001. FSTAT, a program to estimate and test gene diversities and fixation indices, version 2.9.3. http://www2.unil.ch/ popgen/softwares/fstat.htm Accessed May 26, 2009. Hammond, K. 1994. Conservation of domestic animal diversity: Global overview. Pages 423 439 in Proc. 5th World Congr. Genet. Appl. Livest. Prod., Guelph, Ontario, Canada. Univ. Guelph, Guelph, Ontario, Canada. Hillel, J., M. A. M. Groenen, M. Tixier-Boichard, A. B. Korol, L. David, V. M. Kirzhner, T. Burke, A. Barre-Dirie, R. P. Crooijmans, K. Elo, M. W. Feldman, P. J. Freidlin, A. Mäki-Tanila, M. Oortwijn, P. Thomson, A. Vignal, K. Wimmers, and S. Weigend. 2003. Biodiversity of 52 chicken populations assessed by microsatellite typing of DNA pools. Genet. Sel. Evol. 35:533 557. Jamieson, A., and S. C. S. Taylor. 1997. Comparison of three probability formulae for parentage exclusion. Anim. Genet. 28:397 400. Kalinowski, S. T., M. L. Taper, and T. C. Marshall. 2007. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol. Ecol. 16:1099 1106. Lee, E.-J., H. Mannen, M. Mizutani, and S. Tsuji. 2000. Genetic analysis of chicken lines by amplified fragment length polymorphism (AFLP). Anim. Sci. J. 71:231 238. Nei, M., F. Tajima, and Y. Tateno. 1983. Accuracy of estimated phylogenetic trees from molecular data. J. Mol. Evol. 19:153 170.

SPANISH CHICKEN GENETIC DIVERSITY 2525 Okumura, F., T. Shimogiri, K. Kawabe, S. Okamoto, M. Nishibori, Y. Yamamoto, and Y. Maeda. 2006. Gene constitution of South- East Asian native chickens, commercial chickens and jungle fowl using polymorphisms of four calpain genes. Anim. Sci. J. 77:188 195. Osman, S. A.-M., M. Sekino, T. Kuwayama, K. Kinoshita, M. Nishibori, Y. Yamamoto, and M. Tsudzuki. 2006. Genetic variability and relationships of native Japanese chickens based on microsatellite DNA polymorphisms Focusing on the natural monuments of Japan. Jpn. Poult. Sci. 43:12 22. Ota, T. 1993. DISPAN: Genetic Distance and Phylogenetic Analysis. Penn. State Univ., University Park, PA. Plotsky, Y., M. G. Kaiser, and S. J. Lamont. 1995. Genetic characterization of highly inbred chicken lines by two DNA methods: DNA fingerprinting and polymerase chain reaction using arbitrary primers. Anim. Genet. 26:163 170. Ponsuksili, S., K. Wimmers, and P. Horst. 1996. Genetic variability in chickens using polymorphic microsatellite markers. Thai J. Agric. Sci. 29:571 580. Romanov, M. N., and S. Weigend. 2001. Analysis of genetic relationships between various populations of domestic and jungle fowl using microsatellites markers. Poult. Sci. 80:1057 1063. Saitou, N., and M. Nei. 1987. The neighbor-joining method: A new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406 425. Siegel, P. B., A. Haberfeld, T. K. Mukherjee, L. C. Stallard, H. L. Marks, N. B. Anthony, and E. A. Dunnington. 1992. Jungle fowl-domestic fowl relationships: A use of DNA fingerprinting. World s Poult. Sci. J. 48:147 155. Smyth, J. R., Jr. 1976. The inheritance of melanic pigmentation in the fowl. Pages 59 86 in Proc. 25th Poult. Breed. Roundtable., Kansas City, MO. Univ. Missouri, Kansas City. Tadano, R., M. Sekino, M. Nishibori, and M. Tsudzuki. 2007. Microsatellite marker analysis for the genetic relationships among Japanese long-tailed chicken breeds. Poult. Sci. 86:460 469. Takahashi, H., K. Nirasawa, Y. Nagamine, M. Tsuddzuki, and Y. Yamamoto. 1998. Genetic relationships among Japanese native breeds of chicken based on microsatellite DNA polymorphisms. J. Hered. 89:543 546. Takezaki, N., and M. Nei. 1996. Genetic distances and reconstruction of phylogenetic trees from microsatellite DNA. Genetics 144:389 399. Tautz, D. 1989. Hypervariability of simple sequences as a general source for polymorphic DMA markers. Nucleic Acids Res. 17:6463 6471. Twito, T., S. Weigend, S. Blum, Z. Granevitze, M. W. Feldman, R. Perl-Treves, U. Lavi, and J. Hillel. 2007. Biodiversity of 20 chicken breeds assessed by SNPs located in gene regions. Cytogenet. Genome Res. 117:319 326. Wimmers, K., S. Ponsuksili, T. Hardge, A. Valle-Zarate, P. K. Mathur, and P. Horst. 2000. Genetic distinctness of African, Asian and South American local chickens. Anim. Genet. 31:159 165. Ya-Bo, Y., W. Jin-Yu, D. M. Mekki, T. Qing-Ping, L. Hui-Fang, G. Rong, G. Qing-Lian, Z. Wen-Qi, and C. Kuan-Wei. 2006. Evaluation of genetic diversity and genetic distance between twelve Chinese indigenous chicken breeds based on microsatellite markers. Int. J. Poult. Sci. 5:550 556. Yeh, F. C., R. Yang, and T. Boyle. 1999. POPGENE (version 1.31): Microsoft Windows-based freeware for population genetic analysis. Univ. Alberta, Edmonton, Canada. Zhang, X., F. C. Leung, D. K. Chan, G. Yang, and C. Wu. 2002. Genetic diversity of Chinese native chicken breeds based on protein polymorphism, randomly amplified polymorphic DNA, and microsatellite polymorphism. Poult. Sci. 81:1463 1472. Zhou, H., and S. J. Lamont. 1999. Genetic characterization of biodiversity in highly inbred chicken lines by microsatellite markers. Anim. Genet. 30:256 264.