Four of the microsatellite markers that amplified successfully produced anonymous amplification products suggesting a second annealing site in the genome sequence of Blacks. These market forces are what ultimately caused the initial collapse of the profitable ostrich industry. Ostrich feathers, which initially spawned the industry, are today only a by-product of the ostrich industry.
THE SOUTH AFRICAN OSTRICH INDUSTRY
World ostrich production figures for 2004 (Figure 1.1) show that South Africa remains the world leader with 52% of the market (Stewart 2004). Ostrich farms have spread from the Klein Karoo in the Oudtshoorn area of the Western Cape to other provinces. South Africa enjoys a leading position in the ostrich industry due to its long heritage and natural conditions ideal for breeding these unique birds.
THE OSTRICH
However, unlike other birds, ratite sex chromosomes are monomorphic, meaning that the two chromosomes are indistinguishable in appearance as shown in the karyotype in Figure 1.3 (Ogawa et al. 1998; Petitte & Davis 1999). The structural similarity of sex chromosomes in this species is possibly reflected in the lack of sexual dimorphism in the juvenile ostriches (Takagi et al. 1972). However, these birds are larger and tend to be 'wilder' than the blacks, and very little is known about their performance under South African conditions.
EVOLUTIONARY AND DIVERSITY STUDIES
A subsequent study by Bezuidenhout in 1999 used mtDNA to investigate relationships among ostrich subspecies and to estimate genetic diversity among and within populations of the South African ostrich S. A recent genetic diversity study using nuclear DNA and microsatellites instead of mtDNA found the highest genetic variability in blacks and the lowest in reds in the studied population. This study also showed that there is a maximum genetic distance between blacks and reds, as seen in the neighbor-joining tree in Figure 1.5, indicating that the greatest effect of heterosis will be achieved by crossing subspecies (Kawka 2005).
- BREEDING PRACTICES
- MOLECULAR GENETICS IN THE OSTRICH INDUSTRY .1 Introduction
- Applications of fingerprinting
- APPLICATIONS OF FINGERPRINTING IN GENETIC ANALYSES
- AIMS
Polymorphisms are detected by differences in the lengths of the amplified fragments by polyacrylamide gel electrophoresis (PAGE) (Karp et al. 2001). By manipulating the number of nucleotides in the adapters, the number of amplified fragments can be adjusted (Karp et al. 2001). Individual identification requires methods that reveal the highest possible level of variation (Parker et al. 1998).
MATERIALS AND METHODS
INTRODUCTION
MATERIALS
Blood was used as a source of DNA because it is relatively easy to obtain and DNA yield is high due to Aves nucleated erythrocytes. Blood was drawn directly from the heart using a syringe and needle and squeezed into a Vacutainer™ EDTA tube.
METHODS
The supernatant containing the DNA (leaving the precipitated protein pellet behind) was then poured into a clean 1.5 ml microfuge tube containing 600 µl of 100% isopropanol (2-propanol). Then 600 µl of 70% ethanol was added and the tube was inverted to wash the DNA. The tube was inverted to drain on clean absorbent paper and the sample was air dried for 10-15 minutes.
GENOTYPIC ANALYSIS
- Allele identification and sizing
- Construction of genotypes
- Quantification of alleles
CERVUS Version 2.0 (Marshall et al., 1998) was used to calculate allele frequencies of different alleles at different microsatellite loci. These calculations were performed separately for each individual from the breeding pair families and the individuals of the colony. The file containing the genotypic information was named GenotypesCorrected2.csv (BPGenotypes.csv for paired individuals) and was used as the input file.
PARENTAGE ASSIGNMENT
Input files for candidate parents were created by extracting the relevant parental genotypic data from the individual genotype data file into separate data files. Since neither parent was known, CERVUS recommends a two-step analysis, with the first step running the group of parents with fewer candidates, in this case men. For the female lineage analysis, step 1 of the wizard was modified so that the required offspring genotype input file was the male lineage analysis file and the threshold confidence level was set to "Strict" (Figure 2.5).
CONSTRUCTION OF PEDIGREES
DETERMINATION OF BREEDING STATISTICS
RESULTS
- INTRODUCTION
- DNA YIELD
- GENOTYPIC ANALVSIS
- Estimation of fragment sizes
- Genotypes
- PARENTAGE ASSIGNMENT
- PEDIGREES
- BREEDING STATISTICS
The second step in optimizing the amplification conditions required adjusting the individual annealing temperatures for the different microsatellite loci. The sizes of the amplification products were used to determine and name the different alleles at different microsatellite loci. Through careful examination of the fingerprints, snagging bands were excluded from the final genotype pool.
The frequencies of the different alleles of the different microsatellite loci were calculated from the input file generated with the genotypes, named GenotypesCorrect2.csv (BPGenotypes.csv). The output file for the allele frequency analysis, Alle/eFreq.alf (AlleleFreqBP.a/f), was required as an input file for the simulation. The results obtained from the pedigree analyzes of the colony were expressed in number of assignments (Table 3.4).
Pedigrees were constructed using either microsatellite fingerprinting or the results of parentage analyses. The alleles of the offspring correspond to the alleles of their parents within 4 base pairs. For the colony, since the parentage was unknown, the results of the parentage analyzes generated by CERVUS were used.
Breeding statistics were calculated for the production potential of the individual females in the colony.
DISCUSSION AND CONCLUDING REMARKS
However, optimization of the protocol was necessary, as little or no amplification resulted, and therefore the number of cycles was increased. Three of the eleven markers were not amplified across a range of annealing temperatures, indicating a low homology of these primers, so further optimization with respect to other PCR variables is required. Additional amplification products were detected outside the expected product size range for four of the markers tested.
This was not unexpected and provides indications of genome diversity between these two subspecies. It was interesting to note that five loci showed a higher number of alleles than that reported in the literature (Kimwele et al. A comparative study by Kimwele and Graves (2003) on a wild population of Reds using four identical markers showed more high observed heterozygosity values.
Analysis of the genotypes of the colony individuals revealed that 90% of the individuals had a low number of heterozygous loci. Low heterozygosity indicates inbreeding in the population and it remains to be investigated whether this is the causative factor of the high chick mortality commonly observed in these birds. However, it should be noted that the sample size of the breeding pair families was small.
Parentage technology could facilitate the monitoring and evaluation of breeding stock so that low or non-producers can be removed from the breeding system.
Proceedings of the 7th World Congress on Genetics Applied to Livestock Production, Montpellier, France, August 19-23, 23-37. Deeming DC, Ayres L, Ayres FJ (1993) Observations on the commercial production of ostrich (Struthio camelus) in the United Kingdom: rearing chicks. Griffiths R, Orr K (1999) The use of amplified fragment length polymorphism (AFLP) in the isolation of sex-specific markers.
Harlid A, Janke A, Arnason U (1997) The mtDNA sequence of the ostrich and the difference between paleognathous and neognathous birds. Jarvis MJF (1998) The subspecies and breeds of ostriches and their current status in the wild. Kimwele CN, Graves JA, Burke T, Hanotte 0 (1998) Development of microsatellite markers for parentage typing of chicks in the ostrich Struthio camelus.
Lambrechts H (2004) Reproductive efficiency of ostriches (Struthio camelus), PhD thesis, University of the Free State, South Africa. Malecki lA, Martin GB (2002b) Egg fertilization in emu and ostrich - How much sperm do they need. Ogawa A, Murata K, Mizuno S (1998) The location and sequence of Z- and W-linked marker genes on the homomorphic sex chromosomes of the ostrich and emu.
Van Tuinen M, Sibley CG, Hedges SB (1998) Phylogeny and biogeography of ratites inferred from DNA sequences of the mitochondrial ribosomal genes.
APPENDIX A
BUFFERS AND REAGENTS
APPENDIX B OUTPUT FILES
Polymorphic information content (PlC) : 0.744 Average probability of exception Average probability of exception Estimation of null allele frequency: Not done. Average error rate observed across all locations: 0.6658. assumes that all known parent-offspring pairs are equally independent). Polymorphic Information Content (PIC): 0.885 Mean probability of exception Mean probability of exception Hardy-Weinberg equilibrium test: Not done Null allele frequency estimate: 0.7966.
Expected heterozygosity: 0.936 Polymorphic information content (PlC): 0.926 Average exclusion probability Average exclusion probability Hardy-Weinberg equilibrium test: not done Estimated null allele frequency: 0.6266. Polymorphic information content (PlC): 0.969 Average exclusion probability Average exclusion probability Hardy-Weinberg equilibrium test: not performed Estimated null allele frequency: 0.2693. Polymorphic information content (PlC): 0.972 Average exclusion probability Average exclusion probability Hardy-Weinberg equilibrium test: not performed Estimated null allele frequency: 0.2567.
Locus name N matched N unmatched N null Probability of detection. assumes that all known parent-offspring pairs are equally independent).
APPENDIX C
ARTICLES PUBLISHED
In eight triplets, it was possible to tell the difference between the weight of the eggs and assign the eggs to one or the other female. Egg weight was then plotted against production date to see if the eggs of the two females in the trio could be distinguished by egg weight. The ASREML program (Gilmouret al., 1999) was used to adjust for the random effects of the original sire and then the dam to obtain an indication of the reproducibility of egg weight and chick weight.
It was reasoned that agreement of these estimates with the literature would provide an indication of the success achieved in allocating eggs to individual females. It was possible to differentiate between females based on egg weight differences for eight of the 14 trios (Figure 1). In half of the trios examined, no difference could be distinguished between the hatchability of eggs from the two females (Table 2).
This study uses available microsatellite markers to attempt a comprehensive parentage analysis. The origin, background and rearing of the commercial population at the center are described in the literature (van Schalkwyk et al., 1996; Bunter, 2002). Marshallet sod. (998) suggested that the number of loci needed to resolve parentage with a given level of confidence depends on factors such as the degree of variation at the locus (expected heterozygosity), the number of parent candidates, the proportion of parent candidates sampled, and the availability of genetic data from known parent.
The high estimated error rate arises from the high frequencies of null alleles observed at the loci.
