Genetic parameters for reproduction traits and correlation with pre weaning growth traits of Fogera cattle at Metekel ranch, north west Ethiopia

Almaz Bekele, Zewdu Wuletaw¹, Aynalem Haile², Solomon Gizaw³ and Getinet Mekuriaw⁴

Wolaita Sodo University, Ethiopia
bekele.almaz1@gmail.com
¹ Deutsche Gesellschaft fuer Intenationale Zusammenarbeit,(GIZ) GmbH, Bahir Dar, Ethiopia
² International Centre for Agricultural Research in Dry Areas (ICARDA), Aleppo, Syria
³ International livestock research institute, Addis Ababa, Ethiopia
⁴ International livestock research institute, Kenya.

Abstract

The objective of this study was to estimate genetic parameters of reproduction traits of pure Fogera cattle at Metekel Ranch. Metekel cattle Breeding and Improvement Ranch has so far been engaged in maintenance of Fogera cattle population outside their adapted environment (ex-situ conservation). The breeding program has two components: selection and crossbreeding. The establishment of the pure breed unit is meant for the improvement of the Fogera breed and for providing heifers to cross- breed to exotic dairy sires (by Artificial insemination). In cross breeding program; crossbred animals are produced through artificial insemination of Fogera cows with Friesian semen. Around three to six months of pregnancy, the F1 cross heifers are sold to farmers for milk production (Melaku et al., 2011a, b). Very recently the ranch started to distribute non pregnant F1 cross heifers. WOMBAT software was used to estimate genetic parameters. The variance components and heritability were estimated using a Uni-variate animal model using four models which fitted direct additive, dam genetic and permanent environmental effect as a random effect and the fixed effects (year, season, sex and parity). Parameter of age at first calving (AFC) were estimated using Model 1 (Y= Xb + Z₁a + e), Model2 (Y= Xb + Z₁a + Z₃c + e), Model3 (Y= Xb + Z₁a + Z₂m + e (cova, m = 0), and Model4 (Y= Xb + Z₁a + Z₂m + Z₃c + e (cova, m = 0) whereas parameters of reproductive traits including gestation length (GL), calving interval (CI) and days open (DO) were estimated using model 2 and 4 which fit permanent environmental effect due to repeated records per cow. Correlations (genetic and phenotypic) among the different traits were estimated from bi­-variate analysis by using model 1 for growth traits and AFC and model 2 to estimate the correlation between CI, GL and DO and birth weight (BWT) and GL. Correlation between birth weight and gestation length were estimated by treating gestation length as a trait of calf. Due to record limitation correlation between AFC with other reproductive traits were not estimated.

Estimates of direct heritability of reproductive performance traits from the best model were 0.003 ± 0.05 for AFC, 0.00 ± 0.03 for GL and CI and 0.013 ± 0.03 for DO. The phenotypic correlations between reproductive traits were 0.003 ± 0.034 for CI and GL, 0.37 ± 0.041 for CI and DO and 0.167 ± 0.036 for DO and GL and the genetic correlation between reproductive traits ranged from -0.94 ± 0.2 for DO and GL to 0.83 ± 0.579 for CI and DO. Genetic correlation between growth and AFC ranged from 0.77 ± 0.24 for BWT and AFC to 0.87 ± 0.03 for pre-weaning average daily gain (PADG) and AFC. The phenotypic and genetic correlations between BWT and GL were 0.03 ± 0.05 and -0.84 ± 0.013 respectively. The results of genetic correlation between considered traits was ranged from moderate to high and it indicates that selection for one trait would have a significant effect on the other traits not considered however selection must be done with caution. The heritability estimates confirmed the presence of high environmental effect among the study population and it masks the individual animal genetic difference. Given the low heritability estimates obtained, effective improvement in reproduction performance of Fogera cattle could be achieved by improving the production conditions and through crossbreeding.

Key words: correlation, genetic improvement, heritability, selection

Introduction

The process of genetic improvement of animals is systematic and follows several important steps: definition of breeding objectives, development of selection criteria, genetic evaluations, selection of animals and finally design of appropriate mating systems (Kluyts et al 2003; Thomas 2004). The efficiency of genetic improvement program is dependent on animal selection using precise and accurate estimates of performance parameters for prediction of breeding values (Maniatis and Pollott 2003; Sendros et al 2004; Ilatsia et al 2007). Reproductive performances are usually a critical censure of the efficiency of the production system since they affect the herd size and off take. Reproductive performance is commonly evaluated by analyzing female reproductive traits (Wasike 2006). Reproductive performance is a complex trait that has many components and heritability estimates of reproductive traits are often low indicating that genetic improvement may be slow (Gutiérrez et al 2002). Most reproductive traits are heavily influenced by differences in herd management practices and other environmental factors rather than genetic factors (Yosef 2006; Cammac et al 2009).

The Fogera cattle are used for milk, meat and draft power by smallholder farmers and known by their milk production performance (Belete et al 2010). Selection on Fogera cattle is being undertaken in different ranches (IBC 2004). However, the selection activity on the Fogera cattle in Metekel Cattle Breeding and Improvement Ranch has been exclusively based on morphological characters and has not led to genetic improvement in any of the production traits (Fasil 2006). Further development of the genetic improvement programme could be achieved once genetic parameters for the population of interest are known and this study was initiated with the objectives of to estimate genetic parameters of reproductive traits of Fogera cattle.

Materials and methods

Description of the study Area

Metekel Cattle Breeding and Improvement Ranch is found in Guangua district of Awi zone in Amhara National Regional State, and is situated about 505 km North-west from Addis Abeba. The ranch was established in 1986 and became functional in 1988. The annual mean relative humidity is 61.7% and it reaches to high from June to October (76.7-83.8%). The ranch receives an average annual rain fall of 1730.0 mm; average temperature ranges from 13.7 to 29.5⁰ (ENMA 2010). The rain fall distribution is bi-modal. The area has three season; long rainy season (June-October), short rainy season (March-May) and dry season (November-February) (Melaku et al 2011a, b; Addisu and Hegde 2003).

Herd management and Breeding program Metekel cattle Breeding and Improvement Ranch has so far been engaged in maintenance of Fogera cattle population outside their adapted environment (ex-situ conservation). The cattle were herded based on breed, sex and age. For ease of herd management animals were kept at three sites. On the ranch, calves were weighed on the date of birth and identified within 72 hours of birth. Calves were weaned and weaning weights taken at 8 months of age.

The breeding program has two components: selection and crossbreeding. The purchase of cattle is done to build up the number of breeding cows in accordance with the carrying capacity of the ranch and to provide sufficient head for cross breeding. According to the breeding plan of the ranch, Fogera heifers were allowed for mating for the first time when they are 24 months of age and have attained a minimum body weight of 250 Kg. If the heifer has not attained the minimum body weight her age of mating was extended. In cross breeding program; crossbred animals were produced through artificial insemination of Fogera cows with Friesian semen. Around three to six months of pregnancy, the F₁ cross heifers were sold to farmers for milk production (Melaku et al 2011a, b). Very recently the ranch started to distribute non pregnant F ₁ cross heifers.

Data Source and Data management system

Data found in Metekel ranch were collected for this study. Farm records of Metekel ranch from 1990-2011 were used. Parity was classified as 1, 2, 3, 4 and those parities from the fifth and above were considered as parity five because of very few observations available. Season was classified into three (dry season, short and long rainy season) based on the rain fall distribution of the area. The animals that have abnormal calving, i.e., abortion and stillbirths were not included in the analysis of breeding data. AFC above five years and CI above two years were also excluded from the data. Records with irregularity in pedigree information and dates were discarded. Records available for analyses within traits are summarized in Table 1.

Genetic parameter estimation

The variance components and heritability were estimated using a univariate animal model using four models which fitted direct additive, dam genetic and permanent environmental effect as a random effect and the fixed effects by using WOMBAT. Heritability for AFC was estimated using Model 1, 2, 3, and 4 where as parameters of reproductive traits including GL, CI and DO were estimated using model 2 and 4 which fit permanent environmental effect due to repeated records per cow. Correlations (genetic and phenotypic) among the different traits were estimated from bi-variate analysis by using model 1 to estimate correlation between growth traits and AFC and model 2 to estimate the correlations between CI, GL and DO and BWT and GL. Correlation between birth weight and gestation length was estimated by treating gestation length as a trait of calf.

The model equations used for the analysis were:

Model 1 Y= Xb + Z₁a + e

Model 2 Y= Xb + Z₁a + Z₃c + e

Model 3 Y= Xb + Z₁a + Z₂m + e (cova, m = 0)

Model 4 Y= Xb + Z₁a + Z₂m + Z₃c + e (cova, m = 0)

Where, Y = the vector of records, b = vector of fixed effects, X = incidence matrix of fixed effects, a = vector of direct additive genetic effect, m = vector of maternal additive genetic effect, c = vector of permanent environmental effect, Z₁ = incidence matrix for direct additive genetic effect, Z₂ = incidence matrix for maternal additive genetic effect, Z₃ = incidence matrix for permanent environmental effect and e = vector of random errors. The fixed effect included in the model includes year, season, sex of calf and parity. For AFC the sex effect was not fitted because it had no significant effect on GL so it was excluded from the model.

Result and discussion

Variance component and heritability

The result of variance components and heritability estimation for reproductive traits are summarized in Table 2 and 3. The best model for all reproductive traits was Model 4 which includes both maternal genetic and permanent environmental effect. The likelihood changes in model 4 for AFC and GL were not significant but model 4 was selected as a best model because the maternal genetic effect was greater than the direct genetic effect. Estimated heritability value of maternal genetic effect for AFC was higher than direct genetic effect. Similarly estimated heritability value of direct genetic effect for CI was lower than the estimated value of maternal genetic and permanent environmental effects. It is consistent with the reports of Aynalem et al (2009) who found lower direct genetic effect (0.0014 ± 0.04) than permanent environmental (0.03 ± 0.09) effect on CI for Ethiopian Boran Cattle. It shows the existence of permanent environmental effect on reproductive traits. Variations due to direct genetic effects were low and error variance accounted for a higher proportion of total phenotypic variation.

Reproductive traits were highly influenced by the environment; the heritability estimates for all reproductive traits were low. The high environmental variance shows the poor management activity of the ranch decreases the performance of animals and reduces the variation between animals. It results low heritability value for all reproductive traits. Based on review of literatures by Cammac et al (2009) estimates of heritability for many reproductive traits were low, some exist that have moderate heritability but low heritability estimates do not necessarily suggest that genetic control of reproductive traits is less economically important than management inputs or vice versa.

The present result of heritability of direct genetic effect for AFC is comparable with 0·012 (Hailemariam and Kassa 1994) from bi-variate analysis. However it is less than the reports of Oyama et al (2002) (0.215 ± 0.026) for Wagyu cattle, Sendros et al (2004) (0.44 ± 0.05) for Ethiopian Boran; Oyama et al (2004) (0.2) for Wagyu cattle; Wasike (2006) (0.04) for Kenyan Boran; Ilatsia et al (2007) (0.04) for Sahiwal; Aynalem et al (2009) 0.22 ± 0.21 for Ethiopian Boran and Wasike et al (2009) (0.04 ± 0.06) for Kenyan Boran. It can be associated with data sets and model used. Different results of heritability h² for AFC were reported by Hailemariam and Kassa (1994) from original data and selected data for which is 0·062 and 0·075 when estimated from the whole data and selected data from which cows with AFC higher than five years were deleted respectively. Mohamed (2004) found largely different result of heritability for AFC from animal model (0.263 ± 0.06) and sire model (0.75 ± 0.17). When the traits have lower the heritability value there is greater the heterotic response from various crossbreeding mating systems. Improvement for AFC could be achieved by improving of management of the ranch and through cross breeding.

The low heritability estimates for CI indicates that CI is probably not under the influence of additive gene action and is an environmental trait. This could be attributable to the high phenotypic variance arising from high environmental variations. In agreement with the present result, Wasike (2006) and Wasike et al (2009) found zero direct heritability value for CI for Kenyan Boran and Aynalem et al (2009) found very low direct heritability value (0.0014 ± 0.037) for Ethiopian Boran. Estimated heritability of cow effect 0.03 ± 0.03 was comparable with the reports of Oyama et al (2002) 0.04 and 0.04 and Oyama et al (2004) 0·05 and 0.04 for Wagyu cattle and Aynalem et al (2009), 0.03 ± 0.093 for Ethiopian Boran respectively for maternal genetic and permanent environmental effect.

The result of direct heritability for DO (0.01 ± 0.03) was similar for model 2 and model 4. It is higher than the reports made by Aynalem et al (2009) (0.0006 ± 0.044) by using uni-variate model for Ethiopian Boran. Aynalem et al (2009) reported higher direct heritability value (0.1 ± 0.047) with the same model for the Boran crosses with Holstein Frisian. The present result is less than the reports of Oyama et al (2002) (0.04)

Correlations

The estimated phenotypic correlations between AFC and growth traits (BWT, AWWT and PADG) were found low; 0.24 ± 0.33, -0.04 ± 0.04 and -0.04 ± 0.04, respectively (Table 4) and positive correlation was found only for correlation of AFC with BWT. Comparatively low correlation results were found 0.072 ± 0.05 by Wakchaure and Meena (2010) for BWT and AFC for Sahiwal cattle and Wasike (2006) for AFC with WWT (-0.05) for Kenyan Boran cattle. The genetic correlations for AFC with growth traits were high; 0.77 ± 0.24, 0.82 ± 0.02 and 0.87 ± 0.03 respectively for AFC with BWT, AWWT and PADG, respectively (Table 4). Similarly Wasike (2006) found high genetic correlation for AFC and WWT (0.72) for Kenyan Boran cattle. A strong positive genetic relationship between growth traits and AFC indicating that selection for high weight would result large AFC or selection for early AFC would result in diminished growth performance.

High and negative genetic correlation (-0.84 ± 0.01) between BWT and GL was estimated (Table 4). It is suggested that faster growing fetuses may trigger parturition earlier than average or the birth process is initiated at an earlier stage of gestation among faster growing than slower growing fetuses (Bourdon and Brinks, 1982). It indicates that selection for birth weight will result positive genetic gain for GL. In the opposite Van Graan et al (2004) estimate positive genetic correlation between BWT and GL (0.40) for Bonsmara cattle in South Africa.

Low phenotypic correlations were estimated between reproductive traits (Table 5). Consistently, low phenotypic correlations among reproductive traits were reported by Oyama et al (2004); 0·00 for GL and CI and -0·03 for DO and GL but based on his report DO and CI (1.00) were strongly correlated. Van Graan et al (2004) estimate positive phenotypic correlation between BWT and GL (0.22) for Bonsmara cattle in South Africa and it was consistent with the present result.

The genetic correlation between reproductive traits was strong and it was ranged from strong positive 0.83 ± 0.58 and 0.72 ± 0.02 between CI and DO and GL and CI to strong negative -0.94 ± 0.2 between GL and DO (Table 5). Due to the strong genetic relationship between these traits, selection for one of them could have high effect on the other through correlated responses. Similarly Strong positive genetic correlation (1.0) was estimated by Oyama et al (2004) and Goyache et al (2005) between DO and CI. Oyama et al (2004) and Goyache et al (2005) report negative genetic correlation between GL and DO similar to the present study but lower in value -0·11 and -0.089, respectively. Strong positive correlation between reproductive traits indicates that selection for one traits will improve the other but factors to be considered when selecting for reproductive traits, is the strong negative genetic correlation between DO and GL as selection for improving DO will result large GL.

Conclusions

• The low heritability estimates indicate that the low genetic control of the expression of a trait. Therefore producers should decide on for a long-term strategy of achieving change in these traits firstly through improvement of the production environment and then by crossbreeding.
  
  
• Maternal genetic and permanent maternal environmental effects should be accounted for in genetic evaluation to eliminate the upward biasness in estimates of direct heritability.
  
  
• The correlations between traits should be taken care in selection and in designing of breed improvement. The results of genetic correlation between pre-weaning growth traits and AFC indicate that pre-weaning growth traits should be given some consideration in preliminary selection criteria of heifers and selection could be done with caution.
  
  
• Selection for birth weight might result in improvement for GL but heavier calves at birth may cause higher probability of dystocia. Therefore, optimal selection criterion should be strained under a control of large BW.

References

Addisu Bitew and Hegde B P 2003 Evaluation of reproductive and growth performance of Fogera cattle and their F₁-Friesian cross at Metekel Ranch, Ethiopia. Proceedings of the 10^th annual conference of the Ethiopian Society of Animal Production (ESAP) held in Addis Ababa, Ethiopia. ESAP, Addis Ababa.pp119-126. http://www.esap-ethiopia.org/index.php/publications/proceedings.

Aynalem Haile, Joshi B K, Workneh Ayalew, Azage Tegegne and Singh A 2009 Genetic evaluation of Boran cattle and their crosses with Holstein Friesian in central Ethiopia: reproduction traits. Journal of Agricultural Science. 147:81–89. http://journals.cambridge.org/article_S0021859608008095.

Belete Anteneh, Azage Tegegne, Fekadu Beyene and Berhanu Gebremedhin 2010 Cattle milk and meat production and marketing systems and opportunities for market-orientation in Fogera woreda, Amhara region, Ethiopia. IPMS (Improving Productivity and Market Success) of Ethiopian Farmers Project Working Paper 19. ILRI (International Livestock Research Institute), Nairobi, Kenya. pp65. https://cgspace.cgiar.org/bitstream/handle/10568/1686/WP?sequence=1.

Bourdon R M and Brinks J S 1982 Genetic, environmental and phenotypic relationships among gestation length, birth weight, growth traits and age at first calving in beef cattle. Journal of Animal Science 55:543. https://dl.sciencesocieties.org/publications/jas/pdfs/55/3/JAN0550030543.

Cammack K M, Thomas M G, PAS and Enns R M 2009 Review: Reproductive Traits and Their Heritabilities in Beef Cattle. Journal of Animal Science 25: 517–528. http://www.professionalanimalscientist.org/article/S1080-7446(15)30753-1/fulltext.

Fasil Getachew 2006 On-farm phenotypic characterization of cattle genetic resources and their production systems in awi, east and west gojjam zones of amhara region, Ethiopia.M.Sc. Thesis, Alemaya University. Ethiopia.pp153.

Getinet Mekuriaw, Workneh Ayalew and Hegde B P 2009 Growth and reproductive performance of Ogaden cattle at Haramaya University, Ethiopia. Ethiopian Journal of Animal Production. Vol.9 N.1. Pp13-38. http://esap-ethiopia.org/Publications/Journals/EJAP_Volume_9.pdf.

Goyache F, Gutierrez J P, Fernandez I, Royo L J and Alvarez I 2005 Genetic analysis of days open in beef cattle. Livestock Production Science, Vol 93(3): 283- 289. http://www.sciencedirect.com/science/article/pii/S0301622604002362.

Gutierrez J P, Alvarez I Fernandez I, Royo L J, Dıez J and Goyache F 2002 Genetic relationships between calving dates, calving interval, age at first calving and type traits in beef cattle. Livestock Production Science 78, 215 – 222. http://ria.asturias.es/RIA/bitstream/123456789/727/1/geneticrelation.pdf.

Haile-Mariam M and Kassa-Mersha H 1994 Genetic and environmental effects on age at first calving and calving interval of naturally bred Boran (zebu) cows in Ethiopia. Animal Production, 58: 329-334. http://journals.cambridge.org.

IBC (Institute of Biodiversity Conservation) 2004 The State of Ethiopia's Farm Animal Genetic Resources: Country Report. A Contribution to the First Report on the State of the World's Animal Genetic Resources. IBC. Addis Ababa, Ethiopia. ftp://ftp.fao.org/docrep/fao/010/a1250e/annexes/CountryReports/Ethiopia.pdf.

Ilatsia E D, Muasya T K, Muhuyi W B and Kahi A K 2007 Milk production and reproductive performance of Sahiwal cattle in semi-arid Kenya. Tropical Science 47(3):120 – 127. http://onlinelibrary.wiley.com/doi/10.1002/ts.205/pdf.

Kluyts J F, Neser F W C. and Bradfield M J 2003 Review Article: Development of breeding objectives for beef cattle breeding: Derivation of economic values. South African Journal of Animal Science, 33 (3): 142-158. http://www.ajol.info/index.php/sajas/article/download/3768/11833.

Maniatis N and Pollott G E 2003 The impact of data structure on genetic (co)variance components of early growth in sheep, estimated using an animal model with maternal effects. Journal of Animal Science, 81: 101-108. https://www.animalsciencepublications.org/publications/jas/pdfs/81/1/101.

Melaku M, Zeleke M, Getinet M and Mengistie T 2011 Pre-weaning growth performances of Fogera calves at Metekel cattle improvement and multiplication ranch, North West Ethiopia.Livestock Research for Rural Development. Volume 23, Article #182. Retrieved November 4, 2016, from http://www.lrrd.org/lrrd23/9/mela23182.htm.

Melaku Minale, Zeleke Mekuriya, Getinet Mekuriya and Mengistie Taye 2011b Reproductive Performances of Fogera Cattle at Metekel Cattle Breeding and Multiplication Ranch, North West Ethiopia. Online Journal of Animal and Feed Research 1(3): 99-106. http://www.ojafr.ir/main/attachments/article/68/OJAFR-A15.pdf.

Mohammed Aliy 2004 Estimates of genetic parameters of birth weight, age at first calving and milk production traits in Holstein Friesian dairy herds kept in three state farms. An MSc Thesis presented to School of Graduate studies of Alemaya University, p. 89.

Oyama K, Katsuta T, Anada K and Mukai F 2002 Heritability and repeatability estimates for reproductive traits of Japanese Black cows. Asian-Australasian Journal of Animal Science 15(12), 1680-1685. http://www.ajas.info/journal/view.php?number=20323.

Oyama K, Katsuta T, Anada K and Mukai F 2004 Genetic parameters for reproductive performance of breeding cows and carcass traits of fattening animals in Japanese Black (Wagyu). Animal Science 78 (2), 195-201. http://journals.cambridge.org/abstract_S1357729800053984.

Sendros D, Neser F W C and Schoeman S J 2004 Estimates of genetic parameters for Boran, Friesian, and crosses of Friesian and Jersey with the Boran cattle in the tropical highlands of Ethiopia: reproduction traits. Journal of animal breeding and genetics 121(1): 57-65. http://onlinelibrary.wiley.com/doi/10.1046/j.0931-2668.2003.00438.x/full.

Takayuki I B I, Alexander K K and Hiroyuki H 2008 Genetic parameters for gestation length and the relationship with birth weight and carcass traits in Japanese Black cattle. Animal Science Journal 79 (3): 297–302. http://onlinelibrary.wiley.com/wol1/doi/10.1111/j.1740-0929.2008.00530.x/full.

Thomas R 2004 Development of breeding objectives for production systems utilizing the Boran Breed in Kenya. MSc Thesis, Egerton University, Kenya. http://www.borankenya.org/Thesis_for_Mr._T._O._Rewe.pdf.

Van Graan A C, Neser F W C, van der Westhuizen J and Theron H E 2004 Genetic and phenotypic parameter estimates of gestation length and birth weight in Bonsmara cattle. South African Journal of Animal Science, 34 (Supplement 2): 95- 97. http://www.sasas.co.za/sites/sasas.co.za/files/VanGraan1Goudini_0.pdf.

Wakchaure R S and Meena R 2010 Factors affecting, birth weight, age and weight at first calving in sahiwal cattle. Indian Journal of Animal Research, 44 (3): 173 – 177. http://animres.edpsciences.org/

Wasike C B 2006 Genetic evaluation of growth and reproductive performance of the Kenya Boran cattle. MSc Thesis. Egerton University, Kenya.pp108.

Yosef Tadesse 2006 Genetic and Non-Genetic analysis of fertility and production traits in Holetta and Ada’a Berga Dairy herds. MSc.Thesis, Alemaya University, Ethiopia.pp143.

Received 3 November 2016; Accepted 17 July 2017; Published 1 August 2017

Go to top