2.4.1. Phenotypic trait.

Means and standard deviations were calculated for all traits by using the sample sizes as weights. The total number of records for each phenotypic trait was calculated as the sum of number of records in each article that reported the trait. The coefficient of variation in percentage (CV (%)) for each i^th trait was calculated as follows: CV (%) = si/Xi *100, where si is the standard deviation for the i^th trait, and Xi is the trait mean.

2.4.2. Heritability and genetic correlation.

Meta-analysis was performed on the basis of a random-effects model [12], in which the parameter estimates for all traits were analysed by assuming independence and normality.

These assumptions were investigated for each trait by using the Box–Pierce and Shapiro–Wilk tests, available respectively in the box.test and shapiro.test functions of the R software [19]. The meta for package [20] available in the R software [19] was used to fit the following random-effects model for each trait: where is the published parameter estimate in the j^th article, is the weighted population parameter mean, uj the among study component of the deviation from mean, assumed as u_i~(N(0, τ²), where τ² is the variance representing the amount heterogeneity among the studies, e_j is with-in study component due to sampling error in the parameter estimate in the jth article, assumed as e_j~(N(0, σ_e²), where σ_e² is the with-in variance. The I² index [21] used to quantify the degree of heterogeneity among studies (τ²) for each trait can be described as follows: I² = ((Q-df)/Q) x100,

Where Q is the Q statistics [22] given by

Where w_j is the parameter estimate weight (as the inverse of published sampling variance for the parameter, 1/v) in the j^th article; were defined above in the random-effects model; and the d.f. is the degrees of freedom (J-1, where J is the number of used articles) of a chi-squared distribution assumed for expected Q value on the assumption that τ² = 0. Basically, I² values estimated as 25%, 50% and 75% might be considered as low, moderate and high heterogeneity respectively [21]. The 95% lower and upper limits for the estimated parameter would be computed respectively for each trait as follows:

Where the predicted standard error for the estimated parameter ɵ, given by

3.1.1. Adaptation to harsh environmental conditions.

The summary of results of the systematic review of adaptation characteristics of Boran and other zebu cattle breeds were given under this subsection. The focus of the 28 articles reviewed were split into seven categories of adaptation traits, which included morphological (11), behavioural (8), physiological (7), neuro-endocrine (6), blood biochemical (7), metabolic (4) cellular and molecular (9) adaptation traits of Boran and other zebu cattle breeds in the tropics.

Boran cattle have tremendous adaptive capacity to harsh environment through specific morphological and physiological characteristics [6, 7, 23, 24]. On the other hand, heritability (h²) estimates of adaptation traits were not determined for Boran cattle in this study due to scarcity of published data for the traits. In some tropical cattle breeds heritability estimates for heat tolerance (h² = 0.18–0.75) and tick resistance (h² = 0.15–0.44) showed a wide range of variabilities [25–27]. Heritabilities for resistance to nematode (h² = 30–0.38) and for trypanotolerance (h² = 0.30–0.88) vaied from medium to high heritabilities [26, 28]. Available reports [1, 26, 29] show that Boran cattle have adaptive traits to heat stress, external and internal parasites, and as well as to some blood parasites such as trypanssomiasis and east cost fever. These adaptive mechanisms of the Boran cattle to the harsh environmental conditions could be due to morphological, behavioural, physiological, neuro-endocrine, blood biochemical, metabolic, and molecular and cellular responses which are combined to promote survival and production in a specific environment [30].

3.1.2. Morphological adaptation.

Indigenous cattle breeds adapted to arid and semi-arid regions possess special morphological features, such as skin coat, which helps to provide better protection from direct solar radiation [31]. Phenotypically, the typical Boran cattle breed (Qorti subtype) have white color, large dewlap and well-developed hump. But mostly they have light grey or fawn colour with black or dark brown shading on the neck, head, shoulders and hindquarters, shorter and more pendulous sheath, well developed udder, long legs, wide ears and large dewlap, small humps, a short tail and erect and short horns with dominantly thick base [24, 32].

Phenotypic characteristics are resultant of adaptation mechanisms of the breed. Coat colour was one of the important morphological traits which impart adaptive ability to heat stressed livestock. For instance, white/ light grey coloured coats are helpful in thermo-regulation ability in tropical regions as it reflects 50% to 60% of direct solar radiation compared with the dark-coloured animals [1, 33]. Light or white-coloured coated animals are recognized as being advantageous in hot tropical regions. On the other hand, highly pigmented skin protects the deep tissues from direct short wave UV radiation by blocking its penetration. In addition, skin hair coat length, thickness and hair density also affect the adaptive nature of animals in tropical regions, where short hair, thin skin and fewer hair follicles per unit area are directly linked to higher adaptability to warm climate [34, 35]. Indigenous cattle breeds are more adapted to high temperatures, high solar radiation and dry conditions than exotic cattle breed due to their high skin pore density which allows them to successfully regulate their body temperature [36]. The Boran cattle also exhibit long legs to trek long distance (usually more than 60 km/day) in search of feed and water [24, 29] and wide ear and large dewlap to increase surface area for metabolic heat dissipation [7, 37].

In addition, the Boran cattle have higher level of resistance to biting insects and tick infestation due to: (1) A highly sensitive and motile skin with thick and well-developed layer of subcutaneous tissue causes the muscles beneath their skin to contract and move in reaction to insects landing on and biting them. This enables them to vigorously shake off external pests, (2) A very short hair coat makes it difficult for insects to attach onto hides of Boran cattle (3) A waxy and oily secretion from the skins of the Boran cattle makes them less desirable host for ticks and flies and (4) A long tail with a big well-formed twitch used to ride off vector flies for eye infection from Boran cattle [34, 38]. It was also reported that Boran cattle have prominent, protective eyebrows and long eyelashes which protect their eyes from bright sunlight, dust and other irritants factors that predispose cattle to pinkeye infection [1]. The Boran cattle (Kenyan Boran) tend to present some resistance to trypanosomal infection in the tsetse-infested regions of the east coast of North East Kenya [2]. Borana cattle both bulls and cows usually have horns and their horns are relatively smaller in size which was believed to have lower energy requirement for maintenance and have probably more adaptive capacity as compared to cattle breeds with large horns [1]. Morphological traits in livestock are highly important from the adaptation point of view, as they directly influence the heat exchange mechanisms between the animal and the surrounding environment [33].

3.1.3. Behavioural adaptation.

In an effort to adapt to varying environmental conditions, animals exhibit several behavioural responses. The most important behavioural responses studied in tropical cattle include: shade seeking, standing time, feed intake, defecating and urinating frequency, water intake and frequency [39]. Tropical indigenous cattle breeds were observed to be highly adapted to direct heat stress, spending more time for grazing than resting in shade. Zebu cattle have the ability to adapt their grazing behaviour in response to restricted grazing time or when no grazing is allowed during nights [40–43]. Several studies show that different Zebu breeds spend more than 4% of their pasture time resting which varies between 4.4% and 10.12% [41, 42]. In a study involving Boran cattle, the resting time is as low as 2% [40]. The high grazing frequency and low resting time of the Boran show its special ability to utilize the available pasture efficiently during grazing time [43].

Another behavioural response to climate stress in ruminants is the reduction in feed intake and utilization in arid regions as adaptive response to regulate internal metabolic heat production under hot environment [44, 45] and decreasing water excretion by concentrating urine were also alternative strategies to adapt to severe climate stress for livestock in arid and semi-arid of East African rangelands [45, 46]. Increased standing and decreased lying time were also reported to be associated with higher ambient temperatures [47, 48]. Generally, heat stressed animals tend to spend more time standing so that they can reorient themselves in different directions to avoid direct solar radiation and ground radiation. In addition, the standing position also obstructs the conductive heat transfer into the animal body due to the presence of a layer of air adjacent to the skin, and also facilitates the dissipation of body heat load to the surroundings by increasing the amount of skin exposed to air flow or wind.

Boran cattle breed have developed high degree of favourable behavioural adaptation to harsh environment due to genetic basis, human and natural selection for many generations [2]. They have developed ability to withstand periodic shortage of water and feed, ability to walk long distances in search of water and feed and ability to digest low quality feeds from degraded arid rangelands [20]. The Boran cattle are non-selective aggressive grazers and browsers [1]. The Boran herd have a positive social behaviour with good mothering ability which may reduce chance of being attacked by predators which helped them to adapt to free grazing rangelands [41].

3.1.4. Physiological adaptations.

Some of the physiological mechanisms of adaptations to heat stress are respiration rate, rectal temperature, pulse rate, skin temperature and sweating rate. Increased sweating rate, high respiration rate, vasodilation with increased blood flow to skin surface, reduced metabolic rate, decreased dry matter intake and altered water metabolism are the physiologic responses that have negative impact on the production and reproduction of the cows [49]. All these physiologic responses are substantial and prolonged in Bos taurus than in Bos indicus [50]. Hence the consequences of exposure to heat stress for production of milk and meat are less pronounced in the tropical cattle [30].

Excessive heat causes decreased food intake and disturbances in protein and energy metabolism, mineral balance, enzymatic reactions, hormones and metabolites secretion in the blood [30]. Metabolic disorders caused by thermal stress led to reduced milk production, growth and reproductive rates and increase the susceptibility of animal diseases causing economic loss [51–53]. Moreover, climate change causes an increase in average temperature and reduced rainfall, putting the sustainability of the livestock production system in risk, especially in countries such as Ethiopia which already has high air temperature averages and grazing systems dependent on the rainy season [6, 7, 54].

The Boran cattle is one of the most productive indigenous cattle breeds in east Africa which is capable of surviving and reproducing under the prevailing harsh climatic, nutritional and management conditions of the region while maintaining good productivity on poor forage and low water availability [1, 37]. This is because they have physiologically adapted and develop lower maintenance requirements than Bos taurus cattle [55]. Boran cattle have higher level of lipoprotein-lipase enzyme activity in the subcutaneous fat depot, which enables them to survive drought in the Borana rangelands with drastic recoveries after drought years when pasture condition improves [1, 7, 29]. The Boran cattle are non-selective feeders and browsers which give them the ability utilize shrubs, trees and dry unpalatable grasses that often not consumed by other cattle breeds [6, 34].

3.1.5. Neuro-endocrine response.

Several studies of various livestock species clearly established higher plasma cortisol level in ruminants during heat stressed conditions [30, 56, 57] reported that the plasma cortisol level was significantly lower in multiple stressors groups (heat, nutrition and walking) as compared with individual (heat stress/nutritional stress) or combined stresses (heat and nutrition stress). Aldosterone is another steroid hormone released from the cortex of the adrenal glands and is involved in the regulation of water and mineral balance in the body. It is a well-established fact that during heat stress conditions ruminants may undergo severe dehydration, which may result in the activation of renin–angiotensin–aldosterone pathway to restore the water and electrolyte balance [58]. Severe dehydration may lead to increased secretion of antidiuretic hormone (ADH) through activation of renin–angiotensin–aldosterone system. The ADH hormone regulates the blood osmolality by increasing the water absorption in the kidneys, which also assists the excretion of concentrated urine in animals suffering from heat stress [59]. Studies on the responses of neuro-endocrine mechanisms in Boran cattle breed are scarce in Ethiopia and this requires further investigation.

3.1.6. Blood biochemical response.

Blood biochemical constituents and enzymes are fundamental biomarker for climate stress adaptation in tropical livestock. There are several biochemical and enzymes reported to involve in adaptation of livestock to climate stress. One of these events is an increasing trend of total blood haemoglobin (Hb) with an increase in environmental temperature which could be due severe dehydration [60, 61]. Plasma haptoglobin is also one of the most commonly used acute phase proteins to assess the health and inflammatory response of animals [62, 63] reported a significantly higher production of haptoglobin in the blood plasma of Holstein-Frisian dairy cows exposed to high heat load. In several experiments, significantly increased levels of packed cell volume (PCV) were observed in various livestock species suffering from heat stress ([33]. On the other hand, a decreased concentration of plasma protein [50, 64] and cholesterol [50] were recorded in livestock exposed to elevated ambient temperatures. Further, there are reports which also established an increased concentration of free fatty acid in livestock exposed to heat stress [61].

Antioxidant enzymes such as superoxide dismutase (SOD) and glutathione peroxidase (GPx) are synthesized in the body and provide protection from reactive oxygen species generated during heat stress [65]. These antioxidants scavenge both intracellular and extracellular super oxides and inhibit lipid peroxidation of plasma membrane during the challenges of heat stress [61, 66] reported a significantly higher level of plasma malondialdehyde, SOD and GPx activities in Surti buffaloes during hot humid periods and hot dry periods indicating an increased free radical production during periods of heat stress. In addition to this, plasma antioxidant levels in the hot dry period were significantly higher than in the hot humid period indicating more stressful condition may lead to the elevated synthesis of free radicals [61]. There are also reports establishing significantly higher total antioxidant status (TAS) values in a hot dry season in ruminant animals [61]. All these findings establish the significance of blood biochemical responses to be one of the primary means used by animals to cope with adverse environmental conditions.

3.1.7. Metabolic responses.

Metabolic adaptation is another important means through which animals tackle challenges of heat stress, essentially by reducing the metabolic heat production [67]. Thyroid hormones play an important role in regulating the thermogenesis and are also identified as an indicator for assessing the thermo-tolerance of the farm animals [68]. Thyroid hormones, namely triiodothyronine (T₃) and thyroxine (T₄), play a vital role in metabolic adaptation and growth performance of animals [62]. During heat stress, serum and plasma concentrations of T₃ and T₄ reduce and are likely to be due to the direct effect of heat stress on the hypothalamo-pituitary and thyroid axis to decrease the production of thyrotropin-releasing hormone, which will limit basal metabolism [67]. Reduced concentrations of circulating T₃ and T₄, were an indicative of an attempt to reduce metabolic rate and thus metabolic heat production in heifers [69].

During periods of high ambient temperatures, some metabolic enzymes increase their activity, the levels of activity of these enzymes in plasma can be informative of how various organs are responding and adapting to heat load and such enzymes play a vital role in the diagnosis of welfare of animals [65]. Acid phosphatase (AP) and alkaline phosphatase (ALP) are two major enzymes associated with the metabolic activities in animals. The levels of these enzymes are generally low in heat stressed animals, which could be attributed to a metabolic shift in the animals [65]. Likewise [50], reported a decrease in ALP during summer in ruminants, which they attributed to the dysfunction of the liver during heat stress exposure. Aspartate aminotransferase (AST) and alanine aminotransferase are two important metabolic enzymes that increase during heat stress exposure in sheep [70] and goats [65]. These authors concluded that such increase in the activity of these enzymes is due to the higher adaptive capability of the animals to cope with heat stress [71].

Another important metabolic regulator is non-esterified fatty acids (NEFA) in plasma and serum [62]. Low NEFA concentrations are mostly reported in heat stressed dairy cows. It is thought that this is an attempt to increase glucose utilization which will result in lower metabolic heat production [72]. However [73], reported an increase in NEFA production of dairy cows during summer compared with winter, which they attributed to an attempt by the animals to maintain energy balance. In summary, at least in livestock, haptoglobin, NEFA, thyroid hormones (T₃ and T₄) and liver enzymes are considered to be reliable indicators of metabolic adaptation to high heat load [62].

3.1.8. Cellular and molecular responses.

The cellular level of adaptation is one of the acute systemic responses to heat stress and it plays a significant role in imparting thermo-tolerance to animals. Gene networks within and across the cells respond to a higher temperature through both intra-and extracellular signals that result in cellular adaptation. Cattles evolved in hot climates had acquired different thermo-tolerant genes when exposed to a higher temperature [2]. Heat stress was found to alter several molecular functions such as DNA synthesis, replication and repair, cellular division and nuclear enzymes and DNA polymerases functions [74, 75]. It also affects a complex array of cellular and molecular responses in livestock [76]. With the development of molecular biotechnologies, new opportunities are available to characterize gene expression and identify key cellular responses to heat stress [25]. For example, there are changes in the expression patterns of certain genes that are fundamental for thermo-tolerance at the cellular level in animals [65]. Such genes having a cellular adaptation function in animals are considered potential biomarkers for understanding stress adaptation mechanisms [30]. The classical heat shock protein (HSP) genes, apoptotic genes and other cytokines and toll-like receptors are considered to be up regulated on exposure to heat stress. Several reports established the role of HSP70 during heat stress exposure in ruminant livestock and they identified this to be ideal molecular marker for quantifying heat stress response [45, 65]. Apart from this, several other genes such as Superoxidedismutase (SOD), nitric oxide synthase (NOS), thyroid hormone receptor (THR) and prolactin receptor (PRLR) genes were found to be associated with thermo-tolerance in ruminant livestock [30].

Furthermore [45], a higher expression of HSP70 messenger RNA (mRNA) in the adrenal gland of the multiple stressor groups, which could be attributed to the adaptive mechanism of livestock to counter both the heat stress and nutritional stress. The significantly higher expression of adrenal HSP70 in the multiple stressed animals as compared with animals subjected only to heat stress could be attributed to additional nutritional stress in the multiple stresses group. The higher HSP70 expression in the adrenal gland could also be attributed to the hyperactivity of adrenal cortex to synthesize more cortisol as evident from this study [45]. Similarly, the plasma HSP70 and expression pattern of peripheral blood mononuclear cell HSP70 also showed similar trends of significantly higher value in multiple stressor group animals as compared with control and individual (heat/nutritional) stress groups [45]. Studies have also indicated that HSP70s and their associated cochaperones participate in numerous processes essential to cell survival under stressful conditions. They assist in protein folding and translocation across membranes, assembly and disassembly of protein complexes, presentation of substrates for degradation and suppression of protein aggregation [76].

3.3.1. Summary statistics for quantitative traits.

The weighted descriptive statistics and abbreviations for the traits considered in the present study for Boran cattle breed are shown in Tables 1–3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Mean phenotypic traits of reproduction in Boran cattle breed.

https://doi.org/10.1371/journal.pone.0244836.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Mean phenotypic traits of growth and carcass characteristics in Boran cattle breed.

https://doi.org/10.1371/journal.pone.0244836.t002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Mean phenotypic traits of milk yield and composition in Boran cattle breed.

https://doi.org/10.1371/journal.pone.0244836.t003

The number of published articles reported were higher for CI, NSC, DO, AFS, AFC, BW, WW, ADG2, LYD and DMY (Table 2) traits, suggesting their importance for the Boran cattle breeding program. The number of studies as well as the records per articles were comparatively lower for some growth and milk production traits than reproductive traits as these traits are mainly age and sex-specific and difficult to measure periodically. The lowest coefficients of variations were estimated for milk composition traits indicating that the number of studies and records on the specific breed is limited and requires more investigation. In addition, the phenotypic variations for these specific traits are biologically limited [16]. On the other hand, the higher coefficient of variations was observed for WWT (49.6%), Omega-FA (36.8%), ADG (34.9%) and CI (30.1%), showing that there is larger phenotypic variation in these traits than in others (1–3 Tables).

3.3.2. Reproduction and production performances of Boran cattle and other cattle breeds.

3.3.2.1. Reproductive characteristics. Comparisons of the reproductive performance of Ethiopian Boran with other indigenous cattle breeds indicated that Boran cattle have better reproductive performance (Table 4). The breed has shorter female age at first matting (FAFM), age at first calving (AFC), calving interval (CI), days open (DO) while longer reproductive life time for male (RLTM) and reproductive life time for female (RLTF) as well as comparable male age of fertility for mating (MAFM). These traits vary from low to high in heritability and are more challenging to record [83].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Average reproduction and production performances of Boran cattle as compared to other Ethiopian cattle breeds.

https://doi.org/10.1371/journal.pone.0244836.t004

In Ethiopia, breeding mainly relies on natural service and, therefore, acceptable bull fertility is also critical [5]. Factors that determine a bull’s fertility and performance include plane of nutrition [5], structural soundness, capability of the reproductive organs, quality of the semen, libido level and servicing capacity [88]. Under pastoralist management conditions, age at first calving of Borana cattle is about four years of age [89]. At Abernossa ranch in Ethiopia, weight and age at puberty in heifers were found to be about 155 kg and 22 months, respectively. Calving rate under a single-sire mating system was also improved to above 80% [5], compared to about 45% under pastoral management conditions [89]. These results would therefore indicate the improvement that could be achieved through proper selection scheme and better management. On the other hand, Ethiopian Boran had longer calving interval, lower breeding efficiency, delayed age at first service and age at first calving and longer days’ open compared with their Friesian crosses. These differences in reproduction performance between Ethiopian Boran and the crosses are comparable with those obtained for local cattle and their crosses in Ethiopia [84, 90]. Thus, the more advanced age at first calving obtained in the Boran compared with their exotic crosses indicated the potential that could be exploited by merely improving management. However, there was no significant difference in number of services per conception among the genetic groups [5]. This could due the fact that nutrition [84] and inseminator effects [29] are more important contributors to the variation in number of services per conception than genotype.

3.3.2.2. Growth and carcass characteristics. Table 4 indicates comparative summary of birth weight (BW) and mature weight (MW) of six Ethiopian cattle breeds. Boran cattle had better performance in both growth traits (Table 4). Additionally, performance of Boran has been substantially improved through improvement in management and selection under intensive system. For example, the improved Boran had birth weight of 30kg in Kenya [4] and weaning weight of 158 kg at Abernossa ranch in Ethiopia [84]. This variation indicates the potential that can be exploited by within breed selection and improvement in management. Growth performance of Boran and their crosses as beef animal has been evaluated in different parts of the world. For example, the value of the Boran for beef production in the tropics is clearly shown in crossbreeding studies in Kenya (www.borankenya.org). In this experiment the F1 Angus/Boran steer weighed 426 kg when sold at 13 months with a daily gain of 1.36 kg. In Australia (www.boransaustralia.com), it is generally believed that using Boran bulls for crossbreeding is the quickest way of improving the commercial potential of beef herds, because one of the most important attributes of the Boran is its ability to transmit hybrid vigour to the traditional beef breeds of Australia (including Brahman and its’ crosses). The higher growth performance observed for the crossbred calves in comparison with the Ethiopian Boran cattle could be due to the effects of breed and heterosis on the growth performance of crossbred cattle [5]. The average adult live weight differs depending on the status of improvement, level of management and production system. For example, mature bulls of the Improved Boran in Kenya were reported to weigh 550–850 kg and those for cows were 400–550 kg [4]. The birth weight under range condition of Ethiopia is indicated to be 18 kg, while it is nearly 25 kg in research stations of Ethiopia and in the commercial ranches of Kenya [2]. Due to this good beef performance, it is indicated that Boran could be used to improve small East African Zebu [5].

The Boran had higher carcass weight when compared with Angoni and Barotse breeds of the same age group, while Boran and Angoni were comparable with regard to dressing percentage [77]. Carcass characteristics differ between breeds and are influenced by the plane of nutrition and production system [87]. Selection for these traits is greatly influenced by the market demand. In the Ethiopian context, export markets demand lean meat whereas when the target is local market, fattened cattle are required [5, 87]. Therefore, the breeding, feeding and other management conditions should be designed in such a way that the requirements of the specific market are met.

It is established that Boran produces high quality beef by utilizing low quality forage. This is substantiated by data from the FAO/UNDP feedlot trial at Lanet in Kenya, where 7625 Boran and crossbreds were fed between 1968–73 (www.borankenya.org). Similarly, the performance of improved Boran as a beef breed was reported higher in South Africa, USA and Australia [4]. In addition, Boran cattle are less affected by mild drought shocks and have fast recovery rate [24]. In Ethiopia, literature report on beef attributes of Ethiopian Boran is scanty. Some of the literatures on growth performance of Ethiopian Boran cattle presented in Table 3 are based on data from crossbreeding studies where Ethiopian Boran was used for dairy production. There is therefore, a need to investigate the beef qualities of the Ethiopian Boran for future uses.

3.3.2.3. Milk yield and quality characteristics. Milk production performance of Ethiopian Boran compares fairly well with other indigenous Ethiopian breeds (Table 4). Boran cattle had better daily milk yield (DMY), lactation milk yield (LMY) and comparable lactation length (LL) among Ethiopian cattle breed under grazing systems. However, the improved Boran in Kenya had much higher performance than the unimproved Ethiopian Boran. In Kenya, improved Boran cows could produce 1130 kg milk over a 36-week period during with calves suckled three times a day producing butter fat percentage of 5.7, 5.8, 5.9 and 6.1, respectively [4]. But under more favourable production environment, Boran cows produced up to 1657kg of milk per 252-day lactation [2]. On the other hand, converting calf growth up to 7 months in to milk intake and adding recorded values of milk off-take for human consumption; lactation milk yield of Boran cows was calculated to be 843 kg under the pastoralist system in Ethiopia [90]. Moreover, crossbreeding of Ethiopian Boran with Holstein Friesian resulted in improved milk production [5]. For example, 50% Holstein Friesian crosses had a fourfold increase over the Ethiopian Boran breed in terms of lactation milk yield, 305-days milk yield, daily milk yield and life time milk yield; they were also milked for 97 more days than Ethiopian Boran.

In general, meta-analysis for performances of Boran cattle compared with other Zebu cattle breeds as beef and/or dairy cattle are described by four categories of traits namely, reproduction, growth, milk production, and carcass yield and quality traits (1–3 Tables). Although their production potential is less compared with exotic and cross breeds, the level of production of Boran cattle is relatively stable during harsh conditions where high producing animals are at risk [1]. Furthermore, during periods of extreme heat stress, water scarcity and reduced pasture availability, they maintain their reproductive potential due to their smaller body size whereas the larger exotic animals may face reproductive impairments which could be attributed to their higher maintenance energy requirements [5, 24].

3.3.2.4. Heritability of production and reproduction traits in Boran cattle. Direct and maternal–heritability estimates of reproductive and production traits with respective 95% confidence intervals and the I² index to test the significance of heterogeneity among studies, for each trait, for Boran cattle are shown in Table 4. Direct heritability estimates for reproduction traits ranged from lower to medium magnitude. Lower heritability estimates were observed for DO (0.040) and CR (0.042), for which the impudence of environmental factors is more evident. Medium heritability was estimated for longevity (0.08), SPC (0.081), CI (0.03) and AFC (0.215), indicating relatively better response to selection for these traits as compared to the other indicated reproductive traits. Usually, indigenous cattle in Ethiopia are raised on extensive production systems, which are often characterized by high temperatures and periods of feed and water scarcity that might affect the reproductive performances of animals [5].

These phenomena are quite severe for poorly adapted exotic and crossbreds [91], but for Boran cattle breed which had better adaption to tropical environment, the observed heritability estimates were from lower to medium (Table 5), implicating selection response for AFC should be faster than other traits for Boran cattle. Similar trend was observed for reproduction traits of Nellore cattle in Brazil [16].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Number of articles to estimate direct and maternal heritability (N_d and N_mat respectively), direct and maternal heritability (h² and h² mat respectively) estimated from meta-analysis using random-effects model, 95% confidence interval for heritability (95% CI and 95%CI _mat respectively) and the I² index to test heterogeneity in each trait (I² and I² mat respectively).

https://doi.org/10.1371/journal.pone.0244836.t005

Direct heritability estimates for the growth traits ranged from medium in BWT (0.25), WW (0.243), and YW (0.265) to higher in ADG (0.46) showing that relatively higher response to selection for ADG trait than other growth traits in Boran cattle breed (Table 5). It shows that ADG, birth weight and mature weight have the highest heritability among the production traits which had good correlation with FCR explaining that accelerated genetic progress could be obtained for these traits through selections in Boran cattle [23].

Similarly, direct heritability estimates for milk production and composition traits ranged from medium to higher magnitude, indicating better response for selection for milk composition traits compared to milk yield traits in Boran cattle. The heritability estimates for growth for Nellore cattle 16] and other Zebu cattle breeds [92] corroborates the current study for Boran cattle.

Heterogeneity among studies were moderate in milk yield traits by I² index (50 to 75%) and higher in reproduction, growth and milk composition traits with I² index (>75%). These results reinforced the importance of accounting this heterogeneity in selecting random effect model in the meta-analysis reviews of the studies (Table 5).

The weight and milk production performances of Boran calves are also affected by the maternal genetic effect from dam. In this study, maternal-heritability estimates for growth traits had a low to medium magnitude (ranging from 0.089 to 0.20), indicating that the genetic progress for these maternal effects is slow in this breed. Similar trend was observed for milk production traits. It is well known that the initial phases of growth traits are influenced by the maternal effect and ignoring these effects may lead to bias in many situations [95]. In general, higher heritability estimates were reported in the studies for milk composition traits followed by growth traits and milk yield traits but lower heritability estimates were reported for reproduction traits in Boran cattle except trait for AFC. Since meta-analysis brings together published parameter estimates provided by studies based on populations at different stages of selection, with different sample sizes and considering distinct effects in the model, it is expected that the true parameter may vary from study to study. In meta-analysis, the estimated confidence intervals of the studied traits were narrower than those obtained from individual published articles. In Boran cattle, data on heritability estimates for carcass traits were limited for this meta-analysis which requires more studies in the future.

On the other hand, heritability for average daily gain and feed conversion ratio were estimated to be medium for other African cattle breeds (Table 6). The summary of ranges of heritability for tropical cattle breeds indicate that they have lower heritability for reproductive traits except for AFC and lower to medium values for production traits. Similar trends were observed in the current meta-analysis for both reproduction and production traits of Boran cattle.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Summary of ranges of heritability values for reproduction and production traits of Tropical cattle.

https://doi.org/10.1371/journal.pone.0244836.t006

3.3.3.5. Genetic correlations between reproduction and production traits in Boran cattle. Genetic-correlation estimates with the respective 95% confidence interval and the I² index to quantify the degree of heterogeneity among the studies were shown in Table 7 for each trait pairs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 7. Number of articles (N_d), genetic correlation among traits(g_r) estimated from meta-analysis using random model, 95% confidence interval (95% CI), and the values of I² index(I²) in each trait.

https://doi.org/10.1371/journal.pone.0244836.t007

Except correlation involving AFCXCI, CDXCI and DOXPR which assumed negative values, all other correlation estimates were positive. Higher correlation estimates were observed between WWXYW, WWXMW, PXSN, PXTS, LLXYD and SNFXTS trait pairs, while medium correlation estimates for WXWW, BWXMW, and FXP and lower correlation estimates for 305YDXLL, 305YDXLD, 305YDXDMY, LYDXDMY, FXSNF and BWXYW for Boran cattle breed.

Similar to heritability estimates, the estimated confidence interval of genetic correlations was narrower than reported in individual published studies. All correlation estimates presented higher heterogeneity among studies by the I² index (I²>75%) which justify the use of random-effects model. In general, a small number of articles reporting genetic correlation estimates were found for purebred Boran cattle. This highlighted the requirement for studies investigating these estimates, which, in general, also presented high standard errors in relation to heritability estimates (Table 7). Estimated genetic correlations among some reproduction and production traits in these studies are in line with summary of genetic correlations of fertility and production traits for tropical zebu cattle in East Africa (Table 8). Except for correlation between AFC and CD, the other correlation coefficient between pairs of reproduction traits were negative which may explain improvement for one trait may negatively affect the performance of the other traits [93].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 8. Genetic correlations (rg) for specific fertility and production traits in tropical cattle breeds.

https://doi.org/10.1371/journal.pone.0244836.t008

Although genetic correlations estimated in the present study were in favour of the common objectives of genetic improvement programs of Boran cattle in relation to climate change adaptation, it is important to re-estimate the genetic parameters in regular intervals, especially if selection of different traits is made simultaneously [5].