Livestock Research for Rural Development 28 (2) 2016 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Phenotypic and genetic parameters of calf growth traits for Malawi Zebu

Wilson Nandolo1,2, Timothy N Gondwe1 and Mcloyd Banda3

1 Lilongwe University of Agriculture and Natural Resources, Department of Animal Science, P.O. Box 219, Lilongwe, Malawi
2 The University of Natural Resources and Life Sciences, Division of Livestock Sciences (NUWI), Gregor-Mendel-Straße 33, 1180 Vienna, Austria
3 Department of Agricultural Research Services, P. O. Box 30779, Lilongwe 3, Malawi
wilsonandolo@gmail.com

Abstract

Malawi Zebu form the back-bone of the Malawian cattle industry. They are used as the base population for dairy and beef production and development. Very little has been done to improve them genetically, and data on their current productivity are scanty. This study was carried out to refresh information on Malawi Zebu calf growth traits (birth weight, 205-day weight, weaning weight and pre-weaning growth rate) as part of their conservation program. Data from 1985 to 2014 from Mbawa Research Station in Mzimba District were used. Single trait best linear unbiased mixed models were used to estimate fixed and random effects for each trait.

 

Estimated least squares means and standard errors were 19.4±0.532, 82.6±2.26, 89.6±3.25 and 0.388±0.00119 kg for birth weight, 205-day weight, weaning weight and pre-weaning growth rate, respectively. All the traits showed decreasing trends over the years and seasonal effects on growth traits support the idea of having breeding seasons. The estimated direct heritability and standard errors of the traits were, 0.33±0.144, 0.47±0.160, 0.61±0.204 and 0.43±0.211 for birth weight, 205-day weight, weaning weight and pre-weaning growth rate, respectively. The moderate to high heritability of the growth traits imply that selection for these traits is likely to be worthwhile. The estimated breeding values of birth weights have been declining while those of the other traits have been constant, implying lack of genetic improvement. The results from this study have refreshed the phenotypic parameter values for the Malawi Zebu and the estimated genetic parameters may be useful for future genetic evaluation of this important genetic resource.

Key words: birth weight, breeding season, heritability, pre-weaning growth rate, weaning weight


Introduction

Malawi has about 1.4 million cattle, 95% of which are the Malawi Zebu, 15,000 pure dairy animals and 50,000 dairy crosses (Department of Animal Health and Livestock Development 2014). Nearly all the crosses are between the Malawi Zebu and dairy breeds that include Holstein Friesian and Jerseys (Nandolo 2015). Essentially, the majority of the milk and almost all the beef produced in Malawi comes from the Malawi Zebu. However, milk production levels in Malawi are low and declining due various reasons, amongst which are poor animal management practices and the low genetic potential (for milk production) of the Malawi Zebu (Banda et al 2012; Tebug et al 2012). The Malawi Government recommends use of crosses for smallholder dairying, based on the premise that the dairy management levels may not be good enough for high producing dairy animals (Department of Animal Health and Livestock Development 2006). For instance, an average producing dairy animal may require between 50 and 100 litres of drinking water daily, while most farmers do not have running water, and it is not easy for them to fetch so much water just for one animal. Feed is compromised in terms of quantity and quality. So, the current understanding is that the crosses are sufficient for the smallholder farmers, although an appropriate level of crossbreeding has not yet been established. A fundamental flow with this approach is that generally, there is little being done to improve the Malawi Zebu which is the dam line for crossing. Farmers who have difficulties accessing semen from the National Artificial Insemination Service or from other commercial players normally tend to use natural service using Malawi Zebu bulls to keep their cows in production. Very little is done to select the best Zebu cows or bulls for cross breeding. Besides, very little is known about the current production levels of the Malawi Zebu, although the success of the crossbreeding programmes is supposed to depend on the worthiness of the Malawi Zebu.

 

Information on the productive and reproductive parameters of the Malawi Zebu is scanty and old (Butterworth and McNitt 1984; Kasowanjete 1979; Munthali 1982). Namwaza et al (1998) updated figures for growth parameters for the Malawi Zebu, but they used data from 1977 to 1981 only for Chitala and Mbawa Research Stations, and for 1987/1988 for Dzalanyama Cattle Ranch. Since then, these figures have not been updated while climate and ecological changes have evolved. Also, there are no heritability estimates for Malawi Zebu cattle that can be used to facilitate genetic improvement. This study aimed at refreshing information on calf growth traits (birth weight, 205-day weight, weaning weight and pre-weaning growth rate) of Malawi Zebu and estimating basic genetic parameters using available data to fill the existing information gap.


Materials and methods

Description of study sites

 

Data were obtained from Mbawa Research Station in Mzimba District, Northern Malawi. The primary purpose of the station is to conserve the Malawi Zebu cattle by identifying and keeping the best Malawi Zebu animals, multiplying them and distributing the breeding stock or steers to farmers. Breeding is controlled, with a breeding season from January to March every year, so that the calves drop from between October and December, coinciding with the onset of rainy season. Animals that are born between October and December are exposed to favorable growth conditions during their formative months, especially in terms of the availability of forage. Weaning is at 6 months between April and June, depending on whether the calf was conceived at the beginning or end of the breeding season. There is no supplementary breeding season, but a few calves are born out of season, implying that breeding is not perfectly controlled. Regardless of being born out of season, they are weaned at about the same time as the ones born in season. The cattle are grazed throughout the year from around 0800 hours to around 1500 hours and there is minimal feed supplementation during lean periods (August-November).

 

Data preparation

 

A total of 1187 records were available, of which 978, 683 and 595 records had weights at birth, weights at weaning and age at weaning, respectively. Records of dams with less than 3 offspring were discarded, resulting into 640 records from 22 sires and 146 dams with number of records ranging from 3-120 per sire and 3-10 per dam. Each sire was used for an average of 2.7 years (maximum 7 years), and there were on average 4±1.52 sires at any given time (minimum 2, maximum 7 and median 4). Pedigree analysis done with Contribution, Inbreeding (F) and Coancestry (CFC) software (Salgozaei et al 2006) showed that only 5 out of the 739 individuals in the pedigree were inbred. Average inbreeding coefficients in the whole pedigree and among inbred individuals were 0.000677 and 0.100, respectively. Numerator relationships (reciprocals plus self-relationships) averaged 0.030, with an average number of discrete generation equivalents of 1.16.

 

Pre-weaning growth rates were calculated by dividing the difference between birth weights and weaning weights by weaning ages. Adjusted 205-day weights were calculated by adding the pre-weaning growth to the sum of birth weights and adjustment factors based on the age of the dam when the calves were born.  Table 1 shows the descriptive statistics of the growth traits before fitting the models. Birth weights averaged around 20 kg. Weaning was done at an average of 205 days (the same as the number of days for weight adjustment, p (t) = 0.899) with weaning weight of 92 kg and pre-weaning daily growth rate of 375 g.

Table 1: Descriptive statistics for the growth traits

Variable

n

Mean

Range

Standard
deviation

Coefficient
of variation

Birth weight (kg)

623

20.0

10-39

4.96

24.8

205-day weight (kg)

297

99.8

40.9-169

22.9

22.9

Weaning age (days)

357

205

52-302

52.3

25.5

Weaning weight (kg)

414

92.1

45-174

20.3

22.0

Growth rate (kg/day)

297

0.375

0.0922- 0.694

0.111

29.6

Data analysis

 

Two types of best linear unbiased prediction mixed models were used to estimate random and fixed effects for each trait: an animal model (Model 1) and a maternal effects model (Model 2): 

Model 1

Model 2

Where y was a vector of observations for calf growth traits; X was a known incidence matrix relating observations in y to known classes of fixed effects; b was a vector of unknown fixed effects; Z and Z1 were incidence matrices relating  observations to direct genetic effects, u was a vector of unknown direct genetic effects; Z2 was the incidence matrix relating observations to additive maternal genetic effects; m was a vector of unknown maternal effects and e was a vector of unknown random residual effects.

 

Fixed effects included year (1986 to 2014) sex and season according to the conventional Malawian season categorization. The seasons were cool dry (May to August), hot dry (September to November) and hot rainy (December to April). Covariates included dam age at the time of the birth of the calf (for birth weight), birth weight (for weaning weight and growth rates) and weaning age (for weaning weight and growth rate). The models were run in ASREML Version 4.1 (Gilmour et al 2015). Model 1 had lower Akaike Information Criterion (AIC) values in all the traits (Table 2). This is most likely because Model 2 was over-fitted considering the small size of the dataset.  Subsequently, estimation of fixed effects was based on Model 1. Genetic trends were tested using regression analysis.

Table 2: Goodness of fit of the test models

Model

Trait

BWT

ADJWT

WWT

GR

1

1181

1738

1274

-463

2

1182

1740

1274

-461

AIC-diff

1.08

1.82

0

2

BWT = birth weight (kg), ADJWT = 205-day weight (kg), WWT = weaning weight (kg),
GR = pre-weaning growth rate (kg/day); AIC-diff = (Model 2 AIC) – (Model 1 AIC)


Results and discussion

Estimates of fixed effects

 

Estimates of the current values of the four growth traits are given in Table 3. Birth weight was slightly above the 17.7 to18.9 kg range (females and males) estimated by Namwaza et al (1998) but within the range 17.7±2.9 to 22.0±2.5 (females and males) estimated by Butterworth and McNitt (1984). Weaning weight was also within the 83.9 to 90.2 kg, estimated by Namwaza et al (1998), but lower than those of other zebu cattle elsewhere (Diop et al 1999; Lóbo and Filho 2000). Growth rate was much lower than the 0.450 kg/day reported by Butterworth and McNitt (1984) when Malawi Zebu is put to better management. 

Table 3: Estimated least squares means of the traits

Variable

Mean

Standard error

Birth weight (kg)

19.4

0.532

205-day weight (kg)

82.6

2.26

Weaning weight (kg)

89.6

3.25

Pre-weaning growth rate (kg/day)

0.388

0.00119

Sex did not affect any of the growth traits, contrary to what Namwaza et al (1998) found. This could be because the researchers used a fixed effects model that might have overestimated the effects, making some of them apparently significant. Bayou et al (2015) found that sex influenced growth traits up to puberty in Sheko cattle of Ethiopia. Generally, sex is known to account for a small proportion of the variation in growth traits in other non-zebu cattle breeds. Manzi et al (2012) reported that sex accounted for 6% of the variation in Brown Swiss and Jersey calves in Rwanda while Krupa et al (2005) reported 7.8% in European beef breeds.

 

Phenotype trends

 

All the traits were affected by year-season effects (Figure 1). Birth weights declined from 1986 to 2013, but the birth weights of calves born in the hot dry season declined more slowly than those of calves born in other seasons. Birth weights for calves born in the hot rainy season were consistently higher than those of calves born in the other seasons.  Adjusted (205-day) weights and pre-weaning growth rates also showed declining trends, but with more stable values for calves born in the hot rainy season compared to those born in the hot dry seasons, and sharply declining values for calves born in the cool dry season. On the other hand, calves which were born in the cool dry season consistently had higher weaning weights than those born in the hot rainy and cool dry seasons.

Figure 1: Trends in (A) birth weight (kg), (B) 205-day weight (kg), (C) weaning weight (kg) and
(D) pre-weaning growth rate (g) by season. Blue and round= Cool season (May to August)
red and square = Hot season (September to November) and green and diamond = Rainy
season (December to April). The trend lines (dotted lines) were fitted with period 2.

These results confirm the existence of seasonal patterns in calf growth in this area. Consistent with management expectation, calves born around the expected calving time did have higher birth weights, growth rates and adjusted weights (Figure 2).

Figure 2: Overall trends in (A) birth weight (kg), (B) 205-day weight (kg), (C) weaning weight (kg) and
(D) pre-weaning growth rate (kg) within a year.

Higher weaning weights for calves born in the cool dry season may not be because of the season per se but probably because of higher weaning ages, which make it more likely for the calves to survive better in stressful periods of the year. These calves are weaned in June of the following year 8 to 12 months later, together with the calves born in the other reasons. This is consistent with the 8-10 months recommended by some researchers as an appropriate, natural weaning age for calves in extensive cow-calf systems for Bos indicus cattle (Reinhardt and Reinhardt 1981).  

 

The higher weaning weights may also be explained in terms of feed availability. Calves born in the cool dry season benefit from the good condition of their dams due to the availability of higher quality dry matter during the period May to August. The feed shortage in the short dry season (September-November) probably has little effect on the calves, which then take advantage of the pasture growth during the rainy season in December, ending up with higher weaning weights. On the other hand, it is highly probable that the calves that are born in the hot season suffer from undernutrition too early in their lives, so that moderately high growth rates at the beginning of and during the rainy season (November-April) do not cover up for this loss. Similarly, calves born in the rainy season may have the expected advantages of higher birth weights and growth rates, but may most likely face challenges that slow down their growth, leading to low weaning weights. The most likely challenges for this calf crop could be neonatal diseases typical of the wet hot season, compounded by muddy pens during this season.

 

Heritability, phenotypic correlations and genetic correlations

 

Heritability estimates for the traits were moderate to high (Table 4). Weaning weight had the highest heritability while birth weight had the lowest. Maternal heritabilities were low for birth weight, 205-day weight and nearly zero for pre-weaning growth rate. However, maternal heritability for weaning weight was high, implying that weaning weights depend very much on maternal effects. Supriyantono et al (2014) reported maternal heritability of 0.21 in Japanese Black cattle and Diop et al (1999) also reported maternal heritability of 0.21 in Gobra cattle.

 

The heritability for birth weight was within the commonly reported medium range of 25-45% (Lóbo and Filho 2003; Supriyantono et al 2014). Plasse et al (2002) found direct heritability equal to 0.33 for birth weight and 0.14 for 205-day weight for Brahman cattle. Similarly, Plasse et al (2004) found 205-day weights heritability values of 0.13 for Bos indicus cattle upgraded to Brahman, while Szabó and Bene (2013) reported heritability values ranging from 0.18 to 0.61 for some European cattle breeds. Abera et al (2011) reported high heritability of 0.53 for weaning weight in Horro cattle (Sanga-Zebu type of cattle). Heritabilities ranging from 0.31 to 0.84 have been reported in composite tropical beef breeds of Australia (Burrow, 2001). The heritability estimates for pre-weaning growth rate was similar to the one reported by Afroz et al (2011) in the Rec Chittagong Cattle (Bos indicus). Overall, the moderate to high heritability estimates suggest that implementing a systematic Malawi Zebu genetic improvement programme through selection is viable.

Table 4: Heritability estimates

Model

Parameter

Trait

BWT

ADJWT

WWT

GR

1

VA

4.06

159

193

0.0031

VE

8.31

179

124

0.00417

VT

12.4

338

317

0.00727

hd2 (se)

0.328 (0.144)

0.470 (0.160)

0.609 (0.204)

0.426 (0.211)

2

VD

0.748

11.2

45.7

3.78e-09

VA

3.07

151

141

3.10e-03

VE

8.38

174

123

4.17e-03

VT

12.2

337

309

0.00727

hm2 (se)

0.0613 (0.0688)

0.0333 (0.0814)

0.148 (0.113)

0.000 (0.000)

hd2 (se)

0.252 (0.153)

0.448 (0.175)

0.456 (0.235)

0.426 (0.211)

BWT = birth weight (kg), ADJWT = 205-day weight (kg), WWT = weaning weight (kg), GR = pre-weaning growth rate (kg/day); V A = direct (animal) additive variance component, V D = maternal additive variance component; V E = residual variance component, V T = total variance; hd2 = direct heritability; hm2 = maternal heritability; se = standard error

A summary of genetic and phenotypic correlations is shown in Table 5. Generally, the phenotypic correlations between birth weight and the other traits were low and negative, while the phenotypic correlations among weaning weight, adjusted weight and growth rate were very high. This implies that there were a lot of environmental factors influencing the growth traits such that calves with high birth weights did not appear to have a big advantage over the others in terms of their later growth, contrary to expectation in a cow-calf system in which calf management is good. This is confirmed by the low to zero genetic correlations between birth weight and the other traits. This implies that there is a lot that needs to be done to improve the environment.

Table 5: Phenotypic and genetic correlations between the growth traits

Variable

BWT

ADJWT

WWT

GR

BWT

0.328

-0.0430

0.187**

-0.244**

ADJWT

0.140***

0.470

0.577**

0.946**

WWT

0.0933*

0.717***

0.609

0.564**

GR

0.0339

0.831***

0.774***

0.426

Phenotypic correlations are above the diagonal, heritabilities in the diagonal and genetic correlations below the diagonal. BWT = birth weight (kg), ADJWT = 205-day weight (kg), WWT = weaning weight (kg), GR = pre-weaning growth rate (kg/day).
* Significant at p = 0.05; **Significant at p = 0.01; **Significant at p = 0.001

Genetic trends

 

The estimated breeding values of birth weight declined slightly (p = 0.00162) with a mean close to zero by the year 2013, compared to the late 1980s and early 1990s when the mean was slightly above zero (Figure 3). Weaning weight, 205-day weight and growth rate did not change (p-values 0.587, 0.312 and 0.425, respectively).

Figure 3: Genetic trends of birth weight (blue), 205-day weight (red), weaning weight (green) and
pre-weaning growth rate (black). The graphs were plotted using scaled EBVs.

This is worrisome, and it implies that contrary to expectations, the animals are not being improved. The genetic observations confirms observations from characterization study in Mzimba that showed declining trend of productive traits for Malawi Zebu (Nandolo et al 2015). This is a wake-up call to put in place measures to ensure that there is meaningful genetic improvement.


Conclusion


References

Abera H, Abegaz S and Mekasha Y 2011 Genetic parameter estimates of pre-weaning weight of Horro (Zebu) and their crosses with Holstein Friesian and Jersey cattle breeds in Ethiopia. International Journal of Livestock Production. Retrieved October 15, 2015, from http://www.academicjournals.org/journal/IJLP/article-full-text-pdf/D8F54671700

Afroz M A, Hoque M A and Bhuiyan A K F H 2011 Estimation of heritability for growth traits of Red Chittagong cattle in a nucleus herd. The Bangladesh Veterinarian 28(1):39–46

Banda L J, Kamwanja L A, Chagunda M G, Ashworth C J and Roberts D J 2012 Status of dairy cow management and fertility in smallholder farms in Malawi. Tropical Animal Health and Production 44(4):715–27.

Bayou E, Haile A, Gizaw S and Mekasha Y 2015 Evaluation of non-genetic factors affecting calf growth, reproductive performance and milk yield of traditionally managed Sheko cattle in southwest Ethiopia. SpringerPlus 4(568). http://doi.org/10.1186/s40064-015-1340-9

Burrow H 2001 Variances and covariances between productive and adaptive traits and temperament in a composite breed of tropical beef cattle. Livestock Production Science, 70(3): 213–233. http://doi.org/10.1016/S0301-6226 (01)00178-6

Butterworth M H and McNitt J I 1984 The Malawi Zebu. World Animal Review. FAO, Rome.

Department of Animal Health and Livestock Development 2006 Policy Document on Livestock in Malawi. Ministry of Agriculture and Food Security, Malawi Government. Lilongwe.

Department of Animal Health and Livestock Development 2014 Agricultural Production Estimates. Ministry of Agriculture and Food Security, Malawi Government. Lilongwe.

Diop M, Dodenhoff J and Van Vleck L D 1999 Estimates of direct, maternal and grand maternal genetic effects for growth traits in Gobra cattle. Genetics and Molecular Biology, 22(3):363–367. Retrieved October 15, 2015, from http://doi.org/10.1590/S1415-47571999000300013

Gilmour A R, Gogel B J, Cullis B R, Welham S J and Thompson R 2015 ASReml User Guide Release 4.1 Functional Specification, VSN International Ltd, Hemel Hempstead, HP1 1ES, UK www.vsni.co.uk

Kasowanjete M B B 1979 Evaluation of the three breeding systems: within - Malawi Zebu (MZ) breeding, upgrading MZ to Charolais (CH) breed, and rotational crossing between CH and MZ breeds for maximum and efficient beef production under Malawi conditions. Royal Veterin­ary and Agricultural University, Copenhagen.

Krupa E, Oravcova M, Polak P, Huba J and Krupova Z 2005 Factors affecting growth traits of beef cattle breeds raised in Slovakia. Czech Journal of Animal Science 1:14-21

Lóbo R N B and Filho R M 2000 Genetic parameters for growth traits of zebu cattle in the semi-arid region of Brazil. Cięncia Animal. Retrieved October 15, 2015, from http://www.uece.br/cienciaanimal/dmdocuments/Artigo1.2000.1.pdf

Lóbo R N B and Filho R M 2003 Estimates of genetic parameters for weights and maturing traits derived by growth functions in Nellore cattle. Cięncia Animal, 13(1):7-15. Retrieved October 15, 2015, from http://www.alice.cnptia.embrapa.br/alice/bitstream/doc/530056/1/APIEstimatesofgenetic.pdf

Manzi M, Junga J O, Ebong C and Mosi R O 2012: Factors affecting pre and post-weaning growth of six cattle breed groups at Songa Research station in Rwanda. Livestock Research for Rural Development. Volume 24, Article #68. Retrieved December 3, 2015, from http://www.lrrd.org/lrrd24/4/manz24068.htm

Munthali B C 1982 Studies on conception rates, calving rates and calf mortality rates in beef and dairy herds at Bunda College farm. Bunda College of Malawi, University of Malawi, Lilongwe.

Namwaza A G, Wollny C B A, Kamwanja L A and Phoya R K D 1998 Performance of Malawi-Zebu Cattle on-station in Malawi. World Congress on Genetics Applied to Livestock Production. Retrieved October 15, 2015, from http://www.ilri.org/InfoServ/Webpub/fulldocs/AnGenResCD/docs/MalawiZebuPerform/MalawiZebuPerform.htm

Nandolo W 2015 The performance of Jersey cattle at Naming’omba Tea and Coffee Estate in Malawi. Livestock Research for Rural Development. Volume 27, Article #14. Retrieved December 19, 2015, from http://www.lrrd.org/lrrd27/1/nand27014.htm

Nandolo W, Gondwe TN and Banda LJ 2015 Characterisation of breeding systems for Malawi Zebu cattle in Mzimba district, Northern Malawi. Malawi Journal of Agriculture, Natural Resources and Development Studies, 1 (1): 14 – 20.

Plasse D, Arango J, Fossi H, Camaripano L, Llamozas G, Pierre A and Romero R 2004 Genetic and non-genetic trends for calf weights in a Bos indicus herd upgraded to pedigree Brahman. Livestock Research for Rural Development. Vol. 16, Art. #46. Retrieved October 15, 2015, from http://www.lrrd.org/lrrd16/7/plas16046.htm 

Plasse D, Verde O, Fossi H, Romero R. Hoogesteijn R, Bastidas P and Bastardo J 2002 (Co)variance components, genetic parameters and annual trends for calf weights in a pedigree Brahman herd under selection for three decades. Journal of Animal Breeding and Genetics, 119(3), 141–153. http://doi.org/10.1046/j.1439-0388.2002.00321.x

Reinhardt V and Reinhardt A 1981 Natural sucking performnce and age of waening in zebu cattle (Bos indicus). Journal of Agriculture Science 96:309-312

Sargolzaei M, Iwaisaki H and Colleau J J 2006 CFC: A tool for monitoring genetic diversity. Proceedings of the 8th World Congress on Genetics and Applied Livestock Production, CD-ROM Communication 27-28. Belo Horizonte, Brazil, Aug. 13-18, 2006.

Supriyantono A, Tomiyama M, Suzuki K 2014 Estimation of direct and maternal genetic effect on weaning weight and average daily gain to wean in Japanese Black cattle. 10th World Congress of Genetics Applied to Animal Production. Retrieved October 15, 2015, from https://www.asas.org/docs/default-source/wcgalp-posters/597_paper_9249_manuscript_591_0.pdf?sfvrsn=2

Szabó F and Bene S 2013 Population genetic evaluation for weaning weight of different beef cattle breeds. Journal of Central European Agriculture, 14(3).

Tebug S F, Kasulo V, Chikagwa-Malunga S, Wiedemann S, Roberts D J, and Chagunda M G G 2012 Smallholder dairy production in Northern Malawi: production practices and constraints. Tropical Animal Health and Production, 44(1):55–62.


Received 13 January 2016; Accepted 24 January 2016; Published 1 February 2016

Go to top