Servicios Personalizados
Revista
Articulo
Indicadores
- Citado por SciELO
- Accesos
Links relacionados
- Similares en SciELO
Compartir
Revista mexicana de ciencias pecuarias
versión On-line ISSN 2448-6698versión impresa ISSN 2007-1124
Rev. mex. de cienc. pecuarias vol.12 no.1 Mérida ene./mar. 2021 Epub 20-Sep-2021
https://doi.org/10.22319/rmcp.v12i1.5347
Articles
Genetic selection aimed to reduce methane emissions and its effect on milk components
a Universidad Nacional Autónoma de México. Facultad de Medicina Veterinaria y Zootecnia. Av. Universidad 300, 04510, Ciudad de México. México.
b The University of Nottingham. School of Biosciences, Sutton Bonington Campus. Loughborough LE 12 5RD, UK.
c Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP). Centro Nacional de Investigación Disciplinaria en Fisiología y Mejoramiento Animal. Querétaro, México.
d INIFAP. Centro de Investigación Regional Golfo-Centro. Campo Experimental La Posta. Veracruz, México.
e INIFAP. Centro de Investigación Regional Norte-Centro. CE La Campana. Chihuahua, México.
f Red de Investigación e Innovación Tecnológica para la Ganadería Bovina Tropical (REDGATRO).
This study aimed to estimate the response to selection through different selection indices between methane production and milk production and its components in specialized tropical, dual-purpose, and family dairy systems. Methane emissions were sampled during milking using the Guardian-NG gas monitor; milk samples were collected individually during methane sampling. DNA was extracted from the hair follicles of all the animals included in this study. The variance and covariance components were estimated using the mixed model methodology. Due to the incomplete genealogical information, molecular markers were used to build the genomic relationship matrix (Matrix G). The estimated heritability for methane emissions during milking was 0.18 and 0.32 for the univariate and bivariate analysis, respectively. The genetic correlation between the milk fat and protein percentages and methane emissions during milking was negative, -0.09 and -0.18, respectively. The response to selection, estimated through selection indices, demonstrated that it is feasible to reduce methane emissions up to 0.021 mg/L during milking in five generations without detriment to milk components.
Key words Methane; Milk; Heritability; Genetic correlation
El objetivo de este trabajo fue estimar la respuesta a la selección a través de diferentes índices de selección entre producción de metano y producción y componentes de la leche en los sistemas de lechería tropical especializada, doble propósito y lechería familiar. El muestreo de las emisiones de metano se realizó durante la ordeña mediante el equipo Guardian-NG. Se tomaron muestras de leche de manera individual durante el muestreo de metano. La extracción de ADN se realizó de folículos pilosos de todos los animales incluidos en el estudio. La estimación de los componentes de varianza y covarianza se realizó mediante la metodología de modelos mixtos. Se utilizaron los marcadores moleculares para construir la matriz de relaciones genómicas (Matriz G), debido a que no se contaba con información genealógica completa. La heredabilidad estimada para las emisiones de metano durante la ordeña fue de 0.18 y 0.32 para los análisis univariados y bivariados, respectivamente. Los resultados de las correlaciones genéticas entre porcentaje de grasa y proteína en leche con las emisiones de metano durante la ordeña fueron negativas, -0.09 y -0.18 respectivamente. La respuesta a la selección estimada mediante los índices de selección demostró que es factible obtener reducciones de hasta 0.021 mg/l de emisiones de metano durante la ordeña en cinco generaciones; lo anterior sin detrimento en los componentes de la leche.
Palabras clave Metano; Leche; Heredabilidad; Correlación genética
Introduction
In recent years, the Intergovernmental Panel on Climate Change (IPCC)1 and the Food and Agriculture Organization of the United Nations (FAO)2 declared that the agricultural sector is the principal source of short-lived greenhouse gases (GHG), such as methane (CH4) and nitrous oxide (N2O).
Some strategies to mitigate methane emissions from dairy cattle include reducing the herd, changing bovine diet, using supplements, immunization against methanogenic archaea, and selecting animals with lower CH4 production3.
The selection of low methane-producing animals requires knowledge about the genetic correlations between methane production and other characteristics of productive and economic importance4.
A selection index is a methodology that maximizes breeding for a specific trait5. Selection indices have been widely used to estimate the reproduction value of dairy cattle for individual and combined characteristics for selection purposes6.
In cattle and sheep, the variation of CH4 emission has been demonstrated between individuals fed the same diet7. De Haas et al8 mentioned the possibility of selecting cows with low CH4 emissions since genetic variation suggests that the reductions would be 11-26 % in 10 yr and could be even higher in a genomic selection program. However, the available information about the opportunities to mitigate enteric CH4 through genetic improvement is scarce. Still, the genetic selection of animals with low methane emissions could affect economically important production traits.
This study aimed to estimate the response to selection through different selection indices between methane production and milk production and components in three dairy production systems in Mexico.
Material and methods
This study was carried out in three dual-purpose (DP) production units (PUs), two specialized tropical dairy (STD) PUs, and four family dairy (FD) systems (Table 1). Milk components and methane emissions were measured in 274 cows (98, 74, and 102 in the DP, STD, and FD systems, respectively).
Farm | System | n | Localization | Breeds |
---|---|---|---|---|
La Posta | DP | 33 | Veracruz | HOZ and BSZ |
El Zapato | DP | 16 | Veracruz | HOZ |
La Doña | DP | 49 | Puebla | HOZ, BSZ, and SMZ |
Santa Elena | STD | 37 | Puebla | HO, BS, and HOBS |
Aguacatal | STD | 37 | Puebla | HO, BS, and HOBS |
Farm 5 | FD | 16 | Jalisco | HO |
Farm 6 | FD | 32 | Jalisco | HO |
Farm 7 | FD | 24 | Jalisco | HO |
Farm 8 | FD | 30 | Jalisco | HO |
DP= dual-purpose; STD= specialized tropical dairy; FD= family dairy.
HOZ= Holstein x Zebu, BSZ= Brown Swiss x Zebu, SMZ= Simmental x Zebu, HO= Holstein, BS= Brown Swiss, HOBS= Holstein x Brown Swiss.
Two of the three DP PUs are located in the Medellín de Bravo municipality, Veracruz, and have a tropical savanna climate, Aw(o), and an altitude of 12 m asl9. The annual mean temperature and precipitation are 25 °C and 1,460 mm9. The third DP PU and the STD PUs are located in the Hueytamalco municipality, Puebla, at an altitude of 240 m, with a tropical wet climate (Af(c)), mean annual temperature of 23 ºC, and mean annual precipitation ranging from 2,200 to 2,500 mm9.
The four FD PUs are located in the Tepatitlán municipality, Jalisco, at an altitude of 1,927 m. This location has a humid subtropical climate ((A)C(w1) (e)g) with an annual mean temperature and precipitation of 18 ºC and 715 mm9.
The DP systems mainly use cross-bred Bos taurus taurus and Bos taurus indicus. The most common Bos taurus indicus breeds are Brahman, Gyr, and Sardo Negro; as for Bos taurus taurus, the most common breeds are Holstein, Brown Swiss, and Simmental10.
One of the variants of tropical dairy systems is the STD. This system is characterized by using pure breeds, such as Holstein and Brown Swiss. Overall, STD management is similar to DP systems except for calf rearing, which is artificial, and milking, which is carried out without calf support10.
FDs are characterized by small production units that fluctuate from 3 to 30 cows. The production units are conditioned to small areas and adjacent to the housing units, called “backyard.” FDs can be intensive or semi-intensive according to the conditions of the cultivation field. Holstein is the most common breed. The technological level is considered scarce because producers do not carry out adequate feeding, reproductive, preventive, or breeding practices. This system lacks production records and has rudimentary facilities; manual milking is often performed. Feeding is based on grazing or the supply of forages and wastes from the producer's crops11.
Methane sampling
Methane was sampled using the methodology developed by Garnsworthy et al12 and the Guardian-NG gas monitor (Edinburgh Instruments, Scotland, United Kingdom); this methodology measures environmental gas concentrations every second using a non-dispersed dual-wavelength system.
The devices were installed in the feeding troughs where cows were offered feed during milking. Adaptations were made for the different types of feeding troughs to create a closed atmosphere to prevent drafts from skewing CH4 concentrations. These adaptations aimed to generate the least disturbance during routine milking and allow the atmospheric sampling of the trough while the animal was feeding.
An adaptation period of one week was carried out to the presence of the new troughs. CH4 was measured for two weeks during milking; the aim was to have a minimum of 10 effective days of measurement in each PU.
Milk sampling
Milk samples were obtained from each animal during milking. Samples were at least 50 mL and were directly obtained from the weighers at the start of the measurements. After collection, samples were preserved with bronopol and identified with the PU’s number and the animal’s identification number.
Milk samples were analyzed in the milk quality control laboratory of the Asociación Holstein de México A.C. using the mid-infrared technique to measure protein and fat percentages.
DNA sampling and extraction
Hair follicles were collected from the hairs obtained from the tail of all the animals included in this study. Hair samples were labeled and sent to the GENESEEK laboratory (Lincoln, Nebraska). In this laboratory, DNA was extracted, and genotypes were obtained through high-density microarrays. The GGP BOVINE LD V4 array was used for the animals from FDs; with this array, it is possible to get 30,125 SNPs. As for the animals from the STD and DP systems, the GGDP BOVINE 150K array was used to identify 138,962 SNPs per animal; this is because crossed animals require a greater number of markers for the information to be valid. In this study, only the SNPs located in the 29 autosomal chromosomes were included. The quality control of the genotypes was carried out using the PLINK 1.713 software and consisted of 1) removing the individuals with less than 90 % of the genotypic information, 2) removing the animals with a minor allele frequency of less than 5%, and 3) removing the animals with less than 90 % of the useful markers. At the end of the quality control analysis and keeping the markers shared in both platforms, the number of available markers was 20,776 SNPs for each animal.
Statistical analysis
Estimation of the genomic relationships
Variance components were estimated using the mixed model methodology. Due to the lack of complete genealogical information needed to build the additive relationship matrix (A Matrix), molecular markers were used to construct the genomic relationship matrix between all animals (G). The G matrix was built based on the method proposed by VanRaden14. This method creates the M matrix using the dimensions: number of individuals (n) x number of markers (m). The matrix elements were coded as -1 (homozygous for one allele), 0 (heterozygous), and 1 (homozygous for the other allele). The P(nxm) matrix is subtracted from the M matrix; this subtraction results in the Z matrix (Z = M - P). The P(nxm) matrix contains columns with all the 2(pi-0.5) elements, where pi is the frequency of the second allele in the locus i. Finally, the G matrix was calculated as:
Estimation of the variance components
The variance components for CH4 emissions and the milk components (fat and protein percentages) were implemented with the ASReml-R program15.
The model was selected based on the effects of daily milk production during measurements, lactation days, lactation period, lactation number, production system, herd number, and breed on the response variables: methane production during milking, fat percentage, protein percentage. All the logical combinations within the fixed and random effects that converged with the response variables were tested. The resulting univariate model is represented as follows:
Where,
Estimation of the covariance components
The covariance components between CH4 emissions during milking and milk components were calculated with the ASReml-R program15. The variances estimated with the univariate models were used as initial values to estimate covariances.
The bivariate model is represented in matrix terms as follows:
Where,
Estimation of the genetic parameters
The h2 were obtained from the variance components estimated with
the univariate models. The genetic correlations (
The
Selection index
The response to selection was estimated by different selection indices between methane production and milk production and components. A sensitivity analysis identified different scenarios in which methane emissions, fat percentage, and protein percentage could be selected. Thus, it was possible to observe the dynamics between the accuracy of the indices and their genetic gain (Table 2). The selection indices were carried out for five generations. The traits included in the indices were assigned a value based on selection importance and intensity; the sum of the values in absolute quantities must be equal to 100. CH4 emissions were assigned values ranging from 0 to -100; fat and protein percentages were assigned values ranging from 0 to 100.
Index | CH4 | Fat | Protein | Index | CH4 | Fat | Protein |
---|---|---|---|---|---|---|---|
INDEX1 | -100 | 0 | 0 | INDEX34 | 0 | 10 | 90 |
INDEX2 | -90 | 0 | 10 | INDEX35 | -40 | 40 | 20 |
INDEX3 | -90 | 10 | 0 | INDEX36 | -20 | 20 | 60 |
INDEX4 | -80 | 0 | 20 | INDEX37 | -30 | 30 | 40 |
INDEX5 | -80 | 10 | 10 | INDEX38 | -40 | 50 | 10 |
INDEX6 | -80 | 20 | 0 | INDEX39 | -10 | 20 | 70 |
INDEX7 | -70 | 0 | 30 | INDEX40 | 0 | 20 | 80 |
INDEX8 | -70 | 10 | 20 | INDEX41 | -40 | 60 | 0 |
INDEX9 | -60 | 0 | 40 | INDEX42 | -30 | 40 | 30 |
INDEX10 | -70 | 20 | 10 | INDEX43 | -20 | 30 | 50 |
INDEX11 | -70 | 30 | 0 | INDEX44 | -10 | 30 | 60 |
INDEX12 | -60 | 10 | 30 | INDEX45 | -30 | 50 | 20 |
INDEX13 | -50 | 0 | 50 | INDEX46 | -20 | 40 | 40 |
INDEX14 | -60 | 20 | 20 | INDEX47 | 0 | 30 | 70 |
INDEX15 | -50 | 10 | 40 | INDEX48 | -30 | 60 | 10 |
INDEX16 | -40 | 0 | 60 | INDEX49 | -10 | 40 | 50 |
INDEX17 | -60 | 30 | 10 | INDEX50 | -30 | 70 | 0 |
INDEX18 | -60 | 40 | 0 | INDEX51 | -20 | 50 | 30 |
INDEX19 | -50 | 20 | 30 | INDEX52 | -20 | 60 | 20 |
INDEX20 | -30 | 0 | 70 | INDEX53 | -10 | 50 | 40 |
INDEX21 | -40 | 10 | 50 | INDEX54 | -20 | 70 | 10 |
INDEX22 | -50 | 30 | 20 | INDEX55 | 0 | 50 | 50 |
INDEX23 | -20 | 0 | 80 | INDEX56 | 0 | 40 | 60 |
INDEX24 | -40 | 20 | 40 | INDEX57 | -20 | 80 | 0 |
INDEX25 | -10 | 0 | 90 | INDEX58 | -10 | 60 | 30 |
INDEX26 | -30 | 10 | 60 | INDEX59 | 0 | 60 | 40 |
INDEX27 | -50 | 40 | 10 | INDEX60 | -10 | 70 | 20 |
INDEX28 | 0 | 0 | 100 | INDEX61 | -10 | 80 | 10 |
INDEX29 | -50 | 50 | 0 | INDEX62 | 0 | 70 | 30 |
INDEX30 | -20 | 10 | 70 | INDEX63 | -10 | 90 | 0 |
INDEX31 | -40 | 30 | 30 | INDEX64 | 0 | 80 | 20 |
INDEX32 | -30 | 20 | 50 | INDEX65 | 0 | 90 | 10 |
INDEX33 | -10 | 10 | 80 | INDEX66 | 0 | 100 | 0 |
The variance and covariance components used were those obtained with the previously described models for milk components (fat and protein percentages) and CH4 production during milking.
The original specification of the selection index foresees the use of a correlated variable (I) based on the phenotypic performance of each animal for several traits5. Therefore, it is defined as:
I = b p
Where p is a vector of phenotypic values for the selection criteria and b corresponds to the weighting factors used in selection decision making. To maximize the correlation of I with the contribution of any candidate for the selection as a possible parent, the information is combined as:
Ga = Pb
Where G is a nxm matrix of genetic variances and covariances between all the m traits, a is a mx1 vector of values relative to the selection intensity for all traits. P is a nxn matrix of phenotypic variances and covariances between the n traits measured and available as selection criteria and b is a nx1 vector of weighting factors applied to the traits used in selection decision making. Thus, the previous equation is solved as:
P -1 Ga = b
To obtain the weighting factors contained in b, the selection candidates are classified based on the index (I).
The index precision (
Where, b is a vector of weighting factors to be applied to the traits used to decide the selection; P is a matrix of phenotypic variances and covariances between the measured traits used as selection criteria; a is a vector of relative values for all traits; Q is a matrix of the genetic variances and covariances between all the traits considered as part of the system.
The genetic gain
Where: i= selection intensity; G= matrix of the genetic
variance-covariance of the traits; b= is a vector of weighting factors to be
applied to traits used in selection decision making;
The standard deviation of the index was calculated as follows:
Where: b= vector of the weighting factors applied to the traits used in selection decision making; P= matrix of the phenotypic variances and covariances between the measured traits used as selection criteria.
Results
The CH4 emissions in the STD system were 0.08 mg of CH4/ L; FD and DP systems produced 0.06 mg of CH4/ L (Table 3). The average of the three systems was 0.065 mg of CH4/ L. As for milk components, the average fat percentage in the three systems was 4.82 %; the values per system were 3.69 % in STD, 3.72% in FD, and 6.84 % in DP. The protein percentage in the STD, FD, and DP systems was 3.20, 3.29, and 3.21 %, respectively. When combining the three systems, the average protein percentage was 3.23 %.
System | Methane (mg/L) | Fat % | Protein % | |||
---|---|---|---|---|---|---|
Mean | SD | Mean | SD | Mean | SD | |
DP (n=98) | 0.06 | 0.039 | 6.84 | 4.938 | 3.21 | 0.405 |
FD (n=102) | 0.06 | 0.014 | 3.72 | 0.632 | 3.29 | 0.315 |
STD (n=74) | 0.08 | 0.016 | 3.69 | 0.492 | 3.20 | 0.417 |
Average | 0.065 | 0.028 | 4.828 | 3.344 | 3.234 | 0.380 |
DP= dual-purpose system, FD= family dairy, STD= specialized tropical dairy, N= number of observations, SD= standard deviation.
The h2 estimated for CH4 emissions during milking using the univariate model was 0.19. Similarly, the h2 for fat percentage was 0.39 and 0.18 for protein percentage.
However, the h2 estimated using bivariate models was 0.32±0.245 for CH4 emissions during milking and 0.46±0.278 for fat percentage. The h2 for protein percentage was similar to the one estimated with the univariate model. However, the h2 of CH4 is similar to the one found in the bivariate analysis with fat percentage (0.35). The genetic correlations between milk fat and protein percentage and CH4 emissions during milking were -0.090 ± 0.080 and -0.18 ± 0.575, respectively.
Table 4 and Figure 1 show the accuracy of the selection indices (
Index | rHI | CH4 (mg/L) | Fat (%) | Protein (%) | Index | rHI | CH4 (mg/L) | Fat (%) | Protein (%) |
---|---|---|---|---|---|---|---|---|---|
INDEX1 | 19.58 | -0.021 | 0.030 | 0.073 | INDEX34 | 10.72 | -0.013 | 0.015 | 0.117 |
INDEX2 | 18.38 | -0.021 | 0.029 | 0.078 | INDEX35 | 10.63 | -0.021 | 0.030 | 0.086 |
INDEX3 | 17.92 | -0.021 | 0.030 | 0.073 | INDEX36 | 10.41 | -0.018 | 0.023 | 0.110 |
INDEX4 | 17.23 | -0.021 | 0.029 | 0.082 | INDEX37 | 10.32 | -0.020 | 0.027 | 0.100 |
INDEX5 | 16.72 | -0.021 | 0.030 | 0.078 | INDEX38 | 10.12 | -0.021 | 0.031 | 0.078 |
INDEX6 | 16.26 | -0.021 | 0.030 | 0.072 | INDEX39 | 9.95 | -0.016 | 0.020 | 0.115 |
INDEX7 | 16.14 | -0.021 | 0.028 | 0.088 | INDEX40 | 9.70 | -0.014 | 0.016 | 0.117 |
INDEX8 | 15.57 | -0.021 | 0.029 | 0.083 | INDEX41 | 9.70 | -0.021 | 0.032 | 0.069 |
INDEX9 | 15.13 | -0.020 | 0.027 | 0.093 | INDEX42 | 9.62 | -0.020 | 0.029 | 0.094 |
INDEX10 | 15.06 | -0.021 | 0.030 | 0.078 | INDEX43 | 9.56 | -0.018 | 0.025 | 0.107 |
INDEX11 | 14.61 | -0.021 | 0.031 | 0.072 | INDEX44 | 9.01 | -0.017 | 0.022 | 0.113 |
INDEX12 | 14.49 | -0.020 | 0.028 | 0.089 | INDEX45 | 9.01 | -0.020 | 0.030 | 0.087 |
INDEX13 | 14.21 | -0.020 | 0.026 | 0.099 | INDEX46 | 8.77 | -0.019 | 0.027 | 0.103 |
INDEX14 | 13.92 | -0.021 | 0.029 | 0.084 | INDEX47 | 8.70 | -0.014 | 0.018 | 0.117 |
INDEX15 | 13.50 | -0.020 | 0.027 | 0.095 | INDEX48 | 8.49 | -0.021 | 0.032 | 0.078 |
INDEX16 | 13.41 | -0.019 | 0.025 | 0.104 | INDEX49 | 8.12 | -0.017 | 0.024 | 0.111 |
INDEX17 | 13.40 | -0.021 | 0.030 | 0.078 | INDEX50 | 8.09 | -0.020 | 0.033 | 0.067 |
INDEX18 | 12.96 | -0.021 | 0.031 | 0.071 | INDEX51 | 8.05 | -0.020 | 0.029 | 0.097 |
INDEX19 | 12.86 | -0.020 | 0.028 | 0.090 | INDEX52 | 7.41 | -0.020 | 0.031 | 0.089 |
INDEX20 | 12.74 | -0.018 | 0.023 | 0.110 | INDEX53 | 7.28 | -0.018 | 0.026 | 0.107 |
INDEX21 | 12.63 | -0.019 | 0.026 | 0.101 | INDEX54 | 6.90 | -0.020 | 0.033 | 0.078 |
INDEX22 | 12.27 | -0.021 | 0.030 | 0.085 | INDEX55 | 6.78 | -0.016 | 0.022 | 0.114 |
INDEX23 | 12.23 | -0.017 | 0.020 | 0.114 | INDEX56 | 6.78 | -0.015 | 0.020 | 0.116 |
INDEX24 | 11.90 | -0.020 | 0.027 | 0.097 | INDEX57 | 6.52 | -0.020 | 0.034 | 0.064 |
INDEX25 | 11.89 | -0.015 | 0.017 | 0.117 | INDEX58 | 6.52 | -0.019 | 0.029 | 0.100 |
INDEX26 | 11.89 | -0.018 | 0.024 | 0.107 | INDEX59 | 5.89 | -0.017 | 0.025 | 0.110 |
INDEX27 | 11.76 | -0.021 | 0.031 | 0.078 | INDEX60 | 5.86 | -0.020 | 0.032 | 0.091 |
INDEX28 | 11.75 | -0.013 | 0.014 | 0.118 | INDEX61 | 5.35 | -0.020 | 0.034 | 0.077 |
INDEX29 | 11.32 | -0.021 | 0.032 | 0.070 | INDEX62 | 5.08 | -0.017 | 0.028 | 0.104 |
INDEX30 | 11.30 | -0.017 | 0.021 | 0.112 | INDEX63 | 5.02 | -0.019 | 0.036 | 0.057 |
INDEX31 | 11.23 | -0.020 | 0.029 | 0.092 | INDEX64 | 4.40 | -0.018 | 0.032 | 0.092 |
INDEX32 | 11.08 | -0.019 | 0.025 | 0.104 | INDEX65 | 3.90 | -0.018 | 0.035 | 0.073 |
INDEX33 | 10.91 | -0.015 | 0.018 | 0.116 | INDEX66 | 3.67 | -0.017 | 0.037 | 0.044 |
In the indices with greater
Discussion
The CH4 emissions during milking estimated in this study are lower than those reported by Bell et al17; this is possibly due to diet heterogeneity. In the farms analyzed in this study, animals are fed by grazing; in the specialized systems, animals are fed a concentrate-based diet. The h2 of CH4 production during milking estimated in this study are similar to those reported by other authors with metabolic chambers18,19 or even with prediction equations8. These results coincide with those reported by other authors using similar CH4 measuring methodologies20; this suggests that CH4 during milking is a trait that could be potentially used in breeding programs because it can be easily incorporated into production control programs and requires unsophisticated equipment compared to respiratory chambers; additionally, this equipment can be mobilized to places of difficult access. The fat percentage h2 estimated in this study is higher than those observed in other studies21-22. The protein percentage h2 is lower than the 0.23 reported by Othmane et al23; this may be due to the inherent heterogeneity in dairy production units depending on the herd's diet and race.
The genetic correlations between CH4 emissions and milk components suggest genetic antagonism. However, the estimates in this study were not different from zero. The CH4 and grams of milk fat correlation observed by Pszczola et al4 is higher (0.21) than that reported in this study; the correlation between CH4 and grams of milk protein in this study was similar to the one reported by Lassen et al24 (0.39). In both cases, the opposite sign. It is important to mention that this difference is due to the units of measurement since milk production and the percentages of milk components have a negative correlation. In contrast, milk production and the content of its components have a positive correlation25. Currently, the reduction of CH4 emissions through genetic selection has been proposed; this could reduce dairy cattle CH4 emissions in 10 yr between 11 and 26 %8; 5 % in beef cattle26.
The selection indices performed by Kandel et al27 include fat and protein production, which were positively correlated with CH4 production. Considering the units of measurement, their result is similar to the one reported in this study since the correlation between milk production and the percentage of milk components is negative. In contrast, the correlation between milk production and the content of milk components is positive27.
Conclusions and implications
The genetic correlations between CH4 emissions and milk components (fat and protein percentage) suggest that a breeding program aimed to simultaneously decrease CH4 emissions and increase the percentages of milk components is feasible. In other words, these results show that it is possible to genetically select animals to reduce CH4 emissions without negatively affecting milk composition; this is confirmed by the genetic gains per generation predicted by the selection indices. The above, with CH4 reductions during milking in five generations ranging from 0.013 to 0.021 mg/L, without decreasing fat and protein percentages.
Acknowledgments
The authors would like to thank the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias and the Consejo Nacional de Ciencia y Tecnología.
REFERENCES
1. IPCC. Climate Change 2007: Impacts, adaptation and vulnerability. 1a ed. Cambridge University Press, 2007. doi:10.2134/jeq2008.0015br. [ Links ]
2. FAO. Greenhouse gas Emissions from the Dairy Sector. Roma, Italy, 2010. doi:10.1016/S0301-4215(01)00105-7. [ Links ]
3. Wall E, Simm G, Moran D. Developing breeding schemes to assist mitigation of greenhouse gas emissions. Animal 2010;4(03):366-376.doi:10.1017/S175173110999070X. [ Links ]
4. Pszczola M, Calus MPL, Strabel T. Short communication: Genetic correlations between methane and milk production, conformation, and functional traits. J Dairy Sci 2019;102:1-5. doi:10.3168/jds.2018-16066. [ Links ]
5. MacNeil MD, Nugent RA, Snelling WM. Breeding for profit: an introduction to selection index concepts. Range Beef Cow Symposium. 1997. [ Links ]
6. Philipson J, Banos G, Arnason T. Present and future uses of selection index methodology in dairy cattle. J Dairy Sci 1994;77:3252-3261. doi:10.3168/jds.s0022-0302(94)77266-0. [ Links ]
7. Grainger C, Clarke T, McGinn SM, Auldist MJ, Beauchemin KA, Hannah MC, et al Methane emissions from dairy cows measured using the sulfur hexafluoride (SF6) tracer and chamber techniques. J Dairy Sci 2007;90:2755-2766. doi:10.3168/jds.2006-697. [ Links ]
8. De Haas Y, Windig JJ, Calus MPL, Dijkstra J, Haan M De, Bannink A, Veerkamp RF. Genetic parameters for predicted methane production and potential for reducing enteric emissions through genomic selection. J Dairy Sci 2011;94:6122-6134.doi:10.3168/jds.2011-4439. [ Links ]
9. García E. Modificaciones al sistema de clasificación Köppen. Quinta ed. México; Instituto de Geografía-UNAM; 2014. [ Links ]
10. Román PH, Ortega RL, Ruiz LFJ, Medina CM, Vera AHR, Núñez HG, et al. Producción de leche de bovino en el sistema de doble propósito. Libro técnico. INIFAP-CIRGOC. Veracruz, Ver. 2009. [ Links ]
11. Vera A, Hernández A, Espinoza G, Díaz A, Román P, Núñez H, Al E. Producción de leche de bovino en el sistema familiar. Libro técnico. INIFAP-CIRGOC. Veracruz, Ver. 2009. [ Links ]
12. Garnsworthy PC, Craigon J, Hernandez-Medrano JH, Saunders N. Variation among individual dairy cows in methane measurements made on farm during milking. J Dairy Sci 2012;95:3181-3189.doi:10.3168/jds.2011-4606. [ Links ]
13. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira M, Bender D, et al. PLINK (1.07). PLINK: a toolset for whole-genome association and population-based linkage analysis. Am J Human Genet 2007;81:1-293.doi:http://pngu.mgh.harvard.edu/purcell/plink/. [ Links ]
14. VanRaden PM. Efficient methods to compute genomic predictions. J Dairy Sci 2008;91:4414-23. doi:10.3168/jds.2007-0980. [ Links ]
15. Butler DG, Cullis BR, Gilmour AR, Gogel BJ. Analysis of mixed models for S language environments: ASReml-R reference manual. 2009:145. doi:citeulike-article-id:10128936. [ Links ]
16. Cameron N. Selection indices and prediction of genetic merit in animal breeding. CAB international, 1997. [ Links ]
17. Bell M, Saunders N, Wilcox R, Homer E, Goodman J, Craigon J, Garnsworthy P. Methane emissions among individual dairy cows during milking quantified by eructation peaks or ratio with carbon dioxide. J Dairy Sci 2014;97:6536-6546. doi:10.3168/jds.2013-7889. [ Links ]
18. Pinares-Patiño CS, Hickey SM, Young EA, Dodds KG, MacLean S, Molano G, et al. Heritability estimates of methane emissions from sheep. Animal 2013;7(2):316-321.doi:10.1017/S1751731113000864. [ Links ]
19. Pickering NK, Chagunda MGG, Banos G, Mrode R, Mcewan JC, Wall E. Genetic parameters for predicted methane production and laser methane detector measurements. Am Soc Anim Sci 2015;92(1):11-20.DOI:10.2527/jas2014-8302. [ Links ]
20. Lassen J, Lovendahl P, Madsen J. Accuracy of noninvasive breath methane measurements using Fourier transform infrared methods on individual cows. J Dairy Sci 2012;95:890-898.doi:10.3168/jds.2011-4544. [ Links ]
21. Miglior F, Sewalem A, Jamrozik J, Bohmanova J, Lefebvre DM, Moore RK. Genetic analysis of milk urea nitrogen and lactose and their relationships with other production traits in Canadian Holstein Cattle. J Dairy Sci 2007;90(5):2468-2479. doi:10.3168/jds.2006-487. [ Links ]
22. González-Peña D, Guerra D, Evora JC, Portales A, Ortiz J, González S, Ramírez R. Heredabilidad y tendencia genética de la producción de leche y grasa en vacas Siboney de Cuba. Ciencia y Tecnología Ganadera 2009;3(1):27-32. [ Links ]
23. Othmane MH, De La Fuente LF, Carriedo JA, San Primitivo F. Heritability and genetic correlations of test day milk yield and composition, individual laboratory cheese yield, and somatic cell count for dairy ewes. J Dairy Sci 2002;85(10):2692-2698. doi:10.3168/jds.S0022-0302(02)74355-5. [ Links ]
24. Lassen J, Poulsen NA, Larsen MK, Buitenhuis AJ. Genetic and genomic relationship between methane production measured in breath and fatty acid content in milk samples from Danish Holsteins. Anim Prod Sci 2016;56(3):298-303. doi:10.1071/AN15489. [ Links ]
25. Schutz MM, Hansen LB, Steuernagel GR, Reneau JK, Kuck AL. Genetic parameters for somatic cells, protein, and fat in milk of Holsteins. J Dairy Sci 1990;73:494-502. doi:10.3168/jds.S0022-0302(90)78697-3 [ Links ]
26. Hayes BJ, Donoghue KA, Reich CM, Mason BA, Bird-Gardiner T, Herd RM, Arthur PF. Genomic heritabilities and genomic estimated breeding values for methane traits in Angus cattle. J Anim Sci 2016;94:902-908. doi:10.2527/jas.2015-0078. [ Links ]
27. Kandel P, Vanderick S, Vanrobays ML, Soyeurt H, Gengler N. Consequences of genetic selection for environmental impact traits on economically important traits in dairy cows. Anim Prod Sci 2018;58(1966):1779-1787. doi:10.1071/AN1659.2 [ Links ]
Received: April 21, 2019; Accepted: August 26, 2020