Introduction

In recent decades, the complex vertebral malformation (CVM) syndrome has probably been one the most frequently detected autosomal, recessive, hereditary defects in Holstein cattle (^{Patel & Patel, 2014}; ^{Zhang et al., 2012}). CVM is caused by a point mutation from Guanine (G) to Thymine (T) at nucleotide position 559 of the gene SLC35A3 (bovine solute carrier family 35, member 3) in region BTA3 on chromosome 3 (^{Kanae, Endoh, Nagahata & Hayashi, 2005}). The mutation provokes the substitution of valine to phenylalanine in amino acid 180 of the protein which inhibits the functioning of uridine diphosphate N-acetylglucosamine (UDP-GlcNAc), a nucleotide-sugar transporting protein residing in the Golgi apparatus. This peptide is essential for the control of vertebral formation in the mesoderm, and the defective molecule leads to vertebral malformation (^{Thomsen et al., 2006}). Approximately 80% of affected homozygous fetuses are aborted by the day 260 of gestation (^{Ghanem & Nishibori, 2008}).

The defective allele for CVM was spread globally by the over-usage of a high-performance bull, Penstate Ivanhoe Star, and its offspring, Carlin-M-Ivanhoe Bell, which were both carriers of the deleterious allele (^{Agerholm, 2007}). The propagation of recessive genetic defects in domestic animal populations negatively affects the economics of dairy farms and reduces the success of genetic improvement programs (^{Ghanem & Nishibori, 2008}).

The objectives of this study were to identify cows carrying CVM in three Holstein herds in Western Mexico, to identify any differences in first- and second-lactation milk production between carrier cows and normal homozygotes, and to determine whether the carrier genotype has an improving effect on dairy production.

Materials and Methods

This study included 308 Holstein cows with two or more lactations from three industrial dairies in the state of Jalisco, Western Mexico. The first, second, and third herds were comprised of 102, 100, and 106 cows, respectively. Total milk yield adjusted to 305 days in kilograms of milk production records for 2016, for the first and second lactations, in kilograms of milk for each cow, were obtained using DairyCOMP 305® version 5 software (DC305, 2013).

DNA was extracted from blood samples using the Quick-DNA™ Universal Kit (Zymo Research, Orange, CA, USA). For genotyping by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) assay, the Thermo Scientific^™ DreamTaq^™ PCR kit (Takara Bio, Inc., Kusatsu, Shiga, JAP) was used.

The PCR for amplification of 233 bp fragments from cow’s DNA was performed with reaction conditions as follows: 5 pmol of each forward and reverse primer, 0.5 mM dNTP, 1.25 mM MgCl2, 1.6 μl 10X DreamTaq™ Green Buffer, and 0.333U of DreamTaq™ polymerase in 12 μl reaction volumes. Cycling parameters included an initial 5 min denaturation at 94 °C followed by 35 cycles of 30 s denaturation at 94 °C, annealing for 30 s at 56 °C, and a 72 °C extension for 30 s. The cycling parameters were followed by a final extension at 72 °C for 5 min. The restriction enzyme PstI from New England Biolabs™ (New England Biolabs, Inc., Ipswich, MA, USA) in a TECHNE™ TC-5000 (Techne Inc. Burlinon, NJ, USA) thermal oscillator was based on methodologies described by ^{Kanae et al. (2005)} for gene SLC35A3 to identify the CVM polymorphism. If the 233 bp amplified sequence belongs to the wild-type allele, the enzyme cut the fragment into two parts (21 pb and 212 bp); if it belongs to the mutant allele, there was no cutting site and the fragment will remain complete.

To calculate and analyze allelic and genotype frequencies, the POPGENE^© software (version 1.32; POPGENE, 1997) was used. The SPSS software, version 20, 2011, was used to perform a mixed statistical model in each herd separately, in order to identify differences in first- and second-lactation milk production between carrier cows and wild type homozygotes of gene SLC35A3.

A fixed and random effects design was used considering the following mixed statistical model:

where Yij is a cow’s first- and second-lactation milk production adjusted to two milkings at 305 days; µ is the general average milk production; genotype SLC35A3 refers to the genotypes for the SLC35A3 gene in each animal (N/N and N/CVM); _i is 1, 2; animal is the total random genetic component of each animal; j is 1, 2, …, 308; and eij is the random error for each measurement.

where Y_ij is a cow’s first- and second-lactation milk production adjusted to two milkings at 305 days; µ is the general average milk production; genotype SLC35A3 refers to the genotypes for the SLC35A3 gene in each animal (N/N and N/CVM); _i is 1, 2; animal is the total random genetic component of each animal; j is 1, 2, …, 308; and e_ij is the random error for each measurement.

Results

The genotyping of SLC35A3 revealed 260 homozygotes cows for the wild allele and 40 carrier cows with CVM mutation. Table 1 shows the distribution and frequencies of SLC35A3 genotypes in the three herds. The mutant homozygous genotype (CVM/CVM) was absent due to the sampling of only adult cows and the nature of the syndrome, as homozygous animals with this mutation survive for only a few hours after birth. Not a significant difference in milk production was observed between normal and heterozygous genotypes (p > 0.05; Table 2). Differences in milk production by lactation among herds, which can be explained by the different environmental and handling factors present in each daily farm, were found. Milk production did not differ between normal and heterozygous CVM genotypes.

Discussion

In the first decade of this millennium, high allelic frequencies of animals carrying CVM, such as 0.325 (^{Nagahata et al., 2002}) and 0.26 (^{Berglund, Persson & Stalhammar, 2004}), were reported. However, in the second decade of this millennium, the frequency of the mutant allele of the CVM decreased. The genetic frequency of the CVM mutation in our study was 0.06, value which fits into the previously reported values ranging from 0.115 to 0.008 (^{Akyuz, Sariozkan & Bayram, 2015}; ^{Hemati, Gharaie-Fathabad, Fazeli, Namvar & Ranji, 2015}; ^{Mahdipour et al., 2010}; ^{Meydan, Yildiz & Agerholm, 2010}; ^{Paiva et al., 2013}; ^{Sun et al., 2011}; ^{Wang et al., 2011}; ^{Zhang et al., 2012}). This trend suggests gradual selection against the mutation since the year 2000. This trend suggests a gradual selection against the mutant allele; since the early reports in the first decade of this century, the frequency of the mutation has been progressively decreasing; in addition to that, the absence of the mutation in some herds has been reported since 2010 (^{Eydivandi, Sevedabai & Amirinia, 2011}). The low frequency of CVM carriers in the present study does not imply that the frequencies of these diseases cannot be higher in other herds, but it is known that even the low frequencies of autosomal recessive genetic diseases have an economic impact on herds of dairy cattle (^{Uffo & Acosta, 2009}); therefore, it is important to genotype the herds. Not a significant difference in milk production was observed between normal and heterozygous genotypes (p > 0.05). A literature review revealed no previous study examining associations between dairy production parameters and the SLC35A3 genotype causing CVM.

The CVM mutation was detected in three Mexican Holstein cow herds analyzed in this study, since a direct relationship between milk production and the CVM carrier genotype was not found. The use of molecular marker-assisted selection against CVM will not reduce herd productivity in Holstein-Friesian cattle in Mexico. The implementation of programs to monitor hereditary diseases among Holstein cattle ranches is necessary to avoid the spread and future presentation of genetic diseases.