A microsatellite study of bovine solute carrier family 11 a1 (Slc11a1) gene diversity in Mexico in relation to bovine tuberculosis

Bovine tuberculosis, caused by Mycobacterium bovis, is a disease of socio-economic and public health importance and of significance to international trade regulation. Allelic variants of several genes have been implicated in the genetic susceptibility to tuberculosis in some human populations, but little is known in cattle. We surveyed 34 European, 18 Asian, 20 Creole and 23 hybrid bovines for polymorphisms of the bovine solute carrier family 11 a1(Slc11a1) gene, formerly known as natural resistance associated macrophage protein (Nramp1), gene by typing the cattle using two microsatellite loci closely linked to this gene. The microsatellites used were 311-22, located at the 3’ untranslated region (3’ UTR) of the Slc11a1 gene, and ARO28 situated about 0.6 cM upstream of the same gene Based on allele size in base pairs (bp) we determined five 311-223 locus variants (221, 223, 225, 227 and 229 bp) and 12 ARO28 loci. There was marked diversity and a very high level of heterozygosity in most of the cattle surveyed except the Europeans bovines and especially Holsteins in relation to the 3’ UTR microsatellite locus.

Bovine tuberculosis (BTB), caused by Mycobacterium bovis, is a disease of socio-economic and public health importance and of significance to international trade regulation.Because M. bovis infects cattle and other domesticated and wild animals world-wide this disease is difficult to control or eradicate (Cousins 2001) and remains a major problem to the livestock industry since vaccination is not used for the control of the disease or as therapy for infected animals.Resistance to infectious diseases is often influenced by the genetics of the host and identifying the genes involved can lead to marker-assisted selection strategies which can contribute to controlling the spread of M. bovis infection by increasing the frequency of desirable genes.Molecular genetics studies have focused on understanding host susceptibility and resistance factors such as innate, specific and regulatory immune response-related genes and their relationship to infectious pathogens.Allelic variants of several genes have been implicated in the genetic susceptibility to tuberculosis in some human populations, but little is known in cattle (Estrada-Chavez et al. 2001).
The solute carrier family 11 a1 gene (Slc11a1), formerly known as natural resistance associated macrophage protein (Nramp1), has been positionally cloned in mice and it has been established that mutations of this gene are responsible for susceptibility to a number of intracellular microorganisms, including M. bovis/BCG, Salmonella typhimurium and Leishmania donovani (Govoni et al. 1996).The Slc11a1/Nramp1 gene encodes a pH-dependent proton/divalent cation antiporter localized in mature endosome or lysosome membranes, this protein specifically transports iron (Fe) from the cytosol into these acidic compartments and contributes to the generation of antimicrobial reactive oxygen intermediates (ROI) by the mononuclear phagocytic cell system via the Fenton and Haber-Weiss reaction (Goswami et al. 2001).Genetic variation in the Slc11a1 locus affects susceptibility to tuberculosis in West African, Guinea-Conakry, Koreans, and Japa-nese patients (Bellamy et al. 1998;Cervino et al. 2000;Ryu et al. 2000;Gao et al. 2000).Microsatellite polymorphism located in the 5' untranscribed promoter region, with at least 10 identified alleles, has been found to regulate different levels of Slc11a1 expression by the juxtaposition of trans-acting elements which appear to differentially affect the functionality of this gene (Searle et al. 1999).
The bovine Slc11a1 gene has been isolated as the homologue of the mouse and human genes (Feng et al. 1993;Horín et al. 1999), but the influence of this gene on bovine tuberculosis remains undetermined due to insufficient studies (Barthel et al. 2000).In order to study the association between Slc11a1 and susceptibility to disease information must be obtained on the number and frequency of Slc11a1 alleles in different lineages of bovines, such studies being best carried out using the microsatellite approach.
DNA was obtained from semen or blood samples according to previously described methods (Williams, 1997;Miller, 1988) and DNA was quantified by fluorimetry and the concentration adjusted to 20 ng/mL and its quality evaluated using ethidium bromide-stained agarose gels.To genotype the cattle we used two microsatellite loci closely-linked to the Slc11a1 gene, the 311-22 microsatellite located at the 3' untranslated region (3' UTR) of the Slc11a1 gene and the ARO28 microsatellite about 0.6 cM upstream of, but closely-linked to, the same gene (Kappes 1997).Primers were designed to amplify a 223 bp fragment containing the 311-22 microsatellite (GenBank U12862) (5' GGA ATG AGT GGG CAC AGT GGC 3'; 5' CCT TCC AGA ACT CCC TCT CCG 3').A segment located between 256 and 276 bp containing the ARO28 microsatellite was amplified using previously described primers (GenBank L06143) (5' GAT TTC TCT AGT GAG TAA CA 3'; 5' TCT TGC CCA GAT GTT CTT AG 3').
The PCR mix (20 mL) consisted of 10 mM of Tris-HCl (pH 8.3), 50 mM KCl, 0.4 mM each primer, 200 mM each dNTPs, 1.25 mM of MgCl 2 for 311-22 and 2.0 mM of MgCl 2 for ARO28, 0.25 units of Taq-DNA-polymerase (Biogenica, Mexico City) and 40 ng of DNA.Amplification conditions for the 311-22 locus were 30 cycles of 94 °C for 30 s, 62 °C for 30 s and 72 °C for 30 s. Same conditions were used for the ARO28 microsatellite, with 56 °C at the annealing step.
The PCR products were verified using ethidium bromide-stained agarose gels and then 1 mL of each product was mixed with 0.5 mL of the TAMRA 350 molecular marker (Applied Biosystem) and 30 mL of deionized formamide.Microsatellite sizing was carried out with the ABIprism 310 automatic sequencer and fragments analyzed with the GeneScan 2.1 and Genotyper 2.0 (Applied Biosystem) using at least duplicate samples.Allele identification was based on size.
Gene frequencies and individual genotypes were determined by direct allele counting and observed and expected heterozygosity was calculated and a Chi-squared (c 2 ) test with one degree of freedom (df) was performed to assess Hardy-Weinberg equilibrium (HWE) using the computer program GENEPOP 1.2 software (Rymond and Rousset 1995).
There have been no previous reports on Slc11a1 (or the Nramp1 locus as it was previously known) polymorphism frequencies in bovines from different lineages or breeds.Based on allele size in base pairs (bp) we determined a total of five 311-22 locus alleles and twelve ARO28 alleles, of which three 311-22 alleles (225, 227 and 229) and three ARO28 alleles (246, 280 and 288) have not, to the best of our knowledge, been previously described.The number of alleles found indicates that both loci were polymorphic.
The 311-22 and ARO28 frequencies and observed and expected heterozygosities (HS), and fixation indices (Fis) for each bovine population are shown in Table 1.The size of the 311-22 alleles ranged from 221 to 229 bp, of which the 223 allele was the most common in all groups at an average frequency of 57.4% and a range of from 37% to 65%, followed by the 227 and the 225 alleles with a frequency of about 16% for each.The 311-22 microsatellite 229 allele was present in Asian cattle and its hybrids (3/4 Holstein x 1/4 Brahman) but not in pure European or Creole cattle, while the 221 allele was absent from Creole cattle and the 227 allele was not found in Holsteins.The most frequent ARO28 alleles in almost all populations were alleles 260 (36.3%), 262 (21.6%) and 254 (13.6%), with the 246 and 256 alleles being found only in Asiatic cattle (Guzerat), the 270 allele only Holsteins and the 288 allele only in Jersey cattle.
In the 95 cattle studied we found a total of 55 different genotypes, which we designated using the ARO28 allele first and the 311-22 allele second keeping the order in the genome (Table 2).Most of the genotypes had frequencies ranging from 1 to 4% but the homozygous ARO28 260/260 -311-22 223/223 and ARO28 262/262 -311-22 223/223 genotypes were over-represented in European (38%) and Holstein-Frisian (50%) cattle, although these genotypes were almost absent from the other lineages.In contrast to European and Holstein-Frisian cattle, but as expected, we found that Asian, Creole and Creole/Asian hybrid cattle were mainly heterozygous for any genotypes demonstrated for them.The low Slc11a1 diversity may be a consequence of intense selection reduced effective population number and inbreeding, which has been previously reported for dairy cattle (Lindhe et al. 1998).
311-22 microsatellite 223 allele corresponds to the GT 11 dinucleotide (Slc11a1 or Nramp1.2) previously reported in a bovine hybrid obtained by crossing B. taurus (Angus, Jersey and Hereford) and with 1/8th B. indicus American Brahman breeds (Feng et al. 1993).The 311-22 microsatellite 223 allele and the 225 allele (GT 12 : Slc11a1 or Nramp1.3)were over-represented in all the cattle studied by us, while the 311-22 microsatellite 221 allele was not found in Creole cattle.Similarly, the 311-22 microsatellite 227 allele (GT 13 : Slc11a1 or Nramp1.4) and the 229 allele (GT 14 : Slc11a1 or Nramp1.5)were not found by us in Holstein-Frisian cattle.ho/ho: Homozygous for both loci; |he/ho: Heterozygous for ARO28 locus and homozygous for 311-22 locus; † † Hol x Bra: Holstein Frisian x Brahman; ‡ ‡ Guz x Cre:Guzerat x Creols; *: The most frequents genotypes /: Divide alleles; -: Divide loci.Barthel et al (2001) have shown that transfected murine macrophages can be used to measure the functionality of two bovine Slc11a1 alleles, one of which correlates with enhanced expression and the in vitro control of Brucella abortus replication but not with Salmonella enterica.Searle et al. (1999) have reported that polymorphism of the human Slc11a1 promoter (which encodes a Z-DNA forming di-nucleotide repeat) affects the level of gene expression and correlates with autoimmune versus infectious disease susceptibility in different human populations.Similarly, Beaumont et al. (2003) have shown that a polymorphic microsatellite located close to the 3' untranslated region of the Slc11a1 gene is a useful tool in co-segregation studies aimed at elucidating the involvement of this gene in the control of sheep and fowl resistance to Salmonella infection.
Most of the cattle studied by us are sires used by one of the major semen suppliers in Mexico (CONAMEGRA) and our work shows that polymorphism in Mexican cattle populations has not been sufficiently studied.Microsatellite polymorphisms at the Slc11a1 (Nramp1) locus could be very useful in association studies to test the relationship of this gene with resistance to Bovine tuberculosis.

Table 2 -
Genotypes identified in the entire population.Frequency of animals with this genotype; ‡ he/he: Heterozygous for both loci; § ho/he: Homozygous for ARO28 locus and Heterozygous for 311-22 locus; ¶