Genetic causes involved in Leishmania Chagasi infection innortheastern-Brazil

A sample of 502 individuals from 94 families from Jacobina, State of Bahia, Brazil, was investigated to determine the causal mechanisms involved in Leishmania chagasi (the causal agent of visceral leishmaniasis in the American hemisphere) infection, as measured by the intradermic reaction to antigens derived from this parasite, using complex segregation analyses. The results showed evidence of a major genetic mechanism acting on infection, with a frequency of a recessive (or additive) susceptibility gene (q) of approximately 0.45. A small multifactorial component (H = 0.29) acting in conjunction with a major recessive gene (q = 0.37) is not ruled out as a concomitant causative factor. INTRODUCTION Variation in host resistance/susceptibility to infection by intracellular parasites has been demonstrated to be genetically controlled in experimental animals (cf. Skamene and Pietrangieli, 1991; Skamene, 1998). Innate susceptibility to Leishmania donovani infection in mice is controlled by a single autosomal gene (Lsh) segregating for incompletely dominant resistant and recessive susceptible alleles (Bradley, 1977), located on chromosome 1 (Bradley et al., 1979). Acquired resistance is influenced by genes linked to the major (H-2) and minor (H-11) histocompatibility complexes (Blackwell et al., 1980, 1985; Roberts et al., 1989a,b). Experimental infections in mice have demonstrated that gene(s), also on chromosome 1, regulate resistance/susceptibility to Salmonella typhimurium (= Ity) (Plant and Glynn, 1974, 1976), Mycobacterium bovis (= Bcg) (Gros et al., Genetics and Molecular Biology Genetic causes involved in Leishman... http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1415-47571... 1 de 11 3/10/2011 14:4

Recent studies, utilizing high resolution genetic mapping techniques, have revealed that an array of closely linked marker genes has been conserved between murine chromosome 1 and the telomeric end of the long arm of human chromosome 2 (Schurr et al., 1989(Schurr et al., , 1990;;Mock et al., 1990;Malo et al., 1991;Barton et al., 1992).Nevertheless, other studies (Blackwell, 1992(Blackwell, , 1998;;Shaw et al., 1993;Levee et al., 1994) could not provide conclusive evidence of linkage between a putative "leprosy", "visceral leishmaniasis" or "tuberculosis" gene and DNA markers on the 2q35 chromosomic region (Morgan et al., 1994), which is believed to have homology to mouse chromosome 1 (Vidal et al., 1993).
The aim of the present study was to investigate possible genetic mechanisms acting on resistance/susceptibility to Leishmania chagasi (the causal parasite of visceral leishmaniasis in the American continent) infection in a sample of families from Jacobina, Brazil, using complex segregation analysis.

MATERIAL AND METHODS
The data were collected at random from a population living in an area endemic for visceral leishmaniasis, and consisted of nuclear families residing in Jacobina, Bahia State, Brazil (cf.Cabello et al., 1995;Lima et al., 1996).The total sample comprised 1,604 individuals belonging to 243 nuclear families, of which 502 individuals were tested for Leishmania infection.A skin test was used to characterize Leishmania infection.An intradermal injection was made of 50 mg of soluble protein from freeze-thaw preparations of L. chagasi centrifuged at 35,000 g.Skin tests were considered positive if the diameter of the reaction was larger than 5 mm, approximately two days after inoculation (Reed et al., 1986).

Segregation analysis
Segregation analyses were done on the 502 individuals, belonging to 94 nuclear families (Table I).This analysis was performed using the unified model (Lalouel et al., 1983), from the computer program POINTER (Lalouel and Morton, 1981;Morton et al., 1983), which incorporates the transmission frequencies (Elston and Stweart, 1971) into the mixed model (Morton and MacLean, 1974).The mixed model assumes an underlying liability scale to which a major locus, multifactorial component, and random environment contribute independently.The major gene effect results from segregation at a single locus with two alleles (A, a), where genotypes are distributed according to Hardy-Weinberg proportions.The estimated parameters are: the frequency (q) of the allele a responsible for the infection; the displacement between the two homozygous means (t); position of the mean of the heterozygote relative to the means of the two homozygous genotypes (d), and two parameters representing the multifactorial heritabilities in children (H) and parents (HZ).Additional parameters, t AA , t Aa , t aa , can be estimated to test deviations from Mendelian transmission of the major effect from parent to offspring, and denote the probabilities of transmitting allele A for genotypes AA, Aa, and aa, respectively.Under Mendelian transmission, t AA = 1, t Aa = 1/2, t aa = 0, while no major gene transmission is obtained when the tree taus are equal.In order to infer a major locus acting together or not with a polygenic mechanism, three conditions are usually required: i) rejection of the hypothesis: d = t = q = 0 (no major effect); ii) no rejection of Mendelian transmission (when t AA = 1, t Aa = 1/2, t aa = 0), and iii) rejection of the no Mendelian transmission model (when t AA = t Aa = t aa ).Different hypotheses were tested by estimating or fixing parameters of the complete model.Standard tests of hypotheses were performed using the likelihood ratio criterion (-2lnL).Alternatively, the Akaike Information Criterion (AIC) (Akaike, 1974) was used to compare the likelihood of nonnested models.AIC is defined as -2lnL plus twice the number of estimated parameters, and the lowest AIC value is considered to be the best model.
Stepwise multiple regression was applied in order to detect concomitant variation that could confound genetic effects on infection.The independent variables used in the model were: cubic polynomial on age, sex and the interactions between them.Restricting to terms significant at the 5% level, only age in years showed a significant effect, according to the equation: Y = 0.141348 + 0.0073 (age); F (1,500) = 32.88.Thus, the segregation analysis was carried out using liability class values provided by this regression analysis (Table II).= 324.67-318.50 = 6.17,P = 0.046, respectively).A multifactorial component in addition to the major gene was not necessary to explain familial aggregation (model 5 vs. model 1: c 2 1 = 320.58-318.44 = 2.14, P = 0.144).Nevertheless, in addition to a major gene, a small multifactorial cause (H = 0.29) is suggested, when the AIC criterion is utilized.The data were compatible with a recessive or an additive mode of inheritance (model 6 vs. model 5: c 2 1 = 322.84-320.58 = 2.26, P = 0.133; model 7 vs. model 5: c 2 1 = 322.17-320.58 = 1.59;P = 0.207, respectively), while a dominant model was rejected (model 8 vs. model 5: c 2 1 = 327.81-320.58 = 7.23; P = 0.007).A model that allows estimation of taus does not differ significantly from the mixed model with fixed Mendelian transmission probabilities (c 2 3 = 2.24, P = 0.524).Also a model with equal transmission probabilities differs (with borderline significance) from the mixed model (c 2 2 = 5.99, P = 0.050).Unfortunately, several models involving an additive major gene could not be tested since the Mendelian mixed model, with d = 0.5, did not reach convergence.Therefore, the Mendelian recessive mixed model, with a small multifactorial component (H = 0.29) with a gene frequency of approximately 0.4, is the best model as indicated by the (AIC).III DISCUSSION Cabello et al. (1995) observed familial aggregation of infection in the same population of Jacobina.They investigated intrafamilial patterns of infection and found significant associations concerning father x offspring, mother x offspring, and sib x sib, in contrast with a nonsignificant parental association.The present study shows evidence of a major gene for human resistance/susceptibility to L. chagasi infection, with a gene frequency (q) around 0.4.The conditions of transmission probabilities to infer the presence of a major gene were satisfied.
The presence of a major gene was also suggested to exist for a similar parasite (L.peruviana) (Shaw et al., 1995).However, the genetic mechanism, presently suggested, should be taken as tentative, since problems of penetrance and dominance may confound the whole picture.The conclusions should be confirmed in other populations, as well as explored through both genetic linkage and genetic mapping starting on chromosome 2, which has a region homologous to the mouse chromosome 1.Since recent research reported that related traits (leprosy per se) may segregate linked to 2q markers, as a homologous Lsh gene should do (Abel et al., 1998;Sachdeva et al., studies on less environmentally influenced phenotypes, i.e., independent of a previous infection and less dependent on the prevailing environment conditions like the Mitsuda reaction, for instance (Feitosa et al., 1996), must have precedence, due to the existence of apparently conflicting linkage studies (Shaw et al., 1993;Levee et al., 1994;Newport et al., 1995).Other candidate gene regions have been suggested for human susceptibility in addition to NRAMP1 (cf.Blackwell, 1998) and must be investigated.

ACKNOWLEDGMENTS
Research supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico -CNPq and Fundação de Amparo à Pesquisa do Estado de São Paulo -FAPESP.Publication supported by FAPESP.

Table I -
Distribution of studied families, according to family size, from Jacobina, Bahia State, Brazil.

Table II -
Liability values to Leishmania chagasi infection according to specific age groups (in years).Leishmania infection was applied to the sample of 502 individuals in 94 nuclear families.The degree of dominance reached value zero for the Mendelian mixed model (TableIII), suggesting a recessive

Table III -
Segregation analysis of Leishmania chagasi infection.Genetics and Molecular Biology -Genetic causes involved in Leishman... http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1415-47571...The probabilities of being affected for the three genotypes following the nine age-liability classes were determined according to the most parsimonious models, for recessive and additive inheritance (TableIV, based model 6 of Table III and in Table V, based on model 7 of Table III).

Table IV -
Probability of Leishmania chagasi infection based on a model where the susceptibility phenotype is due to a recessive gene*.Data are based on model 6 in TableIII *

Table V -
Probability of Leishmania chagasi infection based on a model where the susceptibility phenotype is due to an additive gene*.
* Data are based on model 7, in Table