Linkagebased genome screens

The affected sib pair method, a variation on that originally described by Penrose (1953), is most frequently used for linkage analysis of multifactorial diseases such as tuberculosis. The families required include two full siblings who have both had disease and preferably also their parents (who are preferably unaffected). As shown in Fig. 1, the inheritance of a highly polymorphic marker can be traced

FIG. 1. Affected sib pair identical by descent analysis. The numbers below each family member represent alleles at a marker locus. In the first family, both offspring inherit allele 1 from the father and allele 3 from the mother and thus share two alleles identical by descent. In the second family they have inherited the same paternal allele but different maternal alleles and so share one allele identical by descent. In the third family they have inherited different alleles from both parents and thus share zero alleles identical by descent. If a large number of families are typed, the expected 2:1:0 sharing frequencies are 25% : 50% : 25%, assuming no linkage between the disease and marker.

FIG. 1. Affected sib pair identical by descent analysis. The numbers below each family member represent alleles at a marker locus. In the first family, both offspring inherit allele 1 from the father and allele 3 from the mother and thus share two alleles identical by descent. In the second family they have inherited the same paternal allele but different maternal alleles and so share one allele identical by descent. In the third family they have inherited different alleles from both parents and thus share zero alleles identical by descent. If a large number of families are typed, the expected 2:1:0 sharing frequencies are 25% : 50% : 25%, assuming no linkage between the disease and marker.

from parents to offspring and the number of alleles that the sibs share (i.e. zero, one or two), identical by descent, can be determined. If the marker is not linked to a tuberculosis susceptibility locus then the sibs should share two alleles in 25% of families, one allele in 50% of families and zero alleles in 25%. If the marker locus is linked to a tuberculosis susceptibility locus then there should be a significant excess of families where sibs share two alleles identical by descent and a reduction in the number sharing zero alleles compared to the expected frequencies. This methodology does not make any assumptions about penetrance of genes, mode of inheritance or number of genes involved.

Approximately 100 sib pair families are generally used to carry out a genome screen on a multifactorial disease, and around 300 microsatellite markers, spaced at approximately 10 cM intervals must be typed. Microsatellites are dinucleotide repeats of a (CA) n sequence. They are highly polymorphic because at a single locus the size of n varies throughout the population. Many thousands of microsatellites have now been isolated, producing a comprehensive map of the entire human genome (Dib et al 1996). Advances in utilizing fluorescence-based typing systems and advanced computer software have allowed semi-automated microsatellite genotyping systems to develop, revolutionizing linkage studies (Reed et al 1994). PCR primers are labelled with one of three fluorescent dyes, FAMTM (blue), HEX™ (yellow) or TETtm (green), and microsatellite PCR products are separated on the basis of size by polyacrylamide gel electrophoresis. A laser is used to detect dye fluorescence as the PCR products migrate through the gel. By utilizing a TAMRATM (red)-labelled internal size standard the size of the microsatellite PCR

U) Biosystems

A Division of Perkin-Elmer

Plots - tbtempset7 Licensed to adrian hill, wtcg

-Ju^JUWLJU

100 150 200

01-Sample File 1 Blue 33.1

1500 1000 -500

Id7s519| ICFTRl

—1 i i i i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—;—i—i—i—i—i—i—i—i—i—i—r

298 300 302 304 306 308 310 312 314 316 318 320 322 324 326 328 330 332 14 3.1-Father 14 Blue 3.1-Father

|313 27| 1317.331

|313 27| 1317.331

900 600 300

15 3.2-Mother 15 Blue 3.2-Mother

1313.091

1313.091

2000 1500 - 1000 "500

1000 - 500

-1000 -500

FIG. 2. Seven microsatellite loci each labelled with FAM™ and run on a single lane of an ABI373 sequencer™ (Perkin-Elmer). The Genotyper™ (Perkin-Elmer) software is used to focus on the inheritance of alleles from a single microsatellite locus in one sib pair family. The offspring have inherited the same paternal allele but as the mother is a homozygote it cannot be determined whether or not they have inherited the same maternal allele.

products can be accurately calculated to less than 0.5 bp. Using sets of microsatellite PCR products that do not overlap in size, it is possible to run eight loci labelled with each colour in a single gel lane, i.e. 24 loci can be simultaneously analysed per lane, producing up to 864 genotypes per gel (Reed et al 1994). Figure 2 shows all the microsatellites, labelled with a single fluorescent dye run in one lane and how the Genotyper™ (Perkin-Elmer) computer software can be used to focus on a single locus and compare genotypes in several family members.

We have used this technology to carry out a tuberculosis genome screen on 282 microsatellite markers on 92 sib pair families from The Gambia and South Africa. Five of these markers showed significant evidence of co-segregation with tuberculosis (Table 1). One of these markers, d6s276, is located at the major histocompatibility complex on chromosome 6p. The absence of strong linkage to a single marker demonstrates that tuberculosis susceptibility is not a monogenic trait and that a large number of genes are probably involved. Due to the large number of microsatellites typed, false positive linkages to regions that do not include a tuberculosis susceptibility gene could have arisen. To overcome this problem we are conducting this genome screen in two stages. Having screened the entire genome in our first 92 sibling pairs we now intend to genotype the five markers that demonstrate evidence of linkage in a second set of families, which we are currently recruiting from Africa. This approach minimizes the risk of false positive linkages.

Strong evidence of linkage between microsatellite markers and disease susceptibility is only the first step in identifying the gene(s) of interest. Even when association is present so that the gene can be mapped to a small chromosomal region, the task can still be considerable. This will become easier as the Human Genome Project produces more high resolution genetic and physical maps. A genome-wide map of expressed sequence tags (ESTs) should facilitate the isolation of genes from the region(s) of interest. Hopefully, this approach will eventually lead to the identification of the genes that exert the largest effects on host variability in tuberculosis susceptibility. However, this approach will be likely to miss genes that exert a moderate effect on risk of tuberculosis, and therefore any serious attempt to identify tuberculosis susceptibility genes should also utilize both a linkage-based genome screen and a candidate gene approach.

TABLE 1 Genome screen for tuberculosis

Marker

2:1:0

1:0

Pvalue

d3sl262

24.1/24.4/9.5

84.9/52.8

0.0024

dl5sl28

21.0/28.9/8.6

87.3/59.3

0.0052

d6s276

17.7/25.2/8.5

79.5/55.9

0.0098

d8s272

19.2/29.6/10

83.9/65.6

0.0160

dxs984

0.0025

The five markers that show significant co-segregation with tuberculosis are shown. 2:1:0 sharing refers to the number of sib pairs sharing two, one or zero alleles identical by descent. 1 : 0 refers to whether each allele was co-inherited or not, and therefore includes families where one parent is a homozygote. The values shown are calculated by the program 'sib pair' (Delepine et al 1997) and are not whole integers, as probabilities are inferred when a parental genotype is not available. Genotype sharing frequencies are not comparable for the X chromosome and are therefore not stated.

The five markers that show significant co-segregation with tuberculosis are shown. 2:1:0 sharing refers to the number of sib pairs sharing two, one or zero alleles identical by descent. 1 : 0 refers to whether each allele was co-inherited or not, and therefore includes families where one parent is a homozygote. The values shown are calculated by the program 'sib pair' (Delepine et al 1997) and are not whole integers, as probabilities are inferred when a parental genotype is not available. Genotype sharing frequencies are not comparable for the X chromosome and are therefore not stated.

0 0

Post a comment