L13: Identifying Human Disease Genes

Question 1

Genetic Based Gene Hunting

Answer 1

Identify 1/more sets of related patients
Connect patients into pedigrees (mode of inheritance important)
Genotype patients and their families
Look for polymorphisms that co-segregate with the disease
When linkage is found, narrow the region with more families →Look for candidate genes (e.g., by computer prediction)
Screen genes for mutations
Prove mutations are causal (proof standards vary)

Question 2

Classical Positional Cloning

Answer 2

Difficult approach
Heavily dependent on size of candidate region and availability of either:
1) patients with visible deletions;
2) good family collection for genetic mapping;

Duchenne myotrophy, cystic fibrosis familial colorectal cancer, polycystic disease genes were cloned using this approach

Question 3

How to Find a Gene

Answer 3

1. Locate chromosomal area associated with or deleted in patients with particular syndrome
2. Screen hundreds of patients to find a “lucky” small deletion or minimal region associated
3. Make a cosmid (PAC) contig OR Retrieve a genomic sequence
4. Map all genes-candidates (experimentally or in silico)
5. Check all candidates (try mutations, expression in cell lines, or knockout in mice)
6. Study gene functions in vivo

Question 4

Strategies

Answer 4

1. Human genome and transcript maps
2. Mouse and other models

Question 5

EST–expressed sequence tag

Answer 5

Partial DNA sequence (“single-pass”) of a cDNA clone
Largest and fastest growing division of GenBank
Derived from some specific RNA source; source is part of record and can be researched

Question 6

UniGene

Answer 6

UniGene (unique gene) goal is to create 1 unique entry for each gene and collect all of the ESTs associated with that gene
UniGene cluster can have from 1 to several hundred ESTs created by similarity searching
Used in development of commercial microarrays (Affymetrix)

Question 7

Genes from candidate region must be prioritized from mutation testing

Answer 7

1) Appropriate expression pattern (at least 1 gene should be expressed in place where pathology is seen)
2) Appropriate function (e.g. ion channel protein in the deafness, as other ion channels are involved)
3) Homology to paralogous gene (Wilson disease – Menkes disease)
4) Homology to orthologous gene (mouse models)

Question 8

Mouse-human homology (syntheny)

Answer 8

1. Mouse genome is ~14% smaller than human genome. The difference probably reflects a higher rate of deletion in mouse.
2. Over 90% of mouse and human genomes can be partitioned into corresponding regions of conserved synteny (with gene order preserved)
3. At the nucleotide level, 40% of human genome can be aligned with mouse genome. The rest was probably deleted in one or both genomes.
4. The proportion of mouse genes with a single identifiable orthologue in human genome is 80%. The proportion of mouse genes w/o any homologue currently detectable in human genome (and vice versa) is < 1%.

Question 9

Confirming Candidate Gene

Answer 9

1) Mutation screening
2) Restoration of expression = restoration of normal phenotype in vitro (cancer)
3) Production of mouse model for disease (Knockout or transgenic)

Question 10

Mutation screening of Rett Syndrome

Answer 10

Rett syndrome: autistic behaviour, ataxia, repetitive hand movements, mostly seen in girls

Combines methylated CpG islands and transcription repression domains -mainly missense and nonsense mutations

Question 11

Restoration of Phenotype

Answer 11

Shaker-2 mouse: deaf –causes circling and head tossing
Cross this mouse type with wild type (maps sh2 onto chromosome 11 which corresponds to candidate region of DFNB3 deafness gene in humans)
Isolate BAC clones and inject into sh2/sh2 fertilized eggs to create transgenic mice
BAC 425 p24 correct sh2 defect
Computer analysis identifies novel myosin gene (myo15)
Myo15 mutation identified in sh2 mouse
Isolate human Myo15 gene and use somatic cell hybrids to check it maps
Myo15 mutations identified in 3 unrelated patients

Question 12

Comparative Genomic Hybridization

Answer 12

1) Extract DNA from cells from control and from patient
2) Label w/ 2 diff fluorochromes
3) Mix in equal quantities, hybridize to microarray of clones
- Red: from control; green=from patient
- if red, not involved in the disease
- if yellow, intermediate

Question 13

Identify Multiple Sulfatase Deficiency Gene
2 methods:

Answer 13

Biochemical approach:
1) Extract of bovine testis: created functional assay
2) Highly enriched enzyme prep: purify sample and complete mass spec
3) Partial peptide seq: database search for cDNA that could encode these peptides
4) Find candidate cDNA: design primer, test for mutations in patients
5) Mutation in SUMF1 identified in patients

Model approach:
1) Panel of mouse-human hybrid cells: treat w mitotic inhibitor cytochalasin B
2) Panel of micro cells (contain only 1 human chromosome): fuse each with cell from patients with deficiency
3) Human chromosome 3 can correct defect: repeat procedure with mouse-human hybrids containing a fragment of human chromosome 3
4) Found a specific region: identify genes in this region
5) 7 candidates genes: screen patients with the deficiency
6) Mutation in SUMF1 identified in patients

Question 14

How BRCA1 gene was found

Answer 14

1500 families and massive analysis
Create genetic susceptibility model
23 selected multi-case families
LOD score analysis
Identified candidate region lead to positional cloning and mutation analysis

Question 15

Identifying Human Disease Genes Model

Answer 15

Model organism: Mapped candidate homology–>candidate chromosomal region in databases–>collect families for mapping–>genetic mapping
Database searching: Candidate chromosomal region–>check databases for genes–>possible candidate gene–>fully characterized?–>collect unrelated patients–>look for mutations