Mammalian Genetics Lecture 1

BIMS 808

Pearson-White, sp3i@virginia.edu, 2-0756

March 29, 2004

Triplet Repeat Disorders in human

Mouse genome

comparative mouse/human genome analysis

positional cloning, obese gene example in mice

congenic strains, speed congenics

Triplet repeat Disorders Exhibit Anticipation

disease increases in severity in successive generations
turns out to be due to expansion of the number of repeats at the unstable locus

Triplet repeat disorders

1. Fragile X - triplet CGG in FMR-1 gene, below example, in coding region
2. Myotonic dystrophy - triplet CTG in 3’ untranslated
3. Spinobulbar muscular atrophy (SBMA) - androgen receptor, triplet CAG in coding region
4. Huntington’s disease - triplet CAG in coding region
5. Spinocerebellar ataxia type 1 (SCA1) - triplet CAG
6. Machado-Hoseph disease (SCA3)
7. Dentatorubral and pallidolyusian atrophy (DRPLA)

Trinucleotide Repeats

Occur naturally in the genome
Are prone to deletions and insertions
By slipped strand mispairing
Above a certain threshold length they are very unstable
Once they are long, they expand with each generation
Size change depends on sex of parent and length of repeat

Mechanism of defect, 3 classes

1) CAG repeats within coding region
2) CGG repeats in non-coding
3) CTG in 3’ untranslated

CAGn repeats within coding region

Translated as polyglutamine tracts
Affect protein function, impairing the ability of the proteasome to elevate activity in response to stress
Do not affect transcription or translation
e.g. Huntington, Kennedy, Spino-cerebellar ataxia1
5 neurodegenerative diseases, neuron loss
Expanded polyglutamine tract protein expression is sufficient to cause defect, dominant

Insertion of human huntingtin with 120 polyglutamine repeats

CGGn repeats in non-coding

Can expand from 10-50 normal repeats up to hundreds or thousands
Expanded repeats affect DNA methylation and chromatin structure
Produce inducible chromosomal fragile sites, e.g. Fragile X

CTGn in 3’ untranslated region

Expands from 5-35 normal, up to 2000 in myotonic dystrophy
Number of repeats correlates with severity of disease
No evident effect on transcription or structure of the gene product, so mechanism unclear
e.g. myotonic dystrophy

Features of mammalian genes

Variation in gene density

Variation in size and exon content

Presence of pseudogenes

Gene densities in HLA region and dystrophin gene

Human genes vary in size and exon content

Clustered gene families, e.g. HLA class I, with non-processed pseudogenes and fragments

Pseudogenes - Nonprocessed

Nonprocessed (conventional) pseudogenes
- nonfunctional copies of the genomic DNA sequence
- may have inappropriate termination codons
- common in clustered gene families, thought to arise by tandem gene duplication
Expressed nonprocessed pseudogenes
- originate from gene duplication events, in which one copy picks up deleterious mutations and loses its original function
- before it has lost its ability to be expressed, there will be a transition period when it is expressed
  e.g. sigma-globin is expressed, but no evidence for its incorporation in functional hemoglobin
- Truncated genes and gene fragments
- often found in clustered gene families
- probably arise from unequal crossover or unequal sister chromatid exchanges
Truncated genes and gene fragments
- often found in clustered gene families
- probably arise from unequal crossover or unequal sister chromatid exchanges

Pseudogenes - Processed

Processed (also called RNA-copy) pseudogenes
- nonfunctional copies of the exon sequences of an active gene
- often found in interspersed gene families
- thought to arise by integration into chromosomes of a natural cDNA sequence generated by
- reverse transcriptase from an RNA transcript
- pol II-derived processed pseudogenes are normally not capable of expression due to lack of promoter sequences
- polII-derived processed pseudogenes such as Alu repeats can reach very high copy numbers
Expressed processed pseudogenes
- processed pseudogenes that are capable of expression because of location near promoter sequences
  e.g. testis-specific expression of pyruvate dehydrogenase gene PDHA2

Comparison of mouse and human genome organization and gene expression

genome size 3000 Mb in both
~35,000-50,000 genes in both
humans 22 pairs of autosomes, X and Y
mouse 19 pairs of autosomes, X and Y
humans metacentric or submetacentric chromosomes
mouse all acrocentric chromosomes
humans ~45,000 CpG islands
mouse ~37,000 CpG islands

Comparison of mouse and human orthologues

usually 70-90% sequence identity between genes in coding DNA, 75-95% aa sequence identity

ribosomal proteins, nuclear regulatory proteins most conserved

host defense ligands and receptors most diverged

Amino acid sequence divergence between human and rodent orthologues

Human and mouse proteins with identical amino acid sequences

Protein name	Coding nucleotide sequence identity (%)
Calmodulin	84.01
GTP-binding protein (RAB1)	95.47
GTP-binding protein (RAC)	91.36
Histone H3	86.13
Histone H4	89.74
Lupus autoantigen (snRNP)	87.50
Ribosomal protein L30	93.97
Skeletal muscle alpha-actin	90.74

The most divergent human and mouse proteins, examples:

Protein name	Aa identity (%)	Cds identity (%)
B7 B ly ag	41.11	60.74
Fas (APO-1)	49.54	66.46
Interferon beta	50.0	68.98
IFN-g Receptor	52.26	68.6
IL-6	41.43	61.77
IL-4	43.57	51.31
CD2 T ly ag	52.34	68.44

Comparison of mouse and human gene organization

synteny is subchromosomal but may be substantial on autosomes

synteny is almost complete on the X chromosome, except for inversions that alter gene order

Comparison of mouse and human gene expression

gene family organization very similar. some examples of gene duplications that have led to differences in gene number

e.g. 2 mouse ß-globin genes and 1 a-globin gene
but 1 human ß-globin gene and 2 a-globin genes

similar gene organization, intron positions often conserved

very similar gene expression, except some differences in the choice of alternative promoters and patterns of splicing between orthologues

some differences in imprinting patterns in autosomal genes

Conservation of orthologous human and mouse linkage groups is limited to subchromosomal regions

Advantage of mouse for genetic studies

short gestation period (19-20 days)

sexual maturity attained quickly (~6 weeks)

large litter sizes (~6-8 pups)

~4-8 litters per breeding female

life span in laboratory (~1.5-2.5 years)

availability of inbred strains

ability to perform controlled matings

outbred - e.g. CD-1

approximate human populations

easier to maintain

heterozygous vigor, often maintain mutants on vigorous outbred background so don’t weaken and die

inbred - e.g. C57Bl/6J

made by brother X sister matings for >20 generations

will approximate isogenicity

at each locus things are becoming homozygous

approach homozygosity, doesn’t reach 100%

decreasing heterozygosity, doesn’t reach zero

Inbred strains

approximates identical genetic backgrounds

variation seen in inbred background probably genetic in origin

good for comparison of groups - controlled experiments

pitfalls

not clones (not isogenic)

subline divergence, due to residual heterozygosity, mutation , or contamination
‘wild type’ is a per locus term

Different inbred strains have different characteristics

C57Bl/6J (B6) mice will consume 20-30 times as much morphine and 10X as much alcohol as DBA/2J (D2) mice

QTL analysis of F2 progeny from a B6XD2 intercross was used to map loci

3 QTL loci on chromosomes 1, 6, and 10 were identified that together explain 85% of the genetic variance in morphine consumption between B6 and D2 mice

2 QTL on chromosomes 2 (Alcp1 in males), and 11 (Alcp2 in females) for alcohol preference

Alcohol and morphine preferences were genetically distinct

More mouse strains

co-isogenic - e.g. C57Bl/6J-C²⁵/C²³

a single mutation on a defined inbred background, a nicely controlled situation

congenic - e.g. C57Bl/6J.AKR-H-2^k

strains backcrossed and selected to put your locus of interest in another inbred background

takes >3.5 years

recombinant inbred - e.g. B X D-n

combination of two inbred strains, backcrossed to segregate, then maintained as separate lines with inbreeding

used for typing for genetic markers, recombination

interspecific backcross strains - e.g. (B6 X SPRET/Ei) X B6

mapping

also mapping in human syntenic regions

Repeated DNA sequences in mouse

~ 8-10% of genome is repeated

five highly repeated DNA sequence families

MIF-1

EC1

Chromosomal location of major repetitive DNA classes

Location of Alu, LINE-1 and (A)_n/(T)_n repeats within Rb gene

Traditional mapping is difficult in mouse

methods rely on ability to discriminate chromosomes cytologically

mouse chromosomes all acrocentric
show a continuous gradient of sizes

somatic cell hybrids that carry single mouse chromosomes are rare

three point crosses are slow

Recombinant inbred strains (RI)

Obtained by crossing two inbred strains, making G1 heterozygous at every locus, then sib matings to segregate at each locus, then backcross to inbred strain, maintaining each as separate lines. In 4-5 years will fix recombination events in inbred background in each line.

BXD are a set of 26 lines derived by over 60 generations of inbreeding from the progeny of a C57BL//6J X DBA/2J cross. They provide unlimited supplies of a panel of chromosomes with fixed recombination points. DNA is available as a public resource, and the strains function rather like the CEPH families do for humans.

limited by low degree of allelic variation among parental laboratory mouse strains

New gene mapping methods led to explosion of mapping the mouse

RFLPs (Restriction fragment length polymorphisms)

PCR - STRP (short tandem repeat polymorphisms), STR, SSRP (simple sequence repeat polymorphisms), SSR, VNTR (variable number of tandem repeats)

interspecific crosses

PCR used to type short tandem repeat polymorphisms (STRPs)

Interspecific crosses

between laboratory strain and distantly related species of Mus

exploit genetic diversity between these strains

most genes or DNA sequences are polymorphic, and can thus be placed relative to other genes in a single cross

DNA from a single cross is sufficient to map thousands of genes by RFLP, tens of thousands by PCR

genes can be mapped simultaneously, gene order is easy to define, at least within a single cross

Pioneered by Francois Bonhomme, Philip Avner, and Jean-Louis Guenet in late 70’s and mid-80’s

now most genes mapped this way

Mus spretus

most genetically divergent

still interbreeds with common lab mice, producing at least one sex (female) that is fertile

Drawbacks

F1 males are sterile, so only females from F1 mice can be used as parents in backcross. this prevents the study of male meiosis
the wide divergence may have permitted the accumulation of small chromosomal inversions that could suppress recombination and possibly hamper fine-structure genetic mapping used in positional cloning

now using Mus musculus castaneus or Mus musculus molossinus, in which both sexes are fertile in the F1

B6 and spretus

Interspecific backcross experimental design

Interspecific backcross PCR data

(scroll down to see all figures)

BSB panel BSB and BSS map, mouse chromosome 3

Mouse resources

http://lena.jax.org/resources/documents/cmdata/index.html

http://www.informatics.jax.org/

http://www.rodentia.com/wmc/index.html

The Whole Mouse Catalog

Anchor loci

anchor loci are sets of PCR primers that have been well characterized and mapped and defined in many backcross mouse panels

to combine mapping data from different crosses and/or different laboratories, now include a common set of anchor loci among probes mapped in each cross

mapping data can then be combined with respect to the anchor loci

anchor loci are spaced evenly every 10 to 20 centimorgans (cM)

highly polymorphic even within intraspecific crosses

easily typed by PCR

well mapped in interspecific crosses

drawback:

the order of loci between anchors can be only indirectly inferred on the basis of distances
such inferences can be unreliable because of variation in recombination frequencies among crosses

Mouse Genetic Map

mutations that cause phenotypic deviations

mapped throughout the century

up-to-date maps published

biologically interesting genes

but, no light shed on biochemical basis of the defect

isozyme loci

cloned genes

important biological information

useful in comparative mapping relating mouse and human and other genomes

but, can be tedious to genotype and may not be polymorphic in crosses between related strains

highly polymorphic anonymous DNA sequences

minisatellites

microsatellites

single-strand conformation polymorphisms

often polymorphic between closely related strains, informative

many can be rapidly mapped using PCR

but, little biological information

can now map a new gene in the mouse and predict its location in the human more easily than can map directly in humans

SSCP-single strand conformation polymorphism

a method to identify mutations in short stretches of DNA

PCR amplify short stretches of DNA

Denature

altered electrophoretic mobility of mutant compared with wild type sequences on non-denaturing thin acrylamide gels
- can add glycerol to the gel to facilitate detection of some mutants

For heteroduplex analysis, allow some of each DNA to reanneal, run on gel

Some mutant DNA remains single stranded, retarded on gel

CF mutation screening by heteroduplex and SSCP analysis (same as above figure)

Positional cloning of the mouse obese gene its human homologue

5 single gene mutations have been described in mice that lead to obesity

ob gene was the first, in 1950

ob leads to morbid obesity and type II diabetes

Positional cloning, 1

ob mapped proximal to Microphthalmia (Mi) and waved-1 (wav-1) loci on proximal mouse chromosome 6

then positioned relative to molecular markers on mouse chromosome 6

then mapped close to a RFLP marker D6Rck 13

closely linked to Pax4

used interspecific and intersubspecific mouse crosses, a total of 835 informative meioses that were segregating ob

Positional cloning, 2

Pax4 was mapped proximal to ob and was recombinant in two animals (111 and 420 in Fig. 1)

no recombination was detected among the first 835 meioses scored between ob and D6Rck 13

isolated and characterized YACs corresponding to Pax4 and D6Rck 13

centromeric and telomeric ends were recovered, and ends mapping closer to ob used to screen for new YACs

YAC ends were recovered using vectorette PCR, or the plasmid end-rescue technique

Positional cloning, 3

one of the ends (labeled 1 in Fig 1) of a D6Rck 13 YAC, 902A0653, was recombinant in animal 257, positioning ob between this YAC end and Pax4

had a ~70 kb gap between YAC ends, bridged with P1 clones

then had a 2.2 Mb contig, ob localized to 900 kb, then 650 kb

Phage P1 vector system: cloning up to 100 kb

Positional cloning, 4

then genotyped an additional 771 meioses from a C57BL/6 ob/ob X DBA/2J intercross and backcross

made SSLP’s for D6Rck13 and Pax4 by sequencing and finding microsatellite sequences
then made primers flanking these microsatellites
found no additional recombinants between ob and Pax4
found one animal recombinant between D6Rck13 and ob
there was a 3-fold lower rate of recombination in this region

Gene identification

isolated genes from the 650 kb region using exon trapping

genomic DNA is subcloned into a splicing plasmid, in a multiple cloning site flanked by a splice donor and a splice acceptor element
DNA isolated from transformants is transfected into COS cells, which are grown and RNA is isolated

RT-PCR is performed on that RNA using primers from the splicing vector
products are cut with BstXI to eliminate sequences that do not contain exons

secondary PCR amplification is performed, PCR products are cloned, colonies are screened using PCR and DNA sequencing, then compared to sequences in the Genbank database
putative exons are screened for the presence of RNA from various tissues using Northern blots and RT-PCR

Exon trapping using the pSPL3 vector

ob gene

six genes were identified, 4 within the region and 2 outside

one of the trapped exons, 2G7 was white fat-specifically expressed

ob product

Ob encodes a 4.5 kb mRNA with a highly conserved 167 aa open reading frame (ORF) and a very long 3’ UT

has features of secreted protein, signal sequence

protein 84% identical between mouse and human

3’ and 5’ untranslated regions only 30% identical

zoo blot shows conservation in all vertebrate species

Cloned and sequenced mouse ob gene, then mutant versions

found nonsense mutation in the original ob mouse strain gene at codon 105

expressed at 20X higher levels in C57BL/6 ob/ob mice, suggesting that the gene product was nonfunctional and that the mRNA was increased as part of a feedback loop

a second ob mutant strain (SM/Ckc- +^Dacob^2J/ob^2J)does not synthesize any ob mRNA, suggesting that ob gene product may function as part of a signaling pathway from adipose tissue that acts to regulate the size of the body fat deposit

Knockout phenotype depends on the genetic background

EGFR

On CF-1 background: perimplantation death due to ICM degeneration
On 129Sv background: midgestation death due to placental defects
On CD-1 or mixed background: mice lived 3 weeks, abnormal skin, kidney, brain, liver and GI tract

p53

p53 -/- mice get malignant lymphomas in 75% B6, 25% 129/Sv, common in B6

p53 -/- mice get malignant lymphomas in 129/Sv, and teratocarcinomas, common in 129/Sv, with earlier onset

129/Sv background is itself mixed

Backcrossing to place the mutation on an inbred, homogeneous background

congenic breeding strategies

3-4 years to get 99% recipient strain

speed congenic = MASP, marker-assisted selection protocol

1-2years to get 99% recipient strain

supersonic congenic

superovulation and embryo transfer combined with MASP to accelerate to 8 months

Backcross generations

Not all animals are equally heterozygous

So can select animals with less heterozygosity for further breeding, to get to the desired homozygous inbred strain faster.

Selection of animals with less heterozygosity at each generation decreases the length of time it takes to get to inbred background
This is most efficient if many animals are produced and tested, especially at the N2 and N3 generations
Fewer genotype determinations need to be performed in later generations, as reduced numbers of segments from the donor are present

Lessons learned from the Human Genome Project

Information is power
Collective action is powerful
Instead of individual laboratories
High volume sequencing is valuable, e.g. evolutionary biology

Promises kept by the Human Genome Project

The power of making connections
- e.g. fused mouse mutant on shelf since 1937
- Mapped with markers
- Transgene insertion with similar phenotype mapped to the same location
- YAC used to clone fused gene
The power of developing model organisms
The freedom to do biology
Escape from cloning and sequencing
Get to functional biology