Behavioral Genetics Topic #12

1

Behavioral GeneticsTopic #12

Animal Models of Alcoholism

Advantages of Animal Models

• Precise control over environmental exposure

• Manipulation of genetic risk

• Explore component phenotypes and their interrelationships

OUTLINE• Component Phenotypes:

– Alcohol Preference -- Inbred strain comparisons

– Alcohol Sensitivity -- Selection studies• Forward Genetics

– QTL Analysis - Withdrawal Severity – Random mutation screens

• Backward Genetics– Targeted mutations (transgenic, null-mutants)

2

Preference/Avidity

• % of times in 14-day period animal selects 10% ethanol solution vs. tap water (both a sweetened with saccharin)

• Marked differences, 0-80%

• Is it heritable? – Inbred strain comparisons

Inbred Strain Comparisons

• Inbreeding – mating between genetically related individuals. Increases homozygosityat every loci

• Inbred Strains – At least 20 generations of brother-sister mating (at least 98% of segregating loci are fixed)

3

Alcohol Preference Comparisons

4

Classical Analysis

P1 P2

F1

F2

Sensitivity

Physiological, biochemical, and behavioral response to ethanol in a drug-naïve animal.

Is it heritable? – Selection studies

-4 -3 -2 -1 0 1 2 3 4

-4 -3 -2 -1 0 1 2 3 4

70”

Parents

Children

Kids’ Mean

72”

Response to Selection = R = Kids’ Mean – Original Mean

Original Mean

5

Realized Heritability

S = Selection differential (mean of parents –mean of original population)

R = response to selection (gain=mean of children – mean of original population) =

h2S

Realized heritability = h2 = R/S

Response to Selection

Genetic Correlation/Pleiotropy

G2

G3

G1

MediatingSystem

BehavioralPhenotype

6

Genetic Correlation

Selection concentrates gene variants underlying, e.g., high sensitivity in the selected line.

Therefore, correlated responses to selectionidentify genetically correlated traits that may explicate the mechanism of genetic influence.

Alcohol Metabolism

ETOH Acetaldehyde Acetate

ADH ALDH

Neurobiological Basis for ETOH Sensitivity?

• LS mice are hyper-sensitive to a wide range of hypnotic drugs including:– Benzodiazepines– Barbituratesγ-amniobutyric acid Type A Receptors (GABA receptors)

7

Correlated Responses

Inbred Strain Comparisons

287 mg/kg430 mg/kgBrAC at righting

138 minutes38 minutesSleep Times

BALB/CC57BL

Forward Genetic Approaches(Phenotype-Driven)

• Top down approach (i.e., from phenotype to genotype) to identify relevant genes– Quantitative Trait Loci (QTL) Analysis– Mutation Screens – Microarray analysis

8

Withdrawal Severity• 72 hour exposure to ethanol vapor and

treatment with pyrazole

• 25 hour withdrawal period where relevant withdrawal symptoms are measured

• Genetic analysis:– Response to selection (realized h2 = .28)– Not correlated to sensitivity– Moderate correlation with preference

QTL Analysis• Linkage analysis in non-human animal

species

• Based on Recombinant Inbred Strains– Inbred strains that are created from the crossing

of two progenitor inbred strains. Each RI strain represents a unique recombination of the alleles that differed in the two parental strains.

• BxD RI derived from C57BL x DBA

BXD Recombinant Inbred Lines

C57BL/6J00

DBA/2J11

F101

F225% - 0050% - 0125% - 11 20 generations

of B-S mating

26 RI Strains

9

Buck et al. (1997)

Replications

Linkage to 1, 4 & 11 have been most consistently reported and have alsobeen implicated in QTL analysis of pentobarbital withdrawal.

Buck & Finn (2000). Addiction, 96: 139-149.

10

Significance of QTL

• Identifies a region in which to look for genes– 1 cM (~ 10-12 genes) is probably limit of

resolution– Expression studies may reduce to 5-6 genes– Sequence analysis would reduce further

• Implicates gene regions in the human

Importance of the Mouse Genome

• Mouse genome (Nature, December 5, 2002):– 2.5Gb– ~27,000 – 30,500 genes

• Relationship to human genome:– ~99% of mouse genes have counterparts (orthologs) in

human– ~96% of human genes have orthologs in mouse– Conservation of some non-coding regions– Synteny

Mouse Genome Sequencing Consortuim (2002). Initial Sequencing and comparative cnalysis of the mouse genome. Nature, 420: 520-562.

11

QTL for Withdrawal Severity

2p24-q37/11p13

Glutamic acid decarboxylase

2

5q32-q35GABAA12%11

9p21-p23/1p32-22.1

26%4

1q21-q3226%1

HumanSynteny

Candidate Genes

% Genetic Variance

Mouse QTL

QTL Methodology• Strengths:

– Systematic and relatively easily implementableapproach to identify linkage for natural variants

– Can help to implicate specific candidate genes and syntenic regions in human genome

• Limitations:– Linkage identifies regions; limited resolution – Not all genetic variants exist in parent stocks– Will not inform us about invariant genes

Random Mutagensis

• Expose a large number (thousands) of male mice to a chemical mutagen (ethylnitrosourea, ENU, produces point mutation rates at about 1/1000)

• Bred with unexposed females to produce many thousands of F1 progeny. Screen F1 for dominant effects (further breeding is needed to screen for recessive effects)

• Confirm outlier is transmitted by further breeding

12

Random Mutagenesis

• Strengths:– Both monomorphic and polymorphic genes– Focusing on large-effect mutations makes gene

identification easier– In theory, greater likelihood of gene identification in

complex systems• Limitations:

– Less powerful when there is large background phenotypic variability due to genes or environ

– Need efficient, sensitive and reliable behavioral screens

Forward Genetic Approaches(Phenotype-Driven)

• Top down approach (i.e., from phenotype to genotype) to identify relevant genes– Quantitative Trait Loci (QTL) Analysis– Mutation Screens – Microarray analysis

Microarray Analysis• Identify gene expression differences (mRNA) from two

samples simultaneously on 1000s of genes

• Human Example: Mayfield et al. (J Neurochem, 2002) compared expression of 10,000 genes in post-mortem brains of alcoholics versus controls found 191 that had at least 1.4-fold difference in expression

• Animal Example: McBride at al. (2002, Alc: Clin & Exp Research) compared expression of more than 3906 genes in hippocampus of P and NP rats; found 25 that differed in expression at p < .0001.

13

Mayfield et al. (2002). Patterns of gene expression are altered in thefrontal and mortor cortices of human alcoholics. J Neurochem, 81: 802-813.

Reverse Genetics(Genotype-Driven)

• Bottom up approach (i.e., from genotype to phenotype) to determine gene effects– Transgenic:

– Knock-out (null mutant):

Transgenics

• Animal where foreign DNA has been stably incorporated into their germ line

14

Overexpression of Dopamine D2 Receptors (DRD2)

(Thanos et al. (2001), J of Neurochem, 78: 1094-1103)

• Low levels of DRD2 appear to be involved in reinforcing effects of alcohol– Reduced reinforcement in KO mouse– Alcoholics have reduced levels of DRD2– Rats selected for ETOH preference have low DRD2

• Thanos et al. experimentally increased expression of DRD2 in (NAc brain region of) rats– Results in sharp reduction in ETOH preference

Thanos et al (2001)

Null-Mutants (Knock out)

• Targeted inactivation of a specific gene

• Conditional knock-outs

15

Behavioral Examples• 5-HT1B Receptor Knock-Outs:

– Saudou et al (Science, 1994) – intruder aggression– Crabbe et al. (Nature Genetics, 1996) – ethanol

consumption and sensitivity• 5-HTT Knock-Outs:

– Kelai et al (Science, 1994) – reduced alcohol consumption

• D4R Knock-outs:– Dulawa et al. (J. Neuroscience, 1999) less responsive to

novelty– Rubinstein et al. (Cell, 1999) – more responsive to

activating effects of ETOH, cocaine, methamphetamine

Limitations of the Standard Knock-out Procedure

• Effects of deleting gene expression may depend on genetic background

• Compensatory mechanisms

• Cannot necessarily extrapolate from absence of gene to consequences of varying level of gene expression

Gene-Environment Interaction

Bennett et al. (2002). Molecular Psychiatry, 7: 118-122.

16

Barr et al. (2004). Archives of General Psychiatry, 61: 1146-1152

Summary• Classical methods for establishing heritability

– Inbred strains– Selective breeding

• Genetic Correlation– Correlated response to selection– Inbred strain correlations

• Preference and Sensitivity are heritable and inversely associated (at least for the depressing effects)

• Metabolism is not assoc with sleep time but is with preference

• Withdrawal susceptibility is largely a genetically independent phenotype

Summary• Forward Genetic Approaches (phenotype-driven)

– QTL Analysis Candidate genes & candidate regions

– Mutation Screens– Microarray analyses

• Reverse Genetic Approaches (genotype-driven)– Transgenics– Knockouts

• Genotype-environment interaction