Mice, chimpanzees and the molecular basis of speech Wolfgang Enard - - PowerPoint PPT Presentation

Mice, chimpanzees and the molecular basis

• f speech

Wolfgang Enard Max-Planck Institute for Evolutionary Anthropology, Leipzig, Germany

Who is the closest ?

Edward Tyson, Orang-outan, sive Homo sylvestris: or the anatomy of a pygmie compared with that of a monkey, an ape and a man, 1699

„it is quite certain that the Ape which most nearly approaches man, in the totality of its organization, is either the Chimpanzee or the Gorilla.“

Thomas Huxley, Evidence as to man's place in nature, 1863

mutation C=>T

mutation C=>T Sequence DNA

Individual 1: AGTTACCATGACTAGACTAGCTGAAGGG Individual 2: AGTTACCATGACTAGACTAGCTGAAGGG Individual 3: AGTTACCATGACTAGACTAGCTGAAGGG GATCCGATCGACTTTTACATTAGCTACGACTACGACTACGAT T GATCCGATCGACTTTTACATTAGCTATGACTACGACTACGAT T GATCCGATCGACTTTTACATTAGCTATGACTACGACTACGAT T GATCGATTATGCTTGTAAACTTACAGCATCGCATACGACTAC A GATCGATTATGCTTGTAAACTTACAGCATCGCATACGACTAC

mutation C=>T

mutation C=>T

mutation C=>T

Human Chimpanzee

1.2 %

Human Chimpanzee

1.2 %

Human Gorilla

1.7 %

Chimpanzee Gorilla

1.7 %

Human Orangutan

3.3 % 3.3 %

Chimpanzee

3.3 %

Gorilla Orangutan Orangutan

3.3 3.3 3.3 3.3 1.7 1.7 1.7 1.2 1.2 0.4

Genome-wide average of point substitutions (%)

Data from Burgress et al, 2008; Caswell et al., 2008

Chimpanzee Bonobo Human Gorilla Orang-Utan time

0.6 % 0.85 % 1.7 %

Average DNA tree

Chimpanzee Bonobo Human Gorilla Orang-Utan time

5-6 mya 7-9 mya 14-18 my

Average DNA tree

Chimpanzee Bonobo Human Gorilla Orang-Utan

Particular DNA tree

Chimpanzee Bonobo Human Gorilla Orang-Utan

Particular DNA tree

Chimpanzee Bonobo Human Gorilla Orang-Utan

Particular DNA tree

Chimpanzee Bonobo Human Gorilla Orang-Utan

DNA tree – Species tree

4 mya

Human 1 Human 2

0.08 %

Average DNA tree

0.08-0.1 %

Average DNA tree

Particular DNA tree

Particular DNA tree

Pääbo, 2003

Human genome as mosaic

Our closest extinct relatives

Green et al. Science, 2010

0.15% 600-800 kya

Average DNA tree

Reference genome French Chinese Mbuti San Neanderthal Chimpanzee

Average DNA tree

Species split

0.3-0.4 mya

Human-specific changes

~6 million years or ~6 substitutions each 1000 bases

Linguistics Bernard Comrie Developmental Psychology Michael Tomasello Primatology Christophe Boesch Palaeoanthropology Jean-Jacques Hublin Evolutionary Genetics Svante Pääbo

Max-Planck-Institut für evolutionäre Anthropologie, Leipzig

genome

human chimpanzee

phenotype

transcripts

genome

methylation proteins

phenotype

transcripts methylation proteins enviroment enviroment

Genomic approaches

genome

human chimpanzee

genome

• ~35 Million nucleotide substitutions
• ~5 Million insertions/deletions of DNA
• hundreds of various rearrangements
• ~60,000 amino acid substitutions
• 2 amino acid differences per protein (median)

Genomic Differences

0.1 0.2 0.3 0.4 0.5 0.6

all (13438) perception of smell (41) spermatogenesis (80) immune response (534) neurogenesis (292) mRNA processing (99) Wingless pathway (68)

Ka/Ks

Gene differences among categories

Lymphocyte activation Immune cell activation 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Primates ka/ks Rodents ka/ks

Spermatogenesis

Gene differences compared to rodents

Chimpanzee Ka/Ks Human Ka/Ks 0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4

Gene changes in humans and chimpanzees

• ~20 million genetic mutations
• Overall typical patterns of

primate/mammalian gene evolution

Genomic approaches

Gene A Gene a

human chimpanzee

phenotype A phenotype a

Model organisms Human disease variants

Candidate gene approaches

• AIDS-susceptibility
• Malaria-susceptibility
• Alzheimer-susceptibility
• Doubled maximal lifespan
• Bipedal walking
• Larger brain/body size
• Language
• Speech/vocal learning

Human-specific changes in the phenotype

Pedigree of the KE-family

Hurst et al. (1990) Dev Med Child Neurol 32, 352-355

male female affected non-affected

The KE family

“Aa...y me aar Shuar, ah ist Bedfond. Aa me ah six.“ My name is Stewart and I live in Bedfont. I am six “D’i nam ist Zoe and’ dyl liv how-woh, I…d’i…d’a…d’i…how- woh – six” My name is Zoe and I live in Hounslow and I am six “Ah i mi-tine. I liv in Ea-ling. I am d’ine.” I am Christine I live in Ealing. I am nine “M’nam i Carol. I live in Ea-ling and I am a-leven.” My name is Carol. I live in Ealing and I am eleven.

Transcription provided by Prof. Faraneh Vargha-Khadem

Phenotype of affected members

• Developmental verbal dyspraxia
• problems with the processing of complex words

and non-words

• impaired orofacial movements
• affected in receptive and expressive verbal

abilities

• MRI studies show a smaller caudate nucleus
• PET scans show a disturbed activation during

language processing Vargha-Khadem lab

Liegeois et al., Nature Neuroscience, 2003

FOXP2

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga

exon (200 bp) intron (10,000 bp)

Chromosomal breakpoint, unrel. patient R553H mutation in KE family Lai et al (2001) Nature 413, 519–523 MacDermot et al (2005), Am.J.H.G. 76, 1074

PolyQ

STOP mutation, unrel. family

Two functional copies of FOXP2 are needed for normal speech and language development

polyQ FOX polyQ FOX polyQ FOX polyQ FOX

human chimpanzee

phenotype a

language/ speech

phenotype A

No language/ speech

FOXP2

Human FOXP2

Mutations of FOXP2 in the KE family

• Chimp. FOXP2

?

Molecular evolution of FOXP2

Molly Przeworski, Victor Wiebe, Svante Pääbo

Enard et al., (2002), Molecular evolution of FOXP2, a gene involved in speech and language. Nature 418, 869-872

Affected KE family member polyQ FOX H polyQ FOX R polyQ FOX R polyQ FOX R Normal / unaffected family member

FOXP2

polyQ FOX R polyQ FOX R S D N N E T human mouse

FOXP2

140 – 220 million years

FOXP2 protein evolution

99,6 % identity

3 substitutions / 715 amino acids

D..N..S E..T..N

D..N..S E..T..N D..T..N 140 – 220 million years

T=>N; N=>S D=>E

FOXP2 protein evolution

A..D..N..S V..D..T..N A..E..T..N A..D..T..N A..D..T..N A..D..T..N 140 – 220 million years

T=>N; N=>S D=>E A=>V

FOXP2 protein evolution

• FOXP2 is the only gene known to

specifically influence speech and language development

• FOXP2 changed more during human

evolution than expected

FOXP2 protein evolution

Hypothesis: The human-specific amino acid changes in FOXP2 were selected due to some advantage in speech and/or language related features

When ?

>1000 humans worldwide

T=>N; N=>S

A..D..N..S A..D..N..S

2 Neandertals

Krause et al., 2007

Hypotheses

Human-specific changes in gene X contribute to human-specific changes in phenotype Y

Genomic data Gene function Phenotype data

b) Genetic engeneering in humans and/or chimpanzees to test the variants

Hypotheses testing

a) Crossing and backcrossing to map the trait

Hypotheses testing

a) Crossing and backcrossing to map the trait b) Genetic engeneering in humans and/or chimpanzees to test the variants c) Finding a natural backmutation in humans d) Study the differences between the ancestral and the human state in vitro d) Study the differences between the ancestral and the human state in human cell lines e) Study the differences between the ancestral and the human state in mouse

Enard et al., (2009), Cell, 137, 961-971

Studying FOXP2 evolution in a mouse model

A mouse model for FOXP2 evolution

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ 2 17

T N

7 8

T N

Knock-in vector

N S

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ

N S

A mouse model for FOXP2 evolution

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ

N S

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ

T N

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ

T N

Foxp2hum/hum Foxp2wt/wt

N S

Cross with Cre mice

Foxp2ko ko/ko wt/wt Foxp2hum

A mouse model for FOXP2 speech deficits

A mouse model for FOXP2 speech deficits

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ

T N

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ

T N

Foxp2wt/ko Foxp2wt/wt

FOX 1 2 3 4 5 6 7 8 9 10 111213 141516 17 atg tga PolyQ

T N

FOX 1 2 3 4 5 6 8 9 10 111213 141516 17 atg tga PolyQ

Stop

“Humanized” Mice

polyQ FOX

polyQ FOX

polyQ FOX

polyQ FOX

N S N S

polyQ FOX

polyQ FOX

polyQ FOX

polyQ FOX

T N T N Human Chimpanzee

“Speech impaired” Mice

polyQ FOX

polyQ FOX

polyQ FOX

polyQ FOX

T N T N

polyQ FOX

polyQ FOX

polyQ FOX

polyQ FOX

T N T N Speech normal Speech impaired

Speaking mice?

Mice homozygous for humanized Foxp2 (Foxp2hum/hum) Mice homozygous for wildtype Foxp2 (Foxp2wt/wt)

• German Mouse Clinic Screen
• Gene expression
• Neuroanatomical analysis
• Neuronal stem cells
• Elektrophysiology
• Vocalisation

Analysis of humanized Foxp2 mice

Molecular function of FOXP2

• FOXP2 is an ~ubiquitously

expressed transcription factor

• FOXP2 is usually expressed in a

subset of cell types in a given tissue

Lai et al. 2001 Shu et al. 2001

Molecular function of FOXP2

• FOXP2 is expressed in postmitotic neurons in the brain

Mouse; Ferland et al, 2003

Questions

• Which of the organic systems/cell types in

which FOXP2 is expressed might be affected by the human-specific amino acid changes?

• Could such changes be related to the

evolution of speech and language and if yes, how?

German Mouse Clinic Consortium; Helmholtz Center; Munich

Mouse Clinic Screen

H e a r t L u n g M e t a b

• l

i s m M

• r

p h

• l
• g

y C l i n i c a l C h e m i s t r y N e u r

• l
• g

y E y e N

• c

i c e p t i

• n

P a t h

• l
• g

y I m m u n

• l
• g

y B e h a v i

• r

Foxp2hum/hum mice are healthy, explore their enviroment more carefully, but show no difference in over 250 cardiovascular, immunological, pathological, morphological or respiratory measurements.

Mouse Clinic Screen

H e a r t L u n g M e t a b

• l

i s m M

• r

p h

• l
• g

y C l i n i c a l C h e m i s t r y N e u r

• l
• g

y E y e N

• c

i c e p t i

• n

P a t h

• l
• g

y I m m u n

• l
• g

y B e h a v i

• r

Foxp2wt/ko mice are healthy, show a variety of subtle differences in lung function, metabolism, blood pressure, hearing and immune function and explore their enviroment slightly more.

German Mouse Clinic Consortium; Helmholtz Center; Munich

Christine Winter and Reinhard Sohr, Charite, Berlin

Neurotransmitter levels

Frontal cortex Cerebellum Caudate-Putamen Nucleus accumbens Globus pallidus Glutamate Serotonin GABA Dopamine

Christine Winter and Reinhard Sohr, Charite, Berlin

Neurotransmitter levels

Foxp2hum/hum mice show reduced dopamine levels (p<0.001) Foxp2wt/ko mice show increased dopamine levels (p<0.05)

0% 50% 100% 150% 200% 250% Striatum Cerebellum Globus pallidus Cortex Nucleus accumbens

• rel. to wildtype

Foxp2hum/hum Foxp2wt/ko

* ** * * * ** **

Cortico-striatal circuits

• Striatum is the major target of

dopaminergic neurons

• Foxp2 is expressed in a large

fraction of striatal neurons Foxp2

Cortico-basal ganglia circuits and Foxp2

Vargha-Khadem et al., PNAS, 1998

The caudate nucleus has less grey matter and is

• veractivated in humans having only one

functional copy of FOXP2

Cortico-basal ganglia circuits and Foxp2

Haesler et al., PlOS Biology, 2007

Reduced FoxP2 expression in striatal neurons impairs vocal imitation in birds

Striatal gene expression patterns

• Analysed gene expression patterns in 70 striatal

biopsies of embryos (E16.5), adolescent (P15- P21) and adult (3 month) mice

• Find significant but very subtle effects that are

not enriched in particular functional categories

• Find tendencies for opposite effects in

Foxp2hum/hum mice and Foxp2wt/ko mice

Striatal neurons in vitro

• In vitro differentiated neurons from Foxp2hum/hum show longer,

more branched neurits

Sabine Gehre, Eun und Johannes Schwarz; Neurology; Leipzig *

100 200 300 400 500 all primary secondary tertiary quartery neurite length (m)

Foxp2wt/wt Foxp2hum/hum * *

Foxp2hum/hum Foxp2wt/wt

Striatal neurons in vivo

• Striatal neurons of adult mice have longer dendritic trees

Sabine Gehre und Johannes Schwarz; Neurology; Leipzig

Foxp2wt/wt Foxp2hum/hum

250 500 750 1000 1250 1500

dendrite length [m] * * Foxp2wt/wt Foxp2hum/hum Foxp2wt/ko

Johannes Schwarz; Neurology; Leipzig

Striatal synaptic plasticity

• Changes in the strength of

cortico-striatal synapses, i.e. synaptic plasticity, is thought to be important for learning action patterns

From cortex Striatal neuron

Johannes Schwarz; Neurology; Leipzig

From Gerdeman, et al., 2003

Striatal synaptic plasticity

From cortex Striatal neuron striatum

Long-term depression is a frequently studied form of synaptic plasticity at cortico- striatal synapses

• Long term depression in medium spiny neurons

Johannes Schwarz; Neurology; Leipzig

Striatal synaptic plasticity

40 50 60 70 80 90 100 110

• 5

5 15 25 time (min) normalised amplitude (%) Foxp2wt/wt

• Striatal Foxp2hum/hum neurons show stronger long term depression

(LTD)

Johannes Schwarz; Neurology; Leipzig

Striatal synaptic plasticity

40 50 60 70 80 90 100 110

• 5

5 15 25 time (min) normalised amplitude (%) Foxp2wt/wt Foxp2hum/hum

* ** ** * * * * * * * * *

Groszer et al., 2008

• Striatal neurons of mice carrying one non-functional FOXP2

allele show weaker long term depression

Striatal synaptic plasticity

Specificity of effects

Sabrina Reimers, MPI-EVA

Layer VI bipol Striatum Thalamus Layer VI Pyr Purkinje cells Layer II/III Pyr Amygdala Piriform Pyr Layer VI LCN Layer VI iPyr Golgi Foxp2

0% 25% 50% 75% 100% 125% 150% 175%

* ** *

wt/wt hum/hum

ANOVA: Foxp2expression*genotype: p<0.05

Reimers et al, in preperation

Specificity of effects

Sabrina Reimers, MPI-EVA

Golgi circuit Layer VI bipol Striatum Thalamus Layer VI Pyr Purkinje cells Amygdala

0% 25% 50% 75% 100% 125% 150% 175%

ANOVA: circuit*genotype: p<0.01 no no yes yes yes yes

wt/wt hum/hum

Specificity of effects

Wulf Hevers, MPI-EVA

40% 60% 80% 100%

• 15

5 25 45

Normalised amplitude (%) Time (min) wt/wt hum/hum

Increased LTD in striatum is confirmed (time*genotype, p<0.05)

Specificity of effects

Wulf Hevers, MPI-EVA

1 1.5 2 2.5 3 0.1 0.2 0.3

Paired pulse ratio Time (sec)

40% 60% 80% 100%

• 15

5 25 45

Normalised amplitude (%) Time (min)

Synaptic plasticity in Purkinje cells is unchanged (time*genotype, p>0.5)

Cortico-striatal circuits, Foxp2 and mice

• Humanizing Foxp2 in mice specifically affects cortico-

striatal circuits

• Mice with only one functional copy of Foxp2 show partly
• pposite effects
• Humans with only one functional copy of FOXP2 have a

reduced size and a different activction of the striatum (Vargha-Khadem et al., 2005)

• Reduced FoxP2 expression in striatal neurons impairs vocal

imitation in birds (Haeseler et al., 2007)

Kurt Hammerschmidt, Julia Fischer, Primatenzentrum, Göttingen

Vocalisation

50 100 75 kHz msec start PF

• max. PF

trend mean PF

Kurt Hammerschmidt, Julia Fischer, Primatenzentrum, Göttingen

Vocalisation

Foxp2hum/hum mice show subtle differences in the structure of ultrasonic isolation calls.

50 100 75 kHz msec start PF

• max. PF

trend mean PF s l

• p

e

• f

t r e n d Foxp2wt/wt Foxp2hum/hum kHz

70 75 80

kHz

• 0.4
• 0.3
• 0.2
• 0.1

m e a n P F m a x . P F s t a r t P F

*** ** ** ***

Conclusion

• The amino acid changes in FOXP2 could

have impacted the evolution of speech by altering properties of cortico-striatal circuits

• Mice models might be a promising system

to study the genetic basis of human evolution

Outlook

• Genomics will bring us catalogues of genetic

changes that occurred on the human and other primate/mammalian lineages

• Interpreting these differences will require

– statistical methods to identify genes/group of genes/genomic elements where changes on particular lineages are different from expectations

• since the power is limited one needs additional

information

– phenotypic changes on such lineages – functional information about genes – experimental systems such as the mouse to investigate resulting hypotheses

DO

• Work on models to investigate specific

genotype-phenotype hypotheses

• Sequence genomes
• Collect genome-wide data on gene

functions

• Specifiy phenotypic changes

precisely

Thank You !

Thank you!

Evolution of the primate transcriptome

Philipp Khaitovich, Michael Lachmann, Svante Pääbo

Khaitovich et al. Genome Research, 2004 Khaitovich et al. Nature Review Genetics, 2006 Enard et al., Science 2002 Khaitovich et al. , Science 2005

0% 5% 10% 15% 20% 25% 30% 35% Brain Heart Kidney Liver Testis

differently expressed genes false positives

Many genes are differently expressed between humans and chimpanzees

Khaitovich et al. Science, 2005

Khaitovich et al. Nature Review Genetics, 2006 Khaitovich et al. , Science 2005

Many genes are differently expressed between humans and chimpanzees

• Prim. Visual Ctx

Similar differences are found among brain regions

B r

• c

a ’ s a r e a P r e f r

• n

t a l C t x A n t . C i n g u l a t e C t x C a u d a t e n u c l e u s C e r e b e l l u m Khaitovich et al. Genome Research, 2004

Enard et al., Science 2002

More expression changes in the brain during human evolution ?

Specificity of effects

Sabrina Reimers, MPI-EVA

* Cerebellum Thalamus Striatum Amygdala Piriform cortex M1 layer II/III M1 layer VI Neuronal marker Foxp2 merged

Specificity of effects

Sabrina Reimers, MPI-EVA

* * * *

0.25 0.5 0.75 1

Foxp2wt/wt Foxp2hum/hum Cerebellum Thalamus Striatum Amygdala Piriform cortex M1 layer II/III M1 layer VI Neuronal marker Foxp2 merged