EVOLUCIÓN: PRINCIPIOS Y CUANTIFICACIÓN. Human nuclear genome Only 3% coding DNA.

EVOLUCIÓN:

PRINCIPIOS Y CUANTIFICACIÓN

Human nuclear genomeOnly 3% coding DNA

The Evolutionary forces:

• Natural selection (Darwin, Bernardi)

• Neutral Theory: Genetic drift (Kimura)

• Small population sizes

• Mutation– Mutationalist theory (Sueoka)– Thermodynamic pressure theory (Zimic & Arévalo)

• Gene flow (migration), horizontal transfer

MUTATIONTHE ULTIMATE SOURCE OF NEW GENETIC VARIATION.

Mutation rates are in the general range of:

Approx 10-7 to 10-8 per nucleotide per generation

Approx 10-5 per gene per generation

Approx 10-3 per generation at microsatellites

Genomes, genes and molecular evolution

A G

TC

purines

pyrimidines

transversions

transitions

transitions

BZM210: E.Willassen

http://www.ncbi.nlm.nih.gov/index.htmlhttp://www.no.embnet.org/Interesting links:

Transitions - transversions

A G

TC

purines

pyrimidines

transversions

transitions

transitions

TS / TV ratios

mtDNA 9.0 and globins 0.66

Expected:

TS / TV = 4 / 8 = 0.5

GENE FLOWSPREAD OF VARIATION OVER SPACE BY MOVEMENT AND/OR INTERMARRIAGE AMONG PEOPLE (‘ADMIXTURE’)

INTRODUCES NEW VARIATION INTO A POPULATION

REDUCES VARIATION BETWEEN POPULATIONS

GENETIC DRIFT

ALLELE FREQUENCY CHANGE DUE TO CHANCE FACTORS IN SEGREGATION, SURVIVAL & REPRODUCTION IN FINITE POPULATIONS.

GENETIC DRIFTINVERSELY RELATED TO POPULATION SIZE

POSITIVELY RELATED TO TIME.

PROBABILITY OF ULTIMATE FIXATION OF AN ALLELE IS ITS CURRENT FREQUENCY

APPLIES TO WHOLE SPECIES, BECAUSE TIME IS LONG

Initial p=0.5, N=25, 80 generations

UP OR OUT IN SMALL POPULATIONS

About ½ get fixed

Initial p=0.5, N=300, 100 generations

CHANGE IS SLOWER IN BIG POPULATIONS

Change is slower in larger populations

Drift reduces variation within populations due to fixation & loss of neutral alleles.

Drift increases variation between populations because different alleles are fixed in each populationLAS RAZAS HUMANAS EXISTEN?

Mean times to fixation and loss for selectively neutral alleles

Loss occurs more rapidly than fixation

Common alleles are generally old alleles

Geographically widespread alleles are usually old

GENE ‘TREES’

What happens to a DNA sequence over time?

. . . . . and why?(THINK OF THE DICE EXPERIMENT)

SPECIATIONA reduction in gene flow between populations accompagnied by divergent selection and/or

genetic drift, can lead to speciation.

Evolutionary history includes the transformation and divergens of lineages

Phylogenetic evolution or anagenesis

NOW

THEN

A2 A3 A1 A1 A4 A5

T1

T2

T3T4T5

ALLELES SAMPLED

A1

mutation

MUTATION HISTORY OF ALLELES

Mutations arise hierarchically

over time, generating a phylogeny

of cladistic (tree-like, branching)

DNA sequence relationships.

DNA sequences have a common ancestor and their variation reflects their descent history

Current sample of DNA sequences

ACTAA AATGA CGAAA CGAAG AGTAG

MRCA of all samples

MRCA of these 3 samples

Population history is reflected in

the pattern of sequence variation,

and the geographic location where

DNA sequence haplotypes are found.

• The great majority of mutations that are fixed are effectively neutral with respect to fitness, and are fixed by genetic drift

• polymorphism within populations is transient and due to the presence of selectively neutral alleles on their way to fixation or loss

The Neutral Theory

The Neutral Theory

• Adaptive Evolution is due to Natural Selection

• Advantageous mutations are rare

• most genetic variation at the molecular level is not selected within a population

• most genetic substitutions at the molecular level are not due to selection

• vertebrates

• fibrinopeptides

• hemoglobin

• cytochromes

• rates depend on functional constraints

Functional Constraint

millions of years since divergence

cytochrome c

hemoglobin

fibrinopeptide

• Mitochondrial gene in mammals

• uniform rate

• rate difference between silent and amino-acid replacement mutations

Functional Constraint

silent

replacement

Molecular Clock: observations

-hemoglobin in vertebrates

• plot amino acid differences against divergence time

• good linear approximation

Molecular Clock: observations

-hemoglobin about

constant rate over time

1 What use is the molecular clock?

• date divergence in phylogeny

• as a first approximation

Molecular Clock

Rates of Nucleotide Substitution

Rate: number of substitutions K between two homologous sequences divided by twice the time of divergence t

Ancestral sequence

Sequence 1 Sequence 2

t t

• Number of substitutions

Rates of Nucleotide Substitution

1 lineageK = r t

2 lineagesfrom splitK = 2 r t

• molecular clock is used to put a time to phylogenies

• construct phylogeny first by clock independent method

• clock based on well established partial phylogenies

• rate tests on reference set and subsets

• estimate times on total data base

Molecular Clock

Orthologous genes or not?

•orthologous - same gene copy

Well matching sequences may not be directly homologous

•paralogous - duplicate gene copy

•xenologous - introgressed gene copy(hybridization, virus) Horizontal transfer

v (3rd)

tran

sversio

ns

F84 distance

0.01

0.03

0.04

0.05

0.07

0.08

0.10

0.11

-0.0581 0.0000 0.0581 0.1163 0.1744 0.2326 0.2907 0.3489

’Multiple hits’ and ’saturation’

Time

Bas

e pa

ir d

iffe

renc

es

AD BCE

G>T

A>T

T>A

G - T - A - T

Reversal to a previous state may be detected as homoplasy. True phylogenetic signal would be masked with time and give false synapomorphies.

Signal depends on mutation rates, r.

v (3rd)

tran

sversio

ns

F84 distance

0.01

0.03

0.04

0.05

0.07

0.08

0.10

0.11

-0.0581 0.0000 0.0581 0.1163 0.1744 0.2326 0.2907 0.3489

Adjusted sequence change

Time

Bas

e pa

ir d

iffe

renc

es

different models have been made with intention to correct for multiple hits by converting observed distances between sequences to actual (expected) distances (under the particlar model)

’correction factor’

We can use genetic differences among populations or species to reconstruct

evolutionary history

Infering on likely evolutionary history from genetic differences

Amino acid sequences of hemoglobin alpha chains No. of Taxa : 6 Gaps/Missing data : Complete Deletion Distance method : Amino: Poisson correction No. of Sites : 140 d : Estimate

1 2 3 4 5 6 [1] Human -[2] Horse 0.13 -[3] Cow 0.13 0.13 -[4] Kangaroo 0.21 0.23 0.20 -[5] Newt 0.57 0.64 0.60 0.64 -[6] Carp 0.66 0.65 0.62 0.71 0.75 -

Human Horse

Cow Kangaroo

Newt Carp

0.1

Divergence can be used for grouping

An example of phyllogeny reconstruction from genetic differences by UPGMA

Molecular clocks:The longer time => The more genetic divergence

Molecular clock

Kimura (1968,1983): •if sequence divergence between humans and horses is scaled for time using fossils •and estimated evolutionary rate, r, is applied to all known protein coding loci•one amino acid substitution has been fixed every second year on average

Interpretation: This is too much for selection to have been influential during evolution of the vertebrates

Zuckerkandl & Pauling (1965): rate of amino acid change appears constant through time

the fate of mutations

mutations can be neutral

mutations can be advantageous and subject to positive selection

mutations can be disadvantageous and subject to purifying selection

selection can be detected by testing sequences against the predictions of neutral theory(for instance synonymous vs non-synonymous codons)

Mutations can be driven by thermodynamic pressure

Evolutionary constraints on DNAs

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Base position

Ent

ropy

G A A G A C C C U AUA A AGC U U

U AUAUUUUAUAUUUA

UUUUUUAUAAAGA A U AUUU

AAAAUU

UU

AUUUAAUUAAAUAU U U

UGUU

GGGG

UGAC

CAUAAGAU

UUAA UA

AA

CUCUUAUAAAUAUUUAACAUUG

AUUAAUG

AA

UUAUUGAUC

CGGUUUU

AUCGAUUAAAAA

UU

UAAGUU

ACUUUAGG

GA

Constraints are associated with functionality, for instance the need for rRNA to base pair and form helices in a secondary molecular structure

helix

loop

Transcription and translation

Translation requires available tRNA with appropriate anticodons to match with each codon on mRNA

codon

anticodon

DNA coding

Codes in organelle genomes differ slightly from the standard code

Anopheles gambiaeAAcid Codon FractionGly GG G 0.14Gly GG A 0.56Gly GG T 0.27Gly GG C 0.03

Glu GA G 0.02Glu GA A 0.98Asp GA T 0.95Asp GA C 0.05

Val GT G 0.02Val GT A 0.50Val GT T 0.45Val GT C 0.02

Ala GC G 0.00Ala GC A 0.28Ala GC T 0.64Ala GC C 0.08

codon bias: all codons are not equally frequentCodon usage

Codon redundancy:synonomous (silent) substitutions give the same amino acids.

synonomous substitutions do not affect the translation product and thus should be neutral in expressed genes

However, availability of specific tRNAs may make some codons more ’fit’

Synonymous codons are expected to be neutral, are expected to occur in equal frequency

Expect 50/50 frequency for two phenylalanine codons

5. Anomalous DNA composition

Codon biases are found in all known prokaryotes

Codon frequencies in E. coli

for instance: codons xxC and xxU can be read by the same anticodon, xxG

Translational efficiencydepends on tRNA availability

some tRNAs may pair with different codons due to: •’wobbles’ on the anticodon •modified nucleotides on the anticodon(possibility of G-U-pairing, Inosine (G’) pairs with A,C and U

codon

anticodonxxG xxGxxC xxU

Consequently some genomes do well with reduced number of tRNA types in the genome: 22 in vertebrate mitochondrial (mtDNA).

Leucine codons in two organisms

tRNAavailability

Usage: highly expressed

Usage: lowly expressed

Factor Analysis of codon usage of B. subtilis genes reveals three classes of genes

Class 3 (13%) genes that were apparently horizontally transferred.

Class 1 comprises the majority of the B. subtilis genes (82%)

Class 2 (5%) genes that are highly expressed under exponential growth conditions

Because some of the genes in this group showed clear relationships with bacteriophage genes, the hypothesis has been proposed that all these genes were alien and have been acquired horizontally from various sources.

Kunst, F et al. Nature (1997) 390 249-256

Mozner I. Current Opinion in Microbiology 1999, 2:524–528

Why do horizontally transferred genes use the genetic code differently?

Rocha EP. Trends Genet 2002 Jun;18(6):291-4

Bacterial species display a wide degree of variation in their overall G+C content

• Distribution of A + T-rich islands along the chromosome of B. subtilis.

• Location of genes from class 3 according to codon usage analysis is indicated by dots at the bottom of the graph.

• Known prophages (PBSX, SPb and skin) are indicated by their names, and prophage-like elements are numbered from 1 to 7.

However, most genes have roughly the same GC content within a genome

Kunst, F et al. Nature (1997) 390 249-256

Synonimous substitutions are not necessarily neutral

lowly expressed genes highly expressed genes

strong selection for translational efficiency

weak selection for translational efficiency

fewer tRNAs usedmore tRNAs used

strong codon biasweak codon bias

high rates of silent (neutral) mutations

low rates of silent mutations:

i.e. synonomous mutations

are not necessarily neutral!purifying selection

Redundancy and rates on codon positions

Code Table: StandardMethod: Nei-Gojobori (1986)S = No. of synonymous sitesN = No. of nonsynonymous sites----- No of Sites Redundancy----- for codon Pos Pos PosCodon S N 1st 2nd 3rdTTT (F) 0.333 2.667 0 0 2 TTC (F) 0.333 2.667 0 0 2 TCT (S) 1.000 2.000 0 0 4 TCC (S) 1.000 2.000 0 0 4 TCA (S) 1.000 2.000 0 0 4 TCG (S) 1.000 2.000 0 0 4 TAA (*) 0.000 3.000 0 0 0 TAG (*) 0.000 3.000 0 0 0TGA (*) 0.000 3.000 0 0 0TGT (C) 0.500 2.500 0 0 2 TGC (C) 0.500 2.500 0 0 2 TGG (W) 0.000 3.000 0 0 0 CTT (L) 1.000 2.000 0 0 4 CTC (L) 1.000 2.000 0 0 4 CTA (L) 1.333 1.667 2 0 4TTA (L) 0.667 2.333 2 0 2 TTG (L) 0.667 2.333 2 0 2

With codon redundancy we would expect less selective constraints on 3rd codon positions.

1st and 2nd position should be under stronger selective pressure.

Consequently evolution rates on 3rd codon positions are usually found to be higher than on 1st and 2nd positions

Even the sacred of sacreds of phylogenetic taxonomy can be violoated:

The problem: different molecules can yield different trees AND may still be telling the truth

ArchaeBacteria

Gene tree A Gene tree B Gene tree C

Kingdoms are not monophyletic in gene tree B and C

HGT possesses two ingredients sure to cause a controversy

1. Challenges the traditional tree-based view of evolution2. Is difficult to prove unambiguously

The solution: Horizontal Gene Transfer (HGT)

The significance of horizontal transfer was first recognized in the 1950’s resistance to multiple antibiotics could be transferred simultaneously from Shigella to Escherichia coli

“Infectious heredity”

Xenologs arise by horizontal transfer

SpeciationOrthologs

DuplicationParalogs

Horizontal TransferXenologs

Ancestral geneti

me

organisms

Xenologs

Paralogs – homologs related by duplicationOrthologs – homologs related by speciation

Xenologs – homologs related by horizontal transfer

1) Transformation – prokaryotes can take up free DNA from their surroundings

2) Conjugation – (bacterial sex) an organism builds a tube-like structure known as the pilus, joins it to its ‘‘mate’’, and transfers a plasmid through the tube. E. coli has been shown to conjugate with cyanobacteria, AND EVEN with S. cerevisiae!

3) Transduction – genes can be moved from one prokaryote species to another via viruses.

Mechanisms of horizontal transfer (also referred to as lateral transfer)

Base composition differences are mostly due to third position of codonsLawrence and Ochman. J Mol Evol (1997) 44:383–397

Horizontally transferred genes retain the sequence characteristics of the donor

genome

4. Conservation of gene order

Gene order is not generally conserved in microbial genomes

E. coli

B. subtilis

V. cholerae

• The presence of three or more genes in the same order in distant genomes is extremely unlikely unless these genes form an operon.

• Each operon typically emerges only once during evolution and is maintained by selection ever after.

• Therefore, when an operon is present in only a few distantly related genomes, horizontal gene transfer seems to be the most likely scenario.