Alternative splicing and gene duplication are interchangeable neofunctionalization mechanisms

1
Alternative splicing and gene duplication are
interchangeable neofunctionalization mechanisms

Naama Kopelman
2
Neofunctionalization
How to acquire a new function (without
losing the old)?
?
3
Two evolutionary mechanisms promoting
neofunctionalization
1. Gene duplication 2. Acquisition of splice
isoforms
4
42 of human genes are members of gene families
Gene duplications are ubiquitous
Family Size
Human Gene Families
5
Alternative splice variants are ubiquitous
Estimations of the extent of AS still rising
with ESTs coverege
Brett, D., Pospisil, H, Valcarcel, J. Bork, P.,
Alternative splicing and genome complexity,
Nature Genetics 30, (2002).
6
What is the relationship, if any, between gene
duplications and the acquisition of splice
variants?
Null hypothesis There is no relationship
between these genic attributes!
Positive correlation
No correlation
Negative correlation
Bigger gene familiy -gt More alternative splicing
Forces are not related by any way
Bigger gene familiy -gt Less alternative splicing
7
Estimating the number of splice isoforms of genes
from ESTs and mRNAs
Blast mRNAs and ESTs against known genes
Filter genes
1
2
3
Filter HSPs (alignments)
Delineate exons introns
1
2
Group transcripts into classes
3
Find max clique in collisions graph
8
A method for estimating a genes number of
splice variants
Maximum clique size is the minimal isoform count
The gene represented by this graph has at least
4 isoforms!
9
Time resolution of gene duplications
MYA
HS SC
1500
HS DM
990
HS FG
450
HS MM
90
Puffer fish
Mouse
Human
Yeast
FruitFly
10
Identifying duplications that follow a specific
speciation event
speciation
duplication
Gene A
Gene A1
Gene A2
speciation
duplications
A1
A2
A3
A1
A2
Remm, M., Storm, C.E.V. Sonnhammer, E.L.L.,
Automatic clustering of orthologs and in-paralogs
from pairwise species comparisons, Journal of
Molecular Biology 314, 1041-1052 (2001).
11
Ignoring duplications that preceded a specific
speciation event
p
Primordial hemoglobin
Duplication event
b
Speciation event
a
Human a
Human b
Mouse b
Mouse a
12
The relationship between gene family size and the
number of splice isoforms
Human duplications lt 450 MYA
10
9
8
7
6
Spliced Variants Count
5
4
3
2
1
1
2
3
4
5
6
7
8
9
10
Gene Family Size
Apparent trend of negative correlation
13
Statistical significance
Human duplications lt 450 MYA
3.5
10
9
3
8
2.5
Plt0.05 - Depletion Plt0.01 -Depletion Not
significant Plt0.05 - Enrichment Plt0.01 -
Enrichment
7
6
2
Spliced Variants Count
5
1.5
4
1
3
2
0.5
1
0
1
2
3
4
5
6
7
8
9
10
14
The relationship between gene family size and the
number of splice isoforms
Human duplications lt 90 MYA
15
Inverse relation between a genes family size
and the number of splice isoforms it produces
In reference to Mouse Puffer fish
Fruit fly Yeast
16
Large gene families tend not to use alternative
splicing

• Implied by this trend
• Genes that belong to large families are the least
  likely to be alternatively spliced according to
  this trend
• 2. Genes that have a large number of splice
  isoforms are the least likely to belong to a
  large gene family

17
Human gene duplicates of age lt 90MYA

• Genes with family 10 have no alternative
  isoforms (12 genes, 0 AS)
• Genes with family 5 hardly have alternative
  isoforms (76 genes, 16 AS)
• Genes with family1 usually have alternative
  isoforms (13156 genes, 52 AS)

18
Human gene duplicates of age lt 450MYA
9 out of the 10 largest gene families are
significantly depleted with alternative splicing
when compared to singletons (P lt 0.0001) Out of
415 genes in these families for which we have
splicing data, less than 3 employ alternative
splicing, as compared to 50 in singletons
19
Can retrotransposition explain this trend?

DNA
Transcription
Pre-mRNA
Processing
Cap
pTpTpT
cDNA
Integration
Genomic DNA
DNA repair
Retrosequence
Adapted from Graur and Li 2000
20
Retrotransposition alone cannot explain this
trend!
In reference to Mouse Puffer fish
Fruit fly Yeast
Excluding genes with less than 3 (?) exons
21

• What processes are responsible for the inverse
  relationship?

The inherent tendency hypothesis

• Let there be two classes of genes
• Prefer to proliferate by gene duplication
• Prefer to proliferate by alternative splicing
• Under this model, an inverse relationship is
  expected

Thus, an inherent tendency causes some genes to
proliferate by gene duplication and others by
isoform acquisition.
22
To test this hypothesis we group duplications
according to four evolutionary scenarios
H
M
H1
M
H
M2
H2
H1
H2
M1
M1
M2
23
Negative correlation revisited
Genes duplicated in both lineages have less
alternative splicing
Singletons have alternative splicing
Evolutionary scenarios
24
Negative correlation cannot be explained by
inherent tendency hypothesis
Scenarios 2 and 3 show that the negative
correlation is not due to inherent tendencies of
genes
25
Do duplicates lose splice variants? Or do
singletons gain them?
26
Staged duplications
Human
Fruit fly
Puffer fish
Mouse
Yeast
27
Duplicates acquire splice isoforms
28
A model to explain the inverse relationship
between gene duplication and alternative splicing
Alternative splicing
Gene duplication
mutual repression is explained by a demand for
proliferation that can be met by both mechanisms
29
Summary

• Negative correlation is found between a genes
  family size and the number of its splice
  variants, i.e. genes tend to utilize mainly one
  of these mechanisms
• A model of mutual repression is explained by a
  demand for proliferation that can be met by both
  mechanisms

30
The End
Thanks
Itai Yanai Doron Lancet Eytan
Domany All my groups memebers Yoram
Groner Adnan Derti (Church lab, Harvard) Aren
Bar-Even (Pilpel lab)