Formation of novel protein-coding genes

1
Formation of novel protein-coding genes

• Level 3 Molecular Evolution and Bioinformatics
• Jim Provan

Patthy Chapter 6
2
De-novo formation of novel protein-coding genes

• Creation of simple structural elements such as
  a-helices, b-sheets and reverse turns seems to be
  rather trivial
• So many alternative ways of forming these
  structures
• Have been invented independently several times
• Proteins with repetitive structure are most
  likely to arise de novo
• Repetitive oligonucleotide sequences can expand,
  forming periodic protein structures
• Collagen-like
• Leucine-rich repeat (LRR)
• Probably arose several times during evolution

3
Evolution of serum antifreeze glycoproteins

• Fish that live in polar waters have serum
  antifreeze glycoproteins (AFGPs) which allow them
  to tolerate temperatures of as low as 1.9C
• It has been shown that fish from the north and
  south poles have evolved very similar AFGPs
  independently
• AFGP of Antarctic fish, made up of a simple
  tripeptide repeat evolved by recruitment of the
  5 and 3 ends of an ancestral trypsinogen gene
  (secretory signal and 3 UTR) and de novo
  amplification of a 9bp Thr-Ala-Ala motif
• Arctic cod also have a Thr-Ala-Ala tripeptide
  repeat-based AFGP but this has no relationship
  with the trypsinogen gene
• Threonines are O-linked to galactosyl-N-acetylgala
  ctosamine and periodicity of repeats matches
  periodicity of water molecules
• Convergent evolution of the tripeptide-based AFGP

4
De novo creation of complex proteins

• Probability of de novo creation of more complex,
  globular proteins is inversely proportional to
  complexity
• Those that consist of a single supersecondary
  structure element (TIM barrel proteins) have
  higher probability of independent creation
• TIM barrel structure likely to have evolved
  several times
• Easier to remodel replicas of old protein folds
  than to invent them from scratch
• Creation of first folded proteins was probably
  the rate-limiting step in protein-based life
• All extant proteins probably arose from a limited
  number of ancestral folds through divergence

5
Evolutionary convergence

• Previous examples highlight convergence to
  similar primary, secondary or tertiary structure
  (structural convergence)
• Unlike structural convergence, functional
  convergence and mechanistic convergence are
  relatively common
• Several types of proteinases that have similar
  function (i.e. they cleave proteins) but have
  different structures and catalytic mechanisms and
  have evolved independently
• Example of mechanistic convergence is the serine
  proteases of the subtilisin and trypsin families
• Similar active sites and catalytic mechanisms but
  no sequence or conformational homology
• Catalytic triad residues (His, Asp, Ser) occur in
  different order in primary structures
• Difficult to prove structural convergence

6
Gene duplications

• Evolutionary significance of gene duplication is
  that it gives rise to a redundant duplication of
  a gene
• Duplicated gene may acquire divergent mutations
  and eventually emerge as a new gene
• Gene duplication is the predominant and most
  important mechanism by which new genes arise
• Genes derived by a duplication event are said to
  be paralogous and are found in different loci of
  the chromosome
• Different from orthologous genes gained by
  speciation events, which are found in different
  loci of the corresponding species

7
Types of DNA duplications

• An increase in the number of copies of a DNA
  segment can be brought about by several types of
  DNA duplication
• Partial, intragenic or internal gene duplication
  only an internal segment of a protein-coding
  gene is duplicated
• Complete gene duplication, including flanking
  regions necessary for expression
• Partial chromosome duplication several adjacent
  genes are duplicated
• Chromosomal duplication (aneuploidy)
• Genome duplication (polyploidy)

8
Mechanisms of gene duplication

• Major mechanisms for short intragenic
  duplications is disengagement of the DNA
  polymerase from the strand that is being copied
  and reattachment at the wrong point (slipped
  strand mispairing)
• Major mechanism for larger duplications involves
  unequal crossing over
• Involves mistaken pairing and recombination
  between homologous chromosomes
• Most likely in already-duplicated regions
• Allows rapid expansion of repeats within genes
  and expansion of gene families
• May facilitate homogenisation of gene sequences
  and thus slow down divergence (concerted
  evolution)

9
Unequal crossing-over
10
Gene duplications in lysozyme

• In ruminants, lysozyme gene has been duplicated
  10 times and is expressed less in
  extra-intestinal tissues
• In mice, intestinal lysozyme is expressed from
  lysP gene, whereas in other tissues it is encoded
  by the lysM gene
• Original gene duplication through unequal
  crossing-over in Alu-like B2 middle repetitive
  elements

11
Retrosequences

• Copies of protein-coding genes may be produced by
  duplicative transposition
• DNA is transcribed into RNA, which is
  reverse-transcribed into a cDNA (retroposition)
• During re-insertion, small segments of host DNA
  (4-12bp) are duplicated, forming direct repeats
• Significant diagnostic features of
  retrosequences
• Lack introns (where parent gene would have
  introns)
• Lack upstream promoter elements of parent gene
• Contain poly(A) stretches at 3 end
• Flanked by short, direct repeats
• Different chromosomal location from original gene

12
Functionality of retrosequences

• Depending on whether the copied gene is
  functional or not, we can distinguish processed
  genes (retrogenes) and processed pseudogenes
  (retropseudogenes)
• Several reasons why functional retrogenes are
  unlikely
• Process of reverse-transcription is very
  inaccurate
• Lacks necessary regulatory elements
• Generally truncated at 5 end (reverse
  transcriptase failure)
• May be inserted in genomic region unsuitable for
  expression
• More likely to form retropseudogenes
• Some examples of processed functional genes have
  been found e.g. human phosphoglycerate kinase
• X-linked gene has 11 exons and 10 introns
• Autosomal PGK gene has no introns and a poly(A)
  tail

13
Alu elements

• Processed pseudogenes of the RNA gene specifying
  7SL RNA which cuts signal sequences of secreted
  proteins
• About 300 bp long
• Around 500,000 copies in the human genome (5-6)
• Named after characteristic AluI restriction site
• Derived from functional 7SL sequence by
  duplication, two deletions and many mutations
• Play a key role in genome plasticity since they
  facilitate unequal crossing-over
• Gene duplication
• Exon shuffling

14
Fate of duplicated genes

• Determined by functional consequences of having
  extra copies of same gene and increased amounts
  of protein
• Duplications can be advantageous, deleterious or
  neutral
• If an organism is exposed to a toxic environment,
  there may be an advantage in overproduction of
  detoxifying enzymes
• Disadvantage will result of overproduction of
  protein upsets regulatory balance
• Most duplications are neutral fate determined
  by selection and drift
• Duplicated gene is unlikely to be fixed unless it
  acquires a novel and useful function
• May specialise in different subfunctions of
  ancestral gene
• May acquire drastically different functions
  (hepatocyte growth factor vs. plasminogen)

15
Formation of gene families

• Recently duplicated gene families are generally
  found in close proximity on the same chromosome
• Some multigene families contain invariant
  repeated genes
• Common when large quantities of protein product
  are required
• Histones have to be synthesised at a high rate
  during a well-defined, short period of cell
  division
• Some members of multigene families serve the same
  function but differ in tissue specificity,
  developmental regulation or biochemical
  properties e.g. isozymes

16
Concerted evolution in multigene families

• Paralogous members of multigene families are very
  similar to each other within one species although
  orthologous members of the same family may differ
  greatly between even closely related species
• Suggests that mechanisms exist which cause gene
  families to evolve together as a unit (concerted
  evolution)
• Process of concerted evolution of multigene
  families under the effects of random genetic
  drift is known as molecular drive
• Gene correction mechanisms may homogenise genes
  difficult to trace true evolutionary history of
  many multigene families

17
Dating gene duplications

• Assuming duplicated genes diverge at a constant
  rate, we can estimate the date of a gene
  duplication, TD, that gave rise to two paralogous
  genes (A and B) if we have sequences of these
  paralogues from two different species (1 and 2)
  and we know the time of speciation TS
• If genes evolved at a constant rate then
• Average number of substitutions per site (KA
  KB/2) in the two orthologue comparisons (A1 vs.
  A2, B1 vs. B2) is proportional to TS
• Average number of substitutions per site KAB in
  the four paralogous comparisons (A1 vs. B1, A2
  vs. B2, A1 vs. B2, A2 vs. B1) is proportional to
  the time since duplication TD
• Thus, the following equation holds
• TD/TS 2KAB/KA KB

18
Dating gene duplications (continued)

• All vertebrates have both myoglobin and
  haemoglobin
• Myoglobin differs from both the a and b subunits
  of haemoglobin more than they differ from each
  other
• Myoglobin diverged (TD 600-800 mya) before the
  a and b genes arose (TD 500 mya)
• Mammals, reptiles, birds, amphibians and bony
  fish all have distinct a and b subunits, whereas
  the most primitive vertebrates, the Agnatha
  (jawless fish), contain only one type of
  haemoglobin subunit
• Myoglobin and haemoglobin diverged prior to the
  separation of agnathans and jawed vertebrates
• Duplication giving rise to a and b subunits
  occurred in the ancestor of all jawed vertebrates
  following its divergence from agnathans

19
Evolutionary history and linkage patterns in a-
and b-globin clusters

• In humans, gene cluster of the a-globin family
  (c16) consists of four functional genes (z, a1,
  a2, q1) and three unprocessed pseudogenes (Yz,
  Ya1, Ya2)
• Embryonic type z is most divergent (estimated TD
  gt 300 mya)
• q1 is less divergent (estimated TD 260 mya)
• Genes a1 and a2 produce identical polypeptide and
  have near-identical nucleotide sequence,
  suggesting recent divergence
• b-globin family (c11) contains five functional
  genes and Yb
• Adult types (b and d) diverged from non-adult
  types (Gg, Ag and e) around 155-200 mya
• Ancestor of both g genes diverged from e about
  100-140 mya
• Duplication that formed Gg and Ag occurred after
  separation of human lineage from New World
  monkeys (35 mya)
• Divergence of adult genes (b and d) occurred
  about 80 mya

20
Intergene distance and time since duplication