Genèse des mutations germinales chez l'Homme Molly Przeworski Cours #1 - Collège de France
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Genèse des mutations germinales chez l’Homme
Molly Przeworski
Cours #1
https://www.rdmag.com/news/2017/02/understanding-genetics-human-height http://www.sbs.utexas.edu/levin/bio213/popgen/popgen.html http://www.pnas.org/content/97/13/7016
1062
diversité génétique in figure
re
ut tend to b
ne
on Two hapl
tati site, accou
u
M
haplotyp
pattern d
is largely
Mutation délétère lotype cla
spans ∼1
al. 1994)
consisten
Mu coalescen
tat contained
ion
bé served in
né
fiq CYP3A4
ue
ples, and
ples that
65% of t
or no pol
portion i
spectively
is indeed
To asse
the skew
our resul
of the Un
cer Resea
focuses o
tests of n
tests, assu
man popu
Comparin
Figure 1 Inferred haplotypes at CYP3A4 and CYP3A5. Neither
singleton sites nor multiallelic indels were included. The chimpanzee pirical dis
sequence was used to infer the ancestral allele at each site. The numbers circumve
on the right indicate the number of haplotypes in each population. tated by
The asterisk (*) indicates the position of CYP3A5*1/*3. Numbers can samp
below the gene names indicate the position of each polymorphic site
in the Sea
relative to the reference sequence for each gene (GenBank accession
numbers AC069294 and AC005020 for CYP3A4 and CYP3A5, the Seattl
respectively). evolution
normaliz
amount o
1062
in figure
tre
neu tend to b
on Two hapl
tati site, accou
u
M
haplotyp
pattern d
is largely
Mutation délétère lotype cla
spans ∼1
al. 1994)
consisten
Mu coalescen
tat contained
ion
bé served in
né
fiq CYP3A4
ue
ples, and
ples that
65% of t
or no pol
portion i
spectively
is indeed
To asse
the skew
our resul
of the Un
cer Resea
focuses o
tests of n
tests, assu
man popu
Comparin
Figure 1 Inferred haplotypes at CYP3A4 and CYP3A5. Neither
singleton sites nor multiallelic indels were included. The chimpanzee pirical dis
sequence was used to infer the ancestral allele at each site. The numbers circumve
on the right indicate the number of haplotypes in each population. tated by
The asterisk (*) indicates the position of CYP3A5*1/*3. Numbers can samp
below the gene names indicate the position of each polymorphic site
in the Sea
relative to the reference sequence for each gene (GenBank accession
numbers AC069294 and AC005020 for CYP3A4 and CYP3A5, the Seattl
respectively). evolution
normaliz
amount o
9 novembre 2018 Cours: « Genèse des mutations germinales chez l'Homme » Séminaire de Michel Georges (Liège): « Le processus de mutagenèse germinale revisité chez le bovin» 16 novembre 2018 Cours: « Evolution des mutations germinales chez les primates et datation de la spéciation humaine » Séminaire de Michel Brunet (Collège de France, émérite) ; « « Paléontologie et phylogénie moléculaire... Réflexions autour de Toumai et la dichotomie Chimpanzés - Famille humaine » 23 novembre 2018 Cours: « Mutation, sélection naturelle et fréquences des allèles pathologiques chez l'Homme » Séminaire en anglais de Guy Sella (Columbia/Pasteur): «A population genetic interpretation of complex trait architecture in humans» 20 mars 2019 Cours: « Causes de la variation du taux de recombinaison chez les vertébrés » Séminaire de Bernard de Massy (Montpellier): « Le contrôle de la distribution de la recombinaison par Prdm9, une intrigante stratégie moléculaire» 27 mars 2019 Cours: « Conséquences de la variation du taux de recombinaison chez les vertébrés » Séminaire de Laurent Duret (Lyon) : « Conversion génique biaisée: la face cachée de la recombinaison » 3 avril 2019 Cours: « À la recherche de la base moléculaire des adaptations » Séminaire en anglais de David Reich (Harvard): “Learning about human adaptation from ancient DNA” 10 avril 2019 Cours: « Empreintes de l'adaptation dans le génome humain » Séminaire de Laure Ségurel (Musée de l’Homme): « Adaptation à la consommation de lait chez l'Homme: un cas d'école pourtant encore bien mystérieux »
A>T
Malaspina et al. 2001 Arch Gen Psychiatry Samocha et al. 2014 Nat Gen Scally et al. 2012 Nature Comprendre l’effet de l’âge des Cartographier les Dater les événements parents sur le risque de mutations qui conduisent dans l’évolution certaines maladies héréditaires à des maladies graves humaine
ATTTCGA ATTTCGA
ATATCGA ATTTGGA
ATTTCTA
ATTTGGA
Roach et al. 2010 Science
https://www.genome.gov/images/content/comparative_genomics_factsheet.jpg
A>T
5’ methylated cytosine
deamination Efficiently
Cytosine Uracil removed
C>T methylation
deamination Not efficiently
5-Methylcytosine Thymine removed
(most in CpG context) (highly mutagenic)
From Gao et al. 2016 PLoS BiologySpermatogenesis ongoing after onset of puberty Oocytogenesis completed by birth
Fertilization 2n
~16-18 divisions
Number of cell divisions
Primordial
germ cell 2n
Sex determination >400 at age 30
~20 divisions ~15 divisions
2n 2n
Primary
oocytes
Birth 23 per year
Onset of pubertySpermatogenic No more DNA
stem cell 2n replication
~37
23 divisions/year
throughout life
~32
n n n n birth puberty age
Production of n
gametes
Male Female
Based on Drost and Lee, (1995)
Slide courtesy of Ziyue Gao (Stanford)Mutations not due to replication errors?
Interpreting the Time-Dependence of Mutation Rates
Ziyue Gao Gao et al. 2016 PLoS BiologyDependencies on age & sex
Inefficient repair Efficient repair or replication-driven
father
father
# mutations
# mutations
mother mother
SD Birth Puberty Reproduction SD Birth Puberty Reproduction
Age of parent Age of parent
Gao et al. 2016 PLoS BiologyGudjonsson , Lucas D. Ward ,
bjorg Jonasdottir1, Aslaug Jonasdottir1,
teinsdottir1,2, Gisli Masson1,
LETTER
3
& Kari Stefansson1,2
doi:10.1038/nature24018
d show strand concordance18. Despite many
dge on how sex differences in germ cell devel-
Parental
nce affect their mutability isinfluence
limited. To assess on human germline de novo
mutations
nd class of DNMs transmitted by inmothers
1,548andtrios from Iceland 2017
whole-genome sequencing (WGS) data from
Hákon Jónsson1, Patrick Sulem1, Birte Kehr1, Snaedis Kristmundsdottir1, Florian Zink1, Eirikur Hjartarson1,
19 1
an average of 35× coverage
Marteinn (Data
T. Hardarson 1
Descriptor
, Kristjan )., Hannes P. Eggertsson1, Sigurjon Axel Gudjonsson1, Lucas D. Ward1,
E. Hjorleifsson
Gudny A. Arnadottir1, Einar A. Helgason1, Hannes Helgason1, Arnaldur Gylfason1, Adalbjorg Jonasdottir1, Aslaug Jonasdottir1,
48 trios, usedThorunn
to identify
Rafnar1,108,778
Mike Frigge1high-quality
, Simon N. Stacey1, Olafur Th. Magnusson1, Unnur Thorsteinsdottir1,2, Gisli Masson1,
nucleotide polymorphisms (SNPs);
Augustine Kong1,3, Bjarni Methods
V. Halldorsson 1,4
, Agnar Helgason1,5, Daniel F. Gudbjartsson1,3 & Kari Stefansson1,2 ATTTCGA ATTTCGA
an average of 70.3 DNMs per proband. ATATCGA ATTTGGA
The characterization of mutational processes that generate sequence maternal origin17, and show strand concordance18. Despite many
diversity in the human genome is of paramount importance both to advances, our knowledge on how sex differences in germ cell devel-
medical genetics1,2 and to evolutionary studies3. To understand how opment and maintenance affect their mutability is limited. To assess
d the age and sex of transmitting parents affect de novo mutations, differences in the rate and class of DNMs transmitted by mothers and
here we sequence 1,548 Icelanders, their parents, and, for a subset fathers, we analysed whole-genome sequencing (WGS) data from
40225, at least one child, to 35× genome-wide coverage. We find
of 14,688 Icelanders with an average of 35× coverage (Data Descriptor19).
108,778 de novo mutations, both single nucleotide polymorphisms This set contained 1,548 trios, used to identify 108,778 high-quality
and indels, and determine the parent of origin of 42,961. The DNMs (101,377 single nucleotide polymorphisms (SNPs); Methods
number
30 of de novo mutations from mothers increases by 0.37 per and Fig. 1), resulting in an average of 70.3 DNMs per proband.
year of age (95% CI 0.32–0.43), a quarter of the 1.51 per year from
8 9 10 17
Counts
fathers (95% CI 1.45–1.57). The number of clustered mutations
pring increases faster with the mother’s age than with the father’s, and the a 800 d
20
genomic span of maternal de novo mutation clusters is greater than 600 40
ATTTCTA
Counts
400
that of paternal ones. The types of de novo mutation from mothers
change substantially with age, with a 0.26% (95% CI 0.19–0.33%)
200
0
30 ATTTGGA
decrease
10 in cytosine–phosphate–guanine to thymine–phosphate– 1 2 3 4 5 6 7 8 9 10 17
Counts
Number of offspring
guanine (CpG>TpG) de novo mutations and a 0.33% (95% CI 0.28– b 20
0.38%) increase in C>G de novo mutations per year, respectively. 100
Remarkably, these age-related changes are not distributed uniformly
Counts
0 50 10
across the genome. A striking example is a 20 megabase region on
0
chromosome 40 8p, with80 a maternal120 160 rate that is up
C>G mutation 1 2 3 4 5 0
toNumber
50-fold greater
of de novothan mutations
the rest of per
the genome.
probandThe age-related
Number of offspring
40 80 120 160
c e
accumulation of maternal non-crossover gene conversions also Number of de novo mutations per proband
Phase mostly occurs within these regions. Increased sequence diversity and Phase
Number of phased DNMs per proband
Paternal linkage disequilibrium of C>G variants within regions affected by Paternal
Maternal
Maternal excess maternal mutations indicate that the underlying mutational 90
process has persisted in humans for thousands of years. Moreover,
the regional excess of C>G variation in humans is largely shared by 60
chimpanzees, less by gorillas, and is almost absent from orangutans.
This demonstrates that sequence diversity in humans results from
evolving interactions between age, sex, mutation type, and genomic
Roach et al. 2010 Science
30
location.
In a previous study, we found that the number of de novo muta- 0
tions (DNMs) transmitted by fathers increases with age, at a rate of 20 30 40 50
∼2 per year, with no significant effect of the mother’s age4. Recently, Age of parent at conception (yr)
studies have shown a maternal age effect5–7 of 0.24 DNM per year Figure 1 | Family relationships and phasing of DNMs in three-
(ref. 6). The greater impact of the father’s age is consistent with repeated generation families. a, b, Number of offspring per parent pair and three-
mitosis of spermatogonia (∼23 per year (ref. 8)), whereas ova do not generation family proband. c, Schematic view of the three-generation
divide30postnatally. Moreover, mothers 50
transmit relatively more C>T, phasing approach. The DNM (star), along with the paternal chromosome
20 40 (blue segment), is transmitted from the proband to an offspring. We
and fewer T>G and C>A, DNMs than fathers6. Nucleotide type2,9,
modelled DNM transmission in three-generation families and used the
Age of parent
sequence at2,9
context conception
, replication (yr)
timing10, functional constraints9,11, resulting prediction to define high-quality DNMs (Methods). d, Number
apolipoprotein B messenger RNA-editing enzyme catalytic (APOBEC) of DNMs per proband. e, Phased DNMs as a function of the parent’s age at
nships and phasing
polypeptide ofactivity
DNMs 12 inepigenetics
, and three- 13,14 have also been reported to conception (restricting to 225 three-generation probands). The lines areLinear increase with paternal age
~¾ of mutations inherited from
father.
~1.5 mutations per year
Must be spontaneous
damage that is inefficiently
repaired
~0.4 mutations per year
Borrowed from Jónsson et al. 2017 Nature“One plausible explanation for the drastic age-related sex differences in transmitted DNMs is the relative lack of mitosis in ageing oocytes compared with spermatogonia, which may enrich for damage-induced DNMs.” Jónsson et al. 2017 Nature
Ratio father: mother
= « alpha »
Inefficient repair Efficient repair or due to replication
father
father
# mutations
# mutations
mother mother
PGC Birth Puberty Reproduction PGC Birth Puberty Reproduction
Age of parent Age of parent
Gao et al. 2016 PLoS BiologyGao et al., 2018 BioRxiv
1) le nombre de divisions de cellules
germinales mâles de la
détermination du sexe à la puberté a
été considérablement sous-estimé
2) après la détermination du sexe, les
divisions des cellules germinales
sont beaucoup plus mutagènes chez
les hommes que chez les femmes;
3) Les dommages contribuent de
manière substantielle aux mutations
de la lignée germinale mâle pendant
la puberté.
Gao et al., 2018 BioRxivRatio of male to female mutations
Direct estimates of alpha, ratio of male to female mutations in mammals
2.6 2.6 2.1
~3
Lindsay et al. Harland et al. Thomas et al.
Lindsay et al.
BioRxiv BioRxiv (Michel 2018 Current
BioRxiv
Georges) BiologyRecap The number of mutations inherited by a child increases with paternal and maternal age. In a typical sample of humans, ¾ of mutations are of paternal origin. This number shows little dependence on parental ages. Some fraction of mutations and possibly most mutations are non- replicative in origin and poorly repaired.
Gao et al., 2018 BioRxiv
Gao et al., 2018 BioRxiv
génome paternel génome maternel
effet de la mère
effet du père
Gao et al., 2018 BioRxivgénome paternel génome maternel
effet de la mère
effet du père
Gao et al., 2018 BioRxivgénome paternel génome maternel
effet de la mère
effet du père
Gao et al., 2018 BioRxivgénome paternel génome maternel
Stronger signal for
C>A mutations
effet de la mère
effet du père
Gao et al., 2018 BioRxivShort oocyte arrest
Long oocyte arrest
Borrowed from Jónsson et al. 2017 Nature
Slide courtesy of Guy AmsterShort oocyte arrest
Long oocyte arrest
Borrowed from Jónsson et al. 2017 Nature
Slide courtesy of Guy AmsterSummary The number of mutations inherited by a child increases with paternal and maternal age. In a typical sample of humans, ¾ of mutations are of paternal origin. This number shows little dependence on parental ages. Some fraction of mutations and possibly most mutations are non- replicative in origin and poorly repaired. Mutations arise at all stages of development, from zygote to sperm/egg. What fraction arise at each stage is still unknown in any mammals, and likely differs among species.
You can also read
