The genetic basis of phenotypic diversity. Virginie Orgogozo Institut Jacques Monod Paris

Transcription

1 The genetic basis of phenotypic diversity Virginie Orgogozo Institut Jacques Monod Paris

2 What makes us different? between individuals between species? How do genotypes map onto phenotypes? Phenotype = observable attributes of an individual Genotype = inheritable genetic material = DNA or RNA

3 1) Identifying the mutations responsible for phenotypic differences 2) Gephe : thinking in terms of differences 3) Genetic hotspots of evolution

4 1) Identifying the mutations responsible for phenotypic differences 2) Gephe : thinking in terms of differences 3) Genetic hotspots of evolution

5 What are the mutations responsible for phenotypic differences? Genomic approach identifies rapidly-evolving genomic regions, new genes, deleted genes? Genomes Phenotypic differences CCTCCTCCATACCCAAATGGATGGTACGGCATTCTTGAATCATCAAAGCT TAGAGCGGGGGAATCGAAGCATATATCATGTCTAGGCGAGCAACTTATAG TGTTCCGTTCCCAAGCTGGTGAAGTTTATATCTTGGATGCGTATTGCCCG CACTTGGGCGCTAATTTGAGTAAGGGAGGTCGAGTTATAGGAGATAATAT TGAATGTCCCTTTCACCACTGGAGCTTTAGAGGCAGTGATGGCATGTGTA CCAATATTCCCTACAGCAGCAATATACACTCATCTACAAAAACTAAAAAA TGGACCTCCACCGAAGTGAATGGATTCATATTTCTTTGGTACAATGTCGA AGAATCTGAAGTTCCGTGGAATATACCAAAATCAGTTGGTGTTGCAAAAA? CCTCCTCCATACCCAAATGGATGGTACGGCATTCTCGAATCATCAAAGCT TAGAGCGGGGGAATCGAAGCATATATCATGTCTAGGCGAGCAACTTATAG TGTTCCGTTCCCAAGCTCGTGAAGTTTATATCTTGGATTCGTATTGCCCG CACTTGGGCGCTAATTNGAGTAAGGGAGGTCGAGTTATAGGAGATAATAT TGAATGTCCCTTTCACCACTGGAGCTTTAGAGGCAGTGATGGCATGTGTA CCAATATTCCCTACAGCAGCAATATACACTCATCTACAAAAACTAAAAAA TGGACTTCCACCTAAGTGAATGGATTCATATTTCTCTGGTACAATGTCGA AGAATCTGAAGTTCCGTGGAATATACCAAAATCAGTTGGTGTTGCAAAAA Phenotypic approach identifies the gene(s) and the mutation(s) responsible for a phenotypic change

6 Pelvic reduction in sticklebacks marine 5mm Paxton Lake, Canada 5mm Gasterosteus aculeatus (Peichel et al., 2001 ; Shapiro et al, 2004 ; Chan et al. 2010)

7 Marine fishes with robust pelvis = ancestral Freshwater fishes with reduced pelvic structures = derived, independently at least 20 times - limited calcium availability - absence of gape-limited predatory fishes - predation by grasping insects Last glacier retreat = years ago

8 QTL mapping lake marine parents F1 gametes 375 F2 individuals QTL? phenotype (Shapiro et al., 2004)

9 Quantitative measurement of the phenotype Length of pelvic girdle Spine length Height of ascending branch

10 1000 microsatellite markers 26 linkage groups (Peichel et al., 2001)

11 One major locus at the end of linkage group 7 Major locus responsible for 65% of the variance

12 One major locus at the end of linkage group 7 A few minor loci Major locus responsible for 65% of the variance

13 Three candidate genes: Pitx1, Pitx2, Tbx4 Screen of a BAC library (Bacterial Artificial Chromosomes) containing kb fragments of stickleback genomic DNA Partial sequencing and generation of new markers Genotyping of 375 F2 individuals

14 One major locus at the end of linkage group 7 A few minor loci

15 Pitx1, responsible for the phenotypic change? Pitx1 null mutations in mice (pelvis reduction, stronger on right side) QTL mapping Same coding sequence in lake and marine forms Pitx1 expressed at stage 29 in marine individuals but not in marine individuals BUT The decrease in Pitx1 expression levels might have evolved due to mutations in an upstream regulatory gene

16 Comparison of allele expression in hybrids

17 What is pyrosequencing?

18 Test of Pitx1 cis-regulatory regions egfp

19 Rescue of a pelvis in freshwater individuals

20 Several independent deletions in the cis-regulatory region of Pitx1 Region sequenced in two lake pop[ulations: a 2-kb deletion in one and a 757-bp deletion in the other one SNP genotyping in 13 populations with reduced pelvis and in 21 populations with complete pelvis 8 different deletions

21 How a Drosophila species adapted to its cactus host Lang et al. Science 2012

22 Picture: L. Matzkin

23 Picture: L. Matzkin Drosophila pachea senita cactus

24 toxic alcaloids Picture: L. Matzkin resistance Drosophila pachea senita cactus

25 toxic alcaloids Picture: L. Matzkin resistance Drosophila pachea senita cactus dependance

26 How did D. pachea become dependent on a cactus? generalist specialist bamboo obligate specialist cactus senita

27 toxic alcaloids resistance 7-sterols senita cactus dépendance dependance Picture: L. Matzkin Drosophila pachea

28 Ineluctable degradation of metabolic activities during evolution arginine histidine isoleucine leucine lysine methionine phenylalanine threonine valine ascorbic acid (vitamin C) biotin (vitamin H) folic acid (vitamin M) riboflavin (vitamin B2) thiamine (vitamin B1) cobalamin (vitamin B12) Romero 2005 Genome Biology 7-sterols Heed 1965 Science Why have these metabolic activities been lost? selective advantage? neutral?

29 The nannoptera group on normal food D. pachea D. acanthoptera D. wassermani D. nannoptera (Ward and Heed, 1970)

30 How did D. pachea become dependent on a cactus? Drosophila pachea 7-sterols cactus senita dépendance dependance

31 D. melanogaster cholesterol steroid hormone (ecdysone) D. pachea cholesterol

32 D. melanogaster D. pachea cholesterol cholesterol 7-dehydrocholesterol 7-dehydrocholesterol steroid hormone (ecdysone) steroid hormone (ecdysone) (Clayton, 1964 ; Heed and Kircher, 1965; Chu et al., 1970; Warren et al., 2001)

33 D. melanogaster cholesterol nvd D. pachea cholesterol X 7-dehydrocholesterol 7-dehydrocholesterol steroid hormone (ecdysone) steroid hormone (ecdysone) Mutation(s) in the nvd gene in D. pachea? (Yoshiyama et al., 2006)

34 Several important amino acid changes in D. pachea NVD E G polar, acidic non polar, neutral

35 In D. pachea nvd is still expressed in prothoracic glands D. melanogaster D. acanthoptera D. pachea 100µm 60MYA D. pachea D. acanthoptera D. melanogaster

36 Novel metabolic pathway in D. pachea? D. melanogaster D. pachea cholesterol HO cholesterol X nvd 7-dehydrocholesterol 7-dehydrocholesterol HO 20-hydroxyecdysone Senita cactus contains lathosterol? HO

37 In vivo tests in D. melanogaster Fly survival WT -normal food +lat + lathosterol +7DHC + 7DHC nvd RNAi nvd RNAi + pa-nvd nvd RNAi + pa-nvd 4mut

38 D. pachea NVD converts lathosterol and not cholesterol rescue on lathosterol no rescue on normal food Fly survival WT -normal food +lat + lathosterol +7DHC + 7DHC nvd RNAi nvd RNAi + pa-nvd cholesterol lathosterol 7-dehydrocholestérol

39 In vitro assay of Nvd activity UAS-nvd Gene transfection of Drosophila S2 cells cholesterol Extraction and analysis by HPLC 7-dehydrocholesterol

40 As expected, D. pachea NVD converts lathosterol but not cholesterol with cholesterol with lathosterol 0 0 GFP B. mori D. mojavensis D. acanthoptera D. pachea

41 with cholesterol with lathosterol GFP B. mori D. mojavensis D. acanthoptera D. pachea Unexpectedly, all the tested enzymes convert lathosterol

42 ANCESTRAL STATE DERIVED STATE D. mojavensis and D. acanthoptera D. pachea diverse food lathosterol NVD cholesterol NVD 7-dehydrocholesterol steroid hormone (ecdysone) Senita cactus cholesterol X lathosterol NVD 7-dehydrocholesterol steroid hormone (ecdysone)

43 Which amino acid changes? D. mojavensis D. robusta D. wassermani D. nannoptera D. acanthoptera D. pachea 19 amino acid changes 5 are predicted to be deleterious SIFT (Ng et al. 2003)

44 These five amino acids are present in all the sampled D. pachea individuals 45 individuals a fly stock a fly stock a fly stock 15 individuals 400 km

45 Three mutations reduce NVD activity to less than 15% with cholesterol with lathosterol 40 0 D. mojavensis 0 D. mojavensis NVD wild-type activity P290C L300I G376T E377G 160 D. mojavensis NVD wild-type activity G250A

46 At least 2 mutations are required to restore NVD activity on cholesterol D. acanthoptera NVD activity HA actin

47 Reverting 4 amino acids restores NVD activity in vivo rescue on normal food Fly survival WT -normal food +lat + lathosterol +7DHC + 7DHC nvd RNAi nvd RNAi + pa-nvd nvd RNAi + pa-nvd 4mut

48 How a few mutations can restrict the ecological niche of a species Most insects D. pachea diverse food lathosterol NVD cholesterol NVD 7-dehydrocholesterol steroid hormone (ecdysone) Senita cactus cholesterol lathosterol X NVD G250A, NVD G376T, E377G 7-dehydrocholesterol steroid hormone (ecdysone)

49 How a few mutations can restrict the ecological niche of a species Most insects D. pachea Senita cactus diverse food lathosterol NVD cholesterol NVD 7-dehydrocholesterol cholesterol X NVD G250A, NVD G376T, E377G 7-dehydrocholesterol Why has this metabolic activity been lost? selective advantage? steroid hormone (ecdysone) lathosterol neutral? steroid hormone (ecdysone)

50 In vitro: No advantage with cholesterol with lathosterol 0 0 GFP B. mori D. mojavensis D. acanthoptera D. pachea

51 Recent selection produces a selective sweep Lower nucleotide diversity Lower recombination rate

52 Collecting flies in Mexico X 400 km Michael LANG

53 Sequencing the D. pachea genome kb nvd

54 Sequencing the D. pachea genome kb nvd > 200 kb

55 nvd is in a low-nucleotide-diversity region Size of the region (bp) Inferred number of recombination events nvd region control loci

56 The nvd region has faced recent positive selection Omega (Kim and Nielsen) C C952 C3665 nac CKIIa nvd acat SpdS

57 The D. pachea nvd allele is advantageous on lathosterol * relative number of flies * n.s. n.s lat +7DHC GAL4 WT 2-286>nvd-RNAi nvd RNAi -normal food +lat + lathosterol +7DHC + 7DHC 2-286>nvd-RNAi nvd>nvd-pa RNAi + pa-nvd 2-286>nvd-RNAi >Nvd-pa4mut nvd RNAi + pa-nvd 4mut * p < 0.04

58 The D. pachea nvd allele is advantageous on lathosterol * relative number of flies * n.s. n.s lat +7DHC GAL4 WT 2-286>nvd-RNAi nvd RNAi -normal food +lat + lathosterol +7DHC + 7DHC 2-286>nvd-RNAi nvd>nvd-pa RNAi + pa-nvd 2-286>nvd-RNAi >Nvd-pa4mut nvd RNAi + pa-nvd 4mut * p < 0.04

59 D. pachea became dependent on a cactus How? via several mutations in a single gene + maybe other genes Why? These mutations are advantageous on lathosterol Losses of metabolic activities can be beneficial. Short-term beneficial mutations can be selected despite long-term detrimental effects. Michael Lang, Sophie Murat, Géraldine Gouppil (Institut Jacques Monod) Catherine Blais, Émilie Guittard, René Lafont, Chantal Dauphin-Villemant (UPMC) Andrew Clark (Cornell U) Luciano Matzkin (U Alabama) Takuji Yoshiyama, Hiroshi Kataoka, Ryusuke Niwa (Japon)

60 cactus generalist ancestor 1) evolved resistance to senita toxic compounds still feeding on multiple cactus species 2) adaptation to senita together with inability to live on other cactus species senita Picture: A. Martin cactus obligate specialist

61 Methods to identify the genes and the mutations responsible for phenotypic evolution

62 Two types of appraoches Genetic mapping Candidate gene no a priori, fewer biais long and tedious rarely ends with identification of the gene Based on an a priori idea can be fast and efficient only with strains/species which produce fertile hybrids will only find known genes In both cases, genes with small effect are more difficult to identify

63 Various methods Genetic which chromosome (ex: autosomal versus sex) QTL mapping Genetic association studies Complementation tests General biology General knowledge of the genes involved in the phenotype Similarity with a known phenotype Correlation with a change in gene expression level/pattern Final test of protein activity in vitro in E. coli, by transgenesis in the studied species or the closest model organism (ex: beta-defensin of dogs tested in mouse) Final test of cis-regulatory regions - with reporter constructs, transgenesis, comparison of both regions - comparison of allele expression levels in hybrids (pyrosequencing)