An evolutionary shift in the regulation of the Hippo pathway between mice and flies
Wouter Bossuyt, Chiao-Lin Chen, Qian Chen,Marius Sudol, Helen McNeill, Duojia Pan,Artyom Kopp, and Georg Halder
Supplemental information
1. Supplemental Figure Legends
Supplemental Figure 1:Conservation, protein domains and protein-protein interactions in Hippo pathway core components.
A)The phylogenetic distribution of the different core components and Merlin and Kibra of the Hippo pathway. Green indicates that an ortholog is present and red marks it’s absence.B)Schematic representation of the domain composition of different Hippo pathway components. Horizontal black lines under the proteins indicate domains that are essential for binding to other members of the Hippo pathway. Black ticks in proteins indicate phosphorylation sites. Human and fruitfly proteins are marked by a human and fly character respectively. The EBI domain nomenclature was used.
Supplemental Figure 2:Protein domains and protein-protein interactions in Hippo pathway inputs.
Schematic representation of the domain composition of different Hippo pathway inputs. The domain that is essential for binding to other members of Hippo signaling are indicated by horizontal black lines below the proteins. Black ticks in proteins indicate phosphorylation sites. Human and fruitfly proteins are marked by a human and fly character respectively. The EBI domain nomenclature was used. The red box in Drosophila Fat indicates the arthropod-specific motif that is necessary for the interaction with Hippo signaling.
Supplemental Figure 3:Multiple sequence alignment of the intracellular domain of Fat proteins form different animals.
We used orthologs of different species, namely for arthropods Drosophila melanogaster (fruitfly), Anopheles gambiae (mosquito), Tribolium castaneum (beetle), and Daphnia pulex (crustacean); for the species outside of arthropods: Homo sapiens (human), Mus musculus (mouse), Branchiostoma floridae (lancelet), and Lottia gigantean (mollusk). Conserved residues are marked by dark background. The transmembrane domain is indicated by a black box. The different deletions used in the UAS-Ft constructs are marked and numbered by red boxes. The residue that is mutated in the ftsum mutant,in which isoleucine 4852 is substituted for asparagine, is highlighted in a blue box. Drosophila Ft deletion constructs span the following residues: FtΔ1: aa 4704-4713, FtΔ2: aa 4744-4770, FtΔ3: aa 4834-4899, FtΔ4: aa 4922-4955, FtΔ5: aa 4973-4994, FtΔ6: aa 5044-5074, FtΔ7: aa 5089-5114, and FtΔ8: aa 5144-5147. Region 3 is highly conserved in arthropods but is not conserved outside of arthropods. The Hippo pathway interacting domains that were discovered by Matakatsu and Blair are indicated with solid black lines above the alignment marked Hippo N and Hippo C.
Supplemental Figure 4:The effect of Ft deletions constructs on wing size.
We tested the tumor suppressor function of the different Ft constructs by analyzing the ability to reduce wing size in ft mutant animals. All Ft overexpression constructs except FtΔ3 reduces the wing size of Ft mutants.Genotypes are indicated below each panel. Only female wings were used.
Supplemental Figure 5: The effects of Ft deletion constructs on ex-lacZ and compartment size in the wing imaginal disc.
All Ft overexpression constructs except FtΔ3 reducesex-lacZ expression in the posterior compartment and the compartment size when overexpressed by hh-Gal4 in a ft-mutant background. Genotypes are indicated below each panel. (J) Quantifications of size of the posterior compartment of third instar wing imaginal discs upon expression of Ft overexpression constructs normalized to the entire wing disc size. (K) Quantifications of ex-lacZ expression in the posterior compartment of third instar wing imaginal discs upon expression of Ft overexpression constructs normalized to ex-lacZ expression in the anterior part. Double asterisk indicate p<0,01 and single asterisk indicate p<0,05 as tested by one-way ANOVA.
Supplemental Figure 6:The effects of Ft deletion constructs on planar cell polarity of the abdominal hairs in pharate adults.
All Ft overexpression constructs except FtΔ3 reduces the planar cell polarity phenotype of abdominal hairs of Ft mutants.Genotypes are indicated below each panel.
Supplemental Figure 7: Polydot plot showing the conservation of Expanded within arthropods and the lower conservation outside of arthropods. All Expanded homologs are aligned to each other and a stretch of 4 similar residues is represented by a dot. All Ex homologs show alignment with the FERM domain of Expanded of Drosophila melanogaster, marked by a red box. Within arthropods, Expanded is highly conserved.
2. Fly Genotypes
Figure 2
2A,C: +/ ftsum
2B,D: ftsum / ftsum
2E: y w, hs-Flp; FRT40A ubi-GFP/FRT40A ft422; diap1-lacZ / +
2F: y w, hs-Flp; FRT40A ubi-GFP/FRT40A ftsum; diap1-lacZ / +
2G: w
2H: w; ft422/ ftfd
2I: w; ft422/ ftfd; tub-Gal4, UAS-Ft FL
2J: ftsum / ftsum
2K: y w, hs-Flp; FRT40A ubi-GFP/FRT40A ft422; m-lacZ / +
2L: y w, hs-Flp; FRT40A ubi-GFP/FRT40A ftsum; m-lacZ / +
Figure 3
3B: y w
3C: w; ft422/ ftG-rv
3D: w; ft422/ ftG-rv; act-Gal4/ UAS-Ft FL
3E: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ3
3H: w; ft422/ ftfd, ex697; hh-Gal4/ UAS-Ft FL
3I: w; ft422/ ftfd, ex697; hh-Gal4/ UAS-FtΔECD
3J: w; ft422/ ftfd, ex697; hh-Gal4/ UAS-FtΔ3
3K: yw, hs-Flp; fj-lacZ, FRT40A, Ft422/FRT40A, ubi-GFP; DE-Gal4/UAS-Ft3only
Figure S4
S4A: w
S4B: w; ft422/ ftG-rv
S4C: w; ft422/ ftG-rv; act-Gal4/ UAS-Ft FL
S4D: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔECD
S4E: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ1
S4F: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ2
S4G: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ3
S4H: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ4
S4I: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ5
S4J: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ6
S4K: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ7
S4L:w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ8
Figure S5
S5A: w; ft422/ ex697, ftfd, FRT40a ; act-Gal4 / UAS-FtΔECD
S5B: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ1
S5C: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ2
S5D: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ3
S5E: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ4
S5F: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ5
S5G: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ6
S5H: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ7
S5I: w; ft422/ ex697, ftfd, FRT40a; act-Gal4 / UAS-FtΔ8
Figure S6
S6A: w
S6B: w; ft422/ ftG-rv
S6C: w; ft422/ ftG-rv; act-Gal4/ UAS-Ft FL
S6D: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ1
S6E: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ2
S6F: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ3
S6G: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ4
S6H: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ5
S6I: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ6
S6J: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ7
S6K: w; ft422/ ftG-rv; act-Gal4/ UAS-FtΔ8
3. Supplemental Methods
Tracing the evolutionary history of genes and protein domains
To test for the presence or absence of each gene or protein domain in each of the candidate taxa, we first conducted BLAST/tblastn searches using human and one or more insect protein sequences as queries under default tblastn parameters. Analysis was restricted to species with fully sequenced genomes. Initial searches were performed against species-specific protein or predicted gene databases. In cases where a gene appeared to be absent from one or more taxa, we used three strategies to either identify this gene or confirm its absence. First, we conducted tblastn searches against the latest (as of June 2011) assembly of the genome sequence, thus eliminating potential false negatives that might result from errors in gene prediction. Second, we repeated tblastn searches using query sequences from several of the closest available relatives of the taxon in question. For example, Amot sequences from Mayetiola, Lutzomyia, and Anopheles were used to search the genomes of Drosophila, Ceratitis, and Glossina; the C-terminal domains of Ex from Daphnia, Nasonia, and Bombyx were used to search the genome of Ixodes, etc. (see Figure 2 for phylogenetic relationships). Third, the E-value cutoff was raised to 10, and any sequences identified under these parameters were subjected to phylogenetic analysis (see below).
In cases where the presence of a particular protein domain, rather than the entire gene, was in question in one or more taxa, we used two strategies to either identify this domain or confirm its absence. First, we conducted tblastn searches against species-specific protein or predicted gene databases using full-length genes from several other taxa as queries, and examined the gene models identified by these searches. Second, we performed tblastn searches against the latest assembly of the genome sequence using isolated sequences of the domain in question from several taxa as queries. If no significant hits were found using this approach, we concluded that this domain was absent or has diverged to the point where homology is impossible to establish.
In cases where homology was uncertain, or where there were multiple paralogs in some or all genomes (e. g. for Dachs), we used phylogenetic analysis to confirm or reject the orthology of genes from different taxa. Protein sequences were aligned and gene trees reconstructed from multiple sequence alignments using the PhyML algorithm ( Genes from different taxa were considered orthologous if they formed a monophyletic clade that did not include related sequences from the same taxa.
Databases used
The following taxa and sequence databases were searched (see Figure 2 for phylogenetic relationships):
Diptera
Drosophila melanogaster, Anopheles gambiae, Aedes aegypti, and Culex pipiens using the Flybase BLAST tool (Lutzomyia longipalpis and Mayetiola destructor using the Baylor College of Medicine Human Genome Sequencing Center (HGSC) genome databases and BLAST search tools ( Glossina morsitans using the Wellcome Trust Sanger Institute database and BLAST tool ( Ceratitis capitata, a draft genome assembly was kindly made available by Dr. Alfred Handler and colleagues, and searched using stand-alone BLAST.
Hymenoptera
Nasonia vitripennis and Apis melifera using the Flybase BLAST tool;Harpegnathos saltator,Camponotus floridanus, Acromyrmex echinatior, and Solenopsis invicta using the NCBI databases and BLAST tools ( Due to the preliminary state of genome assemblies in some hymenopterans, a gene was considered to have been present in the last common ancestor of Hymenoptera if it was identified in at least one hymenopteran species as well as in non-hymenopteran insects.
Other insects
Bombyx mori (Lepidoptera), Tribolium castaneum (Coleoptera), Acyrthosiphon pisum (Hemiptera), and Pediculus humanus (Phthiraptera) using the Flybase BLAST tool.
Non-insect arthropods
Daphnia pulex (Cladocera, Branchiopoda, Crustacea) using the NCBI and wFleaBase ( databases and BLAST tools. Ixodes scapularis (Acari, Arachnidae, Chelicerata) using the Vectorbase databases and BLAST tools (
Nematodes
Caenorhabditis elegans, Brugia malayi, Loa loa, and Ascaris suum using the NCBI databases and BLAST tools
Lophotrochozoan phyla
Capitella teleta and Helobdella robusta (Annelida) and Lottia gigantea (Mollusca) using the Joint Genome Institute (JGI) databases and BLAST tools ( Due to the preliminary state of genome assemblies in these taxa, a gene was considered to have been present in the last common ancestor of Lophotrochozoa if it was identified in at least one lophotrochozoan species as well as in ecdysozoans and/or deuterostomes.
Vertebrates
Multiple mammalian species, Gallus gallus (Aves), Xenopus tropicalis (Amphibia),and Danio rerio (Actinopterygii) using the NCBI databases and BLAST tools.
Non-vertebrate chordates
Branchiostoma floridae (Cephalochordata) and Ciona intestinalis (Urochordata) using the NCBI databases and BLAST tools.
Ambulacraria
Strongylocentrotus purpuratus (Echinodermata) and Saccoglossus kowalevskii (Hemichordata) using the NCBI databases and BLAST tools.
Cnidaria
Nematostella vectensis using the JGI database and BLAST, and Hydra magnipapillata using the Metazome website (
Primers used
ft-DECD-Not-RI-N5'; AAATAATAGCGGCCGCGAATTCAAACCATGGAGAGGCTACTGCTC
ft-DECD-Kpn-C3': ATAATAGGTACCTTACACGTACTCCTCTGGAGC
ft-DECD-D1-C5': GCAGCAACGTCCCCAGCTGATAAGGGAGGATCATCAC
ft-DECD-D1-N3': GATCCTCCCTTATCAGCTGGGGACGTTGCTGCTGCTG
ft-DECD-D2-C5': CATGGGTTCCGAGTACGAGGCTGCTGGCCTACGCAAA
ft-DECD-D2-N3': GTAGGCCAGCAGCCTCGTACTCGGAACCCATGTCCAC
ft-DECD-D3-C5': CCGCACCCATCAGAGCCAAAGTTCCAGTGCCAGCAGG
ft-DECD-D3-N3': TGGCACTGGAACTTTGGCTCTGATGGGTGCGGGATGC
ft-DECD-D4-C5': GCAGCAAACTTCCATGGTGTCGAGCAGCGCTCTGCAT
ft-DECD-D4-N3': GAGCGCTGCTCGACACCATGGAAGTTTGCTGCGCCTG
ft-DECD-D5-C5': AGATGTGGATGCCCATAACAGTCTCAGTGGCGACGGC
ft-DECD-D5-N3': CGCCACTGAGACTGTTATGGGCATCCACATCTCCACC
ft-DECD-D6-C5': CATTCCGCCGCACGCCAAGGCCAATGGAGCCGCATCC
ft-DECD-D6-N3': CGGCTCCATTGGCCTTGGCGTGCGGCGGAATGGGCGG
ft-DECD-D7-C5': GGCCACCACCCTCGGCACAAATGGACCGTCGCAGCAA
ft-DECD-D7-N3': GCGACGGTCCATTTGTGCCGAGGGTGGTGGCCGATGG
ft-DECD-D8-N3': ATAATAGGTACCTTATGGAGCCGCCGGTCCATTCGA
1