ERK1 and ERK2 present functional redundancy in tetrapods despite higher evolution rate of ERK1
© Buscà et al. 2015
Received: 14 July 2015
Accepted: 10 August 2015
Published: 3 September 2015
The Ras/Raf/MEK/ERK signaling pathway is involved in essential cell processes and it is abnormally activated in ~30 % of cancers and cognitive disorders. Two ERK isoforms have been described, ERK1 and ERK2; ERK2 being regarded by many as essential due to the embryonic lethality of ERK2 knock-out mice, whereas mice lacking ERK1 are viable and fertile. The controversial question of why we have two ERKs and whether they have differential functions or display functional redundancy has not yet been resolved.
To investigate this question we used a novel approach based on comparing the evolution of ERK isoforms’ sequences and protein expression across vertebrates. We gathered and cloned erk1 and erk2 coding sequences and we examined protein expression of isoforms in brain extracts in all major clades of vertebrate evolution. For the first time, we measured each isoforms’ relative protein level in phylogenetically distant animals using anti-phospho antibodies targeting active ERKs. We demonstrate that squamates (lizards, snakes and geckos), despite having both genes, do not express ERK2 protein whereas other tetrapods either do not express ERK1 protein or have lost the erk1 gene. To demonstrate the unexpected squamates’ lack of ERK2 expression, we targeted each ERK isoform in lizard primary fibroblasts by specific siRNA-mediated knockdown. We also found that undetectable expression of ERK2 in lizard is compensated by a greater strength of lizard’s erk1 promoter. Finally, phylogenetic analysis revealed that ERK1 amino acids sequences evolve faster than ERK2’s likely due to genomic factors, including a large difference in gene size, rather than from functional differences since amino acids essential for function are kept invariant.
ERK isoforms appeared by a single gene duplication at the onset of vertebrate evolution at least 400 Mya. Our results demonstrate that tetrapods can live by expressing either one or both ERK isoforms, supporting the notion that ERK1/2 act interchangeably. Substrate recognition sites and catalytic cleft are nearly invariant in all vertebrate ERKs further suggesting functional redundancy. We suggest that future ERK research should shift towards understanding the role and regulation of total ERK quantity, especially in light of newly described erk2 gene amplification identified in tumors.
ERKs are the effector kinases of the Ras/Raf/MEK/ERK signaling pathway involved in multiple essential cell processes such as proliferation , differentiation , survival and memory formation [3, 4]. Abnormal activation of this cascade leads to pathologies such as cancer  or cognitive impairments . Since the discovery of two ERK isoforms in mammals, ERK1 (MAPK3) and ERK2 (MAPK1) in 1991 , numerous researchers have strived to understand their respective roles. While some cancers have been associated with isoforms of ERK cascade members such as B-Raf for melanoma , the putative differential involvement of either ERK isoform in cancer or any disease remains unknown.
ERK1 and ERK2 are expressed ubiquitously in mammals where both display the same kinase specific activity in vitro [9, 10] and share a highly similar 3D structure (Additional file 1C). Furthermore in mammals, both translocate to the nucleus upon stimulation by cell surface receptors  and while they do share at least 284 interactors, no isoform-specific substrates have been identified . Indeed, ERK1 and ERK2 share 22 out of 23 amino acids that have been demonstrated to directly interact with substrates [13, 14], the sole difference being a conservative substitution: leucine155ERK2 into isoleucine175ERK1 (Additional file 1A).
ERK2 is regarded by many as essential due to the embryonic lethality of ERK2 knock-out mice [15–17], whereas mice lacking ERK1 are viable and fertile suggesting a dispensable role of ERK1 . Similarly, some studies based on siRNA mediated-invalidations suggest specific functions [19–21]. However, targeted erk1 and/or erk2 gene disruption in mice organs can evoke redundancy . In mouse fibroblasts we showed that solely si-RNA mediated ERK2 knock-down reduced cell proliferation by itself, however when ERK2 levels were clamped down, ERK1 knock-down became effective at reducing cell proliferation. Hence, we and others have hypothesized that the apparent dominant role of ERK2 is only due to its higher expression rather than functional differences [10, 23]. To date, the controversial question of ERKs differential function versus redundancy has not been successfully addressed.
Here we show by a combined approach based on ERK1/2 sequence evolution and ERK1/2 protein expression across vertebrates, that while some tetrapods express both ERK1 and ERK2 proteins, others have lost the erk1 gene and others express either only ERK1 or only ERK2 at detectable levels despite having both genes. Hence our results strongly suggest that ERK1/2 can act interchangeably, a conclusion strengthened by the observation that amino-acids required for function are invariant in ERK1 and ERK2.
One ERK is expressed in agnathans and chondrichthyes
The next major clade in the vertebrate lineage, cartilaginous fishes (class Chondrichthyes) are constituted of three sub-clades: chimeras, sharks and rays. Among chimeras, the genome of the elephant shark (Callorhinchus milii) is sequenced and has only a single erk . To extend this result to the two other classes of chondrichthyes we first analyzed ERK expression in brains. We detected a sole ERK isoform in the brain of winter skate (Raja ocellata), medulla of the small-spotted catshark (Scyliorhinus canicula) and spinal cord of the spiny dogfish (Squalus acanthias, Fig. 1). Using degenerated primers targeting teleost erk1 and erk2, we also amplified and cloned a single erk from winter skate (ray) and spiny dogfish (shark), both of which share a high sequence identity (87–89 % at nucleotide level) to elephant shark erk. We conclude that members of the three clades of Chondrichthyes express a single ERK.
The single ERK of Chondrichthyes is likely to be ERK2, firstly because its sequence segregates within ERK2 group by three independent topological methods (Fig. 2 and Additional file 5), and the bootstrap values of the branches of Fig. 2 that place Chondrichthyes ERK into the ERK2 group support this finding. Secondly, the size of the erk gene from elephant shark is large (55 kb), similar to the large size of teleost and mammal erk2 genes and opposite of the extremely compact erk1 genes present in vertebrates as it will be described at the end of the result section (Fig. 6).
ERK1 and ERK2 from all bony vertebrates arose from a single duplication event
All teleosts genomes sequenced so far (e.g. Danio rerio and Fugu rubripes) possess two ERK genes; and we readily detect two ERK proteins from brain extracts of two representatives of the same teleost clade: Dicentrarchus labrax and Hemichromis bimaculatus (Additional file 2A).
Bichirs are extant representatives of the most ancient branch of ray-finned fishes, diverging from the rest of Actinopterygii prior to the whole genome duplication specific of teleosts . Therefore ERK isoform identification in bichir is critical for determining ERK1 and ERK2 emergence in bony fish. We analyzed brain extracts from bichir (Polypterus senegalus) and detected two ERK isoforms (Fig. 1A and Additional file 2A) and we further cloned two erk genes from P. senegalus. Phylogenetic analysis of bichir ERK isoforms revealed that they already fully segregate at gene and protein levels as ERK1 and ERK2 as classified in teleosts and tetrapods (Fig. 2 and Additional file 5). This demonstrates duplicate ERK1/2 emergence at least 400 Mya prior to the divergence of the bichirs from primitive bony vertebrates . Furthermore, in human and zebrafish genomes ypel3 gene is located downstream of erk1/mapk3 gene in the same orientation than erk1; and ypel1 is located downstream of erk2/mapk1 in the same orientation than erk2; ypel1 and ypel3 being also isoforms therefore the synteny of isoforms is conserved from fish to mammals between these two loci, which is further indicative of a single duplication event that generated ERK1 and ERK2 in vertebrates.
erk1 gene has been lost independently in two tetrapod clades
Birds, frogs and cartilaginous fishes lack erk1 gene in their genome
Proteins detected by western blot
Pituophis catenifer sayi
Trachemys scripta elegans
European pond turtle
African clawed frog
Presence in fishes
European sea bass
Smaller spotted catshark
In reptiles either one or both ERKs are expressed
This raises the interesting question of why ERK2 is not detected in squamates. Quantitative measurements of erk1/2 mRNA expression reveal that erk1 mRNA expression is 12- to 24-fold higher than erk2 in A. carolinensis tissues (mid-gestation embryo or brain, neo-natal brain and adult brain (Fig. 4d). This markedly contrasts with juvenile crocodile brain where erk2 mRNA expression is 8-fold greater than erk1 and with mouse NIH3T3 cells where erk2 mRNA expression is 3- to 4-fold greater than erk1. Conversely, in adult turtle brain, erk1 and erk2 mRNAs are expressed at similar levels (Fig. 4d). In all these animals, protein expression of ERK isoforms correlates well with mRNA expression therefore lizards express undetectable ERK2 protein due to minimal erk2 mRNA expression.
We then compared opposite mammalian versus squamate ERKs expression by cloning mouse and anole lizard erk1 and erk2 promoters upstream of the luciferase reporter gene (1 kb upstream initiating codon). In cells from these organisms, transient transfection revealed that mouse erk1 promoter is much weaker than mouse erk2 promoter; conversely A. carolinensis erk1 promoter is markedly stronger than A. carolinensis erk2 promoter (nearly 20-fold more in lizard fibroblasts, Fig. 4e). Hence, the low expression of ERK1 in mouse and low expression of ERK2 in anole lizard seem to originate from the weakness of their respective promoters. The difference of strength between mouse erk promoters is larger than the protein ratio observed in the same cells; therefore further research is needed to understand the individual contribution of promoters, RNA regulation and protein stability to establish the final ERK1/ERK2 protein ratio.
ERK1 evolve faster, but amino-acids required for kinase function are invariant in ERK1 and ERK2
ERK1s protein sequences evolve faster than ERK2s’, not their nucleotide sequences
Number of codons analysed
Nucleotide substitution rate
Amino acid substitution rate
Nonsynonymous substitution rate
Synonymous substitution rate
Nonsynonymous/synonymous substitution rate
(Ka/Ks Ratio of ERK1) / (Ka/Ks Ratio of ERK2)
The present study reveals that ERK1 and ERK2 identities are well separated in tetrapods, indicative that they arose from a single duplication event. The synteny between ERK isoforms with YPEL isoforms is conserved at least in human and zebrafish indicating that a segment of chromosome has duplicated. Furthermore, genome segments encompassing human ERK1 and human ERK2 display patterns of synteny with a single amphioxus linkage group . Taken together, these elements are indicative that ERK1 and ERK2 likely originated from the vertebrate-specific whole genome duplications (WGDs). Prior to bony fish, all vertebrates express only one ERK. Considering that two rounds of WGD occurred prior to divergence of gnathostome lineages , we conclude that paralogous ERK duplicates were lost in Chondrichthyes. The resolution of phylogenetic analysis in these deep vertebrate branches does not allow to classify the single erk of Agnathans as ERK1 or ERK2, however extant Chondrichthyes seem to express the ERK2 isoform, a conclusion confirmed by the large size of the erk gene of elephant shark fish, a unique characteristic of erk2 genes in all bony vertebrates.
It has been hypothesized that gene duplicates are usually lost during evolution after a WGD if no new function appear among paralogs . The predominant loss of ERK1 protein expression from cartilaginous fishes to birds, and lack of detectable ERK1 expression in crocodiles mimics the lack of major phenotype of mice invalidated for the erk1 gene. This could suggest that ERK2 carries ancestral and essential ERK function and therefore its expression is required in vertebrates. At the protein level, however the results presented here within the reptile clade shatter this hypothesis since we demonstrate that ERK2 is not detected in squamates. The identities of lizard ERK isoforms are kept since the sequence of ERK2 from the lizard A. carolinensis is very similar to the sequence of mouse ERK2 (97 % identity, Additional file 1B) and amino acid position that diverge between lizard ERK2 and human ERK2 are all localized distally from the face of the kinase (Additional file 1D). Similarly lizard ERK1 sequence is 91 % identical to that of mouse ERK1 and all divergent amino acids are located distally from the face of the kinase, except the conservative substitutions serine135-humanERK1 for threonine132-anolisERK1 and glutamic-acid194-humanERK1 for aspartic-acid191-anolisERK1 (Additional file 1E). Hence, in squamates, ERK1 and ERK2 have kept their identity but only ERK1 is significantly expressed. Though undetectable relatively to ERK1 by our methods, squamates’ ERK2 is certainly expressed since it is still under selective pressure to remain free of nonsense mutations. Similarly, in crocodiles ERK1 protein is undetectable relative to ERK2 and its sequence remains fully functional however, it is unlikely to play a specific role since in the same clade of Archosaurians since birds have lost erk1 altogether. The extinction of ERK1 protein expression in crocodiles may even be a premise to the loss of erk1 gene in birds since birds diverged later than crocodiles from their common ancestor. We propose that both ERKs contribute to reach a global threshold concentration which maintains selective pressure on isoforms’ functionality. Even in adults, a minimum threshold of ERK is required to sustain vertebrate life (as demonstrated whereby partial loss of ERK upon gene invalidation mediated by CRE-recombinase induction, leads to rapid death by multiple organ failures) . Considering that in vertebrates expressing both functional ERK proteins, ERK quantity can be overwhelmingly provided by ERK1 or by ERK2, our results suggest that ERK1 and ERK2 are functionally redundant. In mammals, two gene loci may increase the ability to regulate exquisitely the quantity of ERK during all life stages. During the revision of this manuscript, Meloche and co-workers showed that mice completely lacking ERK2 expression were rescued only when the ERK1 was overexpressed, suggesting a full redundancy of ERK1/2 proteins during development up to normal reproduction .
Our study suggests that ERK1 and ERK2 are functionally redundant in tetrapods since either one or both isoforms are expressed in animals from the same sauropsidian clade (reptiles and birds). ERK1 proteins evolve faster than ERK2 proteins, even when the study was restricted to the mammalian clade; however amino acids diverge only at position neutral for function hence we propose that the origin of this difference originates from genomic factors such as the large gene size difference, but not from functional differences. It remains to be determined whether the large gene size differences may also explain why erk1 genes are lost more readily than erk2’s. When evaluating ERK1/2 invalidations, observed phenotypes should now be linked to the quantitative amount of total-ERK reduction, not to functional differences among isoforms. ERKs being the most expressed proteins of this signaling cascade, future research should aim at understanding the regulation and role of total ERK quantity in cells.
Cell culture and reagents
Lizard embryo-fibroblasts preparation was adapted from : Eggs of mid-gestation embryos of A. carolinensis or A. sagrei were sterilized by three washes with 70 % ethanol and cut open by scalpel. Embryos were washed in sterile PBS, and sectioned into approximately 1 mm3 cubes to be incubated at 37 °C in DMEM medium complemented with 0.25 % trypsin (Gibco #15090046), 2 mg/ml collagenase (Sigma # C2674), 2 mg/ml DNAse (Sigma DN25) and 5 mM MgCl2. Gentle pipetting was applied to aid in cell dispersal. After 30 min treatment, the suspended cells were collected and centrifuged at 16.1x g for 5 min while chunks of tissues were incubated for an additional 30 min with fresh medium prior to centrifugation.
Supernatants were drawn off and cellular pellets were re-suspended in DMEM medium (GIBCO 61965–026) supplemented with 5 % chicken serum (sigma # C5405), 10 % FCS (Dominique Dutsher-South American serum), penicillin and streptomycin then plated on 6-well plates (Falcon #353046). Some cells were grown in the same medium supplemented with Nutrient mixture F (Sigma # N6013) with no detectable advantages. Plated cells were incubated at 37 °C in 5 % CO2 for 1 week. Cells were then cultivated in DMEM supplemented with 8 % FCS only. Cell-passage occurred by trypsinizing confluent cells following PBS washes, and plating cells with 3-fold dilution. Cells remain viable for less than 10 cell-passages. Cells observed to be rapidly dividing were frozen in aliquots in presence of 10 % DMSO, DMEM and 50 % serum to be studied later.
Mouse NIH 3 T3 cells, Human A375 and Chinese Hamster CCL39 cells were obtained from the AmericanType Culture Collection.
6 μl of Ribocellin (BioCellChalenge; France) was mixed with siRNAs in 100 μl PBS (each siRNA is transfected at a final concentration of 60nM; 180 nM for the control). This mix was added to lizard fibroblasts cultivated in 1 ml of serum-containing medium in 35 mm plates. Cells were cultivated in presence of siRNA for 3 days. Each extract loaded on gel represents harvesting from two 35 mm plates. siRNA were synthetized and annealed by Eurogentec (Belgium), sequences targeting A. carolinensis are presented in Fig. 4a and the Si-control (si-luciferase) sequence was: CGTACGCGGAATACTTCGA.
Brains (stored at −80 °C) were dissected, lysed in laemmli sample buffer, sonicated and then boiled at 95 °C for 8 min. Protein concentrations of brain extracts and cell lysates were measured by the bicinchoninic acid method (Pierce). Tissue samples from A. carolinenis where incubated in RNA later for shipment, chunks were isolated by centrifugation and frozen in liquid nitrogen prior being smashed into powder. Frozen powder was dissolved into triton lysis buffer containing proteases and phosphatases inhibitor for 30 min, and then centrifuged at 20000 g for 30 min. Supernatant was diluted 2 folds into 3X laemmli sample buffer prior boiling. Proteins were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (10 % acrylamide-bis acrylamide [29:1] gels) loaded with 15 μg to 40 μg of protein per lane depending on gel size. Proteins were transferred onto nitrocellulose membranes: Hybond RPN303D from Amersham. Membranes were blocked by incubating with 2.5 % BSA in PBS containing 0.12 % gelatin, 0.1 % casein. Antibodies were incubated in PBS containing 0.1 % gelatin and 0.08 % casein. Phosphorylated ERK1/2 was detected with the mouse monoclonal anti-phospho-ERK1/2 antibody (Sigma M8159), identical results were obtained with the rabbit monoclonal anti-phospho ERK (Cell Signaling #4370). Total ERK was detected with the rabbit monoclonal anti-ERK (Cell Signaling #4695), or the rabbit anti-rat ERK1 (1:4,000; Fisher 1019–9152) or mouse monoclonal ERK2 (1:5,000; BD Pharmingen #610103) plus mouse monoclonal ERK1 (1:1,000, BD Pharmingen #554100). Secondary antibodies were IR dyes anti-rabbit 800CW (1:1000) and anti-mouse 680RD (1:1000) from Li-cor.
erk coding sequences were gathered from genomic sequencing by searching the Ensembl database or by exon search by BLAST in NCBI database or ambystoma.org database. Hagfish erk sequence was obtained from transcriptomic sequencing. For animals with unknown erk sequences, tissues fragments were dissolved into 1 ml TRIzol reagent (# 15596–018 Life Technologies), aqueous phase containing RNA was separated by adding 20 % chloroform, then RNA isolation was performed with aqueous solution with RNeasy Mini kit (Qiagen # 74124) as per manufacturer specifications. cDNA was generated by oligo-dT or random hexamer priming with either omniscript enzyme (Qiagen # 205111 ), primescript enzyme (Clonetech # 2680A) or revert aid enzyme (Thermo-scientific EP0441) according to manufacturers’ specifications. For RACE PCR and extracts from organisms with high GC content, cDNA was generated by incubating the enzyme at gradually increasing temperature from 42 to 65 °C. First-strand cDNA samples were then incubated with RNaseH (New England Biolabs # M0297S) for RNA degradation prior to use in downstream PCR reactions. RACE PCR: polyA-cDNA was generated by incubating cDNA with terminal deoxynucleotidyl transferase (NEB # M0315S) in presence of dATP (200 μM) for 20 min at 37 °C, and incubation at 70 °C for 10 min.
PCR primers: for 5′RACE PCR, oligo-dT anchored primer was used for initial PCR cycles, GGACTCGAGTCGACATCGATTTTTTTTTTTTTTTTTGG, when nested PCR was required, second amplification required oligo GGACTCGAGTCGACATCGA. This oligonucleotide was also used for 3′RACE after 10 cycles with the primer: GGACTCGAGTCGACATCGATTTTTTTTTTTTTTTTTVN.
Degenerated oligonucleotides to amplify erk from dogfish were : >PL-09-10 CCATCAAAAAGATCAGCCCT; >PL-09-11 GGGTCATAGTACTGCTCCAGGTA; for skate > PL-11-23 CTTCTGCAGAGGACCTGAACTG; >PL-11-24 GGACCAGCTCAATCATATCCTTG;>PL-11-25 GCTGGCAGTATGTCTGGTGTTC > PL-11-26 CTCATGTTTGAAGCGCAGCAG; for Bichir erks a combination of oligonucleotides was used for nested PCR and extension of clones > PL-11-19 gtggccatcaagaagatcagccc > PL-11-20 Gagcaccagacwtactgccagcg; >PL-11-42 GACCTGAAGCCGTCCAACC; >PL-11-43 CATCCACTCAGCAAATGTTCTCC;>PL-11-44 CTTTGGATCTGCTGGATCGAATG; >PL-11-49 GATCTGAAACCCTCTAACTTGCTG; >PL-11-50 CAAATCCTGCGAGGCTTAAAGTA. Combinations of primers were used to clone erk isoforms from cDNAs, primers for Crocodylus niloticus erk1 were: >PL-12-07 GCGGCCATCAAGAAGATCAG; >PL-12-08 ATCGGCATCAACGACATCCT; >PL-12-09 GTACATCCACTCGGCCAACGT; >PL-12-10R GCCTTCATGTTGATGATGCAGTT; >PL-12-11R GCTTGTTGGGGTTAAAGGTCA; >PL-12-12R GGTCGTAGTACTGCTCCAGGTAGG; primers for cloning Trachemys scripta elegans erk1 were: >PL-10-16 GCTGTATCCTGGCGGAGATG; >PL-10-17 ATCATGCTCAACTCCAAGGGCTA; >PL-11-07 GCGGCCTCAAGTACATCCACTCG ; >PL-11-08 GCGATCTCAAGATCTGTGACTTTGGTC ; >PL-11-15 GCGCTACACGGACCTGCAGTACATC ;>PL-11-16 TACACGGACCTGCAGTACAT; >PL-11-17R GAGCTGGTCCAGGTAGTGCTTGCC ; >PL-11-18R CAAGCACTACCTGGACCAG; primers for cloning Anolis carolinensis erk1 were: >PL-12-16 GTAAAGCAGCAGCAGCACTAAGGA; >PL-12-17 GATTCTCCAGCGGATCGGC; >PL-09-04 TCATCGGCAT CAATGACATT; >PL-09-05 TAAAGACCCAGCAACTCAGCAA; >PL-09-06R GTGTGGTCATGGTCTGGATC; >PL-09-07R CCAGGAAAGATGGGTCGGTT; >PL-10-20 TCCGGGTAAAGCAGCAGCAGC; primers for cloning Crocodylus niloticus erk2 were: >PL-10-27 CCAACTATTGAGCAAATGAAAGATG; >PL-10-28 CAGCACCTCAGCAACGACCAC; >PL-11-67 CGTGCAAGATCTCATGGAGACAG; >PL-12-13 CTACAGGCTCATCACTTGGATCGTA; >PL-12-14R GGCTGGAATCTAGCAGTTTCCTC; >PL-11-68R CATCTTTCATTTGCTCAATAGTTGG; >PL-11-69R GTGGTCGTTGCTGAGGTGCTG; >PL-11-70R CTGTCTCCATGAGATCTTGCACG; >PL-11-71 GTGGCCTATCTCAGCTTATCCAC; primers for cloning Anolis carolinensis erk2 were: >PL-12-15 CCTCTTCCTCCTCGCTGTTTC; >PL-09-24 TATTTCCTTTATCAAATCCTGAG; >PL-09-25R CAGCTGATCGAGATAGTGTTTC; >PL-10-21 GAGTTGTGTCTTCCTTCGCCTTCC; >PL-10-22R CCTCTCAGGATYTGATAAAG; >PL-11-11 GCGAGAAATCAAAATCCTCT; >PL-11-12R CGGCACTTTGTTTTTGTAC; pri-mers for cloning Trachemys scripta elegans erk2 were: >PL-10-23 CGGGGGTCCGGAGATGGT; >PL-10-24 TACGGCATGGTCTGTTCTGCCTA; >PL-11-09 CCACACGTTGGTACAGAGCACCTG; >PL-11-10R AAGTATGGTCAGCAGTGGGTGCAC; >PL-11-06 GAAGTGGAGCAAGCTTTGGC. Fragments subcloned for normalization of absolute RT-qPCR were obtained with the following primers: Anolis carolinensis erk1 > PL-12-17 GATTCTCCAGCGGATCGGC and > PL-09-07 CCAGGAAAGATGGGTCGGTT Anolis carolinensis erk2 > PL12-15 CCTCTTCCTCCTCGCTGTTTC and > PL-12-13 CTACAGGCTCATCACTTGGATCGTA; Trachemys scripta elegans erk1 > PL-11-16 tacacggacctgcagtacat and > PL-09-09 GAGCTGGTCCAGGTAGTGCTTG Trachemys scripta elegans erk2: >PL-10-24 TACGGCATGGTCTGTTCTGCCTA and > PL-11-04 GCTTACTAATGCATATGGATGC; Crocodylus niloticus erk1: >PL-12-07 GCGGCCATCAAGAAGATCAG and > PL-12-12 GGTCGTAGTACTGCTCCAGGTAGG; Crocodylus niloticus erk2: >PL-11-67 CGTGCAAGATCTCATGGAGACAG and > PL-12-13 CTACAGGCTCATCACTTGGATCGTA.
RNAs were processed as described in the previous section, for quantification of Fig. 4, RNAse-free DNAse was added in solution prior to extracting RNA from columns (Qiagen #79254) and cDNA was generated with nonamers or hexamers priming by Qiagen Omniscript kit as recommanded by supplier; for quantifications of Additional file 4, cDNA was generated with Qiagen Quantiscript kit with nonamers (genomic DNA diminished by g-removal buffer). Quantitative PCR was performed on Applied Biosystems 7300 and Step One plus Real Time PCR System. For sybr green detection, mouse erk1 was amplified with primers > ERK1-03mForward CCTGCTGGACCGGATGTTA and > ERK1-03mReverse TGAGCCAGCGCTTCCTCTAC, mouse erk2 was amplified with primers > ERK2-02mForward GGAGCAGTATTATGACCCAAGTGA > ERK2-02mReverse TCGTCCACTCCATGTCAAACT, anolis erk1 was amplified with primers > anolisERK1-5forward CAAGATTCGAGCTGCCATCA and > anolisERK1-5reverse GCAGTGTGCGTTGGCAATAA, anolis erk2 was amplified with primers > anolisERK2-Cforward AAACAAAGTGCCGTGGAACAG and > anolisERK2-C reverse GGATGGGCCAAAGCTTCTTC. For TaqMan quantification primer mix were from AppliedBiosystems, Mn01973540-g1 for ERK1 and Mn00442479-m1 for ERK2. Quantity of molecules was determined by normalisation with linearized plasmid. For mouse erk1 and erk2, full length cDNAs cloned were used ; for A. carolinenis, T. scripta elegans and C. niloticus, the plasmids with partial cDNA were cloned into invitrogen PCR2topo vector (sequences of the primers presented in the section DNA sequences). Dilutions of linearized plasmids ranged from 30 to 300 000 molecules per reaction. Each reaction was performed with cDNAs generated from 25 ng of RNA.
ProtTest  was used to evaluate the best model to use with the MaximumLikelihood method and the protein sequences for Fig. 2b. The tree obtained with the best model showed little difference with the tree resulting from the consensus of all model. In addition sequences were analyzed using SeaView . Sequences were aligned as amino-acid sequences and then back converted to nucleotide sequences (as implemented in SeaView) and alignments were manually checked. Trees were then built using the BioNJ (Kimura correction), Parsimony and Maximum likelyhood methods as implemented in SeaView.
For computations of Shannon entropy, an ad-hoc Python program was written. For nucleotide sequences, indels were treated as a 5th character. Since erk1 and erk2 differ in length, positions that are all indels in one of the gene have an energy of 0; this can be clearly misleading, but it allowed direct comparison of other positions.
For the identification of Site-Specific Positive or Purifying selections, we used the Selecton program, locally compiled under Linux and with default values .
For the calculation of global Ka/Ks ratios of Table 2, the mean nucleotide substitution rate, amino acid substitution rate, nonsynonymous substitution rate (Ka), synonymous substitution rate (Ks), and nonsynonymous to synonymous substitution rate ratio (Ka/Ks) were calculated for ERK1 and ERK2 CDS alignments in MEGA 6 . The shark ERK sequences were excluded in this calculation to ensure our calculation for Erk1 and Erk2 covered exactly the same species sampling. 94 mamalian sequences and 28 vertebrate sequences corresponding to the species’ list of Fig. 2 were studied. The Jukes-Cantor correction model  was used when calculating nucleotide substitution rate; the Poisson model  was used when calculating amino acid substitution rate; and the Nei-Gojobori model  was used when calculating Ka and Ks. The mean Ka/Ks value was calculated by calculating Ka/Ks for each pairwise comparison first and then taking the average.
3D structures used to generate still images
3D structures solved from ERK crystals were recovered from Protein Data Bank (PDB): 1ERK (rat ERK2) ; 4QTB and 4QTA (respectively human ERK1 and ERK2 bound to inhibitor SCH772984) . Sequences were opened with the freely available 3D software Cn3D (from NCBI), and amino-acids to be highlighted were colored with the annotation tool. (http://www.ncbi.nlm.nih.gov/Structure/CN3D/cn3d.shtml). Still images were generated when 3D structure was rotating in the software Cn3D.
For all structures, threonine and tyrosine phosphorylated upon activation by MEK are colored in white; the amino-acids interacting with substrates forming the DEJL(KIM) motif are colored in blue, light blue for docking groove and darker blue for acidic patch; the amino-acids interacting with substrates via the DEF domain (FXFP) are colored in green.
We are grateful for the very generous gifts of animal material to: Dr. T. Cabana (Opossum), Dr. E. Houliston (Xenopus), Drs. M. Avella and P. Pierson (European seabass), Dr. J-Y Sire (Bichir and African jewelfish), Dr. Y. Andéol (Axolotl), Dr. P. Fafournoux (Greater Rhea, Chicken, Cattle, Rabbit), Dr. G. Anderson (Winter skate), Drs. R Caldwel and M. Erdmann (Coelacanth), Dr. Cagnol (Mallard Duck, Phaesant), Dr. S. Martin -La ferme aux crocodiles- (Crocodile, Turtles, Common agama), Mr. Daoues -La ferme tropicale- (Carolina Anole, Brown Anole, Corn Snake, Bullsnake, Eastern milksnake, Western terrestrial garter snake), Restaurant”la maison baron lefevre” (Lamprey), Mrs F. Latrache (Dogfish). We also thank F. Paput, J. Paput, Z. Dahmoul and Dr. J-C. Chambard for technical help. This work was supported by grant #333831 from Association pour la Recherche sur le Cancer to PL ,the Ligue Nationale Contre le Cancer (Equipe LNCC Labellisée, JP) and NSERC Discovery grant # 203736 to GG.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Chambard JC, Lefloch R, Pouyssegur J, Lenormand P. ERK implication in cell cycle regulation. Biochim Biophys Acta. 2007;1773(8):1299–310.View ArticlePubMedGoogle Scholar
- Tang N, Marshall WF, McMahon M, Metzger RJ, Martin GR. Control of mitotic spindle angle by the RAS-regulated ERK1/2 pathway determines lung tube shape. Science. 2011;333(6040):342–5.PubMed CentralView ArticlePubMedGoogle Scholar
- Kelleher 3rd RJ, Govindarajan A, Jung HY, Kang H, Tonegawa S. Translational control by MAPK signaling in long-term synaptic plasticity and memory. Cell. 2004;116(3):467–79.View ArticlePubMedGoogle Scholar
- Cui Y, Costa RM, Murphy GG, Elgersma Y, Zhu Y, Gutmann DH, et al. Neurofibromin regulation of ERK signaling modulates GABA release and learning. Cell. 2008;135(3):549–60.PubMed CentralView ArticlePubMedGoogle Scholar
- Torii S, Yamamoto T, Tsuchiya Y, Nishida E. ERK MAP kinase in G cell cycle progression and cancer. Cancer Sci. 2006;97(8):697–702.View ArticlePubMedGoogle Scholar
- Samuels IS, Karlo JC, Faruzzi AN, Pickering K, Herrup K, Sweatt JD, et al. Deletion of ERK2 mitogen-activated protein kinase identifies its key roles in cortical neurogenesis and cognitive function. J Neurosci. 2008;28(27):6983–95.PubMed CentralView ArticlePubMedGoogle Scholar
- Boulton TG, Nye SH, Robbins DJ, Ip NY, Radziejewska E, Morgenbesser SD, et al. ERKs: a family of protein-serine/threonine kinases that are activated and tyrosine phosphorylated in response to insulin and NGF. Cell. 1991;65(4):663–75.View ArticlePubMedGoogle Scholar
- Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417(6892):949–54.View ArticlePubMedGoogle Scholar
- Robbins DJ, Zhen E, Owaki H, Vanderbilt CA, Ebert D, Geppert TD, et al. Regulation and properties of extracellular signal-regulated protein kinases 1 and 2 in vitro. J Biol Chem. 1993;268(7):5097–106.PubMedGoogle Scholar
- Lefloch R, Pouyssegur J, Lenormand P. Single and combined silencing of ERK1 and ERK2 reveals their positive contribution to growth signaling depending on their expression levels. Mol Cell Biol. 2008;28(1):511–27.PubMed CentralView ArticlePubMedGoogle Scholar
- Lenormand P, Sardet C, Pagès G, L’Allemain G, Brunet A, Pouysségur J. Growth Factors Induce Nuclear Translocation of MAP Kinases (p42maPk and p44mapk) but not of Their Activator MAP Kinase Kinase (p45mapkk) in Fibroblasts. J Cell Biol. 1993;122:1079–89.View ArticlePubMedGoogle Scholar
- von Kriegsheim A, Baiocchi D, Birtwistle M, Sumpton D, Bienvenut W, Morrice N, et al. Cell fate decisions are specified by the dynamic ERK interactome. Nat Cell Biol. 2009;11(12):1458–64.View ArticleGoogle Scholar
- Lee T, Hoofnagle AN, Kabuyama Y, Stroud J, Min X, Goldsmith EJ, et al. Docking motif interactions in MAP kinases revealed by hydrogen exchange mass spectrometry. Mol Cell. 2004;14(1):43–55.View ArticlePubMedGoogle Scholar
- Liu S, Sun JP, Zhou B, Zhang ZY. Structural basis of docking interactions between ERK2 and MAP kinase phosphatase 3. Proc Natl Acad Sci U S A. 2006;103(14):5326–31.PubMed CentralView ArticlePubMedGoogle Scholar
- Saba-El-Leil MK, Vella FD, Vernay B, Voisin L, Chen L, Labrecque N, et al. An essential function of the mitogen-activated protein kinase Erk2 in mouse trophoblast development. EMBO Rep. 2003;4(10):964–8.PubMed CentralView ArticlePubMedGoogle Scholar
- Yao Y, Li W, Wu J, Germann UA, Su MS, Kuida K, et al. Extracellular signal-regulated kinase 2 is necessary for mesoderm differentiation. Proc Natl Acad Sci U S A. 2003;100(22):12759–64.PubMed CentralView ArticlePubMedGoogle Scholar
- Hatano N, Mori Y, Oh-hora M, Kosugi A, Fujikawa T, Nakai N, et al. Essential role for ERK2 mitogen-activated protein kinase in placental development. Genes Cells. 2003;8(11):847–56.View ArticlePubMedGoogle Scholar
- Pages G, Guerin S, Grall D, Bonino F, Smith A, Anjuere F, et al. Defective thymocyte maturation in p44 MAP kinase (Erk 1) knockout mice. Science. 1999;286(5443):1374–7.View ArticlePubMedGoogle Scholar
- Zeng P, Wagoner HA, Pescovitz OH, Steinmetz R. RNA interference (RNAi) for extracellular signal-regulated kinase 1 (ERK1) alone is sufficient to suppress cell viability in ovarian cancer cells. Cancer Biol Ther. 2005;4(9):961–7.View ArticlePubMedGoogle Scholar
- Fremin C, Ezan F, Boisselier P, Bessard A, Pages G, Pouyssegur J, et al. ERK2 but not ERK1 plays a key role in hepatocyte replication: an RNAi-mediated ERK2 knockdown approach in wild-type and ERK1 null hepatocytes. Hepatology. 2007;45(4):1035–45.View ArticlePubMedGoogle Scholar
- Shin S, Dimitri CA, Yoon SO, Dowdle W, Blenis J. ERK2 but not ERK1 induces epithelial-to-mesenchymal transformation via DEF motif-dependent signaling events. Mol Cell. 2010;38(1):114–27.PubMed CentralView ArticlePubMedGoogle Scholar
- Fan HY, Liu Z, Shimada M, Sterneck E, Johnson PF, Hedrick SM, et al. MAPK3/1 (ERK1/2) in ovarian granulosa cells are essential for female fertility. Science. 2009;324(5929):938–41.PubMed CentralView ArticlePubMedGoogle Scholar
- Voisin L, Saba-El-Leil MK, Julien C, Fremin C, Meloche S. Genetic demonstration of a redundant role of extracellular signal-regulated kinase 1 (ERK1) and ERK2 mitogen-activated protein kinases in promoting fibroblast proliferation. Mol Cell Biol. 2010;30(12):2918–32.PubMed CentralView ArticlePubMedGoogle Scholar
- Smith JJ, Kuraku S, Holt C, Sauka-Spengler T, Jiang N, Campbell MS, et al. Sequencing of the sea lamprey (Petromyzon marinus) genome provides insights into vertebrate evolution. Nat Genet. 2013;45(4):415–21. 421e411-412.PubMed CentralView ArticlePubMedGoogle Scholar
- Venkatesh B, Lee AP, Ravi V, Maurya AK, Lian MM, Swann JB, et al. Elephant shark genome provides unique insights into gnathostome evolution. Nature. 2014;505(7482):174–9.PubMed CentralView ArticlePubMedGoogle Scholar
- Near TJ, Eytan RI, Dornburg A, Kuhn KL, Moore JA, Davis MP, et al. Resolution of ray-finned fish phylogeny and timing of diversification. Proc Natl Acad Sci U S A. 2012;109(34):13698–703.PubMed CentralView ArticlePubMedGoogle Scholar
- Lefloch R, Pouyssegur J, Lenormand P. Total ERK1/2 activity regulates cell proliferation. Cell Cycle. 2009;8(5):705–11.View ArticlePubMedGoogle Scholar
- Alfoldi J, Di Palma F, Grabherr M, Williams C, Kong L, Mauceli E, et al. The genome of the green anole lizard and a comparative analysis with birds and mammals. Nature. 2011;477(7366):587–91.PubMed CentralView ArticlePubMedGoogle Scholar
- Li M, Liu J, Zhang C. Evolutionary history of the vertebrate mitogen activated protein kinases family. PLoS One. 2011;6(10):e26999.PubMed CentralView ArticlePubMedGoogle Scholar
- Putnam NH, Butts T, Ferrier DE, Furlong RF, Hellsten U, Kawashima T, et al. The amphioxus genome and the evolution of the chordate karyotype. Nature. 2008;453(7198):1064–71.View ArticlePubMedGoogle Scholar
- Aury JM, Jaillon O, Duret L, Noel B, Jubin C, Porcel BM, et al. Global trends of whole-genome duplications revealed by the ciliate Paramecium tetraurelia. Nature. 2006;444(7116):171–8.View ArticlePubMedGoogle Scholar
- Blasco RB, Francoz S, Santamaria D, Canamero M, Dubus P, Charron J, et al. c-Raf, but not B-Raf, is essential for development of K-Ras oncogene-driven non-small cell lung carcinoma. Cancer Cell. 2011;19(5):652–63.View ArticlePubMedGoogle Scholar
- Frémin C, Saba-El-Leil MK, Lévesque K, Ang S-L, Meloche S. Functional Redundancy of ERK1 and ERK2 MAP Kinases During Development. Cell Reports. 2015;12:913-21.View ArticlePubMedGoogle Scholar
- Choi SS, Hannenhalli S. Three independent determinants of protein evolutionary rate. J Mol Evol. 2013;76(3):98–111.View ArticlePubMedGoogle Scholar
- Pal C, Papp B, Hurst LD. Highly expressed genes in yeast evolve slowly. Genetics. 2001;158(2):927–31.PubMed CentralPubMedGoogle Scholar
- Liao BY, Scott NM, Zhang J. Impacts of gene essentiality, expression pattern, and gene compactness on the evolutionary rate of mammalian proteins. Mol Biol Evol. 2006;23(11):2072–80.View ArticlePubMedGoogle Scholar
- Marais G, Nouvellet P, Keightley PD, Charlesworth B. Intron size and exon evolution in Drosophila. Genetics. 2005;170(1):481–5.PubMed CentralView ArticlePubMedGoogle Scholar
- Polazzi E, Alibardi L. Cell culture from lizard skin: a tool for the study of epidermal differentiation. Tissue Cell. 2011;43(6):350–8.View ArticlePubMedGoogle Scholar
- Darriba D, Taboada GL, Doallo R, Posada D. ProtTest 3: fast selection of best-fit models of protein evolution. Bioinformatics. 2011;27(8):1164–5.View ArticlePubMedGoogle Scholar
- Gouy M, Guindon S, Gascuel O. SeaView version 4: A multiplatform graphical user interface for sequence alignment and phylogenetic tree building. Mol Biol Evol. 2010;27(2):221–4.View ArticlePubMedGoogle Scholar
- Stern A, Doron-Faigenboim A, Erez E, Martz E, Bacharach E, Pupko T. Selecton 2007: advanced models for detecting positive and purifying selection using a Bayesian inference approach. Nucleic Acids Res. 2007;35(Web Server issue):W506–11.PubMed CentralView ArticlePubMedGoogle Scholar
- Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol. 2013;30(12):2725–9.PubMed CentralView ArticlePubMedGoogle Scholar
- Jukes TH, Cantor CR. Evolution of protein molecules. New York: Academic; 1969.View ArticleGoogle Scholar
- Zuckerkandl E, Pauling L. Evolutionary divergence and convergence in proteins. New York: Academic; 1965.View ArticleGoogle Scholar
- Nei M, Gojobori T. Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol Biol Evol. 1986;3(5):418–26.PubMedGoogle Scholar
- Zhang F, Strand A, Robbins D, Cobb MH, Goldsmith EJ. Atomic structure of the MAP kinase ERK2 at 2.3 A resolution. Nature. 1994;367:704–11.View ArticlePubMedGoogle Scholar
- Chaikuad A, Tacconi EM, Zimmer J, Liang Y, Gray NS, Tarsounas M, et al. A unique inhibitor binding site in ERK1/2 is associated with slow binding kinetics. Nat Chem Biol. 2014;10(10):853–60.View ArticlePubMedGoogle Scholar