Download Evolution of the Insulin Receptor Family and

Evolution of the Insulin Receptor Family and Receptor Isoform Expression in
Vertebrates
Catalina Hernández-Sánchez,* Alicia Mansilla,* Flora de Pablo,* and Rafael Zardoyaà
*3D Lab (Development, Differentiation & Degeneration), Department of Cellular and Molecular Physiopathology, Centro de
Investigaciones Biológicas, Consejo Superior de Investigaciones Cientı́ficas (CSIC), Ramiro de Maeztu 9, Madrid, Spain; Centro de
Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas (CIBERDEM), Ramiro de Maeztu, 9, Madrid, Spain; and
àDepartmento de Biodiversidad y Biologı́a Evolutiva, Museo Nacional de Ciencias Naturales, CSIC, José Gutiérrez Abascal, 2,
Madrid, Spain
The molecular phylogeny of the vertebrate insulin receptor (IR) family was reconstructed under maximum likelihood (ML)
to establish homologous relationships among its members. A sister group relationship between the orphan insulin–related
receptor (IRR) and the insulin-like growth factor 1 receptor (IGF1R) to the exclusion of the IR obtained maximal bootstrap
support. Although both IR and IGF1R were identified in all vertebrates, IRR could not be found in any teleost fish. The
ancestral character states at each position of the receptor molecule were inferred for IR, IRR þ IGF1R, and all 3 paralogous
groups based on the recovered phylogeny using ML in order to determine those residues that could be important for the
specific function of IR. For 18 residues, ancestral character state of IR was significantly distinct (probability .0.95) with
respect to the corresponding inferred ancestral character states both of IRR þ IGF1R and of all 3 vertebrate paralogs. Most
of these IR distinct (shared derived) residues were located on the extracellular portion of the receptor (because this portion
is larger and the rate of generation of IR shared derived sites is uniform along the receptor), suggesting that functional
diversification during the evolutionary history of the family was largely generated modifying ligand affinity rather than
signal transduction at the tyrosine kinase domain. In addition, 2 residues at positions 436 and 1095 of the human IR
sequence were identified as radical cluster-specific sites in IRR þ IGF1R. Both Ir and Irr have an extra exon (namely exon
11) with respect to Igf1r. We used the molecular phylogeny to infer the evolution of this additional exon. The Irr exon 11
can be traced back to amphibians, whereas we show that presence and alternative splicing of Ir exon 11 seems to be
restricted exclusively to mammals. The highly divergent sequence of both exons and the reconstructed phylogeny of the
vertebrate IR family strongly indicate that both exons were acquired independently by each paralog.
Introduction
Insulin and insulin-like growth factors (IGFs) constitute
a fundamental family of hormone polypeptides common to
all metazoans. These hormones control essential functions
including cell growth, metabolism, reproduction, and longevity (Kimura et al. 1997; Efstratiadis 1998; Tissenbaum
and Ruvkun 1998; Brogiolo et al. 2001; Nakae et al. 2001;
Saltiel and Kahn 2001; Holzenberger et al. 2003; Nef et al.
2003). Dysfunction of these factors in humans is associated
to several pathological disorders such as diabetes, dwarfism,
and cancer. In invertebrates, insulin and IGF have a general
function as mitogenic growth factors (Chan and Steiner
2000). In postnatal vertebrates, the cell proliferation function
has been restricted to IGF1 and IGF2, whereas insulin has
become a metabolic regulatory hormone mainly controlling
homeostasis of different metabolites (most prominently glucose) (Chan and Steiner 2000). However, during embryonic
development, insulin action and regulation in vertebrates appear to be reminiscent of those found in invertebrates
(Hernandez-Sanchez et al. 2006). Physiological functions
of the insulin and IGF polypeptides require specific surface
cell receptors, and subtle differences in the structure and
function of the receptors can account for important variations
in the biological activity of the hormones across metazoans.
Although in invertebrates there are several insulin-like
peptides, only 1 insulin receptor (IR) protein has been described (Fernandez et al. 1995; Pashmforoush et al. 1996;
Kimura et al. 1997; Ruvkun and Hobert 1998). However, in
Key words: insulin receptor, alternative splicing, ancestral character
states.
E-mail: [email protected]
Mol. Biol. Evol. 25(6):1043–1053. 2008
doi:10.1093/molbev/msn036
Advance Access publication February 29, 2008
Ó The Author 2008. Published by Oxford University Press on behalf of
the Society for Molecular Biology and Evolution. All rights reserved.
For permissions, please e-mail: [email protected]
vertebrates, 3 distinct receptors that can bind with highaffinity insulin and the IGF were described based on differences in primary structure and function: the IR (Ebina et al.
1985; Ullrich et al. 1985), the type 1 IGF receptor (IGF1R)
(Ullrich et al. 1986), and the type 2 IGF receptor (IGF2R)
(Morgan et al. 1987). Of these, the IGF2R is in fact the
mannose-6-phosphate receptor that, only in mammals, has acquired a binding domain for IGF2, and it is not a signaling
receptor (Morgan et al. 1987). In addition, an orphan receptor (with an unknown ligand) termed the insulin receptor-related receptor (IRR) was also described as a member of the
IR family based on sequence similarity (Shier and Watt
1989). The IR, IGF1R, and IRR present a rather conserved
protein structure (see known domains in fig. 1) (Ullrich et al.
1986; De Meyts 2004) and belong to the larger tyrosine kinase receptor superfamily (Hubbard and Till 2000). Unlike
other members of this superfamily, the above-mentioned 3
receptors form dimeric (a2/b2) structures in the cell membrane, which can be either homodimers, composed by 2
identical a/b momomers, or heterodimers formed by 2 different a/b monomers (e.g., IRab/IGF1Rab) (Moxham et al.
1989; Soos and Siddle 1989; Schlessinger 2000; Fernandez
et al. 2001). Ligand binding to IR and IGF1R triggers a conformational change that enables autophosphorylation of the
receptor cytoplasmic tyrosine residues and initiates a cascade
of intracellular signaling events that engender diverse biological responses (metabolism, cell proliferation, cell differentiation, survival, and growth), depending on the cell type
and the developmental and functional stage. IR and IGF1R
have different but overlapping physiological functions (reviewed in Nakae et al. [2001]). The evolutionary and molecular mechanism through which the functional specialization
of each receptor was achieved remains an open question.
Thus far, the exact mechanism through which the orphan
IRR is activated and its function are unknown.
1044 Hernández-Sánchez et al.
The genes encoding IR, IGF1R, and IRR share similar
genomic organization. Both the a and b chains are synthesized from a unique mRNA, which is comprised by 22
exons in IR and IRR and by 21 exons in IGF1R (Rosenfeld
and Roberts 1999). In both, Ir and Irr, the extra exon with
respect to Igf1r is exon 11. Strikingly, exon 11 is constitutive in Irr whereas each of the human and murine Ir exon 11
is alternatively spliced, which results in 2 protein isoforms
(IRA and IRB) that differ by the absence or presence of 12
amino acids at the C-terminus of the a subunit, respectively
(Ebina et al. 1985; Ullrich et al. 1985; Seino and Bell 1989;
Seino et al. 1989). Both IR isoforms display differences in
ligand affinity binding, kinase activity, receptor internalization, and recycling as well as intracellular signaling capacity and tissue distribution (Mosthaf et al. 1990; McClain
1991; Vogt et al. 1991; Yamaguchi et al. 1991; Kellerer
et al. 1992; Leibiger et al. 2001).
In the present study, we reconstructed the molecular
phylogeny of the vertebrate IR family in order to establish
homologous relationships among its members. We also
identified evolutionarily conserved and functionally divergent amino acid residues in the 3 vertebrate receptors, as
well as shared derived residues of IR in order to gain insights on the evolutionary mechanisms underlying the functional diversification of the family and to identify those
residues that may be responsible for the specific function
of IR. In addition, we traced the presence of the alternatively spliced Ir exon 11 in the recovered phylogeny in order to characterize the evolution of this extra exon, and
found that it is a novel acquisition of mammals.
Materials and Methods
Animals
Fertilized White Leghorn (Gallus gallus) eggs (Granja
Rodrı́guez-Serrano, Salamanca, Spain) were incubated at
38.4 °C and 60–90% relative humidity for the time periods
indicated, and the embryos were staged according to
(Hamburger and Hamilton 1951). The 10-day posthatching
chickens (P) were from Avı́cola Grau (Madrid, Spain).
Frogs (Xenopus laevis) were kindly supplied by Dr MJ
Delgado (Universidad Complutense de Madrid). The 10-day
and 35-day postnatal mice (C57BL/6) (Mus musculus) were
from Centro de Investigaciones Biológicas stabularium. All
animals were handled according to European Union Guidelines for animal research.
RNA Isolation and Reverse Transcriptase–Polymerase
Chain Reaction
Total RNA from tissues was isolated using Trizol reagent (Invitrogen, Carlsbad, CA). The reverse transcriptase
Fig. 1.—Diagram of the a2/b2 quaternaty structure of the IR
showing the protein domain organization. L1 and L2, large domains 1 and
2 (leucine-rich repeats); CR, Furin-like cysteine-rich domain; FnIII-1,
FnIII-2, FnIII-3, fibronectin type III domains; ID, insert domain in FnIII-2;
TM, transmembrane domain; JM, juxtamembrane domain; TK, tyrosine
kinase domain; and CT, carboxy-terminal tail. Disulphure bonds are
shown. Arrowheads on the left side of the diagram indicate IR shared
derived amino acids, whereas lines on the right side of the diagram indicate
amino acids conserved in all 3 vertebrate members of the IR family.
Evolution of the Insulin Receptor Family in Vertebrates 1045
reaction was typically performed with 5 lg RNA, the Superscript III Kit, and oligo-dT primer (all from Invitrogen),
followed by amplification with the Expand High fidelity
Polymerase (Roche Diagnostics, Mannheim, Germany).
The mouse Ir was amplified using the sense primer 5#GGCCAGTGAGTGCTGCTCATGC-3# (mP1) and the antisense primer 5#-TGTGGTGGCTGTCACATTCC-3#
(mP2). The chicken Ir was amplified using the sense primer
5#-CAGAAGGAGCTGGAGGAGTC-3# (cP1) and the antisense primer 5#-TCTGCTCCTCTGCACTCTC-3# (cP2)
for the first polymerase chain reaction (PCR) and cP1 sense
primer and the antisense (cP4) 5#-GGAGCCCAGGTCTCTTCTCT-3# for the nested PCR. The Xenopus Ir
was amplified using the sense primer 5#-ACCTTCATCCAAGTGCTGTC-3# (xP1) and the antisense primer 5#-CAGAGTTCCATTGGCTACTC-3# (xP2) for the first PCR
and the sense 5#-GCCTTCCAGAACTTGGACTC-3#
(xP3) and the antisense 5#-TGGCTCTGTTTCATCCGGAG-3# (xP4) for the nested PCR.
Sequences
Molecular databases at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov)
were screened for vertebrate IRs using the Blast search program (Altschul et al. 1997) and the human IR
(NP_000199.2) as query. In addition, some additional vertebrate IRs were directly retrieved through searches in ENSEBML (http://www.ensembl.org/index.html).
Phylogenetic Analysis
A total of 39 complete or almost complete IR family
proteins were included in the phylogenetic analyses. Sequences were aligned using MUSCLE version 3.633 (Edgar
2004). Multiple alignments were subsequently refined by
eye and compared with IR alignments at the Receptors
for Insulin and Insulin-like molecules (RILM) database
(http://www.biochem.ucl.ac.uk/RILM; Garza-Garcia et al.
2007). Ambiguous alignments in highly variable (gap-rich)
regions were excluded from phylogenetic analyses (aligned
sequences and the exclusion sets are available from the authors upon request). We used PROTTEST version 1.2.6
(Abascal et al. 2005) to select the substitution model that best
fit the empirical data set (JTT þ I þ C; a 5 0.86, I 5 0.09)
and PHYML version 2.4.4 (Guindon and Gascuel 2003) to
find the maximum likelihood (ML) tree. The robustness of
the inferred tree was assessed using bootstrapping (500
pseudoreplicates) as implemented in PHYML.
The recovered ML tree was used as framework to
identify those protein residues that are shared derived by
IR orthologs. In a first round of analysis, each of the protein
residues was mapped onto the phylogeny, and the ancestral
character states and shared derived amino acid residues
were inferred with parsimony using PAUP* version
4.0b10 (Swofford 2002) and MacClade v4.0 (Maddison WP
and Maddison DR 1992) and taking into account only those
shared derived characters that had a consistency index (CI)
(as a measure of the fit of each character to the tree) above
0.5. In a second round of analysis, the likelihood of the an-
cestral character state of the identified potential shared derived amino acids of IR was estimated using BayesTraits
version 1.0 (Pagel et al. 2004). Finally, the posterior probability for functional cluster-specific (type II; Gu 2006) residues was estimated for IR versus IRR þ IGF1R using
Diverge 2.0 (Gu 2006). No attempt of testing for positive
selection events based on branch-site likelihood ratio tests
(Yang and Nielsen 2002) was made because of the relatively high nucleotide sequence divergences found among
vertebrate IRs and the existence of saturation of silent mutations at third codon positions (data not shown).
Results
Phylogeny of Vertebrate IRs
The phylogeny of the vertebrate IR family was reconstructed in order to establish homologous relationships
among its members. Phylogenetic analyses were based
on an amino acid sequence data alignment including
1,596 positions, of which 335 highly variable positions
were excluded due to ambiguity in positional homology assignment. A total of 250 positions were invariant, and 813
characters were parsimony informative. The alignment was
used to map sequence divergence along the receptor molecule. Although invariant sites were found throughout the
IR sequence (fig. 1), they were particularly abundant
around positions 90–110 and 1000–1250 (human IR sequence; fig. 2). In contrast, positions around 650–810,
890–990, and at the carboxy end (above position 1310)
were relatively variable and showed few invariant positions. The ML analysis of the molecular data set using
the IRs of 2 basal chordates, Ciona intestinalis and Branchiostoma lanceolatum, as outgroup sequences recovered
the phylogenetic tree (logL 5 32422.12) shown in
figure 3. According to the reconstructed tree, vertebrate
IRs could be grouped with maximal bootstrap support into
3 distinct paralogous groups, which correspond to IR, IRR,
and IGF1R, respectively. A sister group relationship between IRR and IGF1R to the exclusion of IR obtained maximal bootstrap support. As expected, the phylogeny of
vertebrates (teleost fish, (amphibians, (birds, mammals)))
was recovered within each paralog. However, although
both IR and IGF1R were found in all vertebrates, IRR could
not be identified in any teleost fish. Mean (±standard deviation [SD]) amino acid sequence divergences between
both outgroups and the 3 vertebrate paralogs, IR, IRR,
and IGF1R, were 0.54 ± 0.06, 0.57 ± 0.04, and
0.54 ± 0.05, respectively. Mean (±SD) amino acid sequence divergences within IR, IRR, and IGF1R were
0.15 ± 0.01, 0.21 ± 0.01, and 0.20 ± 0.01, respectively.
Mean (±SD) amino acid sequence divergences across placentals for IR, IRR, and IGF1R were 0.07 ± 0.05,
0.12 ± 0.03, and 0.04 ± 0.03, respectively.
Ancestral Character State Reconstruction, Shared
Derived Characters of IR, and Cluster-Specific
Functionally Divergent Sites
Ancestral character state reconstruction analyses were
used 1) to identify those residues that being shared derived
of IR could be responsible for its specific function as well as
1046 Hernández-Sánchez et al.
Fig. 2.—Alignment of selected members of the vertebrate IR family. Amino acid positions correspond to the human IR (NP_000199.2). Black
boxes represent IR shared derived amino acids, whereas gray boxes indicate amino acids that are conserved in all 3 members of the vertebrate IR family.
Main domains of the receptor are shown (see also IR alignments at the RILM database: http://www.biochem.ucl.ac.uk/RILM; Garza-Garcia et al. 2007).
2) to determine their distribution along the receptor molecule. In a preliminary filtering analysis, a total of 41 amino
acid residues were identified as potential shared derived
characters of IR under the parsimony criterion (CI
.0.50) (data not shown). The putative character states of
the 41 residues in the common ancestor of IR, IRR þ
IGF1R, and all 3 vertebrate paralogs were estimated under
the likelihood criterion, respectively (table 1). For 18
Evolution of the Insulin Receptor Family in Vertebrates 1047
Fig. 3.—Phylogenetic analysis of the vertebrate IR family. The ML phylogram is shown. Numbers in nodes represent bootstrap values above 70%.
GenBank or Ensembl accession numbers are in brackets. *Ornithorhynchus and Ciona sequences were manually assembled from data available in
Ensembl.
1048 Hernández-Sánchez et al.
Table 1
Ancestral Character State Reconstruction under the Likelihood Criterion
AA position
Ir
137a
M (0.99)
Igf1r þ Irr
T
K
R
G
Ir þ Igf1r þ Irr
(0.64)
(0.16)
(0.14)
(0.06)
192a
G (1)
E
F
R
S
D
K
(0.36)
(0.15)
(0.15)
(0.15)
(0.11)
(0.04)
266
E (0.97)
P
Q
R
K
W
T
A
Q
P
R
W
E
K
(0.33)
(0.30)
(0.16)
(0.07)
(0.04)
(0.04)
(0.04)
(0.33)
(0.26)
(0.15)
(0.11)
(0.06)
(0.05)
F
L
R
K
L
A
283
(0.94)
(0.03)
(0.39)
(0.33)
(0.20)
(0.08)
L
F
K
R
(0.39)
(0.33)
(0.16)
(0.07)
284a
S (1)
287a
(0.51)
(0.49)
(0.46)
(0.27)
(0.25)
E (0.37)
D (0.35)
A (0.28)
Q
Y
S
G
A
A (0.53)
E (0.34)
D (0.10)
S (0.63)
G (0.23)
A (0.1)
T (0.97)
E (0.90)
K (0.04)
S (0.02)
AA position
Ir
735a
Y (1)
806
S (0.99)
839
Y (1)
841a
S (0.86)
T (0.11)
883
L (1)
886a
V (1)
Igf1r þ Irr
F (1)
R (0.99)
F (0.99)
F (0.72)
Y (0.27)
H (0.99)
I (0.92)
L (0.08)
Ir þ Igf1r þ Irr
F(1)
S (0.89)
F (0.05)
R (0.04)
Y (0.99)
Y (0.86)
F (0.12)
L (0.86)
H (0.14)
I (0.97)
a
304a
Y (1)
364a
N (1)
436
R (1)
F (0.99)
K (0.59)
E (0.27)
D (0.14)
Q (0.99)
F (1)
K (0.51)
E (0.36)
D (0.14)
C
G
P
A
911
(0.61)
(0.28)
(0.10)
(0.96)
C (0.60)
G (0.27)
P (0.09)
H
Y
F
N
V
R
A
Q
456a
(0.49)
(0.51)
(0.21)
(0.17)
(0.16)
(0.15)
(0.15)
(0.15)
R (0.99)
A
V
F
N
Q
R
(0.29)
(0.18)
(0.17)
(0.15)
(0.11)
(0.1)
940
T (0.99)
941
Y (0.97)
956
K (0.96)
V
M
I
L
A
(0.27)
(0.26)
(0.24)
(0.19)
(0.04)
S
H
F
V
D
(0.28)
(0.22)
(0.16)
(0.15)
(0.14)
L
I
M
T
V
(0.43)
(0.22)
(0.14)
(0.11)
(0.10)
Y
H
D
T
V
A
F
(0.58)
(0.14)
(0.07)
(0.06)
(0.04)
(0.04)
(0.03)
T
M
L
A
V
F
I
K
F
A
L
I
M
T
V
(0.18)
(0.18)
(0.16)
(0.14)
(0.12)
(0.11)
(0.11)
(0.26)
(0.23)
(0.11)
(0.10)
(0.09)
(0.08)
(0.07)
(0.06)
IR-derived state.
residues, the inferred ancestral character state of the IR paralog was significantly distinct (probability .0.95) with respect to the corresponding inferred ancestral character states
both of IRR þ IGF1R and of all 3 vertebrate paralogs
(table 1). These IR distinct (shared derived) residues were
distributed rather evenly along the IR molecule (figs. 1
and 2). Moreover, 14 out of the 18 residues mapped on
the extracellular portion of the receptor, whereas the remaining 4 residues were located on the intracellular region (figs. 1
and 2). Given the relative proportions of the extracellular
(70%) and intracellular (29%) portions of the receptor, an
expected random distribution of the shared derived residues
would have been 12 and 6, respectively. The difference between the observed and expected frequencies was not statistically significant according to a chi-square test (P . 0.05).
The physicochemical nature of the amino acid change
leading to the ancestral IR character state for each of the
detected IR shared derived positions was characterized.
In 2 instances (positions 137 and 192 of the human IR sequence), the change from the ancestral vertebrate IR character state to the ancestral IR character state implied
a replacement of a polar residue by a nonpolar one, whereas
in another 2 cases (304 and 735), the change was in the
opposite direction. In 3 instances (364, 520, and 1128), a
polar residue was substituted by another polar residue,
whereas in another 3 cases (628, 640, and 886), a nonpolar
residue was replaced by another nonpolar residue. In another 8 cases (284, 287, 456, 531, 841, 995, 1026, and
1141), it was not possible to determine the physicochemical
nature of the change (table 1).
The inferred shared derived positions of IR do not
need to be necessarily conserved in the other paralogous
groups of the family. According to the posterior probabilities estimated in the ancestral character state reconstruction
analyses, in 10 (192, 284, 287, 364, 456, 520, 531, 628,
1026, and 1128) out of the 18 positions presenting an unambiguous shared derived amino acid state of IR, the inferred amino acid residue was conserved in the ancestor
of IR but was variable in the ancestors of the other 2 subsets
(IRR þ IGF1R and IR þ IRR þ IGF1R) of homologous
genes (table 1). In the remaining 8 (137, 304, 640, 735,
841, 886, 995, and 1141) positions, the ancestors of the
other 2 subsets of homologous genes (IRR þ IGF1R and
IR þ IRR þ IGF1R) also showed relatively unambiguous
character states (table 1). Only 1 (640) out of these 8 positions showed a distinct conserved amino acid in each of the
3 subsets, whereas in the other sites, IRR þ IGF1R and IR
þ IRR þ IGF1R shared the same ancestral character state.
Changes in the evolutionary conservation at a particular residue may reflect functional divergence after gene
Evolution of the Insulin Receptor Family in Vertebrates 1049
Table 1
Extended
AA position
Ir
520a
W (1)
529
G (1)
531a
M (1)
566
R (0.97)
620
T (1)
638
I (1)
640a
L (1)
698
N (0.80)
721
K (0.98)
731
T (0.90)
Igf1r þ Irr
Q
R
H
Y
(0.46)
(0.27)
(0.19)
(0.08)
S (0.99)
T
I
V
R
(0.51)
(0.25)
(0.15)
(0.10)
N (0.91)
S (0.09)
A (0.97)
L
V
M
A
L (1)
V (0.98)
D
P
S
T
E
(0.28)
(0.26)
(0.21)
(0.13)
(0.09)
A
M
R
E
(0.44)
(0.23)
(0.19)
(0.10)
N
S
V
I
R
(0.76)
(0.08)
(0.06)
(0.04)
(0.03)
Ir þ Igf1r þ Irr
R (0.51)
Q (0.40)
Y (0.06)
G (0.69)
S (0.31)
T (0.83)
V (0.16)
R (0.96)
T (0.95)
A (0.54)
M (0.36)
V (0.06)
I (0.93)
L (0.07)
P (0.98)
M
K
A
E
(0.65)
(0.17)
(0.09)
(0.07)
T
S
R
I
(0.58)
(0.21)
(0.11)
(0.08)
995a
P (0.51)
S (0.49)
1026a
L (1)
1128a
R (0.98)
1135
E (0.98)
Q (0.41)
S (0.20)
N (0.15)
H (0.07)
T (0.07)
P (0.04)
E (0.04)
1141a
A (0.50)
V (0.49)
AA position
Ir
960
G (1.00)
983
Q (0.81)
I (0.15)
986
G (0.88)
S (0.12)
Igf1r þ Irr
L (0.86)
M (0.06)
G (0.04)
E (0.33)
N (0.33)
S (0.33)
F (0.66)
Y (0.26)
S (0.08)
P
Q
R
M
V
R
990
(0.41)
(0.39)
(0.19)
(0.57)
(0.39)
(0.03)
V (1)
C
S
N
H
M
(0.25)
(0.23)
(0.21)
(0.15)
(0.13)
Q
E
I
L
(0.53)
(0.27)
(0.12)
(0.05)
R (0.92)
E (0.08)
G (0.63)
S (0.37)
Y
G
S
M
(0.63)
(0.20)
(0.10)
(0.07)
Ir þ Igf1r þ Irr
G (0.82)
M (0.15)
Q (0.55)
I (0.38)
E (0.04)
G (0.54)
S (0.45)
R
Q
M
P
(0.47)
(0.16)
(0.15)
(0.13)
V (1)
M
N
S
C
(0.40)
(0.31)
(0.18)
(0.08)
E (0.82)
Q (0.12)
I (0.04)
E (0.76)
R (0.21)
D (0.03)
S (0.87)
G (0.13)
T
Y
G
S
M
(0.33)
(0.31)
(0.19)
(0.09)
(0.07)
duplication. Using a statistical approach that compares
amino acid changes between IR and IRR þ IGF1R, we
found that most sites of the IR molecule were predicted
to be unrelated with cluster-specific (type II) functional divergence. Only 2 residues (positions 436 and 1095 of the
human IR sequence) received the highest posterior ratio
score (2.54) and were identified as radical cluster-specific
sites (posterior probability of 0.72). However, neither position 436 (R in IR and Q in IRR þ IGF1R) nor 1095
(K in IR and Q in IRR þ IGF1R) correspond to any of
the sites identified as shared derived of IR (table 1).
628a
P (0.96)
(0.37)
(0.36)
(0.17)
(0.10)
1362
T (0.99)
only isoform IRB and in brain only isoform IRA were detected (Fig. 4B). Interestingly, when the equivalent approach
was taken for chicken and Xenopus tissues, only a single isoform was detected (fig. 4C and D) after 35 cycles of PCR
amplification. Sequencing analysis of the amplified PCR
bands showed that they corresponded to the isoform IRA.
To discard the possibility that the isoform IRB could be expressed at so low levels that could not be detected in a single
round of amplification, we performed a nested PCR. As
shown in figure 4E and F, only a single amplification product
was again obtained strongly, suggesting that only 1 isoform is
expressed in the analyzed chicken and Xenopus tissues.
Evolution of the Splicing of Exon 11
The human and murine Ir are alternatively spliced in a
tissue-specific manner and produce 2 isoforms, IRA and
IRB. In order to trace the origin of this splicing mechanism,
we analyzed the presence of 2 isoform transcripts in nonmammalian species by reverse transcriptase–polymerase
chain reaction (RT-PCR) analysis using the upstream
primer directed against the 3# end of exon 10 and the downstream primer against the 5# end of the putative exon 12 (fig.
4A). In agreement with previous reports, we found differential
distribution of both Ir isoform transcripts in mouse tissues in
the 2 ages analyzed. Adipose tissue and muscle expressed
both isoform transcripts to different degree, whereas in liver
Discussion
In this study, we provide for the first time a robust phylogenetic framework to understand the molecular evolution
and functional diversification of IRs in vertebrates. According to the reconstructed phylogeny, the 3 described vertebrates IR, IRR, and IGF1R conform each a monophyletic
group and correspond to 3 distinct paralogous groups. A
first duplication of the receptor gene led to the Ir paralog
and the ancestor of Igf1r and Irr, which were both subsequently generated in a second round of duplication. The
presence of 3 or more paralogs in vertebrates but only 1
gene copy in nonvertebrates is a common pattern to other
1050 Hernández-Sánchez et al.
Fig. 4.—IR isoform expression. (A) Schematic representation of the mouse IR gene. White boxes indicate noncoding regions, whereas gray boxes
represent the coding exons, which are numbered. Solid lines represent constitutive splicing, and dashed lines represent alternative RNA processing.
Primers (P) used in PCR are indicated. IR RT-PCR with P1 and P2 primers of RNA from different tissues of postnatal day 10 (P10) and 35 (P35) mouse
(B) embryonic day 19 (E19) and postnatal day 10 (P10) chicken (C) and adult Xenopus (D). Nested PCR using P1 and P4 primers of P10 and P35
chicken (E) and P3 and P4 of adult Xenopus (F) tissues. A, adipose tissue; M, skeletal muscle; L, liver; and B, brain. (G) IR and IRR exon 11
alignments. Amino acids encoded together by exon 11 and its flanking exons are shown in red.
protein families (e.g., hedgehog; Zardoya et al. 1996) and
could be the result of independent duplication events in
each protein family. However, there is increasing evidence
that 2 rounds of whole-genome duplications occurred early
in vertebrate evolution and could be responsible for having
generated the observed higher paralog number of vertebrate
protein families (Meyer and Schartl 1999; Dehal and
Boore 2005), in general, and of the IR family, in particular.
Another genome duplication has been proposed in teleost
fishes (Meyer and Van de Peer 2005; Brunet et al. 2006),
and, for example, Tetraodon nigroviridis presents indeed 2
gene copies of Ir (CAG08022.1 and CAG07190.1) and of
Igf1r (CAG13078.1 and CAG03114.1). Strikingly, Irr was
missing in teleost fish. This may reflect that Irr orthologs in
teleosts might have highly divergent sequences and thus
might have not been detected yet through similarity
searches. Alternatively, it may be possible either that
Irr was lost at least in the common ancestor of teleosts
or that Irr was a novel acquisition of tetrapods. Sequencing
the genomes of basal actinopterygian (e.g., bichir and sturgeon) and sarcopterygian (e.g., lungfishes and coelacanth)
fishes would help in discerning among these competing
hypotheses.
As previously reported for mammals (Ullrich et al.
1986; Shier and Watt 1989; Rosenfeld and Roberts
1999), primary structure was relatively highly conserved
among paralogs of the IR family across vertebrates (22%
invariant sites). Conserved sites are distributed throughout
the molecule, and conserved stretches are particularly abundant around the tyrosine kinase domain. In fact, some sites
of this domain are also conserved in less related members of
the tyrosine kinase superfamily (Hubbard and Till 2000;
Ward et al. 2007) such as the epidermal growth factor receptor and the plate-derived growth factor receptor (data
not shown). The overall evolutionary conservation of the
primary structure of the vertebrate IR indicates that a relatively high proportion of the molecule is under strong selection pressure because it is likely needed to maintain the
general IR and signal transduction functions. In agreement
with this observation, only few sites (2%) were identified as
being shared derived by IR. These positions are maintained
by purifying selection and not need to be fully conserved in
all IR orthologs. They characterize IR and may be particularly important for its related but distinct function. Our results show that the relative distribution of IR shared derived
sites was even along the receptor (i.e., the rate of generation
of shared derived sites along the molecule was uniform).
However, because the extracellular portion of the receptor
more than doubles the intracellular portion, most IR shared
derived sites were located in the extracellular portion,
Evolution of the Insulin Receptor Family in Vertebrates 1051
which may suggest that subtle differences in function
among paralogs of the family are evolutionarily achieved
more through changes in ligand-binding affinity (Schaefer
et al. 1990; Brandt et al. 2001) than by modifications of
the intracellular signal transduction at the tyrosine kinase
domain.
Shared derived residues are best candidates for sitedirected mutagenesis in order to identify which evolutionary changes are responsible for functional divergence of
IR (Jimenez-Jimenez et al. 2006). One of the identified
shared derived IR characters (position 364) corresponds
to a potential site for N-glicosilation. This posttranslational
modification is critical for the correct assembling of tertiary
and quaternary IR structures (Olson et al. 1988). Two other
potential N-glicosilation sites (positions 105 and 651) are
conserved among all paralogs across vertebrates. However,
individual mutation of the 18 N-glicosilation sites of the
human IR showed high functional redundancy of those sites
(reviewed in Adams et al. [2000]). Other important residues
in maintaining the quaternary structure of the receptor are 6
cysteines involved in the formation of disulphure bonds.
These cysteines are highly conserved in all paralogs across
vertebrates and in Branchiostoma. However, 4 out of these
6 cysteines are not conserved in Ciona. A similar pattern is
found in Drosophila (Fernandez et al. 1995) where the receptor assembles into a quatenary structure but only 2 out of
the 6 cysteines described in human IR are conserved.
Amino acid residues that are highly conserved in IR
not need to be necessarily conserved in the other paralogs
and reflect site-specific shift of evolutionary rate (Gu 2006).
However, in 2 positions (436 and 1095), cluster-specific
residues (different between paralogs IR and IRR þ IGF1R,
but otherwise highly conserved within each homologous
group) were identified with statistical support. These residues evidenced radical shifts of amino acid property (type II
functional divergence) (Gu 2006). According to the phylogeny, IR retains the ancestral character state in these 2 positions, whereas IRR þ IGF1R feature a shared derived
character state. Again, these sites should be straightforward
targets for site-directed mutagenesis, in this case, to characterize IRR þ IGF1R functional divergence.
Both Ir and Irr have an extra exon (namely exon 11)
with respect to Igf1r. The Irr exon 11 can be traced back
to amphibians, whereas Ir exon 11 is found exclusively in
mammals (fig. 4). The highly divergent sequence of both
exons (fig. 4) and the reconstructed phylogeny of the vertebrate IR family strongly indicate that both exons were acquired independently by each paralog rather than being
present in the ancestor of the vertebrate protein family and
lost multiple times. Although most functional diversification
of the vertebrate IR family is achieved through gene duplication, alternative splicing is also found to be an important
mechanism that has generated functional diversification during the evolutionary history of the family. In this study, we
show that alternative splicing of Ir exon 11 seems to be restricted exclusively to mammals. The physiological outcome
of the evolutionary acquisition of Ir exon 11 is not fully understood to date. The novel IR isoform (IRB) shows a decreased affinity for IGF2 resulting in a more specific
receptor for insulin and restricting IGF2 signaling. In this regard, 2 additional evolutionary novelties appeared in mam-
mals to fine-tune IGF2 activity. First, a novel receptor for
IGF2, namely IGF2R, was evolved. This receptor is devoid
of signal transduction capabilities, and it acts as a clearance
receptor that modulates levels of circulating IGF2. Second,
imprinting of the Igf2 gene was evolved to prevent expression
of the maternal allele. Alterations of the strict control of IGF2
bioavailability as well as IRA expression have been associated to malignant process (reviewed in Denley et al. [2003]).
A second selective advantage derived from the evolutionary acquisition of Ir exon 11 could be specialization of
IRB as a more metabolic receptor. This isoform is predominantly expressed in insulin target tissues (Seino and Bell
1989; Mosthaf et al. 1990) that are responsible of glucose
homeostasis. Furthermore, patients with myotonic dystrophy type 1 present a 70% decrease in insulin sensitivity
in skeletal muscle that is associated with a switch in alternative splicing from IRB to IRA (Savkur et al. 2001).
The insulin signaling pathway is most complex in vertebrates with both insulin and IGF having acquired important and diversified metabolic functions beyond their
original growth-stimulating function in nonvertebrates.
Our study shows that an important element to understand
the functional diversification of these hormones is the corresponding functional diversification of the vertebrate IRs
with respect to their nonvertebrate counterparts. Specificity
of the ligands of the three vertebrate IR paralogs seems to
have been acquired mostly through gene duplication of the
gene products, as well as through a mechanism of alternative splicing in mammals. Functional divergence among
vertebrate IR paralogs is centered in few amino acid residues along the molecule and future site-directed mutagenesis essays on these residues will be key in disentangling the
complex evolution of new functions within the protein family and, in particular, in deciphering the unknown function
of the orphan IRR.
Acknowledgments
We thank Ms C. Murillo for her excellent technical
support. We thank Dr MJ Delgado (Universidad Complutense de Madrid) for providing the adult Xenopus and Dr R.
Martı́nez-Álvarez for performing preliminary experiments.
We thank Dr J. Rozas and 2 anonymous reviewers for insightful comments on an earlier version of the manuscript.
The studies were financed partially by the grants
BFU2004–2352 and BFU2007-61055 from the Spanish
Ministry of Education and Science (MEC) and the ‘‘Red
de Grupos’’ RGDM G03/212 from the ‘‘Instituto de Salud
Carlos III’’ from MSC (Spain) to F.d.P. and the grant
CGL2004-00401 from MEC to R.Z.; C.H.S. was a holder
of a ‘‘Ramón y Cajal’’ contract and A.M. had a predoctoral
fellowship, both from MEC (Spain).
Literature Cited
Abascal F, Zardoya R, Posada D. 2005. ProtTest: selection of
best-fit models of protein evolution. Bioinformatics.
21:2104–2105.
Adams TE, Epa VC, Garrett TP, Ward CW. 2000. Structure and
function of the type 1 insulin-like growth factor receptor. Cell
Mol Life Sci. 57:1050–1093.
1052 Hernández-Sánchez et al.
Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z,
Miller W, Lipman DJ. 1997. Gapped BLAST and PSIBLAST: a new generation of protein database search
programs. Nucleic Acids Res. 25:3389–3402.
Brandt J, Andersen AS, Kristensen C. 2001. Dimeric fragment of
the insulin receptor alpha-subunit binds insulin with full
holoreceptor affinity. J Biol Chem. 276:12378–12384.
Brogiolo W, Stocker H, Ikeya T, Rintelen F, Fernandez R,
Hafen E. 2001. An evolutionarily conserved function of the
Drosophila insulin receptor and insulin-like peptides in
growth control. Curr Biol. 11:213–221.
Brunet FG, Crollius HR, Paris M, Aury JM, Gibert P, Jaillon O,
Laudet V, Robinson-Rechavi M. 2006. Gene loss and
evolutionary rates following whole-genome duplication in
teleost fishes. Mol Biol Evol. 23:1808–1816.
Chan SJ, Steiner DF. 2000. Insulin through the ages: phylogeny
of a growth promoting and metabolic regulatory hormone.
Am Zool. 40:213–222.
De Meyts P. 2004. Insulin and its receptor: structure, function
and evolution. Bioessays. 26:1351–1362.
Dehal P, Boore JL. 2005. Two rounds of whole genome
duplication in the ancestral vertebrate. PLoS Biol. 3:e314.
Denley A, Wallace JC, Cosgrove LJ, Forbes BE. 2003. The
insulin receptor isoform exon 11 (IR-A) in cancer and other
diseases: a review. Horm Metab Res. 35:778–785.
Ebina Y, Ellis L, Jarnagin K, et al. (12 co-authors). 1985. The
human insulin receptor cDNA: the structural basis for hormoneactivated transmembrane signalling. Cell. 40:747–758.
Edgar RC. 2004. MUSCLE: multiple sequence alignment with
high accuracy and high throughput. Nucleic Acids Res.
32:1792–1797.
Efstratiadis A. 1998. Genetics of mouse growth. Int J Dev Biol.
42:955–976.
Fernandez AM, Kim JK, Yakar S, Dupont J, HernandezSanchez C, Castle AL, Filmore J, Shulman GI, Le Roith D.
2001. Functional inactivation of the IGF-I and insulin
receptors in skeletal muscle causes type 2 diabetes. Genes
Dev. 15:1926–1934.
Fernandez R, Tabarini D, Azpiazu N, Frasch M, Schlessinger J.
1995. The Drosophila insulin receptor homolog: a gene
essential for embryonic development encodes two receptor
isoforms with different signaling potential. EMBO J.
14:3373–3384.
Garza-Garcia A, Patel DS, Gems D, Driscoll PC. 2007. RILM:
a web-based resource to aid comparative and functional
analysis of the insulin and IGF-1 receptor family. Hum Mutat.
28:660–668.
Gu X. 2006. A simple statistical method for estimating type-II
(cluster-specific) functional divergence of protein sequences.
Mol Biol Evol. 23:1937–1945.
Guindon S, Gascuel O. 2003. A simple, fast, and accurate
algorithm to estimate large phylogenies by maximum likelihood. Syst Biol. 52:696–704.
Hamburger V, Hamilton HL. 1951. A series of normal stages in
the development of the chick embryo. J Morphol. 88:49–92.
Hernandez-Sanchez C, Mansilla A, de la Rosa EJ, de Pablo F.
2006. Proinsulin in development: new roles for an ancient
prohormone. Diabetologia. 49:1142–1150.
Holzenberger M, Dupont J, Ducos B, Leneuve P, Geloen A,
Even PC, Cervera P, Le Bouc Y. 2003. IGF-1 receptor
regulates lifespan and resistance to oxidative stress in mice.
Nature. 421:182–187.
Hubbard SR, Till JH. 2000. Protein tyrosine kinase structure and
function. Annu Rev Biochem. 69:373–398.
Jimenez-Jimenez J, Zardoya R, Ledesma A, Garcia de Lacoba M,
Zaragoza P, Mar Gonzalez-Barroso M, Rial E. 2006.
Evolutionarily distinct residues in the uncoupling protein
UCP1 are essential for its characteristic basal proton
conductance. J Mol Biol. 359:1010–1022.
Kellerer M, Lammers R, Ermel B, Tippmer S, Vogt B,
Obermaier-Kusser B, Ullrich A, Haring HU. 1992. Distinct
alpha-subunit structures of human insulin receptor A and B
variants determine differences in tyrosine kinase activities.
Biochemistry. 31:4588–4596.
Kimura KD, Tissenbaum HA, Liu Y, Ruvkun G. 1997. daf-2, an
insulin receptor-like gene that regulates longevity and
diapause in Caenorhabditis elegans. Science. 277:942–946.
Leibiger B, Leibiger IB, Moede T, Kemper S, Kulkarni RN,
Kahn CR, de Vargas LM, Berggren PO. 2001. Selective
insulin signaling through A and B insulin receptors regulates
transcription of insulin and glucokinase genes in pancreatic
beta cells. Mol Cell. 7:559–570.
Maddison WP, Maddison DR. 1992. MacClade: analysis of
phylogeny and character evolution. Sunderland (MA):
Sinauer Associates Inc.
McClain DA. 1991. Different ligand affinities of the two human insulin receptor splice variants are reflected in parallel changes in
sensitivity for insulin action. Mol Endocrinol. 5:734–739.
Meyer A, Schartl M. 1999. Gene and genome duplications in
vertebrates: the one-to-four (-to-eight in fish) rule and the
evolution of novel gene functions. Curr Opin Cell Biol.
11:699–704.
Meyer A, Van de Peer Y. 2005. From 2R to 3R: evidence for
a fish-specific genome duplication (FSGD). Bioessays.
27:937–945.
Morgan DO, Edman JC, Standring DN, Fried VA, Smith MC,
Roth RA, Rutter WJ. 1987. Insulin-like growth factor II receptor
as a multifunctional binding protein. Nature. 329:301–307.
Mosthaf L, Grako K, Dull TJ, Coussens L, Ullrich A,
McClain DA. 1990. Functionally distinct insulin receptors
generated by tissue-specific alternative splicing. EMBO J.
9:2409–2413.
Moxham CP, Duronio V, Jacobs S. 1989. Insulin-like growth
factor I receptor beta-subunit heterogeneity. Evidence for
hybrid tetramers composed of insulin-like growth factor I and
insulin receptor heterodimers. J Biol Chem. 264:13238–13244.
Nakae J, Kido Y, Accili D. 2001. Distinct and overlapping
functions of insulin and IGF-I receptors. Endocr Rev.
22:818–835.
Nef S, Verma-Kurvari S, Merenmies J, Vassalli JD,
Efstratiadis A, Accili D, Parada LF. 2003. Testis determination requires insulin receptor family function in mice. Nature.
426:291–295.
Olson TS, Bamberger MJ, Lane MD. 1988. Post-translational
changes in tertiary and quaternary structure of the insulin
proreceptor. Correlation with acquisition of function. J Biol
Chem. 263:7342–7351.
Pagel M, Meade A, Barker D. 2004. Bayesian estimation of
ancestral character states on phylogenies. Syst Biol. 53:673–684.
Pashmforoush M, Chan SJ, Steiner DF. 1996. Structure and
expression of the insulin-like peptide receptor from amphioxus. Mol Endocrinol. 10:857–866.
Rosenfeld RG, Roberts CT. 1999. The IGF system: molecular
biology, physiology and clinical applications. Totowa (NJ):
Humana Press.
Ruvkun G, Hobert O. 1998. The taxonomy of developmental
control in Caenorhabditis elegans. Science. 282:2033–2041.
Saltiel AR, Kahn CR. 2001. Insulin signalling and the regulation
of glucose and lipid metabolism. Nature. 414:799–806.
Savkur RS, Philips AV, Cooper TA. 2001. Aberrant regulation of
insulin receptor alternative splicing is associated with insulin
resistance in myotonic dystrophy. Nat Genet. 29:40–47.
Schaefer EM, Siddle K, Ellis L. 1990. Deletion analysis of the
human insulin receptor ectodomain reveals independently
Evolution of the Insulin Receptor Family in Vertebrates 1053
folded soluble subdomains and insulin binding by a monomeric alpha-subunit. J Biol Chem. 265:13248–13253.
Schlessinger J. 2000. Cell signaling by receptor tyrosine kinases.
Cell. 103:211–225.
Seino S, Bell GI. 1989. Alternative splicing of human insulin
receptor messenger RNA. Biochem Biophys Res Commun.
159:312–316.
Seino S, Seino M, Nishi S, Bell GI. 1989. Structure of the human
insulin receptor gene and characterization of its promoter.
Proc Natl Acad Sci USA. 86:114–118.
Shier P, Watt VM. 1989. Primary structure of a putative receptor
for a ligand of the insulin family. J Biol Chem.
264:14605–14608.
Soos MA, Siddle K. 1989. Immunological relationships between
receptors for insulin and insulin-like growth factor I. Evidence
for structural heterogeneity of insulin-like growth factor I
receptors involving hybrids with insulin receptors. Biochem J.
263:553–563.
Swofford DL. 2002. PAUP*: phylogenetic analysis using
parsimony (*and other methods). Version 4.0b 10. Sunderland (MA): Sinauer Associates, Inc.
Tissenbaum HA, Ruvkun G. 1998. An insulin-like signaling
pathway affects both longevity and reproduction in Caenorhabditis elegans. Genetics. 148:703–717.
Ullrich A, Bell JR, Chen EY, et al. (15 co-authors). 1985. Human
insulin receptor and its relationship to the tyrosine kinase
family of oncogenes. Nature. 313:756–761.
Ullrich A, Gray A, Tam AW, et al. (14 co-authors). 1986. Insulinlike growth factor I receptor primary structure: comparison
with insulin receptor suggests structural determinants that
define functional specificity. EMBO J. 5:2503–2512.
Vogt B, Carrascosa JM, Ermel B, Ullrich A, Haring HU. 1991.
The two isotypes of the human insulin receptor (HIR-A and
HIR-B) follow different internalization kinetics. Biochem
Biophys Res Commun. 177:1013–1018.
Ward CW, Lawrence MC, Streltsov VA, Adams TE,
McKern NM. 2007. The insulin and EGF receptor structures:
new insights into ligand-induced receptor activation. Trends
Biochem Sci. 32:129–137.
Yamaguchi Y, Flier JS, Yokota A, Benecke H, Backer JM,
Moller DE. 1991. Functional properties of two naturally
occurring isoforms of the human insulin receptor in Chinese
hamster ovary cells. Endocrinology. 129:2058–2066.
Yang Z, Nielsen R. 2002. Codon-substitution models for
detecting molecular adaptation at individual sites along
specific lineages. Mol Biol Evol. 19:908–917.
Zardoya R, Abouheif E, Meyer A. 1996. Evolutionary analyses
of hedgehog and Hoxd-10 genes in fish species closely related
to the zebrafish. Proc Natl Acad Sci USA. 93:13036–13041.
Norihiro Okada, Associate Editor
Accepted January 29, 2008