Please cite this article in press as: Lovejoy, D.A., de Lannoy, L., Evolutioof peptides: Expansion and specialization in the vertebrates. J.neu.2013.09.006
0891-0618/$ – see front matter � 2013 Published by Elsevier B.V.
http://dx.doi.org/10.1016/j.jchemneu.2013.09.006
1. Introduction
The Corticotropin Releasing Factor (CRF) family of peptides arewidely expressed throughout the central nervous system and avariety of peripheral tissues in all vertebrate species studied todate. These peptides play a major physiological role in theregulation of stress and energy-associated homeostasis. Thephysiology and behaviour of CRF has been reviewed in Qdetail(Dunn and Berridge, 1990; De Souza, 1995; Vale et al., 1981;
n and phylogeny of the corticotropin-releasing factor (CRF) family Chem. Neuroanat. (2013), http://dx.doi.org/10.1016/j.jchem-
D.A. Lovejoy, L. de Lannoy / Journal of Chemical Neuroanatomy xxx (2013) xxx–xxx2
G Model
CHENEU 1253 1–7
vejoy and Balment, 1999; Boorse and Denver, 2006; Lovejoy,13). Thus the focus of the current review is on the understanding
the evolution of CRF in light of recent findings.The first attempts to purify CRF were published in 1955,
dependently by Schally and Saffran (1955) and Guillemin andsenberg (1955). The next major finding occurred in 1981 whenle and his colleagues published the sequence of ovine CRF. Valed his colleagues established that the hypothalamus–pituitary–renal/interrenal (HPA/I) axes are a pivotal component of thertebrate stress-response system. In all vertebrates, the HPA/Ies are regulated apically by CRF (Fig. 1.). Shortly after thescovery of CRF, a family of related peptides were discovered:otensin-1 (fish)/urocortin (mammals)/sauvagine (amphibians),n-2 and Ucn-3. These studies opened a new understanding ofe regulation of the physiology of stress. The perception of aressor, something that can challenge an organism’s ability torvive and reproduce, is one of the most highly conservedechanisms associated with intracellular and inter-organismalmmunication.Many neurologically active peptide families possess early
olutionary origins. Short amino acid sequences, peptides, areought to be among the first biogenic macromolecules synthe-ed in the prebiotic earth (Rode, 1999; Rode and Plankensteiner,06). These early peptides may have acted in several different
ructural and physiological capacities. As genes, they have thetential to be reproduced and can carry information at theimary structural level. In addition, differing affinities amongptides allow for the physical interaction between differentecies of peptides and proteins to foster the first receptor–ligand
Please cite this article in press as: Lovejoy, D.A., de Lannoy, L., Evolutiof peptides: Expansion and specialization in the vertebrates.
neu.2013.09.006
interactions. The amphiphilic nature of peptides allows interactionwith both the hydrophobic interior of the plasma membrane andthe hydrophilic nature of the extracellular environment. Moreover,the ionic character of amino acids could act to stabilize the osmoticand pH balance of the early protocells. Given these characteristics,there may have been considerable evolutionary pressure to selectsuch early cell-like systems for ones that expressed these earlypeptides (Lovejoy, 2005). Such a theory may explain, in part, thediversity of peptides found in the Metazoa. Peptide hormonesarguably constitute the largest family of high-affinity and high-specificity hormone systems. The extant expression of this peptidediversity among metazoans appears to be due, in part, to a series ofgenome expansion events that occurred early in metazoanphylogeny. Evidence for this early evolution of peptide hormonescomes from the numerous examples of orthologous peptidesystems identified by the genomic sequencing programmes inthe various model species.
2. CRF-family peptide evolution in metazoans
The presence of CRF-like peptides, receptor and binding-proteinhomologues in both the insects and chordates strongly supportsthe concept that the CRF system evolved in a metazoan speciesancestral to both the deuterostomes (chordates, echinoderms) andprotostomes (molluscs, annelids, arthropods). In jawed verte-brates, the CRF family of peptides include four paralogous lineages;CRF, urotensin-1/urocortin/sauvagine, Ucn2 and Ucn3. In insects,CRF orthologues include the diuretic hormones (DH). A transitionalform, a CRF–DH-like peptide is found in the urochordate, Ciona
intestinalis. Thus, the conservation of the CRF superfamily peptidesystem is indicative of an essential physiological system requiredfor enhanced organismal survival.
2.1. CRF family in chordates
Although the first evidence of a hypothalamic corticotropin-releasing substance was reported in 1955, the first family memberof the CRF family was found not in the hypothalamus, but rather inthe skin of a frog. This peptide, called ‘sauvagine’, consisted of 40amino acids and was effective at releasing ACTH from rat pituitarycells (Montecucchi et al., 1979). About the same time, a 41-aminoacid peptide, called urotensin-I (UI), from the caudal neurosecre-tory system (urophysis) of carp was identified (Lovejoy andBalment, 1999; Lovejoy and Jahan, 2006; Lovejoy et al., 2009). Theactual structure of CRF was reported by Vale and his associates in1981. Over the next 10 years, the discovery of CRF in fishes andamphibians indicated that UI and sauvagine were homologues ofCRF (Lovejoy and Balment, 1999). The rat and human orthologuesof UI were eventually cloned and were termed urocortin (Ucn)(Vaughan et al., 1995; Donaldson et al., 1996). Ucn, like sauvagine,was 40 amino acids long, in contrast to the 41-residue sequencethat characterized CRF and UI. Further studies (Lovejoy andBalment, 1999; Bittencourt and Sawchenko, 2000) suggested thatfour distinct CRF/UI-like peptides were found in chordates. Then in2001, the presence of two additional CRF-related peptides werereported by the Vale and Sawchenko laboratories who called thepeptides urocortin 2 (Ucn2) and urocortin 3 (Ucn3) (Reyes et al.,2001; Lewis et al., 2001), and independently by Hsu and Hseuh(2001) who named the peptides stresscopin and stresscopin-related peptide on the basis of their anxiolytic effects. However,parallel to these studies were a host of additional investigationscharacterizing the role of CRF-family peptides in non-mammalianvertebrates and invertebrates. Together these studies led to amodel of CRF family phylogeny and evolution and established thata common ancestor developed early in the evolution of multicel-lular organisms (Metazoa) (Fig. 2).
on and phylogeny of the corticotropin-releasing factor (CRF) familyJ. Chem. Neuroanat. (2013), http://dx.doi.org/10.1016/j.jchem-
Fig. 2. Phylogeny of the CRF family in multicellular organisms (Metazoa). CRF family peptides have been identified in both deuterostomes and protostomes indicating that a
CRF-like peptide was present before the bifurcation of these two lineages. Currently, there is no evidence of CRF-like peptides in metazoan lineages that evolved before this
time. Note 1: Ucn2 (urocortin 2) is present as a pseudogene in some vertebrate lineages; Note 2: In some insects, the CRF homolog is DH (diuretic hormone) and there may be
some lineage specific gene duplications that led to paralogous forms of DH in insects. Abbreviations: CRF, corticotropin-releasing factor; gd, gene duplication; UI, urotensin-I
(also includes urocortin 1 and sauvagine); Ucn2, urocortin 2; Ucn3, urocortin 3.
D.A. Lovejoy, L. de Lannoy / Journal of Chemical Neuroanatomy xxx (2013) xxx–xxx 3
G Model
CHENEU 1253 1–7
The basic structure of the CRF and Ucn genes are similar but arelocated on chromosomes 8q13 and 2p23-p21, respectively, inhumans and chromosomes 3 and 13 in mice. There are two exonsof unequal sizes where the first encodes for a portion of the 50
untranslated region (UTR) whereas the second exon encodes theentire preprohormone sequence and the entire 30 UTR. The intronlength varies considerably among species, although the Ucnintron is much smaller compared to that of CRF (Lovejoy, 2006).Furthermore, while the CRF cDNA structure remains fairlyconsistent among orthologues, there is considerable variabilityamong the Ucns and UI. The UI mRNA is 3–5 times longer in fishesthan the mammal Ucn mRNA. This length difference is due to amuch shorter gene-associated peptide in the Ucns and extended 30
UTR in fish UI, particularly in rainbow trout. In this species, thereare multiple polyadenylation sites that correspond to thedifferential mRNA processing that occurs in the urophysis andbrain (Barsyte et al., 1999). The gene structures of Ucns 2 and 3 alsoreflect a structural relationship to each other, although these geneshave undergone considerable change. Both Ucn2 and 3 genespossess two exons similar to the arrangement for urocortin andCRF. Ucn2 is found on human chromosome 3p21.3 and on mousechromosome 9. In both cases, the start of the Ucn2 gene is only afew hundred bases from the terminal exon of type VII a1 collagen.Ucn3, on the other hand possesses an extended intron of over 8000bases. The position of the 30 UTR of the mRNA is unclear. The geneis found on human chromosome 10p15.1 and on mouse chromo-some 5 (Lovejoy, 2006).
Please cite this article in press as: Lovejoy, D.A., de Lannoy, L., Evolutioof peptides: Expansion and specialization in the vertebrates. J.neu.2013.09.006
The sequencing of the tunicates, C. intestinalis and C. savignii,genomes has provided considerable insight into our understandingof how the CRF family of peptides evolved in the chordates. Thetunicates, along with the class Urochordata, are the direct taxon tothe Chordata (see Lovejoy and Barsyte-Lovejoy, 2010). The recentcharacterization of the single CRF-like peptide found in thegenomes of each of these species indicates a sequence possessingboth Ucn2/3 and insect diuretic peptide structural attributes(Lovejoy and Barsyte-Lovejoy, 2010). However, the CRF-likepeptide gene of the tunicate is unique among chordates in thatonly a single exon is present. It is not clear what the ancestralcondition was, however because as many as five exons are found inthe insect homologues of CRF, therefore it is likely that at least twoor possibly three exons were inherited by the protochordate andchordate ancestor (Lovejoy, 2006). Thus, the loss of the first exonlikely occurred during the extensive genomic reorganization of theurochordate, or tunicate lineage.
The gene structure of the CRFs and the Ucns in vertebratesindicate that the expansion of the ancestral gene into four distinctparalogues is due to two genome duplications early in chordateancestry. The most recent genome expansion events occurred afterthe appearance of the protochordates, but before the emergence ofthe jawed fish (Lundin et al., 2003; Dehal and Boore, 2005). If therewere two rounds of major genome expansion events before theemergence of jawed fishes and particularly before the sarcopter-ygian (lobe-finned fishes, e.g. lungfish, tetrapods) and actinopter-ygian (ray-finned fishes; e.g. salmon, cod) split, this would explain
n and phylogeny of the corticotropin-releasing factor (CRF) family Chem. Neuroanat. (2013), http://dx.doi.org/10.1016/j.jchem-
D.A. Lovejoy, L. de Lannoy / Journal of Chemical Neuroanatomy xxx (2013) xxx–xxx4
G Model
CHENEU 1253 1–7
e presence of the four paralogous peptides in vertebrates. Thisggests there would be only a single CRF-like peptide that washerited by the early vertebrates. Consistent with this postulationour finding of a single form of a CRF-like peptide in C. intestinalis
d C. savignii.
2. CRF family in non-chordates
The most structurally similar CRF-like peptides found invertebrates are the diuretic hormones (DH) isolated from insectsig. 2). A number of these diuretic peptides possess sequenceentities that indicate a structural relationship with both the CRF/ocortin and Ucns 2/3 lineages. The high metabolic rate and shortneration time of insects promote and increase the probability ofutational events in genes, thus the amino acid sequence isbject to greater rate of change over the same time frame, relative
chordates. Several variants have been identified, although it isfficult to establish orthology or paralogy to the variants. Thecently discovered diuretic hormone-31 (DH31) shows somequence identity to Ucns 2 and 3, although others have suggestedat DH31 may be more related to the calcitonin family of peptidesuruya et al., 2000; Coast et al., 2001). The CRF receptors andlcitonin receptors show more sequence similarity to each otheran either do to other classes of G-protein coupled receptorsredricksson et al., 2003). This observation, along with theentified structural similarity of DH31 to both calcitonin and Ucnsand 3, indicate the possibility that the CRF family of peptide ande calcitonin family of peptides have a common origin that cannot
discounted.
3. Comparison of chordate and non-chordate CRF family peptides
A comparison among CRF orthologue and paralogue structuresroughout the metazoans has provided insight into how theseptide hormones evolved. Indeed, a source of confusion in manyblished studies is the tendency to group Ucn/UI, Ucn 2 and Ucn 3
ithin a peptide clade distinct from CRF. The apparent sequencessimilarity of Ucns 2 and 3 to CRF and Ucn/UI suggested,wever, that the former were not urocortins, per se, but rather theoduct of a gene duplication early in CRF ancestry where Ucn2d Ucn3 represented a separate and distinct lineage of peptides ine chordates. Three sets of independent observations support thispothesis: First, the Ucn2/3 sequences possessed a greaterquence similarity, than the CRF/UI/Ucn lineage, to the insectthologues of the CRF family (Lovejoy and Jahan, 2006). Secondly,ey also retained key amino acid motifs found in a newlyscovered calcitonin-like peptide in insects, confirming a closeolutionary relationship between the CRF and calcitonin peptidemilies. Finally, they possess amino acid sequence motifs found ine Ciona CRF–DH-like peptides that are not present in the CRF–UI/n lineage (Lovejoy and Barsyte-Lovejoy, 2010).The sequence comparisons of the vertebrate CRF –UI/Ucn
ptides with the Ciona and insect DH peptides do suggest thate CRF and UI/Ucn/sauvagine series is the more derived of theF family members in vertebrates. Interestingly, the dogfishark, (Scyliorhinus canicula ) UI and monkey frog (Phylomedusa
uvagii) sauvagine peptides appear more basal in the clustalalysis than what is expected given the phylogenetic position ofe species where these peptides are found. The elasmobranches,though an early bifurcating lineage of the Vertebrata, arelatively slowly evolving and, therefore, many gene sequencesay be expected to reflect this low rate of change. Although
sauvagii is considered to be a phylogenetically recent offshoot the anuran lineage, it is considered to be very highlyecialized and thus this peptide reflects some of the recentanges in the UI/Ucn lineage as well as a number of the ancestral
Please cite this article in press as: Lovejoy, D.A., de Lannoy, L., Evolutiof peptides: Expansion and specialization in the vertebrates.
neu.2013.09.006
characteristics (Lovejoy and Balment, 1999; Lovejoy and Jahan,2006; Lovejoy, 2009; Nock et al., 2011 ).
Taken together, these observations imply that the CRF/UI(including Ucn and sauvagine) clade of peptides represents a morederived version of the CRF peptide family in chordates and thephysiological functions of the Ucn2/3 peptides may resemblethe ancestral functions of this peptide family to a greater degree.The highly conserved nature of CRF in chordates may, therefore,be the result of the utilization of this paralogue for the primarycontrol of the hypothalamic-pituitary adrenal/interrenal axis andthe subsequent selection constraints imposed on this sequence(Lovejoy, 2009).
3. Evolutionary trends of CRF peptides in vertebrates
Thus, once CRF became associated with the regulation of theHPA/I axis and the anticipation of the stress-response, it becameunder a much greater physiological constraint leading to lessvariation in its structure. Thus, the sequence of CRF is compara-tively stable in comparison to the UI/Ucn or the Ucn2/3 lineages(Lovejoy and Jahan, 2006). A notable exception does occur inruminant ungulates where the CRF structure changed at a muchfaster rate than predicted likely due to the distinct evolution ofthis taxon (Shen et al., 1990; Liu et al., 1990; Lovejoy, 1996; Lovejoyand Balment, 1999). UI, in contrast, became specialized forits role in the urophysis in an analogous manner to how CRFbecame associated with the HPA/I axis. The urophysis, or caudalneurosecretory system, is analogous to the neurophysis in that it isa neurohaemal organ where the neurosecretory cells release theirconstituents into the capillaries of the vascular system. The caudalneurosecretory system in found in all actinopterygian fishes, buthas been reduced in the sarcopterygian line and lost entirely in thetetrapods. Thus, this organ system likely evolved in response toadapt to the osmoregulatory stress of rapidly changing ambientconditions such as salinity that occur in estuarine conditions, forexample (see Lovejoy, 2009 for discussion). In a number of ways,the functional role of UI in vertebrates is reflective of the role of thediuretic hormones in insects. The preferential binding of the UI-likepeptides by the CRF binding protein (CRF-BP) (Sutton et al., 1995),in addition to its recent discovery in insects (Huising and Flik,2000), argues strongly that the original CRF-BP evolved to bind apeptide with characteristics more similar to UI than CRF.
However, additional taxon-specific selection pressure hasinduced a number of alternative expression patterns of the CRFfamily genes. Unusually, there are two CRF paralogues distinctfrom UI in this species, an observation that is currently uniqueamong chordates (Nock et al., 2011). Although there are two CRFparalogues in a number of teleost fishes (Lovejoy and Balment,1999), these peptides are almost identical and reflect a relativelyrecent genome duplication, and likely possess similar functions. Incontrast, the sequence differences between the two CRF para-logues in the elephant shark are highly divergent indicating a geneduplication early in this lineage’s 400-million year evolution. Theretention of the same precursor structure in the second paraloguesuggests that this peptide is functional as a CRF but possesses adifferent function and expression pattern in this species. Further, inthe elephant shark, the Ucn2 gene is a pseudogene, although genesfor CRF, UI and Ucn3 are present. Interestingly, like the elephantshark, the chimpanzee also possesses a Ucn2 gene that is apseudogene as the result of a premature stop codon. Moreover, inhumans, this gene lacks the amidation signal that is required forfull activity, although despite this numerous studies indicate thatthis peptide is functional in humans. Thus, in light of the closesequence similarity between Ucn2 and 3, it appears that the Ucn2gene and its peptide function is being lost in some taxa and its
on and phylogeny of the corticotropin-releasing factor (CRF) familyJ. Chem. Neuroanat. (2013), http://dx.doi.org/10.1016/j.jchem-
D.A. Lovejoy, L. de Lannoy / Journal of Chemical Neuroanatomy xxx (2013) xxx–xxx 5
G Model
CHENEU 1253 1–7
function is being replaced by Ucn3 (Ikemoto and Park, 2006;Kuperman and Chen, 2006; Chand and Lovejoy, 2011).
4. Functional evolution of CRF family peptides
CRF is classically involved in the HPA axis in vertebrates. Therole of CRF in homeostasis is mirrored in CRF-like peptides in bothchordates and insects. Studies implicating the CRF and diuretichormone family peptides in feeding and diuresis in chordates(Bergh and Sodersten, 1996; Spina et al., 1996) and insects(Audsley et al., 1997; Coast, 1998), respectively, suggests that theancestral peptide also regulated homeostatic mechanisms associ-ated with ion and water intake and elements of energy-relatedmetabolism. The relationship of CRF to the HPA/I system inchordates may be a mechanism unique to the chordates.
A complete functional HPA/I neuroendocrine system, as foundin vertebrates, does not seem to be in place in any invertebratespecies (Campbell et al., 2004). Steroid feedback is a necessaryfeature of the HPA/I axis. Although a number of invertebratelineages possess some of required enzymes for the initial stages ofsteroid synthesis, a complete pathway is missing (Campbell et al.,2004). Similarly, although some studies indicate a functionalcorticoid-like receptor in C. intestinalis (Maury et al., 2006) andDrosophila (Carton et al., 2002), the requisite enzymes for thesynthesis of vertebrate-like corticoids are missing in these species.In Ciona, although a CRF-like ligand and receptor (Campbell et al.,2004) are present in the genome, other elements of the HPA/I axisare not. Moreover, early studies have shown immunoreactivity ofPOMC-like peptides in tunicates (Fritsch et al., 1982), howeveranalysis of the C. intestinalis or savignii genomes has not revealedthe existence of POMC-like genes (Campbell et al., 2004; Sherwoodet al., 2006). Taken together, these observations indicate that thecomponents of the HPA/I axis amalgamated into a functionalneuroendocrine loop during chordate evolution with CRF acting asthe releasing factor.
The ancient evolution of CRF and its paralogues indicates thatthis peptide family has had a significant period of time for selectivepressure to allow integration into a number of diverse physiologi-cal systems. In vertebrates and humans, CRF and Ucn elicitbehaviours resembling the response to the stress and environ-mental novelty such as increased arousal decreased food intake,diminished sexual and reproductive activity, and increasedgrooming. Thus, generally the CRF family of peptides act tointegrate the autonomic and behavioural responses to stressthrough interactions at the CNS level. These mechanisms areachieved by a CRF1 receptor-mediated activation of the hypotha-lamic–pituitary–adrenal and sympathetic branches of the stress-response. In contrast, the CRF2 receptor has been associated withanxiolytic effects, thus Ucns 2 and 3 have been deemed anxiolyticpeptides as they bind exclusively to this receptor. CRF and Ucn aregenerally associated with anxiogenic actions (Kuperman and Chen,2008; Chand and Lovejoy, 2011). A number of other neurologicalactions of CRF family peptides have been described. CRF familypeptides have well described effects on feeding and weightregulation in both mammals and fishes (Bale and Vale, 2004;Solinas et al., 2006; Backstrom and Winberg, 2013).
The HPA/I axis is ultimately responsible for the regulation ofenergy production to organs and tissues that require activationduring homeostatic challenge. Thus, CRF itself is intrinsicallyinvolved in energy metabolism. In vertebrates, CRF induces therelease of adrenal/interrenal glucocorticoids. The metabolicactions of the glucocorticoids are to provide the organism withenough energy to survive stressful situations that occur duringperturbations of homeostasis. Glucocorticoids possess a number ofactions associated with the utilization and flow of energy. Thisincludes the de novo synthesis of glucose, gluconeogenesis, which
Please cite this article in press as: Lovejoy, D.A., de Lannoy, L., Evolutioof peptides: Expansion and specialization in the vertebrates. J.neu.2013.09.006
is increased in the liver following mobilization of proteins from theskeletal muscle and subsequent deamination of amino acids thatare released during protein breakdown. However, glycogen canalso be deposited in the liver because of a glucocorticoid-stimulated increase in the glycogen synthase reaction. In tandemwith this, the breakdown of glycogen is inhibited by glucocorti-coids (Dunn and Berridge, 1990; Lovejoy, 2005; Lovejoy andBarsyte, 2011). In rats, Ucn is the most potent peptide at decreasingfood intake (Spina et al., 1996; Dautzenberg and Hauger, 2002). Infishes, food intake is also reduced significantly after CRFadministration (De Pedro et al., 1998). A recent report indicatinga role for the CRF-like diuretic hormones in insect feeding suggeststhat this mechanism is phylogenetically ancient (Audsley et al.,1997). Moreover, the over-expression of the CRF-BP in the mousemodel also affects body weight (Lovejoy et al., 1998) presumablyby sequestering free CRF, thereby reducing its receptor activationability in circuits associated with food intake.
The functional role of Ucn in mammals only provides a partialunderstanding of its orthologue, UI, in fishes. Urophyseal UI hasbeen implicated in ion and fluid equilibrium, cardiovascularactivity and participates in inter-renal tissue glucocorticoidrelease, all of which are necessary for the adaptability of fishes.In vertebrates, UI likely became specialized for its role in theurophysis in an analogous manner to how CRF became associatedwith the HPA/I axis. The urophysis in found in all fishes, but hasbeen reduced in the sarcopterygian line and lost entirely in thetetrapods. Thus, this organ system appears to have evolved inresponse to the osmoregulatory stress of rapidly changing ambientconditions that occur in aquatic environments. In addition,because water is a much denser medium than air, integration ofmechanoreception with the stress response is advantageous forsurvival if these compression waves can alert the organism ofpotential prey or predators. Thus, we might speculate that,whereas the pituitary and hypothalamus has been utilized bychordates to link the higher integration of some sensory systemswith the appropriate neural–humoural responses, the urophysishas evolved to link the peripheral osmoreceptor and mechanore-ceptor inputs of fishes to the appropriate stress response. This roleof UI with mechanoreceptors may also be reflected by therelationship urocortin has with the regulation of auditory sensoryprocessing in mammals (Lovejoy, 2009; Lovejoy and Balment,1999).
New studies indicate that the CRF family, and particularly thatof Ucn2 and 3, play a critical role in energy, and especially, glucosehomeostasis. Ucn2 can increase tissue sensitivity to insulin,whereas Ucn3 may act to increase the secretion of pancreaticinsulin. However, these mechanisms appear to involve theintegrated actions of both CNS and peripheral targets (Kupermanand Chen, 2008). Given that Ucn2 and 3 possess a number ofancient structural features of the CRF family, this role in energymetabolism may reflect some of their ancestral functions.
5. Speculation on the evolution of the CRF family of peptides
Currently, there is no evidence of CRF peptides older than thosethat evolved before the deuterostome–protostome bifurcation.However, there is some evidence that CRF-like peptides werepresent before this time (Fig. 3). The logic for this speculation isoutlined in four main premises. First, the insect DH-like peptide(DH31) has a compelling sequence similarity to the Ucns 2 and 3(Lovejoy and Jahan, 2006). Second, the tunicate CRF-DH-likepeptide sequence has primary sequence motifs similar to both theinsect DHs and the CRF family (Lovejoy and Barsyte-Lovejoy, 2010)suggesting that the original CRF-like peptide ancestral todeuterostomes and protostomes possessed both DH family andCRF family characteristics. Third, the insect DH31 peptide
n and phylogeny of the corticotropin-releasing factor (CRF) family Chem. Neuroanat. (2013), http://dx.doi.org/10.1016/j.jchem-
D.A. Lovejoy, L. de Lannoy / Journal of Chemical Neuroanatomy xxx (2013) xxx–xxx6
G Model
CHENEU 1253 1–7
quence has a strong sequence similarity to the calcitonin family peptides (Coast et al., 2001; Brugge et al., 2008). And given thate CRF/DH, calcitonin and pituitary adenylate cyclase-activatingptide (PACAP)/vasoactive intestinal peptide (VIP) familieshich also include secretin, glucagon and growth hormone-
leasing hormone peptides) receptor are closely related in the Bmily of G-protein-coupled receptors (Fredricksson et al., 2003), itlikely that similarly, the ligands also possess a structural
lationship. Fourth, the recently described teneurin C-terminalsociated peptides (TCAP) have structural similarity to both CRFd calcitonin families as well as PACAP-VIP families (Lovejoy and
han, 2006). Recent studies indicate that TCAP was the result of arizontal gene transfer event from prokaryotes (Zhang et al.,12; Chand et al., 2013). Taken together, we offer the followingeculation (Fig. 3). A horizontal gene transfer from prokaryotes to
protistan–metazoan ancestor led to the formation of a TCAP-likecestral peptide which subsequently evolved into a solublenalling peptide. Further gene duplications of this peptide gene
cus led to an ancestor of Family B GPCR ligands. Additional gene genomic duplication events eventually led to the formation ofCAP-VIP, calcitonin and CRF peptide families.Could we expect to discover additional CRF-like peptides?
rtebrate genomes are comparatively well understood, and thusr the only peptide family that show any resemblance to the CRFmily are the TCAP peptides, although they possess only about% sequence identity at the amino acid level. However, TCAP doest bind to either of the CRF receptors or the binding protein
ovejoy, unpublished observations). The closest major taxon toe chordates is the echinoderms (sea urchins, sea stars), but thereno information currently available on the identity of any CRF-likeptides from this group. Similarly, no CRF-like peptides have beenund in molluscs, although CRF-like immunoreactivity has beentected (Gonzalez et al., 1991; Suzuki et al., 2003) and syntheticF can induce norepinephrine release in hemocytes (Ottaviani
al., 1993). Based on the analysis presented in this paper, it would expected that CRF/diuretic hormone-like peptides are present inth mollusc and annelid species.Whether the CRF, calcitonin and TCAP families form a clade of
lated peptides will require additional analysis. If CRF did, indeed,olve in species associated with the genome expansion eventsat occurred after the Radiata, but before the formation of the firstlateral animals, then it would be expected that CRF can be traced
Please cite this article in press as: Lovejoy, D.A., de Lannoy, L., Evolutiof peptides: Expansion and specialization in the vertebrates.
neu.2013.09.006
to a much wider gene family. However, it is unlikely that all genesthat duplicate retain similar functions. Some will retain a similarfunction, however others will be silenced and still others may beincorporated into other genes thus conferring a new or modifiedfunction onto those genes.
The ancestral CRF peptides may have evolved before theircognate receptors. Some studies support a series of genomeexpansion events that occurred after the emergence of the radiates(Ctenophora, Cnidaria) and before that of bilateral organisms.Hydra, for example, possesses a large number of bioactive peptidemodulators yet has few G-protein coupled receptors (GPCR). Apaucity of GPCRs have been noted all through the Ctenophora andCnidaria (Steele, 2002). Moreover, the CRF and calcitonin receptorsappear to be more structurally related to each other than eitherlineage is to other classes of GPCRs. A structural and phylogeneticanalysis of GPCRs indicates that the CRF and calcitonin receptorsdiverged early after the radiation of the GPCRs (Fredricksson et al.,2003). If so, we might surmise that the basic CRF family evolvedaround the period of time that led to the formation of bilateralorganisms and after the proliferation of GPCRs. Given this, ourspeculation of the CRF family occurring as part of a set of geneduplications following an early horizontal gene transfer event maybe a plausible hypothesis.
Uncited reference Q3
Lundin (1993).
Acknowledgements
The authors thank the Natural Sciences and EngineeringResearch Council (NSERC) for funding to complete this work. Inaddition, we thank the University of Toronto for graduate fundingto Louise de Lannoy.
References
Audsley, N., Goldsworthy, G.J., Coast, G.M., 1997. Circulating levels of locust diuretichormone: the effect of feeding. Peptides 18, 59–65.
Backstrom, T., Winberg, S., 2013. Central corticotropin releasing factor and socialstress. Front. Neurosci. 7, 117.
Bale, T.L., Vale, W.W., 2004. CRF and CRF receptors: role in stress responsivity andother behaviours. Annu. Rev. Pharmacol. Toxicol. 44, 525–557.
Barsyte, D., Tipping, D.R., Smart, D., Conlon, J.M., Baker, B.I., Lovejoy, D.A., 1999.Rainbow trout (Oncorhynchus mykiss) urotensin-I: structural differences be-tween urotensins-I and urocortins. Gen. Comp. Endocrinol. 115, 169–177.
Bergh, C., Sodersten, P., 1996. Anorexia nervosa, self starvation and the reward ofstress. Nat. Med. 2, 21–22.
Brugge, V.A., Schooley, D.A., Orchard, I., 2008. Amino acid sequence and biologicalactivity of a calcitonin-like diuretic hormone (DH31) from Rhodnius prolixus. J.Exp. Biol. 211, 382–390.
Bittencourt, J.C., Sawchenko, P.E., 2000. Do centrally administered neuropeptidesaccess cognate receptors? An analysis in the central corticotropin-releasingfactor system. J. Neurosci. 20, 1142–1156.
Campbell, R.K., Satoh, N., Degnan, B.M., 2004. Piercing together the evolution of thevertebrate endocrine system. Trends Genet. 20, 359–366.
Carton, Y., Frey, F., Stanley, D.W., Vass, E., Nappi, A.J., 2002. Dexamethasoneinhibition of the cellular immune response of Drosophila melanogaster againsta parasitoid. J. Parasitol. 88, 405–407.
Chand, D., Lovejoy, D.A., 2011. Stress and reproduction: controversies and chal-lenges. Gen. Comp. Endocrinol. 171, 253–257.
Chand, D., de Lannoy, L., Tucker, R.P., Lovejoy, D.A., 2013. Q4Origin of chordate peptidesby horizontal protozoan gene transfer in early metazoans and protists: Evolu-tion of the teneurin C-terminal associated peptides. Gen. Comp. Endocrinol. (inpress).
D.A. Lovejoy, L. de Lannoy / Journal of Chemical Neuroanatomy xxx (2013) xxx–xxx 7
G Model
CHENEU 1253 1–7
De Pedro, N., Pinillos, M.L., Valenciano, A.I., Alonzo-Bedate, M., Delgado, M.J., 1998.Inhibitory effect of serotonin on feeding behaviour in goldfish: involvement ofCRF. Peptides 19, 505–511.
Dunn, A.J., Berridge, C.W., 1990. Physiological and behavioral responses to cortico-tropin-releasing factor administration: is CRF a mediator of anxiety or stressresponses? Brain Res. Rev. 15, 71–100.
Fredricksson, R., Lagerstrom, L., Lundin, L.G., Schioth, H., 2003. The G protein coupledreceptors in the human genome form five main families, phylogenetic analysis,paralogon groups and fingerprints. Mol. Pharmacol. 63, 1256–1272.
Fritsch, H.A., Van Noorden, S., Pearse, A.G., 1982. Gastrointestinal and neurohor-monal peptides in the alimenatary tract and cerebral complex of Ciona intes-tinalis (Ascidaceae). Cell Tissue Res. 223, 369–402.
Furuya, K., Milchak, R.J., Schegg, K.M., Zhang, J., Tobe, S.S., Coast, G.M., Schooley, D.A.,2000. Cockroach diuretic hormones: characterization of a calcitonin-like pep-tide in insects. Proc. Natl. Acad. Sci. U.S.A. 97, 6469–6474.
Gonzalez, G.C., Martinez-Padron, M., Ko, D., Lukowiak, K., Lederis, K., 1991. Uro-tensin I-like immunoreactivity in the central nervous system of Aplysia cali-fornica. Peptides 12, 787–793.
Guillemin, R., Rosenberg, B., 1955. Humoral hypothalamic control of anteriorpituitary: a study with combined tissue cultures. Endocrinology 57, 599–607.
Ikemoto, T., Park, M.K., 2006. Comparative genomics of the endocrine system inhumans and chimpanzees with special reference to GnRH 2 and UCN2 and theirreceptors. Genomics 87, 459–462.
Hsu, S.Y., Hseuh, A.J.W., 2001. Human stresscopin and stresscopin-related peptideare selective ligands for the type 2 corticotropin-releasing hormone receptor.Nat. Med. 7, 605–611.
Huising, M.O., Flik, G., 2000. The remarkable conservation of corticotropin-releasinghormone (CRH)-binding protein in the honeybee (Apis mellifera) dates the CRHsystem to a common ancestor of insects and vertebrates. Endocrinology 146,2165–2170.
Kuperman, Y., Chen, A., 2008. Urocortins; emerging metabolic and energy homeo-stasis perspectives. Trends Endocrinol. Metab. 19, 122–129.
Lewis, K., Li, C., Perrin, M.H., Blount, A., Kunitake, K., Donaldson, C., Vaughan, J.,Reyes, T.M., Gulyas, J., Fischer, W., Bilezikjian, L., Sawchenko, P.E., Vale, W.W.,2001. Identification of urocortin III, an additional member of the corticotropin-releasing factor (CRF) family with high affinity for the CRF2 receptor. Proc. Natl.Acad. Sci. U.S.A. 98, 7570–7575.
Liu, J.P., Robinson, P.J., Funder, J.W., Engler, D., 1990. The biosynthesis and secretionof adrenocorticotropin by the ovine anterior pituitary is predominately regu-lated by arginine vasopressin (AVP). J. Biol. Chem. 265, 14136–14142.
Lovejoy, D.A., 2005. Neuroendocrinology: An Integrated Approach. John Wiley andSons, Chichester, UK, pp. 50–52.
Lovejoy, D.A., 2006. Corticotrophin-releasing hormone (CRH) peptide family. In:Kastin, A. (Ed.), Handbook of Biologically Active Peptides. Academic Press, SanDiego, pp. 655–662.
Lovejoy, D.A., 2009. Structural evolution of urotensin-I: retaining ancestral func-tions before corticotropin-releasing hormone evolution. Gen. Comp. Endocri-nol. 164, 15–19.
Lovejoy, D.A., Aubry, J.-M., Turnbull, A., Sutton, S., Potter, E., Yehling, J., Rivier, C.,Vale, W.W., 1998. Ectopic expression of the CRF-binding protein: minor impacton HPA axis regulation but induction of sexually dimorphic weight gain. J.Neuroendocrinol. 10 , 483-449.
Lovejoy, D.A., Balment, R.J., 1999. Evolution and physiology of the corticotropin-releasing factor (CRF) family of neuropeptides in vertebrates. Gen. Comp.Endocrinol. 155, 1–22.
Lovejoy, D.A., Barsyte, D., 2011. Sex, Stress and Reproductive Success. Wiley Black-well, Chichester, UK, pp. 188pages.
Lovejoy, D.A., Barsyte-Lovejoy, D., 2010. Characterization of a diuretic hormone-likepeptide from tunicates: Insight into the origins of the vertebrate corticotropin-releasing factor (CRF) family. Gen. Comp. Endocrinol. 165, 330–336.
Please cite this article in press as: Lovejoy, D.A., de Lannoy, L., Evolutioof peptides: Expansion and specialization in the vertebrates. J.neu.2013.09.006
Lovejoy, D.A., Jahan, S., 2006. Phylogeny and evolution of the corticotropin releasingfactor family of peptides. Gen. Comp. Endocrinol. 146, 1–8.
Lundin, L.G., 1993. Evolution of the vertebrate genome as reflected in paralogouschromosome regions in man and house mouse. Genomics 16, 1–19.
Lundin, L.G., Larhammar, D., Hallbook, F., 2003. Numerous groups of chromosomalregional paralogies strongly indicate two genome doublings at the root of thevertebrates. J. Struct. Funct. Genom. 3, 53–63.
Maury, B., Marinand-Mari, C., Chambon, J.P., Soule, J., Degols, G., Sahuquet, A., Weill,M., Berthomieu, A., Fort, P., Mangeat, P., Baghdiguian, S., 2006. Fertilizationregulates apoptosis of Ciona intestinalis extra-embryonic cells through thyrox-ine (T4)-dependent NF-kB pathway activation during early embryonic devel-opment. Dev. Biol. 289, 152–165.
Montecucchi, P.C., Henschen, A., Erspamer, V., 1979. Structure of sauvagine, avasoactive peptide from the skin of a frog. Hoppe-Seyler’s. Z. Physiol. Chem.360, 1178.
Nock, T.G., Chand, D., Lovejoy, D.A., 2011. Identification of members of the gonado-tropin-releasing hormone (GnRH), corticotropin-releasing factor (CRF) familiesin the genome of the holocephalan, Callorhinchus milii (elephant shark). Gen.Comp. Endocrinol. 171, 237–244.
Ottaviani, E., Caselgrandi, E., Franchini, A., Franceschi, C., 1993. CRF provokes therelease of norepinephrine by hemocytes of Viviparus ater (Gastropoda, Proso-branchia): further evidence in favour of the evolutionary hypothesis of themobile immune-brain. Biochem. Biophys. Res. Commun. 193, 446–452.
Reyes, T.M., Lewis, K., Vale, W.W., 2001. Urocortin II: a member of the corticotropin-releasing factor (CRF) neuropeptide family that is selectively bound by type 2CRF receptors. Proc. Natl. Acad. Sci. U.S.A. 98, 2843–2848.
Rode, B.M., 1999. Peptides and the origin of life. Peptides 20, 773–786.Rode, B.M., Plankensteiner, K., 2006. Prebiotic peptides. In: Kastin, A. (Ed.), Hand-
book of Biologically Active Peptides. Academic Press, San Diego, pp. 1481–1486.Schally, A.V., Saffran, M., 1955. The release of corticotrophin by anterior pituitary
tissue in vitro. Can J. Physiol. Pharmacol. 33, 408–415.Shen, P.J., Clarke, I.J., Canny, B.J., Funder, J.W., Smith, A.I., 1990. Arginine vaspression
and corticotropin-releasing factor: binding to ovine anterior pituitary mem-branes. Endocrinology 127, 2085–2089.
Sherwood, N.M., Tello, J.A., Roch, G.J., 2006. Neuroendocrinology of protochordates:insights from Ciona genomics. Comp. Biochem. Physiol. A 144, 254–271.
Solinas, G., Summermatter, S., Mainieri, D., Gubler, M., Montani, J.P., Seydoux, J.,Smith, S.R., Dulloo, A.G., 2006. Corticotropin-releasing hormone directly sti-mulates thermogenesis in skeletal muscle possibly through substrate cyclingbetween de novo lipogenesis and lipid oxidation. Endocrinology 147, 31–38.
Spina, M., Merlo-Pich, E., Chan, R.K.W., Basso, A.M., Rivier, J., Vale, W., Koob, G., 1996.Appetite-suppressing effects of urocortin, a CRF-related peptide. Science 273,1561–1564.
Steele, R., 2002. Developmental signaling in hydra, what does it take to build a‘simple’ animal. Dev. Biol. 248, 199–219.
Sutton, S.W., Behan, D.P., Lahrichi, S.L., Kaiser, R., Corrigan, A., Lowry, P., Potter, E.,Perrin, M., Rivier, J., Vale, W.W., 1995. Ligand requirements of the humancorticotropin-releasing factor binding protein. Endocrinology 136, 1097–1102.
Suzuki, H., Muraoka, T., Yamamoto, T., 2003. Localization of corticotropin-releasingfactor-immunoreactive nervous tissue and colocalization with neuropeptide Y-like substance in the optic lobe and peduncle complex of the octopus (Octopusvulgaris). Cell Tissue Res. 313, 129–138.
Vale, W., Spiess, J., Rivier, C., Rivier, J., 1981. Characterization of a 41-residue ovinehypothalamic peptide that stimulates secretion of corticotropin and b-endor-phin. Science 213, 1394–1397.
Vaughan, J., Donaldson, C., Bittencourt, J., Perrin, M.H., Lewis, K., Sutton, S., Chan, R.,Turnbull, A., Lovejoy, D., Rivier, C., Rivier, J., Sawchenko, P.E., Vale, W., 1995.Urocortin, a mammalian neuropeptide related to fish urotensin-I and to corti-cotropin-releasing factor. Nature 378, 287–292.
Zhang, D., deSouza, R.F., Anantharaman, V., Iyer, L.M., Aravind, L., 2012. Polymorphictoxin systems: comprehensive characterization of trafficking modes, mecha-nism of action, immunity and ecology using comparative genomics. Biol. Direct7, 18.
n and phylogeny of the corticotropin-releasing factor (CRF) family Chem. Neuroanat. (2013), http://dx.doi.org/10.1016/j.jchem-
