Revisiting the gastrin-releasing peptide/bombesin system: A reverse-evolutionary study considering Xenopus

Gastrin-releasing peptide (GRP), first isolated from the porcine stomach, is a neuropeptide that modulates the autonomic system in mammals and has previously been considered to be the mammalian equivalent of bombesin, a fourteen amino acid peptide first isolated from the skin of the European fire-bellied toad, Bombina bombina. Bombesin-like peptides and the related neuromedin B (NMB) have since been identified in mammals. However, the orthologous relationships among GRP/NMB/bombesin and their receptors in vertebrates are still not well understood. Our studies have focused on the GRP system that is widely conserved among vertebrates. We have used phylogenetic analysis and reverse transcription-PCR, quantitative PCR, immunohistochemistry, and Western blotting experiments to examine the expression of both GRP and its receptor (GRPR) in a clawed frog (Xenopus tropicalis) and to understand the derivation of GRP system in the ancestor of mammals. We demonstrate, by phylogenetic and synteny analyses, that GRP is not a mammalian counterpart of bombesin and also that, whereas the GRP system is widely conserved among vertebrates, the NMB/bombesin system has diversified in certain lineages, in particular in frog species. In Xenopus, we found the expression of the mRNA for both GRP and GRPR in the brain and stomach. In addition, our quantitative PCR analysis shows that, in Xenopus, the expression of GRP mRNA is highest in the brain, whereas expression of GRPR mRNA is highest in the spinal cord. Our immunohistochemical analysis shows that GRP-immunoreactive cell bodies and fibers are distributed in several telencephalic, diencephalic, and rhombencephalic regions and spinal cord of Xenopus. Our Western blotting analysis also indicates the presence of GRPR protein in the brain and spinal cord of Xenopus. We conclude that GRP peptides and their receptors have evolved to play multiple roles in both the gut and brain of amphibians as one of the ‘gut-brain peptide’ systems. Author Summary Bombesin is a putative antibacterial peptide isolated from the skin of the frog, Bombina bombina. Two related (bombesin-like) peptides, gastrin-releasing peptide (GRP) and neuromedin B (NMB) have been found in mammals. The history of GRP/bombesin discovery has caused little attention to be paid to the evolutionary relationship of GRP/bombesin and their receptors in vertebrates. We have classified the peptides and their receptors from the phylogenetic viewpoint using a newly established genetic database and bioinformatics. We demonstrate, by phylogenetic and synteny analyses, that GRP is not a mammalian counterpart of bombesin and also that, whereas the GRP system is widely conserved among vertebrates, the NMB/bombesin system has diversified in certain lineages, in particular in frogs. Gene expression analyses combined with immunohistochemistry and Western blotting experiments indicate that GRP peptides and their receptors have evolved from ancestral (GRP) homologues to play multiple roles in both the gut and the brain as one of the ‘gut-brain peptide’ systems of vertebrates, which is distinct from the frog bombesin lineage.


37
Gastrin-releasing peptide (GRP), first isolated from the porcine stomach, is a 38 neuropeptide that modulates the autonomic system in mammals and has previously been 39 considered to be the mammalian equivalent of bombesin, a fourteen amino acid peptide first 40 isolated from the skin of the European fire-bellied toad, Bombina bombina.  peptides and the related neuromedin B (NMB) have since been identified in mammals. 42 However, the orthologous relationships among GRP/NMB/bombesin and their receptors in 43 vertebrates are still not well understood. Our studies have focused on the GRP system that is 44 widely conserved among vertebrates. We have used phylogenetic analysis and reverse 45 transcription-PCR, quantitative PCR, immunohistochemistry, and Western blotting 46 experiments to examine the expression of both GRP and its receptor (GRPR) in a clawed frog 47 (Xenopus tropicalis) and to understand the derivation of GRP system in the ancestor of 48 mammals. We demonstrate, by phylogenetic and synteny analyses, that GRP is not a 49 mammalian counterpart of bombesin and also that, whereas the GRP system is widely 50 conserved among vertebrates, the NMB/bombesin system has diversified in certain lineages, 51 in particular in frog species. In Xenopus, we found the expression of the mRNA for both GRP 52 and GRPR in the brain and stomach. In addition, our quantitative PCR analysis shows that, in 53 Xenopus, the expression of GRP mRNA is highest in the brain, whereas expression of GRPR 54 mRNA is highest in the spinal cord. Our immunohistochemical analysis shows that 55 GRP-immunoreactive cell bodies and fibers are distributed in several telencephalic, 56 diencephalic, and rhombencephalic regions and spinal cord of Xenopus. Our Western blotting 57 analysis also indicates the presence of GRPR protein in the brain and spinal cord of Xenopus. 58 We conclude that GRP peptides and their receptors have evolved to play multiple roles in both 59 the gut and brain of amphibians as one of the 'gut-brain peptide' systems.

60
(300/300 words PLoS Genetics) The fourteen-amino acid peptide, bombesin, was initially described as a possible 82 antibacterial peptide isolated from the skin of the European fire-bellied toad, Bombina 83 bombina, and was shown to have potent bioactivity in the mammalian nervous system [1,2].

84
Subsequently, the mammalian bombesin-like peptides, gastrin-releasing peptide (GRP) [  165 their genomes have been decoded. In addition, the sequences of prepro-bombesin-like 166 peptides, which have been reported for other frog species were also included (Fig 1).

167
The prepro-GRP/NMB/bombesin sequences were divided into two major clades: 168 GRP and NMB/bombesin clades (Fig 1). A single GRP gene was found in almost all animals were also included in the bombesin clade (Fig 1).

179
These results indicate two possibilities for the evolution of NMB/bombesin: one is 219 detectable in these all tissues of both sexes, no sex differences were detected in any of the 220 tissues we examined (blue bars indicate means of males, red bars indicate means of females) 221 ( Fig 5). In both sexes, the expression of GRP mRNA was higher in the 222 diencephalon/mesencephalon/pons/cerebellum (2) and the medulla oblongata (3), than in the 223 telencephalon (1) and the spinal cord (4) (Fig 5a). In contrast, the expression of GRPR mRNA 224 was the highest in the spinal cord (4), and the lowest in the telencephalon (1) (Fig 5b). 234 antigen peptide ( 2 [Ser] form-NMC); in these experiments no immunostaining was seen (Fig 7).
235 In the spinal cord, GRP + fibers and numerous varicosities were found throughout the spinal 236 grey matter area (Fig 7a,

265
Western immunoblot analysis with the polyclonal antiserum against Xenopus GRPR 266 was performed to determine the presence of GRPR protein in homogenates derived from the 267 brain and spinal cord of adult male Xenopus. An intense protein band was observed in the 268 brain and spinal cord extracts, and its electrophoretic mobility was located at ~43 kDa, which 269 is the expected molecular weight of Xenopus GRPR (Fig 10). Preabsorption of the antiserum 270 with an excess of antigen peptides (50 µg/mL) prevented the immunostaining of the ~43-kDa 271 protein band in the brain and spinal cord (Fig 10). Immunoblot analyses were repeated 272 independently three times by using different three frogs and gave similar results. 285 and NMB/bombesin clades (Fig 1). We further found by using synteny analysis that bombesin 286 and NMB are relative orthologs and that specialization of the NMB sequence only in the frog 287 lineage resulted in bombesin (Fig 2). When tetrapods emerged from the sea, they may have 288 started to produce bombesin in the skin as exocrine secretions to protect themselves from 289 bacterial infection and/or predators. To our knowledge, this is the first demonstration of the 290 gene divergence of bombesin and bombesin-like peptides in vertebrates, which appeared to be 291 divided into the two families (i.e. GRP and NMB/bombesin) from the vertebrate ancestor (Fig   292 11).

293
For bombesin-like peptide receptors, we found that most gnathostomes have an 294 orthologous gene in each of the three groups (GRPR/NMBR/BRS-3), although BRS-3 was not 295 found in Teleostei and Chondrichthyes (Fig 3). This suggests that GRPR/NMBR/BRS-3 296 diverged into three branches in the ancestor of gnathostomes but that the BRS-3 genes have 297 been lost in Teleostei and Chondrichthyes lineages. Otherwise, it is possible that there were 298 two genes for GRPR and NMBR respectively at the divergence of cartilaginous fish, and that 325 but also as neuromodulators (neuropeptides); so called 'gut-brain peptides' in mammals [41].
326 GRP appears to be one of gut-brain peptides, and the GRP system might play multiple roles in  Table 1). In particular, we observed many GRP + cell bodies in the hypothalamus and 340 putative limbic system in Xenopus, which corresponds well with the mammalian case [5].
341 GRP + fibers were also distributed widely throughout the CNS of Xenopus (Fig 6 and

354
In conclusion, GRP has long been considered as the mammalian equivalents of 355 bombesin [6, 7] (Fig 11a). We now demonstrate, by phylogenetic and synteny analyses, that 356 GRP is not a mammalian counterpart of bombesin, and that the GRP system is widely 357 conserved throughout vertebrates, whereas the NMB/bombesin system diversified in some 358 lineages (Fig 11b). Furthermore, we demonstrate that the GRP system might play multiple 359 roles both in the gut and in the brain of amphibians as one of the 'gut-brain peptide' systems.    Table 2 Table 2.  Table 3. 525 supervised the whole study. All authors had full access to all the data in the study and take 526 responsibility for the integrity of the data and the accuracy of the data analysis.