The Crustacean Hyperglycemic Hormone Superfamily: Progress Made in the Past Decade

Isoforms of CHH were firstly characterized with regard to their specific functions. For example, CHHs sensu stricto are pleiotropic hormones mainly involved in the regulation of metabolism or water and ion balance, or may even have a negative effect on reproduction and molting. However, MIH/GIH/MOIH seemingly have, in the current state of our knowledge, more limited functions. These functional differences are obviously associated with structural differences which have enabled them to be distinguished and classified as distinct types, such as Type I and Type II, respectively (33). Type I brings together peptides with a cryptic sequence or PRP (precursor related peptide) and a dibasic cleavage site upstream of the mature peptide sequence. Type II is characterized by the absence of this PRP and has a glycine residue in position 5 after the first cysteine (33–35). It is important to keep in mind that if the first sequences were obtained from an identified biological function, this is no longer the case for most member peptides identified subsequently, as most have been characterized mainly by high-throughput sequencing techniques.

CHH-superfamily members are clearly no longer restricted to crustaceans, since ITP (ion transport peptide) was found in locusts (37, 38). ITP has since been found not only in other hexapods but also in chelicerates (39–42, 44, 99–101) and nematodes (102). ITPs were initially believed to be specific to non-crustaceans until they were identified in non-malacostraca crustaceans such as phyllopods (31, 103, 104), copepods (105–107), and remipedes (108). In these species, the presence of ITP seems to be exclusive since no other CHH peptide has been characterized simultaneously. This isoform presents a precursor organization closer to that of CHHs than MIH/GIH/MOIH, i.e., the presence of a PRP and a dibasic cleavage site, which has led to its inclusion in Type I (31). However, phylogenetic studies, carried out on the basis of sequences of the three isoforms, clearly place the ITPs, including those obtained from non-malacostracan crustaceans and hexapods, at the base of Types I and II. This observation led to the proposal to create a third group, Type III, which brings together all these ITP orthologues (31, 45).

The analysis of recent transcriptomes has added another level of complexity by identifying a new form of peptide belonging to the superfamily (45, 109, 110). These new sequences possess the six conserved cysteine residues, but do not have a PRP sequence, nor a dibasic cleavage site or the glycine residue, a signature of MIH (45). They are also significantly longer than other isoforms with more than 80 amino acids. Nevertheless, their BLAST hit clearly clusters them with ITP isoforms of insects and Cladocera crustaceans such as Daphnia rather than CHHs or MIH/GIHs. For this reason, they have been grouped together under the name ITP-like (45). Thus, inclusion of ITP-like sequences in a phylogenetic analysis conducted on more than 200 sequences of family members confirms their belonging to the same group and positions them as an independent group (45). The features of their sequences exclude them from the types currently recognized and once again question the dichotomous classification proposed by Lacombe et al. (33). It constitutes a new set of members in the same way as CHHs sensu stricto or MIHs and would thus form a fourth type (Type IV), if we refer to phylogeny.

The analysis of the different transcriptomes also addresses the number of isoforms, notably CHHs, present within each species. Information collected from these analyses revealed the presence of many additional isoforms within each species, which are growing more numerous as the number of tissues analyzed has increased. While there are so far only one or two isoforms of MIH/GIH (111–116), CHH peptides exhibit much greater variety with commonly four or more different isoforms (45, 117). The presence of multiple chh genes has been demonstrated in decapods, (up to nine) whereas only one copy of the itp gene is present in insects (29–31, 57). Similarly, only two mih genes have been identified in several species (111, 118). Some are clearly derived from recent duplications and show little divergence (119), while others differ substantially from the most represented isoforms and clearly demonstrate that diversity has been underestimated (45). It is very likely that additional transcriptomic analyses will uncover new isoforms, even in species that have already been well studied (119). This diversity supports the importance of this molecular family at the physiological level and could also explain the apparent pleiotropy of the CHHs.

Beyond this genomic diversity, there are mechanisms in some crustaceans that can produce new isoforms by post-transcriptional or post-translational modification, as detailed below.

Alternative splicing of the chh gene in pericardial organs of C. maenas (47) results in a CHH isoform that lacks the defining activity of CHH. Alternative splicing has since been demonstrated for several CHH genes in malacostracan crustaceans (47, 48, 53, 60, 72, 120) but also for insect ITPs (51, 52, 57, 59). These splicing patterns originate in the structure of the chh or itp genes, which have a minimum number of four exons. The splicing alternates two exons (exon 3 or 4) as the terminal part of the transcript, resulting in two transcript forms—the short-splice forms (exons 1, 2, 4) and the long-splice forms (exons 1, 2, 3, 4), in which the stop codon is located at the end of the third exon, leaving the fourth exon untranslated (Figure 1). This process generates two peptides from a single gene that differ in sequence after the 40^th amino acid residue of the mature peptide. Exons 3 and 4 both encode peptide sequences of similar lengths, with two cysteine residues in the same positions, thus suggesting that alternative exon 4 appeared by tandem duplication of exon 3, or vice versa (31). The transcripts of the short-splice forms of CHH peptides were mainly found in neurons of the central nervous system (more precisely in the X-organ located in the eyestalk), whereas the long-splice forms were found in neurons of the peripheral nervous system and non-nervous tissues, thoracic ganglia, Y-organ, mandibular organ or hemocyte according the species (47, 48, 55, 60, 70, 121). Peptides encoded by the long-splice forms of CHH and ITP transcripts are slightly longer than peptides encoded by the short-splice transcripts and are never C-terminally amidated [see (56)]. They are named CHH-L and ITP-L, respectively. In addition to these genes with four exons, there are also genes with only three or even two exons, such as the majority of CHH isoforms of penaeid shrimps. MIH and MOIH are encoded by genes with three exons (29, 31, 122, 123). Alternative splicing mechanisms, as described here, have never been reported for these two- or three-exon genes of penaeid shrimps. It is also interesting to note that this splicing process can generate “truncated” forms of CHH as demonstrated in the hemocyte of the crayfish P. clarkii (72), resulting in a precursor, where the mature peptide sequence is truncated with only the first 40 residues.

The following examples underline the widespread ubiquity of the CHH superfamily members, which demonstrates the surprising diversity of these peptide isoforms and their functions. Indeed, if the first isoforms were characterized at the level of neural ganglia, first in the XO-SG complex in the eyestalks and then followed by the thoracic ganglia, the detection of these peptides in other tissues is becoming increasingly common. The first evidence was provided by immunocytochemical analysis in C. maenas, in cells located at the fore and hindgut level, suggesting a mechanoreceptive function. These are “paraneurones” which secrete CHH into the hemolymph during ecdysis (68, 124). Since then, it has been shown that CHH-related peptides are expressed in a wide variety of tissues including the pericardial organ (120, 125), retina (126), gill (60, 127), spermatophore (71, 128), Y-organ (121, 129), hemocyte (70, 72), ovary (129, 130), stomach (130), and intestine (54).

Until recently, it was thought that MIH/GIH/MOIH peptides were only expressed in neuronal tissues of the XO-SG system. However, MIH expression has been identified in the heart of the shrimp Litopenaeus vannamei (131), and in the gill, ovary, and abdominal ganglion of Macrobrachium nipponense, though at relatively lower level than in the eyestalks (132).

In parallel with gene multiplication or post-transcriptional processes such as alternative splicing, peptide modification in some taxa further increases the diversity of isoforms and functions of Type I or II members. This mechanism, hitherto restricted to Astacidea in crustaceans but present as well in other arthropods such as arachnids and mollusks and a few vertebrates, involves production of peptides in which the chiral form of a single amino acid has been modified, i.e. from the L-enantiomer to the D-enantiomer (61). Thus, lobsters and crayfish have CHH isomers that are differentiated by the configuration of the Phe³, as either an L-form or a D-form. Both isomers exhibit the defining hyperglycemic activities and regulate energy metabolism, but the change in configuration results in modification of the biological activity of the peptide (64, 66, 133). In particular, the hyperglycemic response kinetics are delayed with d-Phe³-CHH (3 to 4 h instead of 2 h for the L-isoform), and the amplitude of the hyperglycemic response is increased (10x higher for d-Phe³-CHH). Additional functional changes for d-Phe³-CHH were reported as having more potent inhibitory activity on the molting gland during ecdysteroidogenesis (65) or having a higher osmoregulatory activity (62, 63).

A similar modification from l-GIH to d-GIH has also been demonstrated in three Nephropidae species (American, European, and Norway lobsters). In this case, the modification is located at the level of the fourth N-terminal amino acid, a tryptophan (134, 135). However, the function of d-Trp⁴-GIH is still unknown, as it does not display significantly inhibitory effect in a heterologous in vivo assay as its counterpart does.

The study of the maturation dynamics of the CHH isomers highlighted the different stages of the post-translational processing of the precursor (Figure 2). It appears that isomerization takes place after cleavage of the pro-peptide and before N-terminal cyclization (136, 137). The production of antibodies specifically directed against the two CHH isomers made it possible to monitor in situ their location during the maturation process in the XO-SG complex of eyestalks of the crayfish O. limosus and the lobster H. americanus (135, 136). Two cell types producing either L-CHH exclusively (L-CHH cell) or a mixture of both L- and D-isomers (D-CHH cell) were identified (Figure 2). Most of the transformation of the L-isomers seems to occur in the cytoplasm, before the granules penetrate the axons (138). A similar approach was conducted with GIH isomers. Differential localization of the two isomers in different cells was characterized, as for CHHs. In the same way, simultaneous monitoring of the four isomers showed co-localizations within the same cells. Only L-GIH and L-CHH could not be found together (135) (Figure 2).

These studies have highlighted a particularly original mechanism of post-translational modification and curiously, so far, this has been demonstrated only in Astacidea crustaceans. Because this L-to D-post-translational modification is subtle and not detectable by most sequence determination approaches, it cannot be excluded that it exists in other taxa, although it has been searched for in other crustaceans, without success. Nevertheless, this process is not specific to crustaceans. As mentioned above, it has not only been found in arachnids, but also in molluscs, tetrapods, and mammals (61).

Another post-translational modification, a C-terminal amidation event, seems particularly important to limit degradation of the peptides by carboxypeptidases (139). The C-terminal amidation concerns only CHH and ITP, while the long-splice form (CHH-L or ITP-L) do not present this modification. In addition to the protective role in resisting degradation, the amidation is also important for the activity of the peptides, since the amidated peptides, compared to the un-amidated ones, have more potent activities (139–141). This post-translational modification has also been infrequently observed in type-II peptides such as GIH or MIH, although the functional significance of the modification is not clear (132, 142, 143). Like the C-terminal extremity, the N-terminus has, again only for CHH and ITP, a post-translational conversion via cyclization of the terminal Glu or Gln residue into pyroGlu (56, 57). This modification would protect the N-terminus from degradation by aminopeptidases, but does not significantly change the biological activities of the peptide, as the N-terminally unblocked and blocked CHHs of C. maenas were shown to be almost distinguishable in the activities of elevating hemolymph glucose levels and of repressing ecdysteroid synthesis (144).

Among the peptides characterized with D-amino acyl residues, it is interesting to note that there are many examples of these molecules that have been extracted from venoms or the nervous system (145, 146). This link between members of the CHH superfamily and venom peptides was further strengthened by the discovery that certain peptides of low molecular weight identified in black widow spider venom (genus Latrodectus), e.g., the latrodectins, show sequence similarities with CHH, particularly in terms of size, around 70 residues, the signature of the six conserved cysteines including similar disulfide bond pairing, and a similar alpha-helical structure (99). Further analyses suggested that CHH-superfamily peptides might have been recruited for venom expression at least three times: once in Hymenoptera, once in scorpions, and at least once in spiders, each lineage independently evolving venom production (44). However, the functional role of the latrodectins remains unclear as they are not insecticidal or toxic to mice (99), nor do they produce hyperglycemic effects when injected into crabs, crayfish, or shrimps (147).

An exception was however first identified in the venom of common agelenid spider Tegeneria agrestis, the U1-agatoxin-Ta1a (148, 149), and then discovered in the venoms of members of the centipede genus Scolopendra, representing convergent recruitment of CHH-superfamily peptides into the toxin arsenals of these myriapods (43). Despite a weak sequence identity, the structure of MIH of the shrimp M. japonicus and CHH-L of the crab S. olivacea are surprisingly topologically similar to that of Ta1a. However, the N-terminal α-helix (α 5), present in both MIH and CHH-L, is absent from Ta1a [Figure 3; also see (43)] and the loss of this helix could be a key step in weaponization of CHH family peptides (43).

Recently, a proteomics approach of the venom components of the parasitoid wasp Tetrastichus brontispae resulted in the identification of, among a large number of diverse proteins or peptides, three ITP-L (ITPLn1-3) as abundant toxins (150). These ITP-Ls are distinct from ITPs by the implementation of a possible alternative splice allowing diversification of the gene products and a potential neo-functionalization that would lead to an evolution towards peptides of the CHHs family with venomous properties.

The mechanisms for recruiting members of the CHH superfamily as toxins are clearly varied and appear to have evolved independently in the different taxa. Nevertheless, the use of ITP/CHHs as evolutionary fodder for these transformations is a consensus among a growing number of species and is certainly not fortuitous. The structural properties (3 disulfide bridges which guarantee great stability), the multiplicity of genes and the post-transcriptional or even post-translational mechanisms are all processes that can lead to neo-functionalization, of which the acquisition of toxicity is one. In this context, one of the consequences of the acquisition of a D-amino acid peptide is a higher resistance to proteases and therefore a longer half-life for these peptides (146). In the future, it would not be surprising to find organisms with venomous ITP/CHHs carrying this type of modification.

CHH and ITP were functionally defined by the hyperglycemic activity and stimulation of trans-epithelial Cl^– transport in the ileum, respectively, which formed the basis (by bioassay) for purification, chemical characterization, and eventual identification of the hormones (19, 22, 37, 38, 151, 152). Studies over the years have however revealed that the two peptides are more functionally conserved than the initial defining functions would have suggested.

The first structural model of the CHH superfamily peptide was that of MIH (Pej-MIH) from M. japonicus resolved by nuclear magnetic resonance (NMR) spectroscopy. It consists of a long N-terminal tail, followed by five alpha helices (α1−α5), and a C-terminal tail (92). Subsequently, two more structures of the crustacean member peptides have been elucidated: Pej-SGP-I-Gly, the glycine-extended precursor of a M. japonicus CHH by X-ray crystallography (91), and Sco-CHH-L, the CHH-L from the pericardial organ of S. olivacea by NMR spectroscopy (Protein Data Bank: 5XS1). Although topologically equivalent helices—α2, α3, and α4 (π4 in Sco-CHH-L)—are present in the core of the three structures, Pej-MIH and Sco-CHH-L are more topologically resemble one another than Pej-SGP-I-Gly (Figure 3). Thus, Pej-MIH and Sco-CHH-L additionally have a topologically similar C-terminal α5, which brings the C- and N-terminal ends sterically close each other (Figure 3). In contrast, Pej-SGP-I-Gly, which lacks an α5, has a relatively long α4 followed by a C-terminal tail that is kept away from the N-terminal end (Figure 3). Because of the limited availability of the C-terminal amidated Pej-CHH, Pej-SGP-I-Gly, a non-amidated precursor was used for structural determination (91). Post-translational amidation at the C-terminal increased the α-helical content of CHH (139, 140, 172), indicating that the C-terminal modification renders structural changes. Thus, while additional structures of the CHH-superfamily peptides (e.g., amidated CHH and ITP) are needed for validation, it is likely that the folding pattern shared by Pej-MIH and Sco-CHH-L is a common theme for the crustacean and insect peptides of the superfamily. Functionally critical residues at the terminal regions of MIH, CHH, and ITP have been demonstrated using mutated recombinant peptides (87–90, 139, 140, 227). It is likely that parts of the structure, consisting of the sterically close C- and N-terminal regions of the peptide, where the functionally critical residues are located, play important roles in forming the binding site for receptor interaction and activation.

Radiolabeled peptides (125-I) have been used in several studies to identify the potential target tissues of CHH, MIH, or CHH-L and to profile the binding characteristics of the presumed receptor (86, 193, 228, 229). Generally, each of the crustacean members of the superfamily binds to a distinct binding site (receptor) with high specificity. In the shore crab C. maenas, Y-organs clearly have separate and highly specific binding sites for CHH and MIH (193). Similarly, in the blue crab C. sapidus, displacement experiments in multiple tissues also revealed that CHH and CHH-L peptide each has discrete specific binding sites (229). Another aspect of CHH binding specificity was illustrated by studies on hepatopancreas membranes from C. maenas and O. limosus (228). Membranes derived from C. maenas had a much lower affinity for O. limosus CHH than for C. maenas CHH (228), reflecting the species-specificity of CHH in terms of hyperglycemic activity (152).

A wealth of literature has been accumulated over the years regarding the signal transduction pathways of the CHH-superfamily peptides, mostly concerning MIH, CHH, and ITP. These include studies in various species utilizing in vitro or in vivo assays with tissue preparations or hormones, cyclic nucleotide analogues, and pharmacological agents inhibiting or stimulating activity of enzymes involved in the signaling pathways. In general, the studies have concluded that cyclic adenosine monophosphate (cAMP), cyclic guanosine monophosphate (cGMP), or both act as intracellular signaling messengers mediating the action of the CHH-superfamily peptides. However, experimental data are in several instances inconsistent and contradictory, particularly for studies involving MIH. Until recently, the nature (receptor guanylyl cyclase vs. adenylyl cyclase-activating G protein-coupled receptor) of the putative receptor for the CHH-superfamily peptides were only speculated upon based on data derived from the early studies. In a more recent study of the silkworm B. mori, orphan G protein-coupled receptors (GPCRs) were determined to be receptors for ITPs (ITP and ITP-L peptides), the characterization of which provided data for building a model of the signaling transduction pathway of ITPs (94), which is in general similar to an earlier one suggested for MIH (230).

Early studies investigating CHH signal transduction pathways implicated both cyclic nucleotides as second messengers mediating the action of the hormone (156, 157). Thus, CHH preparations increased cGMP and cAMP levels in vivo in several target tissues, including the hepatopancreas and muscle of the crayfish, O. limosus. In in vitro incubations of the hepatopancreas, cGMP levels were significantly increased by CHH, which was followed by release of glucose into the tissue incubation media, whereas cAMP levels were already elevated during the incubation and not further increased by hormonal treatment (156). Further, injection of CHH into the eyestalk-ablated crayfish increased cGMP and cAMP levels and decreased glycogen synthase activity in the abdominal muscle. Cyclic nucleotide analogues mimicked the effect of CHH preparations in inhibiting synthase activity (157). In M. japonicus, bilateral eyestalk ablation decreased intracellular cGMP levels of the hepatopancreas, with little effect on cAMP levels; corroboratively, exposure of tissues to recombinant CHH significantly increased the levels of cGMP, but not those of cAMP (231). The CHH-induced increase in cGMP levels in the muscle of H. americanus (232) was due to stimulation of the membrane-bound guanylyl cyclase (mGC), as cGMP increased not only in intact tissue but also in the isolated membrane preparations (233). Involvement of soluble GC (sGC) in the CHH-induced cGMP increase was considered unlikely, as cytoplasmic sGC activity was not stimulated by CHH and the CHH-induced cGMP increase was not blocked by methylene blue, an inhibitor of the nitric oxide (NO)-activated sGC (233). These combined results favor a scenario in which CHH binds and activates an mGC, thus elevating intracellular levels of cGMP as a second messenger, while changes in cAMP, if present, could occur downstream to receptor activation and cGMP elevation. An mGC (PcGC-M2), which was identified by cDNA cloning in the muscle of P. clarkii, was found to contain the signature domains characteristic of receptor GC (rGC), including an extracellular ligand-binding domain, a single transmembrane domain, and intracellular kinase-like and cyclase catalytic domains (234). PcGC-M2 is widely expressed in several target tissues of CHH (234), although the binding ligand for PcGC-M2 has not yet been determined.

The signaling pathway activated by CHH-L is much less characterized. In the blue crab C. sapidus, CHH-L (pericardial organ-CHH) significantly increased cGMP production in several tissues including scaphognathites, heart, midgut, hindgut, and abdominal muscles (229).

Results obtained from studies of the MIH-mediated suppression of ecdysteroidogenesis in the Y-organs in different species are often contradictory. On the one hand, cGMP was shown to be an important second messenger of MIH in several species, including O. limosus, C, sapidus, and P. clarkii. Thus, Y-organ incubated with native or recombinant MIH resulted in an increase of cGMP, but not cAMP, in these species (192, 202, 235). In related experiments, cGMP invariably played a crucial role in the mediation of MIH action: a cGMP analogue (8-Br-cGMP) significantly suppressed ecdysteroid production by Y-organ of C. sapidus, but neither cAMP analogues (db-cAMP or 8-Br-cAMP) nor an activator of adenylyl cyclase (forskolin) had a detectable effect on ecdysteroidogenesis (236); addition of synthetic MIH to the incubation medium increased cGMP levels, but not cAMP levels, in P. clarkii Y-organs (192). Corroborative observations were also reported for C. maenas Y-organ in which treatment with purified MIH produced a large and sustained increase in intracellular cGMP levels (237). These results are consistent with a model in which cGMP functions as a second messenger in the cellular action of MIH and that the receptor for MIH is likely a rGC. On the other hand, studies in Cancer antennarius showed that adding eyestalk extract (which contains a wide variety of bioactive compounds, including MIH) to incubation of Y-organs resulted in an increase in cAMP levels (189). Additionally, db-cAMP and agents (forskolin and choleragen) that increased intracellular cAMP, each mimicked the inhibitory action of MIH, while 3’,5’-cGMP did not. It was thus concluded that cAMP mediates MIH-induced suppression of ecdysteroid production (189).

Data supporting the notion that the MIH receptor is an mGC and that cGMP functions as a second messenger were obtained in follow-up studies in C. sapidus. Thus, a cDNA (CsGC-YO1) encoding an rGC was cloned from C. sapidus Y-organ (238). Immunohistochemical staining using a primary antibody raised against the extracellular domain showed that CsGC-YO1 is located on the Y-organ cell membrane. Y-organ preconditioned with the anti-extracellular domain antibody became refractory to the stimulation of MIH. In addition, sodium nitroprusside (an NO donor) failed to inhibit ecdysteroidogenesis in Y-organ of C. sapidus, suggesting that NO/sGC/cGMP signaling pathway is not involved in the MIH action (196). Finally, the transcript levels of CsGC-YO1 in Y-organ of the inter-molt animals were four- and two-times higher than those in the pre-molt and post-molt Y-organs, respectively (239).

Alternatively, a model hypothesizing that the MIH receptor is a G protein-coupled receptor (GPCR) was proposed (230) based on data obtained mainly from the crab Gecarcinus lateralis (240–243). In combination with previous related studies (230, 243) and the transcriptomic analysis, the G. lateralis Y-organ was proposed to go through a four-stage transition (basal, activated, committed, and repressed stages) from inter-molt (stage C₄), early pre-molt (stage D₀), middle pre-molt (stage D₁, D₂) and late pre-molt stage (stage D₄), respectively (244, 245). The MIH signaling pathway plays a dominant role in the inter-molt and early pre-molt stage to suppress ecdysteroid production. According to the model, ligand activation of MIH receptor leads to, via a G protein, stimulation of adenylyl cyclase and subsequent cAMP-protein kinase A (PKA) pathway. Phosphorylation of enzymes by PKA in turn causes inhibition of ecdysteroidogenesis. Another phosphorylation event activated by PKA leads to an increase in membrane calcium conductance, resulting in Ca²⁺ influx that activates NO-sensitive sGC pathways through Ca²⁺/calmodulin activation of nitric oxide synthase. Increase in cGMP levels and subsequent activation of protein kinase G ultimately lead to inhibition of ecdysteroidogenesis. This working hypothesis is sustained by a series of experimental studies. First, analysis of G. lateralis Y-organ transcriptome indicated the expression of signaling components mentioned above (including adenylyl cyclases, PKA, PKG, calmodulin, NOS, NO-sensitive guanylyl cyclase (GC-I), and GPCRs) (95, 121, 246). Second, the production of ecdysteroids in vitro was repressed by NO donors (SNAP and SE175) and by cAMP and cGMP analogues (247, 248). Third, an adenylyl cyclase (AC) activator (forskolin) inhibited the production of ecdysteroids (248). Fourth, in combination with IBMX, an sGC activator (YC-1) inhibited ecdysteroid secretion in Y-organ (247). Finally, eyestalk ablation led to the up-regulation of NO-independent-NOS, GC-I, and GC-III mRNA levels, causing the Y-organ to be more sensitive to MIH stimuli, presumably a compensatory response to the removal of MIH (240).

Information about GIH and MOIH signaling cascades are still limited. Effects of various pharmacological reagents have been tested in an in vitro assay using vg mRNA expression in ovarian explants from M. japonicus (249). Results showed that db-cAMP, db-cGMP, forskolin, and IBMX mimicked the inhibitory effects of GIH in reducing the levels of vg mRNA in a dose-dependent manner. Similar results were also obtained when A23187 (calcium ionophore) and PMA (an activator of protein kinase C) were added into the incubation medium. These results suggested that cyclic nucleotides, calcium, and protein kinase C are involved in regulating Vg transcription in the ovary (249). However, whether these signaling components were coupled to GIH activation has not been determined. Recently, a study on the mode of GIH action was carried out in L. vannamei using vg expression in the hepatopancreas as the bioassay. According to Chen et al. (250), in vivo injection and in vitro incubation of hepatopancreatic primary cells with recombinant GIH elevated the levels of intracellular cGMP, but not cAMP or nitric oxide, indicating that GIH exerts a cGMP-mediated inhibitory action. Pharmacological reagents were then tested to characterize the GIH signaling pathway. Results showed that GIH employs an rGC/cGMP/PKG signaling pathway in inhibiting vg expression; phosphorylation of c-Jun N-terminal kinases and upregulation of expression of p38MAPK (a mitogen-activated protein kinase) were implicated as signaling events downstream to PKG activation.

For MOIH, the only result has come from a study of C. pagurus, in which an increase in cAMP was observed with hormonal treatment and with cAMP analogues that mimicked the action of MOIH, implying cAMP as a second messenger for MOIH (251).

cAMP, cGMP, and agents that increase cAMP levels individually were able to mimic the effect of ITP on ileal short-circuit current. Additionally, synthetic ITP elevated intracellular levels of both cyclic nucleotides in a dose-dependent manner (36, 252), again implying the involvement of these two cyclic nucleotides in ITP signal transduction. Importantly, receptors of ITP and ITP-L peptides have recently been identified and characterized in the silkworm B. mori (94). B. mori orphan neuropeptide G protein-coupled receptors (BNGRs), obtained through in silico mining of the silkworm genome, were screened using a Ca²⁺-imaging for activation by recombinant ITP and ITP-L peptides in HEK293T cells co-expressing a BNGR and a promiscuous mouse G_α15. Out of the 34 BNGRs tested, three Class-A BNGRs reacted positively to activation by ITPs, with BNGR-A2 and -A34 responding to ITP (EC₅₀: 1.1 x 10^-8 M and 1.3 x 10^-8 M, respectively) and BNGR-A24 to ITP-L (EC₅₀: 2.6 x 10^-8 M). Interaction between ITPs and BNGRs was evaluated by an in vitro assay that showed co-localization of the ligand and BNGR at the cell membrane of CHO cells heterologously expressing the receptor; the interaction occurred in a ligand-receptor-specific manner, consistent with the data for the distinct receptor responses to ligand activation observed using Ca²⁺ imaging. Stimulation of B. mori ovary-derived (BmN) cells by either ITP or ITP-L significantly increased the intracellular cGMP levels. Coupling of the BNGRs to the ITPs-induced cGMP signaling was demonstrated by observations that simultaneous knockdown of bngr-A2 and -A34 significantly decreased the cGMP response to ITP in BmN cells, whereas knockdown of bngr-A24 led to decreased cGMP responses to ITP-L; transient expression of bngr-A24 potentiated the response of BmN cells to ITP-L. The involvement of mGC and sGC in the signaling of ITPs was examined in BmN cells using, respectively, dsRNA targeting BmGyc76c (a B. mori mGC) or ODQ (a selective and potent inhibitor of NO-sensitive sGC), showing that both GC forms are involved in signaling via ITPs through cGMP production (94). A model for the signaling pathway of ITPs in BmN cells was proposed (94), which is in general similar to the MIH model proposed by Chang and Mykles (230), except that it includes the involvement of both mGC and sGC (Figure 5). Thus, activation of BNGR receptors by their respective ligands leads to activation of adenylyl cyclase via G-protein coupling and to activation of mGC. Like the MIH model, one of the events downstream to activation of protein kinase A activation is the increase in membrane calcium conductance, leading to Ca²⁺ influx, which in turn stimulates NO-sensitive sGC by activating Ca²⁺-calmodulin-dependent NO synthase. Increases in cGMP levels, due to the actions of both mGC and sGC, then activates protein kinase G, which together with protein kinase A, result in cellular responses (Figure 5).

Identification of receptor for insect ITPs not only represents an important breakthrough for insect studies but also provides an opportunity to identify the receptors for crustacean member peptides. Previous studies of molecular mass of the CHH- and MIH-binding proteins, respectively, in the hepatopancreas and Y-organ of C. sapidus and M. japonicus showed that the estimated values (51–70 kDa) were within the molecular mass range of typical GPCRs (194, 195). Given the concept that the peptide ligands co-evolve with their receptors, phylogenetic analysis of the crustacean and insect GPCRs has been performed searching for the crustacean orthologs that are phylogenetically clustered with the receptor for insect ITPs. Thus, analysis of transcriptome data derived from P. clarkii tissues has uncovered GPCRs, Procambarus GPCRs A52, A53 and A63, that clustered with BNGR-A34, and Procambarus GPCR A9 with BNGR-A24, with A52 and A63 being abundantly expressed in the hepatopancreas, implying these GPCRs could be the crayfish CHH, MIH, or CHH-L receptor (97). In addition, analysis of the data derived from the spiny lobster Sagmariasus verreauxi revealed 2 annotated GPCRs (Sv-GPCRA11 and 12) that are phylogenetically clustered with BNGR-A34 (96). In the blackback land crab G. lateralis Y organ transcriptome 3 GPCRs (Gl-GPCRA9, Gl-GPCRA10, Gl-GPCRA12) were similarly identified using phylogenetic analysis as potential CHH-like receptors, as these sequences clustered into the putative CHH receptors clade (95). Expression of one of the GPCRs, Gl-GPCRA12, decreased in late pre-molt and post-molt stages, suggesting that it may be the MIH receptor (95). These results are promising and provide receptor candidates for testing of ligand binding and receptor activation to truly establishing the status of these receptors for the crustacean peptides.

The signaling model based on experimental data mainly derived from the studies of G. lateralis and B. mori (94, 230) is attracting and probably applicable to the crustacean and insect members of the CHH superfamily (Figure 5). Several major coupling events however need experimental verification, including coupling of GPCR activation to adenylyl cyclase, GPCR activation of membrane GC (as suggested in the ITPs signaling), PKA phosphorylation-induced Ca²⁺ influx and subsequent NO production. In addition, for the model to be applicable to the crustacean CHH-superfamily peptides, including CHH, the model would have to be refined, based on data from additional studies, to accommodate the apparently contradictory data derived from other studies. For example, in response to hormonal stimulation, the change in cAMP levels was usually small and insignificant (192, 235, 237) and manipulation of cAMP levels or adenylyl cyclase activity did not mimic the effect of hormonal treatment (192). Further, it is interesting note that the lobster CHH stimulated the guanylyl cyclase activity in isolated membrane (233). Thus, GPCR activation of mGC would occur within the membrane without signaling through the cytoplasm. A GPCR-activated mGC event within the membranes, if proven, would probably represent a novel mode of GPCR signaling mechanism. Finally, it is now feasible to conduct experiments for ligand-receptor interaction and activation, with the availability of identified receptors and of information regarding ligand structure and structure-function relationship of several peptides, including MIH, CHH, CHH-L, ITP (87–93).