Gonadotropin-releasing hormone (GnRH), a member of the GnRH/AKH/CRZ superfamily, is an ancient hormone evolved prior to the divergence of protostomes and deuterostomes (1). In vertebrate species, three forms of GnRH have been identified, namely GnRH1, GnRH2 and GnRH3, and believed to be arose from whole genome duplication occurred in ancestral vertebrates (2). Given that multiple forms of GnRH structurally comparable to vertebrate GnRHs can be identified in tunicates, the possibility of GnRH gene duplication occurred during the evolution of protochordate cannot be excluded (3). The three isoforms of GnRH in vertebrate species are highly conserved in a.a. sequence, expressed in a species-specific manner in neurons within different brain areas, and proposed to have overlapping and yet distinct functions caused by subfunctionalization of the respective paralogs (4). In mammals, GnRH1 is the major form and GnRH3 was lost during evolution, whereas GnRH2 can be found in some species (e.g., human and monkey) but not the others (e.g., rodents). In tetrapods like the amphibians, reptiles and birds, GnRH3 was also lost with only GnRH1 and GnRH2 expressed. In fish, different combinations of the three isoforms can be noted, e.g., GnRH1 and GnRH2 in catfish and eel, GnRH2 and GnRH3 in salmon, goldfish and zebrafish, and three forms of GnRH in sea bass, medaka and tilapia (3, 4). At present, our understanding on the neuro-anatomical distribution and differential functions of different GnRH isoforms is mainly based on the studies in fish models (5, 6). In fish with three forms of GnRH (e.g., sea bass), GnRH1 neurons are located in the preoptic area within the brain with notable innervation in the pituitary while GnRH3 neurons are scattering along the terminal nerve, ventral telencephalon and preoptic area with a low level of pituitary innervation. GnRH2 neurons, however, are confined within the midbrain tegmentum with little/no fibres innervating the pituitary (7). Similar to fish models, localization of GnRH1 neurons in the preoptic area (& GnRH2 neurons in the midbrain area of some species) is well-conserved in mammals (8, 9).

In general, the GnRH1 neurons in the preoptico-hypothalamic tract represent the major source of GnRH for gonadotropin regulation at the pituitary level whereas GnRH2 and GnRH3 from other brain areas are commonly accepted to be neurotransmitters/neuromodulators with other functions related to reproduction (4, 9). For examples, GnRH2 from the midbrain area is involved in sexual behaviours, appetite control, steroidogenesis, and implantation/placental functions, and probably also coordinates the interaction between nutritional status and sexual activities (10). Meanwhile, GnRH3 originated from the terminal nerve ganglia has been proposed to play a role in relaying the sensory input from the olfactory bulbs (via olfactory tract) and visual system (from optic tectum) with reproductive functions (4). In medaka, GnRH1 neurons with pituitary innervation exhibit episodic bursting of action potentials related to the reproduction cycle, which is different from the regular pattern of action potentials in GnRH2 and 3 neurons with intrinsic pacemaker activity (11). In the same model, the frequency of action potentials in GnRH1 neurons can be elevated with LH surge (12) and GnRH1 knockout inhibits ovulation and spawning activity in female fish (13). Episodic firing of GnRH1 neurons in the preoptic area has also been reported in tilapia (14), which is consistent with the idea of GnRH pulsatility caused by KNDy neuron network within the hypothalamus leading to episodic release of luteinizing hormone (LH) at the pituitary level (15). In fish models, the functional role of GnRH1 has been replaced by GnRH3 in fish species with only GnRH2 and GnRH3 expression (e.g., the cyprinids and salmonids). In goldfish, a member of the cyprinid family, GnRH3 neurons in the preoptic area have fibres penetrating into the pituitary (16) and GnRH3 is the key mediator for seasonal changes in LH secretion (17) and spawning activity (18). For GnRH2, besides its expression in the midbrain area, it is also expressed in the forebrain (including the preoptic area) as well as in the pituitary (19, 20). In goldfish pituitary cells, interestingly, GnRH2 is more potent than GnRH3 in stimulating LH and follicle-stimulating hormone (FSH) expression (21). GnRH2 fibres (22) and protein signals (23) can also be detected in the pituitary of zebrafish but the source of GnRH2 is still unclear.

The biological actions of GnRH, especially for gonadotropin regulation, are mediated by GnRH receptor (GnRHR) functionally coupled with the PLC/IP₃/PKC, AC/cAMP/PKA, PLA₂/AA, PI3K/Akt, MAPK and Ca²⁺-dependent cascades (24, 25) and the post-receptor signalling involved, despite the species-specific variations, is highly conserved from fish to mammals (26, 27). GnRHR is a member of the rhodopsin-like G protein-coupled receptor (GPCR), and its origin can be traced back to the early members of invertebrates (1). In vertebrate species, two types of GnRHR, GnRHR-I and GnRHR-II, can be identified and believed to be the result of a tandem duplication of GnRHR gene in ancestral gnathostome occurred prior to the multiple rounds of genome duplication during vertebrate evolution (28). Subsequent local duplication of GnRHR-II gene followed by additional gene duplication of GnRHR-I and -II along with the whole genome duplication (with 3 rounds in fish and 2 rounds in tetrapods) probably have led to the multiple subtypes of GnRHR-I and -II found in modern-day vertebrates (2, 29). During the course of evolution, GnRHR-I is lost in bony fish and amphibians, while GnRHR-II is lost in most of the mammals (e.g., human and rodents) but retained in some primates (e.g., monkey) (2). Although GHRHR-I and -II share a notable level of sequence homology, the two receptors are structurally and functionally distinct from each other. By comparing with GnRHR-II, the 3^rd intracellular loop is relatively shorter and the C-terminal tail typical of GPCR is missing in GnRHR-I (24). In expression studies using GnRHR-II with deletion of the C-terminal tail (30) and GnRHR-I with addition of C-terminal tail from GnRHR-II (31), GnRHR-I, unlike GnRHR-II, was found to be not sensitive to receptor desensitization caused by GnRH and this unique property associated with the lack of C-terminal tail has been recently proven to be essential for normal fertility in the mouse model (32). In general, GnRHR-I is specific for GnRH1 (e.g., mammals) (33) while GnRHR-II (e.g., fish species) is either with a preference for GnRH2 (34) or being promiscuous to the three forms of GnRH (35). Using site-directed mutagenesis and domain swapping in human/monkey GnRHRs, the extracellular loop 2 and 3 together with the residues located close to membrane surface in TMD4 and 7 were found to be the structural determinants for ligand selectivity against different forms of GnRH (36, 37).

In fish models (e.g., tilapia), different subtypes of GnRHR-I and -II are expressed in pituitary cells other than gonadotrophs (38), and GnRH, besides its role as the gonadotropin-releasing factor, can also stimulate growth hormone (GH), prolactin (PRL) and somatolactin (SL) secretion at the pituitary level (27). The PRL response induced by GnRH is particularly interesting as PRL is involved in spawning migration (e.g., in salmon), nesting and brooding (e.g., in birds), parental care (e.g., mouth brooding in fish) and nutrient provision for the young (e.g., lactation in mammals and chick feeding in birds) (for reviews on reproductive functions of PRL, see (39, 40)). Apparently, PRL induction by GnRH may serve as a signal input from the hypothalamo-pituitary axis to trigger the reproductive functions of PRL. Although PRL regulation by GnRH has also been reported in mammals, the information is restricted to PRL release and the results obtained are not consistent, as stimulatory (41–43) and no effect have been documented (44, 45). For stimulatory action, GnRH-induced PRL secretion could be associated with the elevation of IP₃ (46) and cAMP levels (47, 48) or indirect action via local signals produced by gonadotrophs (49, 50). Except for the limited studies in tilapia for PRL release (51, 52) and in salmon for PRL mRNA expression caused by GnRH (53, 54), little is known for PRL regulation by GnRH in non-mammalian species. Despite a single report in tilapia confirming the involvement of PLC/IP₃- and Ca²⁺-dependent mechanisms in GnRH-induced PRL secretion in fish model (51), the post-receptor signalling for the effect on PRL gene expression is still unknown. Given that (i) the reproductive functions of PRL are expected to have beneficial effects on induced spawning in commercial fish with GnRH treatment, and (ii) the studies on signal transduction mediating the PRL responses by GnRH are still fragmentary and restricted to PRL release, an in vitro study was initiated in carp pituitary cells to examine the pituitary action of GnRH and the post-receptor signalling for GnRH-induced PRL release and gene expression. In our study, grass carp was used as the animal model as it is a representative of the cyprinids with high market value in Asian countries (~5.8 million tonnes/year and accounts for 12% of global finfish culture, data from 2022 FAO Report for World Fisheries and Aquaculture). Based on our promoter analysis of grass carp PRL gene, one CRE site (with core sequence “CGTCA”), one AP-1 site (with core sequence “TCAGTCA”) and three NFAT binding sites (with core sequence “GGAAA”) can be identified in the proximal promoter upstream of the TATA box ( Supplementary Figure S1 ), implying that PRL expression in carp pituitary is under the control of cAMP/PKA-, PLC/PKC- and/or Ca²⁺-dependent signalling. Using a pharmacological approach with parallel monitoring of second messengers and transcription factors linked with these signalling pathways, we have demonstrated for the first time that the cAMP/PKA, PLC/PKC and Ca²⁺/Calmodulin (CaM)-dependent cascades together with the crosstalk of cAMP/PKA pathway with CaM expression and Ca²⁺ entry via voltage-sensitive Ca²⁺ channels (VSCC) are differentially involved in PRL release and gene expression caused by GnRH via a specific subtype of GnRHR, namely GnRHR₄, expressed in lactotrophs in the carp pituitary.

In vertebrate species, the pivotal role of GnRH in reproduction, especially via LH and FSH regulation, is well-conserved from fish to mammals (for review, see (5, 6)). In mammals, GnRH-induced PRL release has been reported in vivo in ovariectomized monkey (42) and in human female with HMG treatment during assisted reproduction (41). The in vitro studies for PRL regulation via pituitary action of GnRH, however, are not consistent and vary with sex, reproductive stages and neonatal development. For examples, PRL secretion induced by GnRH was demonstrated in pituitary fragments/pituitary cells prepared from human (64), sheep (43) and neonatal rat (65) but not in other species (e.g., in porcine pituitary cells) (44, 45). Of note, the PRL-releasing effect of GnRH was observed in pituitary cells prepared from female sheep during the breeding season but not in other time of the year (43). In male rats, PRL secretion was not affected by GnRH treatment (66) but GnRH-induced PRL release could be noted in pituitary cells prepared from female rats during the neonatal period of day 3–9 after birth (65). For the post-receptor signalling mediating PRL regulation by GnRH, the information available was based on GGH3 cell lines with functional expression of rat GnRHR. In this case, PRL release could be triggered by GnRH with a parallel rise in IP₃ levels (46) and mediated by cAMP signals and protein synthesis coupled with G_s activation (47). Despite the fact that PRL also plays a key role in reproduction (see introduction), no consensus has been reached for the physiological role of PRL regulation by GnRH. For comparative studies on the same topic, only five reports have been published and they are all based on fish models, with two in tilapia, one in sockeye salmon, and the other two in masu salmon. For the studies in tilapia, in vitro culture of RPD/RPD cells prepared from male tilapia were used and the three native forms of GnRHs (with potency of GnRH2 > GnRH1 > GnRH3) were shown to induce PRL secretion (52) via PLC/IP₃- and Ca²⁺-dependent mechanisms (51). For the reports in salmon, the focus was on pituitary gene expression and PRL mRNA levels in the pituitary of masu salmon (67) but not sockeye salmon (54) could be elevated by long-term implantation with GnRH. Similar to the study in sheep model, GnRH could up-regulate PRL mRNA expression in pituitary cells prepared from female masu salmon during pre-spawning phase but not in other reproductive stages (53). To date, the studies on signal transduction for the PRL-releasing effect of GnRH are rather limited and no information is available for the post-receptor signalling mediating the pituitary action of GnRH on PRL gene expression.

In fish models, nerve fibres with GnRH immunoreactivity are located in close proximity to lactotrophs (e.g., in perch) (68) and GnRHR expression can be detected in lactotrophs in the anterior pituitary (e.g., in tilapia) (38). These findings provide the initial support for lactotrophs as the target cells for GnRH within the fish pituitary. In grass carp, two forms of GnRH, GnRH2 and GnRH3, and four types of GnRHR of the GnRHR-II family, namely GnRHR_1–4, have been cloned recently (35). In our study, GnRHR_1–4 expression could be located in the hypothalamo-pituitary axis of grass carp, which is consistent with the role of GnRH as a hypophysiotropic factor for gonadotropin regulation in carp species. Of note, the transcript expression of GnRHR_1–4 could be detected by RT-PCR in intact pituitary but the corresponding signal of GnRHR₃ was missing in parallel study with freshly dispersed pituitary cells. Since the washing steps with cell pelleting by low-speed centrifugation used in our pituitary cell preparation (56) are known to remove cell debris and nerve fibres/terminals, we do not exclude the possibility that GnRHR₃ is expressed in nerve fibres/terminals but not in endocrine cells within the carp pituitary. Using RT-PCR coupled to LCM capture of pituitary cells with PRL immunoreactivity, only the transcript signal for GnRHR₄ but not the other GnRHRs was found in carp lactotrophs. In carp pituitary cells prepared from whole pituitaries, PRL release, PRL mRNA level and PRL protein production could be elevated by sGnRH-A, a GnRH agonist commonly used in induced spawning in fish farms (62), and these stimulatory effects were also mimicked by parallel treatment with the native forms of GnRH, namely GnRH2 and GnRH3. These results suggest that GnRH can act directly at the pituitary level to stimulate PRL secretion, gene expression and protein translation, probably via GnRHR₄ activation in carp lactotrophs. Using functional expression of GnRHR in HEK293 cells, the four GnRHRs in grass carp were shown to have differential selectivity for the two native forms of GnRH, with GnRHR₁ exhibiting a strong preference for GnRH3, GnRHR₂ and GnRHR₃ with higher selectivity for GnRH2, and GnRHR₄ with similar preference for GnRH2 and GnRH3 (35). Our dose-dependence studies with GnRH2 and GnRH3 in carp pituitary cells revealed that the potency and efficacy for PRL release caused by the two forms of GnRH were quite comparable and the promiscuity of the PRL-releasing effect for the two GnRHs is consistent with the ligand selectivity of GnRHR₄. Interestingly, the corresponding effects on PRL mRNA expression were found to be more selective for GnRH3 than GnRH2 in terms of the maximal responses and ED₅₀ values. These results are intriguing and demonstrate for the first time that the ligand selectivity of GnRHR may vary with the biological readout in the same cell model. The higher sensitivity of GnRH3 for PRL gene expression also raises the possibility that GnRH3 may act as the major signal for PRL regulation by GnRH in carp lactotrophs, which corroborates with the role of GnRH3 as the “hypophysiotropic GnRH” for seasonal breeding in cyprinid species (5). Based on previous studies with domain swapping/deletion of C-terminal tail in GnRHRs from GnRHR-I and -II families (see introduction), it is commonly accepted that GnRHR-I without C-terminal tail is resistant to GnRH desensitization and the opposite is true for GnRHR-II with a C-terminal tail. In grass carp, GnRHR₄ expressed in lactotrophs is a member of GnRHR-II with a C-terminal tail typical of GPCR (35) and yet the PRL responses in terms of hormone secretion, transcript expression and protein production could still be noted after a 48-hr treatment with sGnRH-A, GnRH2 and GnRH3 at micromolar dose. These findings are at variance with the results expected for GnRHR-II (i.e., with refractoriness/even loss of response after prolonged induction) and suggest that GnRHR₄ expressed in carp lactotrophs is also resistant to GnRH desensitization similar to GnRHR-I.

In mammals, GnRHR can trigger post-receptor signalling via a single type/multiple types of G proteins in different cells/tissues in a context-dependent manner, e.g., with G_q/11 coupling in pituitary gonadotrophs, G_q/11 and G_s coupling in hypothalamic neurons and G_i coupling in cancer cells (e.g., prostate cancer) (69). Based on sequence mutations/domain swapping in rodent GnRHR expressed in COS-7/GGH3 cell lines, the 2^nd and 3^rd intracellular loops have been confirmed to be the structural domains in GnRHR for functional coupling with G_s and G_q/11 (70, 71) and activation of these G proteins upon ligand binding in GnRHR can lead to signal transduction via the PLC/IP₃/PKC, AC/cAMP/PKA, MAPK and Ca²⁺-dependent cascades (24, 25). For LH release induced by GnRH, the PLC/IP₃/PKC pathway with IP₃-mediated [Ca²⁺]i release and PKC coupling with MAPK forms a major core of the post-receptor signalling for GnRH at the pituitary level and the AC/cAMP/PKA pathway is not involved (27). In fish models, despite the similarity for the involvement of PLC/IP₃/PKC pathway and the associated Ca²⁺ responses (27, 63), discrepancy on the role of cAMP in GnRH signalling has been reported, e.g., the cAMP-dependent signalling is absent in the LH-releasing effect of GnRH in goldfish (26) and catfish pituitary cells (72) but cAMP/PKA involvement in gonadotropin regulation by GnRH has been documented in tilapia (63) and more recently in grass carp (35). Although PRL release induced by GnRH mediated by G_s and cAMP has been reported in GGH3 cells with stable transfection of rat GnRHR (47), whether the phenomenon can occur in primary culture of pituitary cells has yet to be examined. In our study with carp pituitary cells, AC activation by forskolin or increasing functional level of cAMP by 8Br.cAMP was shown to elevate PRL release and PRL mRNA levels and these stimulatory effects could be mimicked by the cAMP analog specific for PKA (6-Bnz-cAMP) but not Epac (pCPT-O-Me-cAMP), implying that activation of AC/cAMP/PKA pathway can trigger PRL secretion and gene expression in carp lactotrophs. This idea agrees with our findings that the PRL responses caused by forskolin could be blocked by AC inhibition with MDL 12330A or PKA inactivation by H89. For the role of cAMP-dependent pathway in PRL regulation by GnRH, GnRH treatment in carp RPD cells could trigger parallel rises in cAMP production and CREB phosphorylation followed by nuclear translocation of CREB, and these signalling events were mimicked by AC activation with forskolin. In parallel experiments with carp pituitary cells, the stimulatory effects of GnRH on PRL secretion and transcript expression were found to be sensitive to G_s blockade by melittin, AC inhibition by MDL 12330A, and PKA inactivation by H89. Consistent with the idea of PKA phosphorylation of CREB followed by its nuclear translocation and recruitment of the co-activator CBP to initiate target gene transcription (73), PRL gene expression induced by GnRH could be negated by inactivating CREB with KG-501 or interfering CREB: CBP interaction by SGC-CBP30. Our findings, as a whole, suggest that PRL release and gene expression caused by GnRH in carp lactotrophs are mediated by G_s activation coupled to AC/cAMP/PKA pathway. As a well-conserved CRE site can be located upstream of the TATA box within the 5’promoter of grass carp PRL gene, the PRL mRNA response caused by GnRH probably is the result of PRL gene transcription mediated by CREB/CBP complex acting at the promoter level. Given that (i) chromatin remodelling is involved in GnRH-induced gonadotropin gene expression (74), (ii) CBP can induce histone acetylation via its intrinsic HAT activity/by recruitment of PCAF (75), and (iii) PRL mRNA expression induced by GnRH in carp pituitary cells was also sensitive to the blockade of HAT activity with C646, we do not exclude the possibility that modulation of chromatin packaging via CREB/CBP complex coupled with the cAMP-dependent pathway may play a role in PRL gene transcription induced by GnRH in carp species.

In RPD culture prepared from tilapia pituitaries, PRL secretion could be induced by GnRH with parallel rises in IP₃ levels and [Ca²⁺]i signals and this PRL-releasing effect was ablated by PLC inactivation, VSCC blockade and inhibition of [Ca²⁺]i mobilization (51). Although the functional role of PKC on PRL secretion was not examined, the results suggest that the PLC/IP₃ pathway and associated Ca²⁺ responses are involved in PRL regulation by GnRH in fish model. In our study, PRL release and PRL mRNA levels could be up-regulated in carp pituitary cells by the DAG analog DiC8 or PKC activator TPA and these effects could be blocked by inactivating PKC with calphostin C, suggesting that PKC activation by DAG can induce PRL secretion and gene expression in carp lactotrophs. In carp pituitary cells, PRL release induced by GnRH was sensitive to G_q/11 inhibition with YM254890, PLC blockade by U73122, and PKC repression using calphostin C. Although G_q/11 inhibition could suppress PRL transcript expression caused by GnRH, similar blockade of PLC and PKC was found to have no effect in this regard. Furthermore, IP₃ antagonism with xestospongin C or blocking IP₃ channel for [Ca²⁺]i mobilization with 2-APB did not alter GnRH-induced PRL release and gene expression. These results suggest that G_q/11 activation induced by GnRH can trigger PRL release in carp lactotrophs via PLC/PKC pathway and the IP₃ component working downstream of PLC is not involved. Apparently, PRL gene expression caused by GnRH is not mediated by the PLC/IP₃/PKC pathway, which is consistent with our findings that the PKC-sensitive transcription factors, namely c-Fos and c-Jun, did not show noticeable changes in carp RPD cells with GnRH treatment. In our study, PKC activation by TPA or DiC8 could elevate PRL mRNA levels but similar response on PRL gene expression by GnRH did not involve PKC signalling and yet PKC involvement has been clearly implicated for the PRL-releasing effect of GnRH. In living cells, multiple forms of PKC are expressed and TPA and DAG are known to activate the conventional (PKC_{α, β & γ}) and novel PKCs (PKC_{δ, ε, η & θ}) but not atypical PKCs (PKC_{ζ & ι)} (76). It is tempting to speculate that a subtype of atypical PKCs may be involved in PRL release but not gene expression induced by GnRH. Of note, G_q/11 inhibition could block PRL transcript expression caused by GnRH and yet the PLC/IP₃/PKC pathway coupled to G_q/11 was not involved, functional mediation by other signalling events working downstream of G_q/11 (e.g., MAPK) for PRL gene transcription cannot be excluded.

Although [Ca²⁺]i mobilization via IP₃ is well-documented for the LH-releasing effect of GnRH, voltage-gated [Ca²⁺]e influx caused by functional interplay of a combination of ion channels also contributes to the LH response via the action potentials and Ca²⁺ oscillations induced by GnRH at the gonadotroph level (77). The involvement of [Ca²⁺]i mobilization and voltage-gated [Ca²⁺]e entry is well-conserved in GnRH actions in fish species, e.g., for GnRH-induced LH and GH release in goldfish pituitary cells (26, 27). In our study, VSCC activation by Bay K8644 was effective in triggering a rapid rise in cytosolic Ca²⁺ in carp RPD cells and this effect was blocked by inhibiting L-type VSCC with nicardipine. Bay K8644 also increased PRL release and PRL mRNA level in carp pituitary cells and these stimulatory actions could be nullified by L-type VSCC inactivation using nicardipine or CaM antagonism by calmidazolium. For the signalling events working downstream of CaM, blocking CaMK-II activity by KN62 was found to ablate the PRL-releasing effect of Bay K8644 but not the corresponding response on PRL mRNA, whereas the opposite was true for parallel inhibition of CaN using cytosporin A. These results imply that [Ca²⁺]e entry via L-type VSCC can trigger PRL responses via Ca²⁺/CaM-dependent signalling, with PRL release caused by CaMK-II activation and PRL gene expression mediated by CaN-dependent mechanisms. In carp RPD cells, GnRH was effective in inducing Ca²⁺ signals, which could be negated by nicardipine inhibition of L-type VSCC. Parallel kinetic chase of the spatiotemporal distribution of the Ca²⁺ signals in immuno-identified lactotrophs also revealed that the Ca²⁺ rise caused by GnRH was initiated close to the plasma membrane and spread into the central region of responsive cells, which is consistent with [Ca²⁺]e entry via VSCC at the membrane level. During the process, the highest level of Ca²⁺ signals was located in the subplasmalemmal space under the plasma membrane, which is well-documented to be the region with the readily releasable pool of secretory vesicles with proper priming/docking for Ca²⁺-sensitive exocytosis in endocrine cells (78). In carp pituitary cells, PRL secretion and gene expression caused by GnRH were also sensitive to VSCC blockade with nicardipine and CaM antagonism with calmidazolium. For the downstream events coupled to Ca²⁺/CaM signalling, the PRL-releasing effect of GnRH but not the corresponding action on PRL transcript could be obliterated by inactivating CaMK-II with KN62. However, the elevation in PRL mRNA levels but not PRL release caused by GnRH was reduced/nullified by blocking CaN with cyclosporin A, interfering CaN:NFAT interaction by INCA-6, or inhibiting nuclear translocation of NFAT using 11R-VIVIT. These results are also consistent with our finding that GnRH treatment was effective in reducing NFAT₂ phosphorylation in carp RPD cells. Since (i) CaMK-II activation by Ca²⁺/CaM complex can phosphorylate exocytosis regulators (e.g., syntaxin and synapsin) for priming/docking of secretory vesicles at membrane level prior to their exocytosis (79), (ii) CaN is a serine/threonine phosphatase and its enzyme activity can be up-regulated by Ca²⁺/CaM signalling to trigger gene transcription via NFAT dephosphorylation (80), and (iii) three NFAT binding sites with the proximal one overlapping with the CRE site upstream of the TATA box can be located in the 5’ promoter of grass carp PRL gene, our findings may imply that Ca²⁺/CaM signalling caused by [Ca²⁺]e entry via L-type VSCC is involved in PRL regulation by GnRH. Apparently, the PRL-releasing effect is mediated by Ca²⁺/CaM activation of CaMK-II and the corresponding response for PRL transcript is caused by PRL gene transcription via parallel activation of CaN/NFAT₂. In grass carp, differential involvement of CaMK-II and CaN/NFAT in negative feedback of GH by IGF-I has also been reported (58). In this case, IGF-I could inhibit GH gene transcription in carp pituitary cells by elevating CaM expression with parallel rises in CaN/NFAT but not CaMK-II signalling. In carp RPD cells, the Ca²⁺ responses for GnRH could be totally negated by nicardipine and this observation suggest that the Ca²⁺ signals in lactotrophs caused by GnRHR activation are mainly of extracellular origin. Together with our findings that IP₃ antagonism/IP₃ channel blockade did not alter the PRL responses by GnRH, it would be logical to assume that IP₃-induced [Ca²⁺]i mobilization is not a major component in the signal transduction for PRL regulation by GnRH. This idea, however, is at variance with the reports on LH release induced by GnRH, in which the PLC/IP₃ pathway coupled to [Ca²⁺]i mobilization forms a major core of the post-receptor signalling in gonadotrophs (25, 27).

In grass carp, the proximal NFAT binding site was found to overlap with the CRE site identified in the 5’promoter of PRL gene, which raises the possibility that a functional crosstalk between cAMP- and Ca²⁺-dependent pathways may be involved in PRL regulation in carp species. In fish models, functional coupling of cAMP/PKA pathway with Ca²⁺-dependent signalling has been reported for pituitary hormone secretion/gene expression. For examples, cAMP signalling coupled with Ca²⁺ entry via VSCC was shown to mediate GH secretion induced by dopamine via D1 receptor activation in goldfish pituitary cells (81, 82). In carp pituitary cells, PACAP via PAC1 receptor activation can act as a potent stimulator for GH (57), PRL (55) and SLα gene expression (83), and these stimulatory effects also involve cAMP/PKA-mediated Ca²⁺ entry via VSCC and Ca²⁺/CaM activation of CaMK-II. In our previous study in grass carp, CaM gene was cloned, and its promoter activity was shown to be under the control of cAMP signalling (84). Furthermore, CaM expression in carp pituitary cells could be up-regulated by PACAP in a cAMP/PKA-dependent manner (55) and contributes to PACAP-induced PRL promoter activition (85) and transcript expression (55). Apparently, the crosstalk of cAMP pathway with Ca²⁺-dependent signalling can occur at two levels in the carp pituitary, with (i) VSCC activation to allow [Ca²⁺]e entry to jack up cytosolic Ca²⁺ and (ii) increasing CaM expression to enhance the sensitivity to Ca²⁺ signals. In our study with carp RPD cells, a rise in Ca²⁺ signal was noted with AC activation by forskolin and this Ca²⁺ response could be blocked by inhibiting L-type VSCC with nicardipine or PKA inactivation using H89, suggesting that L-type VSCC can be activated by cAMP signal in lactotrophs via PKA activation. This idea also corroborates with the reports that the ion channel activity of VSCC can be altered by protein phosphorylation (86), e.g., enhancement of inward Ca²⁺ current/channel opening probability is well-documented for PKA phosphorylation of Ca_V1.4 (87) and Ca_V1.2a,b subtypes of L-type VSCC (88). In carp pituitary cells, VSCC inhibition with nicardipine and CaM antagonism with calmidazolium were both effective in blocking PRL release and PRL gene expression caused by forskolin. Furthermore, the PRL-releasing effect of forskolin but not the parallel stimulation on PRL mRNA could be ablated by CaMK-II inhibition with KN62 while the opposite was true with CaN inactivation by cyclosporin A. Similar to GnRH, forskolin also induced NFAT₂ dephosphorylation in carp RPD cells. In the same cell model, CaM expression could be up-regulated by GnRH and the effect was mimicked by AC activation with forskolin, increasing cAMP level by 8Br.cAMP, and PKA activation using 6-Bnz-cAMP. Consistent with CaM expression at protein level, GnRH treatment could also increase CaM mRNA level in RPD cells, and the effect was sensitive to G_s inhibition using melittin, AC repression by MDL 12330A, PKA blockade with H89, CREB inactivation by KG-501, and blocking CREB:CBP interaction by SGC-CBP30. These findings, as a whole, imply that (i) the Ca²⁺/CaM-dependent signalling with involvement of CaMK-II in PRL secretion and CaN/NFAT in PRL gene expression are working downstream of the cAMP/PKA pathway coupled with G_s signalling, and (ii) the crosstalk between cAMP- and Ca²⁺-dependent cascades induced by GnRH in carp lactotrophs probably is mediated by [Ca²⁺]e entry via PKA activation of L-type VSCC and CaM expression coupled with AC/cAMP/PKA signalling via CREB/CBP-dependent gene transcription.

In summary, using grass carp pituitary cells as a model, we have demonstrated for the first time that the cAMP/PKA-, PLC/PKC- and Ca²⁺/CaM-dependent signalling are differentially involved in PRL secretion and gene expression induced by GnRH at the pituitary level. Based on the results obtained, a working model has been proposed for the signal transduction mediating PRL regulation by GnRH in carp lactotrophs ( Figure 13 ). In this model, GnRHR₄ activation in lactotrophs by GnRH2/3 or their agonist sGnRH-A can initiate post-receptor signalling via G_s and G_q/11 coupling. The AC/cAMP/PKA pathway and [Ca²⁺]e entry via L-type VSCC with subsequent Ca²⁺/CaM-activation of CaMK-II and CaN are coupled to G_s activation while the PLC/IP₃/PKC pathway with downstream PKC-dependent signalling and [Ca²⁺]i mobilization by IP₃ are linked with G_q/11 activation. Interestingly, different components of these signalling events are differentially involved in PRL secretion and gene expression by GnRH, with (i) the AC/cAMP/PKA, Ca²⁺/CaM/CaMK-II and PLC/DAG/PKC signalling responsible for the PRL-releasing effect, and (ii) PKA-induced CREB phosphorylation for nuclear entry and CBP recruitment and Ca²⁺/CaM-mediated NFAT₂ dephosphorylation by CaN to trigger PRL gene transcription with subsequent rise in PRL mRNA and protein synthesis. In carp lactotrophs, the Ca²⁺-dependent signalling via CaMK-II (for PRL release) and CaN activation (for PRL gene expression) is working downstream of the AC/cAMP/PKA pathway probably through (i) VSCC activation by PKA to induce [Ca²⁺]e entry at membrane level and (ii) CaM up-regulation via transcriptional activation by CREB/CBP coupled to cAMP/PKA signalling. Although the other aspects of post-receptor signalling (e.g., MAPK) have not been touched in PRL regulation in our cell model, our study has provided novel information on the complexity of signal transduction involved in PRL release and gene expression by GnRH in a carp species. Given that our study focused on the pituitary actions of GnRH using primary culture of pituitary cells, further investigations using an in vivo approach (e.g., for GnRH interactions with other PRL regulators of central origin/from the periphery or GnRHR knockout in lactotrophs using CRISPR/Cas9) are clearly warranted to establish a more holistic picture for the physiological role of GnRH in PRL regulation. It is also worth mentioning that the pituitary cells used in our study were prepared from pre-pubertal fish, it will be of interest to know if the post-receptor signalling for PRL regulation by GnRH can be modified/“fine-tuned” with sexual maturation/with different stages of the seasonal reproductive cycle in adult fish as in the case of gonadotropin secretion/gene expression.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The animal study was approved by The Committee on the Use of Live Animal for Teaching and Research, the University of Hong Kong, Hong Kong. The study was conducted in accordance with the local legislation and institutional requirements.

WL: Conceptualization, Data curation, Investigation, Methodology, Writing – review & editing, Formal analysis. CY: Data curation, Formal analysis, Investigation, Methodology, Software, Writing – review & editing. MH: Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing. WK: Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – review & editing. CC: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – original draft. YC: Project administration, Supervision, Visualization, Writing – review & editing. AW: Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing.