PfIRR Interacts with HrIGF-I and Activates the MAP-kinase and PI3-kinase Signaling Pathways to Regulate Glycogen Metabolism in Pinctada fucata

The insulin-induced mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3K) pathways are major intracellular signaling modules and conserved among eukaryotes that are known to regulate diverse cellular processes. However, they have not been investigated in the mollusk species Pinctada fucata. Here, we demonstrate that insulin-related peptide receptor of P. fucata (pfIRR) interacts with human recombinant insulin-like growth factor I (hrIGF-I), and stimulates the MAPK and PI3K signaling pathways in P. fucata oocytes. We also show that inhibition of pfIRR by the inhibitor PQ401 significantly attenuates the basal and hrIGF-I-induced phosphorylation of MAPK and PI3K/Akt at amino acid residues threonine 308 and serine 473. Furthermore, our experiments show that there is cross-talk between the MAPK and PI3K/Akt pathways, in which MAPK kinase positively regulates the PI3K pathway, and PI3K positively regulates the MAPK cascade. Intramuscular injection of hrIGF-I stimulates the PI3K and MAPK pathways to increase the expression of pfirr, protein phosphatase 1, glucokinase, and the phosphorylation of glycogen synthase, decreases the mRNA expression of glycogen synthase kinase-3 beta, decreases glucose levels in hemocytes, and increases glycogen levels in digestive glands. These results suggest that the MAPK and PI3K pathways in P. fucata transmit the hrIGF-I signal to regulate glycogen metabolism.


Results
Characterization of an anti-IRR polyclonal antibody. After double-digestion with BamHI and XhoI, the 837-bp cDNA fragment encoding the TK domain of IRR was amplified and cloned into the BamHI/XhoI sites of the pET28a expression vector, which results in the fusion of a histidine tag to the TK domain. Then, the TK domain was expressed in Escherichia coli and purified (Fig. 1a). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis revealed that the molecular weight of the TK domain was approximately 34 kDa (Fig. 1b). Following induction with isopropyl β -D-1-thiogalactopyranoside (IPTG), the TK domain was detected almost exclusively in inclusion bodies. (Fig. 1b). The denatured TK domain had better affinity for a Ni 2+ -NTA column, and a higher concentration of the TK domain was eluted with a 50-200 mM imidazole gradient (Fig. 1c). Finally, 1.3 mg/ml of purified TK domain was obtained (Fig. 1d).
Polyclonal antibodies were generated in rabbits using the purified TK domain. The antisera were purified using a protein A affinity column. Antigen titers were detected by an enzyme-linked immunosorbent assay (ELISA) with 1:1,250, 2,500, 5,000, 10,000, 20,000, 40,000, and 80,000 dilutions of the antisera ( Table 1). The concentration of the polyclonal antibody serum was 6.76 mg/ml. Western blot analyses were conducted with the purified TK domain. A single band with a molecular weight of approximately 34 kDa was detected (Fig. 1e). These results demonstrate that the antibody specifically recognizes the TK domain of pfIRR.
hrIGF-I interacts with pfIRR in P. fucata oocytes. To verify the interaction between hrIGF-I and pfIRR, an in vitro co-immunoprecipitation (Co-IP) analysis using P. fucata oocytes was employed. The results showed that pfIRR was co-immunoprecipitated by hrIGF-I (Fig. 2a), while hrIGF-I was co-immunoprecipitated by pfIRR (Fig. 2b). The interaction between hrIGF-I and pfIRR was dose-dependent, and it increased with increasing hrIGF-I concentrations (10 −11 -10 −6 M) when using anti-hrIGF-I antibodies for the Co-IP (Fig. 2a). A slight dose-dependent increase was observed for hrIGF-I concentrations ranging from 10 −9 to 10 −6 M, and there was a dose-dependent increase when using anti-pfIRR antibodies (from 10 −11 to 10 −9 M) for the Co-IP (Fig. 2b).

IGF stimulation activates the MAPK and PI3K signaling pathways.
To examine the effect of hrIGF-I on the activation of the MAPK and PI3K signaling cascades, P. fucata oocytes were treated with hrIGF-I (10 −7 M) for various periods of time and with different concentrations of hrIGF-I for same time. The phosphorylation of p44/42 MAPK was examined by determining the ratio of phospho-to total p44/42 MAPK. As shown in Fig. 3a1, hrIGF-I-induced MAPK phosphorylation within 5 min, and the induction lasted for 6 h. The activation Scientific RepoRts | 6:22063 | DOI: 10.1038/srep22063 increased in a time-dependent manner from 0 to 180 min (Fig. 3a2). When treated for 30 min, hrIGF-I induced a dose-dependent phosphorylation of p44/42 MAPK at concentrations ranging from 10 −11 to 10 −7 M (Fig. 3b1,b2).
The results of the Akt/PKB phosphorylation indicate that hrIGF-I-induced the phosphorylation of T308 and S473 of Akt/PKB within 0 min, and the induction lasted for at least 6 h (Fig. 4a1). Phosphorylation increased in a time-dependent manner from 0 to 180 min (Fig. 4a2,a3). The maximal activation of S473 phosphorylation occurred 180 min after hrIGF-I treatment.
This hrIGF-I-induced increase in T308 and S473 phosphorylation was dose-dependent at hrIGF-I concentrations from 10 −11 to 10 −9 M (Fig. 4b1). There was no significant difference in T308 phosphorylation at hrIGF-I from 10 −8 to 10 −6 M (Fig. 4b2). The maximal activation of S473 phosphorylation occurred at an hrIGF-I concentration of 10 −8 M, and it decreased dose-dependently from 10 −8 to 10 −6 M hrIGF-I (Fig. 4b3). pfIRR is required for the IGF-stimulated activation of the MAPK and PI3K signaling pathways. To assess the role of pfIRR in the hrIGF-I-mediated MAPK and PI3K signaling pathways, we measured the effects of a pfIRR inhibitor, PQ401, on hrIGF-I-mediated Akt and MAPK phosphorylation. Basal and hrIGF-I-induced phosphorylation of MAPK were lowered by a 1-h treatment with 20 μ M PQ401 and by a 0.5-h treatment with 40 μ M PQ401. The inhibition of MAPK phosphorylation increased in a dose-dependent manner in hrIGF-treated and untreated cells following PQ401 treatment (   The basal and hrIGF-I-stimulated phosphorylation of T308 and S473 of Akt/PKB were similarly suppressed by pretreatment with PQ401. Pre-incubation with 40 μ M PQ401 for 0.5 h completely prevented basal and hrIGF-I-stimulated S473 phosphorylation, whereas T308 phosphorylation was completely prevented after 2 h (Fig. 5, right panel).

MEK is involved in hrIGF-I-mediated PI3K pathway activation.
To determine whether MEK is involved in hrIGF-I-mediated PI3K pathway activation, we assessed the effects of the MEK inhibitor PD98059 on the hrIGF-I-induced phosphorylation of T308 and S473 of Akt/PKB. Basal MAPK phosphorylation was lowered by treatment with 50 μ M PD98059, and it was completely inhibited by treatment with 60 and 80 μ M PD98059 for 1 h. hrIGF-I-stimulated MAPK phosphorylation was reduced by PD98059 at the three concentrations tested (50, 60, and 80 μ M), and a stronger reduction was achieved by treatment with 50 μ M PD98059. Pretreatment  with PD98059 at 50, 60, and 80 μ M concentrations for 1 h completely inhibited the basal and hrIGF-I-stimulated phosphorylation of T308 and S473 of Akt/PKB (Fig. 6a).
To detect the duration of the PD98059-meidated inhibition of PI3K activity, we analyzed the phosphorylation of T308 and S473 after treatment with 60 μ M PD98059 for 0.5, 1.5, and 2 h. Basal MAPK phosphorylation was completely prevented by treatment with PD98059 for 0.5 and 1.5 h, but not for 2 h. hrIGF-I-stimulated MAPK phosphorylation exhibited the greatest decrease after 1.5 h of PD98059 treatment. Pretreatment with 60 μ M PD98059 for 0.5 to 2 h completely inhibited the basal and hrIGF-I-stimulated phosphorylation of T308 and S473 of Akt/PKB (Fig. 6b).

PI3K is involved in the hrIGF-I-mediated stimulation of the ERK pathway.
To determine whether PI3K is involved in hrIGF-I-mediated ERK pathway activation, we assessed the effect of the PI3K inhibitor   wortmannin on hrIGF-I-induced MAPK phosphorylation. Basal and hrIGF-I-stimulated phosphorylation of T308 and S473 were completely prevented by treatment with 20, 40, and 60 μ M wortmannin for 1 h. Pretreatment with wortmannin lowered basal and hrIGF-I-stimulated MAPK phosphorylation, and a slight dose-dependent inhibition was observed (Fig. 6a).
To detect the duration of the wortmannin-mediated inhibition of ERK activity, we analyzed MAPK phosphorylation after treatment with 40 μ M wortmannin for 0.5, 1, and 2 h. Basal and hrIGF-I-stimulated phosphorylation of T308 and S473 were completely prevented by treatment with wortmannin for 1.5 and 2 h. Pretreatment with 40 μ M wortmannin for 1.5 h, but not for 0.5 h, resulted in the greatest decreases in basal and hrIGF-I-stimulated MAPK phosphorylation. HrIGF-I-stimulated, but not basal, MAPK activity was lowered by wortmannin treatment for 2 h (Fig. 6b).
Effect of hrIGF-I on carbohydrate/glycogen metabolism via the PI3K and/or MAPK pathways. To investigate the participation and conservation of the PI3K and/or MAPK pathways in carbohydrate/glycogen metabolism in response to hrIGF-I induction, we analyzed the effects of hrIGF-I on the glycogen content and glucose levels, the phosphorylation of Akt/PKB at T308 and p44/42 MAPK, and the expression of the involved genes. Three hours after the injection of hrIGF-I or a phosphate-buffered saline (PBS) control, the glycogen content in the digestive glands (DGs) and adductor muscles (AMs) significantly increased (Fig. 7a). The glucose levels in the hemocytes of hrIGF-I-treated oysters decreased significantly compared with those of the PBS-injected control 3 h post-injection (Fig. 7b). hrIGF-I-stimulated the phosphorylation of Akt/PKB at T308, and p44/42 MAPK phosphorylation was increased compared with basal phosphorylation levels in vivo (Fig. 7c).
We analyzed four downstream genes of the hrIGF-I signaling pathway: the pfirr, glucokinase (GK), glycogen synthase kinase-3β (GSK-3β ), and PP1 genes. The relative expression levels of the pfIRR, GK, and PP1 genes significantly increased in DGs after hrIGF-I injection, whereas there were no differences in expression levels for hrIGF-I-treated or non-treated AMs (Fig. 7d). However, GSK-3β expression decreased significantly in AMs after the injection of hrIGF-I, but it did not differ in the DGs.
The protein expression of pfIRR, GK, PP1, phospho-GS, GS, phospho-GSK-3β , and GSK-3β was measured by Western blotting. The expression of pfIRR and PP1 increased in DGs after hrIGF-I injection, whereas no difference was observed for hrIGF-I-treated or non-treated AMs (Fig. 7e), which was similar to the expression pattern of the mRNA levels. However, pfIRR expression was higher in DGs than in AMs, which differed from the pattern of mRNA expression (Fig. 7e). GK expression in the AMs was not significantly different after hrIGF-I injection, which was similar to the expression pattern of its mRNA (Fig. 7e). However, GK was not detected in the DGs, although its mRNA was expressed. GS was expressed in DGs and AMs, but phospho-GS was only detected at low levels in DGs and AMs after several hrIGF-I injections (Fig. 7e).

Discussion
In this study, we report the generation and evaluation of a polyclonal antibody against pfIRR by expressing a fragment encoding the TK domain of pfIRR in E. coli. High avidity antisera were obtained, which were functional in ELISA and western blotting assays. The crude antisera that were purified by a protein A affinity column showed that the antibody is highly specific for the TK domain of pfIRR.
Via a Co-IP assay, hrIGF-I and pfIRR were shown to interact with each other. The interaction between hrIGF-I and pfIRR increased in a dose-dependent manner with the hrIGF-I concentration (10 −11 -10 −6 M) when using the anti-hrIGF-I antibody for Co-IP (Fig. 2a); a slight dose-dependent increase was observed at hrIGF-I concentrations ranging from 10 −9 to 10 −6 M, and there was a dose-dependent increase at hrIGF-I concentrations from 10 −11 to 10 −9 M when using the anti-pfIRR antibody for Co-IP (Fig. 2b). In this study, we did not obtain the insulin protein of P. fucata, and in previous studies, hrIGF-I, which is structurally and functionally conserved, has been reported to be used widely in invertebrate species. Our study showed that the pfIRR gene is expressed in ovaries 39 , and a previous study showed that hrIGF-1 effects germinal cell proliferation and maturation associated with the expression of a homologous IRR in Crassostrea gigas 27 . Moreover, because no oyster cell lines are available presently, we used ovary cells to conduct our experiments.
Our experiments indicated that the MAPK and PI3K pathways are activated following hrIGF-I stimulation in P. fucata oocytes, although their threshold concentrations and durations differed. The effect of hrIGF-I on MAPK activation was sustained for at least 6 h, and the maximal activation required a relatively high concentration of hrIGF-I (10 −7 M). The effect of hrIGF-I on Akt activation was also sustained for at least 6 h. The maximal activation of T308 phosphorylation was obtained at 10 −9 M hrIGF-I, while the maximal activation of S473 phosphorylation occurred at 10 −8 M hrIGF-I. Although the maximal activation level of phospho-T308-Akt was achieved at 180 min, the signal was very strong compared with that in the other lanes; thus, it may be an outlier.
Full Akt activation requires the dual phosphorylation of S473 by mammalian target of rapamycin companion 2 (mTORC2)-rapamycin-insensitive companion of mTOR complex (RICTOR), integrin-linked kinase 1 (ILK)-1, or DNA-dependent protein kinase (DNAPK) as well as the phosphorylation of T308 by PDK1 40 . PDK1 is known to function downstream of PI3K. PI3K activation results in the production of phosphatidylinositol 3,4,5 trisphosphate (PIP3), which binds to the pleckstrin homology (PH) domain of PDK1 and thereby recruits PDK1 to the plasma membrane 41 . Thus, we examined T308 and S473 phosphorylation of Akt, and hrIGF-I resulted in a stronger activation of S473 than T308. As is known, the recruitment of Akt to the plasma membrane is required for the phosphorylation of S473, and PI3K activity is thought to be necessary for S473 phosphorylation 42,43 .
In previous studies, a diaryl urea compound, PQ401, was found to antagonize IGF-IR autophosphorylation in cultured human MCF-7 cells, with a half-maximal inhibitory concentration of 12 μ M, as well as the growth of cultured breast cancer cells in serum at 10 μ M 44 . Because the autophosphorylation regions of pfIRR and IGF-IR are highly conserved, we used PQ401 to inhibit pfIRR. Our previous studies have indicated that PQ401 decreases pfIRR mRNA expression and leads to the developmental arrest of P. fucata embryos 39   inhibition of pfIRR by PQ401 decreased the activation of MAPK and Akt, which indicated that pfIRR is required for the basal and IGF-mediated activation of the MAPK and PI3K signaling pathways. The results suggest that treatment with 40 μ M PQ401 for 1.5 h may be the most efficient conditions for inhibiting MAPK activation, and treatment with 40 μ M PQ401 for 2 h may be the most efficient conditions for inhibiting Akt activation among the different doses of PQ401 and different treatment durations used in our experiments.
We used the MEK inhibitor PD98059 and the PI3K inhibitor wortmannin to examine the role of MAPK and PI3K in regulating the hrIGF-I signaling pathway. Treatments with PD98059 inhibited basal and hrIGF-I-mediated MAPK and Akt activities, indicating that hrIGF-I activates the MAPK signaling pathway, and that MEK is involved in hrIGF-I-mediated PI3K pathway activation. In mammals, active ERK can influence the PI3K/Akt pathway via phosphorylation of GAB1 and IRS on several serine residues that are adjacent to p85 PI3K-binding sites (three YXXM motifs) 45 . The GAB1-p85 PI3K complexes alleviate the intrinsic inhibition of PI3K, which further increases PIP3 production. PIP3 in the plasma membrane leads to the membrane recruitment of GAB1 and IRS through their PH domains, thereby generating a positive feedback loop 46,47 . In Ras/ MAPK → PI3K/Akt cross-talk, in some cells, ERK inhibition by U0126 resulted in a decrease in Akt activation after hepatocyte growth factor (HGF) stimulation 48 . However, this positive regulatory effect of Ras/MAPK on PI3K depends on the type of stimulation, as it has the opposite effect on epidermal growth factor (EGF) signaling. The Ras/MAPK cascade negatively regulates PI3K after EGF stimulation 49,50 . In P. fucata, the results of hrIGF-I signaling show that PI3K activity is decreased following MAPK inhibition, as also occurs during HGF signaling.
PD98059 inhibited MAPK activation at 50-80 μ M concentrations. T The reduced but not inhibited MAPK activation by hrIGF-I stimulated demonstrated that MAPK activation is not regulated by only one signaling pathway, and it indicates that Ras/MAPK → PI3K/Akt cross-talk that can amplify MAPK signaling occurs upstream of MEK. The maximal inhibition time for PD98059 in our study was 2 h. The positive influence of the Ras/MAPK pathway on the PI3K/Akt pathway was not affected by the inhibitor concentration or the stimulation time, which may be because the signal strength (hrIGF-I dose) and the stimulation time (0.5 h) in these inhibition experiments were the same. The basal activities of MAPK and Akt were inhibited by different doses of PD98059 and different treatment times, whereas after treatment, hrIGF-I stimulated the phosphorylation of MAPK, but not Akt, at either T308 or S473. This may be because MAPK is more sensitive than Akt to the dose of hrIGF-I used (10 −9 M).
Treatments with the PI3K inhibitor wortmannin reduced basal and hrIGF-I-mediated Akt and MAPK activities, indicating that hrIGF-I activates the MAPK signaling pathway, and that PI3K is involved in the hrIGF-I-mediated induction of the MAPK pathway. Inhibition of Akt activity by wortmannin indicates that it functions by inhibiting the PI3K-AKT pathway in P. fucata. Inhibition of PI3K by wortmannin and LY294002 has been reported to block the activation of ERK1/2 in the rat skeletal muscle cell line L6 51 , Chinese hamster ovary cells 52 , and rat adipocytes 53 , but not in other cell types such as rabbit parietal cells 54,55 . Previous studies have shown that PI3K positively regulates the Ras/MAPK cascade in most cellular systems, which was also shown in our study. In mammals, PI3K affects MAPK signaling primarily at the level of PIP3-mediated Ras guanine nucleotide exchange factor and Ras GTPase activating protein (GAP) signaling. The positive PI3K-PIP3-GAB-PI3K feedback loop leads to the phosphorylation of GAB1 and IRS, which can recruit Grb2 and Son of Sevenless, leading to Ras activation [56][57][58] or bind RasGAP, which catalyzes Ras deactivation 59 . GAB1 association with SHP2 increases SHP2 phosphatase activity by dephosphorylating SHP2 substrates, such as RasGAP and c-Src. The resulting activation of c-Src and Ras leads to an up-regulation of ERK signaling 60,61 .
Wortmannin inhibited Akt activation at 20-60 μ M concentrations. A previous study demonstrated that Ras/ MAPK → PI3K/Akt cross-talk is dependent on the concentrations of growth factors. For example, at saturating EGF doses, PI3K inhibition by wortmannin only slightly attenuates ERK phosphorylation in A431 human epidermoid carcinoma and MCF7 human mammary carcinoma cells (over the 2-60 min response period), whereas PI3K inhibition dramatically decreases ERK phosphorylation at low physiological EGF doses 13 . Wortmannin decreased MAPK phosphorylation in the middle and late time points of signal propagation, but the peak of activation (0.5 h) did not differ from that of control cells. This decreasing sensitivity of MAPK phosphorylation to PI3K inhibition was also observed in human embryonic kidney 293, HeLa, T47D, and BT-474 cells 13 . A 10 −9 M concentration of hrIGF-I dramatically decreased MAPK phosphorylation with increasing wortmannin concentration, but it had a smaller effect on the inhibition time, which indicates that 1 nM hrIGF-I may be a modest dose.
Our experiments on MAPK and PI3K signaling in P. fucata in regard to carbohydrate/glycogen metabolism induction by hrIGF-I demonstrated that the mechanism of the hrIGF-I responsive machinery is conserved in P. fucata. The glycogen content in various tissues (the mantle, gill, AMs, foot, gonad, and DGs) was examined (Supplementary Figure S5). The glycogen content ranged from 1.69 to 4.4 mg/g wet mass, and was highly detectable in DGs and AMs, and at low levels in the mantle, which is similar to a previous study in Crassostrea angulate 62 . Although muscle is not a glycogen storage tissue in oysters, because its level does not exceed 5% of the total biochemical content 63 , muscle is a repository of active glycogenolysis that quickly provides ATP for muscular contraction in many species 64 . In oysters, most tissues, such as in AMs, DGs, and gills, are capable of glycogen hydrolysis and/or glucose formation 63,65 . Therefore, we determined the glycogen contents after hrIGF-I injection in DGs and AMs.
Our data obtained for DGs and AMs of hrIGF-I-treated oysters showed higher glycogen contents 3 h post-injection compared with those of controls, and lower glucose levels in hemolymph than in the controls. This indicates that the increase in glycogen contents in DGs and AMs is correlated with the decreased glucose levels in the hemolymph. Similarly, in the white shrimp P. vannamei, bovine insulin has an effect on glucose uptake in the hepatopancreas 37 . Meanwhile, hrIGF-I stimulation increases the phosphorylation of Akt/PKB at T308 and p44/42 MAPK, which indicates that the PI3K and MAPK pathways may participate in regulating the effect of hrIGF-I on Scientific RepoRts | 6:22063 | DOI: 10.1038/srep22063 carbohydrate/glycogen metabolism in P. fucata. As described above, this may be caused by the cross-talk between PI3K and MAPK, although the mechanism is still unclear and needs to clarify in the future.
The mRNA expression of the four examined genes was higher in AMs than in DGs, although the expression of three of the mRNAs increased in DGs after the injection of hrIGF-I. The mRNA and protein expression of pfIRR increased in DGs after injection, whereas no difference was observed in AMs. pfIRR may have a transient effect in AMs, and long-term regulatory effects in DGs. The three downstream genes of the hrIGF-I signaling pathway were chosen based on the results of an RNA sequencing analysis in our previous study 66 . As is known, insulin inactivates GSK-3β by Akt-mediated phosphorylation of Ser9, reducing the GSK3β inhibitory activity toward GS to facilitate glycogen synthesis 67 . Our data for AMs is exactly the same as in previous reports, and it indicates that the function of GSK-3β is conserved in P. fucata. However, the expression of GSK-3β was not changed in DGs. This may be because GSK3 expression may have changed before or after the 3-h post-injection time of hrIGF-I. A study of the white shrimp P. vannamei has shown that the glycogen content fluctuates in DGs after hrIGF-I injection: the glycogen content in the hepatopancreas fell significantly within 1 h post-injection of hrIGF-I, increased significantly 3 h post-injection, and decreased significantly 5 h post-injection 37 . GS is expressed in DGs and AMs, but phospho-GS was only detected at low levels in DGs and AMs after hrIGF-I injection, as determined by western blotting. This may be caused by the species diversity of the antibody. However, the low level of phospho-GS still proved that the function of GS in glycogen synthesis is conserved in P. fucata.
GK is an enzyme that facilitates the phosphorylation of glucose to glucose-6-phosphate, and it plays an important role in the regulation of carbohydrate metabolism by acting as a glucose sensor, triggering shifts in metabolism in response to rising or falling levels of glucose 68 . PP1 plays a crucial role in the regulation of blood-glucose levels in the liver, as well as glycogen metabolism 35 . The mRNA expression of the GK and PP1 genes increased significantly in DGs, suggesting that their function in regulating glycogen content is conserved. The expression of PP1 protein was consistent with its mRNA expression pattern, but GK was not detected in DGs, which may be caused by the different responses of antibodies in different tissues, and the species diversity of the antibody. However, GK mRNA and protein expression did not change in AMs after the injection of hrIGF-I. We speculate that GK and PP1 may have transient effects in AMs, and long-term regulatory effects in DGs. In mammals, GK is expressed mostly in the liver, pancreas, small intestine, and brain, where it plays crucial roles in responding to rising or falling levels of blood glucose 69 . The functions of GK and PP1 in AMs need to be investigated in the future.
In conclusion, the present study shows that pfIRR interacts with hrIGF-I, and stimulates the MAPK and PI3K signaling pathways in P. fucata oocytes, and that there is cross-talk between the MAPK and PI3K/Akt pathways, in which MEK positively regulates the PI3K pathway and PI3K positively regulates the MAPK cascade. Additionally, the i.m. injection of hrIGF-I stimulates the PI3K and MAPK pathways to increase glucose transport, glycogen synthesis, and it also regulates the expression of specific genes. In the future, the multiple signaling nodes and routes involved in cross-talk between these two pathways, as well as the functions of MAPK and PI3K signaling in P. fucata in other physiological or environmental contexts, should be more thoroughly investigated. Oocytes were obtained directly from the gonad with a Pasteur pipet, filtered through 120-μ m nylon mesh to remove debris, and washed three times with filtered (0.45-μ m filter) natural seawater at room temperature (26 °C).
Western-blot assays. Cell lysates were separated by SDS-PAGE. After transfer to polyvinylidene fluoride membranes (0.22-mm pore size, EMD Millipore, Billerica, MA, USA), the membranes were blocked in 5% bovine serum albumen (BSA) (Thermo Fisher Scientific) in Tris-buffered saline-Tween 20 (TBST). The blots were incubated with 1:1,000 to 1:5,000 dilutions of the indicated antibodies in blocking buffer overnight at 4 °C. Then, blots were washed with TBST and incubated with a 1:5,000 dilution of an HRP-conjugated anti-rabbit secondary antibody in blocking buffer for 1 h at room temperature, followed by further washing. The SuperSignal ® West Pico chemiluminescent substrate was used to detect HRP according to the manufacturer's instructions (Pierce). Positive signals from the target proteins were visualized using a TANON5200 image analyzer (Bio-tanon, Shanghai, China).

MEK inhibition and Akt/PKB inhibition in vitro.
Oocytes were incubated without or with increasing concentrations of PD98059 or wortmannin for 1 h, or with 60 μ M PD98059 or 40 μ M wortmannin for 0.5, 1.5, and 2 h, and then treated with or without hrIGF-I (10 −9 M) for 30 min. Lysates were subjected to western blotting to detect the phosphorylation of Akt/PKB at T308 and S473, as well as the amounts of Akt/PKB, phosphorylated p44/42 MAPK, and p44/42 MAPK.

hrIGF-I stimulation in vivo.
Oysters were injected i.m. with hrIGF-I (0.06 μ g/g wet body weight) or an equal volume of PBS as a control for 3 h. The oysters were removed for hemolymph sampling and then sacrificed. Hemolymph samples were processed immediately. Glucose levels were determined according to the instructions of the manufacturer of the Glucose Assay Kit (Nanjing Jiancheng Bioengineering Institute, China). The DGs and AMs were quickly dissected and divided into three parts. One part was processed immediately for glycogen content according to the instructions of the manufacturer of the Liver/Muscle Glycogen Assay Kit (Nanjing Jiancheng Bioengineering Institute), while another was frozen in liquid nitrogen and stored at − 80 °C for subsequent RNA extraction. Real-time polymerase chain reaction (PCR) was performed as described previously 39 . Primers for four genes involved in the hrIGF-I signaling pathway, the pfIRR, GK, GSK-3β , and PP1 genes, as well as 18 S RNA and β -actin genes, are shown in Table 2. The third portion was subjected to western blotting to detect the phosphorylation of Akt/PKB at T308, as well as the amounts of Akt/PKB, phosphorylated p44/42 MAPK, p44/42 MAPK, pfIRR, GK, PP1, GS, and phospho-GS.  Table 2. Primer sequences for the amplification of target genes selected from an RNA sequencing study and GenBank.