CT-707

Activation of Pyk2 by CaM kinase II in cultured hypothalamic neurons and gonadotroph cells

Shiho Okitsu‐Sakurayama1 | Sayomi Higa‐Nakamine1 | Hidetsugu Torihara1 | Hana Takahashi1 | Shigeki Higashiyama2 | Hideyuki Yamamoto1

Abstract

Gonadotropin‐releasing hormone (GnRH) is secreted from hypothalamic GnRH neurons and stimulates a GnRH receptor in gonadotroph cells and GnRH neurons. The GnRH receptor belongs to the G‐protein‐coupled receptors, and stimulation of the GnRH receptor activates extracellular signal‐regulated protein kinase (ERK). We reported previously that the δ2 isoform of Ca2+/calmodulin‐dependent protein kinase II (CaM kinase IIδ2) was involved in GnRH‐induced ERK activation in cultured GnRH neurons (GT1–7 cells). Recently, we found that GnRH treatment of GT1–7 cells activated proline‐rich tyrosine kinase 2 (Pyk2), and Pyk2 was involved in ERK activation. In the current study, we examined the possibility that CaM kinase IIδ2 might activate Pyk2. Knockdown of CaM kinase IIδ2 and KN93, an inhibitor of CaM kinases, inhibited the GnRH‐induced activation of Pyk2. In the case of cultured gonadotroph cells (αT3‐1 cells), knockdown of CaM kinase IIβ’e inhibited GnRHinduced Pyk2 activation. In addition, our inhibitor studies indicated that Pyk2 and CaM kinase II were involved in the GnRH‐induced shedding of proHB‐EGF in GT1–7 cells. These results suggested that CaM kinase II activated the ERK pathway through Pyk2 activation and HB‐EGF production in response to GnRH.

K E Y W O R D S
calcium‐calmodulin‐dependent protein kinase type 2, gonadotropin‐releasing hormone, Gprotein‐coupled receptors, HB‐EGF protein, PTK2B protein

1 | INTRODUCTION

Gonadotropin‐releasing hormone (GnRH) is secreted from hypothalamic neurons (GnRH neurons) and stimulates anterior pituitary gonadotroph cells to synthesize and secrete two gonadotropins, luteinizing hormone (LH), and follicle‐stimulating hormone. The GnRH receptor belongs to a class of G‐protein‐coupled receptors (GPCRs) that activate phospholipase Cβ (Naor, 2009). The signal transduction pathways after GnRH receptor stimulation have been well studied using immortalized gonadotroph cells, which express a GnRH receptor, such as αT3‐1 cells and LβT2 cells (Holdstock, Aylwin, & Burrin, 1996; Kaiser, Conn, & Chin, 1997; Turgeon, Kimura, Waring, & Mellon, 1996; Windle, Weiner, & Mellon, 1990). Studies on the gene expression of LH subunits indicated that LβT2 cells are more differentiated than αT3‐1 cells (Japon, Rubinstein, & Low, 1994; for review, see Alarid, Windle, Whyte, & Mellon, 1996). In addition to gonadotroph cells, GnRH neurons also have a GnRH receptor, and the autocrine action of GnRH is reportedly involved in the regulation of the functions of GnRH neurons (for review, see Krsmanovic et al., 1990). GnRH treatment of immortalized GnRH neurons (GT1–7 cells) activates extracellular signal‐regulated kinase (ERK) (Naor, 2009; Shah & Catt, 2004). It has been reported that GnRH treatment of GT1–7 cells produces heparin‐binding epidermal growth factor (HB‐EGF) by ectodomain shedding of its precursor (proHB‐EGF), and then HB‐EGF stimulates the ErbB family, leading to ERK activation (Shah, Farshori, & Catt, 2004). However, the molecular mechanisms by which GnRH receptor stimulation induces the shedding of proHB‐EGF are not clear at present.
Ca2+/calmodulin‐dependent protein kinase II (CaM kinase II) is widely distributed in neuronal tissues (for review, see Coultrap & Bayer, 2012). Four subunits of CaM kinase II (α, β, γ, and δ) are encoded by separate genes, with alternative splicing in their variable domains generating various isoforms (for review, see Coultrap & Bayer, 2012). It was reported that the activation of integrin induced the formation of a complex of Raf‐1 and CaM kinase II, leading to ERK activation (Illario et al., 2003). Our inhibitor studies suggested that CaM kinase II was involved in GnRH‐induced ERK activation in GT1–7 cells and αT3‐1 cells (Omoto, Higa‐Nakamine, Higa, & Yamamoto, 2017; Yamanaka et al., 2007). However, the molecular mechanisms by which CaM kinase II activated ERK are still unclear.
Proline‐rich tyrosine kinase 2 (Pyk2) is abundant in the central nervous system, and is activated through phosphorylation at tyrosine 402 (Tyr402) by the elevation of intracellular Ca2+ and/ or the activation of protein kinase C (PKC) (Lev et al., 1995). The involvement of CaM kinase II in the activation of Pyk2 after the elevation of intracellular Ca2+ is controversial in various cell systems. Some studies suggested that CaM kinase II was involved in Pyk2 activation (Ginnan, Pfleiderer, Pumiglia, & Singer, 2004; Liang et al., 2017). Other studies reported that the Ca2+/calmodulincomplex activated Pyk2 through direct binding (Kohno, Matsuda, Sasaki, & Sasaki, 2008; Xie et al., 2008). Recently, we found that protein kinase D1 (PKD1) was activated by a novel type of PKC after the GnRH treatment of GT1–7 cells, and then PKD1 activated Pyk2 (Higa‐Nakamine, Maeda, Toku, & Yamamoto, 2015). Small interfering RNA (siRNA) of Pyk2 decreased ERK activation, suggesting that Pyk2 was involved in GnRH‐induced ERK activation (Higa‐Nakamine et al., 2015).
In the current study, we precisely examined the possibility that CaM kinase II might be involved in the GnRH‐induced activation of Pyk2 in GT1–7 cells and αT3‐1 cells. The involvement of CaM kinase II and Pyk2 in the shedding of proHB‐EGF was also examined.

2 | MATERIAL AND METHODS

2.1 | Materials

The following chemicals and reagents were obtained from the indicated sources: fetal calf serum (FCS) from HyClone (Logan, UT); des‐Gly10, (D‐Ala6)‐LH‐RH ethylamide (GnRH), Poly‐L‐lysine (PLL), mouse IgG, anti‐ERK antibody (M5670), G418, and Dulbecco’s modified Eagle’s medium (DMEM) from Sigma Chemical Co. (St Louis, MO); anti‐Src antibody (number 2108), anti‐phospho‐Src family (Tyr416) antibody (number 2101), anti‐PKD1 antibody (number 2052), anti‐phospho‐PKD1 (Ser744/748) antibody (number 2054), antiPyk2 antibody (number 3292), anti‐Fyn antibody (number 4023), and anti‐phospho‐ERK antibody (number 4370) from Cell Signaling Technologies (Danvers, MA); anti‐phospho‐Pyk2 (Tyr402) antibody (sc‐101790), and anti‐CaM kinase IIβγ antibody (CaMKIIγ[C‐18]) from Santa Cruz (Santa Cruz, CA); monoclonal antibody to the α‐subunit of CaM kinase II (CBα‐2; anti‐CaM kinase IIα antibody) from Life Technology (Tokyo, Japan); KN62 from SEIKAGAKU Corp. (Tokyo, Japan); KN93 from Sigma and SEIKAGAKU Corp.; DynaMaker Protein MultiColor from BioDynamics Lab. (Tokyo, Japan); protease inhibitor (PI) mixture and protein phosphatase inhibitor (PPI) mixture (EDTA free) from Nacalai Tesque (Kyoto, Japan); PF431396 from Abcam (Cambridge, UK); ionomycin from Calbiochem (Darmstadt, Germany); and SEAP reporter Gene Assay from Roche (Basle, Switzerland). An antibody against the δ1–δ4 isoforms of CaM kinase II (anti‐CaM kinase IIδ1‐δ4 antibody) was prepared by immunizing rabbits with a synthesized peptide corresponding to a 15‐amino acid segment from the unique carboxy‐terminal ends of the δ1–δ4 isoforms (Yamanaka et al., 2007). Other chemicals were of analytical grade.

2.2 | Cell culture and preparation of cell extracts

GT1–7 cells were kindly provided by Dr. R. Weiner (University of California, San Diego, CA) and Dr. M. Kawahara (Musashino University, Japan; Koyama, Konoha, Sadakane, Ohkawara, & Kawahara, 2011; Mellon et al., 1990). The cells were grown in DMEM containing 10% FCS and 100 μg/ml gentamicin in 0.02% (w/v) PLL‐coated Petri dishes (NIPPON Genetics, Tokyo, Japan), as described previously (Yamanaka et al., 2007), and maintained at 37°C in an atmosphere of 95% air and 5% CO2. αT3‐1 and LβT2 cells were kindly provided by Dr. P. L. Mellon (University of California) (Turgeon et al., 1996; Windle et al., 1990). These cells were grown in DMEM containing 10% FCS with 100 μg/ml streptomycin and 100 U/ ml penicillin. When we examined the effects of KN62 and KN93, the cells were preincubated for 30 min with KN62 or KN93. We chose the concentrations of KN62 (20 μM) and KN93 (20 μM) as directed by the manufacturer. The cells were incubated with or without GnRH or ionomycin in the presence or absence of KN62 and KN93. Cells were lysed in 1 × sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) sample buffer containing 2% (w/v) SDS, 62.5 mM Tris‐HCl, pH 6.8, 5% (v/v) 2‐mercaptoethanol, 5% (v/v) glycerol, and 0.01% (w/v) bromophenol blue (Laemmli, 1970). The cell lysate was sonicated for 10 s at room temperature, and heated to 98°C for 5 min to use as a cell extract (Higa‐Nakamine et al., 2012). The cell extract was kept at −80°C until use.

2.3 | SDS‐PAGE and immunoblotting analysis

SDS‐PAGE was performed by the method of Laemmli (Laemmli, 1970). Immunoblotting analysis was performed as described previously (Towbin, Staehelin, & Gordon, 1979; Yamanaka et al., 2007). In the case of immunoblotting analysis of ERK, PKD1, and CaM kinase II, the cell extracts were subjected to SDS‐PAGE in 10% (wt/vol) acrylamide gels. The anti‐phospho‐ERK, anti‐ERK, antiphospho‐PKD1 (Ser744/748), anti‐CaM kinase IIδ1‐δ4, and anti‐CaM kinase IIβγ antibodies were used at dilutions of 1:1,000, 1:2,000, 1:850, 1:500, and 1:200, respectively. After the anti‐phospho‐PKD1 (Ser744/748) antibody was stripped away, immunoblotting with an anti‐PKD1 antibody was performed at a dilution of 1:850. In the case of immunoblotting analysis of Src and Fyn, the cell extracts were subjected to SDS‐PAGE in a 9% (wt/vol) acrylamide gel. The anti‐Src, anti‐phospho‐Src family (Tyr416), and anti‐Fyn antibodies were all used at a dilution of 1:850. In the case of immunoblotting analysis of Pyk2, the cell extracts were subjected to SDS‐PAGE in a 7.5% (wt/ vol) acrylamide gel. The anti‐Pyk2 and anti‐phospho‐Pyk2 (Tyr402) antibodies were used at dilutions of 1:400 and 1:850, respectively. Immunoreactive proteins were detected using an enhanced chemiluminescence detection kit (GE Healthcare UK Ltd., Little Chalfont, UK) as directed by the manufacturer, and quantified using an ImageQuant LAS4000 mini (GE Healthcare UK Ltd.) with Multi Gauge software (version 3.1). We loaded MagickMark XP Western Protein Standard (Invitrogen) onto all SDS‐PAGE gels and estimated the apparent molecular weights of the standard proteins and proteins of interest by chemiluminescence. We noticed that the fold variation in phosphorylation of the proteins varied among the experiments. The reasons for these differences are not clear at present but may be due to differences in cell passage and cell batch. Therefore, representative results are shown instead of the mean ± standard error (SE) values of all the repeats of the same experiments.

2.4 | siRNA transfection

siRNAs against mouse CaM kinase IIβ’e and CaM kinase IIδ2, and control siRNA were obtained from Qiagen (Valencia, CA). The siRNAs used were as follows: CaM Kinase IIβ’e, 5′‐CACACGCAAACCAA CAAACAA‐3′; and CaM kinase IIδ2, 5′‐AGCCAAGAGTTTATTGAA GAA‐3′. The siRNAs were introduced into GT1–7 cells and αT3‐1 cells using a Neon Transfection System Kit (Invitrogen) as directed by the manufacturer. We obtained control siRNA, ALLStars Negative Control siRNA (Cat. No. SI03650318), from Qiagen (Valencia, CA). This siRNA is a validated nonsilencing siRNA with no homology to any known mammalian gene. Each siRNA was transfected into 0.5–1 × 106 cells per 35‐mm dish at a concentration of 40 nM. The cells were kept at 37°C in a CO2 incubator for 48 hr.

2.5 | Reverse transcription‐polymerase chain reaction (RT‐PCR)

Total RNA was prepared from LβT2 cells and αT3‐1 cells using TRIzol LS Reagent (Life Technologies, Inc., Gaithersburg, MD) in accordance with the manufacturer’s protocol. Messenger RNA (mRNA) was reverse‐transcribed into single‐stranded complementary DNA (cDNA) using an oligo (deoxythymidine) primer (Promega) and Moloney murine leukemia virus reverse transcriptase (GIBCO BRL). The reaction mixtures were diluted 20‐fold and then subjected to polymerase chain reaction (PCR) amplification. The PCR primers for CaM kinase II were designed based on the published sequences of the α (X14836), β (M16112), γ (J04063), and δ (J05072) subunits of CaM kinase II. For amplification of the α subunit: sense primer (5′‐AATGCCAGGAGGAAACTGAAG‐3′) and antisense primer (5′‐GCCTGGTCCTTCAATGGGGCA‐3′). For amplification of the β subunit: sense primer (5′‐AGACTGTG GAATGTCTGAAGAAGT‐3′) and antisense primer (5′‐TGAAAC CAGGCGCAGCTCTCACTGCAG‐3′). For amplification of the γ subunit: sense primer (5′‐CAACGGTCAACAGTGGCATCC‐3′) and antisense primer (5′‐GGTCACAAATCTTCGTGTAGG‐3′). For amplification of the δ subunit: sense primer (5′‐GCACGAAAGC AAGAGATC‐3′) and antisense primer (5′‐GACGTGGCATGTTGA CAAT‐3′). PCR amplification was performed using a Gene Amp PCR system 2400 (Perkin‐Elmer, Foster City, CA) with 30 cycles for all subunits. The PCR products were separated by electrophoresis on a 1.2% agarose gel and visualized by ethidium bromide staining.

2.6 | Overexpression of alkaline phosphatase‐proHB‐EGF

An alkaline phosphatase (AP) proHB‐EGF (AP‐proHB‐EGF) expression plasmid was produced by fusion of proHB‐EGF cDNA with an AP cDNA (Tokumaru et al., 2000). The plasmid was introduced into GT1–7 cells using a Neon Transfection System Kit (Invitrogen) as described above. The cells were kept at 37°C in a CO2 incubator for 8 hr. The culture medium was changed to DMEM containing 10% FCS, 100 μg/ml gentamicin, and 600 μg/ml G418. G418‐resistant cell lines were isolated as described previously (Yamanaka et al., 2007). Two or three days before experiments, 2 × 105 cells were plated onto a PLL‐coated 24‐well plate (NIPPON Genetics, Tokyo, Japan). When we examined the effect of KN62 and KN93, the cells were preincubated for 30 min with KN62 or KN93. When we examined the effect of PF431396, the cells were preincubated for 60 min with PF431396. The cells were treated with or without GnRH in the presence or absence of KN62, KN93, or PF431396. We centrifuged the culture medium at 15,000 g for 1 min, and the obtained supernatant was used for the measurement of AP activity. The treated cells were lysed in 50 μl of 1 × SDS‐PAGE sample buffer, and the cell extracts were obtained as described above.

2.7 | Measurement of alkaline phosphatase activity

The enzyme activity of AP was measured using a SEAP Reporter Gene Assay kit (Roche). The principle of the SEAP assay is as follows: dephosphorylation by AP of a chemiluminescent substrate (3‐[4‐methoxyspiro[1, 2‐dioxetane‐3, 2′[5′‐chloro]‐tricyclo[3.3.1.13,7] decane]‐4‐yl] phenyl phosphate [CSPD]) generates an unstable dioxetane anion. The dioxetane anions emit light at a wavelength of 477 nm. In the experiment, CSPD was added to the supernatant, and we measured light emission at a wavelength at 477 nm using a luminometer (GLOMAX 20/20; Promega, Co., Madison, WI) in accordance with the manufacturer’s instructions.

2.8 | Statistical evaluation

We repeated the experiments at least three times with reproducible results. We noticed that the enzyme activity of AP varied among the experiments. Therefore, representative results are shown instead of all the repeats of the same experiments. We expressed values as the mean ± SE. Statistical analysis was performed using one‐way analysis of variance plus Duncan’s multiple range test. The p < 0.05 was considered statistically significant. 3 | RESULTS 3.1 | Effects of KN93 on GnRH‐induced activation of ERK and phosphorylation of Pyk2 at Tyr402 in GT1–7 cells Immunoblotting analysis indicated that ERK1 and ERK2 were expressed at similar levels in GT1–7 cells (Figure 1a). When GT1–7 cells were treated with 50 nM GnRH, ERK2 was strongly activated after 10 min, and it decreased after 30 min. ERK2 activation was almost completely inhibited by KN93, as we reported previously (Yamanaka et al., 2007). It has been reported that Pyk2 is activated through phosphorylation of Tyr402 (for review, see Higa‐Nakamine et al., 2015). Phosphorylation of Tyr402 increased approximately 6.5‐fold and 3.9‐fold after the GnRH treatment of 10 and 30 min, respectively, suggesting that GnRH treatment activated Pyk2 (Figure 1). Of note, Pyk2 activation was decreased by ~64% and ~43% in the presence of KN93 after the GnRH treatment of 10 and 30 min, respectively. In our previous study, we reported that PKD1 was activated by GnRH, and knockdown experiments and inhibitor studies indicated that PKD1 was involved in GnRH‐induced Pyk2 activation (Higa‐Nakamine et al., 2015). GnRH activated PKD1 through phosphorylation of Ser744/ Ser748, as we reported previously, and KN93 did not inhibit the GnRHinduced activation of PKD1 (Figure 1a). The protein amounts of ERK1, ERK2, and Pyk2 were not changed by any treatments. As we also reported previously, PKD1 was degraded after the GnRH treatment of GT1–7 cells (Higa‐Nakamine et al., 2015). We found that KN93 did not inhibit the GnRH‐induced degradation of PKD1 (Figure 1a). In the previous study, we reported that dasatinib, an inhibitor of the Src family, inhibited the GnRH induced‐activation of Pyk2 by ~94%, which indicated that Pyk2 activation was mainly conducted by the Src family (Higa‐Nakamine et al., 2015). Phosphorylation of Src or Fyn at Tyr416, numbered according to Src, was observed in the absence of GnRH, suggesting that Src or Fyn was activated in basal conditions (Figure 1b). We found that KN93 did not inhibit the phosphorylation of Tyr416 in the presence or absence of GnRH. The protein amounts of Src and Fyn were not changed with any treatment. These results indicated that KN93 had no effects on the activation of PKD1 and the Src family, and that CaM kinase II activation was necessary for GnRH‐induced Pyk2 activation. 3.2 | Effects of CaM kinase IIδ2 siRNA on GnRH‐induced activation of Pyk2 in GT1–7 cells In the previous study, immunoblotting analysis indicated that CaM kinase IIδ2 was highly expressed, whereas other isoforms were not expressed in GT1–7 cells (Yamanaka et al., 2007). For the next step, we examined whether or not knockdown of CaM kinase IIδ2 inhibited the GnRH‐induced activation of Pyk2 in GT1–7 cells (Figure 2). Transfection of siRNA against CaM kinase IIδ2 decreased the protein level of CaM kinase IIδ2 by 72% (Figure 2a). Consequently, the activation of Pyk2 by GnRH was inhibited by 55 ± 7% (n = 3, p < 0.05) after transfection of siRNA against CaM kinase IIδ2 (Figures 2a,b). The protein amount of Pyk2 was not changed by any of the treatments. 3.3 | Activation of Pyk2 by CaM kinase IIδ2 overexpression and ionomycin For the next step, we examined whether or not overexpression of CaM kinase IIδ2 enhanced Pyk2 activation in GT1–7 cells. For this purpose, we established several cell lines that were stably transfected with the CaM kinase IIδ2 cDNA. Among these cell lines, δ2–15 cells overexpressed CaM kinase IIδ2 more than 24‐fold (Figure 3a). In contrast with CaM kinase IIδ2, the activation and protein levels of GnRH‐induced activation of Pyk2. (a) GT1–7 cells were transfected with 40 nM control siRNA (Cont) or 40 nM CaM kinase IIδ2 siRNA (CaMKIIδ2). After 48‐hr incubation, the cells were treated with or without 50 nM GnRH for 5 min. Cell extracts (30 μg) were subjected to immunoblotting analysis of phospho‐Pyk2 (Tyr402), Pyk2, and CaM kinase IIδ2, as described in Materials and Methods. The positions of Pyk2 (P‐Pyk2 [Y402)], Pyk2), and CaM kinase IIδ2 (CaMKIIδ2) are indicated. (b) The level of Pyk2 activation with control siRNA (Cont) and without GnRH was taken as 100% (control), and other values were calculated from this value. Values are the mean ± SE (three samples per condition). **p < 0.01 (vs. control). The difference in the level of Pyk2 activation between the cells treated with control siRNA and GnRH and the cells treated with CaM kinase IIδ2 siRNA and GnRH was statistically significant (#p < 0.05). CaM kinase II: Ca2+/calmodulin‐dependent protein kinase II; GnRH: gonadotropin‐releasing hormone; Pyk2: proline‐rich tyrosine kinase 2; siRNA: small interfering RNA Src were not increased in δ2–15 cells, compared with wild‐type cells (Figure 3a). Next, we treated GT1–7 cells with ionomycin, an ionophore of calcium, to activate selectively the pathway of CaM kinases. We found that ionomycin activated Pyk2 about 2‐fold after 5 min in wild‐type cells (Figure 3b). The activation of Pyk2 was completely inhibited by KN93, suggesting that CaM kinases were involved in the ionomycin‐induced activation of Pyk2. It is of interest that Pyk2 in δ2–15 cells was activated even in the absence of ionomycin. We confirmed that KN93 strongly inhibited Pyk2 activation both in the presence and absence of ionomycin in δ2–15 cells (Figure 3b). When the activation level of Pyk2 in the presence of ionomycin was estimated as 100%, it was 72.9 ± 16.0% in the absence of ionomycin and KN93, and no significant difference was observed (Figure 3c). KN93 inhibited Pyk2 activation to 8.3 ± 5.2% (p < 0.01) and 26.6 ± 12.3% (p < 0.01) in the absence and presence of ionomycin, respectively. These results further supported our idea that CaM kinase IIδ2 was involved in Pyk2 activation after elevation of intracellular Ca2+. 3.4 | Effects of KN93 on GnRH‐induced activation of Pyk2 in αT3‐1 cells Recently, we found that GnRH‐induced ERK activation was inhibited by KN93 in αT3‐1 cells (Omoto et al., 2017). By analogy with GT1–7 cells, we expected that GnRH treatment of αT3‐1 cells activated Pyk2 through CaM kinase II. When we treated αT3‐1 cells with GnRH, Pyk2 was activated approximately 2.6‐fold after 30 min, and it decreased after 60 min (Figure 4a). Figure 4b shows that the GnRH‐induced activation of Pyk2 was inhibited by about 42% in the presence of KN93. Src was expressed in αT3‐1 Src) are indicated. GnRH: gonadotropin‐releasing hormone; Pyk2: proline‐rich tyrosine kinase 2 cells, and the activation was not changed by any treatments (Figure 4b). These results suggested that CaM kinase II was involved in the GnRH‐induced activation of Pyk2 in αT3‐1 cells as well as GT1–7 cells. 3.5 | Identification of CaM kinase II isoforms in αT3‐1 and LβT2 cells We decided to identify the major CaM kinase II isoforms in αT3‐1 cells for knockdown experiments. In the previous study, we found that KN93 did not inhibit the GnRH‐induced activation of ERK in LβT2 cells, indicating that CaM kinase II was not involved in the reaction (Yamanaka et al., 2007). In the current study, we examined the isoforms of CaM kinase II in LβT2 cells for comparison. First, mRNAs of the isoforms were examined using total RNA from αT3‐1 and LβT2 cells (Figure 5). When we used a primer pair that was able to amplify fragments of the α (568 bp) and αB (601 bp) isoforms, only the αB band was observed as a weak band in both cell lines (Figure 5a). When we used a primer pair that was able to amplify the fragments of all reported isoforms of the β‐subunit with different lengths, only the β’e isoform (746 bp) was expressed abundantly and weakly in αT3‐1 and LβT2 cells, respectively (Figure 5a). With regard to the γ‐subunit, we could not obtain a 1,270‐bp product, which should have been amplified from all isoforms (data not shown). We obtained 513‐bp and 424‐bp products, which corresponded to δ2 and δ6, respectively, from αT3‐1 cells (Figure 5a). In contrast, no products were observed in LβT2 cells. These results suggested that CaM kinase II was much more highly expressed in αT3‐1 cells than in LβT2 cells, and that the major isoforms at the mRNA level in αT3‐1 cells were β’e, δ2, and δ6. When we carried out immunoblotting analysis with a monoclonal antibody against the α subunit, no clear bands were detected in αT3‐1 cells or LβT2 cells (data not shown). When the anti‐CaM kinase IIβγ antibody was used, two bands were observed (Figure 5b). Between the two bands, a 55‐kDa protein was strongly detected in αT3‐1 cells. The anti‐CaM kinase IIδ1–δ4 antibody detected a 55‐kDa protein strongly in αT3‐1 cells, but no clear band was observed in LβT2 cells (Figure 5b). In the previous study, we reported that the apparent molecular masses of the β’e and δ2 isoforms were 55 kDa (Tabuchi et al., 2000; Yamanaka et al., 2007). These results strongly suggested that β’e and δ2 were major isoforms in αT3‐1 cells, and that a small amount of β’e isoform was expressed in LβT2 cells at the protein level. We could not identify the 60‐kDa protein, which appeared to be an isoform of the β or γ subunit. We could also not detect the δ6 isoform at the protein level (data not shown). 3.6 | Effect of siRNA of δ2 and β’e isoforms on GnRH‐induced activation of Pyk2 in αT3‐1 cells We transfected αT3‐1 cells with siRNAs for CaM kinase IIδ2 and CaM kinase IIβ’e to investigate which CaM kinase II isoform was involved in GnRH‐induced Pyk2 activation. Transfection of CaM kinase IIδ2 siRNA into αT3‐1 cells decreased the protein level of CaM kinase IIδ2 to 42.5% (Figure 5c). Transfection of siRNA for CaM kinase IIδ2 was expected to knockdown δ6 as well as δ2. In contrast, the protein level of CaM kinase IIβ’e was decreased to 4.8% (Figure 5c). Knockdown of CaM kinase IIδ2 did not increase the protein level of CaM kinase IIβ’e, and vice versa, indicating no compensation mechanisms existed between the isoforms. It was of interest that the GnRH‐induced activation of Pyk2 was not inhibited after the knockdown of CaM kinase IIδ2, whereas it was inhibited by ~45% after the knockdown of CaM kinase IIβ’e (Figure 5c). These results may suggest that CaM kinase IIβ’e was more involved in the GnRH‐induced activation of Pyk2 than CaM kinase IIδ2 in αT3‐1 cells. 3.7 | Effect of PF431396 for GnRH‐induced Pyk2 and ERK activations For the next step, we examined the involvement of Pyk2 in the GnRHinduced ERK activation using PF431396, a Pyk2 inhibitor (Figure 6a). We found that PF431396 inhibited ERK activation in GT1–7 cells to 38% and 15% at 1 μM and 10 μM, respectively (Figure 6a). We next examined the effects of PF431396 on Pyk2 activation. It was surprising that PF431396 strongly inhibited the GnRH‐induced activation of Pyk2, and 1 μM and 10 μM PF431396 inhibited Pyk2 activation to 64% and 22%, respectively. In the case of αT3‐1 cells, PF431396 inhibited the GnRH‐induced ERK activation to 54.1% and 41.8% at 1 μM and 10 μM, respectively (Figure 6b). We noticed that Pyk2 was slightly activated in basal conditions, and PF431396 inhibited the basal activation. In the presence of GnRH, PF431396 inhibited Pyk2 activation to 37% and 20% at 1 μM and 10 μM, respectively. The protein amounts of ERK1, ERK2, and Pyk2 were not changed by any of the treatments of GT1–7 cells and αT3‐1 cells. These results suggested that PF431396 inhibited GnRH‐induced ERK activation through the inhibition of the GnRH‐induced autophosphorylation of Pyk2. 3.8 | Cleavage of AP‐proHB‐EGF by treatment with GnRH Finally, we asked whether or not CaM kinase II and Pyk2 were involved in the GnRH‐induced shedding of proHB‐EGF. For this purpose, we generated GT1–7 cells that stably expressed APproHB‐EGF. We measured the enzyme activity of AP using a SEAP assay with the supernatant from the culture medium. The enzyme activity of AP was increased by 460 ± 8% (p < 0.01) of the control by GnRH treatment of 60 min (Figure 7a). In contrast, the enzyme activity increased by 383 ± 15% (p < 0.01) of the control in the presence of GnRH plus KN93. Although the effect of KN93 was not so strong, the AP enzyme activity was significantly decreased compared with GnRH alone (p < 0.01). We next examined the effects of KN62, another inhibitor of CaM kinase II, on the GnRHinduced shedding of AP‐proHB‐EGF (Figure 7b). In this experiment, the enzyme activity was increased by 669 ± 34% (p < 0.01) of the control by GnRH treatment. Compared with GnRH alone, the enzyme activity of AP was decreased to 517 ± 21% (p < 0.01) of the control in the presence of both GnRH and KN62. KN93 and KN62 had no effects on the shedding of proHB‐EGF in the absence of GnRH. In addition to the measurement of enzyme activity with the culture medium supernatant, we prepared cell extracts to examine ERK activation (Figures 7c,d). ERK was activated approximately 9.4‐fold in Figure 7c and 21‐fold in Figure 7d after 60‐min treatment of the cells with GnRH. We found that ERK activation was decreased by ~54% and ~85% in the presence of KN93 and KN62, respectively. These results indicated that KN62 more effectively inhibited both the shedding of AP‐proHB‐EGF and ERK activation than KN93. We next examined the effects of PF431396 on the GnRH‐induced shedding of AP‐proHB‐EGF (Figure 7e). In this experiment, the enzyme activity was increased by 845 ± 17% (p < 0.01) of the control by GnRH treatment. In contrast, the enzyme activity in the presence of GnRH plus PF431396 was 419 ± 41% (p < 0.01) of the control, indicating that PF431396 inhibited the GnRH‐induced shedding of AP‐proHB‐EGF (p < 0.01). We found that PF431396 did not decrease the enzyme activity in the absence of GnRH. These results suggested that both CaM kinase II and Pyk2 were involved in the shedding of AP‐proHB‐EGF. 4 | DISCUSSION When Pyk2 was identified from human brain, it was reported that Pyk2 was activated by the elevation of the intracellular Ca2+ concentration as well as by the activation of PKC (Lev et al., 1995). Pyk2 is highly expressed in the central nervous system, and accumulating evidence indicates that Pyk2 is involved in synaptic functions and plasticity (for review, see Giralt et al., 2017). The molecular mechanisms by which the elevation of Ca2+ concentration activates Pyk2 are very important questions, and many studies have been performed in vitro and in various cell systems (for review, see Liang et al., 2017). From in vitro studies, it was reported that Ca2+/calmodulin bound to a FERM (band four‐point‐one, ezrin, radixin, and moesin homology) domain in Pyk2 to release the catalytic domain from autoinhibition (Kohno et al., 2008). Other groups reported that Ca2+/calmodulin bound to a putative calmodulin‐binding motif in the catalytic domain to activate Pyk2 (Xie et al., 2008). Recently, it was reported that periodic mechanical stress activated Pyk2 in cultured chondrocytes, and the activation was inhibited by KN93 and knockdown of CaM kinase II, although the isoforms of CaM kinase II in the cells were not identified (Liang et al., 2017). In the previous study, we found that KN93 inhibited GnRHinduced ERK activation in GT1–7 cells, whereas overexpression of CaM kinase IIδ2 augmented it (Yamanaka et al., 2007). We also found that novel PKC isoforms activated PKD1 after the GnRH treatment of GT1–7 cells, and then, PKD1 activated Pyk2 through the augmentation of phosphorylation of Pyk2 at Tyr402 (Higa‐Nakamine et al., 2015). Because the PKD family belongs to the CaM kinase group and the substrate specificity of PKD1 appears to be similar to that of CaM kinase II (for review, see Rozengurt, Rey, & Waldron, 2005), we considered the possibility that CaM kinase II might activate Pyk2 through the same molecular mechanisms as PKD1. Therefore, we decided to examine whether or not CaM kinase II was involved in the GnRH‐induced activation of Pyk2. We found that KN93 and knockdown of CaM kinase IIδ2 inhibited the GnRHinduced activation of Pyk2 in GT1–7 cells. In contrast, overexpression of CaM kinase IIδ2 augmented Pyk2 activation, which was inhibited also by KN93. These results clearly indicated that CaM kinase IIδ2 was involved in Ca2+‐induced Pyk2 activation in GT1–7 cells. In addition to GT1–7 cells, the GnRH treatment activated Pyk2 in αT3‐1 cells, and KN93 strongly inhibited Pyk2 activation. Next, we decided to identify the major isoforms of CaM kinase II in αT3‐1 cells for knockdown experiments. RT‐PCR and immunoblotting analysis suggested that β’e and δ2 were the major isoforms in αT3‐1 cells. Although RT‐PCR suggested the expression of δ6, immunoblotting analysis with an anti‐CaM kinase II antibody could not detect δ6 at the protein level (data not shown). After the identification of major isoforms of CaM kinase II, we knocked down the β’e and δ2 isoforms. In contrast with GT1–7 cells, our data suggested that the β’e isoform was more involved in Pyk2 activation than the δ2 isoform. However, we could not exclude the possibility that both the β’e and δ2 isoforms were involved in Pyk2 activation, because the knockdown of the δ2 isoform was less remarkable than the β’e isoform. The current study with PF431396 suggested that the contribution of Pyk2 to ERK activation was about 80% both in GT1–7 cells and αT3‐1 cells. These results indicated that Pyk2 was crucial for the GnRH‐induced ERK activation in both cells. We reported that dasatinib inhibited GnRH‐induced Pyk2 activation almost completely (Higa‐Nakamine et al., 2015). From this result, we considered that phosphorylation of Pyk2 was mainly conducted by the Src family. Therefore, it was unexpected that PF431396 inhibited GnRHinduced Pyk2 activation almost completely. It has been reported that autophosphorylation of Pyk2 at Tyr402 is important for the interaction of Pyk2 and Src (Park, Avraham, & Avraham, 2004). Taken together, autophosphorylation of a small amount of Pyk2 at Tyr402 may be prerequisite for the phosphorylation by the Src family of the rest of Pyk2. The elucidation of each role of autophosphorylation of Pyk2 and its phosphorylation by the Src family is worth examining in a future study. We generated GT1–7 cells that stably expressed AP‐proHB‐EGF to examine the possibility that CaM kinase II and Pyk2 might be involved in the GnRH‐induced shedding of proHB‐EGF. The results of the SEAP assay clearly indicated that CaM kinase II and Pyk2 were involved in the GnRH‐induced shedding of AP‐proHB‐EGF. We noticed that PF431396 inhibited the shedding more strongly than KN93 and KN62. These results may suggest that Pyk2 is more involved in the shedding of proHB‐EGF than CaM kinase II. Figure 8 shows the pathways that appear to play roles in ERK activation after GnRH treatment of GT1–7 cells from our previous studies and the current study. We found that the expression level of CaM kinase II in LβT2 cells was much lower than that in αT3‐1 cells. In the previous study, we reported that KN93 did not inhibit the GnRH‐induced activation of ERK in LβT2 cells (Yamanaka et al., 2007). Taken together, CaM kinase II may not play important roles in the GnRH‐induced activation of ERK in LβT2 cells. It is not clear at present whether or not the difference in the involvement of CaM kinase II in ERK activation between αT3‐1 cells and LβT2 cells reflects the developmental changes of gonadotroph cells in the signal transduction pathways after GnRH receptor stimulation. In conclusion, the current study contributes to understanding of the regulation of the functions of GnRH neurons and gonadotroph cells by GnRH. Because the precise regulation of the functions of GnRH neurons and gonadotroph cells are crucial for female reproductive functions, the current study will contribute also to the understanding of the regulation of female reproductive cycles. REFERENCES Alarid, E. T., Windle, J. J., Whyte, D. B., & Mellon, P. L. (1996). Immortalization of pituitary cells at discrete stages of development by directed oncogenesis in transgenic mice. Development, 122, 3319–3329. Coultrap, S. J., & Bayer, K. U. (2012). CaMKII regulation in information processing and storage. Trends in Neurosciences, 35, 607–618. Ginnan, R., Pfleiderer, P. J., Pumiglia, K., & Singer, H. A. (2004). PKC‐delta and CaMKII‐delta 2 mediate ATP‐dependent activation of ERK1/2 in vascular smooth muscle. American Journal of Physiology ‐Cell Physiology, 286, C1281–C1289. Giralt, A., Brito, V., Chevy, Q., Simonnet, C., Otsu, Y., Cifuentes‐Díaz, C., … Girault, J. A. (2017). Pyk2 modulates hippocampal excitatory synapses and contributes to cognitive deficits in a Huntington's disease model. Nature Communications, 8, 15592. Higa‐Nakamine, S., Maeda, N., Toku, S., & Yamamoto, H. (2015). Involvement of protein kinase D1 in signal transduction from the protein kinase C pathway to the tyrosine kinase pathway in response to gonadotropin‐releasing hormone. The Journal of Biological Chemistry, 290, 25974–25985. Higa‐Nakamine, S., Maeda, N., Toku, S., Yamamoto, T., Yingyuenyong, M., Kawahara, M., & Yamamoto, H. (2012). Selective cleavage of ErbB4 by Gprotein‐coupled gonadotropin‐releasing hormone receptor in cultured hypothalamic neurons. Journal of Cellular Physiology, 227, 2492–2501. Holdstock, J. G., Aylwin, S. J., & Burrin, J. M. (1996). Calcium and glycoprotein hormone alpha‐subunit gene expression and secretion in alpha T3‐1 gonadotropes. Molecular Endocrinology, 10, 1308–1317. Illario, M., Cavallo, A. L., Bayer, K. U., Di matola, T., Fenzi, G., Rossi, G., & Vitale, M. (2003). Calcium/calmodulin‐dependent protein kinase II binds to Raf‐1 and modulates integrin‐stimulated ERK activation. The Journal of Biological Chemistry, 278, 45101–45108. Japón, M. A., Rubinstein, M., & Low, M. J. (1994). In situ hybridization analysis of anterior pituitary hormone gene expression during fetal mouse development. Journal of Histochemistry and Cytochemistry, 42, 1117–1125. Kaiser, U. B., Conn, P. M., & Chin, W. W. (1997). Studies of CT-707 gonadotropinreleasing hormone (GnRH) action using GnRH receptor‐expressing pituitary cell lines. Endocrine Reviews, 18, 46–70.
Kohno, T., Matsuda, E., Sasaki, H., & Sasaki, T. (2008). Protein‐tyrosine kinase CAKbeta/PYK2 is activated by binding Ca2+/calmodulin to FERM F2 alpha2 helix and thus forming its dimer. Biochemical Journal, 410, 513–523.
Koyama, H., Konoha, K., Sadakane, Y., Ohkawara, S., & Kawahara, M. (2011). Zinc neurotoxicity and the pathogenesis of vascular‐type dementia: Involvement of calcium dyshomeostasis and carnosine. Journal of Clinical Toxicology, s3, 002.
Krsmanovic, L. Z., Martinez‐Fuentes, A. J., Arora, K. K., Mores, N., Navarro, C. E., Chen, H. C., … Catt, K. J. (1990). Autocrine regulation of gonadotropin‐releasing hormone secretion in cultured hypothalamic neurons. Endocrinology, 140, 1423–1431.
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685.
Lev, S., Moreno, H., Martinez, R., Canoll, P., Peles, E., Musacchio, J. M., … Schlessinger, J. (1995). Protein tyrosine kinase PYK2 involved in Ca(2 +)‐induced regulation of ion channel and MAP kinase functions. Nature, 376, 737–745.
Liang, W., Li, Z., Wang, Z., Zhou, J., Song, H., Xu, S., … Fan, W. (2017). Periodic mechanical stress induces chondrocyte proliferation and matrix synthesis via CaMKII‐mediated Pyk2 signaling. Cellular Physiology and Biochemistry, 42, 383–396.
Mellon, P. L., Windle, J. J., Goldsmith, P. C., Padula, C. A., Roberts, J. L., & Weiner, R. I. (1990). Immortalization of hypothalamic GnRH neurons by genetically targeted tumorigenesis. Neuron, 5, 1–10.
Naor, Z. (2009). Signaling by G‐protein‐coupled receptor (GPCR): Studies on the GnRH receptor. Frontiers in Neuroendocrinology, 30, 10–29.
Omoto, Y., Higa‐Nakamine, S., Higa, A., & Yamamoto, H. (2017). ErbB4 cleavage by gonadotropin‐releasing hormone receptor stimulation in cultured gonadotroph cells. European Journal of Pharmacology, 799, 171–179.
Park, S. Y., Avraham, H. K., & Avraham, S. (2004). RAFTK/Pyk2 activation is mediated by trans‐acting autophosphorylation in a Src‐independent manner. Journal of Biological Chemistry, 279, 33315–33322.
Rozengurt, E., Rey, O., & Waldron, R. T. (2005). Protein kinase D signaling. Journal of Biological Chemistry, 280, 13205–13208.
Shah, B. H., & Catt, K. J. (2004). GPCR‐mediated transactivation of RTKs in the CNS: Mechanisms and consequences. Trends in Neurosciences, 27, 48–53.
Shah, B. H., Farshori, M. P., & Catt, K. J. (2004). Neuropeptide‐induced transactivation of a neuronal epidermal growth factor receptor is mediated by metalloprotease‐dependent formation of heparin‐binding epidermal growth factor. Journal of Biological Chemistry, 279, 414–420.
Tabuchi, H., Yamamoto, H., Matsumoto, K., Ebihara, K., Takeuchi, Y., Fukunaga, K., … Miyamoto, E. (2000). Regulation of insulin secretion by overexpression of Ca2+/calmodulin‐dependent protein kinase II in insulinoma MIN6 cells. Endocrinology, 141, 2350–2360.
Tokumaru, S., Higashiyama, S., Endo, T., Nakagawa, T., Miyagawa, J., Yamamori, K., … Taniguchi, N. (2000). Ectodomain shedding of epidermal growth factor receptor ligands is required for keratinocyte migration in cutaneous wound healing. Journal of Cell Biology, 151, 209–220.
Towbin, H., Staehelin, T., & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proceedings of the National Academy of Sciences of the United States of America, 76, 4350–4354.
Turgeon, J. L., Kimura, Y., Waring, D. W., & Mellon, P. L. (1996). Steroid and pulsatile gonadotropin‐releasing hormone (GnRH) regulation of luteinizing hormone and GnRH receptor in a novel gonadotrope cell line. Molecular Endocrinology, 10, 439–450.
Windle, J. J., Weiner, R. I., & Mellon, P. L. (1990). Cell lines of the pituitary gonadotrope lineage derived by targeted oncogenesis in transgenic mice. Molecular Endocrinology, 4, 597–603.
Xie, J., Allen, K. H., Marguet, A., Berghorn, K. A., Bliss, S. P., Navratil, A. M., … Roberson, M. S. (2008). Analysis of the calcium‐dependent regulation of Molecular Endocrinology, 22, 2322–2335.
Yamanaka, A., Hiragami, Y., Maeda, N., Toku, S., Kawahara, M., Naito, Y., & Yamamoto, H. (2007). Involvement of CaM kinase II in gonadotropinreleasing hormone‐induced activation of MAP kinase in cultured hypothalamic neurons. Archives of Biochemistry and Biophysics, 466, 234–241.