Increased expression of kisspeptin and GnRH forms in the brain of scombroid fish during final ovarian maturation and ovulation

Adult chub mackerel (2-year-old) were caught from the wild using purse seine during autumn 2008 and reared for six months in sea pens at a fish farm in the Oita prefecture, Kyushu Island. During the following spawning season (April-June), fish were transferred to Fishery Research Laboratory of Kyushu University and moved into 3-ton outdoor concrete tanks circulated with running seawater. The fish were acclimated and reared under natural photoperiod and temperature. Our previous studies indicated that female chub mackerel fail to undergo FOM and ovulation spontaneously in this captive system [13, 14]. An induced spawning protocol based on sustained release GnRH delivery system was adopted from previous study [16].

After 3 days of acclimation, fish were anaesthetized with 2-phenoxyethanol (100 mg/l) and females with late vitellogenic oocytes (600–650 μm in diameter) were selected by ovarian biopsy using a plastic catheter tube (2 mm internal diameter), as described previously [13, 16]. Males oozing milt under gentle abdominal pressure were selected. After selection of required number of females and males, intramuscular injection with the GnRH agonist (D-Ala⁶, des-Gly¹⁰)-LHRH ethylamide (Sigma-Aldrich, St. Louis, USA) at 400 μg/kg body weight were performed on April 30^th, May 13^th, May 21^st, and May 22^nd, 2009 to obtain different ovarian stages, namely germinal vesicle migration (GVM), oocyte hydration (HY), ovulation (OV), and post-ovulation (POV), respectively. The sampling times were 13.00, 16.00, 20.00, and 6.00 h of the day, respectively. In all cases, injections were performed at 11.00 h. Sampling times were fixed based on our previous data on time course of FOM and ovulation in chub mackerel induced by GnRHa [16]. Fish sampling for the analysis was performed on day 8, based on previous reports showing the decline in the plasma concentration of GnRH agonist on day 5 after intramuscular injection with the GnRH agonist suspended in coconut oil in Plaice, Pleuronectes platessa[22, 23]. The first spawning was observed 34–36 h post-injection, and subsequent daily spawning occurred between 22.00 and 24.00 h. In the following experimental system, daily spawning of chub mackerel is observed for 20–30 days during the spawning season, when the water temperature ranged between 18-23°C (Yoneda et al., unpublished observations). The late vitellogenic (LV) stage fish were sampled before the start of induced spawning experiment.

Fish used in the experiment were sacrificed in accordance with the guidelines for animal experiments proposed by the Faculty of Agriculture and Graduate Course at Kyushu University and according to the laws (No. 105) and notifications (No. 6) of the Japanese government. The fork length, body, and gonad weights of each individual were measured before tissue sampling. The brain and pituitary of each fish were removed following decapitation, snap-frozen in liquid nitrogen, and stored at −80°C until further analysis. For ovarian histological evaluation, ovary midsections from individual fish were fixed in Bouin’s solution. To analyze the changes in kiss and gnrh mRNA levels in the whole brain and GnRH peptides in the whole brain and pituitary, two experimental sets of fish samples were used (Table 1). The brain tissue from the first set was used for mRNA analysis, and the second set was used for GnRH peptide analysis. Male fish were excluded from the analysis.

After fixation, ovary samples from each fish were dehydrated in a series of ethanol solutions up to 100%, embedded in paraffin, and sectioned at 5–7 μm using a Leica RM 2155 rotary microtome (Leica, Germany). Sections were stained with hematoxylin and counterstained with eosin. The stained tissues were subsequently observed under a light microscope. Chub mackerel show asynchronous ovarian development, and ovarian stages were thus classified based on the developmental stages of first clutch oocytes as (1) LV, (2) GVM, (3) HY, (4) OV, and (5) POV.

Quantitative real-time PCR (qRT-PCR) analysis was performed on an Mx 3000P quantitative PCR system (Stratagene). Total brain RNA was extracted using ISOGEN (Nippon Gene, Japan), according to the manufacturer’s protocol. One microgram of total RNA from each brain sample was digested with DNase I (Invitrogen) and used as template for reverse transcription (RT) reaction. The cDNA synthesis was performed using Superscript III Reverse Transcriptase (Invitrogen) in a 20 μl reaction mixture containing 2.5 mM dNTP mix, random primers (100 ng/μl; Takara Bio Inc., Japan), 5X First-Strand buffer, 0.1 M DTT, and RNase H (2 units). Based on our previous report on full-length cDNA sequences [17, 20], gene specific primers for chub mackerel kiss1 kiss2 (GenBank accession number: GU731672 and GU731673), gnrh1 gnrh2, and gnrh3 (GenBank accession number: HQ108193, HQ108194, and HQ108195) were designed from the open reading frame region of each gene using GENETYX software (Table 2) and validated with RT-PCR and agarose gel electrophoresis. Amplification of the beta (β)-actin (GenBank accession number: GU731674) was used as the endogenous reference gene to correct for differences in reverse transcription efficiency and template quantity. The qRT-PCR was performed using the Brilliant II Fast SYBR Green QPCR Master Mix (Stratagene), following the manufacturer’s protocol. The thermocycling conditions were set as 95°C for 5 min and 40 cycles of 95°C for 10 sec and 60°C for 30 sec. Dissociation curve analysis was also included; one cycle of 95°C for 1 min, 55°C for 30 sec, and 95°C for 30 sec. All transcripts were quantified using a standard curve method [24] and a previously validated qRT-PCR for kiss gnrh, and β-actin mRNAs [17, 20]. The PCR reaction mixture (20 μl) contained 1X Brilliant II Fast SYBR Green QPCR Master Mix, 0.1 μM each of forward and reverse primer, and 1 μl cDNA sample. For negative control, cDNA sample was replaced with autoclaved distilled water. Duplicate reactions were performed for the standards, target and reference genes, from 5–6 fish collected per ovarian stage. The amounts of target and endogenous reference gene in experimental samples were determined from the respective standard curves using MxPrO QPCR Software. Transcript levels of kiss and gnrh mRNAs were normalized to the levels of β-actin; the data are expressed as relative mRNA levels. Based on two qRT-PCR assays, the intra- and interassay coefficients of variation (CV) for kiss and gnrh mRNAs were less than 8%. All qRT-PCR assays were conducted where practically possible according to the MIQE (Minimum Information for Publication of qRT-PCR experiments) guidelines by Bustin et al. [25].

Brain and pituitary extracts were prepared following the protocol described earlier [26, 27]. Brain and pituitary GnRH peptide levels were measured using a previously developed time-resolved fluoroimmunoassay (TR-FIA) system to quantify levels of GnRH1 (sbGnRH form), GnRH2 (cGnRH-II form), and GnRH3 (sGnRH form) in tissue extracts [28, 29]. Parallelism between the typical standard curves of each GnRH peptide and the corresponding competition curves of sample extracts of chub mackerel was confirmed with serially two-fold-diluted standards and sample extracts in TR-FIA assay buffer [20]. The detection range, minimum detectable limit, and cross reactivity data are presented in our recent publication [20]. The intra- and interassay CV values of TR-FIA for GnRH1 were 9.0% and 19.6%, those for GnRH2 were 7.5% and 5.8%, and those for GnRH3 were 7.4% and 10.3%. GnRH peptide levels in the brain and pituitary samples are expressed as ng/mg tissue and ng/pituitary, respectively.

All data are represented as the mean ± standard error of the mean. Changes in the levels of kiss and gnrh mRNAs in the brain and GnRH peptides in the brain and pituitary during different ovarian stages were analyzed by one-way ANOVA, followed by Tukey’s multiple comparison test. p<0.05 was considered significant and different letters in figures represent significant differences between different ovarian stages. All analyses were conducted in GraphPad Prism4.

The histological images of different ovarian stages analyzed in the present study are presented in Figure 1. LV stage oocytes (Figure 1A) were characterized by the presence of yolk globules around centrally located germinal vesicle of first clutch late vitellogenic oocytes. In the GVM stage oocytes (Figure 1B), GV migration to the animal pole was observed with one or two continuous masses of yolk in the central region of the oocyte. HY stage oocytes (Figure 1C) were transparent and enlarged after germinal vesicle breakdown. OV stage (Figure 1D) was characterized by the presence of freshly ovulated eggs in the ovarian cavity. POV stage (Figure 1E) showed the presence of 6- to 8-h old post-ovulatory follicles (POFs) in the ovarian tissue, characterized by hypertrophied follicular granulosa cells [30].

The levels of kiss1 mRNA significantly increased from the LV stage to GVM (P<0.001); declined during HY and then increased during the OV and POV (P<0.001 for OV and P<0.05 for POV) periods (Figure 2A). Similar to kiss1, kiss2 mRNA levels significantly increased during the GVM stage (P<0.001). However, kiss2 mRNA levels declined during the OV and POV periods (P<0.01 for OV and P<0.001 for POV; Figure 2B).

The levels of gnrh1 mRNA showed significant increase during the POV period (P<0.05; Figure 3A). However, gnrh2 and gnrh3 mRNA levels significantly increased during the GVM period (P<0.001) and then decreased during HY period (P<0.001). Again, their levels increased during OV period and then decreased during POV period (Figure 3B,C).

GnRH1, GnRH2, and GnRH3 peptide levels in the brain were significantly elevated during the GVM (P<0.05) and HY (P<0.001) periods (Figure 4A,B,C). GnRH1 levels significantly declined during OV and POV periods (P<0.01; Figure 4A); GnRH2 levels were low during OV and POV periods but did show any significant differences with HY levels (Figure 4B); GnRH3 levels significantly declined during OV (P<0.05) and were low during POV periods (Figure 4C).

Pituitary peptide levels of all three GnRH forms did not show any significant differences among different ovarian stages (See Additional file 1: Figure S1).