This study deals with the localization of crustacean hyperglycemic hormone (CHH, Pej-SGPIII) and molt-inhibiting hormone (MIH, Pej-SGP-IV) in the eyestalk of the kuruma prawn Penaeus japonicus using immunohistochemistry. High-titer and highly specific antisera were raised in rabbits against synthetic Pej-SGP-III C-terminal peptide (Glu-Glu-His-Met-Ala-Ala-Met-Gln-Thr-Val-NH2) and Pej-SGP-IV C-terminal peptide (Val-Trp-Ile-Ser-Ile-Leu-Asn-Ala-Gly-Gln-OH), both of which were conjugated with bovine serum albumin by a cross linker. Eyestalks were removed from mature male prawns at the intermolt stage of the molting cycle and fixed in Bouin's solution. Serial sections stained immunohistochemically showed that neurosecretory cells of Pej-SGP-III and Pej-SGP-IV were located in the same cluster of the medulla terminalis ganglionic X-organ (MTGX), and that three kinds of neurosecretory cells, which were stained with anti-PejSGP-III antiserum and/or anti-Pej-SGP-IV antiserum were present. The number of neurosecretory cells which stained with both antisera was much fewer than that of neurosecretory cells which stained with one of the antisera only. The axon and axon terminals in the sinus gland were also stained and the staining density of the sinus gland was always deeper than that of the neurosecretory cells.
INTRODUCTION
The crustacean neurosecretory system, which is comprised principally of the X-organ/sinus gland complex, is located in the eyestalk of decapod crustaceans and produces a range of neuropeptides. These neuropeptides which are known to have important roles in physiological regulation are first synthesized in the X-organ and then transferred to the sinus gland, from where they are released into hemolymph (Keller, 1992).
Among these eyestalk peptides, CHH (crustacean hyper-glycemic hormone) -family peptides have been most extensively studied. This peptide family consists of some structurally similar peptides with two common characteristics in that they contain 72-76 amino acid residues and have similar amino acid sequences with conserved six cysteine residues. These CHH-family peptides include CHH (Leuven et al.,1982; Tensen et al., 1989; Yasuda et al., 1994), MIH (molt-inhibiting hormone) (Chang et al., 1990; Aguilar et al., 1996; Terauchi et al., 1996; Webster and Keller, 1986), VIH (vitellogenesis-inhibiting hormone) or GIH (gonad-inhibiting hormone) (Soyez et al., 1991; Aguilar et al., 1992) and MOIH (mandibular organ-inhibiting hormone) (Wainwright et al., 1996). Possibly due to structural similarity between MIH and CHH, MIH of Homarus americanus also exhibits a low hyperglycemic activity (Chang et al., 1990) and CHH of Carcinus maenas shows very weak MIH activity (Webster and Keller, 1986).
Immunohistochemical studies on CHH-, MIH- and GIH-producing cells have been carried out in several crustacean species. By using antisera against CHH and MIH of C. maenas, CHH- and MIH-producing cells were localized in the eyestalks of C. maenas (Jaros and Keller, 1979), Liocarcinus puber, Cancer pagurus, Uca pugilator and Maja squinado (Dircksen et al., 1988). CHH-producing cells were found in the eyestalks of H. gammarus, Orconectes limosus, Nephrops norvegicus, Palinurus vulgaris, Macropipus puber, Palaemon serratus (Gorgels-Kallen et al., 1982) and the larvae and postlarvae of Astacus leptodactylus (Gorgels-Kallen and Meij, 1985) by using antiserum against the CHH of A. leptodactylus. CHH and VIH-producing cells were observed in the eyestalks of H. americanus by using antiserum against the CHH of A. leptodactylus and antiserum against the VIH of H. americanus (Kallen and Meusy, 1989). CHH- and GIH-producing cells were located by using antisera raised against CHH or GIH of H. americanus (Rotllant et al., 1993). These neurosecretory cells are located in the medulla terminalis ganglionic X-organ (MTGX) and all of the antisera used in these studies were raised against native peptides.
In the kuruma prawn, Penaeus japonicus, six molecular species of CHH (Pej-SGP-I, II, III, V, VI and VII) and a MIH (Pej-SGP-IV) have been found in the eyestalk (Yang, 1997; Yang et al., 1997). The complete amino acid sequences of these peptides have been determined and their bioactivities were investigated by in vivo or in vitro assay. The homologies between any two of CHHs (Pej-SGP-I, II, III, V, VI, and VII) were about 65-99% and the homologies between MIH (PejSGP-IV) and CHHs were about 30%. As in the case of the CHHs of C. maenas and MIH of H. americanus, some of these peptides exhibited a certain degree of overlapping biological activity (Yang et al., 1996, 1997). The similarities in sequence between CHHs and MIH were considerably low, which permitted us to classify these peptides into two types, Type I and II (Yang et al., 1996). However, it is still unknown where these neuropeptides are synthesized.
In the present study, we examined the localization of CHH (Pej-SGP-III) and MIH (Pej-SGP-IV) neurosecretory cells immunohistochemically. We attempted to produce highly specific antisera by using synthetic partial peptides in order to localize the CHH (Pej-SGP-III) and MIH (Pej-SGP-IV) neurosecretory cells.
MATERIALS AND METHODS
Animals
Mature male kuruma prawns, P. japonicus, were obtained from the Momoshima Station of the Japan Sea-Farming Association in Hiroshima Prefecture, Japan. The average body weight of the prawns was about 35 g and the GSI (gonadosomatic index) was about 1. Prawns were sampled during the daytime. Molt stage was determined according to Hong (1977). Prawns at the intermolt stage C were chosen for this experiment.
Eyestalk preparation
Eyestalks were cut from live prawns and fixed for 72 hr in Bouin's solution. Before dehydration, the exoskeleton was removed from the eyestalk. Dehydration, clearing, and embedding in paraffin were carried out by conventional methods.
Extraction and purification of CHH (Pej-SGP-III) and MIH (PejSGP-IV)
The extraction and purification of CHH (Pej-SGP-III) and MIH (Pej-SGP-IV) were done according to the methods reported previously (Yang et al., 1995).
Preparation and purification of complexes of Pej-SGP-III (63-72)-BSA and Pej-SGP-IV (68-77)-BSA
C-terminal amino acid sequences of Pej-SGP-III (Glu-Glu-HisMet-Ala-Ala-Met-Gln-Thr-Val-NH2) and -IV (Val-Trp-Ile-Ser-Ile-LeuAsn-Ala-Gly-Gln-OH) were chosen as antigens. A cysteine residue was added to the N-terminus of each synthetic peptide. The two peptides were synthesized on a peptide synthesizer (Applied Biosystems model 430A). Deprotection and cleavage from the resin were carried out according to the manufacture's instructions. The resulting peptides were purified by reversed-phase HPLC using a Senshu Pak ODS-H-4151 column (10 φ × 150 mm, Senshu Kagaku, Tokyo). The purified peptides were conjugated with a carrier protein, bovine serum albumin (BSA), which had been coupled with N-succinimidyl 3-(2-pyridyldithio)propionate (Pierce). The reaction products were purified on a PD-10 column (Pharmacia Biotech).
Preparation of antisera
About 200 μg each of two synthetic peptide-BSA complexes were injected into two rabbits at every injection. For the first and second injections, the complexes were dissolved in 0.5 ml of phosphate buffer (10 mM, pH 7.2, 0.9% NaCl) and emulsified with 0.5 ml of complete Freund's adjuvant (Difco Laboratory, Michigan, USA). For subsequent injections, the complexes were prepared by the same methods described above except for emulsifying with incomplete Freund's adjuvant (Wako, Osaka). Injections were done once every 2 weeks for 6 months. All the rabbits were immunized by subcutaneous and intra-dermal routes except for the first injection when the samples were injected into the inguinal lymph nodes of the rabbits. Three days after the last injection, the rabbits were exsanguinated under anaesthesia by injecting pentobarbital sodium (Diamabot LTD, USA). The blood was kept at 4°C over night and centrifuged. The antisera were stored at −80°C.
Enzyme-linked immunosorbent assay (ELISA)
The purified Pej-SGP-III (0.98 – 250 ng/ml) and -IV (0.98 – 250 ng/ml) were serially diluted with coating buffer consisting of 0.015 M Na2CO3, 0.35 M NaHCO3 and 0.05% NaN3, and were coated onto 96-well microplates (MaxiSorp−, InterMed Nunc) at 4°C overnight. The plates were washed four times with washing buffer (0.07 M Na2HPO4, 0.04 M KH2PO4, 0.04 M NaCl and 0.05% Tween-20), and blocked with blocking buffer (10 mM Na2HPO4, 0.15 M NaCl and 0.5% gelatin) for 2 hr at room temperature. The plates were then washed again with washing buffer. Anti-Pej-SGP-III (1:100,000 dilution) and anti-Pej-SGP-IV (1:100,000 dilution) antisera were added to the wells and incubated for 2 hr and then the plates were washed as mentioned above. Biotinylated anti-rabbit IgG(H+L) goat IgG (1:1000 dilution, Vector) was applied into wells, which were incubated for 2 hr. The plates were then washed. A macromolecular complex of avidin and biotinylated horseradish peroxidase (ABC Kit, Vector) was added to each well, which was incubated for 2 hr and then washed. Thereafter, 100 μl of 3,3′,5,5′-tetramethylbenzidine (TMB, 0.42 M) dissolved in 0.1 M acetate buffer containing 1.3 mM H2O2 (pH 5.5) was added to each well and the plates were incubated for 10-15 min. The reaction was stopped by the addition of 50 μl of 2 M sulfuric acid. The optical density of each well was measured at 450 nm on a Microplate Reader (Tosoh).
Immunohistochemistry
Serial paraffin sections (10 μm) were deparaffinized and washed with distilled water by conventional procedures. In order to eliminate the effects of endogenous peroxidase, sections were treated with 0.3% H2O2 dissolved in 10 mM phosphate buffer containing 0.9% NaCl (pH 7.2) for 30 min at room temperature. This phosphate buffer was also used for the dilution of antisera and reagents described below. All procedures were done at room temperature except for incubation with the first antisera. After blocking with 10% normal goat serum (NGS) for 30 min, sections were incubated separately with the anti-Pej-SGPIII or -IV antisera (1:100,000 dilution) overnight at 4°C. Biotinylated goat anti-rabbit IgG (H+L) (1:1000 dilution, Vector) was applied to the slides for 1 hr. Subsequently, avidin and biotinylated horseradish peroxidase (ABC Kit, Vector) was added onto the sections and the slides were incubated for 30 min. Finally, 0.025% of 3,3′-diaminobenzidine (DAB, Sigma) containing 0.006% H2O2 was added as a substrate for color development. Excess substrate was removed by washing with distilled water before dehydrating and mounting slides. The staining reaction was stopped at 5 min for anti-Pej-SGP-III antiserum and at 10 min for anti-Pej-SGP-IV antiserum. Between each step, slides were washed twice for 5 min with phosphate buffer.
Examination of specificity of antisera
In order to investigate the specificity of the antisera, anti-PejSGP-III antiserum (1:100,000 dilution) was pre-incubated separately with 1 μg/ml of Pej-SGP-I, II, III, IV, V, VI, or VII at room temperature for 2 hr. Similarly, anti-Pej-SGP-IV antiserum (1:20,000 dilution) was pre-incubated separately with 1 μg/ml each peptide. Subsequently, these antisera were applied to consecutive eyestalk sections for immunohistochemistry.
To check the non-specific binding of the antisera, the primary antiserum was replaced by each of the pre-immuned rabbit sera, anti-BSA rabbit antiserum, and biotinylated anti-rabbit IgG(H+L) goat IgG for immunohistochemistry as described above.
RESULTS
Specificity of antisera
The specificity of antisera raised against Pej-SGP-III and -IV was examined by ELISA. The optical density of the solution after enzymatic reaction containing anti-Pej-SGP-III antiserum increased in a dose-dependent manner when the concentration of purified Pej-SGP-III increased from 0.98 ng/ml to 250 ng/ml. However, the similar response was not obtained even when the concentration of Pej-SGP-IV was increased to 250 ng/ml (Fig. 1A). Similarly, a dose-dependent response to the anti-Pej-SGP-IV antiserum was obtained only with purified Pej-SGP-IV, but not with purified Pej-SGP-III (Fig. 1B).
Fig. 1
Specificities of two antisera as determined by ELISA. (A) and (B) show the results using of anti-Pej-SGP-III and anti-Pej-SGPIV antisera, respectively. Pej-SGP-III is indicated by open circles and Pej-SGP-IV by closed circles.
Anti-Pej-SGP-III and anti-Pej-SGP-IV antisera could be pre-absorbed completely by their respective peptides (Fig. 2B and D) but not by the heterologous peptide (Fig. 3B and D, for Pej-SGP-III and IV, respectively. Data not shown for others). Non-specific binding was not observed when the primary antibody was substituted by pre-immuned rabbit sera, anti-BSA rabbit antiserum or biotinylated anti-rabbit IgG(H+L) goat IgG (data not shown).
Fig. 2
Pre-absorption test of Pej-SGP-III and -IV antisera by using homologous peptides. (A) Pej-SGP-III-immunopositive neurosecretory cells in the MTGX. (B) control to A, an adjacent section stained by antiserum pre-absorbed with Pej-SGP-III. (C) Pej-SGP-IV-immunopositive neurosecretory cells in the MTGX. (D) control to C, an adjacent section stained by antiserum pre-absorbed with Pej-SGP-IV. Scale bar represents 50 μm.
Fig. 3
Pre-absorption test of Pej-SGP-III and -IV antisera by using heterologous peptides. (A) Pej-SGP-III-immunopositive neurosecretory cells in the MTGX. (B) control to A, an adjacent section stained by antiserum pre-absorbed with Pej-SGP-IV. (C) Pej-SGP-IV-immunopositive neurosecretory cells in the MTGX. (D) control to C, an adjacent section stained by antiserum pre-absorbed with Pej-SGP-III. Scale bar represents 50 μm.
Localization of Pej-SGP-III and -IV neurosecretory cells and the sinus gland
The immunoreactive neurosecretory cells of Pej-SGP-III and -IV formed a cluster, which was situated at the side of the medulla terminalis ganglion (MT) near to the rostrum and attached onto the surface of the MT. There were three kinds of immunopositive neurosecretory cells in this cluster that produced Pej-SGP-III, -IV or both (Fig. 4A and B). Most of the Pej-SGP-IV neurosecretory cells existed at the periphery of a cluster of Pej-SGP-III neurosecretory cells and were more adjacent to the MT. By examining the immunopositive neurosecretory cells on every consecutive section, it was revealed that the number of Pej-SGP-III cells was most abundant (40-60) among the three kinds of immunoreactive neurosecretory cells. Pej-SGP-IV cells were less abundant (10-20) than PejSGP-III cells. The number of neurosecretory cells immunore-active to both antisera was much fewer (1-5) than those immunoreactive to one or the other. The diameter of these immunoreactive neurosecretory cells and nuclei were estimated to be about 30-50 μm and 10-11 μm.
Fig. 4
Pej-SGP-III- and Pej-SGP-IV-immunopositive neurosecretory cells. (A) Pej-SGP-III-immunopositive neurosecretory cells in the MTGX. (B) Pej-SGP-IV-immunopositive neurosecretory cells in the MTGX. Arrows indicate Pej-SGP-III (➡), Pej-SGP-IV (➱) and Pej-SGP-III and -IV (➤) neurosecretory cells. Scale bar represents 50 μm.
The sinus gland was located between the MT and the medulla internalis ganglion (MI). Its position was opposite to the immunoreactive neurosecretory cells across the MT and it did not attach to the MT. Axons originating from the neurosecretory cells of Pej-SGP-III and -IV were observed to use the same track, penetrating the MT and extending to the sinus gland (Fig. 5A and B). Axon profiles and terminals of PejSGP-III neurosecretory cells were larger and much more abundant than those of Pej-SGP-IV neurosecretory cells (Fig. 5A-D). The staining density of the sinus gland was always deeper than that of the neurosecretory cells (Fig. 6A-B).
Fig. 5
Axon profiles and terminals of Pej-SGP-III and Pej-SGP-IV neurosecretory cells. (A) Pej-SGP-III axon profile. (B) Pej-SGP-IV axon profile. (C) Pej-SGP-III axon terminals (the sinus gland). (D) Pej-SGP-IV axon terminals (the sinus gland). SG, sinus gland. Scale bar represents 50 μm.
Fig. 6
The staining densities of neurosecretory cells and the sinus gland. (A) Pej-SGP-III neurosecretory cells and the sinus gland. (B) Pej-SGP-IV neurosecretory cells and the sinus gland. NC, neurosecretory cells; SG, sinus gland. Scale bar represents 50 μm.
A composite diagram of the immunopositive neurosecretory cells in the eyestalk is schematically illustrated in Fig. 7.
DISCUSSION
In this experiment, we undertook a strategy to use C-terminal synthetic peptides rather than intact peptides as antigens for the following two reasons. Firstly, the sequence homology among seven peptides, Pej-SGP-I∼VII, ranged from 30% to 99%. Except for the highest homology of 99% between Pej-SGP-V and -VI, the homologies between any two of these peptides were 30-90%. High similarities were found in the center of these molecules, which harbored six cysteine residues, while relatively low similarities were found in the C-terminal part. Thus, the selection of the C-terminus as an antigen was expected to produce highly specific antibodies. Secondly, we considered the putative three-dimensional structures of these peptides. There are six cysteine residues in each peptide, which were found to form three intra-molecular disulfide bonds between Cys7 and Cys43, Cys23 and Cys39, and Cys26 and Cys52 in Pej-SGP-III (Yang et al., 1995). For Pej-SGP-IV, they were presumed to form disulfide bridges between Cys7 and Cys44, Cys24 and Cys40, and Cys27 and Cys53, though there was no experimental evidence. Thus, these peptides would be expected to form a rather constrained three-dimensional structure from Cys7 to Cys52 (Cys53 in Pej-SGP-IV) resulting from the formation of the three disulfide bonds. By contrast, the other parts outside of Cys7 and Cys52 seemed to be rather flexible. The C-terminal part was thought to be more flexible than the the N-terminal part due to the longer peptide chain and, therefore, more suitable for usage as an antigen. To these authors' knowledge, this was the first application of synthetic peptides as antigens to produce antisera in order to study CHH-family peptides.
Immunostaining by anti-Pej-SGP-III antiserum could be abolished completely by pre-incubation of the antiserum with Pej-SGP-III but not with heterologous peptides, Pej-SGP-I, II, IV, V, VI and VII. Similarly, the immunostaining of anti-Pej-SGP-IV antiserum was inhibited by pre-incubation with Pej-SGP-IV but not with other heterologous peptides, Pej-SGP-I, II, III, V, VI and VII. The specificity of the two antisera was satisfactory.
Pej-SGP-III and -IV neurosecretory cells were found in the same cluster and exclusively in the MTGX. The results were similar to those from other crustacean species such as C. maenas, L. puber, C. pagurus, U. pugilator and M. squinadao (Dircksen et al., 1988). All these data showed that neurosecretory cells producing CHH-family peptides were localized in the MTGX. In this experiment, we found many cell clusters centered around the medulla externalis ganglion (ME), MI and MT, and not all of the neurosecretory cells in the MTGX could be stained by the antisera. It could not be ruled out that neurosecretory cells producing the other five peptides (Pej- SGP-I, -II, -V, -VI and VII) may exist in other clusters.
Three kinds of immuno-positive neurosecretory cells, which were stained by either anti-Pej-SGP-III, anti-Pej-SGP-IV antisera or by both, could be observed in the same cluster. Co-localization of two neuropeptides has been reported in previous the studies of CHH- and GIH- producing cells in H. gammarus (Rotllant et al., 1993) and H. americanus (De Kleijn et al., 1992) by immunohistochemistry and in situ hybridization. In contrast, no co-localization of MIH and CHH was observed at either mRNA or peptide levels in C. maenas (Klein et al., 1993). In P. japonicus, there still remains two possibilities: that either the two peptides co-exist in some cells, or alternatively, that these antisera also recognized other peptides which we have not yet been characterized. More detailed examination at the mRNA level will be required in order to obtain definite conclusions.
The diameters of Pej-SGP-III and -IV neurosecretory cells ranged from 30 to 50 μm and those of the nuclei were 10-11 μm. Their sizes were smaller than the CHH neurosecretory cells of the adult lobster H. gammarus (Max: 78 μm; Min: 55 μm) (Gorgels-Kallen et al., 1982) but close to those of C. maenas (cells: 30 by 40 μm; nuclei: 10 μm) (Jaros and Keller, 1979). The number of the Pej-SGP-III neurosecretory cells was more than that of Pej-SGP-IV neurosecretory cells from examination of consecutive sections. It seemed reasonable as we purified the extract of the sinus gland, that Pej-SGP-III was always more abundant than -IV.
The staining density of the sinus gland was deeper than that of the neurosecretory cell bodies, suggesting that the peptide concentration in axon terminals was higher than that in the cell bodies. Moreover, not all of neurosecretory cells had the same staining density, which suggested differing concentrations of peptides in each neurosecretory cell. Various staining densities was probably reflective of the differing synthetic activity of an individual cell. The same phenomena has also been found in other crustacean species (De Kleijn et al., 1992; Dircksen et al., 1988; Gorgels-Kallen et al., 1982; Kallen and Meusy, 1989).
Acknowledgments
We are grateful to Mr. J. Matsumoto, the Momoshima station, Japan Sea-Farming Association for supplying us with prawns. We also thank Dr. M. N. Wilder of Japan International Research Center for Agricultural Sciences for critical reading of this manuscript. This work was supported in part by Grants-in-Aid for Scientific Research (No. 07456054 and No. 07406009) from The Ministry of Education, Science, Sports and Culture of Japan.
