Members of the crustacean hyperglycemic hormone (CHH) peptide family are differentially distributed both between and within the neuroendocrine organs of Cancer crabs: implications for differential release and pleiotropic function

Research on crustaceans has played a major role in advancing our understanding of neuroendocrine structure and control. Nearly 80 years ago,Koller (Koller, 1925; Koller, 1927) demonstrated that hemolymph-borne agents were responsible for color change in shrimp. In the decade that followed, Hanström(Hanström, 1931; Hanström, 1933; Hanström, 1935; Hanström, 1937)identified a structure located in the eyestalk, the sinus gland (SG) as the source of these substances. Later, Bliss(Bliss, 1951) and Passano(Passano, 1951) independently demonstrated the neurohemal nature of the SG, a first for any neural structure in the animal kingdom. In recent years, work from many laboratories has shown that the SG is one of the major neuroendocrine organs in crustaceans and a number of hormones contained within it have been identified, including a group of structurally related peptides that have collectively been termed the crustacean hyperglycemic hormone (CHH) family.

In crustaceans, the CHH family includes CHHs, moult-inhibiting hormones(MIHs), gonad-inhibiting hormones (GIHs), vitellogenesis-inhibiting hormones(VIHs) and mandibular organ-inhibiting hormones (MOIHs), all of which are large peptides (72-78 amino acids) that share considerable amino acid sequence and structural homology, including six conserved cysteine residues and three disulfide bridges (Keller,1992; de Kleijn and van Herp,1995; Soyez, 1997; van Herp, 1998; Webster, 1998; Chan et al., 2003; Chen et al., 2005; Fanjul-Moles, 2006). As their diverse names imply, members of the CHH family have been shown to influence a broad array of physiological processes including, but not limited to, the regulation of metabolism, somatic growth and reproductive maturation(Sedlmeier and Keller, 1981; Sedlmeier, 1982; Sedlmeier, 1988; Keller, 1992; Charmantier-Daures et al.,1994; Wainwright et al.,1996; Santos et al.,1997; Khayat et al.,1998; van Herp,1998; Webster,1998; Chung et al.,1999; Spanings-Pierrot et al.,2000; Sonobe et al.,2001; Chan et al.,2003; Serrano et al.,2003).

Based on preprohormone organization, the CHH family can be divided into two sub-groups: (1) CHH and (2) MIH/GIH/VIH/MOIH(de Kleijn and van Herp, 1995; Lacombe et al., 1999; Chan et al., 2003; Chen et al., 2005; Fanjul-Moles, 2006). The preprohormones of the CHH subgroup always contain a second peptide, commonly referred to as CHH precursor-related peptide or CPRP, which is C-terminally flanked by the CHH (Weidemann et al.,1989; Tensen et al.,1991; de Kleijn et al.,1994a; de Kleijn et al.,1995; Ohira et al.,1997a; Davey et al.,2000; Toullec et al.,2002; Marco et al.,2003; Chen et al.,2004; Mettulio et al.,2004; de la Iglesia et al.,2005; Hsu et al.,2005a; Toullec et al.,2006). The preprohormones of the MIH/GIH/VIH/MOIH sub-group lack precursor-related peptides (Klein et al.,1993; de Kleijn et al.,1994b; Sun, 1994; Lee et al., 1995; Ohira et al., 1997b; Chan et al., 1998; Gu and Chan, 1998; Umphrey et al., 1998; Tang et al., 1999; Lu et al., 2001; Yang and Rao, 2001; Edomi et al., 2002; Krungkasem et al., 2002; Ohira et al., 2005). Although CPRP has been shown to circulate in the hemolymph, the physiological role(s)it plays in crustaceans remains unknown(Wilcockson et al., 2002).

In brachyuran crabs, particularly those of the genus Cancer, much work has been done in an effort to identify and characterize the actions of the native CHH-family members and CPRPs in the X-organ (XO)-SG system. In this genus, four distinct CHH isoforms have been identified [two from Cancer pagurus and two from Cancer productus(Chung et al., 1998; de la Iglesia et al., 2005; Hsu et al., 2005a); Table 1] as have two isoforms of MIH [one from Cancer magister and the other from C. pagurus (Chung et al.,1996; Umphrey et al.,1998); Table 1] and two isoforms of MOIH [both from C. pagurus(Wainwright et al., 1996); Table 1]. Similarly, nine isoforms of CPRP have been characterized from the SG of Cancerspecies [one from C. pagurus and four each from C. productusand Cancer borealis (Chung et al., 1998; de la Iglesia et al., 2005; Fu et al.,2005a; Fu et al.,2005b; Hsu et al.,2005a); Table 1]. As Table 1 shows, the isoforms of any given group are nearly identical in amino acid sequence, even when comparing isoforms from different species.

Although most studies concerning CHH-family members and CPRPs have focused on the SG, there is evidence that some of these peptides are also present in other regions of the crustacean nervous system(de Kleijn et al., 1995; Sun, 1995; Chang et al., 1999; Dircksen et al., 2001; Lu et al., 2001; Gu et al., 2002; Wilcockson et al., 2002; Basu and Kravitz, 2003; Chen et al., 2004; Fu et al., 2005b; Ohira et al., 2005; Toullec et al., 2006). CHH-and CPRP-like immunoreactivities have been found in the neuroendocrine pericardial organ (PO) of C. pagurus(Wilcockson et al., 2002) and putative CPRP fragments have been sequenced from this structure in C. productus (Fu et al.,2005b). The functional relevance of this dual XO-SG/PO distribution of CHH in these crabs is currently unknown. However, it has been postulated that the signals governing its release from each site are distinct. Moreover, the isoforms present in one versus the other structure may also be distinct (Dircksen et al.,2001; Toullec et al.,2006) and subserving different physiological roles, as has been shown for the CHHs isolated from the SG and PO of the crab Carcinus maenas (Dircksen et al.,2001).

Recently, we began a study to identify and compare the hormone complements present in the neuroendocrine organs of several Cancer species(Christie et al., 2004; Christie and Messinger, 2005; de la Iglesia et al., 2005; Fu et al., 2005a; Fu et al., 2005c; Hsu et al., 2005a; Hsu et al., 2005b; Messinger et al., 2005a; Messinger et al., 2005b; Cruz-Bermúdez et al.,2006). As a first step toward determining the identity, tissue distributions and physiological roles played by each CHH-family member and CPRP in the neuroendocrine systems of these species, we undertook the immunohistochemical mapping studies presented in this report. Specifically,antisera previously generated against native C. pagurus peptides(Webster, 1996; Wilcockson et al., 2002) were used to map the distributions and co-localization patterns of the CHH family and CPRP in the neuroendocrine organs of Cancer antennarius, Cancer anthonyi, C. borealis, Cancer gracilis, Cancer irroratus, C. magister and C. productus. As the data that follow show, most members of the CHH family and CPRP were immunologically detected in multiple neuroendocrine organs, with the complement of peptides present in each site conserved across species. In at least C. productus, the peptides detected in the XO-SG and PO were also differentially distributed among the neurons comprising each site. Collectively, the data presented here represent the most extensive and complete immunohistochemical survey of CHH family members and CPRP in crustacean neuroendocrine organs to date. Moreover, our findings support the hypotheses that: (1) members of CHH family and CPRP are released in response to varied, tissue-specific cues and (2) the isoforms released from non-XO-SG sites probably possess functions distinct from those ascribed to them when secreted from the SG. Some of these data have appeared previously in abstract form (Hsu et al., 2004).

Brown rock crabs Cancer antennarius Stimpson and yellow rock crabs Cancer anthonyi Rathbun were purchased from Coastal Catch (Santa Barbara, California, USA). Jonah crabs Cancer borealis Stimpson were purchased from the Marine Biological Laboratory (Woods Hole, Massachusetts,USA). Graceful crabs Cancer gracilis Dana were collected by hand at False Bay, San Juan Island, Washington, USA. Atlantic rock crabs Cancer irroratus Say were purchased from J&A Seafood (Brunswick, Maine,USA). Dungeness crabs Cancer magister Dana and red rock crabs Cancer productus Randall were collected by hand, trap or trawl at multiple locations throughout the San Juan Archipelago and greater Puget Sound areas of Washington State, USA. Regardless of species, animals were maintained in either aerated natural seawater aquaria chilled to 10°C (Department of Biology, University of Washington, Seattle, Washington, USA) or in flow-through natural seawater tables [Friday Harbor Laboratories, Friday Harbor, Washington, USA (locally caught animals only); ambient water temperature was approximately 8-12°C, depending on season].

For the collection of tissue, crabs were anesthetized by packing in ice for 30-60 min. After anesthesia, the eyestalks were isolated, then the dorsal carapace was removed and the foregut and the lateral walls of the pericardial chamber were dissected free in chilled (approximately 10°C) physiological saline [composition: 440 mmol l^-1 NaCl; 11 mmol l^-1 KCl;13 mmol l^-1 CaCl₂; 26 mmol l^-1MgCl₂; 10 mmol l^-1 Hepes; pH 7.4 (adjusted with NaOH)]. To obtain the XO-SG system, the carapace encasing an eyestalk was split both dorsally and ventrally and one half of the split shell was gently teased away from the other half. The remaining half of the eyestalk was then pinned in a wax-lined Pyrex dish filled with chilled physiological saline and the optic ganglia containing the XO-SG system were subsequently isolated. POs were obtained by pinning the isolated walls of the pericardial chamber in a wax-lined Pyrex dish containing chilled physiological saline. The nerve roots forming each PO were then dissected free from the muscles and connective tissues of the pericardial wall. For isolating the stomatogastric nervous system (STNS), which contains two recently identified neuroendocrine organs(Christie et al., 2004; Messinger et al., 2005a), the anterior cardiac plexus (ACP) and the anterior commissural organ (ACO), the foregut was flattened by making a longitudinal cut from the esophagus to the pylorus on its ventral side followed by a pair of medial cuts directed along the ossicles of the cardiac sac/gastric mill. After its opening, the teeth of the gastric mill were removed and the flattened foregut was pinned, inside down, in a wax-lined Pyrex dish containing chilled physiological saline. The STNS was then dissected free from the muscles and connective tissues of the foregut.

All of the primary antibodies used in our study were rabbit polyclonal antibodies generated against native C. pagurus peptides(Table 1). These peptides were purified from a common pool of approximately 2000 SGs as described in several previous publications (Chung et al.,1996; Wainwright et al.,1996; Webster,1996; Chung et al.,1998). For production of the CHH antibody, native C. pagurus (Capa)-CHH II was used. The development of this antibody is described elsewhere (Webster,1996). For the production of the CPRP antibody, the antigen was C. pagurus (Capa)-CPRP conjugated to bovine thyroglobulin using either glutaraldehyde or 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide. The production of this antibody is also described elsewhere(Wilcockson et al., 2002). For the production of the MIH and MOIH antibodies, methods similar to those employed for the production of the CHH antibody were used. Specifically, New Zealand white rabbits (one for each peptide) were injected subcutaneously at multiple sites with 5 nmol of native C. pagurus (Capa)-MIH or native C. pagurus (Capa)-MOIH I, each dissolved in 0.3 ml of phosphate-buffered saline and emulsified with an equal volume of complete Freund's adjuvant. Four weeks after the initial immunization, a booster of 5 nmol of peptide (Capa-MIH or Capa-MOIH I) in Freund's incomplete adjuvant was injected into each rabbit. Similarly, 8 weeks after the initial immunization,rabbits were boosted again, this time with 3 nmol of peptide (Capa-MIH or Capa-MOIH I) in Freund's incomplete adjuvant. Twelve weeks after the initial immunization, rabbits were terminally exsanguinated under anesthesia and serum was collected from the retracted clots. Production of all antisera used in this study was done at the University of Wales Bangor (Bangor, UK).

The secondary and tertiary antibodies used in our study were kind gifts from either Jackson ImmunoResearch Laboratories, Inc. (West Grove,Pennsylvania, USA) or Molecular Probes (Eugene, Oregon, USA). These included donkey anti-rabbit IgG labeled with Alexa Fluor 488 (Molecular Probes; catalog#A-21206), Alexa Fluor 594 (Molecular Probes; catalog #A-21207), FITC (Jackson ImmunoResearch; catalog # 711-095-152) or Rhodamine Red X (Jackson ImmunoResearch; catalog # 711-295-152); goat anti-rabbit IgG labeled with Alexa Fluor 488 (Molecular Probes; catalog # A11008) or Texas Red (Jackson ImmunoResearch; catalog #111-075-144); unlabeled monovalent goat anti-rabbit IgG antigen-binding (Fab) fragments (Jackson ImmunoResearch; catalog #111-007-003) or donkey anti-goat IgG labeled with FITC (Jackson ImmunoResearch, catalog # 705-095-147).

For single labeling, tissue was fixed overnight in a freshly made solution of 4% paraformaldehyde (EM grade; Electron Microscopy Sciences, Hatfield,Pennsylvania, USA; catalog #15710) in 0.1 mol l^-1 sodium phosphate buffer (pH 7.4) followed by five rinses (at 1 h intervals) in a solution of sodium phosphate buffer containing 0.3% Triton X-100 (P-Triton). After rinsing, tissue was incubated with gentle agitation in a 1:5000 dilution of primary antibody for approximately 72 h. Primary antiserum was diluted in P-Triton, with 10% normal donkey serum (NDS; Jackson ImmunoResearch; catalog#017-000-121) or normal goat serum (NGS; Jackson ImmunoResearch; catalog#005-000-121) added to diminish nonspecific binding. Following incubation in primary antibody, tissue was again rinsed five times at 1 h intervals in P-Triton and then incubated overnight in a 1:300 dilution of secondary antibody. As with the primary antibody, secondary antibody incubation was done in P-Triton with 10% NDS or NGS, using gentle agitation. After secondary antibody incubation, each preparation was rinsed five times over approximately 5 h in sodium phosphate buffer and then mounted between a glass microscope slide and coverslip using Vectashield mounting medium (Vector Laboratories,Inc., Burlingame, California, USA; catalog # H-1000). Fixation and incubation in both primary and secondary antibody was done at 4°C. All rinses were done at room temperature (approximately 20°C) without agitation. Secondary antibody incubation, and all subsequent processing, was conducted in the dark. Likewise, slides were stored in the dark at 4°C until examined.

To increase our confidence that the immunoreactivities reported here were specific for peptides related to the antigen used for the production of each antiserum, we conducted a series of adsorption controls. In these experiments,native Capa-CHH II, Capa-CPRP, Capa-MIH and Capa-MOIH I isolated from the SG of C. pagurus (Table 1) were used as blocking agents. It should be noted that due to a limited supply of these peptides, adsorption controls were only conducted on C. productus tissues and that the adsorptions used for labeling one preparation were sometimes recycled for use in blocking studies on other samples. In each adsorption experiment, the antibody was incubated with a blocking peptide (approximately 10^-5 mol l^-1) for 2 h at room temperature prior to applying the solution to the tissue. For each tissue in which an antibody produced labeling, a complete block of staining was achieved only when the antiserum was adsorbed with the peptide used for its production (three preparations for each peptide and antibody in each tissue;data not shown). Adsorption of an antibody with the other peptides had no appreciable effect on immunolabeling (three preparations for each peptide and antibody in each tissue; data not shown).

For preparations that were double-labeled (C. productus tissues only), the staining protocol for single labeling was modified in order to permit the use of primary antibody pairs that were both developed in the same host species (rabbit in this case). In brief, POs and SGs were dissected,fixed and incubated in the first primary antibody as described above for 1 or 2 days, respectively. Following incubation in the first primary antibody,tissues were rinsed with P-Triton as described for single labeling, then the SGs were incubated overnight in a 1:25 dilution of monovalent goat anti-rabbit IgG Fab fragments that also contained 10% NDS whereas the POs were incubated for 2 days in a 1:10 dilution of goat anti-rabbit IgG Fab fragments containing 10% NDS. For both tissues, this incubation with Fab fragments was done to sterically cover and immunologically convert the first primary antibody from rabbit to goat. After incubation with the Fab fragments, tissues were rinsed every hour for 5 h and then incubated overnight in a 1:300 dilution of FITC-conjugated donkey anti-goat IgG containing 10% NDS. Following this overnight incubation, tissues were rinsed as before then incubated in the second primary antibody for 1 (PO) or 2 (SG) days. After incubation in the second primary antibody, tissues were rinsed and then incubated overnight in a 1:300 dilution of Rhodamine Red X-conjugated donkey anti-rabbit IgG containing 10% NDS. All subsequent processing was the same as for the single labeling described above.

To assess the veracity of our double-labeling protocol, we conducted controls to determine the extent to which the Fab fragments were capable of sterically covering, and hence masking, the first primary antibody from detection via the second secondary antibody. Specifically, a single primary was applied to a tissue, sterically converted to goat viaincubation in Fab fragments and then incubated in the Rhodamine Red X-conjugated donkey anti-rabbit IgG. For all antibodies and all tissues in which double labeling was undertaken (3 preparations for each antibody in each tissue), a complete masking of the primary antibody was achieved, and hence little or no cross-talk between the first and second primary antibody sets should have occurred in our double-labeled preparations (data not shown).

After immunoprocessing, preparations were viewed and data were collected using one of two Bio-Rad MRC 600 laser scanning confocal microscopes (Bio-Rad Microscience Ltd., Hemel Hempstead, UK), a Bio-Rad Radiance 2000 laser scanning confocal microscope or a Nikon Eclipse E600 epifluorescence microscope. Descriptions of the hardware and software used for imaging on these systems are extensively documented in previous publications(Christie et al., 1997; Christie et al., 2003; Christie et al., 2004; Messinger et al., 2005a).

For the production of figures, Bio-Rad.pic files collected using the MRC 600 or Radiance 2000 systems were converted to tif files using ImageJ (available free of charge from the National Institutes of Health at http://rsb.info.nih.gov/ij/). Individual micrographs and composite figures were produced from the tif files using a combination of ImageJ and Photoshop (version 7.0;Adobe Systems Inc., San Jose, California, USA). Schematic diagrams were produced using Canvas (version 8.0; Deneba Systems Inc., Miami, Florida, USA). It should be noted that the contrast and brightness of final figures were adjusted as needed to optimize the clarity of printed images.

As in other brachyurans, the XO-SG system of Cancer species is located within the eyestalk (Figs 1, 2), with the XO (a loosely associated cluster of somata) situated in the proximoventral corner of the medulla terminalis and the SG (the neurosecretory terminals derived from the XO somata) present at the junction of the medulla interna and medulla externa(Cooke and Sullivan, 1982; Beltz, 1988; Fingerman, 1992). As stated earlier, isoforms of all members of the CHH family (i.e. CHH, MIH and MOIH)have been isolated from the XO-SG system of C. pagurus, as has an isoform of CPRP (Chung et al.,1996; Wainwright et al.,1996; Chung et al.,1998). Using antibodies generated against C. paguruspeptides (Webster et al., 1996; Wilcockson et al., 2002; this study), we immunolabeled eyestalks from C. antennarius, C. anthonyi, C. borealis, C. gracilis, C. irroratus, C. magister and C. productus to determine if the XO-SG system of these species also contains members of these peptide groups.

For each antibody, intense labeling was evident in nerve terminals in the SG (Figs 3, 4) as well as in approximately three-dozen somata in the XO (Fig. 3; MIH labeling shown) in each species (N≥4 preparations for each antibody in each species). Within the SG, the nerve terminals labeled by the CHH and CPRP antisera(Fig. 4A,B) had a tendril-like appearance, whereas the terminals labeled by the MIH and MOIH antisera(Fig. 4C,D) appeared more oval and blob-like. Regardless of antibody or species, all labeling in the SG could be traced back to the immunopositive XO somata via labeled axons in the sinus gland tract (sgt), suggesting that these somata are the sole source of the SG staining.

As just described, the SG terminals labeled by the CHH and CPRP antibodies were tendril-like whereas those labeled by the MIH and MOIH antibodies were distinctly blob-like in appearance in all species(Fig. 4). For C. productus, we quantified and compared the number of somata labeled by each antibody. In this species, the numbers of XO somata labeled by the CHH and CPRP antibodies were essentially identical [anti-CHH: mean=40±1.90 somata/XO, range=36-49, 6 preparations; anti-CPRP: mean=41±2.12 somata/XO, range= 37-47, 4 preparations (Student's t-test, P>0.05)] as were the number of somata labeled by the MIH and MOIH sera [anti-MIH: mean=32±1.38 somata/XO, range=25-36, 8 preparations;anti-MOIH: mean=29±1.92 somata/XO, range= 22-35, 6 preparations(Student's t-test, P>0.05)]. By contrast, the number of somata stained by either the CHH or CPRP antibody was statistically different from those labeled by either the MIH or MOIH antibody (ANOVA, P<0.0005; Tukey's test, P<0.05). Collectively, these findings suggested the potential for CHH/CPRP and MIH/MOIH co-localization in the XO-SG system, and simultaneously implied that the former two peptides were contained in subpopulations of neurons distinct from those producing the latter two peptides. To directly assess the patterns of co-localization present among the CHH family members and CPRP in C. productus, we developed a double-immunohistochemistry method utilizing primary antisera generated in a common host species and conducted double-immunolabeling for all possible combinations of the four antisera using this method.

As Fig. 5 shows,co-localization of CHH and CPRP as well as of MIH and MOIH (shown in Fig. 5A) was seen in XO somata in each of the C. productus eyestalks labeled using these antibody combinations (6 preparations for each combination). In none of the eyestalks labeled with any other combination of primary antisera was any co-localization evident in XO somata (Fig. 5B;6 preparations for each antibody combination). All somata that expressed CHH-like immunoreactivity were co-labeled by the CPRP antibody, a finding that is not surprising given their production from the same preprohormone in C. productus (de la Iglesia et al.,2005; Hsu et al.,2005a). Interestingly, complete co-localization was also seen in the double-labels pairing of the MIH and MOIH antibodies(Fig. 5A), suggesting that all XO somata that produce MIH also produce MOIH.

Another crustacean neuroendocrine system known to contain CHH-related peptides is the PO (Dircksen et al.,2001; Wilcockson et al.,2002; Fu et al.,2005b). In brachyuran species, including C. productus,the PO consists of two or more longitudinal nerve trunks that are connected by vertical nerve bars (Fig. 1). The trunks and bars that form each PO are elaborations of the segmental nerves(sns) that originate from the fused thoracic ganglia(Fig. 1). Both intrinsically and extrinsically located cell bodies contribute to the release terminals that lie under the epineurium that covers the PO(Cooke and Sullivan, 1982; Beltz, 1988; Fingerman, 1992). As stated earlier, both CHH and CPRP have been detected in the PO of C. pagurususing immunohistochemistry (Wilcockson et al., 2002). In this species, all labeling for both peptides originates from three intrinsic somata located in or near the anterior bar of the PO (Wilcockson et al.,2002). To assess if CHH-, CPRP-, MIH- or MOIH-like peptides were present in the POs of other species of Cancer and to determine their distributions and patterns of co-localization in C. productus, we immunoprocessed this tissue with the same set of antisera used in our mapping of these substances in the SG.

Immunoprocessing of POs with the CHH, CPRP and MOIH antisera consistently produced labeling within this tissue in each of the Cancer species investigated (Fig. 6; N≥3 preparations for each antibody in each species). By contrast,no staining was seen when POs of any crab were processed with the MIH antiserum (N≥3 preparations for each species). For both the CHH and CPRP antibodies, labeling in the POs of each species consisted of up to three somata (usually two bipolar and one multipolar) in the anterior bar or in the sns just anterior to it, as well as fine processes and peripherally located nerve terminals covering extensive regions of the anterior and posterior bars and portions of all three nerve trunks(Fig. 6A,B). For both antibodies, the vast majority of the immunopositive fine fibers and nerve terminals appeared to originate from the arborizations of the intrinsic somata. In each of the CHH-immunopositive somata, labeling was cytoplasmic,uniformly diffuse and often extended for a considerable distance into the axons emanating from these cell bodies. By contrast, labeling in the CPRP-immunopositive somata in all species was cytoplasmic and granular, with little immunoreactivity evident in their projecting axons.

In all animals, MOIH-like staining was weak in the axons and in patches of superficially located nerve terminals in both the anterior and posterior bars,as well as in the nerve trunks (Fig. 6D). Whereas no MOIH-like staining was seen in any intrinsic PO somata in any of the species, approximately four MOIH-immunopositive cell bodies were present in the fused thoracic ganglia of C. productus (3 preparations; data not shown). As the thoracic ganglia are known to be the source of many of the inputs to the PO(Cooke and Sullivan, 1982; Beltz, 1988; Fingerman, 1992), these somata are one potential source of the MOIH staining seen in the PO of this species,and possibly the others as well.

In all of the species investigated, the distribution of CHH- and CPRP-like immunoreactivity in the POs was essentially identical with the exception of the appearance of labeling in the three intrinsic somata. To determine the extent of colocalization of the two peptides in C. productus,double-immunolabeling, similar to that in the SG, was undertaken. As Fig. 7 shows, a complete overlap of the two immunoreactivities was found in this tissue (6 preparations). CHH/MOIH (6 preparations) and CPRP/MOIH (6 preparations)double-labelings were also conducted. However, the relative weakness of the MOIH label with respect to either the CHH or CPRP labels made assessment of co-localization difficult. In most POs, the immunoreactivities appeared distinct, whereas in others, weak co-localization in a small population of terminals was evident (data not shown). Given the complete overlap of the CHH and CPRP labels and the fact that they appeared to be derived from a set of somata not labeled by the MOIH antiserum, we believe that the profiles labeled by the MOIH antibody are distinct from those containing CHH/CPRP.

The STNS (Fig. 1) of decapod species is situated over the foregut and is responsible for controlling the movement of food items through this portion of the digestive tract(Selverston and Moulins, 1987; Harris-Warrick et al., 1992). Recently, we identified two neuroendocrine organs within the STNS of C. productus (Christie et al.,2004; Christie and Messinger,2005; Messinger et al.,2005a; Messinger et al.,2005b). One of these, the ACP, is located on the anterior cardiac nerves (acns) overlying the cardiac sac region of the foregut(Christie et al., 2004)(Fig. 1). This site consists of a dense collection of nerve terminals located just below the epineurium covering the acns (Christie et al., 2004), with all innervation to the site originating from four somata, one pair in each commissural ganglion (CoG)(Christie and Messinger, 2005; Messinger et al., 2005b). Based on their innervation of the ACP, these neuron pairs were named anterior commissural neurons 1 and 2 (ACN1/2)(Christie and Messinger, 2005; Messinger et al., 2005b). The other site, the ACO, is located within the anterior medial quadrant of the CoG(Fig. 1) and consists of a dense collection of nerve terminals that envelop an extensive network of hemolymph lacunae that invaginate deep into the ganglion(Messinger et al., 2005a). The origin of the fibers that produce the ACO remain unknown, though somata in the thoracic ganglia are likely candidates(Messinger et al., 2005a). Both the ACP and ACO are also present in the STNSs of C. antennarius, C. anthonyi, C. borealis, C. gracilis, C. irroratus and C. magister (A.E.C., D.I.M. and E. Savage, unpublished observations). To determine if the ACP and/or ACO contain any of the CHH-related or CPRP-like peptides, we immunoprocessed the STNSs from each Cancer species with the same complement of antisera used to map the SG and PO.

In each of the investigated species, no immunoreactivity was seen anywhere in the STNSs processed with the CHH, CPRP or MIH antisera (N≥3 preparations for each antibody in each species). By contrast, in all crabs,extensive labeling was produced throughout using the MOIH antibody(N≥3; Figs 8, 9, 10), including staining of the ACP (Fig. 8) but not the ACO(data not shown).

Regardless of species and in all preparations examined, extensive MOIH-like immunoreactivity was seen in the nerve terminals constituting the ACP(Fig. 8A). In each ACP, the immunopositive terminals originated from four axons projecting from the stomatogastric nerve (stn; Fig. 8B), two traveling from the right superior esophageal nerve(son) and two from the left son(Fig. 8C). Although it was not possible to follow the immunoreactivity in the axons back to their somata of origin, they probably originate from two large (approximately 100 μm in major cross-sectional diameter), weakly immunopositive cell bodies located at the junction of the son and CoG, as these somata are of the same size and location as the ACN1/2s, which are known to be the sole source of innervation to the ACP (Christie et al.,2004; Christie and Messinger,2005; Messinger et al.,2005b).

In addition to the staining just described, in all species the MOIH antiserum also labeled an extensive neuropil in the stomatogastric ganglion that appeared to originate from both the arborizations of two intrinsic immunopositive somata and from the arborizations of projection neurons traveling to the ganglion via the stn(Fig. 9). This, and all other MOIH-like labeling in the STNS, is summarized schematically in Fig. 10.

In this study we used antisera generated against Capa-CHH II, Capa-MIH,Capa-MOIH I and Capa-CPRP to assay the neuroendocrine organs of the crabs C. antennarius, C. anthonyi, C. borealis, C. gracilis, C. irroratus, C. magister and C. productus for the presence of isoforms of these peptides. Our results show that in all species, moieties immunologically related to the peptides used for antibody production are present in the SG, PO and ACP, but not in the ACO. Moreover, the patterns of immunoreactivity revealed by these antisera suggest that the native CHH family members and CPRPs are differentially distributed between the neuroendocrine organs of these species, with all members of the CHH family and CPRP present in the SG;CHH, MOIH and CPRP, but not MIH, present in the PO; and only MOIH present in the ACP.

As are all antibodies, the sera used in our study are biological and not chemical reagents and thus are subject to potential cross-reactions with antigens other than that to which they were raised(Saper and Sawchenko, 2003). As Table 1 shows, significant sequence homology is present between the C. pagurus CHH-family members used for antibody production. Despite the sequence similarity of the antigens, the distinct patterns of labeling seen with each antibody strongly suggest that they are detecting different complements of peptides. Furthermore, in all tissues, only adsorption of an antibody by the peptide used for its production completely eliminated labeling. Thus, we feel confident that the staining presented here is an accurate reflection of the distributions of CHH-family members and CPRPs in the neuroendocrine organs of Cancer species.

In addition to observing a differential distribution of CHH-family peptides and CPRP among the neuroendocrine organs of various Cancer species,we also found that these peptides are differentially distributed within both the XO-SG and PO of at least C. productus. Using a protocol that allowed double labeling of whole-mount tissue with primary antibodies generated in the same host species, we found complete overlap of labeling for CHH and CPRP in both the XO-SG and the PO. Similarly, a complete overlap of labeling was seen for MIH and MOIH in the XO-SG, but in a different set of neurons than those containing CHH and CPRP. MOIH was also present in the PO,again in a set of structures distinct from those possessing CHH and CPRP.

Ours is the first study to show co-localization of CHH- and CPRP-like immunoreactivities in either the XO-SG or the PO, but the association of these two peptides is not surprising given that both peptides are produced from a common preprohormone in C. productus(de la Iglesia et al., 2005; Hsu et al., 2005a). More interesting is the complete overlap of the MIH and MOIH labels in the XO-SG,which are encoded by different genes (Lu et al., 2000). As stated earlier, the peptides used for the production of the MIH and MOIH antibodies exhibit significant amino acid sequence identity (approximately 61%; Table 1), and thus it is possible that each serum was cross-reacting with a common peptide. However, we were unable to completely block the MIH staining in the XO-SG with any peptide other than Capa-MIH, including Capa-MOIH I. Similarly, the XO-SG label produced by the MOIH antibody was suppressed only after adsorption by Capa-MOIH and not by Capa-MIH. These results suggest that the MIH- and MOIH-like labelings were specific for their respective antigens. Clearly, molecular and/or biochemical characterization of the complement of native C. productus CHH-family peptides will be necessary to resolve this issue unambiguously.

Biologically, it is interesting to postulate why MIH and MOIH are present in the same set of XO somata. If MIH and MOIH are contained in the same secretory vesicles, then all signals triggering their release would probably result in the coordinated inhibition of both somatic growth (via the inhibition of steroid production in the Y-organ by MIH) and gonadal growth(via the inhibition of methyl farnesoate production in the mandibular organ by MOIH). If, however, the two peptides are contained in distinct sets of vesicles, then the two peptides might well be released in response to distinct physiological cues. Clearly, the former versus the latter situation would have profound physiological consequences for an animal, and with future studies it will be interesting to determine the physiological significance of the co-localized peptides.

As just discussed, we have found that the CHH, MOIH and CPRP are each present in at least two neuroendocrine sites. With such multiple tissue localizations, the potential clearly exists for distinct cues triggering the release of a given hormone from each site. In addition, distinct isoforms of the CHH, MOIH and CPRP may be present in each tissue, as has been shown for CHH in several crab species (Dircksen et al., 2001; Toullec et al.,2006). In Carcinus maenas, the CHH of the PO, unlike its SG counterpart, does not possess hyperglycemic activity nor does it appear to inhibit ecdysteroid synthesis (Dircksen et al., 2001). Instead, this CHH isoform is proposed to serve either a sensory or local modulatory role in this species(Dircksen et al., 2001). Although unproven, the differential distributions of CHH, CPRP and MOIH revealed in our study suggest the possibility of pleiotropic functions for these peptides in Cancer species (e.g. their classically ascribed functions versus possible roles as locally released neuromodulators). As additional studies are undertaken, it will be interesting to see if this hypothesis is borne out.

In conclusion, the data we present show that members of the CHH family and CPRPs are differentially distributed both between and within the neuroendocrine organs of crabs of the genus Cancer. These results represent the most complete immunohistochemical survey of CHH-family peptides and CPRP in the neuroendocrine organs of crustaceans and provide a foundation for future studies directed at the identification of the native isoforms present in each species and tissue, as well as the elucidation of their physiological roles in members of this genus. The finding of CHH, CPRP and MOIH-like labeling in multiple neuroendocrine tissues suggest that these peptides may be released into the circulatory system in response to varied cues and that some of their isoforms may possess distinct tissue-specific functions.

List of abbreviations
 
• acn

  anterior cardiac nerve

 
• ACO

  anterior commissural organ

 
• ACP

  anterior cardiac plexus

 
• CHH

  crustacean hyperglycemic hormone

 
• CoG

  commissural ganglion

 
• CPRP

  crustacean hyperglycemic hormone precursor-related peptide

 
• GIH

  gonad-inhibiting hormone

 
• MIH

  moult-inhibiting hormone

 
• MOIH

  mandibular organ-inhibiting hormone

 
• PO

  pericardial organ

 
• SG

  sinus gland

 
• sn

  segmental nerve

 
• son

  superior esophageal nerve

 
• STG

  stomatogastric ganglion

 
• stn

  stomatogastric nerve

 
• STNS

  stomatogastric nervous system

 
• VIH

  vitellogenesis-inhibiting hormone

 
• XO-SG

  X-organ-sinus gland system

John Edwards, Lilliam Ambroggio, Rachel Latham, Minhui Lin, Christina Ngo and Trendafil Halatchev are thanked for their help with some of the immunohistochemistry and confocal microscopy. John Weller, Curtis Easton and Christopher Goiney are thanked for their assistance in animal collection. Dr William Stegeman (Jackson ImmunoResearch Laboratories) is thanked for his help in designing and trouble-shooting the double-immunohistochemical studies as well as providing the reagents necessary for conducting them. Y.A.H. gratefully acknowledges financial support from the University of Washington Graduate Opportunities & Minority Advancement Program for Graduate Opportunity Program Fellowship. Some of the animals used here were collected under the auspices of Washington Department of Fish and Wildlife Scientific Collection Permits #03-018a and #04-021 (to A.E.C.).