Cell Tissue Res (1991) 266:209 217
Cell and Tissue Research 9 Springer-Verlag1991
Localization of neuropeptides in the nervous system of the marine annelid Sabellastarte magnitica Lucy Diaz-Miranda, Gladys Escalona de Motta, and Jos~ E. Garcia-Arrarfis Department of Biology, Rio Piedras Campus, and Institute of Neurobiology, Medical Sciences Campus, University of Puerto Rico, Rio Piedras, Puerto Rico 00931 Accepted April 29, 1991
Summary. Immunohistochemical studies of the nervous system of Sabellastarte magn~'ca, a sedentary polychaete, showed the presence of neuropeptide expressing cells and fibers within the double ventral nerve cord. Immunoreactivity to cholecystokinin, neuropeptide Y, enkephalins, substance P, and FMRFamide was found to be present in specific populations of cells, identifiable by their location and by the neuropeptide they expressed. Fibers expressing the various neuropeptides were also observed in particular locations within the nerve cord. This characteristic distribution of the various neuron subgroups and fiber pathways may represent functional circuits within the nervous system of this annelid. Key words" Neuropeptides - Annelids Cholecystokinin Neuropeptide Y - Enkephalins - Substance P FMRFamide - Sabellastarte magnifica (Annelida)
Biologically active peptides have emerged as candidates for neurotransmitters and neuromodulators in the nervous system of species in almost every animal phylum. In the annelid nervous system, a diverse number of these peptides has been demonstrated in specific populations of cells. These include molecules related to: cholecystokinin (CCK), enkephalins (ENK), endorphins, substance P (SP), neuropeptide Y (NPY), neurotensin, somatostatin (SS), vasoactive intestinal polypeptide, bombesin, and F M R F amide (FMRFa). The distribution of many of these peptides has been well studied in two classes of annelids: Hirudinea (leeches) and Oligochaeta (earthworms). Immunoreactivity related to peptides such as CCK, SP, FMRFa, enkephalins, vasoactive intestinal polypeptide, somatostatin and bombesin has been demonstrated in the nervous system of the leech Hirudo medicinalis (Zipser 1980; Osborne et al. 1982; Kuhlman et al. 1985a, b; Leake et al. 1986; Norris and Calabrese 1987). Offprint requests to: L. Diaz-Miranda
In addition, in the brain of the leech Theromyzon tessulaturn substances related to opioid peptides were reported (Verger-Bocquet et al. 1987). Similarly, in the nervous system of the earthworm Lumbricus terrestris the presence of immunoreactivity related to NPY, SP, peptide tyrosine, peptide histidine isoleucine, glucagon, gastrinreleasing peptide, calcitonin gene-related peptide, pancreatic polypeptide, vasoactive intestinal polypeptide, methionine-enkephalin and adrenocorticotropic hormone, has been demonstrated (Sundler et al. 1977; Alumets etal. 1979; Aros etal. 1980; Rzasa etal. 1984; Gesser and Larsson 1986; Curry et al. J989). Some of the peptides exert pronounced effects on the muscular activity of the annelids. For example, in the leech Hirudo medicinalis, FMRFa has been implicated in the control of heart beat and modulation of the contractility of the body-wall musculature (Kuhlman et al. 1985b; Norris and Calabrese 1987). In the intestine of the earthworm Lumbricus terrestris, SP has been shown to increase the frequency and amplitude of phasic contractions (Kaloustian and Edmans 1986). While considerable evidence has accumulated on the presence of neuropeptides in members of the Hirudinea and Oligochaeta, little is known about their presence and distribution in members of the third class of this phylum, the Polychaeta. Immunoreactivity similar to that of neurotensin, somatostatin, SP and CCK has been shown in the nervous system of two species of errant polychaetes, Nereis fucata and Nereis diversicolor (Carraway etal. 1982; Engelhardt et al. 1982; DhainautCourtois et al. 1985 a, b; Porchet and Dhainaut-Courtois 1988). Even less is known regarding peptides present within the sedentary species of this class. In view of the important functions and widespread occurrence of peptides in the annelid nervous system, we have investigated the cellular localization of several of these peptides in the nerve cord of the sedentary polychaete Sabellastarte magnifica using immunocytochemical techniques. Our results extend the available information on the neuroanatomical organization of the annelid nervous system and provide the histological basis for
210
ongoing physiological and pharmacological studies in our laboratory. Materials and methods Adult specimens of S. magnifica, up to 11 cm in length, were collected in coral reefs off the east coast of Puerto Rico and maintained at the facilities of the Institute of Neurobiology of the University of Puerto Rico in San Juan. The facilities include 20 1 sea water-aquaria where the animals were kept until used.
Immunohistochemistry Histochemical studies were carried out on isolated lengths of the ventral body wall that contains the paired nerve cords, and a portion of the ventral longitudinal and circular muscles, from a total of 10 animals. The worm was removed from its tube and anesthesized in 7.5% M g C I ~ - d H 2 0 at room temperature for 2 h. It was then decapitated with a blade knife, pinned with its dorsal side up in a Petri dish and covered with the M g C 1 2 - d H 2 0 solution. The anterior part of the ventral trunk, containing portions of the double ventral nerve cord and the ventral longitudinal and circular muscles, was dissected, pinned in a Sylgard-coated Petri dish and fixed in picric acid-formaldehyde mixture (Zamboni and De Martino 1967), p H 7.4, at 4 ~ for approximately 24 h. The tissues were then dehydrated in an ascending series of ethanol solutions to xylene, rehydrated to H 2 0 and placed overnight at 4 ~ C in 30%
T a b l e 1. Details of antisera. All antisera
were polyclonal antisera produced in rabbits
sucrose-0.1 M phosphate-buffered saline (PBS) as described by Keast and colleagues (1984). Tissues were embedded in O.C.T. compound (Miles Inc.) for sectioning on a cryostat at - 2 0 ~ C. Ten-gin transverse sections were mounted on gelatin-chrome alumcoated slides and dried with cool air for I h. For immunohistochemistry, the indirect immunofluorescence method of Coons et al. (1955) was used. Tissue sections were first incubated in 2 - 4 % nonimmune bovine or goat serum for 30 rain, washed in 0.1 M PBS, and incubated overnight in a humid atmosphere with antisera to the different substances diluted in 0.1% Na azide 0.1 M PBS. Antisera against the neuropeptides F M R F a , SP, CCK, NPY and E N K were used (see Table 1 for details). The following day, after three 15-min washes in 0.1 M PBS, the sections were incubated for 1 h with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit immunoglobulin (TAGO) diluted 1:50 in 0.1% Na azide 0.1 M PBS. Slides were rinsed three times (20 rain/rinse) in 0.1 M PBS, mounted with buffered glycerol pH = 8.6, and examined in a Nikon epifluorescence microscope equipped with a V filter (IF 395-425 nm excitation) for the detection of the FITC molecule. Controls were carried out by substituting 0.1 M PBS for the primary antibody and by incubation of tissue sections with the primary antiserum that had been preabsorbed with an excess of its specific antigen, or with peptides with similar amino acid sequences. Cells stained by the specific antisera in each section were counted and classified by localization of the cell body, morphology and size. Only cells with visible nuclei were counted. The size of cells was determined by measuring the length and width of the cell body with an ocular micrometer. The number of labelled cells was determined both at the commissure (i.e., the area of transverse
Antigen
Code
Working dilution
Source
Molluscan cardioexcitatory neuropeptide: F M R F amide (FMRFa)
Dockray
1:250
Dockray et al. (1983)
CRB
1:1000
Schot JGA
1 : 1000 1 : 500
Cambridge Research Biochemicals, Wilmington, Del., USA see Reuter et al. (1986) Garcia-Arrarfis, University of Puerto Rico, Rio Piedras, Puerto Rico (in preparation)
no. 3 2i 2s
1 : 1500
Lugo-Chinchilla (1989)
no. 4 2i 2s no. 4 3i2s
1 : 1000 1:1000
Lugo-Chinchilla (1989) Lugo-Chinchilla (1989)
no. 1 3i 2s
1 : 1000
no. 2 3i 2s
1 : 1000
Garcia-Arrarfis et al. (1991) Garcia-Arrarfis et al. (1991) Rehfeld (1981)
Porcine neuropeptide Y (NPY)
Cholecystokinin fragment 26-33 (CCK)
no. 4562
1 : 1500
Leucine-enkephalin (Leu-ENK)
JO
I : 100
Dr. J. Oliver Flinders Med. Cr. Bedford Park, S. Australia (unpublished)
Methionine-enkephalin (Met-ENK)
Tramu
1:100
Verger-Bocquet et al. (1987)
Substance P (SP)
no. 101
1:1000
no. 102
1 : 1000
Fontaine-P6rus et al. (1985) Fontaine-P&us et al. (1985)
211 connections between the cords) and in the region between the commissures by counting the cells present in the double ventral cord of at least one section from each of three different specimens. Comparisons were made between the numbers of cells expressing each neuropeptide in the commissural and non-commissural areas. Comparisons were also made between the number of cells expressing each specific neuropeptide. Statistical analysis consisted of nonparametric tests: Mann-Whitney U Test Statistic and Kruskal-Wallis One-Way analysis of variance.
processes. Due to the large number of fibers and varicosities, it was not possible to detect branching of neuronal processes. Processes were also observed within the longitudinal muscle layer of the body wall, either as nerves running in between the muscle bundles or as more discrete fibers in apposition with muscle cells. Elongated cell bodies are also observed outside the cords, within the ventral segmental nerves.
Results
Immunohistochemistry
Organization of nervous system
Histochemical studies revealed the presence-of immunoreactivity to CCK, SP, F M R F a , N P Y and ENK. Each substance presented a particular distribution pattern throughout the nervous system.
The localization of specific neuropeptides in the nerve cords of S. magnifica is described separately below. However, general features of the organization of the nervous system may be briefly mentioned, and a diagram representing the organization of the ventral nerve cords and body wall musculature is provided in Fig. 1 to facilitate the description. Most of the cells expressing neuropeptides are distributed in the periphery of the cords, as has been shown for other invertebrate nervous systems (Kandel 1975). The central region, the neuropile, consists of an extensive network of nerve fibers with large numbers of varicosities. In the cord, two neuronal cell morphologies are distinguished: (1) elongated cells with a small central nucleus, located mostly ventrolaterally, facing the ventral longitudinal muscle, and (2) pyramidal neurons, forming a large cluster, located mainly ventrally and ventromedially. The nuclei of the pyramidal cells are located near the apex of the cells, the region of origin of a single cell process. On the average, elongated neurons are 21.9+0.6 Ixm tong and 9.1 • txm wide (mean -_4-_SE; n = 107), while pyramidal neurons measure 18.5+_0.7 Ixm in length and 11.2_+0.3 mm across the base (n = 66). From the somata of the two types of cells the single axon-like fibers extend toward the central region of the cord or toward the commissures. Some of these fibers can be traced from the cell body to the neuropile or through the peripheral nerves, but most of them can not be followed once they intermingle with other
9~
v
DSN
#
F M R F amide ( FMRFa )-like immunoreactivity Several antibodies against F M R F a showed a similar distribution of the FMRFa-like immunoreactivity (FMRFa-LI) even though the intensity of their labelling varied. Most of our results were obtained using Dockray's (see Table 1) anti-FMRFa serum (1:250). Preincubation of the antibody with the peptide F M R F a at 100 IxM blocked the observed immunoreactivity, while lower concentrations of F M R F a (10 and 2 txM), only reduced it. F L R F a m i d e (10 txM), a related peptide with one amino acid substitution, neither blocked nor reduced the immunoreactivity. Somata showing F M R F a - L I were detected in the cords, especially ventromedially, an~ dense', networks Of immunoreactive varicose processes ~vere observed in the neuropile and segmental nerves (Fig. 2A). Cells expressing F M R F a - L I showed the pyramidal (56%) and elongated (44%) morphologies (Fig. 2B, Table2). Fibers that contained F M R F a - L I were observed along the longitudinal muscle cells, and some of these appeared as discrete fibers close to the muscle cells (Fig. 2C). There was no significant difference between the average number of cells per section at the commissural level (13.2+3.4, mean+_SE; n = 6 ) and the cell number per section between the commissures (6.9_+1.6, n = 1 5 ) . A few cell bodies showing F M R F a - L I were observed in the ventral segmental nerves, with a distinct elongated shape.
Neuropeptide Y ( NP Y)-like immunoreactivity
VSN
Fig. 1. Diagram of transverse section of Sabellastarte magnifica. Components of ventral portion of organism and, in particular double ventral nerve cord are shown. C Commissure; DSN dorsal segmental nerve; GA giant axon; N neurons; NC nerve cord; VLM ventral longitudinal muscle; VgN ventral segmental nerve
All N P Y antisera used produced a similar pattern of labelling; however, antiserum no. 4 2i 2s (see Table 1) showed the best labelling. Treatment of this N P Y antiserum with the synthetic peptide N P Y (2 IxM) caused a marked reduction in the observed fluorescence. F M R F a (2 gM) neither blocked nor reduced the observed NPYLI. Unlike the other peptide-immunoreactive neurons, the cells expressing N P Y - L I were distributed throughout the ventral and dorsal regions of the cords (Fig. 2D, E). Fifty-six percent of the cells showed the pyramidal
212
Fig. 2 A-F. F M R F a - L I and NPY-LI in nerve-celi bodies and fibers. A FMRFa-like immunoreactive neurons in section at commissural level and ventral segmental nerve. Note ventromedial localization of the cells (arrows) and numerous immunoreactive fibers in neuropile and in peripheral nerve (arrowhead). B Pyramidal neurons expressing F M R F a - L I . C Varicose nerve fibers expressing F M R F a -
LI in longitudinal muscle area. D Cells expressing NPY-LI distributed throughout cords at non-commissural level. Irnmunoreactive fibers in neuropile. E Upper right section of nerve cord with neuron expressing NPY-LI. F NPY-like immunoreactive fiber in the longitudinal muscle area. A x 3 2 0 ; B x 8 0 0 ; C x 6 0 0 ; D x 3 2 0 ; E x 6 0 0 ; F x600
213 Table 2. Details of peptide-like immunoreactivity. CN, Central nervous system; PN, peripheral nervous system; LM, longitudinal muscle Substance
Cells/ section (%)
Location
Elongated Pyramidal Elongated Elongated Pyramidal
44 56 100 55 45
NPY
Elongated Pyramidal
44 56
ENKs
Elongated Pyramidal
76 24
FMRFa SP CCK
Morphology
Fibers
Distinctive characteristics
CN
PN
LM
Ventromedially
+
+
+
Ventrolaterally Widely distributed, throughout the cord, except in the dorsal region of the NC
+ +
+ +
+ §
Widely distributed throughout the NC, including the dorsal region Ventromedially
+
+
+
+
+
-
morphology and 44% were elongated (Table 2). A few NPY-immunoreactive cell bodies with an elongated morphology were observed within the ventral segmental nerves. Numerous NPY-like immunoreactive varicose nerve fibers were found in the neuropile, segmental nerves and longitudinal muscle area. Some of the NPY-like immunoreactive fibers in the longitudinal muscle area had nerve terminals in close apposition to muscle cells (Fig. 2 F). The average number of cells per section which expressed N P Y - L I did not vary significantly between the commissural, (8.7 __0.2, n = 4) and non-commissural, (5.0_+ 1.1, n = 14) regions.
Elongated ceils in PN No cells in PN A distinctive CCK-like fiber tract in the NC; cells in PN, and; fibers in nerves in the muscle area Elongated cells in PN
No cells in PN
immunoreactive fibers observed in the muscle area, the CCK-like containing fibers did not appear to innervate the muscle cells. Fifty-five percent of the CCK-immunoreactive cell population were elongated while the remaining 45% showed a pyramidal morphology (Table 2). Elongated CCK-LI neurons were also located within the ventral segmental nerves (Fig. 3 E). Fibers from these peripheral neurons projected toward the cord. There were significantly more cell bodies per section at the commissural level (15.6+_3.3, n = 8 ) than between the commissures (3.1 _0.7, n = 1 0 , T = 3 . 4 2 ; P
Cell and Tissue Research 9 Springer-Verlag1991
Localization of neuropeptides in the nervous system of the marine annelid Sabellastarte magnitica Lucy Diaz-Miranda, Gladys Escalona de Motta, and Jos~ E. Garcia-Arrarfis Department of Biology, Rio Piedras Campus, and Institute of Neurobiology, Medical Sciences Campus, University of Puerto Rico, Rio Piedras, Puerto Rico 00931 Accepted April 29, 1991
Summary. Immunohistochemical studies of the nervous system of Sabellastarte magn~'ca, a sedentary polychaete, showed the presence of neuropeptide expressing cells and fibers within the double ventral nerve cord. Immunoreactivity to cholecystokinin, neuropeptide Y, enkephalins, substance P, and FMRFamide was found to be present in specific populations of cells, identifiable by their location and by the neuropeptide they expressed. Fibers expressing the various neuropeptides were also observed in particular locations within the nerve cord. This characteristic distribution of the various neuron subgroups and fiber pathways may represent functional circuits within the nervous system of this annelid. Key words" Neuropeptides - Annelids Cholecystokinin Neuropeptide Y - Enkephalins - Substance P FMRFamide - Sabellastarte magnifica (Annelida)
Biologically active peptides have emerged as candidates for neurotransmitters and neuromodulators in the nervous system of species in almost every animal phylum. In the annelid nervous system, a diverse number of these peptides has been demonstrated in specific populations of cells. These include molecules related to: cholecystokinin (CCK), enkephalins (ENK), endorphins, substance P (SP), neuropeptide Y (NPY), neurotensin, somatostatin (SS), vasoactive intestinal polypeptide, bombesin, and F M R F amide (FMRFa). The distribution of many of these peptides has been well studied in two classes of annelids: Hirudinea (leeches) and Oligochaeta (earthworms). Immunoreactivity related to peptides such as CCK, SP, FMRFa, enkephalins, vasoactive intestinal polypeptide, somatostatin and bombesin has been demonstrated in the nervous system of the leech Hirudo medicinalis (Zipser 1980; Osborne et al. 1982; Kuhlman et al. 1985a, b; Leake et al. 1986; Norris and Calabrese 1987). Offprint requests to: L. Diaz-Miranda
In addition, in the brain of the leech Theromyzon tessulaturn substances related to opioid peptides were reported (Verger-Bocquet et al. 1987). Similarly, in the nervous system of the earthworm Lumbricus terrestris the presence of immunoreactivity related to NPY, SP, peptide tyrosine, peptide histidine isoleucine, glucagon, gastrinreleasing peptide, calcitonin gene-related peptide, pancreatic polypeptide, vasoactive intestinal polypeptide, methionine-enkephalin and adrenocorticotropic hormone, has been demonstrated (Sundler et al. 1977; Alumets etal. 1979; Aros etal. 1980; Rzasa etal. 1984; Gesser and Larsson 1986; Curry et al. J989). Some of the peptides exert pronounced effects on the muscular activity of the annelids. For example, in the leech Hirudo medicinalis, FMRFa has been implicated in the control of heart beat and modulation of the contractility of the body-wall musculature (Kuhlman et al. 1985b; Norris and Calabrese 1987). In the intestine of the earthworm Lumbricus terrestris, SP has been shown to increase the frequency and amplitude of phasic contractions (Kaloustian and Edmans 1986). While considerable evidence has accumulated on the presence of neuropeptides in members of the Hirudinea and Oligochaeta, little is known about their presence and distribution in members of the third class of this phylum, the Polychaeta. Immunoreactivity similar to that of neurotensin, somatostatin, SP and CCK has been shown in the nervous system of two species of errant polychaetes, Nereis fucata and Nereis diversicolor (Carraway etal. 1982; Engelhardt et al. 1982; DhainautCourtois et al. 1985 a, b; Porchet and Dhainaut-Courtois 1988). Even less is known regarding peptides present within the sedentary species of this class. In view of the important functions and widespread occurrence of peptides in the annelid nervous system, we have investigated the cellular localization of several of these peptides in the nerve cord of the sedentary polychaete Sabellastarte magnifica using immunocytochemical techniques. Our results extend the available information on the neuroanatomical organization of the annelid nervous system and provide the histological basis for
210
ongoing physiological and pharmacological studies in our laboratory. Materials and methods Adult specimens of S. magnifica, up to 11 cm in length, were collected in coral reefs off the east coast of Puerto Rico and maintained at the facilities of the Institute of Neurobiology of the University of Puerto Rico in San Juan. The facilities include 20 1 sea water-aquaria where the animals were kept until used.
Immunohistochemistry Histochemical studies were carried out on isolated lengths of the ventral body wall that contains the paired nerve cords, and a portion of the ventral longitudinal and circular muscles, from a total of 10 animals. The worm was removed from its tube and anesthesized in 7.5% M g C I ~ - d H 2 0 at room temperature for 2 h. It was then decapitated with a blade knife, pinned with its dorsal side up in a Petri dish and covered with the M g C 1 2 - d H 2 0 solution. The anterior part of the ventral trunk, containing portions of the double ventral nerve cord and the ventral longitudinal and circular muscles, was dissected, pinned in a Sylgard-coated Petri dish and fixed in picric acid-formaldehyde mixture (Zamboni and De Martino 1967), p H 7.4, at 4 ~ for approximately 24 h. The tissues were then dehydrated in an ascending series of ethanol solutions to xylene, rehydrated to H 2 0 and placed overnight at 4 ~ C in 30%
T a b l e 1. Details of antisera. All antisera
were polyclonal antisera produced in rabbits
sucrose-0.1 M phosphate-buffered saline (PBS) as described by Keast and colleagues (1984). Tissues were embedded in O.C.T. compound (Miles Inc.) for sectioning on a cryostat at - 2 0 ~ C. Ten-gin transverse sections were mounted on gelatin-chrome alumcoated slides and dried with cool air for I h. For immunohistochemistry, the indirect immunofluorescence method of Coons et al. (1955) was used. Tissue sections were first incubated in 2 - 4 % nonimmune bovine or goat serum for 30 rain, washed in 0.1 M PBS, and incubated overnight in a humid atmosphere with antisera to the different substances diluted in 0.1% Na azide 0.1 M PBS. Antisera against the neuropeptides F M R F a , SP, CCK, NPY and E N K were used (see Table 1 for details). The following day, after three 15-min washes in 0.1 M PBS, the sections were incubated for 1 h with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit immunoglobulin (TAGO) diluted 1:50 in 0.1% Na azide 0.1 M PBS. Slides were rinsed three times (20 rain/rinse) in 0.1 M PBS, mounted with buffered glycerol pH = 8.6, and examined in a Nikon epifluorescence microscope equipped with a V filter (IF 395-425 nm excitation) for the detection of the FITC molecule. Controls were carried out by substituting 0.1 M PBS for the primary antibody and by incubation of tissue sections with the primary antiserum that had been preabsorbed with an excess of its specific antigen, or with peptides with similar amino acid sequences. Cells stained by the specific antisera in each section were counted and classified by localization of the cell body, morphology and size. Only cells with visible nuclei were counted. The size of cells was determined by measuring the length and width of the cell body with an ocular micrometer. The number of labelled cells was determined both at the commissure (i.e., the area of transverse
Antigen
Code
Working dilution
Source
Molluscan cardioexcitatory neuropeptide: F M R F amide (FMRFa)
Dockray
1:250
Dockray et al. (1983)
CRB
1:1000
Schot JGA
1 : 1000 1 : 500
Cambridge Research Biochemicals, Wilmington, Del., USA see Reuter et al. (1986) Garcia-Arrarfis, University of Puerto Rico, Rio Piedras, Puerto Rico (in preparation)
no. 3 2i 2s
1 : 1500
Lugo-Chinchilla (1989)
no. 4 2i 2s no. 4 3i2s
1 : 1000 1:1000
Lugo-Chinchilla (1989) Lugo-Chinchilla (1989)
no. 1 3i 2s
1 : 1000
no. 2 3i 2s
1 : 1000
Garcia-Arrarfis et al. (1991) Garcia-Arrarfis et al. (1991) Rehfeld (1981)
Porcine neuropeptide Y (NPY)
Cholecystokinin fragment 26-33 (CCK)
no. 4562
1 : 1500
Leucine-enkephalin (Leu-ENK)
JO
I : 100
Dr. J. Oliver Flinders Med. Cr. Bedford Park, S. Australia (unpublished)
Methionine-enkephalin (Met-ENK)
Tramu
1:100
Verger-Bocquet et al. (1987)
Substance P (SP)
no. 101
1:1000
no. 102
1 : 1000
Fontaine-P6rus et al. (1985) Fontaine-P&us et al. (1985)
211 connections between the cords) and in the region between the commissures by counting the cells present in the double ventral cord of at least one section from each of three different specimens. Comparisons were made between the numbers of cells expressing each neuropeptide in the commissural and non-commissural areas. Comparisons were also made between the number of cells expressing each specific neuropeptide. Statistical analysis consisted of nonparametric tests: Mann-Whitney U Test Statistic and Kruskal-Wallis One-Way analysis of variance.
processes. Due to the large number of fibers and varicosities, it was not possible to detect branching of neuronal processes. Processes were also observed within the longitudinal muscle layer of the body wall, either as nerves running in between the muscle bundles or as more discrete fibers in apposition with muscle cells. Elongated cell bodies are also observed outside the cords, within the ventral segmental nerves.
Results
Immunohistochemistry
Organization of nervous system
Histochemical studies revealed the presence-of immunoreactivity to CCK, SP, F M R F a , N P Y and ENK. Each substance presented a particular distribution pattern throughout the nervous system.
The localization of specific neuropeptides in the nerve cords of S. magnifica is described separately below. However, general features of the organization of the nervous system may be briefly mentioned, and a diagram representing the organization of the ventral nerve cords and body wall musculature is provided in Fig. 1 to facilitate the description. Most of the cells expressing neuropeptides are distributed in the periphery of the cords, as has been shown for other invertebrate nervous systems (Kandel 1975). The central region, the neuropile, consists of an extensive network of nerve fibers with large numbers of varicosities. In the cord, two neuronal cell morphologies are distinguished: (1) elongated cells with a small central nucleus, located mostly ventrolaterally, facing the ventral longitudinal muscle, and (2) pyramidal neurons, forming a large cluster, located mainly ventrally and ventromedially. The nuclei of the pyramidal cells are located near the apex of the cells, the region of origin of a single cell process. On the average, elongated neurons are 21.9+0.6 Ixm tong and 9.1 • txm wide (mean -_4-_SE; n = 107), while pyramidal neurons measure 18.5+_0.7 Ixm in length and 11.2_+0.3 mm across the base (n = 66). From the somata of the two types of cells the single axon-like fibers extend toward the central region of the cord or toward the commissures. Some of these fibers can be traced from the cell body to the neuropile or through the peripheral nerves, but most of them can not be followed once they intermingle with other
9~
v
DSN
#
F M R F amide ( FMRFa )-like immunoreactivity Several antibodies against F M R F a showed a similar distribution of the FMRFa-like immunoreactivity (FMRFa-LI) even though the intensity of their labelling varied. Most of our results were obtained using Dockray's (see Table 1) anti-FMRFa serum (1:250). Preincubation of the antibody with the peptide F M R F a at 100 IxM blocked the observed immunoreactivity, while lower concentrations of F M R F a (10 and 2 txM), only reduced it. F L R F a m i d e (10 txM), a related peptide with one amino acid substitution, neither blocked nor reduced the immunoreactivity. Somata showing F M R F a - L I were detected in the cords, especially ventromedially, an~ dense', networks Of immunoreactive varicose processes ~vere observed in the neuropile and segmental nerves (Fig. 2A). Cells expressing F M R F a - L I showed the pyramidal (56%) and elongated (44%) morphologies (Fig. 2B, Table2). Fibers that contained F M R F a - L I were observed along the longitudinal muscle cells, and some of these appeared as discrete fibers close to the muscle cells (Fig. 2C). There was no significant difference between the average number of cells per section at the commissural level (13.2+3.4, mean+_SE; n = 6 ) and the cell number per section between the commissures (6.9_+1.6, n = 1 5 ) . A few cell bodies showing F M R F a - L I were observed in the ventral segmental nerves, with a distinct elongated shape.
Neuropeptide Y ( NP Y)-like immunoreactivity
VSN
Fig. 1. Diagram of transverse section of Sabellastarte magnifica. Components of ventral portion of organism and, in particular double ventral nerve cord are shown. C Commissure; DSN dorsal segmental nerve; GA giant axon; N neurons; NC nerve cord; VLM ventral longitudinal muscle; VgN ventral segmental nerve
All N P Y antisera used produced a similar pattern of labelling; however, antiserum no. 4 2i 2s (see Table 1) showed the best labelling. Treatment of this N P Y antiserum with the synthetic peptide N P Y (2 IxM) caused a marked reduction in the observed fluorescence. F M R F a (2 gM) neither blocked nor reduced the observed NPYLI. Unlike the other peptide-immunoreactive neurons, the cells expressing N P Y - L I were distributed throughout the ventral and dorsal regions of the cords (Fig. 2D, E). Fifty-six percent of the cells showed the pyramidal
212
Fig. 2 A-F. F M R F a - L I and NPY-LI in nerve-celi bodies and fibers. A FMRFa-like immunoreactive neurons in section at commissural level and ventral segmental nerve. Note ventromedial localization of the cells (arrows) and numerous immunoreactive fibers in neuropile and in peripheral nerve (arrowhead). B Pyramidal neurons expressing F M R F a - L I . C Varicose nerve fibers expressing F M R F a -
LI in longitudinal muscle area. D Cells expressing NPY-LI distributed throughout cords at non-commissural level. Irnmunoreactive fibers in neuropile. E Upper right section of nerve cord with neuron expressing NPY-LI. F NPY-like immunoreactive fiber in the longitudinal muscle area. A x 3 2 0 ; B x 8 0 0 ; C x 6 0 0 ; D x 3 2 0 ; E x 6 0 0 ; F x600
213 Table 2. Details of peptide-like immunoreactivity. CN, Central nervous system; PN, peripheral nervous system; LM, longitudinal muscle Substance
Cells/ section (%)
Location
Elongated Pyramidal Elongated Elongated Pyramidal
44 56 100 55 45
NPY
Elongated Pyramidal
44 56
ENKs
Elongated Pyramidal
76 24
FMRFa SP CCK
Morphology
Fibers
Distinctive characteristics
CN
PN
LM
Ventromedially
+
+
+
Ventrolaterally Widely distributed, throughout the cord, except in the dorsal region of the NC
+ +
+ +
+ §
Widely distributed throughout the NC, including the dorsal region Ventromedially
+
+
+
+
+
-
morphology and 44% were elongated (Table 2). A few NPY-immunoreactive cell bodies with an elongated morphology were observed within the ventral segmental nerves. Numerous NPY-like immunoreactive varicose nerve fibers were found in the neuropile, segmental nerves and longitudinal muscle area. Some of the NPY-like immunoreactive fibers in the longitudinal muscle area had nerve terminals in close apposition to muscle cells (Fig. 2 F). The average number of cells per section which expressed N P Y - L I did not vary significantly between the commissural, (8.7 __0.2, n = 4) and non-commissural, (5.0_+ 1.1, n = 14) regions.
Elongated ceils in PN No cells in PN A distinctive CCK-like fiber tract in the NC; cells in PN, and; fibers in nerves in the muscle area Elongated cells in PN
No cells in PN
immunoreactive fibers observed in the muscle area, the CCK-like containing fibers did not appear to innervate the muscle cells. Fifty-five percent of the CCK-immunoreactive cell population were elongated while the remaining 45% showed a pyramidal morphology (Table 2). Elongated CCK-LI neurons were also located within the ventral segmental nerves (Fig. 3 E). Fibers from these peripheral neurons projected toward the cord. There were significantly more cell bodies per section at the commissural level (15.6+_3.3, n = 8 ) than between the commissures (3.1 _0.7, n = 1 0 , T = 3 . 4 2 ; P
