Introduction

Vertebrates produce a variety of molecules that are important in controlling and regulating the life processes. Initially, a clear distinction was made between the peptide hormones produced by the endocrine system, and the other major class of regulatory substances, the neurotransmitters, produced by the nervous system. Emphasis was placed on the contrasts between these two types of molecules, and comparisons were made of the differences in the sites of production, methods of release, transportation within the body and the duration of action. On the basis of these comparisons the endocrine and nervous systems were considered to be completely different.

Gradually, however, these two systems have come to be seen as the extremes in a wide spectrum of control mechanisms over biological processes. In between, we have a subtle arrangement and interplay of molecules which cannot, usefully, be categorised as either hormonal or nervous.

A major factor in the shift away from this traditional and separatist view of biological control systems is the finding that the brain produces a great variety of peptides. Indeed, the term ‘endocrine brain’ has been coined (Motta 1980). The discovery of the peptide hormones of the hypothalamus, firstly oxytocin and vasopressin and shortly afterwards the releasing hormones, provided the first clear evidence that certain neurones were peptidergic. Subsequently the phenomenon of neurosecretion has been studied in a wide range of vertebrates and invertebrates.

Nerve cells use various chemical messengers, including acetylcholine, monoamines, amino acids and neuropeptides. In the past 10 years it has become clear that there is a great molecular diversity of such messengers particularly amongst the amino acids. However, the molecular diversity of neuropeptides exceeds by far that of the amino acids. There is now evidence that in one animal species as many as 100-200 different peptide molecules may serve as neuronal messengers (Kandel 1983; Nieuwenhuys 1985; Joosse 1986).

Neuropeptides occur in nervous systems of all animals, even in the most simple such as coelenterates (Schaller et al. 1984). For many years neuronal peptides were thought to have only a neuroendocrine role. The present view however is that, in addition to their role as neurohormones, they may act as typical synaptic neurotransmitters and as paracrine or neurocrine regulators at non-synaptic sites (Buma & Roubos 1985; Nieuwenhuys 1985).

Immunocytochemical and biochemical studies provide substantial evidence that structurally related peptides are found in many different phyla. Well-known examples are insulin-like peptides and FMRFamide. An important consequence of these findings is that some of the peptides found so far may have arisen from a smaller number of ancestral molecules, possibly at the prokaryote stage (Joosse 1987).

A large part of the present volume concerns the molecular diversity and functions of neuropeptides in invertebrates. In view of the great diversity in structure and function of neuropeptides and their presence throughout the animal kingdom, the question arises as to the adaptive value of peptide variety.

Introduction

Coelenterates have the simplest nervous system in the animal kingdom, and it was probably within this group of animals that nervous systems first evolved. Extant coelenterates are diverse and comprise two phyla. The classes Hydrozoa (for example hydroids and their medusae), Cubozoa (box jellyfishes), Scyphozoa (true jellyfishes), and Anthozoa (for example sea anemones and corals) constitute the phylum Cnidaria. A companion phylum is that of the Ctenophora (comb jellies), or Acnidaria. The general plan of the coelenterate nervous system has often been described as a nerve net. This is an oversimplification because many species show condensation of neurones to form linear or circular tracts. Linear tracts occur in the stem and tentacles of physonectid siphonophores (Mackie 1973; Grimmelikhuijzen et al. 1986), between the ocelli and marginal nerve rings of anthomedusae (Singla & Weber 1982; Grimmelikhuijzen & Spencer 1984), and at the bases of mesenteries in sea anemonies (Batham et al. 1960). Circular tracts, or nerve rings, have been found in the bell margin of hydrozoan medusae (Hertwig & Hertwig 1878; Jha & Mackie 1967; Spencer 1979; Grimmelikhuijzen et al. 1986) and at the base of the hypostome in some hydrozoan and cubozoan polyps (Werner et al. 1976; Grimmelikhuijzen 1985). Presumably, these tracts have evolved to form pathways for rapid conduction of information. The marginal nerve rings of hydromedusae, in addition, form a circular CNS, which is capable of both integrating a variety of sensory inputs, and of transmitting the input rapidly throughout the margin (Spencer & Arkett 1984).

The crustacean neuro-endocrine system has an extensive and diverse organisation (GABE 1966). It includes the Y-organs, mandibular organs and gonads as well as all parts of the nervous sytem (NS) (Echalier 1954; Le Roux 1968; Charniaux-Cotton 1954; Juchault 1966).

This chapter will concentrate on the peptidergic and aminergic areas of the NS, utilizing the isopod model as a framework upon which to base comparative data obtained from decapods.

The neurosecretory cells

The Isopod brain comprises 3 major areas: protocerebrum, innervating the compound eyes; deutocerebrum, innervating the antennule; and tritocerebrum, innervating the antennary.

Neurosecretory products are often revealed by standard histological techniques such as paraldehyde fuchsin (PF) and chrome hematoxylin phloxine (CHP). We have used these techniques to locate 4 types and subtypes of neurosecretory cells (NSC) in the NS (Martin 1981), and the ones described here will be restricted to those terminating in the sinus gland (SG), following the pattern established by the cobalt back-filling experiments of Chiang and Steel (1985a). NSC are classified according to the terminology introduced by Matsumoto (1959).

β-cells: These comprise a group of heterogeneous cells, located in the anterior part of the protocerebrum, on each side of the midline (Fig. 1a; 1c).

Subgroup β1 contains 6 cells, which are polygonal, 14–28 μm in diameter which stain intensely with both PF and CHP. Ultrastructure shows an ovoid nucleus, stacks of RER, Golgi-derived electron dense granules (160-180 nm) and glycogen particles associated with electron lucent vacuoles (Fig. 1d).

In vertebrate endocrinology, significant progress has been made over the past 30 years due to the availability of immunochemical techniques. Marshall (1951) obtained the first immunohistological result in the adenohypophysis of several mammalian species using an anti-ACTH antiserum. These immunohistological investigations were made possible because of the early isolation of vertebrate hormones and neurohormones.

Invertebrate neuropeptides, however, were not purified and sequenced until more recently because of the small size of central nervous systems (CNS) or neurohaemal organs. The first invertebrate neuropeptide to be isolated was proctolin from the cockroach, Periplaneta americana (Brown & Starratt 1975). Invertebrate neurosecretory cells have however been visualized by histochemical staining methods for about forty years.

Most antibodies used in vertebrate immunocytochemistry are raised against mammalian hormones or neurohormones. They nevertheless give positive immunoreactions at the level of lower vertebrate hypophyseal and hypothalamic cells. It was tempting therefore to discover if such antisera could also generate positive results in invertebrates. The first evidence of the existence of an immunochemical relationship between vertebrate and invertebrate neuropeptides was produced in 1975 by Grimm-Jorgensen & MacKelvy. Using radioimmunoassays, they found an immunoreactive thyrotropin releasing hormone (TRH)-like substance in gastropod ganglia.

In 1977, some invertebrate neurosecretory cells synthesizing neurosecretory products related to vertebrate neuropeptides or vertebrate gastro-entero-pancreatic peptides, were visualized by immunocytochemical techniques, in earthworm ganglia (Sundler et al. 1977) and insect suboesophageal ganglia (Rémy et al. 1977). These early findings caused surprise and even scepticism among several scientists. There are now over two hundred different immunocytochemical results in this area, about seventy five of which concern insects.

Introduction

In molluscs, bioactive peptides are produced by peptidergic neurons, endocrine glands and other tissues, such as cells of the intestinal tract. These peptides function as neurotransmitters/neuromodulators and (neuro)hormones, and control a wide range of events concerned with behaviour, reproduction, and metabolism. Particular attention has been paid to the peptidergic model systems in Lymnaea and Aplysia, to FMRFamide and related peptides which also exhibit an extra-molluscan distribution, and to the presence and function of vertebrate peptides in molluscs.

Our knowledge of biologically active peptides in molluscs is expanding rapidly due to the introduction, among other things, of sophisticated chromatographic and sequence techniques, and the methods of molecular biology. A review of this length must, of necessity, be selective. We have attempted to give a critical account of the data concerning the physiological role and the nature of (presumed) bioactive peptides, and avoided discussions of non-relevant details. Recent reviews present a wealth of complementary data (e.g. Joosse & Geraerts 1983; Geraerts & Joosse 1984; Roubos 1984; Rothman et al 1985; Geraerts et al. 1987. See also Chapters 2 and 14 this volume).

The FMRFamide family, and opioid peptides

The FMRFamide family

The neuropeptide FMRFamide was isolated originally from the clam Macrocallista nimbosa (Price & Greenberg 1977). In addition to FMRFamide, various related peptides have subsequently been isolated from the brain of a number of species belonging to different classes of the molluscs (Table 1).

Introduction

A seemingly bewildering array of factors with putative neurohormonal function have been described in (mainly decapod) Crustacea (see Kleinholz & Keller 1979; Cooke & Sullivan 1982; Keller 1983; Kleinholz 1985). They are implicated in almost every aspect of crustacean physiology, including pigment dispersion and concentration, inhibition of moulting, limb regeneration and gonad development, cardiac control, blood glucose, metabolism and respiratory control, ion and water balance, endogenous rhythmicity and locomotion. Several of these factors are produced by neurosecretory structures in the eyestalk, which can be easily ablated. This accessibility has unfortunately led to a tendency to assign hormonal regulation of physiological mechanisms based solely upon the results of eyestalk removal, often without the rigorous application of deficiency and replacement protocols using physiologically relevant doses of extracts or further purification of the active principle. Thus, apart from the well known neuropeptides, it is not known how many of these described ‘factors’ genuinely control individual processes and little is known of their precise chemical identity.

Evidence from immunocytochemical studies suggests that many neuropeptides classically known as ‘vertebrate’ peptides and also neuropeptides that have originally been found in invertebrates (e.g. FMRF amide, proctolin) are ubiquitous in crustaceans (Mancillas et al. 1981; Jacobs & Van Herp 1984; Jaros et al. 1985; Van Deijnen et al. 1985; Stangier et al. 1986). However, there is at present little information concerning the role of ‘vertebrate-type’ peptides in physiological integration.

Amongst invertebrates, few bioactive peptides have been sequenced and, of these, most are from molluscs or arthropods and there are, as yet, no sequenced neuropeptides from the annelids. In spite of this paucity of information, we expect that a range of model systems will be developed within this phylum. The failure to sequence any annelid peptide is therefore unfortunate in view of the experimental opportunities provided by the nervous system (NS) of various species (leeches and polychaetes). From an evolutionary point of view, annelids are basic to the phylogeny of numerous animal groups. Knowledge of the structure of their neuropeptides is therefore of interest for comparison with those of invertebrates either more primitive (Hydra, Schaller et al 1984, see also Chapter 10 this volume) or more advanced (molluscs and arthropods, see Chapters 5-9 and 13-15 this volume).

Studies on neuropeptides are approached in two ways: the first involves a neurophysiological and neurobiological analysis directed towards an understanding of neurone physiology and more particularly the role of peptides in neurotransmission and neuromodulation (O'Shea & Schaffer 1985; see also Chapters 8 and 14 this volume). The pioneering work of Stent has established the leech as an important annelid model for the study of the cellular basis of behaviour (Stent & Kristan 1981) and as a model for neuronal development (Stent et al. 1982). The second approach is neuroendocrinological and is based upon a study of the production of peptides by neurosecretory centres in the central nervous system (CNS). Annelids have no discrete endocrine glands except for the infracerebral region (ICR), a neurohaemal area (fig. 5) found in several polychaetes (Dhainaut-Courtois 1970; Golding & Whittle 1977; Olive 1980; Durchon 1984).

The name ‘femerfamide’ (FMRFamide) has been given (Greenberg 1982) to the molluscan neuropeptide Phenylalanyl–methionyl–arginyl–phenylalanyl–amide (Price & Greenberg 1977). It is a mnemonic for the contained amino acids and was suggested to avoid prejudicing views about its natural role(s). (During the period of its discovery it was called Cardioexcitatory Neuropeptide because this was one notable action of the pep tide. See Price & Greenberg 1977.)

Price and Greenberg and their associates have shown that in molluscs there are several peptides chemically related to FMRFamide (see eg Price 1986): these are FLRFamide, pQDPFLRFamide, GDPFLRFamide and SDPFLRFamide. It is convenient to refer to these peptides with one name, as for instance are the opioid peptides, because like the opioid peptides, they appear to require a particular sequence of amino acids for activity and show a range of effects suggesting that they operate through several receptor types. Consequently we shall refer to the molluscan molecules as the femerfamide peptides.

With the opioid peptides the sequence Tyr–Gly–Gly–Phe (YGGF–) is required for full opioid activity, as with for example Met– and Leu–enkephalin, endorphin and dynorphin. This C–terminal sequence of amino acids has been called the message sequence by Charkin and Goldstein (1981). By contrast, in the femerfamide peptides the C–terminal sequence appears essential for activity, i.e. serves as the ‘message-sequence’: –Phe (or Tyr)–Met (or Leu)–Arg–Phe–NH2 (Price 1986).

Introduction

In insects a variety of functions (e.g. intermediary metabolism, ion and osmoregulation, developmental and neuronal processes) are regulated by peptides from different parts of the nervous system. This diversity necessitates the restriction of the present chapter to a particular group of peptides. We have been interested in the structure and biological functions of peptide hormones regulating intermediary metabolism and fluid secretion in insects and these studies are used to highlight the problems encountered in characterizing these peptides.

Lipid and carbohydrate metabolism in insects can be regulated by adipokinetic (AKH) and so-called hyperglycaemic (HGH) hormones present in the corpora cardiaca (Mayer & Candy 1969; Steele 1961). These peptides increase concentrations of haemolymph lipids or carbohydrates. The physiological functions and modes of action of these hormones are reviewed in Chapter 7. Diuretic hormones (DH) present in the corpora cardiaca but also in other parts of the nervous system, e.g. brain, suboesophageal and thoracic ganglia (Proux et al. 1982; Morgan & Mordue 1984a; Aston & Hughes 1980) regulate fluid secretion.

In recent years investigations have been conducted to isolate and characterize prothoracicotropic hormones (PTTH) which stimulate the release of ecdysone from the prothoracic glands (Williams 1947). In Bombyx mori, the PTTHs have been found to exhibit significant homologies in their N-terminal sequences with insulin and insulin-like growth factors (Nagasawa et al. 1984). These peptides are discussed in Chapter 6 which is devoted to the presence of Vertebrate* peptides in insects. The involvement of peptides in the functioning of neurones as neurotransmitters and/or neuromodulators is covered in Chapter 8.

Ultrastructure of invertebrate nervous systems

Neuroecretory and other neurons

Examination of invertebrate nervous systems reveals that many are richly endowed with neurones resembling classical neurosecretory cells in cytology and ultrastructure. Such cells are clearly specialized for peptide secretion. They contain an abundance of rough endoplasmic reticulum (RER), and secretory granules (variously known as elementary granules, large dense-cored vesicles, etc.) generated by Golgi bodies, accumulate in large numbers within the perikarya. Although many are doubtless endocrine cells, others (Figs. 1-3) have axons which extend not to blood cavities, but into the central neuropile where the secretory material is discharged.

Furthermore, some secretory granules are evident in virtually all neurones (Golding & Whittle 1977), and this is consistent with the finding that many and perhaps all neurones, including those with conventional transmitters, also secrete peptides (review by Hokfelt, Johansson & Goldstein 1984).

Nerve terminals: vesicles and granules

In most nerve terminals, whether in the central or peripheral nervous systems, large numbers of synaptic vesicles are encountered (Figs. 4 & 5). Measuring 20-50nm in diameter, the vesicles in all but a small minority of terminals (Fig. 6) have lucent contents, except following exposure to a mixture of Zinc iodide and Osmium tetroxide (the ZIO reagent) (Fig. 7) which deposits extremely electrondense material within them (May & Golding 1982). The vesicles cluster densely adjacent to sites of specialized contact with other cells. Pre- and postsynaptic thickenings are present and the synaptic clefts are wider and often more regular in form than other intercellular spaces, and contain moderately dense material.

Introduction

The discovery that many central neurones utilize peptides as extracellular chemical messengers has revolutionized our understanding of neuronal signalling. Studies to characterize the structure and functions of neuropeptides have taken various approaches, including purification and biochemical analysis of the peptide products and molecular genetic studies of the genes encoding precursor proteins which give rise to peptide products. These investigations have been greatly aided by the use of non-neuronal tissues, such as epithelial tissue or digestive organs, which are often rich sources of bioactive peptides. Many peptides initially identified in peripheral tissues have been found subsequently in the central nervous system. One preparation, frog skin, has been particularly useful in this regard, and has facilitated the discovery of mammalian peptides related to frog bombesin (Orloff et al. 1984).

Invertebrate nervous systems offer unique advantages in the study of neurotransmitter function. Our understanding of the molecular mechanisms underlying neurotransmitter actions have been greatly facilitated by the use of invertebrate systems due to the smaller number of neurons, their simpler organization, and the often large size of their cell soma (see also Chapter 8). In terms of neuropeptide biology and chemistry a number of questions arise: Can neuropeptides related to vertebrate neuropeptides be found in invertebrates? Can neuropeptides characterized in invertebrate systems be used to identify homologous peptides in mammalian systems? Can invertebrate systems be used to gain further insight into the function, regulation, and evolution of neuroendocrine systems?

Introduction

The collective title protochordates is essentially a convenient ‘umbrella’ term for a diverse assembly of animals sharing relatively few, although important, common features. The protochordate group is usually thought to comprise three subphyla, the hemichordata, urochordata (tunicata) and cephalochordata, although some authorities consider the hemichordates a separate phylum. There are, however, no published reports on the occurrence of peptides or amines in hemichordates and they will not be considered further here.

The tunicates and cephalochordates are highly specialized marine organisms whose close relationship is perhaps rather superficial and, in reality, based more upon similar, ciliary powered, particle feeding mechanisms which have developed around the possession of a perforated pharynx, rather than any genuine and common phylogenetic background. Notwithstanding such reservations the protochordates are probably the only available extant representatives of ancient groups which “bridged the gap” between invertebrates and vertebrates, and as such could provide useful clues to the origins of certain vertebrate features.

The features which have attracted most attention and which have provided the most useful information are the central nervous system (CNS) and the gastrointestinal tract. The tunicate CNS has a relatively simple organisation which is in many respects rather more comparable with simple invertebrate neurosecretory centres than any part of the vertebrate CNS. Similarly the cephalochordate CNS is a simple anterior elaboration of the dorsal nerve cord (characteristic of these animals), and its structure and function reflects the highly specialized nature of the adult organism as well as its unusual and asymmetric development.

Introduction: Hormones and development

Since Kopec's demonstration of a humoral role of the brain in insect development, peptide hormones have had a central historic importance in the study of insect endocrinology: the brain produces prothoracicotropic hormones (PTTH's; see Bollenbacher & Granger 1985), thus stimulating moulting by initiating ecdysone synthesis and release by the prothoracic glands. A second neuropeptide, eclosion hormone (see Reynolds & Truman 1983), initiates the necessary behaviour patterns associated with ecdysis and its timing. Subsequently, a third neuropeptide, bursicon (see Reynolds 1983; 1985), controls the tanning of the new cuticle and stimulates endocuticle deposition. Neuropeptides are also implicated in the control of corpus allatum activity; both allatohibins and allatotropins being identified in various insects (see Tobe & Stay 1985). Unfortunately, chemical characterisation of these peptides has progressed slowly. The physiological actions of most of these peptides concerned with development have been the subject of numerous and extensive reviews (see for example Downer & Laufer 1983; Kerkut & Gilbert 1985) and will not be dealt with here.

Recently considerable progress has been made in the characterisation of PTTH's from Bombyx mori (Nagasawa et al. 1984, 1986). Characterisation of these polypeptides has been hampered by their heterogeneity (in several species high and low molecular weight forms exist) but also, no doubt, due to the impurity of the starting material used. The sequence of 4K-PTTH-II one of the small Bombyx hormones, has now been determined (Nagasawa et al. 1986).