Phagocytosis and Encapsulation: Cellular Immune Responses in Arthropoda 1

Transcription

1 AMER. ZOOL., 23: (1983) Phagocytosis and Encapsulation: Cellular Immune Responses in Arthropoda 1 STUART RATNER AND S. BRADLEIGH VINSON Department of Entomology, Texas A&M University, College Station, Texas SYNOPSIS. The least understood aspects of the cellular immune reactions of arthropods are the earliest events: the initial recognition of foreignness, and the resulting changes in hemocyte behavior and morphology. There are indications that the recognition of a surface as foreign is based primarily upon its electrostatic charge, but more specific criteria may be utilized in some situations. Phagocytes are capable of recognizing foreignness without the intervention of soluble opsonins but, in some arthropods, there is in vitro evidence that opsonins can increase the efficiency of phagocytosis. It has been hypothesized, on the basis of ultrastructural evidence that encapsulation, a multicellular immune response, is induced when labile hemocytes rupture upon encountering a foreign object. The released products may promote the formation of a sheath of ameboid hemocytes around the object. This hypothesis is now supported by the results of experiments performed on an in vitro encapsulation system. This system may prove useful in the purification of encapsulation promoting factors; in determining the mechanism of their release from hemocytes; and in investigating the possibility that the various cellular immune reactions of arthropods have a common underlying mechanism. INTRODUCTION The reactions of arthropodan hemocytes against foreign bodies can be considered to constitute an "immune system" only in the most general sense of the term: they aid in keeping the animals free of parasites and pathogens. The connotations which have become irreversibly bound to the term "immunity" as it has been used by vertebrate-oriented immunologists specificity, memory, the involvement of immunoglobulins do not apply to the situation in arthropods. No immunoglobulins are present, nor do the hemocytes possess receptors for them (Anderson, 1977). Discrimination is generally so poor that allografts, and sometimes even xenografts, are tolerated (Lackie, 1981a). Although there are indications of specific memory in some humoral reactions (Karp and Rheins, 1980), no convincing evidence of cellular immune memory has been produced. Nevertheless, the cellular immune system evidently does an adequate job of eliminating foreign material from the arthropodal hemocoel, judging from the success of the phylum. This review concerns phagocytosis, the 1 From the Symposium on Comparative Aspects of Inflammation presented at the Annual Meeting of the American Society of Zoologists, December 1981, at Dallas, Texas. 185 unicellular defense reaction which clears the hemolymph of bacteria and other microscopic foreign objects; and cellular encapsulation, one of the multicellular reactions, which is directed against larger invaders of the hemocoel (i.e., those greater than 10 Mm in diameter [Salt, 1970]). The morphological aspects of these reactions are becoming increasingly clear, but our understanding of their physiology and biochemistry lags far behind (see recent reviews: Schapiro, 1975; Ratcliffe and Rowley, 1979; Baerwald, 1979; Lackie, 1980). Some of the most difficult unresolved questions deal with the earliest events in the cellular reactions: the actual recognition of foreignness and the processes directly triggered by this recognition. In this review we will survey the progress which has been made toward answering some of these questions. We will also consider the possibility that phagocytosis, encapsulation, and other cellular immune responses share common underlying mechanisms. Recent findings we have made using an in vitro encapsulation system will be incorporated wherever necessary. Arthropodan hemolymph contains an array of humoral immune factors, many of which are antimicrobial (see Boman and Hultmark, 1981). In this review, only humoral factors associated with cellular immunity will be considered.

2 186 S. RATNER AND S. B. VINSON PHAGOCYTOSIS Arthropodan phagocytes are morphologically diverse, and have been granted such names as granulocyte, amebocyte, lamellocyte and, perhaps most commonly, plasmatocyte. For convenience, the term "plasmatocyte" will be used in this paper to denote all ameboid, phagocytic hemocytes. Are humoral factors involved in phagocytosis'? Opsonins are serum substances which, when attached to foreign objects, increase their susceptibility to phagocytosis. In vertebrates, immunoglobulin G and the third component of complement act as opsonins. Arthropod serum lacks immunoglobulins and complement (Whitcomb et al., 1974), but does it contain opsonic factors? The standard test for serum opsonins is the addition of serum-treated or untreated test particles (often mammalian erythrocytes) to hemocyte suspensions or monolayers. If an opsonin is present, serum-pretreatment will increase the binding or uptake of the particles by the hemocytes. This test has revealed the presence of opsonins in some crustaceans (McKay et al., 1969; McKay and Jenkin, 1970; Paterson and Stewart, 1974; Schapiro et al., 1977). There is also evidence that a component of the serum of the crayfish Parachaeraps bicarinatus aids in the elimination of bacteria in vivo (Tyson and Jenkin, 1973) although it was not definitely determined that phagocytosis was the means of elimination. In most cases, some degree of phagocytosis or binding can occur even in the absence of serum. In contrast to the situation in Crustacea, all reported attempts to demonstrate opsonins in insects have been unsuccessful (Rabinovitch and DeStefano, 1970; Scott, 1971; Anderson et al., 1972; Wago and Ichikawa, 1979; Rowley and Ratcliffe, 1980). This is true even in insects whose serum agglutinates the test particles (agglutinins are opsonic in some molluscs [Hardy et al., 1977]). The discrepancy in opsonin occurrence between crustaceans and insects may be the result of different selective pressures exerted on the two classes, as suggested by Scott (1971); or it may be merely a reflection of the use of different techniques, in a small number of experiments, upon a limited range of species. Recently, there have been preliminary indications of humoral mediators of phagocytosis in insects which could operate by means other than opsonization. Mohrig and Schittek (1979) have isolated from the hemolymph of Galleria mellonella (L.) a complex of factors induced by the injection of readily encapsulated particles (latex beads). These factors confer upon a naive larva the ability to clear its hemolymph of the normally unphagocytizable Bacillus thuringiensis subtoxicus. It remains to be proven, however, that this clearance is accomplished by phagocytosis. Wago (1980) found that a component of Bombyx mori (L.) serum causes the elongation of hemocytic filipodia, an action which could facilitate phagocytosis. Regardless of the role of serum factors it seems certain, from the evidence presented above, that phagocytosis in arthropoda can proceed in their absence. Recognition of foreignness must be made by a component of the plasmatocyte membrane. The possible mechanisms of this recognition will be discussed in a later section. CELLULAR ENCAPSULATION Most arthropods have the ability to form a tight, multilayered hemocytic capsule around macroscopic parasites and pathogens. This process, called encapsulation, often kills the offending organism, or at least restricts its growth and movement. Theories advanced to explain the mortality of encapsulated organisms include asphyxiation, buildup of wastes, and toxic action of quinones (Salt, 1963;Nappi, 1977; Poinar et al., 1979). The quinones are precursors of melanin which often forms in the capsule. The structure of hemocytic capsules is fairly consistent in all arthropods in which it has been described. A typical capsule consists of 5-30 layers of hemocytes packed tightly together with few or no intercellular spaces. Desmosomes and gap and sep-

3 CELLULAR IMMUNE RESPONSES IN ARTHROPODA 187 tate junctions are sometimes observed between the cells (Baerwald, 1979). The innermost cell layer adheres tightly to the foreign surface, and is often extremely flattened against it. Many of the overlying layers are similarly flattened, so that the inner portions of some capsules resemble a myelin sheath. Capsule morphology varies from species to species in relatively minor aspects: extent of cell flattening, cell necrosis, presence or absence of melanin, and amount of extracellular material (Crompton, 1967; Grimstone et al., 1967; Norton and Vinson, 1977; Poinar and Hess, 1977; Schmit and Ratcliffe, 1977). Capsule development The cells which comprise a capsule are mainly ameboid phagocytic cells which resemble (and may actually be) plasmatocytes. Ultrastructural studies of the early phases of encapsulation, however, indicate that a different class of hemocyte is responsible for the actual recognition of foreignness and initiation of encapsulation. These labile, generally nonphagocytic hemocytes have been termed hyaline cells, cystocytes, spherule cells, or, as they will be referred to in this paper, granulocytes. The use of a single term is for convenience; these cells may not all be homologous (Price and Ratcliffe, 1974). Ultrastructural evidence indicates that granulocytes rupture upon encountering a foreign surface. The released granular and/or electron-dense flocculent material seems to cause plasmatocytes to adhere to the foreign surface, and may contain the chemical signal which induces them to flatten and conjoin to form the capsule. The material may also contain enzymes and precursors for melanin synthesis (Reik, 1968, as summarized by Salt, 1970; Nappi and Streams, 1969; Unestam and Nylund, 1972; Brehelin et al., 1975; Schmit and Ratcliffe, 1977, 1978). The material released by the granulocytes will thus be referred to as "encapsulation-promoting factors" (EPF). What is the nature of EPF? Despite the important effects of EPF upon cellular behavior, their composition has been characterized in only the most general terms. The labile hemocytes of arthropoda have been demonstrated histochemically to contain acid or neutral mucopolysaccharides and glyco- or mucoproteins (Ashhurst and Richards, 1964; Dumont et al., 1966; Gupta and Sutherland, 1967; Wood?/al., 1971). Reik (1968; summarized by Salt, 1970) implicated mucopolysaccharides in the early stages of capsule formation in Ephestia kuehniella Zeller. A precise determination of the nature of EPF and their action upon target cells awaits purification of EPF, a feat which would require an unambiguous bioassay. We have recently developed a technique which may fill this requirement, and would like to digress briefly in order to describe it. An in vitro encapsulation system Short-term culture of fresh hemolymph of Heliothis virescens (F.) and H. zea (Boddie) can be employed in the study of cellular encapsulation in vitro. The hemolymph is diluted 3-1 OX in Grace's insect tissue culture medium containing phenylthiourea at 8% of saturation. If foreign objects (e.g., cotton thread) are present in the medium, flocculent hemocytic aggregations form around them almost immediately. The aggregations become progressively denser and more sharply defined until, at ca. 2 hr, a classic hemocytic capsule surrounds the foreign object. At 5 hr, the capsule bears strong ultrastructural resemblance to that formed in vivo by H. virescens (Ratner and Vinson, in preparation). When no foreign object is present, the hemocytes aggregate spontaneously, eventually forming smooth, rounded nodules. This response may be triggered by the foreign surface of the culture dish, or by factors exuded from injured tissues during the bleeding process (see Cherbas, 1973). If the addition of a foreign object is delayed until after nodule formation is well under way (ca. 30 min), the object is not encapsulated (Fig. 1 A). Thus, the system rapidly becomes exhausted, probably because of the commitment of cells, and perhaps other components, to the nodules. Both encapsulation and nodule forma-

4 188 S. RATNER AND S. B. VINSON ADD FOREIGN OBJECT ENCAPSULATION OF OBJECT BEGINS FRESHLY DRAWN HEMOLYMPH + MEDIUM +0.05% TRYPS1N + FOREIGN OBJECT OBJECT NOT ENCAPSULATED I ADDTRVPSIN I INHIBITOR I INCUBATE I 2 KR ] OBJECT NOT ENCAPSULATED ' FRESHLY DRAWN HEMOLYMPH + MEDIUM ADO LYSATE AGGREGATION OF HEMOCYTES BEGINS OBJECT NOT ENCAPSULATED INCUBATE 15 M1N ENCAPSULATION OF OBJECT BEGINS WITHHOLD FOREIGN OBJECT ADD FOREIGN,.OBJECT ADO SERUM OR MEDIUM INCUBATE 12 HR OBJECT NOT ENCAPSULATED FIG. 1. In vitro demonstration of an encapsulationpromoting factor released from hemocytes of Heliothis virescens. A. Behavior of an untreated culture. B. Behavior of a gently trypsinized culture (Ratner and Vinson, in preparation). tion can be prevented, however, by a brief, mild trypsinization of the culture (soybean trypsin inhibitor is used to control the duration of trypsinization) (Fig. IB). The ability to encapsulate can subsequently be restored to the culture by the addition of lysate prepared by the ultrasonic disruption of fresh, whole hemolymph (Fig. IB). The addition of freshly prepared serum or Grace's medium has no restorative effect. It therefore appears that mild trypsinization prevents encapsulation not by altering components of the hemocyte membrane, but by digesting a protein released by the hemocytes an EPF (stronger trypsinization does appear to damage the hemocyte membrane, since it renders the culture unable to encapsulate, even after the addition of lysate). This is, to our knowledge, the first experimental evidence that such a factor can actually play a role in encapsulation. Neuraminidase and hyaluronidase do not prevent encapsulation so, if the in vitro system can be taken as a guide, the crucial components of Heliothis EPF are proteins, not mucopolysaccharides. It should be possible to assay hemolymph fractions for EPF activity, and eventually purify an EPF, using trypsin-inactivated in vitro systems. What is the mechanism of EPF release? It is desirable to know what cellular processes cause granulocytes to rupture after the initial recognition of foreignness. This knowledge might be used in devising a method of inducing inappropriate EPF release in pest arthropods, in effect turning their own immune systems against them. Some preliminary findings about the process have emerged from the in vitro Heliothis system. When the hemolymph culture is held at 4 C, the usual aggregation of hemocytes still forms and clings to a foreign object, but it will not mature into a tight, smoothcontoured capsule unless the culture is subsequently warmed to room temperature. Inhibitors of glycolysis and oxidative phosphorylation also prevent capsule maturation, but not the initial aggregation of hemocytes. It seems, then, that the recognition of foreignness and consequent release of EPF are passive processes, while the subsequent consolidation of the capsule requires metabolic energy. Capsule initiation is also unaffected by the presence of membrane-stabilizing drugs such as corticosteroids and local anesthetics (Ratner and Vinson, in preparation). Direct microscopic observation confirms that these compounds are ineffective in preserving granulocyte structure in vitro. Since the drugs are thought to act by displacing membrane-bound calcium and ATP-ase (Seeman, 1970), these substances are apparently not involved in granulocyte rupture in Heliothis. Obviously, our knowledge of the mechanisms of EPF release are at present rudimentary. Much may be revealed by tests of other membrane-active compounds. Such tests are best done in vitro, to ensure that the compounds reach the

5 CELLULAR IMMUNE RESPONSES IN ARTHROPODA 189 hemocytes in known forms and concentrations. WHAT ARE THE PRINCIPLES OF NON-SELF RECOGNITION IN ARTHROPODA? We have seen evidence that two types of arthropodan hemocytes react to a foreign surface: the phagocytic plasmatocytes and the labile granulocytes. What aspects of a surface do these cells actually recognize as foreign? Salt (1970) hypothesized that, to an insect immune system, "non-self is equivalent to anything other than intact, native basement membrane. His hypothesis stemmed from the fact that, except for parasites and pathogens which have developed specific counteracting strategies, few foreign surfaces prove immunologically inert to arthropods. Even materials such as glass and polyfluorocarbons elicit a response. Normally tolerated allografts, however, are encapsulated if their basement membrane is damaged. Salt further hypothesized that the immunologically important components of basement membrane are acid mucopolysaccharides. There have been some criticisms (Crossley, 1975), but Salt's hypothesis is still the best available. It does not, however, address the specific molecular principles of non-self recognition. Initial recognition offoreignness Since the arthropodan immune system can react against inorganic surfaces, it is obvious that the initial recognition of foreignness cannot be explained completely by models involving lock-and-key interactions between complex biogenic molecules (e.g., the glycosyl transferases of Roseman, 1974, and Parish, 1977). Such highly specific factors may come into play in secondary reactions, or in special cases, as we shall see below; but the initial recognition of foreignness must involve less specific, "physicochemical" properties of a surface. Electrostatic charge seems to be such a property. A growing body of evidence suggests that hemocytes form capsules on positively charged surfaces, but not neutral or negatively charged ones (Walters and Williams, 1966; Vinson, 1974; Dunphy and Nolan, 1980). This finding is consistent with Salt's hypothesis since, at physiological ph, basement membrane is negatively charged (Vinson, 1974). Also negatively charged are some of the coatings and membranes which protect parasites from encapsulation (e.g., Brennan and Cheng, 1979). It is not known whether surface charge similarly affects phagocytosis by arthropodan hemocytes. It is interesting that cationic substances increase the phagocytosis of bacteria by human neutrophils in an opsonin-free system (Pruzanski and Sato, 1978). In a different system, bacteria with surfaces more hydrophobic than neutrophils were phagocytized more readily than those with more hydrophilic surfaces (Van Oss and Gilman, 1972). It should be possible to use similar experiments to determine the effect of surface charge on phagocytosis by arthropodan hemocytes. A model of non-self discrimination based on nonspecific physicochemical factors can adequately account for all known properties of arthropodan cellular immune systems. For example, in a single host, the severity of encapsulation of transplanted tissue can vary according to the species of the donor (Lackie, 1981a; S. E.Jones, personal communication). These differences in severity could be related to the number of granulocyte ruptures provoked by each transplant. This in turn could be a reflection of different distributions and strengths of charged groups on the basement membranes of the various donors (Lackie, 1981a). Recognition in a nonspecific system need not occur only at the hemocyte membrane. Soluble factors could stick to surfaces with foreign physicochemical characteristics, and bind to hemocytes by cytophilic groups. The fact remains, though, that arthropodan hemolymph does contain molecules which have highly specific binding affinities. Hemagglutinins, with their saccharide-specific binding, are an example (Yeaton, 1981). What role could such molecules play in an immune system which shows so little in vivo specificity? One possiblity is that they are secondary factors, released or activated only after the initial recognition of foreignness. For example, one would

6 190 S. RATNER AND S. B. VINSON expect that EPF must bind specifically to some components of the plasmatocyte membrane, since capsules are composed almost exclusively of these cells. This specific binding could not occur, however, until the EPF is released by granulocyte rupture, i.e., after foreignness has been recognized. Since granulocytes are extremely fragile, it would not be surprising if their released factors showed up as serum components in extracted hemolymph. Another possibility is raised by Lackie (1981a), who suggests that specifically binding factors actually can participate in the initial recognition of foreignness. In Lackie's "two-tier" hypothesis, the nonspecific recognition system the first "tier" could be supplemented by factors which trigger cellular responses against more specific stimuli. A two-tier system is an entirely plausible concept. It might arise if the physicochemical surface properties of a damaging parasite were close enough to those of the host to prevent recognition by host granulocytes. One possible consequence of the resulting selective pressure upon the host might be the development of a receptor which could recognize other, possibly more specific, aspects of the parasite's surface. This new receptor could be soluble or bound to hemocyte membranes. What components of arthropodan hemolymph might act as specific recognition factors in the type of system proposed by Lackie? One possibility is a group of enzymes, the phenoloxidases, which catalyze the oxidation of tyrosine and other phenolic compounds into dopa, a precursor of melanin. A connection between phenoloxidases and encapsulation has long been suspected, since inhibitors of melanin formation have sometimes proven inhibitory to encapsulation as well (Salt, 1956; Brewer and Vinson, 1971; Nappi, 1973; Beresky and Hall, 1977). The hemolymph of some arthropods contains an inactive prophenoloxidase which can be activated by fairly specific stimuli such as fungal wall polysaccharides and yeast zymosan (Pye, 1974; Soderhall and Unestam, 1979: Ashida and Dohke, 1980). Interestingly, the activated enzyme adheres readily to fungal cell walls and abiotic surfaces such as glass. The inactive proenzyme is not adhesive (Soderhall et al, 1979). Phenoloxidases, then, exhibit the type of behavior one would expect from an arthorpodan nonself recognition factor, although it remains to be proven whether they actually perform this function. The fact that the phenoloxidases and their substrates are often found localized within hemocytes may mean that they come into play only after granulocyte rupture. Another class of molecules which might act as the "specific tier" in a two-tier system is the hemolymph agglutinins. Many agglutinins bind polyvalently to specific oligosaccharide residues, which is why they are often detected by their ability to clump mammalian erythrocytes and bacteria. In some molluscs, agglutinins have been proven to act as opsonins (Hardy et al., 1977). In arthropoda, only circumstantial evidence links agglutinins with cellular immune response. The most convincing evidence was obtained by Lackie (19816, c). Comparing an insect with low xenograft tolerance (Periplaneta americana [L.]) to one with greater tolerance (Schistocerca gregaria Forsk.), Lackie found that the insect with better non-self discrimination possesses a wider spectrum of specific agglutinins. Even more intriguing was the finding that the ability to agglutinate Hymenolepis diminuta oncospheres in vitro is correlated with the ability to encapsulate them in vivo. Lackie suggests that the agglutinins might act as hemocyte-bound receptors, an idea partly inspired by the immunofluorescent localization of Leucophaea maderae Fabr. agglutinin on hemocyte membranes (Amirante and Mazzalai, 1978). Definitive proof of the involvement of arthropodan agglutinins in the recognition of foreignness is still lacking. The blocking of phagocytosis by antibody to purified agglutinins would constitute such a proof. A third possibility must be considered: The specifically binding hemolymph factors may not be involved in cellular immunity at all. Instead, they may constitute a strictly humoral antimicrobial system, or they may perform presently unsuspected functions unrelated to immunity.

7 CELLULAR IMMUNE RESPONSES IN ARTHROPODA 191 INTERRELATIONSHIPS BETWEEN CELLULAR IMMUNE RESPONSES Arthropodan hemocytes participate in a variety of multicellular reactions other than encapsulation. In nodule formation, conglomerates of bacteria or other particulates are sealed into a hemocytic sheath (Ratcliffe and Gagen, 1977). In certain types of coagulation, hemocytes and extracellular materials form a mass which can be an early stage in the process of wound repair (Clark and Harvey, 1965; Gregoire, 1974). The formation of so-called "tumors" in insects is, in most cases, the walling off of pathological tissues by hemocytes (Perotti and Bairati, 1968). It has often been suggested that these reactions are interrelated, since they all show similar patterns of development (reviewed in insects by Ratcliffe and Rowley, 1979; for crustaceans, see Schapiro, 1975, and Ghidalia et al., 1981). The first event is usually the release of the contents of labile hemocytes at or near the site of the triggering stimulus. Mucopolysaccharides and mucoproteins have been detected histochemically in the released material. Nonlabile hemocytes are subsequently ensnared in, or perhaps attracted to, this material. These hemocytes eventually become organized into a multi-layered sheath. There are exceptions (Crossley, 1975), but that is the general pattern, based primarily on electron microscopic observations. There are also experimental results which suggest that a variety of cellular immune reactions are all potentiated by substances liberated from hemocytes. Hemocyte lysate promotes encapsulation in the in vitro Heliothis system (Ratner and Vinson, in preparation); simulates in vitro clotting of the hemolymph of the crab Caranus maena (Bang, 1971); and, when coated on bacteria, speeds their clearance (presumably phagocytic) from the circulation of the crayfish P. bicarinatus (Tyson and Jenkin, 1973). None of this information provides an answer to a fundamental question: To what extent do all the cellular reactions, including phagocytosis, share a common underlying mechanism? This is an important question, for if distinctions between the reactions are only superficial, then coordination between experimenters is needlessly inhibited. Experimental designs may even be distorted. For example, in the in vitro Heliothis system, there seems to be equivalence between the immune and hemostatic reactions. If no foreign object is present, a nebulous clot forms and eventually contracts into a nodular mass. If a foreign object is present, the clot forms around it and matures into a capsule. Thus, if the "clot"were allowed to form before the start of an experiment (a practice not unknown in in vitro studies of hemocyte behavior), the remaining culture would be seriously depleted of immunoreactive cells. One way to test for similarities among the mechanisms of the cellular responses is to look for similarities among the extracellular factors associated with those responses. Two types of factors which have often been analyzed are the clottable proteins and the hemagglutinins, so they might be good subjects for comparison. At present, though, it is hard to draw comparisons, because of the diverse analytical methods used by different groups of experimenters. The only generalization that emerges from a survey of the literature is that both types of factor tend to be proteins of molecular weight 200, ,000, composed of noncovalently linked subunits with weights of 20,000-50,000 (Marchalonis and Edelman, 1968; Doolittle and Fuller, 1972; Hall and Rowlands, 1974; Donlon and Wemyss, 1976; Shishikura et al., 1977; Brehelin, 1979). Obviously, meaningful comparisons are impossible at present. They await the purification of more factors, including EPF, as well as a coordination of analytical techniques. Another way to approach the question of interrelationships between cellular immune response mechanisms is to compare the metabolic activities associated with them, especially the activation of biochemical pathways. This strategy was employed by Anderson et al. (1973) in their comparison of phagocytotic processes in invertebrates and mammals. It has not yet been applied to a comparison of arthropodan multicellular reactions.

8 192 S. RATNER AND S. B. VINSON PROSPECTS FOR FUTURE RESEARCH It must by now be obvious to the reader that our present understanding of the early events in arthropodan immune response is rudimentary. We have a fairly good ultrastructural picture of what is going on, but tentativeness creeps into even the most general statements about the molecular nature of non-self recognition, and consequent hemocytic changes. Tools now becoming available can and should be used to deepen our knowledge of the mechanisms of arthropodan immune response. A variety of charged groups can now easily be attached to inert substrates (see, for example, Carlsson et al., 1979). This should make possible a finer determination of the effect of surface charge on phagocytosis and encapsulation. Investigations of reactions against transplants are growing more numerous, and more subtle. It would be helpful if biochemical analyses of the transplant surfaces could be coordinated with these investigations. Another research opportunity is afforded by the viruses associated with some hymenopterous parasitoids of insects. These viruses suppress the host's immune response to the parasitoid's egg (Edson et al., 1981). The determination of their mode of action may reveal underlying mechanisms of that response. In vitro systems have proven valuable in the detection and study of the serum and cellular factors involved in phagocytosis. A high priority should now be placed upon the improvement of methods of delaying and controlling the multicellular immune reactions in vitro, so that their associated factors, such as EPF, can be isolated, purified, and tested. Such in vitro systems will also make it possible to determine whether substances which inhibit multicellular reactions in vivo (e.g., phenoloxidase inhibitors and viruses) act directly upon the hemocytes, or by some indirect mechanism. ACKNOWLEDGMENTS Approved as TA by the Director, Texas Agricultural Experiment Station. All programs and information of the Texas Agricultural Experiment Station are available without regard to race, ethnic origin, sex, or age. Supported in part by NSF Grant PCM REFERENCES Amirante, G. A. and F. G. Mazzalai Synthesis and localization of hemoagglutinins in hemocytes of the cockroach Leucophaea maderae L. Devel. Comp. Immunol. 2: Anderson, R. S Biochemistry and physiology of invertebrate macrophages in vitro. In L. A. Bulla, Jr. and T. C. Cheng (eds.), Comparative pathobiology, Vol. 3, pp Plenum Press, New York. Anderson, R. S., N. K. B. Day, and R. A. Good Specific hemagglutinin and a modulator of complement in cockroach hemolymph. Infect. Immun. 5: Anderson, R. S., B. Holmes, and R. A. Good Comparative biochemistry of phagocytizing insect hemocytes. Comp. Biochem. Physiol. 46B: Ashhurst, D. E. and A. G. Richards Some histochemical observations on the blood cells of the wax moth, Galleria mellonella L. J. Morphol. 114: Ashida, M. and K. Dohke Activation of prophenoloxidase by the activating enzyme of the silkworm, Bombyx mori. Insect Biochem. 10: Baerwald, R. J Fine structure of hemocyte membranes and intercellular junctions formed during hemocyte encapsulation. In A. P. Gupta (ed.), Insect hemocytes, pp Cambridge Univ. Press, Cambridge. Bang, F. B A cellular factor in crab amoebocytes which stimulates in vitro clotting of crab blood. J. Invertbr. Pathol. 18: Beresky, M. A. and D. W. Hall The influence of phenylthiourea on encapsulation, melanization, and survival in larvae of the mosquito Aedes aegypti parasitized by the nematode Meoplectana carpocapsae. J. Invertbr. Pathol. 29: Boman, H. G. and D. Hultmark Cell-free immunity in insects. Trends in Biochem. Sci. 6: Brehelin, M Hemolymph coagulation in Locusta migratoria: Evidence for a functional equivalent of fibrinogen. Comp. Biochem. Physiol. 62B: Brehelin, M., J. A. Hoffman, G. Matz, and A. Porte Encapsulation of implanted foreign bodies in Locusta migratoria and Melontha melontha. Z. Zellforsch. 160: Brennan, B. M. and T. C. Cheng Resistance of Moniliformis dubius to the defense reactions of the American cockroach Periplaneta americana. J. Invertbr. Pathol. 26: Brewer, F. D. and S. B. Vinson Chemicals affecting the encapsulation of foreign material in an insect. J. Invertbr. Pathol. 18: Carlsson, J., D. Gabel, E. Larsson, J. Ponten, and B. Westermark Protein-coated agarose surfaces for attachment of cells. In vitro 15:

9 CELLULAR IMMUNE RESPONSES IN ARTHROPODA 193 Cherbas, L The induction of an injury factor in cultured hemocytes from satuniid pupae. J. Insect Physiol. 19: Clark, R. M. and W. R. Harvey Cellular membrane formation by plasmatocytes of diapausing cecropia pupae. J. Insect Physiol. 11: Crompton, D. W. T Studies on the haemocytic reaction of Gammarus species and its relationship to Polymorphic minutus (Acanthocephala). Parasitol. 57: Donlon, W. C. and C. T. Wemyss Analysis of the hemagglutinin and general protein element of the hemolymph of the West Indian leaf cockroach, Blaberus craniifer. J. Invertbr. Pathol. 28: Doolittle, R. F. and G. M. Fuller Sodium dodecysulfate-polyacrylamide gel electrophoresis studies on lobster fibrinogen and fibrin. Biochem. Biophys. Acta 263: Dumont, J. N., E. Anderson, and G. Winner Some cytologic characteristics of the hemocytes of Limulus during clotting. J. Morphol. 119: Dunphy, G. B. and R. A. Nolan Response of eastern hemlock looper hemocytes to selected stages of Entomophthora egressa and other foreign particles. J. Invertbr. Pathol. 36: Edson, K. M., S. B. Vinson, D. B. Stoltz, and M. D. Summers Virus in a parasitoid wasp: Suppression of the cellular immune response in the parasitoid's host. Science 211: Ghidalia, W., R. Vendrely, C. Montmory, Y. Coirault, and M. O. Brouard Coagulation in decapod Crustacea. Comparative study of the clotting process in species from groups A, B, and C. J. Comp. Physiol. 142: Gregoire, Ch Hemolymph coagulation. In M. Rockstein (ed.), The physiology of Insecta, Vol. 5, pp Academic Press, New York. Grimstone, A. V., S. Rotheram, and G. Salt An electron-microscope study of capsule formation by insect blood cells. J. Cell Sci. 2: Gupta, A. P. and D.J. Sutherland Phase contrast and histochemical studies of spherule cells in cockroaches. Ann. Ent. Soc. Am. 60: Hall.J. C. and D. T. Rowlands, Jr Heterogeneity of lobster agglutinins: I. Purification and physicochemical characterization. Biochemistry 13: Hardy, S. W., T. C. Fletcher, andj. A. Olafson Aspects of cellular and humoral defense mechanisms in the Pacific oyster, Crassostrea gigas. In J. B. Solomon and J. D. Horton (eds.), Developmental immunobiology, pp Elsevier/North Holland, Amsterdam. Karp, R. D. and L. A. Rheins A humoral response of the American cockroach to honeybee toxin demonstrating specificity and memory. In M. J. Manning (ed.), Phytogeny of immunological memory, pp Elsevier/North Holland, Amsterdam. Lackie, A. M Invertebrate immunity. Parasitol. 80: Lackie, A. M. 1981a. Immune recognition in insects. Devel. Comp. Immunol. 5: Lackie, A. M The specificity of the serum agglutinins of Periplaneta americana and Schistocerca gregaria and its relationship to the insect's immune response. J. Insect Physiol. 27: Lackie, A. M. 1981c. Humoral mechanisms in the immune response of insects to larvae of Hymenolepis diminuta (Cestoda). Parasite Immunol. 3: Marchalonis.J. J. and G. M. Edelman Isolation and characterization of a hemagglutinin from Limulus polyphemus. J. Molec. Biol. 32: McKay, D. and C. R.Jenkin Immunity in the invertebrates. The role of serum factors in the phagocytosis of erythrocytes by haemocytes of the freshwater crayfish (Parachaeraps bicarinatus). Austral. J. Exp. Biol. Med. Sci. 48: McKay, D., C. R. Jenkin, and D. Rowley Immunity in the invertebrates. I. Studies on the naturally occurring haemagglutinins in the fluid from invertebrates. Austal. J. Exp. Biol. Med. Sci. 47: Mohrig, W. and D. Schittek Phagocytosis stimulating mediators in insects. Acta Biol. Med. Germ. 38: Nappi, A. J The role of melanization in the immune reaction of larvae of Drosophila algonquin against Pseudocoila bochei. Parasitology 66: Nappi, A. J Comparative ultrastructural studies of cellular immune reactions and tumorigenesis in Drosophila. In L. A. Bulla, Jr. and T. C. Cheng (eds.), Comparative pathobiology, Vol. 3, pp Plenum Press, New York. Nappi, A. J. and F. A. Streams Haemocytic reactions of Drosophila melanogaster to the parasites Pseudocoila mellipes and P. bochei. J. Insect Physiol. 15: Norton, W. N. and S. B. Vinson Encapsulation of a parasitoid egg within its habitual host: An ultrastructural investigation. J. Invertbr. Pathol. 30: Parish, C. R Simple model for self-non-self discrimination in invertebrates. Nature 267: Paterson, W. D. and J. E. Stewart In vitro phagocytosis by hemocytes of the American lobster (Homarus americanus).]. Fish. Res. Board Can. 31: Perotti, M. E. and A.J. Bairati Ultrastructure of the melanotic masses in two tumorous strains of Drosophila melanogaster.]. Invertbr. Pathol. 10: Poinar, G. O., Jr. and R. T. Hess Cellular responses in decapod crustaceans to Ascarophis spp. (Spirurida:Nematoda). In L. A. Bulla,Jr. and T. C. Cheng (eds.), Comparative pathobiology, Vol. 3, pp Plenum Press, New York. Poinar, G. O.,Jr., R. T. Hess.andJ.J. Petersen Immune responses of mosquitoes against Romanomermis culicivorax (Mermithida:Nematoda). J. Nematol. 11: Price, C. D. and N. A. Ratcliffe A reappraisal of insect haemocyte classification by the examination of blood from fifteen insect orders. Z. Zellforsch. 147: Pruzanski, W. and S. Sato The influence of

10 194 S. RATNER AND S. B. VINSON natural and synthetic cationic substances on phagocytic activity of human polymorphonuclear cells. Exp. Cell Res. 117:1-13. Pye, A. E Microbial activation of prophenoloxidase in immune insect larvae. Nature 251: Rabinovitch, M. and M. DeStefano Interactions of red cells with phagocytes of the wax moth (Galleria mellonella L.) and mouse. Exp. Cell Res. 59: Ratcliffe, N. A. and S. J. Gagen Studies on the in vivo cellular reactions of insects: An ultrastructural analysis of nodule formation in Galleria mellonella. Tissue and Cell 9: Ratcliffe, N. A. and A. F. Rowley In vitro phagocytosis of bacteria by insect blood cells. Nature 252: Ratcliffe, N. A. and A. F. Rowley Role of hemocytes in defense against biological agents. In A. P. Gupta (ed.), Insect hemocytes, pp Cambridge University Press, Cambridge. Reik, L Contacts between insect blood cells, with special reference to the structure of capsules formed about parasites. M.Sc. Diss., University of Cambridge. Roseman, S Complex carbohydrates and intercellular adhesion. In A. A. Moscona (ed.), The cell surface in development, pp John Wiley and Sons, New York. Rowley, A. F. and N. A. Ratcliffe A histological study of wound healing and hemocyte function in the wax moth Galleria mellonella. J. Morphol. 157: Rowley, A. F. and N. A. Ratcliffe Insect erythrocyte agglutinins. In vitro opsonization experiments with Clitumnus extradentatus and Periplaneta americana haemocytes. Immunology 40: Salt, G The defense reactions of insects to metazoan parasites. Parasitology 53: Salt, G The cellular defense reactions of insects. Cambridge University Press, Cambridge. Schapiro, H. C, J. F. Steenburgen, and Z. A. Fitzgerald Hemocytes and phagocytosis in the American lobster, Homarus americanus. In L. A. Bulla, Jr. and T. C. Cheng (eds.), Comparative pathobiology, Vol. 3, pp Plenum Press, New York. Schtnit, A. R. and N. A. Ratcliffe The encapsulation of foreign tissue implants in Galleria mellonella larvae. J. Insect Physiol. 23: Scott, M. T Recognition of foreignness in invertebrates. II. In vitro studies of cockroach haemocytes. Immunology 21: Seeman, P Membrane expansion and stabilization by anesthetics and other drugs. In L. Bolis, A. Katchalsky, R. D. Keynes, W. R. Lowenstein, and B. A. Pethica (eds.), Permeability and function of biological membranes, pp North Holland, Amsterdam. Shishikura,F.,J.Chiba,andK. Sekiguchi Two types of hemocytes in localization of clottable protein in Japanese horshoe crab, Tachypleus tridentatus.]. Exp. Zool. 201: Soderhall, K., L. Hall, T. Unestam, and L. Nyhlen Attachment of phenoloxidase to fungal cell walls in arthropod immunity. J. Invertbr. Pathol. 34: Soderhall, K. and T. Unestam Activation of serum prophenoloxidase in arthropod immunity. The specificity of cell wall glucan activation by purified fungal cell wall glycoproteins of crayfish phenoloxidase. Can. J. Microbiol. 25: Tyson, C.J. and C. R.Jenkin The importance of opsonic factors in the removal of bacteria from the circulation of the crayfish (Parachaeraps bicarinatus). Austral. J. Exp. Biol. Med. Sci. 51: Unestam, T. and J.-E. Nylund Blood reactions in vitro in a crayfish against a fungal parasite, Aphanomyces astacus. J. Invertbr. Pathol. 19: VanOss, C.J. and C. F. Gillman Phagocytosis as a surface phenomenon. I. Contact angles and phagocytosis of non-opsinized bacteria. J. Reticuloend. Soc. 12: Vinson, S. B The role of foreign surface and female parasitoid secretions on the immune response of an insect. Parasitol. 68: Wago, H Theroleof hemolymph in the initial cellular attachment to foreign cells by the hemocytes of the silkworm Bombyx mori. Devel. Comp. Immunol. 5: Wago, H. and Y. Ichikawa In vitro analyses of cyto-adherence phenomenon between Bombyx mori hemocytes and goose erythrocytes in relation to the larval development. Appl. Ent. Zool. 14: Walters, D. R. and C. M. Williams Reaggregation of insect cells as studied by a new method of tissue and organ culture. Science 154: Whitcomb, R. F., M. Shapiro, and R. R. Granados Insect defense mechanisms against microorganisms and parasitoids. In M. Rockstein (ed.), The physiology of Insecta, Vol. 5, pp Academic Press, New York. Wood, P. J., J. Podlewski, and T. E. Sherk Cytochemical observations of hemolymph cells during coagulation in the crayfish Orconectes virilis.]. Morphol. 134: Yeaton, R. W Invertebrate lectins. I. Occurrence. Devel. Comp. Immunol. 5: