Bivalve mollusc hemocyte behaviors: characterization of hemocyte aggregation and adhesion and their inhibition in the California mussel (Mytilus californianus).
Cell aggregation (cell clumping) in bivalves was first reported by Geddes (1880, cited by Narain, 1973). In molluscs, it has been presumed that clump formation is involved in hemostasis and wound healing (Bang, 196l; Sparks, 1972; Sminia, 1981). However, hemocyte aggregation in bivalves differs from blood clotting in vertebrates in that no extracellular fibers are formed. The aggregation of hemocytes in bivalves is reversible, and most aggregated cells later disperse, re-entering the circulatory system as wound repair progresses (Feng and Feng, 1974). Mussel hemocytes will spontaneously aggregate in vitro. When hemolymph is removed from a mussel, the free hemocytes form aggregates during the bleeding or immediately thereafter. When the aggregates and the free cells settle down on a foreign surface, they adhere (cell-substratum adhesion) and usually spread, and later migrate away. Although the phenomena of clump formation and hemocyte adhesion and spreading have been described both in vivo and in vitro in bivalves (Drew, 1910, cited by Narain, 1973; Dundee, 1953; Bang, 1961; Sparks, 1972; Cheng, 1981), few attempts have been made to unravel the operative mechanisms (Seifert, 1983). To study this rapid, spontaneous cell adhesion, it was first necessary to find effective inhibitors that could reversibly block the reaction. In this study, we used hemocytes of the California mussel, Mytilus californianus, to (1) identify inhibitors of hemocyte aggregation or adhesion in vitro, and then (2) investigate differences between hemocyte clump formation (cell-cell interaction) and hemocyte adhesion (cell-substratum interaction) in vitro.
Materials and Methods
The following chemicals, tested at the indicated concentrations for their abilities to inhibit aggregation and adhesion of mussel hemocytes, were obtained from Sigma Co. Protease inhibitors: phenylmethylsulfonylfluoride (PMSF, 20 [[micro]gram]/ml), soybean trypsin inhibitor (20 [[micro]gram]/ml), aprotinin (10 [[micro]gram]/ml), leupeptin (10 [[micro]gram]/ml), bovine [[Alpha].sub.2]-macroglobulin (1 mg/ml); glycosaminoglycans (each at 1 mg/ml): hyaluronic acid, chondroitin sulfate (types A and C). Others: heparin (10 units/ml), protamine (1 mg/ml), poly-L-lysine (1 mg/ml), poly-L-glutamic add (1 mg/ml). Ethylenediaminetetraacetic acid (EDTA), caffeine, cytochalasin B and nor-ethyl-maleimide (NEM) were tested at several concentrations stated in the results. Most of the suspected inhibitors were dissolved in CMTBS ([Ca.sup.++], [Mg.sup.++] Tris-buffered saline: 10 mM Ca[Cl.sub.2], 60 mMM Mg[Cl.sub.2], 50 mM Tris HCl, NaCl to 960 mOsm), but PMSF, pepstatin, aprotinin, leupeptin, and cytochalasin B were first dissolved in DMSO, then mixed with CMTBS. The final concentration of DMSO in the medium was [greater than]0.5%. Peptides containing the Arg-Gly-Asp (RGD) amino acid sequence were kindly provided by Dr. D. W. Barnes, Biochemistry, Oregon State University. BCA protein assay reagents were purchased from Pierce. All other chemicals were purchased from Sigma Co. The suspected inhibitors were dissolved in CMTBS ([Ca.sup.++], [Mg.sup.++]Tris-buffered saline: 10 mM Ca[Cl.sub.2], 60 mM Mg[Cl.sub.2], 50 mM Tris HCl, NaCl to 960 mOsm). The EDTA was dissolved in calcium- and magnesium-free Tris-buffered saline (CMFTBS). Hydrophobic chemicals were first dissolved in DMSO, and then mixed with CMTBS. The final concentration of DMSO in treated hemolymph was lower than 0.5%. The pH value of solutions was adjusted to 7.4 with NaOH, and osmolarities were checked by freezing-point osmometer before they were mixed with hemolymph.
Individual Mytilus californianus larger than 8 cm in length were collected monthly from the rocky intertidal zone at Seal Rock State Park (15 miles south of Newport, Oregon). On the same day, these mussels were transferred to a filtered, recirculating, continuously aerated seawater system that was maintained at pH 7.6, nitrate [less than] 10 ppm, nitrite [less than] 0.2 ppm, 15 [degrees] C, and close to normal salinity. Mussels were held in this system for at least three days before being used as a source of hemolymph.
A plastic rod (3 mm diameter) was inserted between the two shells to prevent them closing, and seawater was drained from the mantle cavity. Animals were then transferred to a cold room (4 [degrees] C). Hemolymph was collected from the posterior adductor muscle using a cooled sterile syringe with 18G 1 1/2[inches] needle. With the needle removed, the colorless (slightly opalescent) hemolymph was immediately transferred to cooled sterile tissue culture tubes (Falcon) for further treatment. Hemolymph was collected from each animal only once.
Coating of surfaces for cell adhesion assay
Except for agarose, the coating process was as follows: the solution of coating material (see Table I) was loaded on the clean surface for 30 min at room temperature. Then the excess was washed and replaced by CMTBS. For agarose coating, melted agarose was added to washed eight-well slides (5 [[micro]liter]/well; Cel-Line, New Jersey), and then mounted by a cover slip which was held off the slide by two small pieces of coverslip. The coverslips were removed after keeping the slide at 4[degrees]C for a few minutes. The thin layer of agarose allowed satisfactory microscopic observation.
Inhibitor screening assay
The solutions of suspected inhibitors (see Results) were aliquoted into siliconized microtubes (PGC Scientifics, Maryland), and kept at 4 [degrees] C. Slides with eight wells, cleaned by immersion in acid alcohol overnight, were dried and placed in a humidity chamber that was also kept at 4 [degrees] C before use. As soon as the hemolymph had been transferred to a cooled sterile tissue culture tube, it was aliquoted and mixed (1: 1) with test solutions by vortexing for about 3 s. The treated hemolymph was then loaded onto the undersurface of an inverted, cooled eight-well slide (50 [[micro]liter]/well). Each sample was loaded in duplicate onto the same position on a different slide. These were maintained as hanging-drops during gyration at 100 rpm performed by Gyrotory[R] Shaker-Model G2 (New Brunswick Scientific Co., Inc., New Jersey) at room temperature (22 [+ or -] 1 [degrees] C). The purpose of using hanging-drops was to avoid cell adhesion to glass during gyration. Following 15 min gyration, the slide was carefully turned over. Two 1-mm-thick spacers were placed on the slide, followed by a coverslip (24 x 50 mm). Cell aggregation, cell adhesion, and cell spreading were observed under phase contrast microscopy.
Those inhibitors that interfered with cell aggregation or cell adhesion were selected for further studies. Due to occasional individual differences between mussels, each reagent was tested on at least three hemolymph samples taken from separate mussels at different times.
Assay for inhibition of cell aggregation
This assay followed the same protocol used in the screening test. After the slide was turned so the hemolymph was on the upper side, 100% formalin was added (1 [[micro]liter]/well) to fix the cells. The final concentration of formalin in hemolymph was 2%. Cells at the center of each preparation were then immediately dispersed in situ by gentle pipetting five times. Samples were then mounted by a coverslip (24 x 50 ram) held on two 1-mm-thick spacers. The extent of cell aggregation was assessed by counting the total free cell numbers in five different areas in each well. To obtain values for "no aggregation," we proceeded as follows: fresh hemolymph was mixed with 20% formalin (1: 1) to immediately block cell aggregation. The total (control) free cell number was counted in hemolymph fixed this way in 10% formalin as soon as it was taken from the mussel.
The percentage of cells remaining free = Total free cell numbers in test treatment/Total free cell numbers in fixed fresh hemolymph x 100
Assay for inhibition of cell adhesion
After hemolymph had been mixed with test solutions, it was then loaded (50 [[micro]liter]/well) into wells of a pre-cooled (4 [degrees] C), flat-bottom 96-well tissue culture plate (Corning, New York). Each treatment was loaded in triplicate. After 15 min at room temperature for cell adhesion to occur, the plasma and unattached cells were removed and each well was washed three times with CMTBS (100 = [[micro]liter]/well). As a measure of the number of remaining (adherent) cells, protein concentrations were measured by means of the BCA protein assay (Pierce). The adherent cells were first lysed in 20 [[micro]liter] cell lysis solution (2% [Na.sub.2]C[O.sub.3], 0.1 M NaOH) for at least 30 min at RT with occasional shaking by vortex, and then the fresh Pierce BCA protein assay reagent was added (100 [[micro]liter/well). After 30 min incubation at 60 [degrees] C, the plate was cooled to RT, and the absorbance of each well was read at 550 nm using a microtiter plate reader (Titertek Multiskan MCC/340).
The percentage of cell adhesion was calculated using the following equation:
[([A.sub.i] - [A.sub.b])/([A.sub.c.] - [A.sub.b])] x 100
[A.sub.i]: Absorbance of inhibitor-treated well
[A.sub.b] Absorbance of background well (lysis solution and BCA reagent only)
[A.sub.c]: Absorbance of CMTBS-treated well
According to this equation, the absorbance in background wells indicates 0% adhesion, and the absorbance in CMTBS-treated wells indicates 100% adhesion.
To correlate cell number with protein values, serial two-fold dilutions of hemolymph in CMTBS (50 [[micro]liter] each) were loaded in quintuplicate into wells of 96-well plates. After 15 min for adherence, all wells were washed three times with CMTBS. Then three were used for protein determinations as described above, and Hoechst 33342 (50 [[micro]liter], 20 [[micro]gram]/mL CMTBS) was added to the other two. Using an inverted fluorescence microscope, nuclei were counted in the Hoechst-stained wells corresponding to those for which protein values were also obtained.
Each experiment was repeated at least three times, using fresh hemolymph samples from different animals. Data are presented as mean [+ or -] S.D. Statistical analysis of data was performed using paired or unpaired Student's t-test as appropriate. Differences were considered significant when P [less than] 0.05.
Following the adhesion/spreading and aggregation aso says, hemocyte viability was determined by visual observation of hemocyte spreading, and by cell exclusion of propidium iodide. This test was performed only on the samples in which cell aggregation or adhesion was interrupted. Treated hemolymph (40 [[micro]liter]) was gently removed from well slides or culture plates, and replaced by 40 [[micro]liter] CMTBS. If the remaining cells had not aggregated or spread in 10 min, two possibilities were considered; either the treated cells had been killed by the chemical or the inhibitory effect of the chemical was irreversible. To determine whether the treated cells were alive or not, the propidium iodide staining protocol was followed: 1 [[micro]liter] propidium iodide solution (500/[[micro]gram]/ml in CMTBS) was added to each sample (50 [[micro]liter]). In cells which are dead, the dye enters and binds to nuclear DNA with high affinity. Stained nuclei generate bright red fluorescence under epifluorescence using a Zeiss microscope with appropriate filter set.
Normal hemocyte aggregation, adhesion, and spreading in vitro
Mytilus hemolymph contains three types of hemocyte (Moore and Lowe, 1977). Preliminary studies showed that hemocytes of distinct densities, separated on 60% Percoll, were all competent in adhesion and aggregation assays. All the following experiments were conducted with freshly taken, unseparated hemocyte populations.
When hemocytes were first removed from the mussels, they remained dispersed and were round [ILLUSTRATION FOR FIGURE 1A OMITTED]. However, they changed from round to elongated in one minute or less [ILLUSTRATION FOR FIGURE 1B OMITTED]. Sometimes aggregation occurred extremely rapidly during the bleeding, or while the hemolymph was being dispensed into the test tube. After initial cellular contact, cells were seen to pull together before the cells or aggregates began to spread. Aggregates formed this way without shaking were small [ILLUSTRATION FOR FIGURE 1C OMITTED], and the cell-cell binding strength was weak. The individual cells were distinguishable and could be easily dispersed from the aggregates by pipetting.
In contrast, clumps that formed during 10 or more minutes of shaking often contained more than ten thousand cells [ILLUSTRATION FOR FIGURE 2A OMITTED]. This kind of clump was resistant to mechanical dispersion by pipetting or vortexing. When cells or aggregates were allowed to settle down onto glass or plastic, they attached, flattened, and developed pseudopodia. During subsequent incubation, cells located at the surface of the clump and in contact with the substratum adhered to the substratum and migrated radially out of the clump [ILLUSTRATION FOR FIGURE 2B, C OMITTED]. The attached and spread cells moved by amoeboid locomotion. The cells possess a strong ability to adhere, spread, and migrate on foreign surfaces. Only on agarose substrata were such kinds of cell behaviors inhibited (Table I).
Assay of cell adhesion
The numbers of cells in individual wells and protein quantified in these wells were dependably correlated. When 50/[[micro]liter] of hemolymph (serially diluted twofold in CMTBS to 1/16) was held in wells of flat-bottomed 96-well plates for 15 min at 22 [+ or -] 1 [degrees] C, nuclear counts and protein values indicated 1215 cells yielding 69.5 pg protein in 1/16 strength hemolymph, 1597 cells yielding 134 pg protein in 1/8 strength hemolymph, and 6636 cells yielding 478.5 pg protein in 1/4 strength hemolymph. These values, from one experiment representative of two runs, indicate a protein value of approximately 71 pg protein per 1000 cells.
[TABULAR DATA FOR TABLE I OMITTED]
[TABULAR DATA FOR TABLE II OMITTED]
Influences of suspected inhibitors on hemocyte aggregation, adhesion, and spreading
From data obtained in screening assays, the tested reagents could be allocated into four categories. The first includes caffeine (25 mM) and N-ethyl maleimide (0.1 mM), which strongly inhibited both hemocyte aggregation and adhesion. The second category of chemicals poorly blocked hemocyte aggregation but significantly inhibited hemocyte adhesion. They include EDTA (60mM), and poly-L-glutamic acid (MW 50,000-100,000; 1 mg/ml). Cytochalasin B (5 [[micro]gram]/ml)is the only chemical in the third category: it effectively inhibited hemocyte aggregation, but poorly blocked hemocyte adhesion. The fourth group of chemicals had no detectable inhibitory effect on either hemocyte aggregation or adhesion. They included RGD- (Arg-Gly-Asp-) containing peptides, glycosaminoglycans (GAGs), poly-L-lysine, heparin, protamine, protease inhibitors, and colchicine (Table II). Because inhibited hemocytes recovered their aggregation and adhesion competence after caffeine, EDTA, poly-L-glutamic acid, or cytochalasin B was replaced by CMTBS, the inhibitory effects of such chemicals, unlike NEM, were reversible (data not shown). The extent of hemocyte aggregation can be easily distinguished at three different states, namely no aggregation, weak aggregation, and cohesive aggregation [ILLUSTRATION FOR FIGURE 3 OMITTED]. Five different stages of hemocyte adhesion/spreading were scored [ILLUSTRATION FOR FIGURE 4 OMITTED].
Hemocytes aggregated and adhered in 10 mM sodium azide (Na[N.sub.3]). However, when the cells were co-incubated in 10 mM Na[N.sub.3] with 25 mM caffeine for 30 min, cell aggregation and adhesion were still affected after the caffeine/azide was replaced by azide solution. Such azide-poisoned cells aggregated weakly after the Na[N.sub.3] + caffeine solution was replaced by Na[N.sub.3]. When the Na[N.sub.3] + caffeine was replaced by CMTBS, the cells recovered eventually and formed cohesive aggregates.
Low temperature (4 [degrees] C) delayed cell adhesion for the first few minutes, but after 15 min no differences from controls could be seen [ILLUSTRATION FOR FIGURE 5 OMITTED]. In cell aggregation assays, cells still aggregated weakly at 4 [degrees] C. With an increase in temperature, the aggregates contracted and cohered to form tight masses as if no interruption had occurred [ILLUSTRATION FOR FIGURE 6 OMITTED].
Osmolarity seems not to be critical for M. californianus hemocyte aggregation and adhesion in vitro. There were no significant differences in hemocyte adhesion and aggregation when hemolymph was mixed (1:1) with CMTBS in the range of 800-1100 mOsm.
The inhibitory effects of caffeine, NEM, EDTA, and cytochalasin B on cell aggregation
Hemocytes exposed to 25 mM caffeine, 5 [[micro]gram]/ml cytochalasin B, or 0.1 mM NEM were strongly inhibited from aggregating (95%, 71%, and 72%, respectively, remained free). When hemocytes were treated with EDTA, the morphology of aggregates varied. Under the microscope, some colorless material was seen among the aggregates. Its presence seemed to restrict hemocyte contact. Though caffeine inhibition at 25 mM or 50 mM was reversible and rapid, it did not kill the cells. At a caffeine concentration of 15 mM, cells formed weak aggregates. The difference between the percent free cells in 25 mM and 15 mM caffeine (means of 61.5% and 7.1%) was statistically significant (P [less than] 0.05) [ILLUSTRATION FOR FIGURE 7A OMITTED]. Cytochalasin B (5 [[micro]gram]/ml) in 0.5% DMSO resulted in 70% of the hemocytes remaining free. DMSO (0.5%) did not influence hemocyte aggregation and adhesion [ILLUSTRATION FOR FIGURE 7B OMITTED]. At 0.1 mM, NEM significantly inhibited hemocyte aggregation (P [less than] 0.05) [ILLUSTRATION FOR FIGURE 7C OMITTED]. EDTA at concentrations 15-60 mM significantly inhibited hemocyte aggregation [ILLUSTRATION FOR FIGURE 7D OMITTED]. However, most cells were observed forming weak aggregates; in three experiments, values for percent free cells were 9.5, 18.2, and 21.8 at 60 mM; 8.0, 19.7, and 23.1 at 30 mM; and 8.4, 20.4, and 13.4 at 15 mM.
The inhibitory effects of caffeine, NEM, EDTA, and cytochalasin B on hemocyte adhesion and spreading
When hemolymph was treated with 5 [[micro]gram/ml cytochalasin B or 0.1 mM NEM, the hemocytes remained round, thus resembling naive cells. Cytochalasin B (1 [[micro]gram]/ml) could retard hemocyte spreading, affect the morphology of spread hemocytes, and inhibit hemocyte migration. Cells exposed to 25 mM caffeine were irregular in outline [ILLUSTRATION FOR FIGURE 8A OMITTED]. EDTA in the range 0.6-60 mM inhibited cell adhesion [ILLUSTRATION FOR FIGURE 8B OMITTED].
The concentration of caffeine necessary to inhibit cell adhesion was similar to that needed to inhibit aggregation. At 15 mM caffeine, cell adhesion was significantly greater relative to that in 25 mM caffeine [ILLUSTRATION FOR FIGURE 9A OMITTED]. Cytochalasin B significantly inhibited cell adhesion at concentrations between 1 and 5 [[micro]gram]/ml (P [less than] 0.05), though 60% of the cells were still adherent in 5 [[micro]gram]/ml cytochalasin B treatment [ILLUSTRATION FOR FIGURE 9B OMITTED]. The inhibitory activity of NEM (0.01 mM to 10 mM) was significant [ILLUSTRATION FOR FIGURE 9C OMITTED]. Concentrations in this range inhibited 90% of cell adhesion on culture plates (polystyrene). EDTA ([greater than or equal to]0.6 mM) influenced cell adhesion on culture plates also [ILLUSTRATION FOR FIGURE 9D OMITTED]. When EDTA-treated hemolymph was replaced by 20 mM or a higher concentration of buffered isotonic [Ca.sup.++] or [Mg.sup.++] solution, 80% of hemocytes recovered their adhesion competence [ILLUSTRATION FOR FIGURE 10 OMITTED]. Therefore, Mytilus hemocyte adhesion is [Ca.sup.++] or [Mg.sup.++] - dependent.
A variety of buffers have been used in mussel hemocyte studies (Moore and Lowe, 1977; Renwrantz et al., 1985; Dageforde et al., 1986). Although the buffer osmolarity can affect phagocytic activity in M. edulis (Renwrantz, pers. comm, 1989), pH in the range of 7.4 to 8.4 or osmolarity in the range of 800 to 1050 mOsm did not interfere with hemocyte aggregation in studies reported here (data not shown). To resemble the seawater in which the mussels were held, a pH of 7.4 and salt content of 960 mOsm were selected for the CMTBS.
The fact that naive hemocytes circulating in vivo remain free in suspension is taken to imply an absence of mutual adhesiveness. This could be due to the effects of surface charge. Most cells are negatively charged on their surfaces (often by sialic acids), and the resulting repulsive forces between blood cells have been postulated to separate cells in their naive state (Bell, 1983). Cell activation may reduce the negative surface charge by loss of sialic acids (Rutishauser et al., 1988). To evaluate possible charge effects on cell aggregation and adhesion, heparin and protamine were added to cell suspensions. These reagents failed to influence cell behaviors (data not shown). Heparin is widely used to inhibit platelet clotting, however, it did not affect hemocyte aggregation in this study, in accordance with results in insects (Ratner and Vinson, 1983) and Limulus (Bryan et al., 1964). In this study, hemocytes adhered to glass coated with a variety of compounds, and only 1% agarose was inhibitory. Agarose has been used to inhibit cell adhesion of various cell types; the inhibitory mechanism is unknown.
In both platelet clotting (Guyton, 1986) and the prophenoxidase (proPO) system of anhropods (Lackie, 1988; Johansson and Soderhall, 1989), serine-protease cascade reactions have been implicated. These reactions are controlled by various protease inhibitors (Hergenhahn et al., 1987). Protease inhibitors can affect hemocyte phagocy-tosis also (Fryer et al., 1991). However, protease inhibitors tested here failed to block mussel hemocyte aggregation or adhesion, leading us to conclude that these behaviors are probably not dependent on proteases in the hemolymph.
Both GAGs and RGD-containing peptides are involved in cell-substratum and cell-cell interactions (Underhill, 1982; Roseman, 1985). Furthermore, a sponge aggregation factor has been revealed that is a sulfated polysac-charide (Coombe and Parish, 1988). In our hemocyte aggregation and adhesion tests, none of the GAGs or RGD peptides had significantly inhibitory effects, though they did retard hemocyte spreading.
N-Ethyl-maleimide has been reported to irreversibly inhibit the aggregation of horseshoe crab hemocytes (Bryan et al., 1964). Similarly, NEM interfered with Mytilus hemocyte aggregation and adhesion. Thus, it can be inferred that, in both these species, normal hemocyte behaviors require intact disulfide bond structures in cell proteins.
Inhibition of cytoskeleton assembly has been reported to disrupt insect plasmatocyte encapsulation (Davies and Preston, 1987), and mollusc hemocyte chemotaxis and cell migration (Cheng and Howland, 1982), as well as cell surface receptor redistribution (Cheng and Howland, 1982; Dageforde et al., 1986). Because bivalve hemocytes alter their shapes in both cell aggregation and cell adhesion (Jones and Gillett, 1976; Feng, 1988), it is inferred that the cytoskeleton plays an important role in these two processes. Cytochalasin B and colchicine have been widely used to reduce the rate of actin polymerization into microfilaments and the assembly of microtubules from monomeric tubulins, respectively. In this study, cytochalasin B was more inhibitory towards hemocyte aggregation than towards hemocyte adhesion. It also retarded hemocyte spreading, and restricted hemocyte migration. These data imply that actin microfilaments are involved in shape change during cohesion. Hemocyte spreading and migration are actin-dependent, but hemocyte adhesion may not be associated with actin.
Caffeine has been used as an anticoagulant in slime molds (Brenner and Thoms, 1984) and in several invertebrates (Bertheussen and Seljelid, 1978; Ratner and Vinson, 1983), but its inhibitory mechanism is not clear. The pharmacological effect of caffeine is due to phosphodi-esterase inhibition and the consequent increase in intra-cellular cAMP (Brenner and Thoms, 1984; Snyder, 1984). In M. californianus, inhibition by caffeine is a dose-dependent, reversible reaction that does not affect cell viability. We suspect that the inhibitory effect of caffeine on cell aggregation and adhesion is receptor-associated. Hemocytes treated with caffeine for 30 min were able to recover their aggregation and adhesion competences immediately when caffeine was replaced by CMTBS. And previously spread hemocytes rounded up immediately when they were exposed to caffeine. If the inhibitory targets of caffeine were inside the cells, the cell behaviors would not be affected by caffeine so quickly.
Cell spike formation is involved in both platelet aggregation (Born, 1962) and slime mold aggregation (Garrod and Born, 1971). In this study, some EDTA-treated hemocytes formed spikes, but still remained free. Similar results were reported with the eastern oyster (Fisher, 1986) and limpet hemocytes (Davies and Partridge, 1972). Therefore, spike formation seems not to be sufficient to cause aggregation in these molluscan hemocytes.
[Ca.sup.++] and [Mg.sup.++] are important in many ligand-receptor binding mechanisms, and in maintaining normal functions of various cell adhesion molecules (Muller, 1980; Springer, 1990). In hemocyte aggregation studies, [Ca.sup.++] and [Mg.sup.++] are essential for normal cell behaviors (Davies and Partridge, 1972; Kenney et al., 1972; Jumblatt et al., 1980; Kanungo, 1982; Shozawa and Suto, 1990). For M. californianus hemocytes, 0.6 mM EDTA could completely block hemocyte adhesion. Normally, mussel hemolymph contains 10 mM calcium, 60 mM magnesium, and trace amounts of other divalent cations (Bayne et al., 1976). Apparently, 0.6 mM EDTA chelating capacity did not remove all divalent cations from mussel hemolymph. However, its inhibitory effect was significant. It has been suggested that EDTA may not only serve as chelator, but also influence membrane permeability to divalent cations (Kenney et al., 1972; Kanungo, 1982). In the present studies, EDTA-treated hemocytes aggregated weakly, and the inhibitory effect of EDTA on cohesive aggregation could be overcome by [Ca.sup.++] or [Mg.sup.++] supplements. Therefore, Mytilus hemocyte aggregation may proceed through two sequential stages. Weak aggregation is the early stage, and is [Ca.sup.++]/[Mg.sup.++]-independent. Cohesive aggregation is the late stage, and is [Ca.sub.++]/[Mg.sup.++]-dependent. Similar results were described for hemocyte aggregation in limpets (Davies and Partridge, 1972), and in Limulus (Kenney et al., 1972), as well as for coelomocyte aggregation in holothurians (Fontaine and Lambert, 1977), and in sea stars (Kanungo, 1982). All of these hemocyte or coelomocyte aggregations are two-stage reactions. One is [Ca.sup.++]/[Mg.sup.++] - independent, and the other is [Ca.sup.++]/[Mg.sup.++]-dependent.
Retardation of hemocyte aggregation has been observed at reduced temperature (4[degrees]C) in Mytilus (this study), Limulus (Kenney et al., 1972), and limpets (Davies and Partridge, 1972). It remains to be determined if this is due to altered fluidity of cell membranes, altered kinetics of metabolic reactions, or alternative mechanisms.
In summary, Mytilus hemocyte aggregation is a two-step process that is retarded at low temperature, and requires endogenous metabolic energy and divalent cations. Based on these results and on aggregation studies in several other cell models (Davies and Partridge, 1972; Jumblatt et al., 1980; Kanungo, 1982; Gibralter and Turner, 1985; Pizzey et al. 1988), two-step cell aggregation mediated by divalent cations and temperature appears to have been well conserved during evolution.
A major goal of this study was to characterize distinctive features of self recognition (hemocyte-hemocyte binding) and non-self recognition (hemocyte-matrix binding). Because no aggregation-specific inhibitor was found, these results do not constitute strong evidence that cell aggregation and adhesion are two independent cell activities. However, some observations imply that this is likely. First, higher concentrations of inhibitor are necessary to block hemocyte aggregation than to affect hemocyte adhesion. Second, when reversible inhibitors such as caffeine and EDTA were removed or diluted, hemocyte aggregation occurred earlier than hemocyte adhesion.
Anderson, R. S., and R. A. Good. 1976. Opsonic involvement in phagocytosis by mollusk hemocytes. J. Invertebr. Pathol. 27: 57-64.
Bang, F. B. 1961. Reaction to injury in the oyster (Crassostrea virginica). Biol. Bull. 161: 57-68.
Bayne, B. L., J. Widdow, and R. J. Thompson. 1976. Physiology: II. Pp. 207-260 in Marine Mussels Their Ecology and Physiology, B. L. Bayne ed. Cambridge Univ. Press, Cambridge.
Bell, G. I. 1983. Cell-cell adhesion in the immune system. Immunol. Today 4: 237-240.
Born, G. V. R. 1962. Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature 194: 927-929.
Bertheussen, K., and R. Seljelid. 1978. Echinoid phagocytes in vitro. Exp. Cell Res. 111: 401-412.
Brenner, M., and S. D. Thoms. 1984. Caffeine blocks activation of cyclic AMP synthesis in Dictyostelium discoideum. Dev. Biol. 101: 136-146.
Bryan, F. T., C. W. Robinson, G. F. Gilbert, and R. D. Langdell. 1964. N-ethylmaleimide inhibition of horseshoe crab hemocyte aggregation. Science 144: 1147-1148.
Cheng, T. C. 1981. Bivalves. Pp. 233-300 in Invertebrate Blood Cells, N. A. Ratcliffe and A. F. Bowley, eds. Academic Press, New York.
Cheng, T. C., and K. H. Howland. 1982. Effects of colchicine and cytochalasin B on chemotaxis of oyster (Crassostrea virginica) hemocytes. J. Invertebr. Pathol. 40: 150-152.
Coombe, D. R., and C. R. Parish. 1988. Sulfated polysaccharide-mediated sponge cell aggregation: the clue to invertebrate self/nonself-recognition? Pp. 31-54 in Invertebrate Historecognition, R. K. Grosberg, D. Hedgecock, and K. Nelson, eds. Plenum Publishing Co., New York.
Dageforde, S., A. Schmucker, and L. Renwrantz. 1986. Capping of cell surface receptors on blood cells from the molluscs Helix pomatia (Gastropoda) and Mytilus edulis (Lamellibranchiata). Eur. J. Cell Biol. 41: 113-120.
Davies, D. H. and T. M. Preston. 1987. Effect of disruption of plasmatocyte microfilaments on encapsulation in vitro. Dev. Comp. Immunol. 11: 353-362.
Davies, P. S., and T. Partridge. 1972. Limpet haemocytes I. Studies on aggregation and spike formation. J. Cell Sci. 11: 757-769.
Dundee, D. S. 1953. Formed elements of the blood of certain fresh-water mussels. Trans. Am. Microsc. Soc. 72: 254-264.
Feng, S. Y. 1988. Cellular defence mechanisms of oysters and mussels. Pp. 153-168 in Disease Processes in Marine Bivalve Molluscs, W. S. Fisher, ed. Am. Fish. Soc. Special Publication 18. Bethesda, Maryland.
Feng, S. Y., and J. S. Feng. 1974. The effect of temperature on cellular reactions of Crassostrea virginica to the injection of avian erythrocytes. J. Invertebr. Pathol. 23: 22-37.
Fisher, W. S. 1986. Structure and functions of oyster hemocytes. Pp. 25-35 in Immunity in Invertebrates, Chap. 3, M. Brehelin, ed. Springer-Verlag, Berlin, West Germany.
Fontaine, A. R, and P. Lambert. 1977. The fine structure of the leu-cocytes of the holothurian, Cucumaria miniata. Can. J. Zool. 55: 1530-1544.
Fryer, S. E., R. C. Bender, and C. J. Bayne. 1991. Proteinase inhibitory activity in the plasma of the gastropod mollusc, Biomphalaria glabrata. Dev. Comp. Immunol. 15(supplement I): S45.
Garrod, D. R, and G. V. R. Born. 1971. Effect of temperature on the mutual adhesion of pre-aggregation cells of the slime mold, Dictyostelium discoideum. J. Cell Sci. 8: 751-765.
Gibralter, D., and D.C. Turner. 1985. Dual adhesion systems of chick myoblasts. Dev. Biol. 112: 292-307.
Guyton, A. C. 1986. Hemostasis and blood coagulation. Pp. 76-86 in Textbook of Medical Physiology, chap. 8, A. C. Guyton, ed. W. B. Saunders Co., Philadelphia.
Hergenhahn, H.-G., A. Aspan, and K. Soderhall. 1987. Purification and characterization of a high-[M.sub.r] proteinase inhibitor of prophenol oxidase activation from crayfish plasma. Biochem. J. 248: 223-228.
Johansson, M. W., and K. Soderhall. 1989. Cellular immunity in crustaceans and the proPO system. Parasitol. Today 5: 171-176.
Jones, G. E., and R. Gillett. 1976. Rapid modification of the morphology of cell contact sites during the aggregation of limpet haemocytes. J. Cell Sci. 22: 21-33.
Jumblatt, J. E., V. Schlup, and M. M. Burger. 1980. Cell-cell recognition: specific binding of Microciona sponge aggregation factor to homotypic cells and the role of calcium ions. Biochemistry 19: 1038-1042.
Kanungo, K. 1982. In vitro studies on the effects of cell-free coelomic fluid, calcium, and/or magnesium on clumping of coelomocytes of the sea star, Asterias forbesi (Echinodermata: Asteroidea). Biol. Bull. 163: 438-452.
Kenney, D. M., F. A. Belamarich, and D. Shepro. 1972. Aggregation of horseshoe crab (Limulus polyphemus) amoebocytes and reversible inhibition of aggregation by EDTA. Biol. Bull. 143: 548-567.
Lackie, A. M. 1988. Immune mechanisms in insects. Parasitol. Today 4: 98-105.
Moore, M. N., and D. M. Lowe. 1977. The cytology and cytochemistry of the hemocytes of Mytilus edulis and their responses to experimentally injected carbon particles. J. Invertebrt. Pathol. 29: 18-30.
Muller, W. E. G. 1980. Cell membranes in sponges. Int. Rev. Cytol. 77: 129-181.
Narain, A. S. 1973. The amoebocytes of lamellibranch molluscs, with special reference to the circulating amoebocytes. Malacol. Rev. 6: 1-12.
Pizzey, J. A., G. E. Jones, and F. S. Walsh. 1988. Requirements for the [Ca.sup.2+]-independent component in the initial intercellular adhesion of C2 myoblasts. J. Cell Biol. 107: 2307-2317.
Rather, S. and S. B. Vinson. 1983. Encapsulation reactions in vitro by haemocytes of Heliothis virescens. J. Insect Physiol. 29: 855-863.
Renwrantz, L., J. Daniels, and P.-D. Hansen. 1985. Lectin-binding to hemocytes of Mytilus edulis. Dev. Comp. Immunol. 9: 203-210.
Roseman, S. 1985. Studies on specific intercellular adhesion. J. Blochem. 97: 709-718.
Rutishauser, U., A. Acheson, A. K. Hall, D. M. Mann, and J. Sunshine. 1988. The neural cell adhesion molecule (NCAM) as a regulator of cell-cell interaction. Science 40: 53-57.
Seifert, R. A. 1983. On the mechanism of oyster amoebocyte aggregation. Ph.D. thesis. University of Washington.
Shozawa, A., and C. Suto. 1990. Hemocytes of Pomacea canaliculata: I. Reversible aggregation induced by [Ca.sup.2+]. Dev. Comp. Immunol. 14: 175-184.
Sminia, T. 1981. Gastropods. Pp. 191-232 in Invertebrate Blood Cells, N. A. Ratcliffe and A. F. Rowley, eds. Academic Press, New York.
Snyder, S. H. 1984. Adenosine as a mediator of the behavioral effects of xanthines. Pp. 129-141 in Caffeine: Perspectives From Recent Research, P. B. Dews, ed. Spring-Verlag.
Springer, T. A. 1990. Adhesion receptors of the immune system. Nature 346: 425-434.
Sparks, A. K. 1972. Reaction to injury and wound repair in invertebrates. Pp. 22-133 in Invertebrate Pathology. Noncommunicable Diseases, Academic Press, New York.
Underhill, C. B. 1982. Interaction of hyaluronate with the surface of simian virus 40-transformed 3T3 cells: Aggregation and binding studies. J. Cell Sci. 56: 177-189.
JIUN-HONG CHEN, Department of Zoology, National Taiwan University, Taipei, Taiwan, R.O.C.
CHRISTOPHER J. BAYNE, Department of Zoology, Oregon State University, Oregon 97331-2914
|Printer friendly Cite/link Email Feedback|
|Author:||Chen, Jiun-Hong; Bayne, Christopher J.|
|Publication:||The Biological Bulletin|
|Date:||Jun 1, 1995|
|Previous Article:||Cytoskeletal architecture and organelle transport in giant syncytia formed by fusion of hexactinellid sponge tissues.|
|Next Article:||Cytological analysis of the urn cell complex of Sipunculus nudus before and after serum-induced secretion.|