Algal cells grown in the green hydra Chlorohydra viridissima were shown to possess characteristic antigenic determinants not found in algae cultured in vitro. These antigenic determinants, including those localized on the algal cell surface, were shown to be responsible for the phagocytic recognition of potential algal symbionts by digestive cells of Chlorohydra. The results of this study indicate the existence of two systems governing phagocytosis in Chlorohydra, one specific for algal cells grown in hydra, another governing the uptake of other particles by the hydra digestive cells.

Digestive cells of Chlorohydra viridissima readily phagocytose cells of the green alga Chlortlla (Muscatine, Cook, Pardy & Pool, 1975). Phagocytosis of algae occurs at the distal ends of the animal cells, and the digestive cells respond in one of two ways to these ingested algae. With 2 exceptions algae belonging to a variety of strains of Chlorella are accumulated within one or more large vacuoles which remain at the distal end of the animal cell. Within 24 h after ingestion these algae disappear from the digestive cell. In contrast, the fate of native hydra algal symbionts and of symbiotic algae from Paramecium bursaria (syngen 1, NC64A strain) is very different from that of other strains of Chlorella. Following ingestion hydra algae and NC64A. remain in small vacuoles (1 algal cell per vacuole) in the digestive cell. These algal-containing vacuoles migrate to the proximal end of the digestive cell, where the algal cells reside in a stable symbiotic association with the hydra host. These responses to ingested algae constitute the algal recognition system of green hydra cells (Muscatine et al. 1975).

The features of native hydra algae and NC64A algae that elicit this specific recognition response by hydra cells are unknown. Neither the production of oxygen nor the selective release of maltose or glucose by symbiotic algae is involved in the recognition process (Muscatine et al. 1975; Pardy & Muscatine, 1973).

There is much evidence that intercellular recognition involves the interaction of components of the surfaces of apposed cells (Burger, Turner, Kuhns & Weinbaum, 1975; Edelman, 1976; Roth, 1973). Cell surface components are known to be involved in gamete recognition in animals (Martinez-Palomo, 1970), yeast (Crandall & Brock, 1968) and bacteria (Sneath & Lederberg, 1961), and in morphogenesis in multicellular organisms (Edelman, 1976). The interaction of cell surface components in symbiotic associations has also been reported, including the role of bacterial surface antigens in the response of mammalian cells to invading bacteria (Medearis, Camitta & Heath, 1968; Davis et al. 1974).

In the present study the role of antigenic features of symbiotic algae, including those of the cell surface, in the recognition of these algae by hydra cells was examined. Antisera specific for native hydra algae were produced, and using these antisera it is demonstrated that algae grown in Chlorohydra possess antigenic determinants unlike those of algae cultured in vitro, and that these determinants are involved in triggering the phagocytic recognition of algae by digestive cells of Chlorohydra.

Chlorohydra viridissima (Florida strain) (Muscatine, 1974) were cultured in ‘M’ solution and fed daily with Artemia (Lenhoff & Brown, 1970). Aposymbiotic C. viridissima were raised from algae-free embryos. The hydra were maintained at 18°C with an irradiance of 2 · 1 × 10−5 einsteins m−2 s−1 delivered by fluorescent tubes (Cool-White, Sylvania) and with a photoperiod of 12 h light/12 h dark.

Algae native to C. viridissima (Florida) were isolated from mass culture of hydra just prior to experiment use (Muscatine, 1965). These algae are termed ‘F/F algae’ in this study. A variety of other strains of Chlorella were also used (Table 1). The algae were cultured in modified Loefer’s medium (Karakashian, 1963) in conditions of temperature and irradiance similar to those used for the hydra. The NC64A strain alga was cultured in either modified Loefer’s medium or in C. viridissima-, the former algae are termed NC/L algae, the latter NC/F algae.

Table 1.

Chlorella strains used in this study

Chlorella strains used in this study
Chlorella strains used in this study

Antisera against F/F and NC/L algae were raised in New Zealand albino rabbits; 0·2 ml of a 20 % suspension (in 0·85 % NaCl) of intact cells was injected intravenously every third day for a total of 24 days. Five days after the final injection the rabbits were bled, injected intravenously with another 0·2 ml of a 20% suspension and bled 5 days later. Sera from either of these bleedings were used in experiments. Antisera were heat-treated (60°C for 20 min) to destroy rabbit complement.

Immunological comparisons of algal cells using these antisera were performed using quantitative microcomplement fixation (Levine, 1973). Guinea-pig complement and sheep erythrocytes were obtained from Mission Laboratories, Rosemead, California. In all cases whole algal cells were used as antigens in the complement fixation assays.

Agglutination of algae by anti-algal serum was measured as described by Nowotny (1969). Serial 2-fold, dilutions of the various sera were used in the assays. Ten microlitres of each dilution were placed in the wells of concavity plates and 5 μl of algal suspensions (7·5 × 107 cells/ml) added to each. Preimmune control sera and anti-algal sera were run in parallel for each assay. The concavity plates were gently rocked for 3–5 min at room temperature, and the results scored as either agglutinated (all algae tightly clumped at the centre of the well) or as nonagglutinated (algae scattered over the bottom of the well).

Algae were also treated with antisera prior to injection into Chlorohydra. Antiserum was diluted to the desired concentration using complement fixation buffer (0·14 M NaCl, 0·15 mM CaCl2, 0·5 mM MgSO4, 0·1 % bovine serum albumin, 0 ·01 M Tris buffer, pH 7 · 4). Algae (5 × 106 cells) were suspended in 10 ml of diluted antiserum and incubated at 37 °C for 1 h with constant shaking. Control groups, consisting of 5 × 10 6 cells suspended in 10 ml of complement fixation (CF) buffer, were always run in parallel for each experiment.The treated and control cells were then washed with CF buffer and then with M solution and injected into Chlorohydra.

Algal pellets (5 × 10 6 cells) were suspended in 50 μl of M solution and injected into the enteron of aposymbiotic C. viridissima as previously described (Pardy & Muscatine, 1973).

The number of algae per host digestive cell was determined by maceration of the hydra (David, 1973). The gastric and budding regions of the body column of each hydra were excised, macerated on gelatin-coated microscope slides and stained with 0 · 25 % toluidine blue. The digestive cells of hydra have a characteristic morphology in macerated preparations (David, 1973) and the algae are easily distinguished from host intracellular organelles. The macerated preparation of each of 5 hydra was examined, and the average number of algae/digestive cell in 50 randomly selected, infected digestive cells was determined for each preparation; these 5 values were then averaged. The percentage of host digestive cells containing algae was determined for each macerated preparation by scoring 200 randomly selected digestive cells with respect to the presence or absence of algal cells.

The uptake of algae by aposymbiotic hydra was also measured by a fluorometric technique developed by C. F. D’Elia (personal communication). Individual hydra were slit open and gently washed with M solution to remove any uningested algal cells. The hydra tissue was transferred to a test tube and 2–4 ml of 90 % CH30H (made alkaline with Na2CO3) was added. The air in each tube was displaced with N2, the tubes sealed with parafilm and incubated in the dark at 4 °C for 24 h. Chlorophyll fluorescence of each sample was determined using a G. K. Turner Associates fluorometer (model in) fitted with a 550-nm excitation filter and a Wratten 25 barrier filter. The fluorescence (expressed in arbitrary units) of each of 15 animals was determined and the average fluorescence calculated. No attempt was made to convert arbitrary fluorescence units into numbers of algae, although the fluorescence of the sample increases linearly with the number of algae present in the sample (D’Elia, personal communication).

Specificity of anti-F/F serum for hydra algae

To determine the specificity of the rabbit antiserum produced against F/F algae, the following experiment was performed. Various strains of Chlorella cultured in Loefer’s medium were reacted with antiserum prepared against F/F algae, and their reactivity was measured by microcomplement fixation. Only algae isolated from hydra were reactive with the anti-F/F serum; therefore, as compared to the other algae, the anti-F/F serum is highly specific for the symbionts of C. viridissima.

Influence of hydra on the antigenic composition of the algae

The algal symbionts of hydra have never been cultured in vitro (Muscatine, 1974); therefore it is not possible to determine if the antigenic determinants characteristic of native hydra algae are always expressed, regardless of where these algal cells are grown, i.e. in situ or in vitro. However, aposymbiotic C. viridissima can be infected with NC64A algae (Pardy & Muscatine, 1973), thus allowing a comparison of the antigenicity of NC64A algae grown in vitro (NC/L algae) with NC64A algae grown in Florida strain green hydra (NC/F algae). Such a heterologous association was established. Because of the very low initial infections (see below) and because of the initially very slow growth rates of the NC64A algae in hydra, the resultant NC/F algae were not examined using complement fixation until 6 months after the initial infection. Fig. 1 shows the result of a typical assay performed using anti-F/F serum and F/F, NC/F, NC/L algae and NC/L algae contaminated with aposymbiotic C. viridissima tissue. The latter group was included since the anti-F/F serum was prepared using native hydra algae contaminated with some host cellular debris, and reaction of these host-contaminants with anti-hydra antibodies in the antiserum would contribute to the observed complement fixations. The contaminated NC/L algae were prepared by homogenizing 0 · 75 ml of wet-packed aposymbiotic C. viridissima in the presence of 0·1 ml of wet-packed NC/L algae. This volume (0 · 75 ml) of green hydra routinely yields 0·1 ml of wet-packed hydra algae. These NC/L algae were then isolated from the hydra suspension using the method for isolation of hydra algae. As shown in Fig. 1 NC/L algae (uncontaminated by hydra cellular debris) do not react with this antiserum. Hydra-contaminated NC/L algae are reactive with anti-F/F serum; however the percentage of complement fixed due to hydra contamination is significantly less than the percentages of complement fixed by either F/F or NC/F algae. As compared to the F/F algae the peak fixation for the NC/F algae is displaced both vertically and laterally. Lateral shifts in peak fixation reflect differences in antigen quality, vertical shifts differences in antigen quantity (Levine, 1973). Thus NC64A algae, grown in Chlorohydra, possess antigenic determinants similar, but not identical, to those characteristic of native hydra algae.

Fig. 1.

Microcomplement fixation using anti-F/F serum and: F/F algae, △; NC/F algae, ○; NC/L algae, □; and NC/L algae contaminated with hydra tissue, ◼. Antiserum dilution = 1/15 000.

Fig. 1.

Microcomplement fixation using anti-F/F serum and: F/F algae, △; NC/F algae, ○; NC/L algae, □; and NC/L algae contaminated with hydra tissue, ◼. Antiserum dilution = 1/15 000.

To determine whether the NC/F algae have lost any of the antigenic determinants found in NC/L algae, NC/F algae were reacted with anti-NC/L serum (Fig. 2). Reaction of NC/L algae with this antiserum did not yield a typical complement fixation curve, i.e. there was no reduction in fixation when the antigens (algae) were in excess as compared to the antiserum concentration. This atypical pattern is not uncommon when whole cells are used as antigens (for example see Pfeiffer et al. 1971). The data for reaction of anti-NC/L serum with NC/F algae are quite typical and demonstrate that the NC/F algae are reactive with this antiserum. The vertical displacement of the NC/F curve relative to the NC/L curve indicates that NC/F algae have far fewer of those antigenic determinants characteristic of NC/L algae. F/F algae are not measurably reactive with this antiserum.

Fig. 2.

Microcomplement fixation using anti-NC/L serum and: NC/L algae, □; NC/F algae, ○; and F/F algae, △-Antiserum dilution = 1/6000.

Fig. 2.

Microcomplement fixation using anti-NC/L serum and: NC/L algae, □; NC/F algae, ○; and F/F algae, △-Antiserum dilution = 1/6000.

Reactivity of anti-algal sera with the algal cell surface

Even though intact algal cells were used to produce the antisera and as test antigens in the complement fixation assays, it is possible that the antisera produced are not reactive with the algal cell surface, rather that they react with intracellular antigenic determinants. In order to determine if the antisera produced against F/F algae or against NC/L algae contain antibodies that react with the cell surfaces of F/F algae and NC/L algae respectively, the ability of these antisera to agglutinate these algae was determined. Anti-F/F serum agglutinated F/F algae at a dilution of 1/64, but control serum at dilutions of as low as 1/2 did not. NC/L algae were agglutinated by anti-NC/L serum at a dilution of 1 /128, while control serum did not agglutinate these cells even at a dilution of 1 /2. Because agglutination is due to the interaction of antibodies with the surface of antigens (Nowotny, 1969), the above data indicate that both anti-F/F and anti-NC/L sera contain antibodies against the surfaces of F/F algae and of NC/L algae respectively.

Effect of prior growth in hydra on uptake of algae by hydra cells

Data presented above (Fig. 1) indicated that NC64A algae, when grown in C. viridissima, possess antigenic determinants similar to those of algae native to C. viridissima. In order to determine if residence in green hydra also affects recognition of NC/F algae by digestive cells of Chlorohydra the following experiment was performed. Control aposymbiotic Chlorohydra were injected with the Florida strain symbionts. Experimental aposymbiotic hydra were injected with either NC/L or NC/F algae. Twenty-four hours after injection all hydra were macerated and numbers of algae per infected digestive cell and percentages of host digestive cells harbouring algae were determined (Table 2). The response of the digestive cells to F/F and NC/L algae was very different. Approximately twice as many F/F algae were taken up per infected digestive cell as compared to the NC/L algae. Further, approximately 10 times as many digestive cells contained F/F algae as compared to digestive cells in those hydra injected with NC/L algae. The uptake of NC/F algae was similar to the uptake of the F/F algae. These results are interpreted to mean that infection and residence of NC64A algae in Chlorohydra has resulted in a change in the ability of NC64A algae to be recognized and sequestered by the digestive cells of the hydra.

Table 2.

Uptake of algae by digestive cells of Chlorohydra

Uptake of algae by digestive cells of Chlorohydra
Uptake of algae by digestive cells of Chlorohydra

To determine if the inclusion of host cellular material in the injection mixture influenced algal uptake and caused the observed differences, the following experiment was performed. Aposymbiotic hydra were homogenized in the presence of NC/L algae. Care was taken to use a quantity of aposymbionts that, were they to harbour algae, would yield upon homogenization as many NC64A algae as were added to the aposymbiotic hydra. The algae were then isolated as described previously and injected into aposymbiotic Chlorohydra. A second batch of NC/L algae was suspended in a symbiont-free hydra homogenate obtained from C. viridissima that harbour NC/F algae (removed from the homogenate by centrifugation prior to the addition of the NC/L algae). This suspension was thoroughly mixed at room temperature for 15 min and the algae were then isolated. These latter NC/L algae, contaminated with host cellular material from hydra that harboured algae, were then injected into aposymbiotic hydra. Twenty-four hours later the uptake of NC/L algae contaminated with homogenized hydra tissue was compared to the uptake of untreated NC/L algae and to the uptake of NC/F algae (Table 3). The uptake characteristics for the NC/F algae and for untreated NC/L algae, treated with either aposymbiotic hydra homogenate or NC/F-hydra homogenate, exhibited no enhancement of uptake. These data demonstrate that constituents of hydra tissue, present as contaminants of the injected algae, do not promote the uptake and sequestration of algae by digestive cells of Chlorohydra.

Table 3.

Uptake by digestive cells of Chlorohydra of algae contaminated with hydra cellular debris

Uptake by digestive cells of Chlorohydra of algae contaminated with hydra cellular debris
Uptake by digestive cells of Chlorohydra of algae contaminated with hydra cellular debris

Effect of anti-F/F serum on algal uptake by hydra

To determine if the similar antigenic determinants of F/F and NC/F algae are involved in the enhanced uptake and sequestration by digestive cells that are characteristic of F/F and NC/F algae, the following experiment was performed. Freshly isolated F/F algae and NC/F algae were treated with anti-F/F serum and the uptake of these treated algae was compared with the uptake of untreated F/F and NC/F algae. The extent of uptake and sequestration was assayed by measuring the fluorescence of the injected hydra 24 h after injection (Table 4). These data show that anti-F/F serum is effective in reducing uptake and sequestration of either F/F or NC/F algae by digestive cells. In the case of the F/F algae, uptake was reduced to 35% of control levels; NC/F uptake was reduced to 41 % of control levels.

Table 4.

Uptake by Chlorohydra of F/F and NC/F algae treated with anti-F/F serum prior to injection

Uptake by Chlorohydra of F/F and NC/F algae treated with anti-F/F serum prior to injection
Uptake by Chlorohydra of F/F and NC/F algae treated with anti-F/F serum prior to injection

The specificity of the anti-F/F serum in the reduction of uptake and sequestration of F/F and NC/F algae was demonstrated by examining the influence of anti-NC/L serum on the uptake of F/F, NC/F and NC/L algae. This antiserum has no affinity for F/F algae, but it is reactive with NC/F and NC/L algae (Fig. 2). Treatment of F/F, NC/L and NC/F algae with this antiserum has no inhibitory influence on the uptake and sequestration of these algae by hydra digestive cells (Table 5). Thus the reduction of uptake and sequestration of anti-F/F serum-treated F/F and NC/F algae was due to a specific action of this antiserum. It is important to note that fluorometric data for uptake of one algae strain (e.g. F/F) cannot be directly compared with another (e.g. NC/F). The fluorescence developed is due to the amount of chlorophyll present, and F/F and NC/F algae differ in the chlorophyll content per cell (Pool, unpublished data).

Table 5.

Uptake by Chlorohydra of NC/L, F/F and NC/F algae treated with anti-NC/L serum prior to injection

Uptake by Chlorohydra of NC/L, F/F and NC/F algae treated with anti-NC/L serum prior to injection
Uptake by Chlorohydra of NC/L, F/F and NC/F algae treated with anti-NC/L serum prior to injection

Anti-F/F antiserum was prepared using F/F algae containing some hydra cellular debris. The inhibitory action of anti-F/F serum could be due to the action of antihydra antibodies on the cells of the host hydra. The inhibition of uptake then would be independent of the algal variety present in the enteron of the hydra and would be dependent on the presence of these anti-hydra antibodies in the suspensions of algae used for injections. This hypothesis was tested in 2 ways. First, NC/L algae were treated with anti-F/F serum and the uptake of the treated NC/L algae compared to the uptake of untreated NC/L algae. The results of this experiment (Table 6) show that such treatment did not influence the uptake of NC/L algae.

Table 6.

Uptake by Chlorohydra of NC/L algae treated with anti-F/F serum prior to injection

Uptake by Chlorohydra of NC/L algae treated with anti-F/F serum prior to injection
Uptake by Chlorohydra of NC/L algae treated with anti-F/F serum prior to injection

The second experimental approach was to wash the antiserum-treated algal pellet repeatedly to remove any anti-hydra antibodies present with the algae. In this experiment the uptake of 2 groups of anti-F/F serum-treated F/F algae was compared to that of untreated F/F algae. After incubation with antiserum one group was washed once with M solution. The second group was washed 5 times in CF buffer and once in M solution. The uptake and sequestration of both groups by aposymbiotic Chlorohydra was measured fluorometrically 24 b after injection (Table 7). Even with repeated washing the uptake of anti-F/F serum-treated F/F algae was not changed from that of algae washed a single time. These results demonstrate that the reduction of uptake and sequestration of F/F and NC/F algae by this antiserum is not due to the effect of anti-hydra antibodies, specific for the hydra cells, that might have been carried along in the injection mixtures.

Table 7.

Uptake by anti-F/F serum Chlorohydra of F/F algae treated with and then washed prior to injection

Uptake by anti-F/F serum Chlorohydra of F/F algae treated with and then washed prior to injection
Uptake by anti-F/F serum Chlorohydra of F/F algae treated with and then washed prior to injection

The process of uptake and sequestration of algae by hydra digestive cells involves first the phagocytosis of the algae and then the transport of these algae to the proximal region of the digestive cell (Muscatine et al. 1975). The process of phagocytosis occurs in the first 2 h after injection of algae and their transport occurs during the next 24 h.

Algal cells that are not transported to the proximal region of the digestive cells disappear from the digestive cells within 24 h after injection (Muscatine et al. 1975). The time difference between uptake and sequestration makes it possible to determine if anti-F/F serum inhibits the phagocytosis of treated algae or if it inhibits the transport of ingested algae. F/F algae were treated with anti-F/F serum and injected into aposymbiotic Chlorohydra. The presence of algae was measured by fluorometry 3 h and 24 h after injection, and compared to animals injected with untreated F/F algae (Table 8). The reduction in uptake due to anti-F/F serum treatment is complete by the third hour after injection, and the extent of infection observed at 3 h post-injection is the same as that observed at 24 h post-injection. Thus the treatment with anti-F/F serum has its effect during the period of phagocytosis of algae from the enteron, and not during the period when algal uptake has ceased and algae either disappear from the digestive cells or are sequestered in the proximal region of the digestive cells.

Table 8.

Uptake by Chlorohydra of anti-F/F treated F/F algae measured 3 and 24 h after injection

Uptake by Chlorohydra of anti-F/F treated F/F algae measured 3 and 24 h after injection
Uptake by Chlorohydra of anti-F/F treated F/F algae measured 3 and 24 h after injection

The results of this study demonstrate that algae native to Chlorohydra are characterized by antigenic determinants not found in algae cultured in vitro. It is further demonstrated that the expression of these antigenic determinants, at least in the case of the NC64A strain of Chlorella, is a function of the residence of the algae in Chlorohydra. Thus NC64A algae grown in Chlorohydra exhibit a quantitative reduction in antigenic determinants characteristic of cells of the same strain grown in vitro and concomitantly possess antigenic determinants similar to those found in native Chlorohydra algae.

The changes in antigenicity of the NC64A algae accompanying residence in hydra might originate through one, or a combination, of 3 mechanisms. The first mechanism is based on the assumption that in vitro cultures of NC64A algae are heterogenous populations with respect to antigenicity. The hydra digestive cells would selectively phagocytose or sequester those algal cells with a specific antigenic composition. This hypothesis assumes that the algae are passive, the observed antigenic changes being produced by the selective response of the host digestive cells. Pardy & Muscatine (1973) and the present study present evidence that different algae are sequestered at different rates and in differing numbers by C. viridissima.

A second mechanism is by either induction of antigenic changes in response to the host cellular environment, or by donation by the host cell of specific antigenic determinants to the newly ingested algal cells. In the former case the algae sequestered by hydra digestive cells would be antigenically plastic, the antigenicity of the algal cells changing in response to changed environmental conditions. Chlorella is known to undergo extensive morphological changes when grown in different environmental conditions. Karakashian (1970) finds a high degree of morphological plasticity in Chlorella symbiotic with Paramecium bursaria, including the NC64A strain, particularly in the ultrastructure of the cell wall, chloroplast and intracellular inclusions. Pardy (1976) reports on similar morphological changes in symbiotic algae grown in different strains of Chlorohydra. The changes in antigenicity observed in this study might thus be due to an antigenic plasticity comparable to the morphological plasticity of Chlorella. Evidence supporting antigenic transfer from host to symbiont in a variety of associations has recently been reviewed (DeVay & Adler, 1976).

A third possible mechanism is through mutation of the algae in hydra and the increased ‘fitness’ of specific mutant forms for replication and survival within the host digestive cells. The algae would be taken up by digestive cells regardless of their antigenic composition. With time mutant algal cells would appear in the algal population, the mutations expressed as altered surface antigenicity. Those mutant algae with the characteristic surface antigenicity would eventually become the dominant forms in the algal population. Watkins (1964) has proposed such a mechanism to explain the changes in antigenicity characteristic of parasitic populations of Trypanosoma. Experiments are currently under way to elucidate the mechanism of antigenic modification of algae in response to residence in Chlorohydra.

This study has also demonstrated a role for the antigenic determinants found in algae grown in hydra, namely in triggering the phagocytic recognition of algae by the hydra digestive cell. Thus treatment of native hydra algae and of NC64A algae grown in Chlorohydra with anti-hydra algae serum results in a reduction in uptake and sequestration of the treated algae. This reduction of uptake is not due to antibodies against hydra cell components in the antisera, nor to some nonspecific coating of the algal cells by antibodies (the anti-NC/L serum, while reacting with NC/F algae, did not reduce uptake and sequestration of NC/F algae). The conclusion most consistent with these results is that anti-F/F antibodies react with and mask, or otherwise make inoperative, some feature(s) of the F/F and NC/F algae that is (are) involved in recognition of these algae by the host cells. The precise loci of action of these specific antibodies are not yet known, although evidence from the present study indicates that the anti-F/F antibodies react with the algal cell surface.

The action of a symbiont-specific antibody interfering in the interaction of the host cell with prospective intracellular symbionts is not novel to C. viridissima. Toxoplasma gondii, an intracellular coccidian protozoan symbiotic with vertebrate macrophages, is able to inhibit host lysosomal fusion with the peri-Toxoplasma vacuoles and is able to resist digestion within the macrophage. Toxoplasma treated with heat-inactivated anti-Toxoplasma antiserum is unable to inhibit lysosomal fusion and is subsequently digested. Similar findings on this response of macrophages to symbionts treated with heat-inactivated antisera have been reported for Rickettsiae (Gambrill & Wisseman, 1973), the bacterium Chlamydiapsittaci (Friis, 1972) and for vaccinia virus (Silverstein, 1970).

The action of anti-symbiont antisera in all of the above examples involves inhibition of symbiont mechanisms which interfere with host intracellular responses to infection. Evidence presented here indicates that the action of anti-hydra algae sera is during the process of ingestion of the algae rather than after the algal cells are within hydra digestive cells. These data in the present study lead me to propose that digestive cells in Chlorohydra possess at least 2 recognition systems for phagocytosis. The first is a specific recognition system that is cued by antigenic determinants of algal cells and results in an enhanced rate of entry for such algae, e.g. native hydra algae. The second system is a nonspecific one. This system governs the uptake of other particulate material, e.g. algae cultured in vitro. Anti-F/F serum is postulated to contain antibodies that interfere with the algal features triggering the first, specific system. Under these conditions the anti-F/F serum-treated algae would enter the host cell through interaction with the nonspecific system.

One immediate problem with this hypothesis is that while anti-F/F serum-treated cells are taken up to a significantly lesser extent than are untreated cells, their uptake is never as low as that observed for NC64A algae cultured in vitro. This rather vexing difference could result from the use of whole hydra to study the uptake of symbionts. Hydra is known to secrete proteolytic enzymes into the enteron (Lenhoff, 1968). These enzymes could act to digest away antibody molecules from algae within the enteron, exposing the features that trigger the symbiont-specific uptake system.

Mammalian macrophages are known to possess receptors that govern the phagocytosis of specific particles (Silverstein, Steinman & Cohn, 1977). These receptors are distinct from those involved in the indiscriminant phagocytosis of other sorts of particles (Silverstein et al. The specific phagocytosis receptors of macrophages are not completely analogous to those proposed for Chlorohydra in that the former are triggered by specific host antibodies that attach to the surface of the particle. Holland et al. (Holland, Holland & Cohn, 1972) have shown that erythrocytes coated with anti-erythrocyte antibodies are recognized by a system different from the system that recognizes other, non-antibody-coated particles. When the former recognition system is inhibited, resulting in a great reduction in endocytosis of antibody-treated red cells, the uptake of other particles (e.g., latex spheres) by the macrophage is unperturbed.

As proposed in the present study, Chlorohydra possess a separate recognition system for specific particles, i.e., prospective symbionts, but instead of this system being triggered by host-produced factors, the hydra system is triggered by features inherent to the prospective algal symbionts.

I am grateful to Dr L. Muscatine for his advice, encouragement and support. I also wish to thank Dr H. Herschman for advice on serology and Ms K. Bolles for technical assistance. This research was supported, in part, by National Science Foundation grant PCM75-03380 to Dr L. Muscatine.

Burger
,
M. M.
,
Turner
,
R. S.
,
Kuhns
,
W. J.
&
Weinbaum
,
G.
(
1975
).
A possible model for cell-cell recognition via surface macromolecules
.
Phil. Trans. R. Soc. Ser. B
71
,
379
393
.
Crandall
,
M. A.
&
Brock
,
T. D.
(
1968
).
Molecular aspects of specific cell contact
.
Science, N. Y
.
161
,
473
475
.
David
,
C. N.
(
1973
).
A quantitative method for maceration of hydra tissue
.
Wilhelm Roux Arch. EnttvMech. Org
.
171
,
268
295
.
Davis
,
B. D.
,
Dulbecco
,
R.
,
Eisen
,
H. N.
,
Ginsberg
,
H. S.
,
Wood
,
W. B.
&
Mccarty
,
M.
(
1974
).
Microbiology
, 2nd edn, p.
633
.
Hagerstown
:
Harper & Row
.
Devay
,
J. E.
&
Adler
,
H. E.
(
1976
).
Antigens common to host and parasites
.
A. Rev. Microbiol
.
30
,
147
168
.
Edelman
,
G. M.
(
1976
).
Surface modulation in cell recognition and cell growth
.
Science, N. Y
.
192
,
218
226
.
Frhs
,
R. R.
(
1972
).
Interaction of L cells and Chlamydia psittaci: Entry of the parasite and host responses to its development
.
J. Bact
.
110
,
706
721
.
Gambrill
,
M. R.
&
Wisseman
,
C. L.
(
1973
).
Influence of human immune serum and complement on the fate of Rickettsia mooseri within human macrophages
.
Infect. Immun
.
8
,
631
641
.
Holland
,
P.
,
Holland
,
N. H.
&
Cohn
,
Z. A.
(
1972
).
The selective inhibition of macrophage phagocytic receptors by antimembrane antibodies
.
J. exp. Med
.
135
,
457
475
.
Karakashian
,
S. J.
(
1963
).
Growth of Paramecium bursaria as influenced by the presence of algal symbionts
.
Physiol. Zoöl
.
36
,
52
68
.
Karakashian
,
S. J.
(
1970
).
Morphological plasticity and the evolution of algal symbionts
.
Ann. N. Y. Acad. Sci
.
175
,
474
, 487.
Lenhoff
,
H. M.
(
1968
).
Chemical perspectives on the feeding response, digestion and nutrition of selected coelenterates
.
Chem. Zool
.
2
,
157
221
.
Lenhoff
,
H. M.
&
Brown
,
R. D.
(
1970
).
Mass culture
of hydra
:
an improved method and its application to other aquatic invertebrates
.
Lab. Animals
4
,
139
154
.
Levine
,
L.
(
1973
).
Micro-complement fixation
.
In Handbook of Experimental Immunology
(ed.
D. M.
Weir
), 2
nd
Edn
, p.
22
.1.
Oxford
:
Blackwell Scientific
.
Martinez-Palomo
,
A.
(
1970
).
The surface coats of animal cells
.
Int. Rev. Cytol
.
29
,
29
-^75.
Medearis
,
D. N.
,
Camitta
,
B. M.
&
Heath
,
E. C.
(
1968
).
Cell wall composition and virulence in Escherichia coli
.
J. exp. Med
.
128
,
399
414
.
Muscatine
,
L.
(
1965
).
Symbiosis of hydra and algae III. Extracellular products of the algae
.
Comp. Biochem. Physiol
.
16
,
77
92
.
Muscatine
,
L.
(
1974
).
Endosymbiosis of cnidarians and algae
.
In Coelenterate Biology: Reviews and Prospectives
(ed.
L.
Muscatine
&
H. M.
Lenhoff
), pp.
359
395
.
New York
:
Academic Press
.
Muscatine
,
L.
,
Cook
,
C. B.
,
Pardy
,
R. L.
&
Pool
,
R. R.
(
1975
).
Uptake, recognition and maintenance of symbiotic Chlorella by Hydra viridis
.
Symp. Soc. exp. Biol
.
29
,
175
203
.
Nowotny
,
A.
(
1969
).
Basic Exercises in Immunochemistry
.
New York
:
Springer-Verlag
.
Pardy
,
R. L.
(
1976
).
The morphology of green Hydra endosymbionts as influenced by host strain and host environment
,
J. Cell Sci
.
20
,
655
669
.
Pardy
,
R. L.
&
Muscatine
,
L.
(
1973
).
Recognition of symbiotic algae by Hydra viridis
.
A quantitative study of the uptake of living algae by aposymbiotic H. viridis. Biol. Bull. mar. biol. Lab., Woods Hole
145
,
505
579
.
Pfeiffer
,
S. E.
,
Herschman
,
H. R.
,
Lightbody
,
J. E.
,
Sato
,
G.
&
Levine
,
L.
(
1971
).
Modification of cell surface antigenicity as a function of culture conditions. J
.
cell. Physiol
.
78
,
145
152
.
Roth
,
S.
(
1973
).
A molecular model for cell interactions
.
Q. Rev. Biol
.
48
,
541
563
.
Silverstein
,
S. C.
(
1970
).
Macrophages and viral immunity
.
Seminars in Hemat
.
7
,
185
214
.
Silverstein
,
S. C.
,
Steinman
,
R. M.
&
Cohn
,
Z. A.
(
1977
).
Endocytosis
.
A. Rev. Biochem
.
46
,
669
722
.
Sneath
,
P. H. A.
&
Lederberg
,
J.
(
1961
).
Inhibition by periodate of mating in Escherichia coli
.
Proc. natn. Acad. Sci. U. S. A
.
47
,
86
90
.
Watkins
,
J. F.
(
1964
).
Observations on antigenic variation in a strain of Trypanosoma brucci growing in mice
.
J. Hyg., Camb
.
62
,
69
80
.