(Jap. J. Parasit., Vol. 32, No. 6, 489-500, December, 1983]

Antibody-Dependent Adherence and Damage to the Young

Adult Worms of Angiostrongylus cantonensis by Rat Eosinophils in Vitro

Kentaro YOSHIMURA*!, Kikue UCHIDAt, Keiko SATOt AND HlROSHI OYAt

(Received for publication; June 13, 1983)

Key words: Angiostrongylus cantonensis, rat-eosinophils, cell-adherence, IgG antibody, in vitro, electron microscopy

that the protective immunity could be Introduction transferred a certain extent to naive rats by intraperitoneal injections of serum Angiostrongylus cantonensis is a neuro- (Yong and Dobson, 1982b) or lymph node tropic and pulmonary parasite in rats. The cells (Yong and Dobson, 1982c) from im parasite occasionally infects man, causing a mune rats. Specific worm stage(s) vulner condition known as eosinophilic meningo- able to the host protective immunity has encephalitis (Alicata and Jindrak, 1970). to be determined, however. When infective third-stage larvae are eaten On the other hand, peripheral eosino- by rats, they migrate to and develop in philia occurs in infected rats 6-7 weeks the rat brains for 26-29 days after infec postinfection, when oviposition by female tion, when they move to the pulmonary worms or hatching of the first-stage larvae arteries, mature, mate and lay eggs. The usually occurs in the lungs (Au and Ko, infected rats usually acquire protective 1979a, b). Although the possible role of immunity to the parasite (Lim et al., 1965; theeosinophilia in A. cantonensis infec Yoshimura et «/., 1979a, b; Au and Ko, tion in rats has not been identified, it has 1979a; Yong and Dobson, 1982a), although been revealed that eosinophils would play the mechanism(s) of the protective immu an important role in killing of young nity and the major sites of worm destruc adult worms surgically transferred into the tion remain to be determined (Au and Ko, pulmonary arteries of nonpermissive hosts 1979a). It has been recently established (Yoshimura et al, 1980, 1983). In an attempt to gain an insight into the possible effector role of the eosinophils This study was supported in part by a grant-in- aid for Scientific Research (No. 457119) from the in A. cantonensis infection in rats and Ministry of Education, Science and Culture of nonpermissive hosts as well, we investi Japan. gated the ability of the cells to adhere to * Present Address: Department of Parasitology, and cause damage to the parasite in vitro. Akita University School of Medicine, Hondo, Akita, 010, Japan. t Department of Parasitology, Juntendo University Materials and Methods School of Medicine, Hongo, Bunkyo-ku, Tokyo, 113, Japan. X Laboratory of Parasitology, Azabu University, Parasite: The life cycle of a Hawaiian Scfiool of Veterinary Medicine, Fuchinobe, Saga- mihara-shi, Kanagawa-ken, 229, Japan. strain of A. cantonensis has been main-

( 1 ) 490 tained in our laboratory through Wistar Tokyo) supplemented with 100 units /ml rats and Biomphalaria glabrata snails. The penicillin and 100 /ig/nil streptomycin. adult worms of a Japanese sheep liver Preparation of eosinophil-rich peritoneal fluke (Fasciola sp.) were collected from the exudate cells (PEC): Five-week A. canto bile ducts of a cow at a local slaughter nensis (60 third-stage larvae/rat)-infected house, and used for obtaining intrauterine F344 rats were intraperitoneally injected eggs for miracidial hatching. Lymnea with 10 ml of 0.9% saline 48 hr prior to ollula snails exposed to the miracidia were cell harvest. The rats were killed with maintained in an aquarium, from the wall ether, the abdominal skin cut open and of which the metacercariae were harvested 50 ml of TC 199 (Nissui Seiyaku, Tokyo) 2-3 months later. supplemented with heparin (5 units/ml) Animals and injection with A. canto- injected into the peritoneal cavity. After nensis and Fasciola sp.: Male outbred gentle massage for 1 min, the fluid and Wistar rats weighing 250-300 g and inbred cells were then withdrawn using a Pasteur F344 rats weighing 200 g were utilized. pipette. Cells obtained from 2-3 rats were The experimental infection of rats with suspended in RPMI 1640 supplemented A. cantonensis was performed as described with 10% heat-inactivated fetal calf serum, elsewhere (Yoshimura and Soulsby, 1976). penicillin (100 units/ml) and streptomycin Rats were also given Fasciola sp. meta (lOO^g/ml) (10% FCS/RPMI), and incu cercariae orally. bated in plastic tissue culture dish (Falcon Antisera: Sera were obtained at ap 3002, Oxnard, Ca.) for 2 hr at 37 C in 5% propriate intervals from F344 rats infected CO2 in air to remove the adherent cells. with 60 third-stage larvae of A. cantonensis Then, the dish was gently agitated and and also from Wistar rats infected with 30 the supernatant removed and centrifuged metacercariae of Fasciola sp. as described at 1500 rpm for 10 min. Following re previously (Yoshimura and Soulsby, 1976), moval of the supernatant, the number of and stored at —80 C until use. cells and eosinophils remaining in the Serology: Phosphate buffered saline (pH sediments was determined by Turk's stain 7.2) extracts of A. cantonensis and Fasciola and Discombe's stain respectively using a sp. adult worms were prepared by the hemocytometer. Differential cell counts of method of Yoshimura and Soulsby (1976). the non-adherent cells were also made on Gel diffusion for assessing precipitin titers Giemsa stained smear preparations. These of sera from rats infected with A. canto cell preparations (PEC) (N=22) contained nensis or Fasciola sp. was carried out as a mean % eosinophil count of 64.6±9.5, described elsewhere (Ouchterlony and neutrophil count of 5.0±3.3, lymphocyte Nillsson, 1973). Forty-eight hr-passive cu count of 11.0±3.2, macrophage count of taneous anaphylaxis (PCA) reaction to 17.0d=7.1, and mast cell count of 2.4±2.8. assess reaginic antibody was performed as In vitro cell adherence assay: Eight to previously described (Yoshimura et ah, 10 worms of Weeks-2 or 3 intracranial A. 1979b). cantonensis were added to 0.2 ml of PEC Preparation of intracranial worms: suspension, 0.2 ml of heat-inactivated (hi) Weeks 2 and 3 intracranial worms of A. infected rat serum (IRS) or hi-normal rat cantonensis were harvested from the sub- serum (NRS), or hi-RPMI 1640, and 0.2 ml arachnoid spaces of rats infected with of 10% FCS/RPMI in Nunc multidish 160-180 third-stage larvae, as described wells (Cat. No. 143982, Inter Med, Den previously (Yoshimura et ah, 1980), and mark) and incubated at 37 C in 5% COo collected in RPMI 1640 (Nissui Seiyaku, in air. The third-stage larvae for cell ad- 491 herence assay were harvested from infected In order to determine the end titer for B. glabrata snails by digestion with arti cell adherence, a serum pool from 8-week ficial gastric juice and separated by a infected rats was used; the study indicated slight modification of the method of that a 1:384 dilution was still capable of Oshima (1976). For 3rd-stage larvae, 0.05 inducing significant cell adherence to Week ml of 10% FCS/RPMI containing 30-40 3 worms. The final serum dilution of larvae was added to 0.05 ml of PEC sus 1:24 wras thus utilized throughout this pension plus 0.05 ml of IRS or NRS in study. microslide culture chambers (Bellco Glass, Column chromatography: Two ml of Vineland, NJ.), and incubated as described IRS obtained from 8-week infected rats above. were gel-filtrated on a Sephadex G-200 At the termination of incubation, all column (25x720 mm) as described previ worms of each well were transferred to a ously (Yoshimura and Yamagishi, 1976), glass slide with a small amount of 0.9% and fractions obtained were concentrated saline solution for light microscopic ex using a collodion bag (Sartorius) and then amination. Cell attachment to the parasite used for cell adherence assays. was quantitated by assessing cellular ad Inhibition of cell adherence by Protein A: herence to the exterior surface of the Protein A (Staphylococcus aureus) (Phar worms, and scored as follows: (a) 4-f, total macia) was dissolved in RPMI 1640 to worm surface was covered with cells; (b) 2 mg/ml. 0.2 mg of Protein A was then 3+, 60-90% of the surface covered; (c) added to each well containing IRS or its 2+, 30-60% of the surface covered; (d) IgG fraction, PEC suspension, and Week 1+, 10-30% of the surface covered; and 3 worms. (e) 0, less than 10% of the surface covered. Transmission electron microscopy: After A preliminary study on PEC: worm their incubation with PEC and IRS, or (Week 3 worms) ratio, using IRS obtained NRS for 2 or 20 hr, the Week 3 worms from 8-week infected rats and 2 hr-incu- were washed several times in 0.9% saline bation system, indicated that significant and then fixed for electron microscopy in cell adherence occurred at a ratio of 2.0% cold glutaraldehyde in 0.1 M phos 100,000:1 (mean score ± SD = 2.3+0.7) or phate buffer, and post-fixed in 2.0% osmi 200,000:1 (2.6+0.8) but not at a ratio of um tetroxide; the specimens were then 20,000:1 (0.9+0.8). The ratio of 100,000:1 dehydrated and embedded in epon 812. was thus employed throughout this study. The preparation of ultrathin sections, A ratio of 20,000:1 was, however, em staining with uranyl acetate and lead ployed for the assay of the 3rd-stage larvae, citrate, and examination with a Hitachi since their size is extremely small as com H-500 electron microscope were carried pared with that of the Week 3 worms. out as described elsewhere (Yoshimura et The time course of PEC adherence to al, 1983). Week 3 worms in the presence of IRS was In order to determine the localization followed up to 48 hr after incubation. A of antibody on the worm surface, Week 3 peak cell adherence (score=2.8±0.5, N= worms incubated in IRS or NRS plus 90) was noted at 2 hr of incubation and PEC for 1 hr at 37 C were washed five gradually decreased to control (NRS or times in 0.9% saline and re-incubated for RPMI 1640) levels at 48 hr. The evalu an additional hour in a ferritin-conjugated ation of cell adherence in the present IgG fraction rabbit anti-rat IgG (1:5 dilu study was thus performed at 2 hr follow tion) (Cappel Laboratories, Dawningtown, ing incubation. Pa.). The worms were then washed several

( 3 ) 492 times in saline, fixed in glutaraldehyde and processed as described above. Scanning electron microscopy: Worms 3.0 incubated with PEC and NRS or IRS for 2 hr were washed five times in 0.9% saline and fixed as described for transmission electron microscopy. The worms were de hydrated in ethanol and dried in a Hitachi a> 2.0 o HCP-1 critical point dryer. Specimens were then coated with gold-palladium in an ion coater (Type IB-3, Eiko Engineering, Tokai-mura, Ibaragi-ken) and examined in a Hitachi S-450 scanning electron micro scope. Pulmonary arterial transfer of worms: In order to determine whether cell ad herence reactions to Week 3 worms in vitro affect their subsequent survival in 0 2 4 6 8 10 12 . 14 16 18 vivo, the worms incubated with PEC and Weeks after infection IRS, or NRS (or RPMI 1640) for 20 hr wTere washed several times in 0.9% saline Fig. 1 Time course of the production of anti and surgically transferred into the pulmo body capable of promoting in vitro cell ad herence to Week 3 intracranial worms in rats nary arteries of naive recipient rats by the infected with A. cantonensis. The values were method of Yoshimura et al. (1979b). The obtained from three independent experiments. rats were necropsied 24 days posttransfer •, mean score for male worms; O, mean score for female worms; X with vertical baf, mean for determining worm recovery. ± SD for male plus female worms. Shaded Statistical analysis: Mean cell adherence area denotes mean score ± SD for worms incu scores and mean worm recovery rates were bated only with eosinophil-rich peritoneal calculated on individual groups and sta exudate cells in RPMI 1640. tistical differences in these values between groups were assessed by Student's t-test. served even when they were incubated for 20 hr in the presence of either IRS (ob Results tained from 2-, 4-, and 8-week infected rats) or NRS. The PEC adherence, how Specific antibody-dependent adherence of ever, occurred on Week 2 intracranial PEC to Weeks 2 or 3 worms worms in the presence of IRS (obtained An initial experiment was attempted to from 4-, 8- and 12-week infected rats) and determine when PEC adherence-promoting their cell adherence scores were compara antibody to Week 3 worms appears in sera ble to those of Week 3 worms. The data of rats infected with A. cantonensis. Fig. 1 are thus suggestive of the stage specificity indicates that this antibody could not be of the antibody-dependent PEC adherence. detected in sera from rats 2 weeks follow In an attempt to determine the antibody ing infection but appeared by 4 wTeeks and specificity of this PEC adherence, sera then being consistently present throughout from 8-, 10-, and 12-week Fasciola sp. in the course of the observations. fected rats were used for assessing their The antibody-dependent PEC adherence capability of promoting PEC adherence to the third-stage larvae could not be ob- to Week 3 worms. These sera, however, 493

Table 1 Opsonizing effect of infected rat serum on the in vitro cell adherence reactions to Week 3 intracranial Angiostrongylus cantonensis

2nd Washing Cell adherence score* 1st incubation in incubation added with RPMI Male Worms Female worms Mean fresh

IRS + Worms Yes PEC 2.1 + 1.0 (41) 3.0±0.9 (41) 2.6±l.la (82) IRS + Worms No PEC 3.5±0.6 (33) 3.8±0.4 (42) 3.7±0.5b (75) None None IRS + Worms + 3.5±0.6 (41) 3.7+0.5 (42) 3.6+0.6c (83) PEC RPMI + Worms No PEC 0.5+0.6 (43) 0.5±0.6 (42) 0.5±0.6d (85) IRS + PEC Yes Worms 0.8+0.6 (43) 1.0±0.7 (42) 0.9+0.7e (85) IRS+PEC No Worms 3.5±0.7 (41) 3.5+0.6 (40) 3.5±0.6f (81)

IRS=8-week infected rat serum. PEC = Eosinophil-rich peritoneal exudate cells. =RPMI 1640. * Mean±SD (N) of duplicate assays in three independent experiments. a, b, c and f differed significantly from d. Differences between a and e, and between d and e, were also significant (p<0.001).

Table 2 Immunoglobulin class involved in the in vitro cell adherence reactions to Week 3 intracranial Angiostrongylus cantonensis

Cell adherence score*

Male worms Female worms Mean

Gel filtration IgM 1.1+0.8 (29) 1.5±1.2 (32) 1.3±l.la (61) IgG 2.9±1.1 (29) 2.4+0.7 (32) 2.6±0. 9b (61) Albumin 1.0+0.7 (30) 1.0±1.0 (32) 1.0±0. 8C (62) IRS 3.6+0.6 (30) 2.9±1.0 (33) 3. 3±0. 9d (63) PBS 0. 4±0.6 (29) 0.4+0.6 (30) 0. 4±0.6 (59) RPMI 1640 0.4+0.6 (30) 0.4+0.7 (30) 0.4±0.6 (60)

Inhibition by Protein A (PA) IgG 2.1±1.0 (30) 3.0±1.2 (30) 2.5±1.2e (60) IgG + PA ].4±0.9 (30) 1.5±1.1 (30) 1.5±1.0f (60) IRS 2.8±0.8 (30) 2.8±0. 8 (30) 2.8±0. 8^ (60) IRS+PA 2.3±1.0 (29) 2.6±0. 7 (31) 2. 5±0. 8h (60) NRS 0.5 + 0.5 (15) 0.5 ±0.6 (15) 0.5 + 0.6 (30) RPMI 1640 0.4 + 0.6 (23) 0.3 ±0.5 (38) 0.4 + 0.5 (61)

IRS=8-week infected rat serum. PBS=Phosphate buffered saline (pH. 7. 2) used for eluate of gel filtration. * Mean±SD (N) of duplicate assays in two independent experiments. d differed significantly from a, b, and c. Other statistical differences: a vs b p<0. 001; b vs c P<0. 001; e vs f p<0. 001; g vs h p<0. 05.

failed to induce PEC adherence, in spite worms and cells were then washed three of the presence of a.nti-Fasciola sp. anti times in large volumes of RPMI 1640. body (precipitin titer=l:32). Fresh PEC were added to the preincubated The opsonic or cy tophi lie nature of the worms whereas fresh worms were added to antibodies was further studied by incuba the preincubated PEC. Both preparations tion of IRS or RPMI 1640 with Week 3 were incubated in the absence of any worms, or with PEC for 2 hr at 37 C. The added rat serum for additional 2 hr. As

( 5 ) 494 a control, the group of simultaneous incu ferri tin-conjugated rabbit anti-rat IgG bation of worms and PEC in the presence demonstrated that the label was found on of IRS was also provided. Table 1 indi the surface of the Week 3 worms at the cates that significant PEC adherence (score sites of eosinophil-parasite contact (Fig. 2). =2.6) was observed when preincubated To the contrary, no label could be found worms were used, whereas only slight PEC on the surface of fresh parasite immedi adherence (0.9) was observed with the use ately after recovery from the rat brains. of preincubated PEC. In order to examine the possible in Determination of antibody class involved volvement of IgE antibody in the ad in the PEC adherence herence reaction, PCA positive serum A serum pool from 8-week infected rats samples from rats taken 6 and 8 weeks was fractionated by Sephadex G-200 gel after infection were heated at 56 C for filtration. Table 2 indicates that 7S IgG 2 hr and resulting PCA negative sera were fraction (score=2.6) contained most of the tested for adherence-promoting activity. activity present in the whole IRS (3.3), PCA negative sera were then found to be although a slight activity was noted in still capable of inducing PEC adherence both 19S IgM and albumin fractions, the reaction. activities of which were probably due to Damage to Week 3 worms by PEC ad the tailing contamination of IgG. An ad herence reaction ditional approach to identify the class of As the PEC adherence did not affect immunoglobulin involved in the PEC ad the survival of worms during the period herence reaction was made by assessing of incubation, the in vitro toxic effect of the inhibitory effect of Protein A on anti the PEC adherence on Week 3 worms was body-cell reaction, since this agent is confirmed by assessing their in vivo survi known to bind to the Fc portion of rat val, when they were transferred into the IgG (Kronvall et al, 1970). Table 2 shows lungs of naive recipient rats. Table 3 that the addition of Protein A to 7S IgG shows that the survival of both male fraction or whole IRS caused a significant (Exp. I) and female (Exp. II) worms incu depression of the adherence reaction. bated with PEC plus IRS was significantly These data suggest that the mechanism of less than that of those worms incubated PEC adherence is mediated through the with IRS alone, or with PEC plus RPMI Fc portion of anti-A. cantonensis IgG. 1640 or with PEC plus NRS. This is indic Immunoelectron microscopic study using ative of the possibility that the antibody-

Explanation of Figures

Fig. 2 Transmission electron micrograph showing interaction between Week 3 intracranial worm and an eosinophil (eo). Ferritin particles (arrow) are seen on the cuticular (c) surface of the parasite (x 45,000). Fig. 3 Scanning electron micrograph showing the attachment of many cells to the surface of Week 3 intracranial worm. Note flattened cells and pseudopodia extending from the cells to cover the worm surface (x2,160). Fig. 4 Transmission electron micrograph showing an eosinophil (eo) adhering to the cuticle (c) of Week 3 intracranial worm. Note electron dense material (arrow) between the parasite and cell sur face. Large cytoplasmic vacuoles (arrowhead) containing dense, coarsely granular materials are seen in the eosinophil (X25,2OO). Fig. 5 Transmission electron micrograph showing an eosinophil (eo) adhering to the cuticle (c) of Week 3 intracranial worm after 20hr in culture. Note ragged losses of granule cores (arrow) (X 12,600). 495

Wf*H

8 <+ \ ^

( 7 ) 496

Table 3 The survival of Week 3 intracranial Angiostrongylus cantonensis incubated with eosinophil-rich peritoneal exudate cells plus infected rat or normal rat serum, when transferred into the pulmonary arteries of naive recipient rats

Exp. I Exp. II

Worms No. of % Worm recovery* No. of % Worm recovery* incubated with rats rats implanted Male Female implanted Male Female

PEC + IRS 18 39. 5±24. 0a 32.9 + 33. 1 11 53.5 + 28.9 40. 0±23. 9d PEC + NRS — 9 50. 8±28. 9 54. 5±21.2e IRS 15 58.2±20.9b 25.9 + 28.0 PEC + RPMI 1640 14 72. 9±27. 2C 25. 2±24. 9 62. 5±32. 0 59. 3+22. 2f

Experiments I and II were carried out in quintuplicate and triplicate respectively. PEC=Eosinophil-rich peritoneal exudate cells. IRS=8-week infected rat serum. NRS=Normal rat

serum. * Mean±SD. Statistical differences: avsbp<0.05; avscp<0.001; dvsep<0. 02; dvsfp<0. 001. dependent PEC adherence caused damages known, the mechanism of eosinophilia and to the worms, leading to less worm survival the functional role of the cells have only in vivo. recently begun to be elucidated. The Electron microscopy of PEC-pamsite inter evaluation of eosinophilia in parasitic in action fections was stimulated by in vivo infor Scanning electron microscopy indicates mation that the acquired resistance of mice that, when the Week 3 worms were incu to Schistosoma mansoni was diminished bated with PEC in the presence of IRS by depletion of eosinophils in these mice for 2 hr, the surface of the worms was with monospecific anti-eosinophil serum attached with numerous cells, some of (Mahmoud et al.t 1975) and also by an ad which were flattened and extended pseudo- ditional in vitro evidence that the eosino podia over the surface of the parasite phil could be an effector cell capable of (Fig. 3). With transmission electron micro inducing antibody-dependent damage to scopy, most of the cells adhering to the schistosomula of S. mansoni (Butterworth surface of the worms were found to be et ah, 1975). Following these pilot studies, eosinophils, and some extracellular elec IgG antibody-dependent eosinophil ad tron dense material was found deposited herence to and killing of the newborn at the sites of eosinophil-parasite contact larvae of Trichinella spiralis (Kazura and (Fig. 4). It is also of interest to note that Grove, 1978; Kazura and Aikawa, 1980) the eosinophil included large cytoplasmic and of the microfilariae of Dipetalonema vacuoles containing electron dense, coarse viteae (Rudin et al., 1980) and Onchocerca ly granular materials (Fig. 4). At 20 hr volvulus (Greene et al., 1981) have been after incubation, some degenerating eosino identified. phils with ragged losses of granule cores The current study clearly indicates that were found attached to the surface of the specific IgG antibody-dependent eosinophil worms (Fig. 5). adherence occurs on Weeks 2-3 intra cranial worms of A. cantonensis but not Discussion on the third-stage larvae, and also that the cell adherence caused marked damages to Although the association of eosinophilia the worms as confirmed by worm transfer with helminth infections has been well experiments. Although the in vivo sig-

( 8 ) 497 nificance of this in vitro PEC adherence determined. Electron microscopic evidences is still unclear, the finding is suggestive of showing the ragged losses of granule cores the possibility that the eosinophil would and the presence of large cytoplasmic play an important role in protective im vacuoles containing lysosomal granules are, munity in A. cantonensis infections in rats however, suggestive of the killing mecha (Lim et al, 1965; Yoshimura et al, 1979a, nisms by the eosinophils. It has been well b; Au and Ko, 1979a; Yong and Dobson, established that major basic protein (MBP) 1982a). Our data show that in at least in granule core is highly cytotoxic to in vitro systems, Weeks 2 or 3 intracranial schistosomula of S. mansoni (Butterworth worms were vulnerable to the antibody- et al, 1979), the newborn larvae of T. dependent eosinophilic attack. Although spiralis (Wassom and Gleich, 1979), tumor no data are available as to the in vivo cells (Butterworth et al, 1979) and various vulnerability of these particular stages of normal cells (Gleich et al., 1979). Ultra- worms, the current data may provide an structural evidences indicating similar insight that the infective third-stage larvae losses of granule cores have been reported would possibly be refractory to immuno- in those cells adhering to the larvae of T. logical killing mechanisms in vivo. spiralis in the presence of immune serum Additionally, the present results may (Perrudet-Badoux et ah, 1978) and adher serve as an indirect evidence of potential ing to the microfilariae in the lymph cytotoxic action of the eosinophil to A. nodes of onchocerciasis volvulus patients cantonensis in nonpermissive hosts. Con (Racz et al., 1982), and those cells in the sistent with this supposition is the obser ileum of Crohn's disease (Dvorak, 1980), vation that the eosinophil would be closely in all cases of which cytotoxic action of associated with the killing mechanism of MBP has been suggested to be involved. young adult worms surgically transplanted The involvement of lysosomal enzymes into the pulmonary arteries of the non- from granule matrices, e.g., peroxidase permissive hosts (Yoshimura et al, 1980, (Jong et al., 1981; Kazura et al, 1981; Buys 1983). et al, 1981) cannot be also excluded, since The present work revealed that the the formation of the cytoplasmic vacuoles eosinophil adherence could be mediated in the eosinophils adhering to the cuticular by IgG class antibody, the opsonic (but not surface of A. cantonensis, and the deposit cytophilic) action of which was responsi of electron-dense, secretion material be ble for the adherence reaction (Table 1). tween the eosinophil and the parasite, are When one considers that both eosinophils consistent with those observations in the (PEC) and intracranial worms were har cells attached to schistosomula of S. man- vested from infected rats, it seems likely soni and the newborn larvae of T. spiralis that the cells and parasites might have in vitro (McLaren et al, 1977; McLaren been sensitized with specific IgG antibody et al, 1978; Kazura and Aikawa, 1980). in vivo. It may be impossible, therefore, In addition to the IgG antibody-depen to deny the involvement of the cytophilic dent eosinophil adherence, IgE-dependent effect of the IgG antibody in the present adherence and damage to S. mansoni schis PEC adherence. This assumption may be tosomula (Capron et al, 1981), and com supported by the fact that slight cell ad plement-dependent eosinophil adherence herence (scores0.5) occurred even in the to and killing of S. mansoni (Rahalho- absence of IRS (Table 1). Pinto et al, 1978; McLaren and Rahalho- Cytotoxic factor(s) of eosinophils to Pinto, 1979; Anwar et al, 1979; McKean Week 3 young adult worms remains to be et al, 1981) have been also evaluated. In

( 9 ) 498

the present investigation, IgE antibody- to the worms were demonstrated by the dependent cell adherence could be ex reduction of worm recovery rates when cluded, but the presence of complement- those worms with adherent PEC were dependent adherence and damage to A. surgically transferred into the pulmonary cantonensis remains to be evaluated, since arteries of naive recipient rats. These data all the sera tested in this study were heat- suggest that the antibody-dependent eosino- inactivated. phil adherence would possibly be associ When A. cantonensis infects nonpermis- ated with the protective immunity to A. sive hosts such as mice and guinea pigs, cantonensis infection in rats. the parasite migrates to and develops in the brain for about 20 days after infection Acknowledgements

but dies there without moving to the lungs We are grateful to Mr. T. Murai for skilled (Yamashita et al., 1975). The mechanism technical assistance in preparing ultrathin sections. of this worm death is still unclear, but it Thanks are also due to Mrs. Keiko Yamashita for assistance in preparing figure. is of interest to note that marked eosino- philia can be usually observed in the References

cerebrospinal fluid of human angiostron- 1) Alicata, J. E. and Jindrak, K. (1970): Angio- gyliasis cantonensis patients (Alicata and strongylosis in the Pacific and Southeast Asia. Jindrak, 1970). In addition to their regu Charles C. Thomas, Springfield, Illinois. 2) Anwar, A. R. E., Smithers, S. R. and Kay, A. latory functions of allergic inflammations B. (1979): Killing of schistosomula of Schisto- (Gleich, 1977; Kay, 1979), therefore, eosino- soma mansoni coated with antibody and/or phils may have more crucial roles in kill complement by human leukocytes in vitro: ing A. cantonensis in the brains of the Requirement for complement in preferential nonpermissive hosts than in those of the killing by eosinophils. J. Immunol., 122, 628- 637. rat. 3) An, A. C. S. and Ko, R. C. (1979a): Changes in worm burden, haematological and serologi- Summary cal response in rats after single and multiple Angiostrongylus cantonensis infections. Z. Para- Eosinophil-rich rat peritoneal exudate sitenkd., 58, 233-242. 4) Au, A. C. S. and Ko, R. C. (1979b): Cross- cells (PEC) containing 65±9.5% eosino- resistance between Trichinella spiralis and phils were harvested from 5-week Angio- Angiostrongylus cantonensis in laboratory rats. strongylus cantonensis infected F344 rats. Z. Parasitenkd., 59, 161-168. In the presence of heat-inactivated serum 5) Butterworth, A. E., Sturrock, R. F., Houba, V., from infected rats, PEC adhered to 2-3- Mahmoud, A. A. F., Sher, A. and Rees, P. H. (1975): Eosinophils as mediators of antibody- week old intracranial worms but not to dependent damage to schistosomula. Nature, the third-stage larvae of A. cantonensis. 256, 727-729. This PEC adherence could not be induced 6) Butterworth, A. E., Wassom, D. L., Gleich, G. by znti-Fasciola sp. rat serum, being sug J., Loegering, D. A. and David, J. R. (1979): gestive of specific antibody dependence. Damage to schistosomula of Schistosoma man soni induced directly by eosinophil major basic The PEC adherence was also dependent protein. J. Immunol., 122, 221-229. on the opsonic nature of IgG antibody 7) Buys, J., Wever, R., van Stigt, R. and Ruiten- that appeared by 4 weeks after A. canto berg, E. J. (1981): The killing of newborn nensis infection and persisted until at least larvae of Trichinella spiralis by eosinophil 18 weeks. Immunoelectron microscopic peroxidase in vitro. Eur. J. Immunol., 11, 843-845. study indicated that the IgG antibody was 8) Capron, M., Bazin, H., Joseph, M. and Capron, localized at the areas of cell-parasite con A. (1981): Evidence for IgE-dependent cyto- tact on the surface of the worms. Damages toxicity by rat eosinophils. J. Immunol., 126,

( 10 ) 499

1764-1768. B. (1981): Scliistosoma mansoni: Complement 9) Dvorak, A. M. (1980): Ultrastructural evidence and antibody damage, mediated by human for release of major basic protein-containing eosinophils and neutrophils, in killing schisto crystalline cores of eosinophil granules in vivo: somula in vitro. Exp. Parasit., 51, 307-317. cytotoxic potential in Crohn's disease. J. Im 22) McLaren, D. J., Mackenzie, C. D. and Ramalho- munol., 125, 460-462. Pintc, F. J. (1977): Ultrastructural observa 10) Gleich, G. J. (1977): The eosinophil: New tions on the in vitro interaction between rat aspects of structure and function. J. Allergy eosinophils and some parasitic helminths Clin. Immunol., 60, 73-82. (Schistosoma mansoni, Trichinella spiralis and 11) Gleich, G. J., Frigas, E., Loegering, D. A., Nippostrongylus brasiliensis). Clin. Exp. Im Wassom, D. L. and Steinmuller, D. (1979): munol., 30, 105-118. Cytotoxic properties of the eosinophil major 23) McLaren, D. J. and Ramalho-Pinto, F. J. basic protein. J. Immunol., 123, 2925-2927. (1979): Eosinophil-mediated killing of schisto 12) Greene, B. M., Taylor, H. R. and Aikawa, M. somula of Schistosoma mansoni in vitro: Syner- (1981): Cellular killing of microfilariae of gistic effect of antibody and complement. J. Onchocerca volvulus: Eosinophil and neutro- Immunol., 123, 1431-1438. phil-mediated immune serum-dependent de 24) McLaren, D. J., Ramalho-Pinto, F. J. and struction. J. Immunol., 127, 1611-1618. Smithers, S. R. (1978): Ultrastructural evidence 13) Jong, E. C, Mahmoud, A. A. F. and Klebanoff, for complement and antibody-dependent dam S. J. (1981): Peroxidase-mediated toxicity to age to schistosomula of Schistosoma mansoni schistosomula of Schistosoma mansoni. J. Im by rat eosinophils in vitro. Parasitology, 77, munol., 126, 468-471. 313-324. 14) Kay, A. B. (1979): The role of the eosinophil. 25) Oshima, T. (1976): Observations of the age J. Allergy Clin. Immunol., 64, 90-104. resistance, eosinophilia, and larval behavior 15) Kazura, J. W. and Aikawa, M. (1980): Host in the helminth-free beagles infected with defense mechanisms against Trichinella spiralis Toxocara canis. Jap. J. Parasit., 25, 447-455. infection in the mouse: Eosinophil-mediated 26) Ouchterlony, 6. and Nillsson, L. A. (1973): destruction of newborn larvae in vitro. J. Im Immunodiffusion and immunoelectrophoresis. munol., 124, 355-361. In Handbook of Experimental Immunology, 16) Kazura, J. W., Fanning, M. M., Blumer, J. L. Vol. 1, ed. by D. M. Weir, Blackwell, Oxford, and Mahmoud, A. A. F. (1981): Role of cell- 19.1-19.39. generated hydrogen peroxide in granulocyte- 27) Perrudet-Badoux, A., Anteunis, A., Dumitrescu, mediated killing of schistosomula of Schisto S. M. and Binaghi, R. A. (1978): Ultrastruc soma mansoni in vitro. J. Clin. Invest., 67, tural study of the immune interaction between 93-102. peritoneal cells and larvae of Trichinella 17) Kazura, J. W. and Grove, D. I. (1978): Stage- spiralis. J. Reticuloendothel. Soc, 24, 311-314. specific antibody-dependent eosinophil-mediated 28) Racz, P., Tenner-Racz, K., Buttner, D. W. and destruction of Trichinella spiralis. Nature, 274, Albiez, E. J. (1982): Ultrastructural evidence 588-589. for eosinophil-parasite adherence (EPA) re 18) Kronvall, G., Seal, U. S., Finstad, J. and Wil action in human onchocercal lymphadenitis in liams, R. C. Jr. (1970): Phylogenetic insight the early period following diethylcarbamazine into evolution of mammalian Fc fragment of treatment. Tropenmed. Parasit., 33, 213-218. yG globulin using staphylococcal protein A. }. 29) Ramalho-Pinto, F. J., McLaren, D. J. and Immunol., 104, 140-147. Smithers, S. R. (1978): Complement-mediated 19) Lim, B. L., Ow-Yang, C. K. and Lie, K. J. killing of schistosomula of Schistosoma mansoni (1965): Natural infection of Angiostrongylus by rat eosinophils in vitro. J. Exp. Med., 147, cantonensis in Malayan rodents and intermedi 147-156. ate hosts, and preliminary observations on 30) Rudin, W., Tanner, M., Bauer, P. and Weiss, acquired resistance. Am. J. Trop. Med. Hyg., N. (1980): Studies on Dipetalonema viteae 14, 610-617. (Filarioidea). 5. Ultrastructural aspects of the 20) Mahmoud, A. A. F., Warren, K. S. and Peters, antibody-dependent cell mediated destruction P. A. (1975): A role for the eosinophil in ac of microfilariae. Tropenmed. Parasit., 31, 194- quired resistance to Schistosoma mansoni in 200. fection as determined by antieosinophil serum. 31) Wassom, D. L. and Gleich, G. J. (1979): Dam J. Exp. Med., 142, 805-813. age to Trichinella spiralis newborn larvae by 21) McKean, J. R., Anwar, A. R. E. and Kay, A. eosinophil major basic protein. Am. J. Trop.

( 11 ) 500

Med. Hyg., 28, 860-863. 37) Yoshimura, K., Aiba, H. and Oya, H. (1979b): 32) Yamashita, T., Sato, Y., Shiraki, T., Otsuru, M. Transplantation of young adult Angiostrongylus and Suzuki, T. (1975): Development of Angio cantonensis into the rat pulmonary vessels and strongylus cantonensis in rats, mice and guinea its application to the assessment of acquired pigs. Jap. J. Parasit., 24, 114-121. resistance. Int. J. Parasitol., 9, 97-103. 33) Yong, W. K. and Dobson, C. (1982a): The 38) Yoshimura, K., Aiba, H., Oya, H. and Fukuda, biology of Angiostrongylus cantonensis larvae Y. (1980): Angiostrongylus cantonensis: De in immune rats. Southeast Asia. J. Trop. Med. velopment following pulmonary arterial trans Pub. Hlth., 13, 244-248. fers into permissive and nonpermissive hosts. 34) Yong, W. K. and Dobson, C. (1982b): Antibody Exp. Parasit., 49, 339-352. responses in rats infected with Angiostrongylus 39) Yoshimura, K. and Soulsby, E. J. L. (1976): cantonensis and the passive transfer of protec Angiostrongylus cantonensis'. Lymphoid cell tive immunity with immune serum. Z. Para- responsiveness and antibody production in rats. sitenkd., 67, 329^-336. Am. J. Trop. Med. Hyg., 25, 99-107. 35) Yong, W. K. and Dobson, C. (1982c): The 40) Yoshimura, K., Uchida, K., Sato, K. and Oya, passive transfer of protective immunity against H. (1983): Ultrastructural evidence for eosino- Angiostrongylus cantonensis with immune phil-mediated destruction of Angiostrongylus lymph node cells from different lymphoid cantonensis transferred into the pulmonary tissues in rats. Int. J. Parasit., 12, 423-425. artery of nonpermissive hosts. Parasite Im S6) Yoshimura, K., Aiba, H., Hirayama, N. and munol., (in press). Yosida, T. H. (1979a): Acquired resistance and 41) Yoshimura, K. and Yamagishi, T. (1976): Re- immune responses of eight strains of inbred aginic antibody productions in rabbits and rats rats to infection with Angiostrongylus canto infected with Angiostrongylus cantonensis. Jap. nensis. Jap. J. Vet. Sci., 41, 245-259. J. Vet. Sci., 38, 33-40.

In vitro iC

(Angiostrongylus cantonensis) ^ . PEC

EC) - t.1Z,

igG

pec

, PEC \$ 2-3 . PEC

c. t.ft, ro pec in vitro , #