Development of the serological response in rabbits infected with Toxocara canis and Toxascaris leonina

TRAN~A~~ONS OF THE ROYAL SOCXETY OF TROPICAL. MEDICINE AND HYGIENE, Vor. 76, No. 1, 1982 89 Development of the serological response in rabbits infected with Toxocara canis and Toxascaris leonina Huw V. SMITH, ROSALIND QUINN, ROBERT G. BRUCE' AND ROBERT W. A. GIRDWOOD** Dept. of Zoology, University of Glasgow, Glasgow G12 8QQ, UK; Dept. of Bacteriology, Stobhill General Hospital, Glasgow G21 3UW, UK Summary The indirect fluorescent antibody test using frozen sections of infective Toxocara canis and Toxascaris leon&z eggs, and the enzyme linked immunosorbent assay using homogenized Toxocara canis embryonated egg extract and T canis excretory-secretory products as adsorbed antigens were used to determine the specificity and development of circulating antibodies in rabbits. Frozen sections were subdivided into four morphologically distinct compartments for analysis of the development of the circulating antibody response. The fluid surrounding the larva was the most reactive up to 21 days after infection, and this material was found to be predominantly excretory-secretory in nature. As the infection progressed antibodies directed against somatic tissue materials increased. Cross reactions between sera from rabbits infected with T. canis eggs and Toxascaris leonina frozen sections, and rabbits infected with T. leonina eggs and Toxocara canis frozen sections occurred between both the excretory-secretory fluid and somatic components of the infective eggs. These results were substantiated using the enzyme linked immunosorbent assay. When T. canis excretory-secretory antigen was used, an earlier response (peak day 21) was detected than when using T. canis embryonated egg extract (peak day 35). However, cross reactions between T. canis excretorysecretory antigen and sera from rabbits infected with Toxascaris leonina occurred, indicating that the serodiagnosis of visceral larva migrans using Toxocara canis excretory-secretory antigen may still prove unsatisfactory when considering the role of Toxascaris as a possible causative agent. Introduction The description by BEAVER et al. (1952) of the visceral larva migrans syndrome in man initiated the search for sensitive and specific serological tests for the diagnosis of this condition. Much of the ensuing research concentrated on the detection of circulating antibodies to Toxocara canis. Numerous techniques have been employed and antigenic preparations derived from the L2 larval stage or its products have produced the most promising results. Thus MITCH- ELL (1964) showed tha!, in the experimentally infected rabbit, using the indirect fluorescent antibody technique, antibodies to T. canis larvae could be detected as early as the second week post infection. In additiol, ALJEBOORI & IVEY (1970) using the haemagglutinauon technique with adult, larval, embryonated egg and hatching fluid antigens derived from T. canis found a high degree of specific reactivity to larval antigen with sera from experimentally infected rabbits, and recorded only low titre cross reactivity with sera from rabbits infected with Ascaris suum. The role of T. cati and Toxascaris leonina in the aetiology of human visceral larva migrans remains unknown. The desirability of defining the role of these parasites in the pathogenesis of this condition is obvious both from epidemiological and public health considerations. HOGARTH-SCOTT (1966) working with experimentally infected rabbits, and using the in vitro larval fluorescent precipitin technique, demonstrated interspecies cross-reactions between sera obtained from Toxocara canis and T. cati infections with T. canis and T. cati larval antigen preparations. No such interspecies cross reactions were detected with Toxascaris ieonina antisera or antigen preparations and indeed T. leonina larvae also failed to react with homologous antisera. The present study was undertaken in an attempt to clarify the nature of the early antibody response in rabbits experimentally infected with Toxocara canis and Toxascaris leonina. It was hoped that such a study would indicate which parasite components could be used most usefully to detect human infections and, conversely, it was felt that such studies might suggest some limitations in the interpretation of the results obtained by the serological tests currently in use for the diagnosis of human visceral larva migrans. Materials and Methods Animals New Zealand White x Sandylop rabbits, eight months old, were infected orally with either 4 x 5,000 Toxocara canis infective eggs administered on alternate davs or 50.000 Toxascaris leonina infective eggs as a single dose. The rabbits were bled at weekly intervals and the serum stored at -20 C until used. Preparation of Antigens Eggs of Toxocara canis and Toxascaris leonina were obtained from the uteri of adult worms recovered from the gastro-intestinal tracts of dogs at autopsy. These were washed and stored at room temperature (R.T. 21 C) in 4% formalin. Following embryonation and development to the L2 larva within the egg, the eggs were washed in phosphate buffered saline (PBS) at ph 7.1 before use. Somatic Antigen Infective eggs were either pelleted by centrifugation, mounted in O.C.T. (Miles Laboratories, Stoke *Address for correspondence: Dr. R. W. A. Girdwood, Dept. of Bacteriology, Stobhill General Hospital, Glasgow G21 3UW, UK

90 DEVELOPMENT OF SEROLOGICAL Poges, England) tissue mounting medium and frozen in 2 methyl butane cooled to - 70 C in liquid nitrogen or disrupted as a 20% wet wt/v in PBS ph 7.1 in an LKB X press, allowed to solubilize overnight at 4 C and centrifuged at 12?000 g for 30 min at 4 C and the pellet discarded. This was designated embryonated egg extract (EEE). ExcretoySecretory Antigen Fully embryonated Toxocaru canis eggs were hatched and placed in in vitro culture in Eagle s minimal essential medium containing Hanks salts according to the method of DE SAVIGNY (1975) and DE SAVIGNY & TIZZARD (1977). The culture fluid was replenished weekly and the expended fluid containing excretory and secretory products was passed through an 0.45 pm Amicon millipore filter (Amicon Ltd., High Wycombe, Bucks, England) to exclude any accidentally transferred larvae. This exhausted culture fluid was concentrated X 10 in an Amicon stirred cell using an UM2 Ultra Filter (exclusion range 2,000 daltons). This was designated T. canis excretorysecretory (ES) antigen. All antigen preparations were stored either at -70 C or in liquid nitrogen until used. Antiserum Antiserum to T. cunis ES antigen (1 mg/ml) was prepared by emulsifying T. cunis ES in Freund s complete adjuvant. The emulsion was injected into multiple subcutaneous sites in the interscapular area of rabbits. Rabbits were challenged with T. cunis ES alone one month after the initial injection, and bled nine to 14 days later. Indirect Fluorescent Antibody Test (IFAT) Frozen sections of.both T. cunis and Toxuscuris leoninu infective eggs were cut at 4 pm. They were air dried, and fixed in absolute methanol for 30 min. Serial dilutions of test sera were incubated with the sections in a humid chamber for 30 min at R.T. and the sections washed in three changes of PBS ph 7.1. FITC-conjugated sheep anti-rabbit immunoglobulin serum (Institut Pasteur, Paris, France) titred to its end point, was incubated on the sections for 30 min at R.T. and the slides washed again in three changes of PBS ph 7.1. Known positive and negative controls were run in conjunction with each test. Slides were viewed on a Leitz Ortholux 11 equipped with a Plijempak epi-illuminator and a GG 475 filter. Photographs were taken on colour reversal film ASA 200 (GAF Ltd., Colnbrook, Slough, England). Enzyme Linked Immunosorbent Assay (ELBA) The enzyme linked immunosorbent assay method used was as described by VOLLER et al. (1976). Antigen was optimally diluted (8 ug/ml for EEE antigen and 7.9 pg/ml for ES antigen in carbonate buffer ph 9.6) and 40 ~1 of antigen were coated on to wells of disposable polystyrene plates for three hours at 37 C. The plates were washed in phosphate buffered saline containing 0.5% Tween 20 (PBS- Tween) ph 7.4 to remove any unadsorbed antigen. Replicate test sera were diluted in PBS-Tween by doubling dilutions from 1:lOO to 1:204,800 and 50 ~1 added to each well, and incubated for three hours at room temperature. Wells were washed with PBS- RE SPONSE To T. CUniS IN RODENTS Tween to remove any unreacted serum proteins, and bound antibody was assayed by adding 50 ul of alkaline phosphatase labelled anti-rabbit immunoglobulin at 10 C for 18 hours. After six washes in PBS-Tween, 50 ~1 of p-nitrophenyl phosphate (1 mg/ ml) were added. The reaction was stopped after a set period by addition of 50 ul of 1M NaOH and the enzyme product assessed visually and spectrophotometrically at 405 mm, and Em5 over 0.5 were regarded as positive. 7-6- 5-4 - 3 - I I I I tttit Fig. 1. Fluorescent antibody times of rabbits infected with either Toxocara canis (-) or Toxascaris leonina (- - - -) infective eggs using frozen sections of Toxocara canis eggs as substrate. Arrows indicate days of infection. 8 - / / \ \ ---- I / / :....... _. /I..,,,......Y I I I I I I Fig. 2 Fluorescent antibody titres of rabbits infected with either Toxnscati leonina (- -) or Toxocara canti (...) infective eggs using frozen sections of Toxoscaris leonina eggs as substrate.

H. v. SMITH et al. Results Indirect Fluorescent Antibody T&t (IFAT) In order to determine the components against which circulating antibodies were produced, frozen sections of fully developed eggs of Toxocara canis and Toxascaris leonina were used. In these sections, four distinct components of the infective eggs were detailed. These were: (a) the fluid surrounding the larva, contained within the eggshell; (b) the eggshell; (c) the larval cuticle; (d) the larval tissues, excluding the cuticle. Both Toxocara canis and Toxascaris leonina embryonated eggs were sectioned in order to compare the degree of cross reactivity encountered in both i I c: A TlTRE 1:10 1:lO 1:20 1:lO 1:20 4"s POST 1FECrlOll o 7 14 21 28 35 42 TOXOCARA CANIS INFECTED RABBITS Toxocara canis and Toxascaris leonina experimental infections. Serum from rabbits infected with Toxocara canis eggs, when reacted against sectioned T. canis eggs produced detectable fluorescent antibody titres seven days after the initial dose of 5,000 infective eggs (Fig. 1). By day 14, titres had risen to I:80 (log2 6.34) and on days 21 and 28 titres were at 1:160 (log2 7.35) but by day 35 they had fallen to I:80 (log, 6.34). Cross-reacting antibodies against Toxascaris Zeonina infective egg components were first detected on day 14 (Fig. 1) at a titre of 1:lO (log1 3.33) and fluctuated to 1:20 (log2 4.33) until day 42. Sera from rabbits infected with T. leonina eggs when tested against sectioned T. leonina eggs produced detectable fluores- A TITRE 1:10 1:10 1:20 1:20 1:lO ws POST HFECTlO" TOXASCARIS LEONINA INFECTED RABBITS Fig. 3. The reactivity of compartments of the fully embryonated eggs of Toxocara cant and lbxa~caris leatna as shown by fluorescence when reacted with antisera from rabbits. Each bar represents the degree of reactivity of one rabbit as 113,213 or 313. The compartments are A-the fluid surrounding the larva within the eggshell, B-the eggshell, C-the cuticle of the 2nd stage larva and D-the larval tissues excluding the cuticle. Titre represents maximum mean titre of rabbits on days indicated, derived from Figs. 1 and 2.

92 DEVELOPMENT OF SEROLOGICAL cent antibody titres seven days after infection (Fig. 2). The titres increased to 1:80 (log, 6.34) by day 21 and then dropped to 1:40 (logr 5.34). Cross-reacting antibodies to sectioned Toxocara canis eggs were first detected on day 14 reaching a peak titre of 1:20 (log2 4-33) on days 28 and 35. Analyses of the egg components against which the sera reacted are summarized in Fig. 3. Sera from rabbits infected with T. canis eggs when tested against sections of T. canis eggs reacted mainly to the fluid surrounding the larva and weakly to the inside of the eggshell with day 7 serum. The reaction to the fluid surrounding the larva increased until day 2 1. At this time weaker reactions to the eggshell were also noticed and by day 28 all the components of the infective egg were fluorescing equally with sera at a titre of 1: 160. Using sectioned Toxascaris Zeonina eggs and rabbit antisera produced in response to a Toxocara canis infection, cross-reacting antibodies first detected on day 14 were directed against the fluid surrounding the larva. By day 21 cross-reacting antibodies to the egg shell and larval tissues were present, and these persisted at low titres until day 42. Antisera derived from a Toxascaris leonina infection and reacted with Toxocara canis sections revealed antibodies which were mainly cross-reactive to the fluid surrounding the larva on day 14, but by day 28 all the components were fluorescing up to a titre of 1:20 (log2 4.33). Pretreatment of both T. canis and Toxascaris leonina infective egg sections with antiserum to Toxocara canis ES antigen reduced the fluorescence of the fluid surrounding the larva and the eggshell when tested with sera derived from both T. canis and Toxascaris leonina-infected rabbits, indicating that this extraembryonic fluid within the eggshell is excretorysecretory in nature. Species specificity of this response was determined by the extinction of fluorescence of the ES component above a titre of 1:20. Enzyme Linked Immunosorbent Assay (ELISA) In sera from rabbits infected with Toxocara canis infective eggs and tested against T. canis ES antigen, circulating antibody was first detected at 15300 on day seven, rose sharply to a maximum of 1:6,400 on day 21 and remained constant thereafter at a titre of 1:3,200 (Fig. 4). Rabbits infected with Toxascaris Zeonina infective eggs showed some cross reactivity but the titre did not exceed 1:800 throughout the course of the experiment. Using Toxocara canis embryonated egg extract as antigen, rabbits infected with T. canis infective eggs demonstrated a low level of circulating antibody early in the infection (1:200 to 1:800), but by day 35 the titre had increased to 1:6,400 at which level it remained until the termination of the experiment (Fig. 5). Sera from rabbits infected with Toxascaris leonina eggs when tested against Toxocara canis EEE antigen, showed a low level of circulating antibody (1:800) with the exception of day 21 when the titre was 1:3,200 (Fig. 4). Discussion Tissue migratory nematode parasites stimulate the host to produce circulating antibodies to both excretory-secretory antigens and somatic antigens during the course of an infection. CATTY (1969) demonstrated that the first mosaic of antigens to which RESPONSE TO T. CfZniS IN RODENTS Fig. 4. Mean circulating antibody titres (ELISA) of rabbits infected with either Toxocara canis (-) or Toxascaris leonina (- - - -) infective eggs using Toxocara canti ES antigen. Fig. 5. Me& circulating antibody &es (ELBA) of rabbits infected with either Toxocara canis (-) or Toxascnris leaim~ (- - - -) infective eggs using Toxocara canis EEE as antigen. guinea-pigs infected with Trichinella spiralis responded are excretory-secretory in nature. Numerous workers have demonstrated circulating antibodies to somatic components of parasites, and SMITH & TONKIN (1979) have demonstrated a degree of somatic antigen stage specificity of circulating antibodies in response to the larval development stages of Hyostrongylus rubidus in pigs. As natural infections with

H. V. SMITH et al. 93 Toxocara canis and Toxascaris leonina in paratenic hosts present such animals with a range of eggshell, ES and somatic antigens derived from established larvae and those which die during or immediately after establishment, the use of embryonated eggs should detect a wide range of antibodies produced. Using the IFAT on frozen sections of both Toxocara canis and Toxascaris leonina infective eggs a distinct compartmentalization and development of the early circulating antibody response could be seen in the homologous systems. Initially, the early antibody response was directed against the fluid surrounding the infective larva in both Toxocara and Toxascaris infections (Fig. 3). This reaction was maximal on day 21 post-infection, but by days 35 and 42 postinfection somatic components were fluorescing as strongly, although by this time the over-all titres were declining. The cross reactions in the heterologous systems which were observed on day 14 post-infection in both systems again were initially directed against the fluid surrounding the infective larva indicating an early identification of the excretory-secretory component. Pretreatment of both Toxocara canis and Toxascaris sections with Toxocara canis ES antiserum reduced the fluorescence of the fluid surrounding the infective larva confirming its excretory-secretory nature. Progressively from day 14, cross reactions to all the components of the fully developed eggs in both heterologous systems were detectable at low titres. When T. canis ES antigen and T. canis embryonated egg extract were used as antigens in the ELISA, a difference in the time scale for peak detection of circulating antibodies occurred (Figs. 4 and 5). With ES antigen a titre of 1:800 was detected on day 7, peaked on day 21 at 1:6,400 and remained at 1:3,200 thereafter, a similar early response to the ES components to that observed in IFAT. Cross reactions to Toxascaris leonina infections occurred but were present at lower levels (1:800), a situation which was-also encountered in the IFAT (Fig. 1). The use of Toxocara canis embryonated egg extract in the ELISA (Fig. 5) showed only low levels of circulating antibodies in T. canis-infected rabbits UD to dav 28 (1:2,000), but after this titres rose * dramatically (1:6,400 on days 35 and 42). When T. canis embryonated egg extract was used, a greater degree of cross reactivity occurred when sera from Toxascati leoninainfected rabbits were tested (maximum titre 1:3,200 on day 21), a situation which also occurred in the IFAT. Analysis of Toxocara canis embryonated egg extract has shown a variety of cross-reacting somatic components to Toxascaris leonina, Ascaris lumbricoides and A. suum embryonated egg extracts and that the majority of these components in Toxocara canis embrvonated egg extract are not recognized bv a T. canis BS antiserum (Smith et al. unpublished observations). The seauential detection of circulating antibodies using both IFAT and ELISA may reflec i upon the timing of antigen presentation either ES or somatic, to the host. As can be seen from Figs. 4 and 5, the peaks of circulating antibodies detected in response to either ES or EEE antigen are about two weeks apart. Following artificial hatching, second-stage T. canis larvae do produce large amounts of ES antigen for the first seven to 14 days, after which time the amount decreases, and it might be that this early increase in anti-es antibody followed by its rapid decline is due to intial high release of ES antigen in viva, followed by a subsequent decrease. The delay in detecting high levels of circulating antibodies to T. canis embryonated egg extract on ELISA in rabbits infected with T. canis could be due to a slower release of somatic material later on in the infection, either due to a chronic exposure to the host whilst migrating through the tissues, or due to exposure during larval entrapment. Thus, using T. canis embtyonated egg extract a greater degree of cross reactivity occurs-with sera from rabbits infected with Toxascaris leonina than when using Toxocara canis ES, although cross reactions do occur using the latter (Fig. 4). This raises the auestion of the value of T. canis ES antigen in s erodiagnosis of VLM when not only T. canis but also Toxascaris leonina might be implicated. It is known that both Toxocara canis and Toxascaris leonina occur as infective eggs in the environment (QUINN et al., 1980) and nreliminarv exoeriments in rabbits have demonstrated that T. -leonina can invade through the small intestine and migrate into various tissues, inducing VLM-like symptoms. Acknowledgements This work was suooorted bv M.R.C. Grant Number 975/259/T. We w&h to thank Dr. J. R. Kusel and Dr. S. H. Bartelmez for constructive criticism and discussions, and Professor D. R. Newth for facilities. References Aljeboori, T. I. & Ivey, M. H. (1970). An improved hemagglutination technique for detecting antibody against Toxocara canis. American 7ournal of Trob i&l Medicine and Hygiene, 19, h-248. a mr Beaver, P. C., Snyder, C. H., Carrera, G. M., Dent, J. H. & Lafferty, J. W. (1952). Chronic eosinophilia due to visceral larva migrans. Pediatrics, 9, 7-19. Catty, D. (1969). The immunology of nematode infections, trichinosis in guinea pigs as a model. In: Monographs in Allergy. Vol. 5. Kallos, P., Hasek, M., Inderbitzin, T. M., Miescher, P. A. & Waksman, B. H. (Editors). Base1 and New York: S. Karger, pp. 87-102. de Savigny, D. H. (1975). In vitro maintenance of Toxocara canis larvae and a simple method for the production of Toxocara ES antigen for use in serodiagnostic tests for visceral larva migrans. Journal of Parasitology, 61, 781-782. de Savigny? D. H. & Tizzard, I. R. (1977). Toxocara larva rmgrans: the use of larval secretory antigens in haemagglutination and soluble antigen fluorescent antibody tests. Transactions of the Royal Society of Tropical Medicine and Hygiene, 71, 501-507. Hogarth-Scott, R. S. (1966). Visceral larva migrans: an immunofluorescent examination of rabbit and human sera for antibodies to the ES antigens of the second stage larvae of Toxocara canis, Toiocara cati and Toxascaris leonina (Nematoda). Immunology, 10, 217-223. Mitchell, J. R. (1964). Detection of Toxocara canis antibodies with the fluorescent antibodv technique. Proceedings of the Society for Ex&rimental Biology and Medicine, 117, 267-270.

94 DEVELOPMENT OF SEROLOGICAL RESPONSE TO T. CaniS IN RODENTS Quinn, R., Smith, H. V., Bruce, R. G. 81 Girdwood, R. W. A. (1980). Studies on the incidence of Toxocara and Toxascatis spp. ova in the environment: (1) A comparison of flotation procedures for recovering Toxocara spp. ova from soil. Journal of Hygiene, 84, 83-89. Smith, H. V. & Tonkin, C. H. (1979). Antigens of Hyostrongylus rubidus the red stomach worm of pigs. Analyses by means of passive haemagglutina- tion and unmediate hypersensitivity tests. Research in Veter-inay Science, 26, 302-307. Voller, A., Bidwell, D E. & Bartlett., A. (1976). Enzyme immunoassays for parasite diseases. Transactions of the Royal Society of Tropical Medicine and Hygiene, 70, 98-106. Accepted for publication 9th June, 1981.