EXPERIMENTAL

PARASITOLOGY

75,

329-339 (1992)

Schistosoma mansoni: Control of Hepatotoxicity and Egg Excretion by Immune Serum in Infected lmmunosuppressed Mice Is Schistosome Species-Specific, but Not S. mansoni Strain-Specific HCIPEWELL

M. MURARE, DAVID

School of Bioiogical

W. DUNNE, JEAN BAIN, AND MICHAEL J. DOENHOFF

Sciences, University College of North Wales, Bangor, Gwynedd, U.K. LW7 2UW

MURARE, H. M., DUNNE, D. W., BAIN, J., AND DOENHOFF, M.J. 1992. Schisrosoma mansoni: Control of hepatotoxicity and egg excretion by immune serum in infected immunosuppressed mice is schistosome species-specific, but not S. mansoni strain-specific. Experimental Parasitology 75, 329-339. lmmunosuppressed mice infected with Schistosoma mansoni suffer from an acute hepatotoxicity reaction, and they fail to excrete as many parasite eggs as comparably infected immunologically intact control animals. The hepatotoxicity was shown here to be preventable, and egg excretion rates were enhanced, by transfer of serum from donors with chronic S. naansoni infections, but not by serum from donors with heterologous infections of Schistosoma haematobium, Schistosoma bovis, or Schistosoma japonicum. The effects of the transferred sera are considered to be due to specific antibody, but the possibility of cytokine involvement is discussed. A high degree of serological cross-reactivity was found between sera from mice infected with the different schistosome species and unfractionated egg homogenate (SEA) in ELBA. Cross-reactivity of the heterologous sera was, however, reduced against CEF6, a partially purified fraction of S. mansard eggs that contains the putative hepatotoxin and has serodiagnostic potential, S. mansoni isolates from Puerto Rico, Brazil, Egypt, and Kenya shared similar characteristics with respect to the immune dependence of egg excretion and hepatotoxicity in immunosuppressed mice. The S. mansoni geographic isolates were also indistinguishable serologically, in terms of both the capacity of respective infection sera to neutralize hepatotoxicity and in their capacity to promote egg excretion of the other isolates in vivo. Complete immunological cross-reactivity of the geographically distinct isolates was also observed in ELBA with both CEF6 and SEA. Utilization of CEF6 for serodiagnosis of schistosomiasis mansoni is therefore unlikely to be restricted by geographical considerations. @ 1992 Academic Press, Inc. INDEX DESCRIPTORS AND ABBREVIATIONS: Schistosoma mansoni; Trematode; Hepatotoxicity; Serodiagnosis; Egg excretion; Passive serum transfer; Enzyme-linked immunosorbent assay (ELBA); Soluble fraction of schistosome egg homogenates (SEA); Cation exchange fraction 6 (CEF6); Glufamic oxalate transaminase (GOT); Ethylene diamine tetracetic acid (EDTA).

cally intact controls (Doenhoff et al. 1978a; Dunne et al. 1983). Immunosuppressed mice with heavy, reWhen immune serum taken from immucently patent Schistosoma mansoni infec- nologically intact mice with light chronic tions suffer from an acute hepatotoxicity homologous infections was injected into the reaction, evidence for which derives from infected, immunosuppressed hosts the hepboth histopatholo~c~ examination and an atotoxicity reaction was completely preelevation of circulating transaminase levels vented, and egg excretion rates were par(Buchanan et al. 1973; Byram and Von tially restored (Doenhoff et al. 1981). Both Lichtenberg 1977; Byram et al. 1979; Lucas of the humoral immune effector activities et al. 1980; Doenhoff et al. 1981). Further- (i.e., prevention of hepatotoxicity and facilmore, these mice do not excrete as many itation of egg excretion) depended on the parasite eggs in their feces as immunologi- serum donor mice being sensitized to egg 329 0014.4894192 $5.00 Copyright

0 19% by Academic Press, Inc. All rights of reproduction in any form reserved.

330

MURARE

antigens, and the effects were found to be egg-stage specific (Doenhoff et al. 1981). The experiments reported here were performed to test in vivo the degree of schistosome species-specificity of these phenomena. This was done by assaying the capacity of passively transferred sera from donor mice infected with Schistosoma haematobium, Schisfosoma bovis, or Schistosoma japonicum to prevent S. mansoni egg-induced hepatocellular damage and to enhance S. mansoni egg excretion in heavily infected T-cell deprived mice. The opportunity was also taken to extend the work on S. mansoni, so far done solely on a Puerto Rican isolate, to encompass strains of the parasite isolated more recently from infected patients from Brazil, Egypt, and Kenya. This was to confirm that earlier observations were not spuriously dependent on use either of a particular geographic isolate of S. mansoni, or on one that had been maintained in the laboratory for many years. Antibodies in the sera that were transferred in vivo were assayed in ELISA using as antigen either crude egg homogenates of the different schistosome species, or CEF6, a fraction purified from S. mansoni eggs. CEF6 contains an antigen oi, previously identified as the putative hepatotoxin, together with another cationic antigen a1 (Dunne et al. 1981, 1991; Dunne and Doenhoff 1983). CEF6 has shown promise as an antigen preparation for serodiagnosis of schistosomiasis (Mclaren et al. 1981; Mott and Dixon 1982), and the sera derived from mice infected with the different geographic isolates of S. mansoni were used to test the extent to which CEF6 prepared from one geographic isolate might be used for diagnosis of the disease in another endemic area. MATERIALSANDMETHODS Parasite mansoni

species and strains. Four isolates of S. were used, all passaged in random bred TO

ET AL. strain mice as definitive host. An isolate from Puerto Rico was maintained in albino Biomphalaria glabrata snails from Puerto Rico. An isolate from Brazil was maintained in pigmented B. glabrata snails from the same country. Isolates of S. mansoni from Egypt and Kenya, respectively, were maintained in Biomphalaria alexandrina snails from Egypt. The period over which the Puerto Rican, Brazilian, and Egyptian isolates had been maintained in the laboratory prior to this work is not known, but on the best information available was considered to be over 20 years for the Puerto Rican parasite and approximately 5 years for each of the Brazilian and Egyptian isolates. The Kenyan isolate was used within 2 years of eggs being obtained from a human donor. S. japonicum was from Checkiang, Chinese mainland, and was maintained in TO mice and Oncomelania hupensis hupensis snails (Moloney and Webbe 1982). S. haematobium cercariae were obtained from Bulinus truncatus roh[fsi snails which had been collected in the Volta Lake region of Ghana, and infected with miracidia hatched from the urine of S. haemato&m-infected patients from the same area. The snails with prepatent infections were shipped by air to the U.K. and the cercariae were used after development of patency in the laboratory (Agnew et a/. 1988). S. bovis was passaged in the laboratory in Bulinus truncatus afiicanus snails taken originally from the HasaHeisa area of Sudan and in Mastomys natalensis multimammate rats (Murare and Doenhoff 1987). Mouse serum recipients and donors. Inbred CBAH/T6T6 mice were bred on site. Serum donors were infected when between 8 and 10 weeks old. Recipients of serum were deprived of their T-cells when 8-10 weeks old by a combination of adult thymectomy and subcutaneous injections of rabbit anti-mouse thymocyte serum as described previously (Doenhoff et al. 1979). The deprived mice were rested for approximately 30 days before infection. Infections

and parasitology.

S. mansoni,

S. bovis,

and S. haematobium cercariae were applied percutaneously to the shaven abdomens of mice (&r&hers and Terry 1965). S. japonicum cercariae were injected intraperitoneally as described by Moloney and Webbe (1982). CBA-H/T6T6 mice which acted as serum donors were given 25 S. mansoni, 200 S. haematobium, 100 S. bovis, or 20 S. japonicum cercariae. (Experience indicated that these respective numbers of cercariae resulted in chronic infections of approximately equivalent intensity for all four schistosome species.) The infections were allowed to progress for a minimum of 12 weeks before weekly serial bleeding commenced to build up pools of chronic infection serum (CIS) for transfer to S. mansoni-infected deprived recipients. Only mice with confined patent infections were used as serum donors, the confirmation being through detection of anti-egg antigen precipitating an-

s. m~nsofZi:

HEPATOTOXICITYAND

tibody in immunodiffusion in gel, high antibody activity against homologous egg antigen in ELISA (see below), or observation of obvious granulomatous intlammatory activity in the liver at laparotomy. The serum pools were stored at -20°C until used. About 15 weeks after infection the blood of a representative number of serum donors of each type was sampled for ELISA reactivity against homologous and heterologous antigens. T-cell deprived mice due to receive immune sera were infected with 150-200 cercariae of a specified S. mansoni isolate. On the day of termination of the experiments a 30- to 50-mg fecal pellet was taken from each of the serumrecipient and control mice for processing and ninhydrin staining to display excreted eggs, as previously described (Bell 1%3; Doenhoff et al. 1978a). Worm burdens were estimated by portal perfusion (Smithers and Terry 1965’; Doenhoff et al. 1978b). The ventral median lobe of the liver was removed for histological examination. The number of S. mansoni eggs in the remainder of the liver and the intestines (from duodenum to rectum) of perfused mice was estimated after alkali digestion as described (Cheever 1968; Doenhoff et al. 1978a). Histopathology. After form01 saline fixation, wax embedding and 5u sectioning, the livers from deprived mice in experiments in which heterologous sera had been transferred (Table I) were stained with haematoxylin and eosin. The degree of microvesicular damage in each liver was scored as described previously (Doenhoff et al. 1981). (Results from livers of control immunologically intact mice in these experiments are not presented, since in earlier experiments only mice with the heaviest S. mansoni infections were found to have microvesiculated hepatocytes, and none were seen in the present experiments.) Scores of damage in the deprived mice ranged from 3 + (every hepatocyte showing evidence of egg-induced microvesicular damage) to 0 (no damaged cells). Serum transaminases and ELISA. On the day of perfusion of the serum recipients and controls a sample of blood was taken into EDTA for estimation of glutamic oxalate transaminase (GOT) as described (Karmen 1955; Doenhoff et al. 1979). S. mansoni eggs were isolated from the tissues of heavily infected mice, and crude saline-solubie egg antigen extracts (SEA) were isolated as described (Doenhoff et al. 1981; Dunne et al. 1981). Purified fraction CEF6, which contains two glycosylated cationic antigens, o1 and ai of 31 and 41 kDa, respectively (Dunne et al. 1991). was obtained in a one-step cation exchange chromatography procedure as described previously (Dunne ef al. 1984). Eggs for preparation of SEA of other schistosome species were obtained from infected mice or other life-cycle passage animals by adaptation of methods used for S. mansoni eggs. Protein concentrations in antigen preparations used for

EGG EXCRETION

331

ELISA were estimated as described by Lowry et al. (1951). ELISA was performed as described previously (McLaren et al. 1981). Crude egg antigens were used to coat the microtitration plates at a concentration of 10 &ml, and CEF6 at a concentration of 1 )~g/ml. Mouse sera to be tested in ELISA were diluted 1:200 before use. All reactions were terminated after 15 min by addition of 25 )*I H$O,. Statistics. Results are given as group mean values + 1 standard deviation. The degree of significance of the differences between experimental groups was estimated using Student’s t test, with P > 0.05 considered not significant. Experimental design. T-cell-deprived mice due to receive serum were infected with S. mansoni cercariae on Day 0. Age- and sex-matched immunologically intact mice were included as a control to confirm efftcacy of T-cell deprivation, and a group of deprived mice which was infected, but received no further treatment, served as untreated controls. From Day 40 after infection, the deprived mice received daily intraperitoneal injections of 0.5 ml heterologous or homologous chronic infection sera (CIS). Experiments were terminated between Days 44 and 46 after infection, depending on the degree of morbidity of the infected, deprived, but otherwise untreated, control mice, and during which time fecal pellets were taken for egg counts. Before autopsy the different groups were bled for estimations of GOT, and after lethal anesthesia the mice were perfused via the portal system for adult worm counting. Tissues were removed for histological examination and tissue egg counts. RESULTS

Table I presents the results of four replicate experiments in which CIS pools obtained from mice chronically infected with S. mansoni, S. haematobium, S. japonicum, or S. bowis were transferred to T-celldeprived mice with heavy S. mansoni infec-

tions. Heterologous infection sera of each species were transferred in two separate experiments, and in each experiment immunologically intact infected mice and deprived mice reconstituted with S. mansoni infection serum served as positive controls, and unreconstituted deprived mice were negative controls. The results in Table I can be summarized as follows: (i) With one exception there were no significant differences in S. mansoni worm

42.0 47.3 45.1 55.0 52.5 44.3 36.9 41.1 37.3 39.3 26.3 40.5 33.3 30.5

NMS S. bovis S. haem. S. mans.

S. jap. S. mans.

S. bovis S. mans.

8 7 7 7

10 8 8 9

29.8 37.1 37.6 34.4 36.5

” ? f 2

+ f k f

+ 2 + 2 2 2

f 2 ” + ”

D&S

5.9 8.5 8.8 9.8

8.8 8.9 6.6 3.8

6.0 8.1 4.3 10.2 8.3 17.2

11.7 6.9 6.4 6.4 9.8

No. of worm

6 6 7 6 6 6

S. jap. S. haem. S. mans.

-

Donor serum

19.9 25.4 18.9 20.1

21.5 24.7 20.9 27.1

32.7 29.7 27.2 32.8 35.2 37.6

23.0 30.5 27.8 27.9 31.4

+ f f f

2 + + 2

2 ” 2 f f 2

+ 2 2 f f

4.8 7.8 5.6 4.7

4.4 5.2 5.0 4.7

7.6 5.5 3.8 10.7 9.4 1.3

10.9 6.9 4.2 4.8 9.8

No. of eggs in liver X lo3

412 1899 2009 224

516 1456 2410 229

475 1464 1394 665 1330 238 177 672 1360 74

241 1011 1106 315 62 52

-1- 121 + 1112 f 1447 + 72

+ + 2 -c

k f + 2 2 -t

289 2 76 1377 + 866 1100 2 525 1164+909 339 + 158

GOT units/ml

I

2.1 2.4 0.4 2.0 1.9 0.3

1.8 2.0 1.5 2.2 0.2

2.2 2.1 2.1 0.2

-

2.2 2.2 0.2

Microvesicular damage

54.7 54.8 50.2 38.9

52.7 50.4 50.8 55.6

61.9 46.0 49.9 54.0 63.5 57.4

53.7 56.1 64.6 54.6 60.3

f ” f t

f 2 + +

+ -t + 2 5 +

* + f + i

15.4 18.9 19.1 8.3

18.5 6.2 12.0 6.2

14.2 6.6 6.7 7.5 14.3 19.3

20. 9.6 7.5 13.1 20.4

No. of eggs in gut X lo3

-

(46) 215 + 148 26 + 36 18 2 32 33 + 29 168 k 115 (45) 534 + 99 222 927 9+7 6+3 145 2 118

(45) 311 + 184 13 + 5 18 + 14 98 + 51

(47) 540 f 210 14 + 13 36 2 29 115 + 42

(46) 550 2 346 6?9 721 165 + 62

(45) 124 f 84 18 -+ 16 17 2 20 14 2 14 122 -t 74 (42) 52 * 38 120 2+2 0 3+3 35 * 40

No. of eggs/l00 feces

Note. Normal and deprived mice were each infected with 200 S. mansoni cercariae of a Puerto Rican isolate. From Day 40 after infection deprived recipients of serum were each given six daily intraperitoneal injections of 0.5 ml serum from pools prepared from normal donors infected with S. bovis, S. haematobium (S. haem), or S. japonicum (S. jap). Fecal egg counts were estimated on the day after infection indicated in the brackets. Circulating GOT concentrations were estimated on a blood sample taken on Day 46. Mice were perfused for worm counts on Day 46. In experiment 3 microvesicular damage in histopathological sections of liver was estimated in duplicate by two observers working independently of each other. Results are given as the group mean values +/- 1 standard deviation, apart from the microvesicular damage readings which are the group mean alone.

Normal Deprived Deprived Deprived Deprived Deprived (3) Normal Deprived Deprived Deprived (4) Normal Deprived Deprived Deprived

(2)

Normal Deprived Deprived Deprived Deprived

6 7 6 8 7

JZOUD

Mice

(1)

No. in

(Expt)

TABLE

F

2

Ii

5

s.

mansoni:

HEPATOTOXICITY

burdens or tissue egg burdens between the different groups in each of the four respective experiments. The exception was in experiment 4, where the number of worm pairs in the normal controls was less than that in the deprived controls (P < 0.001). (ii) Unreconstituted control deprived mice had higher GOT levels than the normal controls, Except for experiment 2 the differences were significant (P < 0.01). (iii) Deprived mice reconstituted with S. mansoni CIS had significantly lower GOT concentrations than unreconstituted deprived mice (P < 0.05 in all experiments). (iv) GOT concentrations in deprived mice reconstituted with CIS from mice infected with S. bovis, S. haematobium, or S. japonicum were not significantly different from GOT levels in unreconstituted deprived mice. (v) Histological evidence of hepatocyte microvesicular damage reflected GOT concentrations in so far as deprived mice reconstituted with S. mansoni CIS showed least evidence of hepatocyte damage. In contrast most of the livers of mice receiving heterologous CIS appeared as damaged histologically as those of unreconstituted deprived mice. (vi) With one apparent exception, normal mice and deprived mice reconstituted with S. mansoni CIS had significantly greater fecal egg counts than the respective unreconstituted deprived controls (P < 0.05). The exception to this was the Day 42 fecal egg count of experiment 2 in which the group of deprived mice given S. mansoni CIS included a mouse with a count of 112 eggs/100 mg feces. This gave rise to a standard deviation that was greater than the group mean value (i.e., 35 4 40). However, exclusion of this animal from the calculations gave a Day 42 group mean of 19 2 13.2 eggs/100 mg, which was significantly greater than the mean values for other groups of deprived mice in this experiment (P < 0.01). (vii) Deprived mice injected with CIS

AND

EGG

EXCRETION

333

from heterologous species had fecal egg counts which were not significantly different from those of unreconstituted deprived mice. The conclusions from the results in Table I are, therefore, first, that hepatocyte damage suffered by S. mansoni-infected deprived mice was effectively prevented only by serum from homologously infected donors, and second, that the reconstitution of egg excretion rates was also achieved with S. mansoni infection serum alone. Figure 1 shows the results of testing sera from control and variously infected mice in ELISA with homologous and heterologous egg antigens. (The samples for ELBA were from the same mice that had acted as donors for the reconstitution experiments in Table I.) Antibody activity in ELISA was in general greatest against egg antigens of the infecting species, but there was considerable cross-reactivity with egg antigens of heterologous species, as indicated by all four sets of infection sera giving higher ODAg2 values against the four SEAS than normal mouse sera (P < 0.001). Nevertheless, reactivities against the SEAS of S. mansoni and S. bovis were to some extent specific in as much as the respective homologous sera gave significantly greater ELISA reactivity than the three sets of heterologous infection sera (P < 0.0001). Paradoxically, S. mansoni, S. haematobium, and S. bovis sera were relatively highly cross-reactive with S. japonicum SEA, while the S. japonicum infection sera were least cross-reactive with other species of SEA. When S. mansoni egg antigen fraction CEF6 was used as the antigen in ELISA the reactivity of S. mansoni infection sera remained at the same level (in terms of group mean OD,g, values) as against the crude S. mansoni egg preparation. The reactivity of the sera from mice with heterologous infections against CEF6 was, however, reduced relative to their reactivity with S. mansoni SEA, but the latter were still significantly

334

MURARE

ET AL.

1.50 1.25 z

1 .oo

c

2

0.75

: :

0.50 0.25 0.00

CEF6 SmSEA ShSEA SbSEA SjSEA FIG. 1. Sera from mice infected with Schistosoma mansoni (H), S. hnematobium (H), S. bovis (N), S. japonicum @I), and from uninfected normal mice (Cl) were tested individually in ELISA against crude SEA antigen preparations from each of the different species or antigen fraction CEF6 prepared from S. mansoni SEA (Puerto Rican isolate). Results are the mean from 12 or more samples, and variability about the group mean OD4, absorbance values is indicated by + 1 standard deviation.

more reactive against CEF6 than normal mouse sera (P < 0.01). Table II gives the results of experiments designed to examine whether geographic isolates of S. mansoni could be distinguished with respect to hepatotoxicity and immune-dependent egg excretion. As with the experiments in Table I, in all four experiments in Table II there were few differences in parasitogical parameters (worm and egg counts) between the normal mice and the various groups of deprived mice. One exception was the group of deprived mice given Egyptian CIS in experiment 3, which had significantly fewer gut eggs than the normal controls (P < 0.005). The unreconstituted infected deprived mice in all four experiments had significantly elevated circulating GOT concentrations (P < O.Ol), indicating that the capacity to induce a hepatotoxicity reaction in immunosuppressed murine hosts was common to different geographic isolates of S. mansoni. Furthermore, with respect at least to mice infected with isolates from Puerto Rico, Brazil, and Egypt (experiments 1, 2, and 3, Table II) sera from do-

nors infected with the same or a different isolate were approximately equivalent in their capacity significantly to reduce GOT concentrations to levels below those found in unreconstituted deprived hosts (P < 0.05). With respect to fecal egg counts, deprived mice infected with the same three isolates had significantly reduced fecal egg counts compared with comparably infected intact normal controls (P < 0.05). The sera from donors infected with the different isolates were not obviously distinguishable with respect to their ability to reconstitute egg excretion rates. The group of unreconstituted T-celldeprived mice infected with the Kenyan isolate of S. mansoni (Table III, experiment 4) also had higher mean GOT concentrations (P < 0.001) and lower mean fecal egg counts (P = 0.067, not significant) than the intact controls. These results were therefore consistent with the preceding three experiments in Table III. Transfer of infection sera to these animals did not, however, have as great an effect in ameliorating hepatotoxicity (by reducing GOT levels) as in

s. mansani:

HEPATOTOXICITY TABLE

Expt Mice 1 Nor W DeP W Dep 2 Nor Dep DeP Dep DeP Dep 3 Nor Dep Dep Dep Dep Dep 4 Nor Dep W Dep Dep Dep

No. of worm n

Serum

pkS

335

AND EGG EXCRETION II

No. of liver eggs X IO3

SGOT

units/ml

No. of eggs/ 100 mg feces

5

-

5 5 6 5

PR Br Eg

41.6 37.4 36.8 48.3 27.8

IL 2 f f +

9.7 12.1 13.2 13.6 13.6

34.7 28.1 35.0 35.7 20.9

+- 12.3 t 10.6 t 9.6 +- 8.7 -t- 5.2

303 1979 143 369 150

2 + * f f

5 5 5 5 5 5

PR Br Eis Ke

48.0 45.2 36.3 49.0 so.4 51.6

+ + 2 + 2 k

11.5 9.4 14.2 16.7 8.6 13.5

22.2 22.3 18.9 30.2 30.6 28.6

it t2 tt

987 3853 448 222 301 321

2 569 2 739 + 204 2 75 -r- 78 2 87

6 6 5 5 7 5

PR Br Eg Ke

15.2 49.7 43.0 48.8 39.0 44.8

+ 8.4 k 16.4 k 23.9 + 9.7 + 12.5 zk 15.1

8.6 15.0 19.1 22.7 20.2 22.4

+- 5.3 -c 7.4 t 5.8 2 3.5 +- 7.2 +- 7.9

316 1809 194 205 230 218

k f f f 2 t

95 1602 27 40 159 28

32.7 57.3 49.5 76.6 71.6 69.9

+ 24.4 t 20.5 -’ 22.8 -+ 17.2 2 19.4 5 21.6

60 k 223 66 2 57 * 302 39 f

70

6 6 6 6 6 6

PR Br Eg Ke

33.7 t 43.0 k SO.7 k 41.5 * 45.6 * 53.3 2

23.0 23.0 32.4 23.3 27.9 46.6

t 9.2 t 4.3 -1- 5.0 1- 4.8 t- 5.9 2 23.3

383 1908 636 397 440 940

I t f 2 k f

68 960 321 229 167 452

72.6 65.0 84.6 65.9 68.4 97.9

2 25.0 -t- 22.8 k 17.6 k 20.5 t 18.0 4 22.1

182 i 14 k 10 + 18 2 53 k 6a3

200 8 9 3 18

8.5 10.9 6.1 9.1 3.7 4.8

4.2 12.5 9.9 8.7 9.9 3.2

61 1212 30 447 54

No. of gut eggs X lo3 72.1 55.8 56.5 52.8 38.8

+ k ? + f

15.8 17.1 16.2 14.1 10.1

266 9 89 99 98

2 121 ?z 20 zk 46 2 103 2 76

109.5 109.7 76.5 98.3 99.1 96.1

2 k + f f t

17.4 52.9 31.1 18.8 15.6 20.0

552 2 281 72 11 81 2 60 38 +- 10 40 2 45 77 ‘- 45

32 22 16 18

Note. Mice were infected with 175 S. mansoni cercariae from Puerto Rico (Expt I), or with 150 cercariae of the isolates from Brazil (Expt 2), Egypt (Expt 3), or Kenya (Expt 4). Injections of infection sera were given as in the protocol for Table I on a daily basis until the day before perfusion, the sera being derived from groups of normal mouse donors infected with S. mansoni from Puerto Rico (PR), Brazil (Br), Egypt (Eg), or Kenya (Ke). Fecal egg counts were performed on Day 46 (Expts 1 and 2), Day 43 (Expt 3), or Day 45 (Expt 4). Mice were bled for GOT estimates and perfused for worm counts on Day 46 (1 and 2), Day 44 (3), or Day 45 (4). Results are given as the group mean values +/- 1 standard deviation.

the preceding three experiments. Thus, GOT levels in the serum-treated deprived mice in the fourth experiment were reduced by only 5&80% compared with unreconstituted deprived mice, and the reconstituted mouse groups had higher GOT concentrations than the normal controls. In contrast, in the preceding three experiments serum treatment reduced group mean GOT concentrations by 80% or more compared with unreconstituted deprived controls, and with two exceptions, mean GOT concentra-

tions in the reconstituted deprived groups were lower than those in normal controls. Furthermore, of the four sera transferred in experiment 4, only the pool from mice infected with the Egyptian isolate enhanced egg excretion rates significantly above the unreconstituted deprived mouse level (P < 0.005). Table III gives the results of testing the reactivity of sera from individual normal mice infected with the four S. munsoni isolates in ELISA with batches of CEF6 and

336

MURARE

ET AL.

TABLE

III

Origin of SEA Serum

8 week

16 week

Origin of CEF6

n

PR

Br

Eg

Ke

PR

Br

Q

Ke

PR

11

11

Ke

11

PR

10

Be

10

Eg

11

Ke

7

0.84 (0.10) 0.04 (0.02)

0.91 (0.10) (Et)

0.78 (0.10) 0.86 (0.12) 0.92 (0.12) 0.66 (0.16) 0.92 (0.11) 0.96 (0.09) 0.98 (0.W 0.94 (0.07) (22,

0.88 (0.13) 0.95 (0.10) 0.80 (0.12) 0.71 (0.14) 0.98 (0.09) 0.98 (0.14) 1.02 (0.08) 0.98 (0.07) 0.07 (0.03)

0.91 (0.12) 0.98 (0.08)

Eg

0.75 (0.12) 0.83 (0.14) 0.68 (0.11) 0.62 (0.14) 0.89 (0.13) 0.99 (0.12) 0.98 (0.05) 0.92 (0.08) (Z,

0.93 (0.15)

11

0.77 (0.11) 0.88 (0.12) 0.72 (0.12) 0.63 (0.14) 0.92 (0.11) 0.98 (0.10) (::Z)

0.89 (0.14)

Br

0.66 (0.10) 0.77 (0.10) 0.60 (0.13) 0.54 (0.13) 0.84 (0.10) 0.90 (0.14) (E,

NMS

10

(::Z) 0.74 (0.15) 0.65 (0.13) 0.98 (0.11) 0.97 (0.13) 1.01 (0.11) 0.89 (0.15) 0.05 (0.02)

(X) 0.73 (0.15) 0.98 (0.10) (Z) 1.03 (0.08) 0.97 (0.06) 0.08 (0.03)

(& 0.85 (0.12) 0.78 (0.11) 1.02 (0.12) 1.01 (0.07) (E) (Z) 0.10 (0.05)

Nore. Serum samples from mice infected with S. mansoni isolates from Puerto Rico (PR), Brazil (Br), Egypt (Eg), or Kenya (Ke), or from normal uninfected mice, were obtained 8 or 16 weeks after percutaneous administration of 25 cercariae, and tested individually in ELISA with batches of CEF6 or SEA prepared from eggs of each of the four geographic isolates. Results are given as the group mean OD,, value + /- 1 standard deviation.

SEA derived from the same isolates. The serum samples were from mice that had acted as donors of the serum pools used to reconstitute infected deprived mice in Table II. With respect to reactivity in ELISA, there was no significant distinction between the batches of CEF6 and SEA derived from the different geographic isolates of S. mansoni. The ELISA reactivities of the four batches of antisera were also similar, with the possible exception of the sera from mice with &week-old Kenya strain infections, which had somewhat (but not significantly) less intense antibody activity than the other isolates. DISCUSSION

The results indicate that there is a high degree of schistosome species-specificity in the capacity of sera from chronically infected mice to prevent egg-induced hepato-

toxicity and to reconstitute fecal egg counts in S. mansoni-infected T-cell-deprived mice. Thus, while homologous infection serum reduced both circulating transaminase levels and the degree of hepatocyte microvesiculation, and also caused egg excretion rates to increase, serum pools from mice infected with S. haematobium, S. bovis, or S. juponicum were ineffective on both counts. There was little difference between the four geographically distinct isolates of S. mansoni with respect to their capacity to induce the hepatotoxicity reaction in immunosuppressed hosts and the degree of immune dependence of the processes involved in excretion of their eggs. The reasons serum pools from mice infected with the Puerto Rican, Brazilian, or Egyptian isolates were relatively ineffective in preventing hepatotoxicity or enhancing egg excretion of the Kenyan isolate are not clear at present, particularly as serum from mice

,!?. mUnSOni:

HEPATOTOXICITY

infected with the Kenyan parasite was effective in these two respects in deprived mice infected with the other three S. mansoni isolates. The species-specificity of the prevention of hepatotoxicity by immune serum is consistent with the failure of T-cell-deprived mice infected with either S. haematobium or S. bovis to suffer from a hepatotoxicity reaction (Mm-are et al. 1987; Agnew et al. 1989). This was the case even when S. bovis-infected deprived mice had liver egg densities which were up to twofold more than those found in the S. mansoni-infected livers (Agnew et al. 1989). Absence of hepatotoxicity in deprived mice infected with S. bovis or S. haematobium may, however, be due to susceptibility to hepatotoxicity being a factor which is under host control (i.e., mouse liver cells may not be as susceptible to hepatotoxicity induced by these two species of schistosome as liver cells of their respective normal hosts.) The hepatotoxicity reaction has been shown to be S. mansoni egg stage-specific (Doenhoff et al. 1981). The hepatotoxic agent was identified as antigen oi by virtue of antisera with demonstrable antibody to this antigen being capable of protecting infected immunosuppressed mice against liver damage, while sera reactive with other S. mansoni egg antigens were ineffective (Dunne et al. 1981, 1991). Antigen oi is one of the constituents of CEF6 and the relative inactivity of sera from mice infected with the other schistosome species is consistent with oi being a S. mansoni-specific antigen. Antigen wi is a 31-kDa glycosylated cationic constituent of S. mansoni eggs (Dunne and Doenhoff 1983; Dunne et al. 1986; Dunne et al. 1991). Immunochemical characterization also showed it to be species-specific, but present in the eggs of S. mansoni isolates from different geographic areas (Dunne et al. 1984). Little is known about the degree of species-specificity or any other properties relating to the second antigen, a,, a 41-kDa glycosylated hetero-

AND

EGG

EXCRETION

337

dimer that is present in CEF6 (McLaren et al. 1981; Dunne et al. 1991). Fraction CEF6 is potentially useful for the serodiagnosis of human schistosomiasis (McLaren et al. 1981; Mott and Dixon 1982; Doenhoff et al. 1985). The results in Table III are of interest in this context in so far as they indicate that the geographic origin of the parasite isolate used to prepare antigens for serodiagnosis by ELISA seems of little consequence. The validity of this conclusion is now being more rigorously tested using human infection sera. The antigens involved in the immune dependence of egg excretion have still to be identified, but the evidence that immune reconstitution of S. mansoni egg excretion has a high degree of schistosome speciesspecificity is a useful step toward this goal. Tumor necrosis factor has recently been shown marginally to enhance the rate of S. mansoni egg excretion in mice with severe combined immunodeficiency (Amiri et al. 1992), and this and other cytokines could be partially responsible for the effects achieved here with transfers of homologous infection sera. Cytokines seem unlikely to be the sole initiators of the egg excretion activity in the transferred sera, however, since the egg-induced immunopathologies of S. bovis and S. haematobium are similar to that of S. mansoni in being T-cell dependent (Murare and Doenhoff, 1987; Agnew et al. 1988, 1989). S. haematobium and S. bovis infections seem therefore likely to generate the same cytokine profiles as S. mansoni, yet infection sera of the former two species did not reconstitute S. mansoni egg excretion rates. The fact that egg excretion rates can be enhanced by sera from mice previously immunized with S. mansoni egg antigens, but not worm or larval antigens (Doenhoff et al. 1981), also argues for a role for schistosome species- and egg stage-specific antibodies in the egg excretion process. Tumor necrosis factor and other cytokines may, of course, be generated as a consequence of antigen-antibody

MURARE

338

reactions after transfer to the serum reciw ients. The relative contributions of specific antibodies and cytokines to the activities of transferred sera deserve further analysis in experiments involving transfer of purified reagents. ACKNOWLEDGMENTS We are very grateful to Dr H. Bushara and Professors M. G. Taylor and M. Hussein who made the S. bovis isolate available, Drs K. Y. Chu and S. T. Wen and Prof. G. Webbe for S. haematobium, and Drs H. Schick, A. Butterworth, and J. Ouma, and Col. M. G. Radke for supplying, respectively, the Egyptian, Kenyan, and Brazilian isolates of S. mansoni. Drs. Anne Moloney and Alison Agnew kindly supplied S. japonicum and S. haematobium cercariae for infection of serum donors, and Padraic FalIon kindly helped with the statistical analysis. This work was-supported in part by the Wellcome Trust, the British Medical Research Council, and the British Council. REFERENCES A. M., LUCAS, S. B., AND DOENHOFF, M. J. 1988. The host-parasite relationship of Schistosoma haematobium in the mouse. Parasitology 97, 403424.

AGNEW,

AGNEW, A. M., DOENHOFF,

MURARE,

H. M.,

LUCAS,

S. B., AND

M. J. 1989. Schistosoma bovis as an immunological analogue of S. haematobium. Parasite Immunology 11, 329-340. AMIRI, P., LOCKSLEY, R. M., PARSLOW, T. G., SADICK, M., RECTOR, E., RITTER, D., AND McKERROW, J. H. 1992. Tumor necrosis factor a restores granulomas and induces parasite egg-laying in schistosome-infected SCID mice. Nature 356, 604607. BELL, D. R. 1%3. A new method for counting Schistosoma mansoni eggs in faeces: With special reference to therapeutic trials. Bulletin of the World Health Organization 29, 525-539. BUCHANAN, R. D., FINE, D. P., AND COLLEY, D. G. 1973. Schistosoma mansoni infection in mice de-

pleted of thymus-dependent lymphocytes. II. Pathology and altered pathogenesis. American Journal ofPathology 71, 207-218. BYRAM, J. E., DOENHOFF, M. J., MUSALLAM, R., BRINK, L. H., AND VON LICHTENBERG, F. 1979. Schistosoma mansoni infections in T-cell deprived mice, and the ameliorating effect of administering chronic infection serum. II Pathology. American Journal

of Tropical

Medicine

and Hygiene

28, 274-

285. BYRAM, J. E., AND VON LICHTENBERG, F. 1977. Altered schistosome granuloma formation in nude

ET AL. mice. American Journal Hygiene 26,944-956.

of Tropical

Medicine

and

CHEEVER, A. W. 1968. Conditions affecting the accuracy of potassium hydroxide digestion techniques for counting Schistosoma mansoni eggs in tissues. BulZetin

of the World

Health

Organization

39: 328

331. DOENHOFF, M., BICKLE, Q., LONG, E., BAIN, J., AND MCGREGOR, A. 1978b. Factors affecting the acquisition of resistance against Schistosoma mansoni in the mouse. I. Demonstration of resistance to reinfection using a model system that involves perfusion of mice within three weeks of challenge. Journal of Helminthology 52, 173-186. D~ENHOFF, M. J., DUNNE, D. W., BAIN, J., LILLYWHITE, J. E., AND MCLAREN, M. L. 1985. Serodiagnosis of mansonian schistosomiasis with CEF6, a cationic antigen fraction of Schistosoma mansoni eggs. Developments ?* I>.

in Biological

Standards

62, 63-

M., MUSALLAM, R., BAIN, J., AND McA. 1978a. Studies on the host-parasite relationship in Schistosoma mansoni-infected mice: The immunological dependence of parasite egg excretion. Immunology 35, 771-778. DOENHOFF, M., MUSALLAM, R., BAIN, J., AND McGREGOR, A. 1979. Schistosoma mansoni infections in T-cell deprived mice, and the ameliorating effect of administering chronic infection serum. I. Pathogenesis. American Journal of Tropical Medicine and

D~ENHOFF, GREGOR,

Hygiene 28, 268-273. DOENHOFF, M. J., PEARSON, S., DUNNE, D. W., BICKLE, Q., LUCAS, S., BAIN, J., MUSALLAM, R., AND HASSOUNAH, 0. 1981. Immunological control

of hepatotoxicity

and parasite egg excretion in infections: Stage specificity of the reactivity of immune serum in T-cell deprived mice. Transactions of the Royal Society of Tropical Medicine and Hygiene 75, 41-53. DUNNE, D. W., AGNEW, A. M., MODHA, J., AND Schistosoma

DOENHOFF,

mansoni

M. J. 1986. Schistosoma

mansoni

egg

antigens: preparation of rabbit antisera with monospecific immunoprecipitating activity, and their use in antigen characterization. Parasite Immunology 8, 575-586. DUNNE, D. W., BAIN, J., LILLYWHITE, J., AND D~ENHOFF, M. J. 1984. The stage-, strain- and speties-specificity of a Schistosoma mansoni egg antigen fraction (CEF6) with serodiagnostic potential. Transactions of the Royal Society of Tropical Medicine and Hygiene 78, 460-470. DUNNE, D. W., AND D~ENHOFF, M. J. 1983. Schistosoma mansoni egg antigens and hepatocyte damage in infected T cell-deprived mice. Contributions in Microbiology and Immunology 7, 22-29. DUNNE, D. W., HASSOUNAH, O., MUSALLAM, R., LUCAS, S., PEPYS, M. B., BALTZ, M., AND DOEN-

s.

HEPATOTOXICITY

i?‘UZnSOni:

HOFF, M. J. 1983. Mechanisms of Schistosoma mansoni egg excretion: Parasitological observations in immunosuppressed mice reconstituted with immune serum. Parasite Immunology 5, 47-60. DUNNE, D. W., JONES, F. M., AND DOENHOFF, M. J. 1991. The purification, characterization, serological activity and hepatotoxic properties of two cationic glycoproteins (q and ai) from Schistosoma mansoni eggs. Parasitology

103, 225-236.

DUNNE, D. W., LUCAS, S., BICKLE, Q., PEARSON, S., MADGWICK, L., BAIN, J., AND DOENHOFF, M. 1981. Identification and partial purification of an antigen (q) from Schistosoma mansoni eggs which is putatively hepatotoxic in T-cell deprived mice. Transactions of the Royal Society icine and Hygiene IS, 54-l 1.

for

Tropical

Med-

KARMEN, A. 1955. A note on the spectrophotometric assay of glutamic-oxalacetic transaminase in human blood serum. Journal of Clinical Investigation 34, 126-133. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. 1951. Protein measurement with the folin phenol reagent. Journal of Biological Chemistry

193, 265-275.

LUCAS, S., MUSALLAM, R., BAIN, J., HASSOUNAH, O., BICKLE, Q., AND DOENHOFF, M. 1980. The pathological effects of immunosuppression of Schistosoma mansoni-infected mice, with particular reference to survival and hepatotoxicity after thymectomy and treatment with antithymocyte serum, and treatment with hydrocortisone acetate. Transac-

AND

EGG

EXCRETION

tions of the Royal Society Medicine 74, 633-643.

339

of Tropical

Hygiene

and

MCLAREN, M. L., LILLYWHITE, J. E., DUNNE, D. W., AND DOENHOFF, M. J. 1981. Serodiagnosis of human Schistosoma mansoni infection: Enhanced sensitivity and specificity in ELISA using a fraction containing S. mansoni egg antigens o, and a,. Transactions of the Royal Society Medicine and Hygiene IS, 72-79.

of Tropical

MOLONEY, N. A., AND WEBBE, G. 1982. A rapid method for the infection of laboratory mice with Schistosoma japonicum. Transactions of the Royal Society of Tropical Medicine and Hygiene 16, 200-

203. MOTT, K. E., AND DIXON, H. 1982. Collaborative study on antigens for immunodiagnosis of schistosomiasis. Bulletin of the World Health Organization 60,129-153.

MURARE, H. M., AGNEW, A. M., BALTZ, M. N., LuCAS, S. B., AND DOENHOFF, M. J. 1987. The response to Schistosoma bovis in normal and T-cell deprived mice. Parasitology 95, 517-530. MURARE, H. M., AND DOENHOFF, M. J. 1987. Parasitological observations of Schistosoma bovis in normal and T-cell deprived mice. Parasitology 95, 507-516. SMITHERS, S. R., AND TERRY, R. J. 1965. The infection of laboratory hosts with cercariae of Schistosoma

mansoni.

Parasitology

55, 695-700.

Received 16 March 1992; accepted with revision 28 July 1992