- Email: [email protected]
TNF and inhibition of growth of Plasmodium falciparum
Immunology Letters, 25 (1990), 175-178 Elsevier
IMLET 01449
TNF and inhibition of growth of Plasmodium falciparum I. A. C l a r k , W. B. C o w d e n a n d G. A. B u t c h e r Zoology Department and John Curtin School of Medical Research, Australian National University, Canberra, Australia
1. S u m m a r y The mechanism of intra-erythrocyte death of Plasmodium chabaudiin vivo has not yet been elucidated. Here we summarise recent experiments in which serum from mice undergoing a successful immune response to this parasite did not inhibit Plasm o d i u m falciparum in vivo unless the P chabaudi infection and T N F levels were high enough to cause illness in the host. This was true for the 556KA and DS strains of P.. chabaudi in intact mice, but not for 556KA in nude mice, which did not generate inhibitory activity at any parasitaemia. Tumour necrosis factor (TNF) inhibits malaria parasites via some undefined secondary mediator. 10 mg of r hu T N F generated this inhibitory activity, as measured against P falciparum in vitro, in the serum of mice only if they were pretreated with Corynebacterium parvum, which activates macrophages and sensitises the mice to the toxic effects of T N E This implies a role for activated macrophages downstream from T N F in the process involved in intra-erythrocytic death of parasites.
vivo [2]. This observation was soon confirmed in vitro in other mouse models [3], and the same crude material subsequently proved to inhibit Plasmodium falciparum [4]. Once recombinant T N F became available, however, it soon was shown to lack direct in vitro toxicity against Plasmodium yoelii [5] or P. falciparum [6]. Nevertheless in vivo activity could be demonstrated against murine malaria parasites [5, 7], implicating a role for some undefined mediator induced by T N F (which is present in serum from mice or patients with malarial illness) [8, 9] in control of the parasite during the acute disease. To investigate this further we have tested the ability of serum from mice given TNF, with or without prior macrophage activation by C. parvum, to kill P. falciparum in vitro. In addition, we have compared the effectiveness of serum from mice infected with P. ehabaudi 556KA of different virulence, and in nude or intact mice, to inhibit growth of cultured P falciparum. 3. Materials and Methods 3.1. Treatment o f donor mice
2. Introduction
Some time ago we argued a role for tumour necrosis factor (TNF) and other factors of macrophage origin in the intra-erythrocytic degeneration of malaria parasites [I, 2]. In these studies malaria was found to prime mice for T N F production as effectively as did Cornyebacterium parvum, and crude T N F was observed to inhibit Plasmodium vinckei in Key words." Tumour necrosis factor; Crisis form; Plasmodium chabaudi; Plasmodiumfalciparum; Activated macrophages Correspondenceto: I. A. Clark, Zoology Dept. and John Curtin School of Medical Research, Australian National University, Canberra, ACT 2601, Australia. Fax: +61-62-490102.
C B A / H male or female mice of 6 - 8 weeks old were infected with 106 P. chabaudi (556 KA strain or DS strain; from D. Walliker, Edinburgh, and A. Low, Melbourne, respectively) and the course of infections followed on Giemsa-stained smears. They were bled in groups of three to five at regular intervals under sterile conditions and serum separated and stored at - 7 0 ° C until required. The 556 KA strain of P chabaudi usually induces parasitaemias not rising much above 30%, but with passage becomes more virulent and parasitaemias go higher, until eventually it may become lethal. The groups of mice described in the following experiments were infected when the parasites were in both the early, less
0165-2478 / 90 / $ 3.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
175
virulent phase and in the later more virulent phase that develops over some months of regular passage. On recovery from P. chabaudi 556 KA some mice were challenged with high doses (10 9) of the homologous parasite and bled at 2, 24, 48 or 72 h later. Other mice were allowed to recover, and bled several weeks or months later, but were not challenged during the intervening period. A group of 21 nude (nu/nu) mice were also infected with the 556 KA P. chabaudi strain and bled out after several days of high parasitaemia. Other groups of C B A / H mice were infected with Plasmodium vinckei vinckei strains and bled when most animals reached peak parasitaemia levels. In two sets of experiments, two groups of 4 or 8 C B A / H mice 6 - 8 weeks old were given an i.v. injection of 1 mg killed C. parvum (Wellcome) followed 11 days later by 10 ~tg of recombinant h u m a n T N F by i.v. injection. They were also bled two hours later. Control mice were given either C. parvum, or T N F alone. 3.2. In vitro cultures o f P. falciparum P. falciparum (a Papua New Guinea strain, FCQ27) was maintained as previously described, in 10% v/v h u m a n group 0 serum in RPMI-1640 [10]. Mouse sera were tested at various concentrations (usually at 0.8 and 2.4% v/v of total culture volume) in flat-well microculture trays containing 100 ml of a 1% cell suspension in the usual culture medium; 20 ml of 10 m C i / m l of [3H]hypoxanthine ( 1 - 5 C l / m M , A m e r s h a m Radiochemicals) was added to wells. Cultures were run for 24 hours in the gas mixture used previously [10]. Percentage inhibition of parasite growth was calculated as previously described [10]. [3H]Hypoxanthine uptake in wells containing normal mouse serum was taken as 100% growth. For morphological observations, the [3H]hypoxanthine was omitted and smears were made to assess parasite growth and estimate the numbers of damaged forms. These cultures were synchronised by the sorbitol-treatment method of Lambros and Vanderburg [11]. Some sera (from mice in the same group) were pooled and dialysed against culture medium minus h u m a n serum, using dialysis tubing with a cut-off of 14 kDa. The dialysed mouse sera were tested against 176
P. falciparum, and results compared with those from undialysed sera. To ensure normal parasite growth, normal non-dialysed h u m a n serum was added to the culture wells. Each set of mouse sera was tested (individually or pooled for dialysis experiments) in several culture experiments. 4. Results
4.1. Post-TNF mouse serum added to P. falciparum cultures In initial experiments, serum from normal mice that had received 10 mg r hu T N F 2 h before being bled was no more toxic to cultures of P. falciparum than was normal mouse serum. This was reminicent of our earlier experiments with endotoxin [2], so we considered the possibility that a small dose of TNF, like endotoxin, elicited anti-parasitic activity only in mice with activated macrophages i.e., animals that were sensitised to its toxic effects [12]. Accordingly, the same 10/zg dose of T N F was given to mice previously treated with C. parvum (see Materials and Methods for details). Their serum was, in contrast, highly toxic to P falciparum, equalling the activity observed in the serum of mice that received 10 tzg of endotoxin instead of T N F (G. A. Butcher and I. A. Clark, manuscript submitted). Sera from mice given C. parvum alone or endotoxin alone were, under the same conditions, non-inhibitory for P. falciparum, as had been serum from mice given T N F alone. We noted that this dose of r hu T N F is not toxic in normal mice, but markedly so in those pretreated with C. parvum. Thus it appears that T N F generates a serum factor(s) toxic for P. falciparum only under conditions where this cytokine is also toxic for the host, such as when systemic macrophage activation is present. This implies, but does not prove, that these cells are the source of the toxic factor. 4.2. Serum f r o m mice infected with less and more virulent P. chabaudi 556 K A added to P. falciparum cultures Over several months' passage a particular isolate of P. chabaudi 556 KA gradually increased in virulence to the host. Early in this period it was attaining a peak parasitaemia of about 30°-/0, producing no
evident illness in the host, and contained neither detectable serum TNF nor inhibitory activity against Pfalciparum at any time during the course of infection, whether tested at 2.407o or 6°70 (v/v). In contrast, sera from mice carrying more virulent infections (peak parasitaemias about 50%, TNF detectable, and mild to moderate illness) inhibited growth by as much as 92°7o at 2.4°7o (v/v) (G. A. Butcher and I. A. Clark, manuscript submitted). Inhibitory activity was highest on day 9, two days after peak parasitaemia, and fell to less than 30°70 by day 11, when the mice were clinically normal again. Crisis forms were routinely observed as both the less or more virulent infections were being cleared from the circulation. 4.3. Serum from mice infected with lethal P. chabaudi DS added to P. falciparum cultures For a given rising parasitaemia, serum from mice infected with 10 6 of this parasite had considerably less activity against P. falciparum than did serum from mice infected with 10 6 of the 556 KA strain. Eventually inhibition rose to 85°70, but only on day 11, when the mice were near death. 4.4. Serum from nude mice infected with P. chabaudi 556 KA Infections in outbred mice reached at least 50°70 parasitaemia, but, in keeping with earlier reports, did not decline. At no stage over several days at these high parasitaemias did their serum contain inhibitory activity, or detectable TNF (G. A. Butcher and I. A. Clark, manuscript submitted).
4.5. Serum from mice re-challenged with P. chabaudi 556 KA
ia parasites in vivo via as an yet undefined final mediator. Lipid peroxidation products, to which malaria parasites are very susceptible [13], have been implicated [14, 15]. Another possibility is that metabolic products of prostaglandins, which are induced by TNF, could help inhibit parasite growth. This is consistent with a recent report [16] that oligomeric prostaglandin derivatives inhibit growth of P chabaudi in vivo. Whatever the final mediator, clearly it has to have a cellular source. Our experiments with post-TNF mouse serum point to the activated macrophage, a cell that is compatible with both radical species and prostaglandins, as well as a range of other mediators. A requirement for a T cell product must also be incorporated into the model. In keeping with the illness produced in C. parvumpretreatment followed by TNF, the capacity of serum from P chabaudi-infected mice to inhibit P.falciparum in vitro correlated not with how well the infection was controlled, but with how ill this infection was making the mice at the time of serum collection. These data also indicate that crisis forms can develop in the absence of circulating inhibitory activity in the serum, as evidenced by the avirulent infections of P. chabaudi 556 KA and the rechallenged mice. Membrane-bound cytokines may explain this phenomenon. Alternatively, certain nontransferable mediators may cause intra-erythrocytic death of circulating parasites at low parasitaemias, and additional factors become operative as parasitaemias rise above the threshold required for illness. This pattern is consistent with the most recent data from Jensen [17], who now reports that the best donors of CFF-positive sera are those with acute malaria, and that CFF activity declines with restoration of health and acquisition of immunity.
Acknowledgements Several weeks or months after recovery from P chabaudi 556 KA mice were challenged with 109 parasites of the same strain, and bled 2 - 7 h later. No illness or inhibitory activity was detected in this serum, although crisis forms were observed in the circulation.
5. Discussion
This study was supported by the malaria component of the UNDP/World B a n k / W H O Special Programme for Research and Training in Tropical Disease, and by the Australian National Health and Medical Research Council. We are grateful to Sarah May and Kylie Place for technical assistance, and to Karen Gray for typing the manuscript.
Current evidence points to TNF inhibiting malar177
References [1] Clark, I. A. (1978) Lancet ii, 75. [2] Clark, I. A., Virelizier, J.-L., Carswell, E. A. and Wood, P. R. (1981) Infect. Immun. 32, 1058. [3] Taverne, J., Dockrell, H. M. and Playfair, J. H. L. (1981) Infect. Immun. 33, 83. [4] Haidaris, C. G., Haynes, J. D., Meltzer, M. S. and Allison, A. C. (1983) Infect. lmmun. 42, 385. [5] Taverne, J., Tavernier, J., Fiers, W. and Playfair, J. H. L. (1987) Clin. Exp. lmmunol. 67, 1. [6] Jensen, J. B., Vande Waa, J. A. and Karadsheh, A. J. (1987) Infect. Immun. 55, 1722. [7] Clark, I. A., Hunt, N. H., Butcher, G. A. and Cowden, W. B. (1987) J. lmmunol. 139, 3493. [8] Clark, I. A. and Chaudhri, G. (1988) Brit. J. Haematol. 70, 99. [9] Kern, P., Hemmer, C. J., Van Damme, J., Gruss, H.-J. and Dietrich, M. (1989) Am. J. Med. 87, 139.
178
[10] Butcher, G. A., Clark, I. A. and Crane, G. (1987) Trans. R. Soc. Trop. Med. Hyg. 81, 568. [11] Lambros, C. and Vanderburg, J. R (1979) J. Parasitol. 65, 418. [12] Clark, 1. A., Cowden, W. B., Butcher, G. A. and Hunt, N. H. (1987) Am. J. Pathol. 129, 192. [13] Clark, I. A., Butcher, G. A., Buffinton, G. D., Hunt, N. FI. and Cowden, W. B. (1987) Biochem. Pharmacol. 36, 543. [14] Buffinton, G. D., Hunt, N. H., Cowden, W. B. and Clark, 1. A. (1988) Biochem. J. 249, 63. [15] Rockett, K . A . , Targett, G. A. T. and Playfair, J. H. L. (1988) Infect. Immun. 56, 3180. [16] Ohnishi, S. T., Ohnishi, N., Oda, Y. and Katsuoka, M. (1989) Cell Biochem. Funct. 7, 105. [17] 3ensen, J. B. (1989). In: Malaria: Host Responses to Infection (M. M. Stevenson, Ed.) pp. 109-126. CRC Press, Boca Raton, FL. (Accepted for publication 15 May 1990)
IMLET 01449
TNF and inhibition of growth of Plasmodium falciparum I. A. C l a r k , W. B. C o w d e n a n d G. A. B u t c h e r Zoology Department and John Curtin School of Medical Research, Australian National University, Canberra, Australia
1. S u m m a r y The mechanism of intra-erythrocyte death of Plasmodium chabaudiin vivo has not yet been elucidated. Here we summarise recent experiments in which serum from mice undergoing a successful immune response to this parasite did not inhibit Plasm o d i u m falciparum in vivo unless the P chabaudi infection and T N F levels were high enough to cause illness in the host. This was true for the 556KA and DS strains of P.. chabaudi in intact mice, but not for 556KA in nude mice, which did not generate inhibitory activity at any parasitaemia. Tumour necrosis factor (TNF) inhibits malaria parasites via some undefined secondary mediator. 10 mg of r hu T N F generated this inhibitory activity, as measured against P falciparum in vitro, in the serum of mice only if they were pretreated with Corynebacterium parvum, which activates macrophages and sensitises the mice to the toxic effects of T N E This implies a role for activated macrophages downstream from T N F in the process involved in intra-erythrocytic death of parasites.
vivo [2]. This observation was soon confirmed in vitro in other mouse models [3], and the same crude material subsequently proved to inhibit Plasmodium falciparum [4]. Once recombinant T N F became available, however, it soon was shown to lack direct in vitro toxicity against Plasmodium yoelii [5] or P. falciparum [6]. Nevertheless in vivo activity could be demonstrated against murine malaria parasites [5, 7], implicating a role for some undefined mediator induced by T N F (which is present in serum from mice or patients with malarial illness) [8, 9] in control of the parasite during the acute disease. To investigate this further we have tested the ability of serum from mice given TNF, with or without prior macrophage activation by C. parvum, to kill P. falciparum in vitro. In addition, we have compared the effectiveness of serum from mice infected with P. ehabaudi 556KA of different virulence, and in nude or intact mice, to inhibit growth of cultured P falciparum. 3. Materials and Methods 3.1. Treatment o f donor mice
2. Introduction
Some time ago we argued a role for tumour necrosis factor (TNF) and other factors of macrophage origin in the intra-erythrocytic degeneration of malaria parasites [I, 2]. In these studies malaria was found to prime mice for T N F production as effectively as did Cornyebacterium parvum, and crude T N F was observed to inhibit Plasmodium vinckei in Key words." Tumour necrosis factor; Crisis form; Plasmodium chabaudi; Plasmodiumfalciparum; Activated macrophages Correspondenceto: I. A. Clark, Zoology Dept. and John Curtin School of Medical Research, Australian National University, Canberra, ACT 2601, Australia. Fax: +61-62-490102.
C B A / H male or female mice of 6 - 8 weeks old were infected with 106 P. chabaudi (556 KA strain or DS strain; from D. Walliker, Edinburgh, and A. Low, Melbourne, respectively) and the course of infections followed on Giemsa-stained smears. They were bled in groups of three to five at regular intervals under sterile conditions and serum separated and stored at - 7 0 ° C until required. The 556 KA strain of P chabaudi usually induces parasitaemias not rising much above 30%, but with passage becomes more virulent and parasitaemias go higher, until eventually it may become lethal. The groups of mice described in the following experiments were infected when the parasites were in both the early, less
0165-2478 / 90 / $ 3.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
175
virulent phase and in the later more virulent phase that develops over some months of regular passage. On recovery from P. chabaudi 556 KA some mice were challenged with high doses (10 9) of the homologous parasite and bled at 2, 24, 48 or 72 h later. Other mice were allowed to recover, and bled several weeks or months later, but were not challenged during the intervening period. A group of 21 nude (nu/nu) mice were also infected with the 556 KA P. chabaudi strain and bled out after several days of high parasitaemia. Other groups of C B A / H mice were infected with Plasmodium vinckei vinckei strains and bled when most animals reached peak parasitaemia levels. In two sets of experiments, two groups of 4 or 8 C B A / H mice 6 - 8 weeks old were given an i.v. injection of 1 mg killed C. parvum (Wellcome) followed 11 days later by 10 ~tg of recombinant h u m a n T N F by i.v. injection. They were also bled two hours later. Control mice were given either C. parvum, or T N F alone. 3.2. In vitro cultures o f P. falciparum P. falciparum (a Papua New Guinea strain, FCQ27) was maintained as previously described, in 10% v/v h u m a n group 0 serum in RPMI-1640 [10]. Mouse sera were tested at various concentrations (usually at 0.8 and 2.4% v/v of total culture volume) in flat-well microculture trays containing 100 ml of a 1% cell suspension in the usual culture medium; 20 ml of 10 m C i / m l of [3H]hypoxanthine ( 1 - 5 C l / m M , A m e r s h a m Radiochemicals) was added to wells. Cultures were run for 24 hours in the gas mixture used previously [10]. Percentage inhibition of parasite growth was calculated as previously described [10]. [3H]Hypoxanthine uptake in wells containing normal mouse serum was taken as 100% growth. For morphological observations, the [3H]hypoxanthine was omitted and smears were made to assess parasite growth and estimate the numbers of damaged forms. These cultures were synchronised by the sorbitol-treatment method of Lambros and Vanderburg [11]. Some sera (from mice in the same group) were pooled and dialysed against culture medium minus h u m a n serum, using dialysis tubing with a cut-off of 14 kDa. The dialysed mouse sera were tested against 176
P. falciparum, and results compared with those from undialysed sera. To ensure normal parasite growth, normal non-dialysed h u m a n serum was added to the culture wells. Each set of mouse sera was tested (individually or pooled for dialysis experiments) in several culture experiments. 4. Results
4.1. Post-TNF mouse serum added to P. falciparum cultures In initial experiments, serum from normal mice that had received 10 mg r hu T N F 2 h before being bled was no more toxic to cultures of P. falciparum than was normal mouse serum. This was reminicent of our earlier experiments with endotoxin [2], so we considered the possibility that a small dose of TNF, like endotoxin, elicited anti-parasitic activity only in mice with activated macrophages i.e., animals that were sensitised to its toxic effects [12]. Accordingly, the same 10/zg dose of T N F was given to mice previously treated with C. parvum (see Materials and Methods for details). Their serum was, in contrast, highly toxic to P falciparum, equalling the activity observed in the serum of mice that received 10 tzg of endotoxin instead of T N F (G. A. Butcher and I. A. Clark, manuscript submitted). Sera from mice given C. parvum alone or endotoxin alone were, under the same conditions, non-inhibitory for P. falciparum, as had been serum from mice given T N F alone. We noted that this dose of r hu T N F is not toxic in normal mice, but markedly so in those pretreated with C. parvum. Thus it appears that T N F generates a serum factor(s) toxic for P. falciparum only under conditions where this cytokine is also toxic for the host, such as when systemic macrophage activation is present. This implies, but does not prove, that these cells are the source of the toxic factor. 4.2. Serum f r o m mice infected with less and more virulent P. chabaudi 556 K A added to P. falciparum cultures Over several months' passage a particular isolate of P. chabaudi 556 KA gradually increased in virulence to the host. Early in this period it was attaining a peak parasitaemia of about 30°-/0, producing no
evident illness in the host, and contained neither detectable serum TNF nor inhibitory activity against Pfalciparum at any time during the course of infection, whether tested at 2.407o or 6°70 (v/v). In contrast, sera from mice carrying more virulent infections (peak parasitaemias about 50%, TNF detectable, and mild to moderate illness) inhibited growth by as much as 92°7o at 2.4°7o (v/v) (G. A. Butcher and I. A. Clark, manuscript submitted). Inhibitory activity was highest on day 9, two days after peak parasitaemia, and fell to less than 30°70 by day 11, when the mice were clinically normal again. Crisis forms were routinely observed as both the less or more virulent infections were being cleared from the circulation. 4.3. Serum from mice infected with lethal P. chabaudi DS added to P. falciparum cultures For a given rising parasitaemia, serum from mice infected with 10 6 of this parasite had considerably less activity against P. falciparum than did serum from mice infected with 10 6 of the 556 KA strain. Eventually inhibition rose to 85°70, but only on day 11, when the mice were near death. 4.4. Serum from nude mice infected with P. chabaudi 556 KA Infections in outbred mice reached at least 50°70 parasitaemia, but, in keeping with earlier reports, did not decline. At no stage over several days at these high parasitaemias did their serum contain inhibitory activity, or detectable TNF (G. A. Butcher and I. A. Clark, manuscript submitted).
4.5. Serum from mice re-challenged with P. chabaudi 556 KA
ia parasites in vivo via as an yet undefined final mediator. Lipid peroxidation products, to which malaria parasites are very susceptible [13], have been implicated [14, 15]. Another possibility is that metabolic products of prostaglandins, which are induced by TNF, could help inhibit parasite growth. This is consistent with a recent report [16] that oligomeric prostaglandin derivatives inhibit growth of P chabaudi in vivo. Whatever the final mediator, clearly it has to have a cellular source. Our experiments with post-TNF mouse serum point to the activated macrophage, a cell that is compatible with both radical species and prostaglandins, as well as a range of other mediators. A requirement for a T cell product must also be incorporated into the model. In keeping with the illness produced in C. parvumpretreatment followed by TNF, the capacity of serum from P chabaudi-infected mice to inhibit P.falciparum in vitro correlated not with how well the infection was controlled, but with how ill this infection was making the mice at the time of serum collection. These data also indicate that crisis forms can develop in the absence of circulating inhibitory activity in the serum, as evidenced by the avirulent infections of P. chabaudi 556 KA and the rechallenged mice. Membrane-bound cytokines may explain this phenomenon. Alternatively, certain nontransferable mediators may cause intra-erythrocytic death of circulating parasites at low parasitaemias, and additional factors become operative as parasitaemias rise above the threshold required for illness. This pattern is consistent with the most recent data from Jensen [17], who now reports that the best donors of CFF-positive sera are those with acute malaria, and that CFF activity declines with restoration of health and acquisition of immunity.
Acknowledgements Several weeks or months after recovery from P chabaudi 556 KA mice were challenged with 109 parasites of the same strain, and bled 2 - 7 h later. No illness or inhibitory activity was detected in this serum, although crisis forms were observed in the circulation.
5. Discussion
This study was supported by the malaria component of the UNDP/World B a n k / W H O Special Programme for Research and Training in Tropical Disease, and by the Australian National Health and Medical Research Council. We are grateful to Sarah May and Kylie Place for technical assistance, and to Karen Gray for typing the manuscript.
Current evidence points to TNF inhibiting malar177
References [1] Clark, I. A. (1978) Lancet ii, 75. [2] Clark, I. A., Virelizier, J.-L., Carswell, E. A. and Wood, P. R. (1981) Infect. Immun. 32, 1058. [3] Taverne, J., Dockrell, H. M. and Playfair, J. H. L. (1981) Infect. Immun. 33, 83. [4] Haidaris, C. G., Haynes, J. D., Meltzer, M. S. and Allison, A. C. (1983) Infect. lmmun. 42, 385. [5] Taverne, J., Tavernier, J., Fiers, W. and Playfair, J. H. L. (1987) Clin. Exp. lmmunol. 67, 1. [6] Jensen, J. B., Vande Waa, J. A. and Karadsheh, A. J. (1987) Infect. Immun. 55, 1722. [7] Clark, I. A., Hunt, N. H., Butcher, G. A. and Cowden, W. B. (1987) J. lmmunol. 139, 3493. [8] Clark, I. A. and Chaudhri, G. (1988) Brit. J. Haematol. 70, 99. [9] Kern, P., Hemmer, C. J., Van Damme, J., Gruss, H.-J. and Dietrich, M. (1989) Am. J. Med. 87, 139.
178
[10] Butcher, G. A., Clark, I. A. and Crane, G. (1987) Trans. R. Soc. Trop. Med. Hyg. 81, 568. [11] Lambros, C. and Vanderburg, J. R (1979) J. Parasitol. 65, 418. [12] Clark, 1. A., Cowden, W. B., Butcher, G. A. and Hunt, N. H. (1987) Am. J. Pathol. 129, 192. [13] Clark, I. A., Butcher, G. A., Buffinton, G. D., Hunt, N. FI. and Cowden, W. B. (1987) Biochem. Pharmacol. 36, 543. [14] Buffinton, G. D., Hunt, N. H., Cowden, W. B. and Clark, 1. A. (1988) Biochem. J. 249, 63. [15] Rockett, K . A . , Targett, G. A. T. and Playfair, J. H. L. (1988) Infect. Immun. 56, 3180. [16] Ohnishi, S. T., Ohnishi, N., Oda, Y. and Katsuoka, M. (1989) Cell Biochem. Funct. 7, 105. [17] 3ensen, J. B. (1989). In: Malaria: Host Responses to Infection (M. M. Stevenson, Ed.) pp. 109-126. CRC Press, Boca Raton, FL. (Accepted for publication 15 May 1990)