Recombinant Tumor Necrosis Factor Alpha Does Not Inhibit the ...

INFECTION AND IMMUNITY, Apr. 2002, p. 2210–2214 0019-9567/02/$04.00⫹0 DOI: 10.1128/IAI.70.4.2210–2214.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Vol. 70, No. 4

Recombinant Tumor Necrosis Factor Alpha Does Not Inhibit the Growth of African Trypanosomes in Axenic Cultures† Hiroshi Kitani,1* Samuel J. Black,2 Yoshio Nakamura,1 Jan Naessens,3 Noel B. Murphy,4 Yuichi Yokomizo,5 John Gibson,3 and Fuad Iraqi3 Japan International Research Center for Agricultural Sciences,1 and National Institute of Animal Health,5 Tsukuba, Ibaraki, Japan; Department of Veterinary and Animal Sciences, Paige Laboratory, University of Massachusetts, Amherst, Massachusetts2; International Livestock Research Institute, Nairobi, Kenya3; and Smurfit Institute of Genetics, Trinity College Dublin, Dublin, Ireland4 Received 3 October 2001/Returned for modification 31 October 2001/Accepted 17 December 2001

Mice whose tumor necrosis factor alpha (TNF-␣) genes were disrupted developed higher levels of parasitemia than wild-type mice following infection with Trypanosoma congolense IL1180 or T. brucei brucei GUTat3.1, confirming the results of earlier studies. To determine whether TNF-␣ directly affects the growth of these and other bloodstream forms of African trypanosomes, we studied the effects of recombinant mouse, human, and bovine TNF-␣ on the growth of two isolates of T. congolense, IL1180 and IL3338, and two isolates of T. brucei brucei, GUTat3.1 and ILTat1.1, under axenic culture conditions. The preparations of recombinant TNF-␣ used were biologically active as determined by their capacity to kill L929 cells. Of five recombinant TNF-␣ lots tested, one lot of mouse TNF-␣ inhibited the growth of both isolates of T. brucei brucei and one lot of bovine TNF-␣ inhibited the growth of T. brucei brucei ILTat1.1 but only at very high concentrations and without causing detectable killing of the parasites. The other lots of mouse recombinant TNF-␣, as well as human TNF-␣, did not affect the growth of any of the test trypanosomes even at maximal concentrations that could be attained in the culture systems (3,000 to 15,000 U of TNF-␣/ml of medium). These results suggest that exogenously added recombinant TNF-␣ generally does not inhibit the growth of African trypanosomes under the culture conditions we used. The impact of TNF-␣ on trypanosome parasitemia may be indirect, at least with respect to the four strains of trypanosomes reported here. brucei AnTat1.1 suspended in phosphate-buffered saline (13, 15) and T. brucei gambiense OK in culture medium (4), its effect on T. congolense has not been studied. Furthermore, TNF-␣ preparations from different mammalian species have not been compared with respect to trypanocidal or trypanostatic activities for any of the African trypanosomes, many of which have a wide host range. The purpose of this study was to examine the effects of recombinant murine, human, and bovine TNF-␣ on the growth of bloodstream forms of T. congolense and T. brucei brucei under in vitro conditions. To evaluate the roles of TNF-␣ in trypanosomiasis in mice, TNF-␣-deficient mice (20) and wild-type C57BL/6 mice at 8 to 12 weeks of age were infected by an intraperitoneal injection of 104 trypanosomes. After challenge, blood samples were taken by tail snip, and the number of trypanosomes per milliliter of blood was scored by the hemocytometer method. As shown in Fig. 1A, TNF-␣-deficient mice infected with T. congolense IL1180 showed a higher peak of parasitemia on day 11 than did the wild-type mice. Parasite numbers then decreased, and there was no significant difference in the numbers between mouse strains from day 14 to day 21. Between day 24 and day 35, however, parasite numbers in the TNF-␣-deficient mice significantly increased again whereas those in the wild-type mice remained low. In addition, TNF-␣-deficient mice died significantly earlier than the wild-type mice after T. congolense infection (data not shown), confirming the results of a previous report (10). TNF-␣-deficient C57BL/6 mice that were infected with T. brucei brucei GUTat3.1 also developed higher parasitemia levels than those of the wild-type mice (Fig. 1B), but survival times were similar (data not shown). These results

Several lines of evidence suggest that tumor necrosis factor alpha (TNF-␣) contributes to the control of trypanosomiasis in mice (2, 14, 21). First, mapping of trypanosomiasis resistance loci to determine quantitative effects in mice infected with Trypanosoma congolense suggests that the locus with the largest effect encompasses the TNF-␣ gene (9, 11, 12). Second, TNF-␣ gene-disrupted mice with the C57BL/6 genetic background are highly susceptible to T. congolense infections and die significantly earlier than wild-type C57BL/6 mice (10). Concomitant with the shorter survival times, TNF-␣-deficient mice develop significantly higher levels of parasitemia than wild-type mice (J. Naessens, H. Kitani, E. Momotani, K. Sekikawa, J. M. Nthale, and F. Iraqi, submitted for publication). Finally, identical TNF-␣-deficient mice (bred in the same lab) also develop much higher levels of parasitemia than do wild-type mice when infected with Trypanosoma brucei brucei (16). However, in the case of T. brucei brucei infections, survival times of the TNF␣-deficient mice do not differ from those of wild-type mice (16). These observations suggest that TNF-␣ plays an important role in the control of parasitemia in murine trypanosomiasis, but its contributions to the control of pathogenesis and survival may differ between T. congolense and T. brucei brucei infections. Although TNF-␣ was demonstrated to be lytic to T. brucei * Corresponding author. Mailing address: National Institute of Agrobiological Sciences (NIAS), 3-1-5 Kannondai, Tsukuba, Ibaraki 305-8602, Japan. Phone and fax: 81-298-38-7801. E-mail: kitani@affrc .go.jp. † International Livestock Research Institute publication no. 200160. 2210

VOL. 70, 2002

NOTES

2211

TABLE 1. Biological activities of recombinant TNF-␣ prepared in T. congolense and T. brucei brucei media Recombinant TNF-␣

Standard LD50b

T. congolense

Mouse lot A Mouse lot B Human Bovine

T. brucei brucei

Mouse lot A Mouse lot B Human Bovine

Organism

LD50a for organism incubated in medium for: 0h

2h

8h

48 h

0.18 1.15 0.20 0.17

0.18 3.10 0.20 0.17

0.18 2.95 0.20 0.18

0.18 1.80 0.20 0.18

0.18 0.95 0.90 0.20

0.18 1.15 0.20 0.17

0.18 2.10 0.20 0.17

0.18 1.70 0.30 0.18

0.18 1.15 NDc 0.17

0.18 0.60 ND 0.22

a Values are the doses at which 50% of the L929 cells were lysed in units per milliliter. b Recombinant TNF-␣ preparations were prepared in L929 cell culture medium and used for bioassay. c ND, not done.

FIG. 1. Changes in mean numbers of parasites in the peripheral blood in wild-type (■) and TNF-␣-deficient (䊐) mice after infection with T. congolense IL1180 (A) and T. brucei brucei GUTat3.1 (B). Mice were infected by intraperitoneal injection of 104 parasites at day 0. Data are mean values ⫾ standard errors of the means (SEM) from experiments with three male and three female mice. Significant differences at P values of ⬍0.05 (ⴱ), ⬍0.01 (ⴱⴱ), and ⬍0.001 (ⴱⴱⴱ) between the two strains are indicated.

confirm the finding of Magez et al. (16) for identical mice (bred in the same lab) infected with T. brucei brucei AnTat1.1. However, the difference in the levels of T. brucei brucei GUTat3.1 parasitemia between the TNF-␣ gene-disrupted and wild-type C57BL/6 mice was not as great as that in the levels of T. brucei brucei AnTat1.1 parasitemia (16). Notwithstanding this variation in results, the data obtained indicate that TNF-␣ contributes to the control of T. congolense IL1180 and T. brucei brucei GUTat3.1 as well as T. brucei brucei AnTat1.1 (16), thus validating the use of the former strains in assays aimed at evaluating whether TNF-␣ has a direct impact on the parasites. Commercially available recombinant mouse TNF-␣ (lots A and C, catalog no. 1271156, 5 ␮g/ml with a specific activity of ⱖ6 ⫻ 107 U/mg) was obtained from Boehringer Mannheim Biochemica, Mannheim, Germany. Another recombinant mouse TNF-␣ preparation (lot B, catalog no. TNF-M, 10 ␮g/ml with a specific activity of ⱖ5 ⫻ 107 U/mg) was obtained from Genzyme Corporation. Recombinant human TNF-␣ (catalog no. 1371843, 10 ␮g/ml with a specific activity of at least 108 U/mg) was obtained from Boehringer Mannheim Biochemica. These TNF-␣ preparations were produced in Escherichia coli, purified by standard chromatographic techniques, and supplied in sterile phosphate-buffered saline containing bovine serum albumin at 1 mg/ml. Recombinant bovine TNF-␣ (approximately 1 mg/ml in Dulbecco’s minimal essential medium [DMEM] with a specific activity of ⱖ2 ⫻ 105 U/mg) was produced in the Bacillus brevis host-vector system (22) by Higeta Shoyu Co., Ltd., Choshi, Japan. All of these recombi-

nant TNF-␣ preparations have similar polypeptide primary structures, which are identical to those in natural mouse, human, and bovine TNF-␣, but they are not glycosylated. To examine whether trypanosome culture conditions would affect the functions of the TNF-␣ preparations, we diluted them in trypanosome culture media, incubated them for 0, 2, 8 and 48 h at 37°C, and subsequently stored them at ⫺80°C until testing (18). The murine fibroblast cell line L929 was cultured in DMEM supplemented with 10% heat-inactivated fetal calf serum. L929 cells were plated in 96-well microtiter plates (2 ⫻ 104 cells in 100 ␮l of medium per well) and incubated overnight at 37°C in a humid atmosphere of 5% CO2–95% air. Test and standard TNF-␣ preparations were serially diluted and added to the L929 microtiter plates (50 ␮l per well). After addition of 50 ␮l of actinomycin D solution (4 ␮g/ml) to each well, the L929 cells in microtiter plates were incubated at 37°C for 18 h, after which the cells were washed with phosphate-buffered saline (150 ␮l per well) and stained with 0.05% crystal violet dissolved in 20% ethanol (50 ␮l per well) for 10 min at room temperature. After thorough rinsing in tap water, the plates were dried in air overnight. An aliquot of 100 ␮l of absolute methanol was added to each well to elute stain from cells, and the absorbance was measured with a microtiter plate reader set at a wavelength of 595 nm. All cells in wells treated with the highest dose of TNF-␣ (final concentration at 83 U/ml) died, and eluates from these cells were used as blanks (100% lysis), whereas all cells in wells treated only with medium survived, and eluates from these cells were used as controls (0% lysis). The 50% lethal dose (LD50) of each TNF-␣ preparation, i.e., the concentration at which 50% of L929 cells were lysed, was obtained from the dose-response curves. The results presented in Table 1 show that prolonged incubation at 37°C in trypanosome culture media had little or no impact on the LD50s (units per milliliter) of recombinant mouse, human, and bovine TNF-␣. The recombinant mouse TNF-␣ (lot A) and human and bovine TNF-␣ retained almost full biological activities after 48 h of incubation in both types of trypanosome culture media routinely used in our lab. There was a 30 to 60% decrease in the biological activity of recombinant mouse TNF-␣ of lot B after incubation for 0 and 2 h.

2212

NOTES

However, we observed almost full biological activity for the same TNF-␣ preparation after incubation for 8 or 48 h in both media. These results show that preparations of the recombinant TNF-␣ used in this study had biological activity to lyse L929 cells. This activity was expressed in the media used in the investigation of trypanostatic and trypanocidal activities as described below, and it was retained under the incubation conditions required for the growth of bloodstream stage African trypanosomes in vitro. T. congolense (IL1180 and IL3338) and T. brucei brucei (GUTat3.1 and ILTat1.1) were passaged at 3-day intervals in vitro according to the culture protocols previously described (7, 8). The culture medium for T. congolense was Iscove’s modified DMEM supplemented with 0.05 mM bathocuproine sulfonate, 1.5 mM L-cysteine, 0.5 mM hypoxanthine, 0.12 mM 2-mercaptoethanol, 1 mM sodium pyruvate, 0.16 mM thymidine, 2.0 mM L-glutamine, and 20% heat-inactivated goat serum. The culture medium for T. brucei brucei was Iscove’s modified DMEM supplemented with 0.05 mM bathocuproine sulfonate, 1.5 mM L-cysteine, 1.0 mM hypoxanthine, 0.2 mM 2-mercaptoethanol, 1 mM sodium pyruvate, 0.16 mM thymidine, 2.0 mM L-glutamine, and 20% heat-inactivated fetal calf serum. Trypanosomes were suspended in the appropriate culture medium at 104 parasites per ml, seeded into a 96-well plate in triplicate (100 ␮l per well), with or without various concentrations of recombinant TNF-␣, and cultured at 34°C in a humidified atmosphere of 4.5% CO2–95.5% air. Numbers of live parasites in the wells were enumerated by hemocytometer counting at intervals noted below for specific experiments. The impacts of the different lots of recombinant mouse TNF-␣ as well as human and bovine TNF-␣ on the growth of T. congolense IL1180 and IL3338 after 72 to 97 h of culture are shown in Fig. 2. Recombinant mouse and bovine TNF-␣ at levels up to 3,000 U per ml and recombinant human TNF-␣ at levels up to 15,000 U per ml had little or no effect on the growth of either strain of T. congolense (Fig. 2). Observation of the parasites with an inverted phase-contrast microscope after 12 h of incubation showed that all test organisms were motile at all concentrations of TNF-␣ tested and that no dead or dying organisms were present in the cultures. Similarly, as shown in Fig. 3, the growth of both T. brucei brucei GUTat3.1 and ILTat1.1 after 62 to 72 h of culture was not significantly affected by the recombinant TNF-␣ preparations tested, except in the case of one lot of recombinant mouse TNF-␣, lot C. This lot inhibited the the growth of both isolates of T. brucei brucei at a level of 3,000 U per ml but had no effect on the parasites at lower concentrations and did not induce detectable killing of parasites at any concentration. It is not clear at present why only this lot of recombinant mouse TNF-␣ had an inhibitory effect, while the other lots did not. While the inhibitory effect may have been due to this recombinant TNF-␣, it is also possible that another, unknown T. brucei brucei growth inhibitory factor was present in this TNF-␣ stock solution but absent in all others tested. Recombinant bovine TNF-␣ at a level of 3,000 U per ml slightly affected the growth of T. brucei brucei ILTat1.1 but not of T. brucei brucei GUTat3.1. Similar to T. congolense parasites, parasites of both strains of T. brucei brucei remained motile in all of the TNF-␣-containing media at 12 h after cultures were initiated. Our results confirm a previous observation that exogenously

INFECT. IMMUN.

FIG. 2. Effects of recombinant mouse, human, and bovine TNF-␣ on the growth of T. congolense IL1180 and IL3338 under in vitro culture conditions. The parasites were suspended in the growth medium at 104 parasites per ml and seeded into a 96-well plate (100 ␮l per well in triplicate), with or without various concentrations of recombinant TNF-␣. After culture at 34°C in humidified 4.5% CO2–95.5% air for 72 to 97 h, numbers of live parasites per milliliter were counted with a hemocytometer. Data are mean values ⫾ SEM determined from three culture wells. Statistical differences were determined by analysis of variance.

added recombinant TNF-␣ does not inhibit the growth of T. brucei brucei cultured in vitro (1), and they extend this earlier observation by showing that supraphysiological concentrations of recombinant TNF-␣ from three mammalian species do not directly kill, or affect the growth of, bloodstream stage T. congolense and T. brucei brucei under axenic culture conditions. To our knowledge, this is the first demonstration that recombinant TNF-␣ does not affect the growth of bloodstream forms of T. congolense cultured in vitro. In contrast to our observations, the observations of other investigators have been that recombinant mouse TNF-␣ can directly affect trypanosomes. Studies have found that it lyses T. brucei brucei AnTat1.1 suspended in phosphate-buffered saline supplemented with 1% glucose (PBSG) and 1% normal mouse serum (13, 15) and that it also lyses T. brucei gambiense OK suspended in culture medium (4). In the case of T. brucei brucei AnTat1.1, approximately half of the parasites were lysed at 1,000 U of murine recombinant TNF-␣/ml of medium after 5 h of incubation in PBSG at 30°C (15). In this study, neither the growth nor the morphological

VOL. 70, 2002

FIG. 3. Effects of recombinant mouse, human, and bovine TNF-␣ on the growth of T. brucei brucei GUTat3.1 and ILTat1.1 under in vitro culture conditions. The parasites were suspended in the growth medium at 104 parasites per ml and seeded into a 96-well plate (100 ␮l per well in triplicate), with or without various concentrations of recombinant TNF-␣. After culture at 34°C in humidified 4.5% CO2–95.5% air for 62 to 72 h, numbers of live parasites per milliliter were counted with a hemocytometer. Data are mean values ⫾ SEM determined from three culture wells. Significant differences at P values of ⬍0.01 (ⴱⴱ) and ⬍0.001 (ⴱⴱⴱ) from control groups are indicated, as determined by analysis of variance.

integrity of T. congolense and T. brucei brucei parasites was impaired by mouse and bovine TNF-␣ at levels up to 3,000 U per ml and by human TNF-␣ at levels up to 15,000 U per ml, although the recombinant TNF-␣ retained all or most of its biological activity in killing L929 cells after dilution in the trypanosome culture media. The divergent results obtained for the lysis of African trypanosomes by recombinant TNF-␣ may reflect differences in the trypanosome species and strains, in the physiological conditions of test trypanosomes, in the recombinant TNF-␣ preparations, or in assay conditions, or they may be due to combinations of these factors. In regard to species and strain differences, T. brucei brucei AnTat1.1 and T. brucei gambiense OK may be particularly sensitive to recombinant murine TNF-␣ in vitro while the T. congolense IL1180 and IL3338 and T. brucei brucei GUTat3.1 and ILTat1.1 parasites used in this study may not be. It has been reported that TNF-␣-mediated lysis of T. brucei brucei occurs only when the parasites are isolated during the peak of parasitemia, not in early stages (15). All the trypanosomes tested in this study were maintained

NOTES

2213

under continuous passage every 3 days prior to their testing, so most of them were in the proliferative stage when they were exposed to TNF-␣. Thus, it is possible that these in vitromaintained trypanosomes are refractory to TNF-␣-mediated lysis because of their growth stage. However, the cultured trypanosomes do accumulate in the G1 stage of their cell division cycle when peak levels of parasites are reached in vitro, and these growth-arrested organisms die during the ensuing 25 to 30 h. Consequently, if TNF-␣ selectively affects trypanosomes at this stage of growth, it would have been expected to accelerate the deaths of the G1-arrested parasites. It did not do so. The peak levels of parasites obtained in vitro (Fig. 2) and the survival of the organisms at peak parasitosis of the medium were not affected (data not shown). The interaction of recombinant murine TNF-␣ with T. brucei brucei AnTat1.1 is via a lectin-like domain that reacts with conserved sugar moieties of variant surface glycoprotein that become exposed in a parasite’s flagellar pocket (17). The lectin-like domain of TNF-␣ is spatially and functionally distinct from the TNF-␣ receptor-binding domain which mediates tumor cell killing (13, 15). It is therefore possible that the four strains of trypanosomes studied here lack conserved TNF-␣binding moieties, that these do not become exposed in their flagellar pockets, or that the trypanolytic, but not the tumoricidal, domain of recombinant TNF-␣ is selectively inactivated by the culture conditions tested. While we have not excluded these possibilities, we consider them unlikely. The variant surface glycoproteins of the cultured parasites were glycosylated as determined by Western blotting with concanavalin A, and there is no reason to consider the possibility that culture medium inactivates the trypanolytic activity of TNF-␣, as it did not do so in the case of T. brucei gambiense OK (4). Native TNF-␣ exists as several differentially glycosylated isoforms (19), whereas the recombinant material is not glycosylated. Thus, there remain the possibilities that native TNF-␣ affects the parasites tested here but that recombinant cytokine does not. Medium collected from mouse spleen cells stimulated with lipopolysaccharide did not show lysis of T. brucei brucei (1). However, while lipopolysaccharide is known to be a potent stimulator of TNF-␣ production by macrophages, the TNF-␣ levels were not measured in that study. Magez et al. (17) have demonstrated ex vivo trypanolysis by TNF-␣ released from the peritoneal exudate cells of mice infected with T. brucei brucei, which suggests that native mouse TNF-␣ has trypanolytic activity at a physiological concentration. However, a number of investigators have shown that these activated macrophages also produce trypanodestructive molecules (3, 5, 6), so synergy may be required between several macrophage products to kill the parasites. There are no data to exclude the possibility that native TNF-␣ is more potently trypanolytic than recombinant TNF-␣, but further comparisons are necessary to test this. Lastly, we would like to draw attention to differences in assay conditions. The direct lytic effect of recombinant TNF-␣ on T. brucei parasites was first reported for organisms in PBSG at pH 8.0 (13, 15) and later for organisms in culture medium (4). T. brucei parasites may survive for certain periods in PBSG, but they are likely to be stressed. In our experience, prolonged culture in PBSG results in limited replication, a loss of population growth potential, and death in less than 24 h. T.

2214

NOTES

INFECT. IMMUN.

congolense IL1180 parasites are even more sensitive to culture conditions than T. brucei brucei. They die within a few hours of incubation at 34°C in PBSG. In this study, complete trypanosome growth medium was used to ensure that the parasites were as metabolically active and unstressed as possible. Under these culture conditions, recombinant TNF-␣ from different mammalian species generally did not affect the growth of Salivarian trypanosomes, which suggests that TNF-␣ may not act directly on metabolically active and healthy trypanosomes. In consideration of these observations, the direct lytic effect of TNF-␣ on Salivarian trypanosomes should be investigated further with different trypanosome species and isolates, during different proliferation stages, and with the use of recombinant as well as native TNF-␣ from different mammalian species. Such comparative studies will contribute to a clearer understanding of how TNF-␣ affects trypanosome parasitemia in infected mammals.

14.

We thank John Wando for his excellent technical assistance. This study was conducted under the Collaborative Research Project between JIRCAS and ILRI.

15.

7. 8. 9. 10. 11. 12. 13.

REFERENCES 1. Black, S. J., M. Murray, S. Z. Shapiro, R. Kaminsky, N. K. Borowy, R. Musanga, and F. Otieno-Omondi. 1989. Analysis of Propionibacterium acnesinduced non-specific immunity to Trypanosoma brucei in mice. Parasite Immunol. 11:371–383. 2. Darji, A., A. Beschin, M. Sileghem, H. Heremans, L. Brys, and P. De Bactselier. 1996. In vitro simulation of immunosuppression caused by Trypanosoma brucei: active involvement of gamma interferon and tumor necrosis factor in the pathway of suppression. Infect. Immun. 64:1937–1943. 3. Daulouede, P. S., M. C. Okomo-Assoumou, M. Labassa, C. Fouquet, and P. Vincendeau. 1994. Defense mechanisms in trypanosomiasis. Bull. Soc. Pathol. Exot. 87:330–332. 4. Daulouede, S., B. Bouteille, D. Moynet, P. De Baetselier, P. Courtois, J. L. Lemesre, A. Buguet, R. Cespuglio, and P. Vincendeau. 2001. Human macrophage tumor necrosis factor (TNF)-␣ production induced by Trypanosoma brucei gambiense and the role of TNF-␣ in parasite control. J. Infect. Dis. 183:988–991. 5. Gobert, A. P., S. Daulouede, M. Lepoivre, J. L. Boucher, B. Bouteille, A. Buguet, R. Cespuglio, B. Veyret, and P. Vincendeau. 2000. L-Arginine availability modulates local nitric oxide production and parasite killing in experimental trypanosomiasis. Infect. Immun. 68:4653–4657. 6. Gobert, A. P., S. Semballa, S. Daulouede, S. Lesthelle, M. Taxile, B. Veyret,

Editor: J. M. Mansfield

16. 17.

18. 19. 20. 21. 22.

and P. Vincendeau. 1998. Murine macrophages use oxygen- and nitric oxidedependent mechanisms to synthesize S-nitroso-albumin and to kill extracellular trypanosomes. Infect. Immun. 66:4068–4072. Hirumi, H., and K. Hirumi. 1994. Axenic culture of African trypanosome bloodstream forms. Parasitol. Today 10:80–84. Hirumi, H., and K. Hirumi. 1991. In vitro cultivation of Trypanosoma congolense bloodstream forms in the absence of feeder cell layers. Parasitology 102:225–236. Iraqi, F., S. J. Clapcott, P. Kumari, C. S. Haley, S. J. Kemp, and A. J. Teale. 2000. Fine mapping of trypanosomiasis resistance loci in murine advanced intercross lines. Mamm. Genome 11:645–648. Iraqi, F., K. Sckikawa, J. Rowlands, and A. Teale. 2001. Susceptibility of tumour necrosis factor-alpha genetically deficient mice to Trypanosoma congolense infection. Parasite Immunol. 23:445–451. Kemp, S. J., A. Darvasi, M. Soller, and A. J. Teale. 1996. Genetic control of resistance to trypanosomiasis. Vet. Immunol. Immunopathol. 54:239–243. Kemp, S. J., F. Iraqi, A. Darvasi, M. Soller, and A. J. Teale. 1997. Localization of genes controlling resistance to trypanosomiasis in mice. Nat. Genet. 16:194–196. Lucas, R., S. Magez, R. De Leys, L. Fransen, J. P. Scheerlinck, M. Rampelberg, E. Sablon, and P. De Baetselier. 1994. Mapping the lectin-like activity of tumor necrosis factor. Science 263:814–817. Lucas, R., S. Magez, B. Songa, A. Darji, R. Hamers, and P. De Baetselier. 1993. A role for TNF during African trypanosomiasis: involvement in parasite control, immunosuppression and pathology. Res. Immunol. 144:370– 376. Magez, S., M. Geuskens, A. Beschin, H. del Favero, H. Verschueren, R. Lucas, E. Pays, and P. De Baetselier. 1997. Specific uptake of tumor necrosis factor-␣ is involved in growth control of Trypanosoma brucei. J. Cell Biol. 137:715–727. Magez, S., M. Radwanska, A. Beschin, K. Sekikawa, and P. De Baetselier. 1999. Tumor necrosis factor alpha is a key mediator in the regulation of experimental Trypanosoma brucei infections. Infect. Immun. 67:3128–3132. Magez, S., M. Radwanska, B. Stijlemans, H. V. Xong, E. Pays, and P. De Baetselier. 2001. A conserved flagellar pocket exposed high mannose moiety is used by African trypanosomes as a host cytokine binding molecule. J. Biol. Chem. 276:33458–33464. Shahan, T. A., P. D. Siegel, W. G. Sorenson, W. G. Kuschner, and D. M. Lewis. 1994. A sensitive new bioassay for tumor necrosis factor. J. Immunol. Methods 175:181–187. Sherry, B., D.-M. Jue, A. Zentella, and A. Cerami. 1990. Characterization of high molecular weight glycosylated forms of murine tumor necrosis factor. Biochem. Biophys. Res. Commun. 173:1072–1078. Taniguchi, T., M. Takata, A. Ikeda, E. Momotani, and K. Sekikawa. 1997. Failure of germinal center formation and impairment of response to endotoxin in tumor necrosis factor ␣-deficient mice. Lab. Investig. 77:647–658. Titus, R. G., B. Sherry, and A. Cerami. 1991. The involvement of TNF, IL-1 and IL-6 in the immune response to protozoan parasites. Immunol. Today 12:A13–A16. Udaka, S., and H. Yamagata. 1993. High-level secretion of heterologous proteins by Bacillus brevis. Methods Enzymol. 217:23–33.