Received 19 August 1996/Returned for modification 30 September 1996/Accepted 19 February 1997

Transcription

1 INFECTION AND IMMUNITY, May 1997, p Vol. 65, No /97/$ Copyright 1997, American Society for Microbiology Experimental Infection of a Nonhuman Primate with Loa loa Induces Transient Strong Immune Activation Followed by Peripheral Unresponsiveness of Helper T Cells ERIC LEROY,* SYLVAIN BAIZE, GOETZ WAHL, THOMAS G. EGWANG, AND ALAIN J. GEORGES Centre International de Recherches Médicales de Franceville, Franceville, Gabon Received 19 August 1996/Returned for modification 30 September 1996/Accepted 19 February 1997 Human infection with the parasite Loa loa is characterized by a good adaptation between the parasite and its host. One portion of the human population harbors only adult worms in subcutaneous tissues, whereas another portion also harbors the L1 microfilarial stage in peripheral circulation. This study was undertaken to understand the mechanisms by which the parasite evades or modulates host immunological attack. The cellular responses, based on T-cell proliferation, to the production of various cytokines (interleukin-2 [IL-2], gamma interferon [IFN- ], IL-4, and IL-5) and to expression of cytokine (IL-2, IFN-, IL-4, IL-5, IL-10, and IL-12) mrnas were investigated during the experimental infection with human parasite L. loa of a nonhuman primate which has been shown to display a spectrum of disease similar to that found in humans. Our results indicate that a T-cell unresponsiveness occurs when female worm products are released into the peripheral circulation, preceded by a transient period of strong T-cell proliferation, cytokine production, and cytokine mrna expression. In the unresponsive state, only IL-10 mrna is expressed, suggesting a role for IL-10 in down-regulation and maintenance of unresponsiveness. Taken together, these results indicate that both IL-10 production, which is known to inhibit B7 expression on monocytes, and the massive release of female products in the blood where T cells encounter antigens presented by nonactivated B lymphocytes, which lack costimulatory signals, should contribute to the inactivation of T cells. * Corresponding author. Mailing address: CIRMF, BP 769, Franceville, Gabon. Phone: (241) Fax: (241) eleroy@cauchy.cirmfrv.fr. Loa loa, a filarial nematode transmitted by species of tabanid flies, infects approximately 10 million people within the regions of the Central West African Forest block where infection is endemic (13). About 30% of affected individuals are microfilaremic (mf ), and twice as many harbor adult worms without showing patent microfilaremia (mf ). Although not usually associated with severe clinical symptoms, L. loa infestation causes a mild to moderate pathology, consisting of pruritis, localized angioedema, arthralgia, or ocular problems caused by subconjunctival migration of adult worms (32). Individuals with parasitologically occult loiasis did not differ from mf subjects, either in the severity of symptoms or in eosinophil counts, which were always high. The immunological status of these two groups has not been extensively characterized, and the cellular mechanisms which control microfilaremia (MF) are still unknown. Previous studies have shown that opsonizing antibodies (immunoglobulins G and E) (34), which mediate adherence of microfilariae (mf) to leukocytes (mainly neutrophils and eosinophils), were found in significantly higher levels in mf persons (33), indicating that antibody-mediated cytotoxicity could represent a major mechanism controlling MF. The immune mechanisms activated during infection with other filariae have been better characterized. A number of observations suggest that the immune response to filarial parasites, after initial exposure, is activated during the prepatency phase and is subsequently modulated when mf appear in the blood (patency phase). In filariasis due to Brugia pahangi, Brugia malayi, and Onchocerca volvulus (3, 43), the patent period is associated with a depressed T-cell proliferation and production of interleukin-2 (IL-2) in response to specific parasite antigen (Ag). Recent reports suggest that in human onchocerciasis, ivermectin eliminates mf from the skin and concomitantly induces reactivation of parasite-specific cellular immune response and a sustained production of IL-2 (44). Taken together, these data suggest that the assimilation of these parasites by their host may be mediated by a state of induced immunological unresponsiveness, or tolerance. Despite the selective inhibition of the T-cell subset in filariasis, the antibody responses remain intact, which has been interpreted as Th1- subset anergy (24). Research on the L. loa parasite has been hampered by the lack of a suitable animal model. Natural infection occurs only in humans; however, the mandrill (Mandrillus sphinx) can be experimentally infected with human L. loa, and the resulting pattern of infection, with different parasitic status, closely resembles that seen in humans (35). Using this primate model, we have examined the early cellular responses after inoculation of L. loa infective third-stage larvae (L3). For this purpose, we have assayed the in vitrospecific production of gamma interferon (IFN- ) and IL-2 (to detect Th1 functions), as well as IL-4 and IL-5 (measures of Th2 activation) and lymphoproliferative responses, to examine the implication of mf in the modulation of the host immune system leading to the survival of the parasite in the host. To confirm the data obtained by enzyme-linked immunosorbent assay (ELISA) and lymphoproliferative assays, and to obtain a more complete description of the cytokines produced during the immune responses, we also analyzed cytokine (IL-2, IL-4, IL-5, IL-10, IL-12, and IFN- ) mrna expression by reverse transcriptase-mediated PCR (RT-PCR) assays. Our results show that microfilaremic mandrills develop a hyperresponsiveness during early prepatency, with increases in both lymphocyte proliferation and IL-2 and IL-5 production when peripheral blood mononuclear cells (PBMC) are stimulated by mf Ag and expression of IL-2, IL-5, IL-10, and IL-12 mrnas. In addition, this study indicates that a T-cell unre- 1876

2 VOL. 65, 1997 INFECTION OF NONHUMAN PRIMATE WITH L. LOA 1877 sponsiveness occurred concurrent with the appearance of mf in blood (suppression of lymphocyte proliferation, cytokine production, and cytokine mrna expression), whereas IL-10 mrna expression was unaffected. This finding suggests that IL-10 contributes to the induction and maintenance of T-cell unresponsiveness. Mandrill TABLE 1. Parasitological data for mandrills that received freshly isolated L3 L. loa Maternal status L3 inoculum Day of patency MATERIALS AND METHODS Source and maintenance of mandrills. The seven mandrills (3 to 6 years old) used were raised and housed at the CIRMF Primate Centre. Mandrills 5A2 and 5A3 were born to a mandrill infected with human L. loa mf. During the 24 months of the study, all mandrills were kept in groups of five or six animals in open-air enclosures. Before infection, none of the mandrills had abnormal biochemical or hematological parameters. None were found to harbor a naturally acquired infection with simian L. loa mf. The mandrills were fed twice daily on a mixed diet of local fruits and vegetables, supplemented by protein biscuits, with water available ad libitum. Infection of mandrills. Mandrills 5A2 and 12I were inoculated subcutaneously with 250 freshly infective L3, and mandrill 5A3 was inoculated with 25 L3 every month for 4 months. Mandrills 5F and 10A1 received 1,000 L3 in one infection, and mandrills 24 and 16C received one infection of 100 L3. These viable L3 were obtained via the Baermann technique (37) and came from wild-caught Chrysops silacea. Parasitologic and clinical examinations. Blood samples were collected from the femoral vein into Vacutainers containing EDTA under ketamine (10 mg/kg) anaesthesia every 1 or 4 weeks. MF was assessed in triplicate on 0.5-ml samples, using a modified Knott s technique (4). The prepatent period was estimated to be the midpoint between the last negative sample and the first showing of mf (10). MF is expressed as mf/milliliter. At the same time, blood was collected for determination of the hematocrit and leukocyte count and for further immunologic studies. Plasma was stored at 80 C in 200- l aliquots. Ag preparation. Ag preparations were made from mf obtained (i) from heavily microfilaremic human patient donors, purified on Percoll gradients and retained on 5- m-pore size filters (mf Ag), (ii) from L3, obtained from C. silicea by using Baermann s technique (L3 Ag), and (iii) from adult worms (AW Ag), surgically removed from patients eyes. The parasites were fragmented in phosphatebuffered saline and homogenized on ice. This preparation was then centrifuged (4 C, 1,300 g, 20 min). The final supernatant was filtered through a 0.2- mpore-size filter (Millex; Millipore, Molsheim, France), and the protein concentration was determined with a MicroBCA kit (Pierce, Rockford, Ill.). Separation of PBMC. PBMC were separated from whole blood by using Histopaque-1077 (Sigma Diagnostics, La Verpillière, France) sedimentation. After centrifugation, cells were washed twice in RPMI 1640 with gentamicin (10 g/ml). Cells were then resuspended at a concentration of 10 6 /ml (for proliferation assay) or /ml (for ELISA and RT-PCR assays) in RPMI 1640 with gentamicin (10 g/ml) supplemented with 25 mm HEPES buffer, 2 mm L-glutamine, 1% nonessential amino acids, 1 mm sodium pyruvate (Gibco BRL, Cergy-Pontoise, France), and 10% human AB serum (CNTS, France) for cytokine assay. In vitro lymphocyte proliferation assay. Mitogen- or Ag-induced in vitro lymphocyte proliferation was performed in flat- or round-bottom microtiter plates (Nunc, Roskilde, Denmark), respectively, using a total volume of 200 l. PBMC were stimulated for 3 days with concanavalin A (10 g/ml; Sigma Chemical Co., St. Louis, Mo.), phytohemagglutinin (PHA; 10 g/ml; Sepracor, Villeneuve-la-Garenne, France), or medium alone and for 6 days with mf Ag (5 g/ml), L3 Ag (2.5 g/ml), AW Ag (2.5 g/ml), or medium alone. Plates were incubated at 37 C in an atmosphere of 5% CO 2. The cells were then pulsed for 10 h with 1 Ci of [ 3 H]thymidine (Amersham, Les Ulis, France) in 10 l of medium per well and harvested onto a paper filter. The incorporation of [ 3 H]thymidine was measured by liquid scintillation counting. Data were expressed as the stimulation index obtained by dividing the mean counts per minute of triplicate cultures for Ag- or mitogen-stimulated cultures by the mean counts per minute for unstimulated control cultures. Cytokine assay. Cells were cultured in 24-well plates in RPMI 1640 (as described above) with PHA (10 g/ml), mf Ag (10 g/ml), L3 Ag (5 g/ml), AW Ag (5 g/ml), or medium alone. Culture supernatants were collected at 24 h (PHA) and 48 h (Ag) for determination of IL-2 and IL-4 and at 48 h (PHA) and 92 h (Ag) for IFN- and IL-5 assays. Levels of IL-2, IL-4, and IL-5 in culture medium were measured by two-site ELISA. Wells of 96-well microtiter plates (Maxisorp; Nunc) were incubated overnight at 4 C with coating buffer (0.1 M bicarbonate, ph 8.5) with 2.5 g of mouse anti-il-2 monoclonal antibody (MAb) (Genzyme, Belgium), 4 g of mouse anti-il-4 MAb 8D4 (Pharmingen, San Diego, Calif.), and 2 g of mouse anti-il-5 MAb TRFH5 (Pharmingen) per ml. After a 1-h blocking step with 3% bovine serum albumin (Sigma) at room temperature, test culture supernatants were incubated for2hat37 C (IL-2) and overnight at 4 C (IL-4 and IL-5). Wells were then incubated for 1hat37 C with dilution buffer (0.1% bovine serum albumin in phosphate-buffered saline) with 1.25 g of rabbit anti-il-2 polyclonal antibody (Genzyme), 2 g of rat anti-il-4 MAb MP4 (Pharmingen), and 1 g of rat anti-il-5 MAb JES1 (Pharmingen) per ml, followed by streptavidin-peroxidase conjugate (0.5 g/ml; Extravidin; Sigma) for 20 min at 37 C and o-phenylenediamine substrate (Sigma). The optical density was measured at 492 nm, using an ELISA reader (Diagnostics Pasteur, France). Levels of IFN- were assessed by using commercial reagents (R&D Systems Europe, England) as instructed by the manufacturer. The results are expressed as picograms per milliliter (IFN-, IL-4, and IL-5) or units per milliliter (IL-2), using standards with known cytokine concentrations. Sensitivities of the different assays were 0.06 U/ml for IL-2, 10 pg/ml for IL-4 and IL-5, and 3 pg/ml for IFN-. Isolation and purification of mrna. RNase-free plastic labware and water were used throughout the assay. By using an RNA purification kit (Qiagen, Hilden, Germany), RNA was isolated from PBMC collected from the mandrills during preinfection, prepatency, and microfilaremic phases of infection. RNA was resuspended in diethylpyrocarbonate-treated water containing 1 mm EDTA and quantitated spectrophotometrically. Gel electrophoresis was then performed to confirm that the RNA was intact. RT-PCR detection of cytokine mrna. RT-PCR was performed to determine the relative quantities of IL-2, IL-4, IL-5, IL-10, IL-12, IFN-, and -actin mrnas. Reverse transcription of RNA was performed with 200 ng of RNA in a 20- l final volume containing 5 l of5 reverse transcription buffer (Gibco BRL), 1 l of RNase inhibitor (Boehringer, Mannheim, Germany); 5 mm dithiothreitol (Gibco BRL); 1 M oligo(dt) 15 (Boehringer), 500 M (each) datp, dctp, dgtp, and dttp (USB, Amersham), and 200 U of reverse transcriptase (Gibco BRL). The reaction mixture was incubated at 42 C for 90 min and heated to 65 C for 5 min to denature the reverse transcriptase. The following pairs of 5 and 3 oligonucleotide primers were synthesized by the Genset (Paris, France) lab: IL-2, TACAACTGGAGCATTTACTG and GTTTCAGATCCCT TTAGTTC (268-bp product); IL-4, GCCTCACAGAGCAGAAGACT (344-bp product) and TCAGCTCGAACACTTTGAAT; IL-5, CTGATAGCCAATGA GACTCT and TATTATCCACTCGGTGTTCA (250-bp product); IL-10, TGC CTGGTCCTCCTGACTGG and GCCTTGCTCTTGTTTTCACA (389-bp product); IL-12, CGGATGCCCCTGGAGAAATG and CTCTTGCCCTGGAC CTGAAC (727-bp product); IFN-, TGGAAAGAGGAGAGTGAC and ATT GCTTTGCGTTGGACA (224-bp product); and -actin, CAGGCACCAGGG CGTGAT and GCCAGCCAGGTCCAGACG (500-bp product). All PCRs were carried out in a thermocycler (Perkin Elmer, Paris, France), using a 50- l incubation volume containing 5 lof10 PCR buffer (Appligene-Oncor, Illrich, France), 200 M each deoxynucleoside triphosphate, 0.5 U of Taq polymerase (Appligene), 0.4 M each 5 and 3 oligonucleotide primer, and 1 l of cdna, for 40 cycles (cytokines) or 30 cycles ( -actin). Each cycle consisted of 94 C for 1 min, with an annealing temperature of 72 C for 2 min. Both positive and negative controls were included in each run to confirm that only cdna PCR products were detected and that no contamination with cdna or previous PCR products had occurred. Annealing temperatures were optimized for each cytokine: IL-2 (52 C), IL-4 (50 C), IL-5 (50 C), IL-10 (60 C), IL-12 (60 C), IFN- (58 C), and -actin (62 C). RESULTS Day of mf peak MF at the mf peak (mf/ ml) 5A2 mf A3 mf F mf 1, ,000 10A1 mf 1,000 12I mf C mf mf ,300 Parasitologic, serologic, and clinical manifestations. As shown in Table 1, six of the seven mandrills developed patent L. loa infection, with onset of patency (defined above) occurring around at 150 days (5F, 24, 12I, and 16C [mandrills born to nonmicrofilaremic mothers]) or 200 days (5A2 and 5A3 [mandrills born to a microfilaremic mother]) postinoculation (Fig. 1). In one of the seven infected animals (10A1), no mf were detectable at any time during the course of infection. The number of circulating mf showed a first peak around day 200 (5F, 24, 12I, and 16C) to day 225 (5A2 and 5A3) which varied from 20 mf/ml (12I and 16C) to 100 (5A3), 500 (5A2), 2,000

3 1878 LEROY ET AL. INFECT. IMMUN. FIG. 1. Microfilaremia in three representative mandrills that received infective L3 of L. loa on day 0. Mandrill 10A1 is amicrofilaremic. Mandrill 24 developed a typical pattern of microfilaremia which closely resembles that seen in 5F, 12I, and 16C; these animals differ only in the number of circulating mf. Mandrill 5A2 developed a pattern of microfilaremia similar to that of 5A3 after a longer prepatent period than was seen in the other mandrills. (24), and 10,000 (5F) mf/ml; the microfilaremia subsequently decreased to 10 to 100 mf/ml and remained at stable levels. At no time during the study were significant leukocytosis, eosinophilia, or other clinical symptoms noticed. Proliferative responses of PBMC to Ag and PHA. The proliferative responses to PHA did not show any significant changes during the 12 months of the infection. All of the cell preparations remained highly responsive to PHA throughout the study (data not shown). As shown in Fig. 2, the PBMC proliferative responses to mf Ag and to AW Ag followed similar patterns. Cell proliferation was enhanced during the prepatent period, reaching a peak around 100 days after inoculation with L3, corresponding to sexual maturity of the adult worm with laying of the first mf in subcutaneous tissues and secretion and excretion of circulating filarial products. The response then declined sharply at the end of the prepatent period to a moderately elevated level similar to that of the mf animal. Interestingly, neither mf Ag nor AW Ag could stimulate T-cell proliferation at 80 days in the mf animal compared to the other mandrills. Moreover, the moderate peak occurred later during the course of infection. Finally, PBMC from all seven animals were consistently unable to respond to L3 Ag. Kinetics of cytokine production during the course of infection. The Th1 and Th2 cytokine responses to different antigens (mf Ag, L3 Ag, and AW Ag) were compared by assessing both IFN- and IL-2 levels and IL-4 and IL-5 levels. The secretion patterns were found to change significantly with time after exposure to AW products. As shown in Fig. 3, cells from mf animals produced similar patterns of cytokine response to mf Ag. This production increased during the prepatent period and reached a peak level at around 80 days after inoculation. At this time, increases in IL-2 and mainly IL-5 responses, compared to those seen for IFN- and IL-4, are evident. In these instances, maximal mf Ag-specific levels of cytokines detected were of the order of 400 to 1,600 pg of IL-5 per ml, 200 to 800 pg of IL-2 per ml, 50 to 100 pg of IFN- per ml, and low levels of IL-4. These responses declined thereafter and appeared to be totally absent as soon as the mf were detectable in the peripheral circulation. Interestingly, Fig. 3 shows a correlation between the magnitude of IL-5 and IL-2 secretion during the prepatent period (significantly higher levels in 5A2 and 5A3) and the time when mf appeared in the circulation (day 200 for 5A2 and 5A3, compared to day 150 for the other mf animals). As shown in Fig. 4, a markedly different and subtle inverse response pattern was observed for the mf mandrill. Although IL-5 rose during the first weeks after the inoculation with L3, this cytokine continued to be secreted later during infection. Moreover, IL-2 secretion tended to increase, reaching levels of 700 pg/ml, while IL-5 production tended to decrease and finally remained at a low level (200 pg/ml) over the same period. A similar pattern of increasing and decreasing response to AW Ag was observed in the IL-5 production of the same cell populations for the mf animals. Secretion of the other cytokines remained at low levels throughout the study (data not shown). In contrast, for mandrill 10A1, IL-2, IL-4, IL-5, and IFN- were produced by the cultured PBMC with AW Ag. In both mf and mf mandrills, the data show the inability FIG. 2. In vitro lymphocyte proliferation in response to mf antigens. PBMC were obtained from mf mandrills born to microfilaremic (mf MM [5A2 and 5A3]) or noninfected (mf MN [5F, 24, 12I, and 16C]) mothers or obtained from mf mandrill 10A1 before infection (PI [preinfection]) and during infection (PP [prepatency]; MF1 [day 100] and MF2 [day 200]) with L. loa. The degree of lymphocyte stimulation is expressed as described in the text. Results represent the mean stimulation index for each group of animals. Standard deviations of the mf animals are indicated by the error bars.

4 VOL. 65, 1997 INFECTION OF NONHUMAN PRIMATE WITH L. LOA 1879 Downloaded from FIG. 3. Kinetics of lymphocyte cytokine responses before and during infection with human L. loa of two mf mandrills (5A2 and 5A3) born to a microfilaremic mother and four other mf mandrills (5F, 24, 12I, and 16C) born to noninfected mothers. Data shown are the amounts of IL-2 and lymphokines IFN-, IL-4, and IL-5 present in supernatants after mf Ag stimulation of PBMC. The data are the means of duplicate assays consisting of two ELISAs. of L3 antigen preparations to stimulate cytokine production by lymphocytes (data not shown). PBMC cytokine responses to PHA. Cytokine levels in the supernatant of PHA-stimulated cells were very high throughout the study, and the levels of cytokines measured were greater than those seen after antigen stimulation, except for the secretion of IL-5 (data not shown). Expression of cytokine mrna by noncultured PBMC. Cytokine mrna expression in noncultured PBMC was studied by reverse RT-PCR. Cytokine mrna expression levels were normalized on the basis of the level of mandrill -actin mrna expression. The results revealed a pattern of cytokine mrna expression consistent with the in vitro lymphocyte proliferation and cytokine production studies. As shown in Fig. 5, significantly elevated IL-2 and IL-5 mrna expression was observed in the PBMC of all mandrills at the prepatent time, while IL-4 and IFN- mrna expression was weak. Thereafter, consistent with ELISA measurements of cytokines, only low levels of mrna expression of these four cytokines were found when mf were present in the circulation. In addition, IL-10 and IL-12 mrna expression was also examined. As was the case with other cytokine mrnas, IL-12 transcripts were significantly expressed at the prepatent point and returned to near the baseline level with the appearance of mf in the blood. Moreover, as shown in Fig. 5, IL-10 mrna expression remained elevated throughout the course of infection even when mf were present in the blood. DISCUSSION Mammalian CD4 Th cells can be divided into two distinct subsets (23, 30), designated Th1 and Th2 cells, that mediate their effector functions through the production of distinct cytokines. Development of an appropriate Th cell response is critical for the outcome of infection. Although previous studies have reported a global Th2 response in resistant patients who have chronic helminth infections (19) or in experimentally on January 16, 2019 by guest

5 1880 LEROY ET AL. INFECT. IMMUN. infected animals, the T helper status may not be irrevocable during the course of infection. In loiasis in particular and in other filarial infections in general, there has been little investigation of the underlying cellular mechanisms, since very few animal models have been available. In the present study, the L. loa-mandrill model has allowed us, for the first time, to assess the roles of key Th1 (IL-2, IFN-g, and IL-12) and Th2 (IL-4, FIG. 5. Pattern of cytokine mrna expression in the PBMC of mandrills infected with L. loa. Total mrna was isolated from PBMC and assayed for expression of cytokine mrna by RT-PCR. PCR products were analyzed by electrophoresis on 1.5% agarose gels and visualized by ethidium bromide staining. PBMC were obtained from mf1 mandrills born to microfilaremic (5A2 and 5A3) or noninfected (5F and 16C) mothers or obtained from mf2 mandrill 10A1 before infection (PI [preinfection]) and during infection (PP [prepatency]; MF [microfilaremic phase]) with L. loa. IL-5, and IL-10) cytokines in the development of immunity to a filarial parasite. Our findings describe critical aspects of the possible regulatory effects of distinct parasite stages on the outcome of T-cell responses. The parasite apparently elicits a vigorous cellular immune response, shown by strong PBMC proliferation, production of high quantities of IL-2 and mainly IL-5, and elevated cytokine mrna expression. Both responses are poor, even defective, when cells are stimulated with L3 Ag at any time after infection, suggesting that L3 has induced the molts to develop immune responses and that the infective stage is a poor target for host immune mechanisms. In contrast, an enhancement of the response occurred early in the prepatent period, indicating that the later L. loa stages provide the necessary stimulus, which implicated common antigenic epitopes shared by postmolting L3, AW, mf, and excretory/secretory products released by the female worm. Although infectious agents frequently cause polarization of responses to either the Th1 or Th2 pattern, our data indicate clearly that L. loa infection leads to the expression of an atypical pattern of cytokines. In our system, the PBMC simultaneously produced IL-2 and mainly IL-5, which seemed to be independent of IFN-g and IL-4 production. Previous studies have suggested that a Th2-like or Th1-like pattern may be manifested depending on the animal host and the helminth studied. Examples supporting either hypothesis are found in both human and rodent studies, whereas many helminth infections evoke predominantly Th2 responses. The inoculation of BALB/c mice with live B. malayi (25) or with radiation-attenuated L3 of B. pahangi (2) results in the induction of a Th2 subset of cells. Similar observations made in mice infected with other helminthic parasites such as the gut-dwelling nematodes Trichuris muris (10), Trichinella spiralis (13), and Heligmosomoides polygyrus (35) showed that an early Th2-restricted cytokine pattern enhanced expulsion of the worms. Likewise, in rat (14) and human (15) schistosomiasis, antibody-dependent cell cytotoxicity, indicative of Th2 responses, is the main mechanism for killing parasite larvae. Nevertheless, the phenotype observed in this study seems to delay the onset of MF since elevated levels of these cytokines appear to correlate with increased levels of resistance (high amounts of IL-5 and IL-2 associated with a longer prepatent period for 5A2 and 5A3). The highest cellular responsiveness was observed in both of the mandrills born to a microfilaremic mother. This result strongly suggests that a transfer of mf may have occurred across the placenta, leading to prenatal sensiti- FIG. 4. Kinetics of lymphocyte cytokine responses before and during infection with L. loa of the mf2 (10A1) mandrill. Data shown are the amounts of cytokines produced by PBMC stimulated with mf Ag or AW Ag.

6 VOL. 65, 1997 INFECTION OF NONHUMAN PRIMATE WITH L. LOA 1881 zation, as has been shown with infection of rats by Dipetalonema vitae (16). The production of both IL-2 and IL-5 in the host s defense mechanisms against parasitic helminths is well documented, and IL-5 has been shown to be an important mediator of the eosinophilic responses seen in filariasis (18), schistosomiasis (4), and other helminth infections (5). By contrast, in our study, no eosinophilic response was demonstrated. Surprisingly, our results indicate an apparently paradoxical pattern of cytokine secretion, with high levels of IL-5 and small quantities of IL-4, which is known to support the differentiation of Th0 into Th2 cells (1). This discrepancy is consistent with the direct correlation observed between the production of IL-2 and IL-5. In fact, data have recently emerged regarding the role of IL-2 in the induction of IL-5 secretion in schistosomiasis (21) and in onchocerciasis (34) accompanied by a weak secretion of IL-4, although the mechanisms mediating this direct stimulatory effect of IL-2 remain unknown. This pattern of secretion failed to prevent the onset of patency and the appearance of mf in the peripheral circulation. However, the mf mandrill displayed a response different from that of the others, composed of IL-5, IFN-, and mainly IL-2. This pattern of secretion is associated with the continuing clearance of mf in blood. Furthermore and surprisingly, cellular responsiveness (both lymphocyte proliferation and cytokine secretion) was significantly lower in this animal than in the others. The most striking finding of this study is the marked general down-regulation of immune reactivity in patently mf mandrills, as revealed by L. loa-specific lymphoproliferation and both Th1 and Th2 responses. The beginning of these changes correlates with the formation of the mature worms; further, the total suppression of immune reactivity may be associated with the onset of patency, where it is impossible to say that the cytokine increase affects the decrease in MF or vice versa. The down-regulation of these responses is comparable to that found for gerbils (3) and dogs (31) infected with B. pahangi and for chimpanzees infected with O. volvulus (32), where the parasites suppressed cellular proliferation and IL-2 production in patent animals. Similar results have been found in humans infected with L. loa, where mf status correlates with cell proliferation and both Th1 and Th2 responses, whereas mf status is characterized by a hyporesponsiveness of the same cell population to mf Ag and AW Ag (1a). It is known that a key facet of parasite persistence appears to be the enhancement of down-regulation of the immune system inducing peripheral tolerance or anergy in the face of repeated infections (19). Such a phenomenon should occur in our L. loa-mandrill system, in which a general unresponsive state of T cells has been demonstrated for the first time in parasitic infection. Beside these data, RT-PCR analysis indicates that both IL-10 and IL-12 mrnas are strongly expressed in PBMC of mf mandrills in the prepatent period, while they are weakly expressed in those of the mf mandrill. Furthermore, it is interesting that for mf mandrills, IL-10 mrna was still being strongly expressed at the time when mf appeared in the bloodstream, which is in contrast with the almost complete lack of IFN-, IL-2, IL-4, IL-5, and IL-12 mrna expression. These observations raise some questions, and several hypotheses may be considered in attempting to explain the mechanisms operating during the shutdown of the immune response. Why does a transient cellular reactivity stage lead to T-cell unresponsiveness in mf mandrills and not in the mf animal? This situation is in agreement with recent observations that induction of T-cell anergy in vivo is always preceded by a transient period of proliferation, suggesting that the unresponsive state can be induced only in previously activated cells (24). IL-10 may be involved in this process, since only IL-10 mrna is expressed at this time. Other reports have suggested that IL-10 is an important negative regulator of T-cell responses, inhibiting T-lymphocyte proliferation and cytokine production (36) mainly by suppression of antigen-presenting cell (APC) function (12), although a direct but minor effect on CD4 T cells has been described (8). IL-10 has been shown to inhibit APC function by down-regulating class II major histocompatibility complex antigens on monocytes (7), by selectively downregulating the expression of B7-BB1 (CD80) on macrophages (9), and by blocking the production of IL-12 from macrophages (6). Finally, our results show that IL-12 mrna expression during prepatency was markedly stronger in mf mandrills than in the one mf animal and decreased as mf appeared in peripheral circulation. IL-12, produced essentially by monocytes after pathogen stimulation, enhances growth of T cells, as well as IFN- production by natural killer and T cells (20), and strongly potentiates proliferation and cytokine production by acting synergistically with B7-CD28 interaction (17). Furthermore, IL-12 has been demonstrated recently to directly induce the production of its own inhibitor (IL-10), which down-regulates cellular immune responses initiated by IL-12 and restricts ongoing T-cell activation (22). Such observations may be consistent with our data which clearly show an inverse correlation between IL-12 and IL-10 mrna expression. Under these latter conditions, the massive release of Ag (excretory/secretory products or mf) into the bloodstream to confront predominately nonprofessional APC, as represented by nonactivated B lymphocytes, and the high levels of IL-10 may help to explain the unresponsiveness of T cells. ACKNOWLEDGMENTS We thank S. Evereaere for helpful technical advice and E. J. Wickings for help in writing the manuscript. CIRMF is supported by the Government of Gabon, Elf Gabon, and le Ministère de la Coopération Française. REFERENCES 1. Abehsira-Amar, O., M. Gibert, M. Joliy, J. Thèze, and D. L. Jankovic IL-4 plays a dominant role in the differential development of Th0 into Th1 and Th2 cells. J. Immunol. 148: a.Baize, S. Submitted for publication. 2. Bancroft, A. J., R. K. Grencis, K. J. Else, and E. Devaney Cytokine production in BALB/c mice immunized with radiation attenuated third stage larvae of the filarial nematode, Brugia pahangi. J. Immunol. 150: Bosshardt, S. C., S. U. Coleman, C. S. McVay, K. L. Jones, and T. R. Klei Impact of microfilaremia on maintenance of a hyporesponsive cellular immune response in Brugia-infected gerbils. Infect. Immun. 63: Capron, M., and A. Capron Immunoglobulin E and effector cells in schistosomiasis. Science 264: Coffman, R. L., B. W. P. Seymour, S. Hudak, J. Jackson, and D. Rennick Antibody to interleukin-5 inhibits helminth-induced eosinophilia in mice. Science 245: D Andrea, A., M. Aste-Amezaga, N. M. Valiante, X. Ma, M. Kubin, and G. Trinchieri IL-10 inhibits human lymphocyte IFN -production by suppressing IL-12 synthesis in accessory cells. J. Exp. Med. 178: de Waal Malefyt, R., J. Haanen, H. Spits, M.-G. Roncarolo, A. te Velde, C. Figdor, K. Johnson, R. Kastelein, H. Yssel, and J. E. de Vries IL-10 and viral IL-10 strongly reduce antigen-specific human T cell proliferation by diminishing the antigen-presenting capacity of monocytes via down-regulation of class II major histocompatibility complex expression. J. Exp. Med. 174: de Waal Malefyt, R., H. Yssel, and J. E. de Vries Direct effects of IL-10 on subsets of human CD4 T cell clones and resting T cells. J. Immunol. 150: Ding, L., P. S. Linsley, L.-Y. Huang, R. N. Germain, and E. M. Shevach IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the up-regulation of B7 expression. J. Immunol. 151: Else, K. J., F. D. Finkelman, C. R. Maliszewski, and R. K. Grencis Cytokine-mediated regulation of chronic intestinal helminth infection. J. Exp. Med. 179:

7 1882 LEROY ET AL. INFECT. IMMUN. 11. Fain, A Les problèmes actuels de la loase. Bull. OMS 56(2): Fiorentino, D. F., A. Zlotnik, P. Vieira, T. R. Mosmann, M. Howard, K. W. Moore, and A. O Garra IL-10 acts on the antigen-presenting cell to inhibit cytokine production by Th1 cells. J. Immunol. 146: Grencis, R. K., L. Hültner, and K. J. Else Host protective immunity to Trichinella spiralis in mice: activation of Th cell subsets and lymphokine secretion in mice expressing different response phenotypes. Immunology 74: Grezel, D., M. Capron, J. M. Grzych, J. Fontaine, J. P. Lecocq, and A. Capron Protective immunity induced in rat schistosomiasis by a single dose of Sm28GST recombinant antigen: effector mechanisms involving IgE and IgA antibodies. Eur. J. Immunol. 23: Grzych, J. M., D. Grezel, C. B. Xu, J. L. Neyrinck, M. Capron, J. H. Ouma, A. E. Butterworth, and A. Capron IgA antibodies to a protective antigen in human schistosomiasis mansoni. J. Immunol. 150: Haque, A., and A. Capron Transplacental transfer of rodent microfilariae induces antigen-specific tolerance in rats. Nature 299: Kubin, M., M. Kamoun, and G. Trinchieri IL-12 synergizes with B7/CD28 interaction in inducing efficient proliferation and cytokine production of human T cells. J. Exp. Med. 180: Limaye, A. P., J. S. Abrams, J. E. Silver, E. A. Ottesen, and T. B. Nutman Regulation of parasite-induced eosinophilia: selectively increased interleukin-5 production in helminth-infected patients. J. Exp. Med. 172: Maizels, R. M., D. A. P. Bundy, M. E. Selkirk, D. F. Smith, and R. M. Anderson Immunological modulation and evasion by helminth parasites in human populations. Nature 365: Manetti, R., F. Gerosa, M. G. Giudizi, R. Biagiotti, P. Parronchi, M.-P. Piccinni, S. Sampognaro, E. Maggi, S. Romagnani, and G. Trinchieri IL-12 induces stable priming for IFN production during differentiation of human T helper cells and transient IFN production in established Th2 cell clones. J. Exp. Med. 179: Metwali, A., D. Elliott, R. Mathew, A. Blum, and J. V. Weinstock IL-2 contributes to the IL-5 response in granulomas from mice infected with Schistosoma mansoni. J. Immunol. 150: Meyaard, L., E. Hovenkamp, S. A. Otto, and F. Miedema IL-12- induced IL-10 production by human T cells as a negative feedback for IL-12-induced immune responses. J. Immunol. 156: Mosmann, T., and R. Coffman TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 7: Editor: J. M. Mansfield 24. Mueller, D. L., and M. K. Jenkins Molecular mechanisms underlying functional T-cell unresponsiveness. Cur. Opin. Immunol. 7: Pearlman, E., F. E. Hazlett, W. H. Boom, and J. W. Kazura Induction of murine T-helper-cell responses to the filarial nematode Brugia malayi. Infect. Immun. 61: Pinder, M Loa loa: a neglected filaria. Parasitol. Today 4(10): Pinder, M., A. Dupont, and T. G. Egwang Identification of a surface antigen on Loa loa microfilariae the recognition of which correlates with the amicrofilaremic state in man. J. Immunol. 141: Pinder, M., A. Leclerc, and S. Everaere Antibody-dependent cellmediated immune reactions to Loa loa microfilariae in amicrofilaremic subjects. Parasite Immunol. 14: Pinder, M., S. Everaere, and G. E. Roelants Loa loa: immunological responses during experimental infections in mandrills. Exp. Parasitol. 79: Salgame, P., J. S. Abrams, C. Clayberger, H. Goldstein, R. L. Modlin, and B. R. Bloom Differing lymphokine profiles of functional subsets of human CD4 and CD8 T cell clones. Science 254: Schreuer, D., and B. Hammerberg Modulation of cellular and humoral immunity, and disease manifestation during onset of patency in Brugia pahangi-infected dogs. Immunology 79: Soboslay, P. T., C. M. Dreweck, H. R. Taylor, B. Brotman, P. Wenk, and B. M. Greene Experimental onchocerciasis in chimpanzees. Cellmediated immune responses, and production and effects of IL-1 and IL-2 with Onchocerca volvulus infection. J. Immunol. 147: Soboslay, P. T., C. G. K. Luder, W. H. Hoffmann, I. Michaelis, G. Helling, C. Heuschkel, C. M. Dreweck, C. H. Blanke, S. Pritze, M. Banla, and H. Schulz-Key Ivermectin-facilitated immunity in onchocerciasis; activation of parasite-specific Th1-type responses with subclinical Onchocerca volvulus infection. Clin. Exp. Immunol. 96: Steel, C., and T. B. Nutman Regulation of IL-5 in onchocerciasis. J. Immunol. 150: Svetic, A., K. B. Madden, X. D. Zhou, P. Lu, I. M. Katoma, F. D. Finkelman, J. F. Urban, and W. C. Gause A primary intestinal helminthic infection rapidly induces a gut-associated elevation of Th2-associated cytokines and IL-3. J. Immunol. 150: Taga, K., and G. Tosato IL-10 inhibits human T cell proliferation and IL-2 production. J. Immunol. 148: Wahl, G. Personal communication.