Mice Genetically Deficient in Immunoglobulin E Are More Permissive Hosts than Wild-Type Mice to a Primary, but Not Secondary, Infection with the Filarial Nematode Brugia malayi

Abstract

Primary and secondary murine and human infections with Brugia malayi are characterized by substantial increases in levels of immunoglobulin E (IgE). To investigate whether this is necessary for worm clearance, IgE^−/− mice were subjected to primary- and secondary-infection protocols. Following a primary infection, IgE^−/− mice displayed a profound deficit in their ability to clear an intraperitoneal injection of L3 infective-stage larvae in comparison to wild-type counterparts and maintained substantial worm burdens as late as 10 weeks postinfection. Although viable adult parasites were recovered at this late time point from IgE^−/− mice, the majority of the mice remained free of microfilariae. IgE^−/− cohorts subjected to a secondary-infection protocol were able to clear the challenge inoculation in an accelerated manner, with kinetics similar to that observed in the wild-type animals. Analysis of the humoral response in IgE^−/− mice following infection demonstrates a defect in IgG1 and IgG2a production, in addition to the expected lack of IgE. The IgG1 deficiency is no longer evident following a secondary infection. These data imply that deficiencies other than IgE production (i.e., IgG1 production) deficiency may be responsible for the increased permissiveness of IgE^−/− mice as hosts following infection with B. malayi.

Human lymphatic filariasis is caused by the parasitic nematodes Wuchereria bancrofti, Brugia malayi, and Brugia timori. Patients exhibit enhanced Th2-like responses, accompanied by Th1 nonresponsiveness and substantially elevated levels of the Th2-associated isotypes immunoglobulin G4 (IgG4) and IgE (20, 22, 26, 37). Murine models of lymphatic filariasis have been used extensively to dissect the mammalian response to these nematodes. Immunocompetent mice on several backgrounds are able to clear an intraperitoneal (i.p.) injection of B. malayi prior to the onset of patency, providing an excellent example of a successful mammalian host response to a human-infective parasite. Like human infections, murine infections with B. malayi and its close relative Brugia pahangi are characterized by an increase in the amount of circulating parasite-specific and nonspecific IgE as well as IgG1, the murine counterpart to human IgG4 (2, 29). As of yet, a definitive role has not been established for these antibodies in control of infection.

IgE binds preferentially to the high-affinity FcɛRI, which is expressed on mast cells, basophils, and eosinophils in humans and on mast cells in the mouse. Cross-linking of bound IgE by antigen leads to activation, cytokine production, and degranulation in these target cells (6, 15, 19). IgE has been implicated in the expulsion of nematode parasites from the gut and respiratory tract, in part by enhancement of eosinophil cytotoxic activities in an antibody-dependent cell-mediated cytotoxicity mechanism (4, 18, 30). In vitro data also suggest a role for IgE-mediated killing of helminths (3). Despite a presumed role for IgE in parasitic infections, neutralization of IgE in vivo has not been shown to dramatically affect parasite expulsion. Similarly, in a mouse model of murine filariasis, in vivo neutralization of IgE was found to have no effect on worm clearance abilities (35). The inability to determine the degree of removal of cytophilic IgE is a limitation of this approach. To circumvent this caveat, we have utilized mice with an isolated null mutation of the Cɛ gene encoding the IgE heavy chain constant region domains. These mice failed to produce detectable IgE or ɛ mRNA following lipopolysaccharide stimulation of B cells (24).

Here we revisit the question of in vivo significance of IgE production in host protection against a primary infection with Brugia. In addition, we investigate for the first time a role for IgE in the accelerated clearance seen upon challenge infection.

MATERIALS AND METHODS

Mice.

BALB/c Pkrdc^scid/Pkrdc^scid+/+ and BALB/c By^+/+ animals were obtained from the Jackson Laboratory (Bar Harbor, Maine). BALB/c IgE^−/− mice were originally obtained as a gift from M. Oettgen (Harvard University School of Medicine) and subsequently bred at our facility. All mice used were males between 6 and 12 weeks of age. The SCID phenotype was confirmed through serum Ouchterlony tests. IgE deficiency of IgE^−/− animals was periodically confirmed using an IgE-specific enzyme-linked immunosorbent assay (ELISA) of serum.

Parasite.

B. malayi L3 infective-stage larvae (hereafter referred to simply as L3 larvae) were harvested at the insectarium of Thomas Klei (Louisiana State University, Baton Rouge) from infected Aedes aegypti mosquitoes and shipped overnight in RPMI containing antibiotics and fluconazole. B. pahangi L3 larvae were harvested from infected mosquitoes at the University of Georgia and shipped in a similar manner.

Experimental infection and parasite recovery.

Mice were inoculated with 35 to 50 B. pahangi or B. malayi L3 larvae i.p. using a 5/8-in. 25-gauge needle for a primary infection. For challenge infections, 50 L3 larvae of the same species were injected i.p. into mice sensitized with 35 to 50 L3 larvae 11 weeks previously. Animals were sacrificed at 7 or 14 days postinfection, and viable parasites remaining from initial and challenge infections, easily distinguishable by size, were counted separately. Mice were sacrificed at various time points postinfection and subjected to a cardiac bleed for retrieval of serum. Peritoneal lavages were performed using RPMI medium supplemented with heparin (5 U/ml). At time points 4 weeks and later, lavage fluid was extracted from the peritoneal cavity using a soft plastic pipette to prevent shearing of the adult worms. Following lavage, intestines were removed and soaked in phosphate-buffered saline (PBS). Testes were cut, and carcasses were placed into PBS for further soaking. Carcasses were then rinsed several times with PBS. Viable worms were counted in peritoneal lavage fluid, intestinal wash fluid, and carcass soak fluid under a dissecting microscope.

Analysis of peritoneal exudate cells (PEC).

The cellular profile of lavage fluid was determined using CellQuest software with the FACSCalibur device (Becton Dickinson) to acquire information on cellular size and morphology. WinMDI software (J. Trotter, Scripps Institute) was utilized to generate forward and side scatter plots and distinguish cell types as previously described (32).

ELISA.

Total levels of IgE, IgG1, and IgG2a in serum were determined by isotype-specific sandwich ELISA following the standard Pharmingen (BD Pharmingen, San Diego, Calif.) protocol. For measurement of IgE, purified anti-mouse IgE R35-72 (catalog no. 02111D; Pharmingen) was used as a capture antibody and alkaline-phosphatase conjugated anti-mouse IgE (catalog no. 02133E; Pharmingen) was used as a detecting antibody. For measurement of IgG1, purified anti-mouse IgG1 A85-3 (catalog no. 02241D; Pharmingen) was used as a capture antibody and biotinylated anti-mouse IgG1 A85-1 (catalog no. 02232D; Pharmingen) was used as a detecting antibody. IgG2a was measured using clone R11-89 (catalog no. 02251D; Pharmingen) as a capture antibody and clone R19-15 (catalog no. 02012D; Pharmingen) as a detection antibody. Total concentrations of IgE in serum were determined using a purified mouse IgE standard (catalog no. 03121D; Pharmingen). Total concentrations of IgG1 and IgG2a were determined using naïve BALB/c wild-type (WT) mouse serum as a standard. Published data on IgG1 and IgG2a levels in mouse serum were used to interpret the ELISA data for these two isotypes (31).

Statistical analysis.

Statistical significance was determined by Student's t test. P values less than 0.05 were considered statistically significant.

RESULTS

IgE^−/− mice are more permissive hosts than WT mice in a primary infection with B. malayi.

When BALB/c WT mice receive i.p. injections with 50 L3 larvae of B. malayi, parasite-specific and nonspecific levels of IgE increase, beginning approximately 2 weeks postinfection (Fig. 1A). By approximately 5 to 7 weeks postinfection, at least 90% of the parasite load has been eliminated in these animals. In contrast, BALB/c IgE^−/− mice still harbor significant worm burdens as late as 10 weeks postinfection. Figure 2 describes one representative experiment of five in which WT cohorts harbored significantly fewer viable parasites than IgE^−/− mice. At 1 week postinfection identical numbers of viable parasites were recovered from WT and IgE^−/− mice (58% ± 4% and 58% ± 6%, respectively). By 2 weeks postinfection, WT mice had decreased their parasitic burdens to 40% ± 6%, while IgE^−/− mice still harbored an average burden of 78% ± 11% (P = 0.015). Differences in parasite recoveries were significantly different at 3 weeks postinfection as well. By 6 weeks postinfection, only 11% ± 3% of the parasites were recovered from WT mice, while 30% ± 4% still remained in IgE^−/− mice (P = 0.006). Ten weeks postinfection, as much as 30% of the initial injection has been recoverable as live worms from IgE^−/− animals.

FIG. 1.

Analysis of humoral response to filarial infection in IgE^−/− mice. Four each of WT or IgE^−/− mice were bled at each time point by cardiac puncture just prior to necropsy. Sandwich ELISA was performed to determine total serum concentrations of IgE (A), IgG1 (B), or IgG2a (C). Error bars represent SEM.

FIG. 2.

IgE^−/− animals are more permissive hosts to infection with B. malayi than WT mice. Thirty each of BALB/c WT (solid bars) and BALB/c IgE^−/− (open bars) mice were inoculated i.p. with L3 larvae of B. malayi. Five animals from each cohort were sacrificed at weeks 1, 2, 3, 4, and 6 postinfection, and viable worms were recovered from peritoneal lavage fluid and counted. Values represent the percentage of the initial inoculum recovered as live worms. Error bars represent SEM. Time points at which differences in recoveries from WT and IgE^−/− mice were statistically significant are indicated (*).

IgE^−/− mice do not become microfilaremic.

Approximately 10 weeks following entry into a susceptible mammalian host, Brugia parasites develop to a mature adult form able to mate and produce their offspring, Mf. Our laboratory has previously demonstrated that highly permissive hosts such as completely immunodeficient SCID mice become microfilaria-positive (Mf⁺) at this time point (23). In addition, mice deficient in interleukin-4 (IL-4) production or signaling also become Mf⁺ by 10 to 12 weeks postinfection (1, 33). Despite the existence of live adult parasites at 10 and 12 weeks postinfection, we have rarely observed Mf in IgE^−/− animals. In one representative experiment, although a live parasite recovery (mean ± standard error of the mean [SEM]) of 34% ± 8% was obtained from IgE^−/− mice at 10 weeks postinfection, no Mf was observed in any of these animals. In this same experiment, four out of five IL-4^−/− mice, with a live parasite recovery of 23% ± 4%, were found to be Mf⁺. This is representative of three independent experiments.

Composition of PEC is equivalent in WT and IgE^−/− mice.

Our laboratory has carefully analyzed and documented the cellular influx into the peritoneal cavity of WT mice throughout infection with B. malayi (32). With other mutant animals found to be susceptible to B. malayi infection, i.e., IL-4 receptor^−/− and Stat6^−/− mice, we have noted significant alterations in the cellular composition at the infection site correlating with their increased permissiveness (33). In contrast to these other mutants, Fig. 3 illustrates that no obvious variations are observed in the composition of PEC from WT and IgE^−/− mice at 2 weeks postinfection, although at this time point a significant difference was observed in average live worm recoveries between the two cohorts (8% ± 3% from WT and 48% ± 10% from IgE^−/− mice).

FIG. 3.

Cellular profile of the infection site is equivalent in WT (solid bars) and IgE^−/− (open bars) mice. Five each of BALB/c WT and BALB/c IgE^−/− mice were inoculated i.p. with L3 larvae of B. malayi. Two weeks postinfection animals were sacrificed, and peritoneal lavage fluid was collected and analyzed. Identification of cell types was based upon forward and side scatter properties as described in Materials and Methods. Error bars represent standard deviation.

IgE^−/− mice mount an accelerated clearance of a secondary infection.

Previously exposed WT animals exhibit an acceleration in clearance kinetics of a secondary infection with B. malayi (2, 5, 7-9, 12, 13, 17, 25, 36). To investigate the functional significance of IgE production in a challenge infection, we infected cohorts of naïve or previously exposed WT or IgE^−/− mice with L3 larvae. In the experiment shown in Fig. 4, which is representative of two independent experiments, naïve WT mice harbored 20% ± 4% of the injected worms at 7 days postchallenge, while only 8% ± 3% were recovered from previously exposed WT mice (P = 0.049). Previously exposed IgE^−/− mice in this same experiment also exhibited a profound acceleration in their ability to clear their parasitic infection in comparison to previously naïve IgE^−/− mice. Naïve and previously exposed IgE^−/− animals harbored 36% ± 4% and 12% ± 5% of their initial parasite inoculum, respectively (P = 0.007). At 14 days post-challenge infection, 8% ± 4% of the injected worms were recovered from naïve WT mice, while only 0.5% ± 0.5% were recovered from previously exposed WT mice (P = 0.005). At this same time point, 20% ± 3% and 0.8% ± 0.8% were recovered from naïve and previously exposed IgE^−/− mice, respectively (P < 0.01).

FIG. 4.

IgE^−/− mice display accelerated clearance of a challenge infection of B. malayi similar to WT mice. Ten each of naïve (solid bars) or previously exposed (open bars) BALB/c WT and IgE^−/− mice were challenged with 50 L3 larvae of B. malayi i.p. 11 weeks after the primary infection. Five animals per group were sacrificed at 7 (A) or 14 (B) days postchallenge, and viable worms recovered from peritoneal lavage fluid and counted. Worm recoveries reflect only the percentage of the challenge inoculum recovered as live worms. Error bars represent SEM.

IgE^−/− mice are profoundly deficient in IgG1 and IgG2a production in response to B. malayi infection.

The discrepancy in data generated through use of IgE^−/− versus antibody-treated mice may be explained by incomplete neutralization of IgE, especially with regard to tissue levels or the existence of additional deficiencies in the genetically manipulated animals. Analysis of the humoral immune response to B. malayi infection in IgE^−/− mice highlighted a significant decrease in both the Th2-associated isotype IgG1, as well as the Th1-associated isotype IgG2a. As shown in Fig. 1B, IgE^−/− animals exhibit lower concentrations of IgG1 in serum as early as 1 week postinfection (365 ± 30 μg/ml in WT and 298 ± 4 μg/ml in IgE^−/− mice) although the values are not significantly different at this time point (P = 0.207). By 2 weeks postinfection, levels of IgG1 in the WT spiked to 800 ± 150 μg/ml, while remaining essentially constant at 350 ± 25 μg/ml in IgE^−/− animals (P = 0.026). Levels of IgG1 continue to rise through week 3 postinfection in WT animals (2,490 ± 485 μg/ml) before declining by week 4 to 1,730 ± 300 μg/ml. In IgE^−/− mice, IgG1 levels remain constant at these same time points (540 ± 125 μg/ml at week 3 and 410 ± 80 μg/ml at week 4 postinfection). P values between WT and IgE^−/− animals at 3 and 4 weeks postinfection are 0.010 and 0.005, respectively.

Similar to levels of IgG1, the concentration of IgG2a in serum is significantly decreased in IgE^−/− mice. Figure 1C shows IgG2a levels in IgE^−/− mice in comparison to WT counterparts at weeks 2 through 4 postinfection. While serum concentrations in WT mice remain between 2,025 ± 322 and 3,548 ± 358 μg/ml throughout the experiment, serum IgG2a concentrations are barely detectable in IgE^−/− cohorts at the same time points, ranging from undetectable levels to 270 ± 30 μg/ml (P values range from <0.01 to 0.006).

IgE^−/− mice display a normal level of IgG1 production in response to a challenge infection with B. malayi.

Humoral analyses were again performed following a challenge infection protocol, and are shown in Fig. 5. In contrast to the primary infection, IgE^−/− mice are able to mount a significant IgG1 response following a challenge infection. In the representative experiment shown in Fig. 5A (representative of two independent experiments), IgG1 levels were nearly threefold greater in IgE^−/− mice in comparison to WT mice, although statistically these values were not different (P = 0.125). Similarly, IgG2a levels in IgE^−/− mice, while consistently lower, were statistically similar to those seen in WT mice (P = 0.174).

FIG. 5.

Humoral analysis of IgE^−/− mice following challenge infection. Results for IgG1 (A) and IgG2a (B) are shown. For each isotype, four each of WT or IgE^−/− mice were bled by cardiac puncture at day 14 post-challenge infection. Sandwich ELISA was performed to determine total serum concentrations of the relevant isotype. Each dot represents an individual mouse, and bars represent the mean concentration for each group.

DISCUSSION

Despite IgE's known association with parasitic disease in both human and murine infections, and in vitro evidence of parasite-killing activity, demonstrating a definitive functional role for IgE in vivo in B. malayi infection has been difficult. Previous attempts through antibody-mediated neutralization of IgE have demonstrated no difference in host protection. In vivo antibody neutralization faces limitations in that the depletion may not be complete. Levels of IgE in serum below the level of detection may prove to be functionally significant, and levels in serum are not necessarily representative of levels of expression in tissue. To circumvent these issues, we have utilized mice rendered genetically deficient for IgE by a null mutation of the Cɛ gene. Our data shown here indicate that BALB/c IgE^−/− mice are significantly more permissive hosts to infection with B. malayi than BALB/c WT mice. WT mice generally reduce their parasitic load to ≤10% of the original infective larvae by 5 or 6 weeks postinfection. In contrast, IgE^−/− mice generally maintain approximately 20% of the infective larvae by as late as 10 weeks postinfection.

At approximately 10 weeks postinfection, adult parasites within a permissive host have developed to full maturity, at which point they produce Mf through sexual reproduction. In B. pahangi infections of IL-4^−/− mice (10) and B. malayi infections of IL-4 receptor^−/− mice (unpublished observations from our laboratory), the absence of IL-4 correlates with a significant increase in the production of Mf. It is intriguing that the IgE knockout mice do not have Mf in their peritoneal cavities, despite the presence of significant numbers of adult male and female worms. In this context, it is worth noting that we have made a similar observation in globally immunoglobulin-deficient mice, such as the Igh-6^null and JhD knockout mice (27, 28). These strains, in common with the IgE knockout mice, have deficits in immunoglobulins (global or isotype specific) but manifest intact T-cell functions. In contrast, we find that mice in whom T-cell function is deficient (SCID, TCRβ knockout, and nude mice) or in whom the IL-4 or IL-4 signal transducing pathway is missing (IL-4, IL-4 receptor, or Stat6 knockout mice) do become microfilaremic at comparable adult worm burdens. For instance, in the experiment described above, despite the fact that IgE^−/− mice have 34% ± 8% worms, a higher mean than the 2% ± 4% in IL-4 knockout mice, only the latter bear Mf. Thus, it would appear that the control of microfilaremia is accomplished by T cells through a mechanism that requires IL-4, IL-4R, and Stat6 but which does not appear to need immunoglobulins. A similar observation linking IL-4 with anti-Mf immunity has been made in experimental infections with the mouse filarial parasite Litomosoides sigmodontis (34).

Despite the significant defect in host protection, the cellular profile at the site of infection appears nearly identical between WT and IgE^−/− mice. This is in contrast to observations comparing BALB/c IL-4 receptor^−/− and Stat6^−/− with WT mice, where the absence of IL-4 receptor signaling was associated with a significant reduction in eosinophils at the infection site (33).

WT mice previously exposed to live B. malayi L3 larvae mount a rapid immune response when confronted with a challenge infection of L3 larvae. While clearance of a primary infection from BALB/c WT mice generally requires 5 to 7 weeks, clearance of a challenge injection is accomplished by 2 to 3 weeks. At variance with the increased permissiveness of IgE^−/− mice in comparison to WT during a primary infection, previously exposed IgE^−/− mice are able to clear a challenge injection of L3 larvae with kinetics very similar to that seen in the challenged WT cohorts. Therefore, in contrast to the observed impairment of IgE^−/− mice to clear a primary infection, the ability of previous parasite exposure to elicit a more rapid response upon secondary challenge appears unhindered. Interestingly, despite the accelerated clearance of a challenge infection achieved by the IgE^−/− cohort, at the time of necropsy these mice still harbored adult parasites remaining from the priming injection (data not shown). Similar observations have been described following challenge of IL-4^−/− mice (33) and support the concept that immunity in filarial infections is a stage-specific phenomenon (14, 21). Induction of successful host-protective mechanisms in the challenge infection of IgE^−/− mice may be indicative of unique mechanisms of parasite killing in primary and secondary infections, with the former being dependent upon IgE while in the latter an IgE-independent mechanism predominates. Alternatively, while an IgE-dependent mechanism may be sufficient to mediate parasite clearance in a primary infection, given the time involved in a challenge infection protocol, IgE-independent compensatory mechanisms (such as WT levels of IgG1) may be able to mediate host protection.

While the clearance of the secondary infection is accelerated in IgE^−/− mice and this is correlated with an increase in the IgG1 titers in the mice as a group, there is no correlation between IgG1 titers and worm burdens in individual mice. This is consistent with our experience over the last several years, during which we have not been able to find any single parameter that is predictive of worm burdens in a given mouse, attesting to the complexity of the protective host immune response. Thus, even individual WT mice may have higher worm burdens than the mean worm burdens of a group of highly permissive SCID mice infected with the same batch of infective larvae and necropsied at the same point after infection. Conversely, individual SCID mice often have lower worm burdens than the mean in WT mice injected with the same batch of larvae and necropsied at the same time point after infection.

A previous report demonstrated that BALB/c WT mice treated from birth with an anti-ɛ monoclonal antibody displayed no difference in host protection against B. malayi infection when compared to untreated mice (35). One explanation for the incongruity would be the existence of phenotypic abnormalities in addition to the absence of IgE in IgE^−/− mice. Humoral analyses reveal just such an abnormality. In response to a primary infection with B. malayi, IgE^−/− animals are impaired in their production of both the Th2-associated isotype IgG1 and in production of the Th1-associated IgG2a. A role for antibody has been hypothesized in the control of B. pahangi Mf (11), as well as in host immunity to L3 and L4 stages of B. malayi (32). Thus, an overall impairment in antibody production, rather than an isolated deficiency in IgE, may be responsible for the increased permissiveness seen in a primary infection of IgE^−/− mice.

Contrary to a primary infection, no significant difference is observed in the overall concentrations of circulating IgG1 or IgG2a following a challenge infection. In the case of Th2-associated IgG1, levels measured in IgE^−/− mice are nearly threefold higher, while Th1-linked IgG2a levels tend to remain slightly lower than levels measured in WT mice. Thus, the restoration of host protection seen in the challenge infection in contrast to the primary infection correlates with the reestablishment of IgG1 production.

A similar finding was previously described in an infection of IgE^−/− mice with another parasitic organism, Schistosoma mansoni. In this study IgE^−/− mice were found to have increased worm burdens and reduced granuloma formation following a primary infection when compared to WT mice. As in our infection model, IgG1 levels in IgE^−/− mice were significantly decreased when compared to WT controls. Contrary to the primary infection, IgG1 levels returned to normal in IgE^−/− mice following a challenge infection, and the parasite was cleared with significantly increased kinetics (16). The simultaneous restoration of IgG1 and host protection observed in both models following a challenge infection may highlight the importance of Th2-associated antibody to host protection.

The present study describes an increase in permissiveness of IgE^−/− mice in response to a primary infection with B. malayi, unaccompanied by any significant alterations in the cellular profile at the infection site. Analyses of the humoral response in these animals in response to infection reveal a defect in production of IgG1 and IgG2a in addition to the absence of IgE. In contrast to the primary infection, IgE^−/− mice are able to clear a challenge infection of the parasite with accelerated kinetics indistinguishable from that of previously exposed WT mice. Coincident with the induction of host immunity is the reinstatement in particular of IgG1 production, although the functional significance of this Th2-associated isotype will be the subject of further investigation.