T Cells Are Required for Host Protection against Brugia malayi but Need Not Produce or Respond to Interleukin-4

Abstract

T cells are known to be required for host protection in mouse models of Brugia malayi infection. Several independent studies in murine models have also highlighted the rapid induction of Th2-like responses after infection with B. malayi or B. pahangi. Previous data from our laboratory have described a significant increase in permissiveness in the absence of interleukin-4 (IL-4), the “prototypical” Th2 cytokine, involved in both the induction and maintenance of Th2 responses. These observations led to our hypothesis that T cells involved in murine host protection would respond to IL-4 signaling and differentiate into cells of the “type 2” phenotype. As such, these cells would presumably also act as major sources of IL-4. To investigate these hypotheses, we performed several adoptive transfers in which we controlled the cell population(s) able to produce or respond to IL-4. We show here that, in contrast to our original hypotheses, IL-4 production and IL-4 receptor expression by T cells are both dispensable for T-cell-mediated host protection. Instead, our data imply that T cells may be required for eosinophil accumulation at the site of infection.

T cells have been shown to be essential for host protection in murine infections with Brugia malayi and the closely related feline parasite B. pahangi (3, 41, 51-53). The precise role of T cells in host protection has, however, not been definitively established. In the murine model of infection, T-cell responses to Brugia are predominantly type 2 in nature. Bancroft et al. suggested that the critical function for CD4⁺ T cells in host protection against a challenge infection with Brugia is the production of Th2 cytokines (i.e., interleukin-4 [IL-4], IL-5, and IL-9) (6). Several more recent observations lend support to this hypothesis. Loke and colleagues have demonstrated that antigen-presenting cells recruited by the parasite induce Th2 differentiation of naive T cells (25). Further, in vitro culture of dendritic cells with an excretory or secretory protein isolated from B. malayi causes the dendritic cells to prime naive T cells to differentiate along the Th2 pathway (54).

Downstream effector functions of Th2 cells have also been described in brugian infections, namely, eosinophilia and production of immunoglobulin E (IgE) and IgG4 in humans (22, 31) and IgE and IgG1 in murine models (5, 13, 23). We have previously demonstrated that the presence of this IgE (and/or IgG1) may promote parasite killing (40).

Interleukin-4 (IL-4) has been identified as the major cytokine involved in promoting Th2 differentiation and, as such, has become the hallmark characteristic of type 2 responses (8, 17, 24, 36, 42, 43). In line with observations of induction of a type 2 response following brugian infection, Devaney and coworkers have described an increase in IL-4 message in the draining lymph nodes within 24 h postinfection with B. pahangi, which they later demonstrated was derived from NK-T cells (4, 30). Our laboratory has shown that, in the absence of IL-4 production or when signaling through the IL-4 receptor (IL-4R) or subsequent Stat6 activation is blocked, mice become significantly more susceptible to infection with either B. malayi or B. pahangi (2, 40).

In light of these observations, we hypothesized that in our infection model T cells contribute to host protection through differentiation to the Th2 phenotype and, as such, serve as a major source of IL-4 throughout the infection. We further surmised that the observed requirements for IL-4R and Stat6 expression reflect the necessity of these molecules in mediating Th2 differentiation. To investigate these hypotheses, we performed several adoptive-transfer experiments utilizing lymphocyte-deficient mice as recipients of various lymphocyte populations from wild-type (wt), IL-4^−/−, IL-4R^−/−, or Stat6^−/− donors. Our data do not confirm our hypothesis. Transfer of T cells from IL-4^−/−, IL-4R^−/−, and Stat6^−/− donors are capable of inducing host protection comparable to that mediated by T cells of wt origin. Thus, while T cells are required for host protection, it appears that they need not produce or respond to IL-4. The most notable defect we find in T-cell-deficient mice, such as T-cell-receptor β (TCRβ) knockout (KO) or SCID mice, is the failure of accumulation of eosinophils at the site of infection. This defect is reversed by adoptive transfer of T cells from wt, IL-4 KO, or IL-4R^−/− mice. Our data imply that one of the functions of T cells in host defense in this model may be to recruit inflammatory cells, including eosinophils, to the site of infection.

MATERIALS AND METHODS

Mice.

BALB/c-Prkdc^scid/Prkdc^scid (SCID), BALB/cByJ^+/+, BALB/cBy-Tcrb^null (TCRβ^−/−), BALB/cBy-Fox1^nu (nude), BALB/c-Il4^tm2Nnt (IL-4^−/−), and BALB/c Stat6^−/− animals were obtained from the Jackson Laboratory (Bar Harbor, Maine). BALB/c-Il4ra^tm1Sz (IL-4R^−/−) mice were initially obtained from the Jackson Laboratory and were subsequently bred and housed under specific-pathogen-free conditions in microisolator cages in the American Association for Accreditation of Laboratory Animal Care-accredited vivarium of the UConn Health Center. All mice used were males between 6 and 12 weeks of age. Confirmation of the genotype of the IL-4R^−/− animals was carried out by PCR analysis of tail DNA from randomly selected mice, with primers specific for the neomycin resistance gene, as well as specific segments of the disrupted IL-4 receptor gene, as described previously (29). Leakiness of the SCID phenotype was ruled out by serum Ouchterlony tests.

Parasite.

L3 larvae of B. malayi were harvested at the insectarium of Thomas Klei (Louisiana State University, Baton Rouge) from infected Aedes aegypti mosquitos and shipped overnight in RPMI containing penicillin, streptomycin, gentamicin, and fluconazole.

Experimental infection.

Mice were inoculated with 50 B. malayi L3 infective-stage larvae (L3) intraperitoneally (i.p.) by using a 25-gauge 5/8-in. needle. Mice were euthanized at various time points postinfection and then subjected to a cardiac bleed for retrieval of serum. Peritoneal lavages were performed by using RPMI medium supplemented with 5 U of heparin/ml. Lavage was extracted from the peritoneal cavity by using a soft plastic pipette to prevent shearing of the adult worms. After lavage, intestines were removed and soaked in phosphate-buffered saline (PBS). Scrotal sacs were opened, and carcasses placed in PBS for further soaking. Carcasses were then rinsed several times with PBS. Viable worms were counted from peritoneal lavage and carcass soaks under a dissecting microscope.

Ex vivo cell culture.

Total peritoneal exudate cells (PEC) were harvested from BALB/c wt mice 7 or 14 days after infection with B. malayi or PBS injection. To obtain T-cell-depleted or -enriched populations, cells were incubated with magnetic beads directly conjugated to anti-Thy1.2 monoclonal antibody on ice for 20 min. After several washes, cells were added to magnetic cell sorting (MACS) negative selection separation columns (Miltenyi Biotec, Auburn, Calif.) in a magnetic field. Cells that were not retained in the column were collected as the T-cell-depleted population. T-cell-enriched populations were recovered through elution of bound cells after removal of the magnetic field. T-cell numbers in each population were determined by fluorescence-activated cell sorting (FACS) analysis of CD3⁺ cells. Unfractionated, T-cell-depleted or -enriched populations were cultured in RPMI supplemented with 10% fetal bovine serum (Gibco) and 1% Pen-Strep antibiotic (Gibco) in the presence or absence of 4 μg of ionomycin/ml. Cells were seeded at a density of 10⁶ cells in 200 μl per well and then incubated at 37°C with 6% CO₂ for 24 h. After incubation, supernatants were centrifuged to remove cellular debris and stored at −80°C.

IL-4 ELISA.

Levels of IL-4 in supernatants collected from the ex vivo cultures were measured by a standard sandwich enzyme-linked immunosorbent assay (ELISA). Anti-mouse IL-4 (clone 11B11; Pharmingen 18191D) was used as a capture antibody, and biotinylated anti-mouse IL-4 (clone BVD6-24G2; Pharmingen 18042D) was used as a detection antibody. Total concentration of IL-4 was determined by comparison to serial dilutions of recombinant mouse IL-4 (Pharmingen 19231T).

Cell preparations and adoptive transfers.

Cells were isolated from the spleens of naive donor mice by teasing the cells through a 450-μm metal screen. For cell depletion prior to reconstitution, total splenocytes were incubated with biotinylated anti-CD19 or anti-CD3 monoclonal antibodies (BD Pharmingen, San Diego, Calif.). Unbound antibody was washed away, and cells were incubated with streptavidin-conjugated magnetic beads. Cells were then washed again and added to MACS negative selection separation columns in a magnetic field. Cells that were not retained by the column in this field were collected as the depleted population and analyzed for efficiency of depletion by FACS analysis. For TCRβ^−/− recipients of CD19- or CD3-depleted populations, 5 × 10⁶ or 4 × 10⁶ cells, respectively, were transferred. For adoptive transfers of total splenocytes, 10⁷ cells were transferred to each SCID recipient. All cells were injected i.p.

FACS.

Peritoneal lavage cells were collected and passed through nylon mesh to remove debris. Conjugated monoclonal antibodies (CD19-phycoerythrin, CD3-fluorescein isothiocyanate, and TCRβ-biotin) were obtained from Pharmingen and used at dilutions of 1:100 unless otherwise stated. Streptavidin-CyChrome (Pharmingen 554062) was used at a dilution of 1:400 with biotin-conjugated antibodies. Stained cells were fixed with 0.5% paraformaldehyde. Data were acquired on a FACSCalibur (Becton Dickinson), by using CellQuest software. Data were subsequently analyzed by using WinMDI.

Statistical analysis.

Statistical significance was determined by the Student t test by using Microsoft Excel. P values of <0.05 were considered statistically significant.

RESULTS

T cells are required for host protection.

It is well established that T cells are necessary for successful host protection against infection with B. malayi or the closely related feline parasite B. pahangi. As shown in Fig. 1, α/β⁺ T cells in particular are critical to a successful host response. A total of 50 B. malayi L3 infective-stage larvae were injected i.p. into groups of BALB/c wt (n = 9) or TCRβ^−/− mice (n = 12), and the animals were sacrificed 6 weeks postinfection. It has been previously established that larvae injected in this manner do not migrate out of the peritoneal cavity (9); therefore, at the time of necropsy all remaining larvae were recovered by peritoneal lavage and counted for an accurate determination of the total parasite burden. In this representative experiment, 10% ± 3% of the injected parasites were recovered from wt mice, while the TCRβ^−/− cohort harbored 27% ± 5% (P = 0.024).

FIG. 1.

TCRα/β⁺ T cells are required for host protection. BALB/c TCRβ^−/− (n = 12) or BALB/c wt mice (n = 9) were injected i.p. with 50 B. malayi L3 infective-stage larvae. At 6 weeks postinfection the animals were necropsied, and live parasites were recovered from peritoneal lavage. Data are shown as a percentage of the initial inoculum recovered as viable parasites at the time of necropsy. The P value between wt and TCRβ^−/− mice is 0.024. The error bars represent the standard error of the mean (SEM).

IL-4 signaling is required for host protection.

More recently, we have shown that IL-4 and signaling through the IL-4 receptor is also required for host protection. In Fig. 2A, which is representative of more than 25 independent experiments performed at various time points with either B. malayi or B. pahangi, parasite recoveries were significantly greater from IL-4^−/− and IL-4R^−/− mice than from wt mice. In this experiment, at 12 weeks postinfection 2% ± 0.6% of the parasites were recovered from wt mice, while 16% ± 3% and 17% ± 6% were recovered from the IL-4^−/− and IL-4R^−/− cohorts, respectively (P = 0.001 and 0.025 between wt and IL-4^−/− mice and between wt and IL-4R^−/− mice, respectively). A comparison of worm burdens between wt and IL-4^−/− mice at 6 weeks postinfection also demonstrated a significantly increased worm burden in the mutant mice (Fig. 2B). Downstream of IL-4 receptor signaling, activation of Stat6, a transcription factor implicated in Th2-cell differentiation (20, 38, 44), also promotes parasite killing, as demonstrated in Fig. 2B. Stat6^−/− T cells are generally unable to induce expression of GATA-3 and c-maf, two transcription factors that appear to be indispensable for Th2-cell development (21, 32). In the present study, at 6 weeks postinfection, wt mice had substantially cleared the parasites (0.4% ± 0.4%), while IL-4^−/− and Stat6^−/− mice still harbored significant numbers of viable parasites: 9% ± 2% and 17% ± 3%, respectively (P = 0.002 and 0.001 between wt and IL-4^−/− mice and between wt and Stat6^−/− mice, respectively).

FIG. 2.

IL-4, IL-4R signaling, and Stat6 are all required for murine host protection (A) BALB/c IL-4^−/− mice (n = 5), BALB/c IL-4R^−/− mice (n = 4), or BALB/c wt mice (n = 5) were injected i.p. with 50 B. malayi L3 infective-stage larvae and then necropsied at 12 weeks postinfection. Live parasites recovered from peritoneal lavage were quantified and are expressed as a percentage of the original infective dose. The P values between wt and IL-4^−/− mice or between wt and IL-4R^−/− mice are 0.001 and 0.020, respectively. (B) BALB/c wt (n = 5), IL-4^−/− (n = 5), or Stat6^−/− (n = 5) mice were injected i.p. with 50 B. malayi L3 infective-stage larvae and sacrificed 6 weeks postinfection. The P values between the recoveries from wt and IL-4^−/− mice or between wt and Stat6^−/− mice are 0.002 and 0.001, respectively. The data are shown as the percentage of the initial inoculum recovered as viable parasites at the time of necropsy. The error bars represent the SEM.

T cells provide the majority of IL-4 at the site of infection.

As described in greater detail in the introduction, data from several laboratories have demonstrated a rapid induction of type 2 CD4⁺-T-cell differentiation after infection with Brugia. In line with these observations, CD4⁺ T cells isolated from secondary lymphoid tissue were shown to contain significant levels of mRNA transcripts of Th2-associated cytokines such as IL-4, IL-5, IL-9, and IL-10 after primary and challenge infections (4-6, 30). Despite these observations, the cytokine repertoire of T cells at the infection site has not been elucidated.

To determine whether T cells provide a significant source of IL-4 at the site of infection, PEC from wt mice were isolated at 1 or 2 weeks postinfection with B. malayi. A control group of animals receiving a PBS injection in place of the parasites were included to differentiate background from parasite-induced responses. In order to ensure equal opportunity for non-T-cell sources of IL-4 to be observed, the calcium ionophore ionomycin was used in place of antigen restimulation. PEC recovered from parasite-infected or sham-infected mice were cultured in the presence or absence of ionomycin for 24 h. After incubation, IL-4 levels were determined from the culture supernatants by ELISA. The relative contribution of T cells to the production of IL-4 was evaluated by comparison of IL-4 production by T-cell-depleted and -enriched fractions of the total PEC. Data from a representative experiment are shown in Table 1. In the absence of any stimulation, a 30- or 10-fold increase versus PBS-induced levels of IL-4 was observed 1 or 2 weeks after infection, respectively. Treatment with ionomycin further increased these levels nine- and fivefold, respectively. When T cells were enriched threefold prior to culture, the levels of IL-4 production at both time points similarly increased threefold. When T cells were depleted prior to culture (<0.5% T-cell contamination), IL-4 production decreased 10- and 7-fold at days 7 and 14 postinfection, respectively.

TABLE 1.

Ex vivo IL-4 production by PEC from BALB/c wt mice^a

Day postinfection	Fraction	% T cells	Mean IL-4 concn (pg/ml) ± SD with:
			Medium alone	Medium + ionomycin
NAb (PBS     control)	Whole PEC	4	3 ± 5	35 ± 47
7	Whole PEC	5	85 ± 23	733 ± 58
	Thy 1.2+	17	220 ± 113	1,500 ± 0
	Thy 1.2−	0.3	20 ± 8	91 ± 33
14	Whole PEC	3	30 ± 13	140 ± 26
	Thy 1.2+	10	74 ± 1	450 ± 71
	Thy 1.2−	0.3	13 ± 14	80 ± 0

^aPEC were recovered by peritoneal lavage from PBS-injected or L3-infected wt mice (n = 5) at days 7 and 14 postinfection. Equal total cell numbers of whole PEC, Thy 1.2-depleted, or Thy 1.2-enriched populations were cultured in the presence or absence of ionomycin as described in Materials and Methods. The cellular composition before and after separation was determined by CD3 staining for T cells and by intermediate Gr-1 staining and forward scatter-side scatter properties for eosinophils. IL-4 was measured from culture supernatants by sandwich ELISA as described in the text and is expressed as an average of multiple wells ± the standard deviation. (Six wells were analyzed for whole PEC samples, 11 wells for were analyzed Thy 1.2-depleted samples, and 2 wells were analyzed for Thy 1.2⁺ samples.)

^bNA, not applicable.

IL-4^−/− T cells are sufficient to mediate host protection.

To determine the relative importance of T-cell-derived IL-4 to host protection, we utilized an adoptive transfer model in which we created chimeras differing in the ability of the T-cell population to produce IL-4. Total splenocytes were isolated from naive wt or IL-4^−/− animals. Using a magnetic bead separation technique, we isolated CD19-depleted cells from each of these groups of animals and transferred the depleted populations into TCRβ^−/− recipients. All of our transferred populations contained ca. 60 to 70% T cells, with <1% B-cell contamination, as shown by flow cytometric analysis (Fig. 3A), and reconstitution of the T-cell compartment was efficient, as demonstrated in Fig. 3B. TCRβ^−/− mice receiving CD3-depleted splenocytes from IL-4^−/− donors were included as a negative control. Three days posttransfer the TCRβ^−/− animals were challenged with 50 L3 infective-stage larvae of B. malayi, and all animals were necropsied at 6 weeks postinfection. The data shown in Fig. 3C are representative of three similar experiments. Groups of unmanipulated wt and IL-4^−/− mice were included to verify a significantly higher parasite burden in IL-4^−/− mice at this time point, with this batch of larvae (P = 0.004). At the time of necropsy, 35% ± 4% of the injected parasites were recovered as viable worms from control animals (recipients of CD3-depleted IL-4^−/− splenocytes). In contrast, significant protection was observed in mice receiving CD19-depleted splenocytes prior to infection, regardless of whether the donor cells were from wt or IL-4^−/− animals: 13% ± 3% and 18% ± 2%, respectively (P = 0.003 between control mice and recipients of wt cells, and between control mice and recipients of IL-4^−/− cells).

FIG. 3.

IL-4^−/− T cells are sufficient to mediate significant host protection. CD19-depleted splenocytes from naive BALB/c wt or BALB/c IL-4^−/− donors were transferred into BALB/c TCRβ^−/− recipients (n = 5) 3 days prior to i.p. infection with B. malayi L3. Control TCRβ^−/− animals received naive CD3-depleted splenocytes from IL-4^−/− donors. (A) Anti-CD3 and anti-CD19 staining of depleted cell populations prior to adoptive transfer. (B) Anti-TCRβ and anti-CD19 staining of PEC recovered at the time of necropsy. Total PEC are gated on lymphocytes. (C) Animals were sacrificed 6 weeks postinfection. The data are shown as the percentage of the initial inoculum recovered as viable parasites at the time of necropsy. The P value between control TCRβ^−/− mice and recipients of CD19-depleted wt cells and between control animals and recipients of CD19-depleted IL-4^−/− cells is 0.003 for both comparisons. The error bars represent the SEM. The P value between recoveries from unmanipulated wt and IL-4^−/− mice is 0.004.

IL-4R^−/− T cells are sufficient for host protection.

With the finding that T-cell production of IL-4 is not essential to host protection, we investigated the importance of IL-4 receptor and Stat6 expression by T cells to host protection by using a similar adoptive-transfer approach. Due to the availability of mice, immunodeficient SCID mice were used as recipients. Figure 4A shows the efficiency of the adoptive transfer protocol to reconstitute the T-cell compartment of SCID mice. As demonstrated by the open bars in Fig. 4C, unmanipulated IL-4R^−/− mice harbor greater numbers of viable parasites than wt mice at 6 weeks postinfection, although this difference does not reach statistical significance (17% ± 5% and 5% ± 1% from IL-4R^−/− and wt mice, respectively, P = 0.083). At 10 weeks postinfection, wt mice have substantially cleared the parasites 1% ± 0.8%, whereas IL-4R^−/− mice still support 12% ± 4% (P = 0.032). Regardless of the higher parasite burdens seen in unmanipulated IL-4R^−/− mice, transfer of either IL-4 R^+/+ or IL-4R^−/− splenocytes was sufficient to induce similar levels of host protection at both of these time points. At 6 weeks postinfection, SCID mice reconstituted with splenocytes from wt donors harbored 5% ± 2% of the original infective dose of B. malayi, whereas SCID mice receiving IL-4R^−/− splenocytes supported 7% ± 2%. Both of these worm burdens are significantly lower than the 63% ± 8% recovered from SCID mice receiving no donor splenocytes. P values between unreconstituted SCID mice and recipients of wt splenocytes or IL-4R^−/− splenocytes were both <0.001. Similarly, at 10 weeks postinfection, SCID recipients of wt splenocytes supported 3% ± 1% of the parasites, and recipients of IL-4R^−/− splenocytes harbored 5% ± 1%. Unreconstituted SCID mice at this time point still harbored 45% ± 8% (P values are both <0.01). This experiment is representative of five independent experiments.

FIG. 4.

SCID recipients of wt, Stat6^−/−, or IL-4R^−/− T lymphocytes are equally well protected. (A) Anti-CD3 and anti-CD19 staining of PEC recovered at time of necropsy of recipient mice. The plots represent total PEC gated on lymphocytes. (B) Total splenocytes from naive wt or Stat6^−/− mice were transferred to SCID recipients (n = 7 to 10) prior to infection with B. malayi. The data are shown as the percentage of the initial inoculum recovered as viable parasites at 6 weeks postinfection. The solid bar represents BALB/c SCID animals receiving a control injection of PBS. The P values between unreconstituted SCID mice and those receiving wt cells and between unreconstituted animals and recipients of IL-4R^−/− cells are 0.001 and 0.002, respectively. The error bars represent the SEM. (C) Total splenocytes from naive wt or IL-4R^−/− mice were transferred to SCID recipients (n = 20) prior to infection with B. malayi. Ten mice from each cohort were sacrificed each at 6 and 10 weeks postinfection. The data are shown as a percentage of the initial inoculum recovered as viable parasites at the time of necropsy. The P value between unreconstituted SCID mice and those receiving wt cells is <0.001, and the P values between unreconstituted mice and recipients of IL-4R^−/− cells are <0.001 and 0.001. The error bars represent the SEM.

Utilization of Stat6^−/− mice as splenocyte donors yielded similar results (Fig. 4B). Again, although unmanipulated Stat6^−/− mice harbored significantly more viable parasites than wt mice at 6 weeks postinfection (20% ± 4% and 5% ± 2%, respectively, P = 0.017), SCID recipients of Stat6^−/− splenocytes were protected comparably to SCID recipients of wt splenocytes. (22% ± 3% and 18% ± 3% recoveries from recipients of Stat6^−/− and wt cells, respectively, compared to 40% ± 4% from unreconstituted SCID mice [P = 0.002 and 0.001, respectively]).

T cells induce eosinophil influx into the infection site.

The data presented above suggest that both IL-4 production and IL-4 receptor expression by T cells are dispensable for host protection, prompting us to ask what T-cell function is required for parasite killing. Representative forward scatter-side scatter plots of PEC recovered from wt or TCRβ^−/− mice at 6 weeks postinfection with B. malayi are shown in Fig. 5A. In these plots the x axis is a measurement of the forward scatter or relative size of the cell, while the y axis plots the complexity or relative granularity of the cell. We have previously demonstrated that the population highest along the y axis and midway along the x axis (circled in Fig. 5A) represents a population highly enriched for eosinophils (34, 40). As illustrated by comparison of the two plots shown in Fig. 5A, this population is notably absent in TCRβ^−/− and SCID mice at this time point. Several lines of in vivo and in vitro evidence support a potential role for eosinophils in parasite killing, including their presence in close proximity to dying worms encased in host cell granulomas (10, 18, 19, 26, 34, 45). The absence of these cells in T-cell-deficient mice suggested a role for T cells in eosinophil recruitment. To confirm the involvement of T cells, CD19-depleted or total splenocytes were transferred from wt donors into TCRβ^−/− or SCID recipients 3 days prior to infection with B. malayi. At 6 weeks postinfection, the animals were sacrificed, and the PEC were recovered and analyzed for the presence of eosinophils. As seen in the representative experiment shown in Fig. 5A, transfer of T cells induced significant eosinophil influx into the infection site of TCRβ^−/− recipient mice.

FIG. 5.

Eosinophil recruitment requires T cells and IL-4R expression by nonlymphocytes. (A) Representative plots of PEC isolated at 6 weeks postinfection from a wt or TCRβ^−/− mouse and from a TCRβ^−/− recipient of CD19-depleted splenocytes from a wt donor. (B) Representative plots of PEC isolated at 6 weeks postinfection from an IL-4R^−/− or SCID mouse and from a SCID recipient of splenocytes from an IL-4R^−/− donor. Circles denote eosinophil population.

Eosinophil recruitment requires IL-4 receptor expression on nonlymphoid cells.

Our laboratory has previously demonstrated that IL-4R^−/− mice are also impaired in their ability to recruit and/or maintain eosinophils at the infection site (Fig. 5B). To determine whether IL-4R^−/− T cells were impaired in their ability to induce eosinophil transmigration, we repeated the adoptive-transfer protocol utilizing splenocytes from IL-4R^−/− donors. Despite the severe deficit in eosinophils observed at the site of infection in IL-4R^−/− mice, transfer of IL-4R^−/− T cells was sufficient to induce eosinophil recruitment in the SCID recipients (Fig. 5B). Table 2 presents the total numbers of eosinophils recovered from the infection site of IL-4R^−/− mice and SCID recipients of IL-4R^−/− splenocytes at 6 weeks postinfection. In this representative experiment, nearly three times more eosinophils were recovered from SCID recipients of IL-4R^−/− splenocytes than from IL-4R^−/− mice (P = 0.004), suggesting that IL-4R expression by nonlymphoid cells induces significant eosinophil recruitment. Eosinophil recruitment is similarly restored to SCID and TCRβ^−/− mice reconstituted with IL-4^−/− T cells (data not shown) although, intriguingly, BALB/c IL-4^−/− mice do not exhibit the deficit in eosinophil recruitment observed in IL-4R^−/− and Stat6^−/− mice after infection with B. malayi.

TABLE 2.

Eosinophil numbers in control and reconstituted animals^a

Animal group	Total no. of eosinophils (106)	Mean cells per mouse ± SD
IL-4R−/− mice	0.1, 0.12, 0.93, 0.65, 0.36	0.43 ± 0.36
SCID mice recipients of IL-4R−/− lymphocytes	0.89, 2, 1.1, 1.3, 0.8, 0.7, 1.6, 1.6, 1.6	1.28 ± 0.45

^aBALB/c SCID mice were reconstituted with total splenocytes from wt or IL-4R^−/− donor animals prior to infection with B. malayi as in Fig. 5B. Total numbers of eosinophils were determined on the basis of their distinctive size and granularity profile as previously described (34, 40). Thus, T cells, even from IL-4R^−/− mice, are able to reconstitute the ability of the recipient mice to recruit eosinophils to the peritoneal cavities in response to B. malayi infection.

DISCUSSION

As mentioned in the introduction, it is well established that T cells, and in particular TCRα/β⁺ T cells, are required for host protection in murine models of infection with both B. malayi and its feline-infective counterpart B. pahangi (3, 41, 51-53). Several laboratories have demonstrated that T-cell responses within the murine host are of the Th2 phenotype (6, 25, 54), and Th2-induced effector mechanisms, such as eosinophilia and production of Th2-associated antibody isotypes, are elicited in both human and murine infections (5, 13, 23).

It has also been established that the prototypical Th2 cytokine IL-4, IL-4R, and Stat6 are all required for murine host protection against infection with B. malayi or B. pahangi (2, 40). This requirement for IL-4 signaling has been well documented in rodent models of infection with several gastrointestinal nematodes (7, 14, 46-48) and, in the case of at least one of these nematodes (Trichinella spiralis), has been shown to be T cell dependent (50).

Despite a significant body of literature demonstrating preferential production of Th2 cytokines by T cells within secondary lymphoid tissue after B. malayi infection, information concerning the cytokine profile of T cells at the site of infection in this model has been lacking. We have investigated here the ex vivo potential of PEC from infected mice to produce IL-4 and demonstrate that, even in the absence of antigen restimulation, T cells provide the majority of IL-4 at 1 and 2 weeks postinfection.

To determine the relevance of this T-cell-derived IL-4 to host protection, we reconstituted T-cell-deficient mice with T cells isolated from IL-4^+/+ or IL-4^−/− donors. TCRβ^−/− animals were used as recipients of CD19-depleted splenocytes from either wt or IL-4^−/− donor mice. Control TCRβ^−/− animals received CD3-depleted splenocytes from IL-4^−/− donors. As shown in Fig. 3A, the CD19-depleted donor cell populations contained ca. 60 to 70% T cells, with <1% B-cell contamination. Regardless of whether the T cells originated from a wt or an IL-4^−/− donor, mice receiving T cells were significantly more protected than those receiving control CD3-depleted donor cells. Further, the level of protection between the two reconstituted cohorts was statistically similar, indicating that the ability of the donor-derived T cells to produce IL-4 did not offer any incremental host protection at this time point (Fig. 3C).

In several gastrointestinal nematode models of infection, IL-13 has been shown to compensate for the absence of IL-4 in mediating host protection (27, 48). Two types of receptors have been reported for IL-4. The first is composed of the IL-4R α-chain dimerized to the common γ-chain, and the second is made up of the IL-4R α-chain and the IL-13R α1-chain. IL-13 has been shown to mimic many of the effects of IL-4 by binding to the second of these receptors. In IL-4R^−/− mice the IL-4R α-chain is disrupted; therefore, signaling by both IL-4 and IL-13 is abolished. Although occasionally parasite burdens recovered from IL-4R^−/− or Stat6^−/− mice are greater than those recovered from IL-4^−/− mice (as in Fig. 2B), this appears to be the exception rather than the rule. More often, statistically identical parasite burdens are recovered from all three mutant strains (as in Fig. 2A), suggesting that IL-13 signaling does not compensate for the absence of IL-4 in a primary infection with B. malayi. Therefore, it is unlikely (although not impossible) that results observed in our adoptive transfer model are solely a reflection of a role for IL-13.

After the demonstration that IL-4 production is a T-cell function dispensable in host protection, we investigated the importance of T cells as targets of IL-4 signaling by using a similar adoptive transfer model. Due to mouse availability, completely immunodeficient SCID mice (lacking both B- and T-cell populations) were used as recipients, although one experiment utilizing T-cell-deficient recipients of IL-4R^−/− or wt splenocytes yielded similar results (data not shown). SCID mice have been shown to be highly permissive to infection with both B. malayi and B. pahangi (28). Since the absence of B cells results in a permissive phenotype after infection with B. pahangi (33), the total splenocytes were transferred in experiments utilizing SCID animals as recipients. As illustrated in Fig. 4B and C, lymphocyte reconstitution of these animals results in significant host protection, regardless of whether the donor population is from wt, IL-4R^−/−, or Stat6^−/− mice.

It is tempting to speculate from these data that although B cells are needed they also do not need to express the IL-4R. We should be cautious with this interpretation, since reconstitution of SCID animals with splenocytes has been shown to lead to inefficient B-cell repopulation of the recipient animals (55). In line with this observation, we routinely observed a lower frequency of repopulation of the B-cell compartment, in comparison to normal B-cell distribution, with splenocyte transfers (data not shown). Therefore, any conclusions as to the relative contribution of IL-4R expression by B cells to host protection exceeds the limitations of the experimental design (40).

One mechanism of host protection against B. malayi infections appears to be through formation of host cell “granulomas” that surround dead or dying parasites. These granulomas have been described in the literature and are made up predominantly of macrophages, giant cells, and eosinophils (18, 19, 26, 34, 45). Due to their involvement in granuloma formation, we have previously suggested that eosinophil recruitment to or accumulation at the infection site may be critical for host protection after infection with B. malayi or B. pahangi. Eosinophil numbers at the infection site are significantly lower in both TCRβ^−/− and SCID mice, suggesting that T cells may be involved in eosinophil recruitment and/or maintenance at the site of infection. Such a role for T cells is confirmed by the induction of eosinophil recruitment after the reconstitution of the T-cell compartment in these animals (Fig. 5A).

Similar to the defect observed in T-cell-deficient hosts, IL-4R^−/− mice also appear to experience a significant impairment in eosinophil recruitment, indicating a role for IL-4R expression in eosinophil recruitment as well. As demonstrated by the ability of IL-4R^−/− T cells to induce eosinophil influx in SCID mice (Fig. 5B), the requirements of IL-4R expression and T cells in mediating host protection are not congruent. Rather, whereas eosinophil recruitment appears to depend upon T cells, IL-4R expression seems to be required on nonlymphoid cells (Table 2). This may indicate a direct role for IL-4 signaling to eosinophils to induce their chemotaxis into the infection site, which would be supported by the finding of constitutive expression of the IL-4R on human eosinophils (12). It is also possible that IL-4 is acting on another cell type, such as the macrophage, to produce an eosinophil chemoattractant, a phenomenon previously described that occurs after B. malayi infection (15). Still another possibility is that IL-4 acts on the endothelial cells to induce transmigration of eosinophils, a phenomenon also supported in the literature (11).

Despite the preferential induction of a Th2 milieu after infection with B. malayi and the observation that T cells produce the majority of IL-4 at the site of infection, we demonstrated here that IL-4 production, as well as IL-4R expression are dispensable T-cell functions with respect to host protection. Rather, induction of eosinophil chemotaxis into the site of infection may be a more integral function of T cells in a murine model of B. malayi infection.

It is noteworthy to mention that several other models of murine nematode infections have demonstrated requirements for IL-4R signaling distinct from T cells in host protection. For example, host protection against Nippostrongylus brasiliensis and Heligosomoides polygyrus infections of mice requires IL-4R expression on intestinal epithelial cells (37, 49). It is tempting to speculate from these data that Th2 cells, per se, play no role in host protection. Further studies, namely, the determination of the cytokine profile produced by IL-4^−/−, IL-4R^−/−, and Stat6^−/− donor T cells after antigen exposure, would be required before this conclusion can be made. It is likely that these cells are still capable of producing Th2-like cytokines such as IL-5, IL-13, and IL-10, as has been previously demonstrated (16, 29). A more telling experiment would be the adoptive transfer of GATA3^−/− T cells into T-cell-deficient hosts prior to infection. However, the alternative possibility, namely, a T-cell-independent initiation of type 2 effector functions, is intriguing, especially in light of recent findings by Shinkai et al. demonstrating the induction of a polarized type 2 response in the absence of an adaptive immune system in mice exposed to N. brasiliensis (39). In that study, IL-4⁺, non-T, non-B cells were recruited to the lungs of infected, lymphocyte-deficient mice with kinetics similar to their wt counterparts, albeit in significantly lower numbers. CD4⁺-T-cell reconstitution of these animals resulted in the degranulation of eosinophils at the infection site, regardless of whether the T cells were isolated from wt, IL-4^−/−, IL-4/IL-13^−/−, or Stat6^−/− donors. This concept of initiation of a polarized response independent of adaptive immune cells is reminiscent of the induction of type 1 immunity through Toll-like receptor-mediated mechanisms (1, 35).

These results do not preclude the possibility that in a natural infection IL-4 signaling of and production by T cells is indeed an important component of the host immune response. Rather, these results demonstrate that, in the absence of T-cell-derived IL-4 or IL-4R expression by T cells, additional mechanisms exist to achieve parasite killing.