Underdeveloped nations are relatively protected from the worldwide asthma epidemic; the hygiene hypothesis suggests this is due to suppression of Th2-mediated inflammation by increased exposure to pathogens and their products. Although microbial exposures can promote Th2-suppressing Th1 responses, even Th2-skewing infections, such as helminths, appear to suppress atopy, suggesting an alternate explanation for these observations. To investigate whether induction of regulatory responses by helminths may counter allergic inflammation, we examined the effects of helminth infection in a murine model of atopic asthma. We chose Heligosomoides polygyrus, a gastrointestinal nematode, as the experimental helminth; this worm does not enter the lung in its life cycle. We found that H. polygyrus infection suppressed allergen-induced airway eosinophilia, bronchial hyperreactivity, and in vitro allergen-recall Th2 responses in an IL-10-dependent manner; total and OVA-specific IgE, however, were increased by worm infection. Finally, helminth-infected mice were protected against eosinophilic inflammation induced by adoptive transfer of OVA-stimulated CD4+ cells, and transfer of cells from helminth-infected/OVA-exposed mice suppressed OVA-induced eosinophilic inflammation, suggesting a role for regulatory cells. Increased CD4+CD25+Foxp3+ cells were found in thoracic lymph nodes of helminth-infected/OVA-exposed mice. Helminthic colonization appears to protect against asthma and atopic disorders; the regulatory cytokine, IL-10, may be a critical player.
Characteristic features of asthma include eosinophilic airway inflammation and airway hyperresponsiveness, both orchestrated by Th2 cytokines (1). Numerous studies show a substantial increase in the prevalence and morbidity of asthma and atopic disorders over the past few decades, especially in Westernized countries (2). Interestingly, as developing countries become more urbanized, and as populations migrate to Western nations, their asthma prevalence increases. In addition, within countries, asthma rates are often lowest among children raised on farms (3, 4). These data suggest that exposure to an extremely hygienic environment may increase the risk of children to develop atopic disorders such as asthma later in their life, which has been codified as the hygiene hypothesis (5). This hypothesis is supported by reduced atopic disorders among children with increased numbers of older siblings (6), those attending day care from a very early age (7), and those subjected to fecal-oral infections, such as salmonellosis, during infancy (8).
Mechanistic explanations for the hygiene hypothesis have focused on the role of microbes and microbial products on suppression of proatopic immune responses. In general, bacterial and viral infections promote protective Th1 cytokines; as Th2 and Th1 responses are counterregulatory (9), reduced Th1 could promote development of Th2 responses to Ags in the environment. Deviation toward Th1 responses by exposure to microbial products does not, however, adequately explain all aspects of the hygiene hypothesis. Importantly, the prevalence of type I diabetes (10), inflammatory bowel disease (11), and multiple sclerosis (12), all Th1-mediated diseases, have increased in concert with allergic disorders. Finally, a consistent negative relation between helminth infections and the prevalence of atopic responses has been seen (13, 14, 15, 16, 17), despite the fact that helminth infections strongly induce Th2-type responses, including high levels of IgE, and circulating eosinophils and mast cells (18). The induction of regulatory responses, characterized by IL-10-releasing lymphocytes, has been speculatively linked to immune modulation by helminths in a number of disease models (19, 20, 21).
Intestinal helminths that infect humans include roundworms (nematodes) and flatworms (cestodes): these are a diverse group of multicellular organisms that often use intermediate hosts during larval stages and a mammalian host for adult worms. Unlike other nematodes, such as hookworms and Ascaris, whipworms (Trichuris) do not require a pulmonary phase in their life cycle, nor is an intermediate host required: eggs are passed into the intestinal lumen, and fecal contamination of food allows ingestion of the infectious eggs and a new cycle of infection. Heligosomoides polygyrus, a naturally occurring murine nematode, is similarly restricted to the intestines of its host. Its infective stage is the third-stage larva (L3)4 that enters the muscular layer of the proximal small intestine within 24–48 h of ingestion (22); within 7–10 days, adult worms exit into the gut lumen, residing in the proximal small intestine and living in reasonable harmony with its host until being expelled (23).
In this study, we examined the effect of infection with intestinal helminths on manifestations of a murine model of asthma, to determine whether these unhygienic influences can suppress the development of atopic responses. We hypothesized that such protection would be characterized by induction of regulatory responses. We chose the organism H. polygyrus, a parasite that does not enter the lung during its life cycle and yet induces systemic inflammatory responses. We found that helminthic colonization prevents the subsequent development of allergen-induced eosinophilic airway inflammation and allergen-specific Th2 cytokine production, despite promoting overall increases in the general Th2 inflammatory milieu. These effects are dependent on the regulatory cytokine, IL-10.
Materials and Methods
Six-week-old female BALB/c mice were obtained from The Jackson Laboratory. Mice on a BALB/c genetic background with a disrupted gene for IL-10 were a gift from D. Berg, University of Iowa (Iowa City, IA) (24). All protocols were reviewed and approved by the University of Iowa Animal Care and Use Committee; all animal care and housing requirements of the National Institutes of Health Committee on Care and Use of Laboratory Animals were followed.
Infective, ensheathed L3 of H. polygyrus (U.S. National Helminthological Collection no. 81930) were propagated and obtained by the modified Baermann method and stored at 4°C until used.
Murine models of asthma and H. polygyrus infection
Helminth infection and atopic asthma model.
Mice were infected with 200 L3 of H. polygyrus in 200 μl of saline by gastric gavage on day −14 (H. polygyrus-colonized mice) or equal volumes of saline alone (noncolonized mice). All mice were sensitized to OVA (Sigma-Aldrich; 10 μg with 1 mg of alum, days 0 and 7). Mice were challenged with aerosolized OVA (1% solution, 30 min) or saline (sham challenge) on days 14 and 16, and sacrificed on day 18. Mice were grouped according to treatment: control, sham-challenged mice; worm, H. polygyrus-colonized/sham-challenged mice; OVA, OVA-challenged mice; worm/OVA, H. polygyrus-colonized/OVA-challenged mice (Fig. 1⇓A).
Modulation of eosinophilic airway inflammation induced by transfer of OVA-activated CD4+ cells
Donor BALB/c mice were sensitized by OVA/alum and challenged with aerosolized OVA, as above. At the time of sacrifice (day 18), splenocytes were isolated and stimulated in vitro with OVA (100 μg/ml) for 72 h, after which CD4+ cells were separated using anti-mouse CD4 mAb (clone GK1.5; Miltenyi Biotec) in a magnetic cell sorting system (Miltenyi Biotec). Recipient BALB/c mice were inoculated with 200 L3 of H. polygyrus or saline on day −14, as above. The CD4+ cells (5 × 106/mouse) isolated from donor mice were transferred to the recipients by tail vein injection on day 0. All recipient mice were challenged with inhaled OVA (1% solution, 30 min/day) on days 0–2, then sacrificed on day 4 (Fig. 1⇑B).
Adoptive transfer of splenocytes before allergen inhalation challenge
Donor mice (control, worm, OVA, worm/OVA) were prepared, as described above; splenocytes were isolated on day 18 of the protocol, and 1 × 107 splenocytes/mouse were transferred to OVA-sensitized recipient mice on day 13 of the protocol; all mice were then challenged with inhaled OVA and sacrificed, as above (Fig. 1⇑C).
General murine treatment
At the time of euthanasia, the trachea was cannulated, and saline washings, consisting of a total of 3.0 ml of 0.9% NaCl delivered by gravity feed in three aliquots, were collected; these bronchoalveolar lavages (BAL) were processed for BAL fluid (frozen for subsequent analysis) and total and differential cell counts (using Diff-Quik staining). Differential cell counts were performed twice by independent investigators who were blind to the grouping of the mice. Lungs of representative mice from each group were Formalin fixed, and sections were stained with H&E stain for examination by light microscopy. Phlebotomy was performed, and serum was isolated.
Some mice were evaluated for nonspecific airway hyperreactivity to inhaled methacholine. Airway physiology was assessed using a computer-controlled small-animal ventilator (Flexivent; Scireq) (25). For these studies, 48 h after the final OVA inhalation, mice were anesthetized with xylazine (12.5 mg/kg) and ketamine (87.5 mg/kg), and their trachea cannulated using a 20-gauge angiocath connected to the ventilator. After ventilation (at a rate of 150 breaths/min and a tidal volume of 10 ml/kg, with passive exhalation through a water trap maintained at a positive end-expiratory pressure of 2–3 cm H2O) was initiated, mice were paralyzed with pancuronium (1 mg/kg). The Flexivent system measures respiratory system impedance through application of a sinusoidal perturbation in inspiratory airflow. Measurements of ventilator piston position and airway pressure are used in a constant phase model to produce an index of resistance in the major conducting airways (Newtonian resistance). This forced oscillation technique correlates well with direct measures of airway resistance (26, 27). Mice were exposed to increasing concentrations of aerosolized methacholine (0, 12.5, 25, 50 mg/ml), delivered in-line, and Newtonian resistance was measured at each dose level, allowing for determination of the methacholine concentration that doubles the airway resistance (methacholine PC200).
In vitro assays
Single-cell suspensions of splenocytes (5 × 106 cells/ml/well) were plated in 24-well tissue culture plates in RPMI 1640 supplemented with 2 mM glutamine, 25 mM HEPES, 10% heat-inactivated FCS, 50 mM 2-ME, 100 U/ml penicillin, and 100 μg/ml streptomycin, and cultured alone or with OVA (100 μg/ml) or Con A (5 μg/ml; Sigma-Aldrich) at 37°C in an atmosphere of 5% CO2 in air. The supernatants were harvested after 72 h of culture and stored at −70°C until the measurement of cytokines.
Murine IL-5, IFN-γ, IL-10, total IgE, and total IgG2a were measured by ELISA (OptEIA; BD Biosciences), according to the manufacturer’s instructions. Serum levels of OVA-specific IgE and IgG2a were measured by ELISA in which plates were coated with OVA, and biotinylated rat anti-mouse IgE (clone R35-118) or anti-mouse IgG2a (clone R19-15, which recognizes an epitope in the CH3 domain of mouse IgG, with strong reactivity to the Igh-1a allotype) were the detection Abs (BD Biosciences). Units were OD readings at 405 nm.
Statistical significance was evaluated using the program SPSS 11.0 for Windows (SPSS). All data are expressed as the mean ± SE. Statistical differences between the groups were determined by a single factor ANOVA, followed by a post hoc Tukey test. Correlation was assessed using a bivariate analysis with calculation of Spearman’s coefficient. A p < 0.05 was considered significant.
H. polygyrus infection suppresses atopic asthma
As helminths are known to induce strong systemic eosinophilic inflammatory responses, it is counterintuitive to link infection with suppression of Th2-mediated atopic inflammation. We first examined whether enteral infection (Fig. 1⇑A) with the gastrointestinal nematode, H. polygyrus, would modulate airway eosinophilia in a murine model of OVA-induced atopic asthma. In OVA-sensitized mice, OVA inhalation induces marked airway eosinophilia; infection with H. polygyrus alone did not induce airway inflammation, but in OVA-challenged mice significantly blunted airway eosinophilia (Fig. 2⇓A). BAL eosinophil levels paralleled histopathologic demonstration of peribronchial airway inflammation (representative sections shown in Fig. 2⇓, B–E). Airway neutrophilia (modestly increased by OVA inhalation) was also suppressed in the helminth-infected mice (data not shown).
To correlate airway inflammation with the local cytokine milieu, we measured BAL fluid cytokine levels, and found that OVA-stimulated BAL IL-5 was significantly suppressed in mice infected with H. polygyrus (control, 20.1 ± 7.4 pg/ml; worm, 35.3 ± 7.4 pg/ml; OVA, 61.6 ± 7.0 pg/ml; worm/OVA, 35.5 ± 7.8 pg/ml; p = 0.004 for OVA vs control and p = 0.038 for OVA vs worm/OVA; n = 8/group in two separate experiments). BAL IL-10 and IFN-γ levels were not significantly changed.
Although eosinophilia is a critical marker of atopic inflammation, the diagnosis of asthma also requires bronchial hyperreactivity. OVA-challenged mice demonstrate significantly increased methacholine-induced airway resistance, compared with control mice (Fig. 2⇑F). This increase is suppressed by infection with H. polygyrus, resulting in a methacholine doubling dose (PC200) no different from control mice (methacholine PC200 control, 56.7 ± 7.3 mg/ml; OVA, 27.5 ± 4.8 mg/ml; worm/OVA, 63.9 ± 13.9 mg/ml; p = 0.03, OVA vs worm/OVA group). Comparing parameters of airway inflammation and bronchial hyperreactivity, we found a nearly significant inverse correlation between BAL eosinophilia and methacholine PC200 with a Spearman’s ρ of −0.32, at a significance level of 0.10. In aggregate, these data show that enteral H. polygyrus infection confers protection in a murine model of atopic asthma.
Infection with H. polygyrus modulates Igs
Atopy is characterized by elevations in specific IgE, and modulation of IgG subsets is associated with successful immunotherapy; thus, we examined the effect of H. polygyrus infection on Ig levels, specifically Th2-associated IgE and Th1-associated IgG2a. In contrast to the suppression of OVA-induced eosinophilia and, as expected, we found that mice infected with the helminths demonstrated significant elevations of IgE, substantially overshadowing the increase associated with OVA inhalation in these OVA-sensitized mice (Fig. 3⇓A). IgG2a, in contrast, was significantly suppressed by helminth infection (Fig. 3⇓B).
Some have suggested that polyclonal induction of IgE may prevent the effect of Ag-specific IgE; in that case, we may not see differences in Ag-specific Igs between helminth-infected and uninfected atopic subjects, but these specific Igs would have reduced functional effect. Thus, we next examined the concentrations of OVA-specific IgE and IgG2a. Unexpectedly, we found that OVA-specific Igs were modulated similarly to total Ig levels, with significant induction of OVA-specific IgE (Fig. 3⇑A) and suppression of OVA-specific IgG2a (Fig. 3⇑B) among the mice infected with H. polygyrus. OVA-specific IgG1, another Th2-associated Ig, was increased both by helminth infection and by OVA inhalation, but there was no significant difference between the OVA and worm/OVA groups in this case (data not shown). Clearly, the protective effects of helminth infection in atopic asthma cannot be accounted for by modulation of the Ig response, and suppression of airway inflammation by helminth infection occurs despite the setting of increased levels of Ag-specific IgE.
H. polygyrus infection modulates splenocyte cytokine responses
Suppression of eosinophilia by helminth infection might reflect modulation of the systemic inflammatory milieu, such as suppression of IL-5 release. We examined both nonspecific (following stimulation with the mitogen Con A) and Ag-specific (after OVA rechallenge) cytokine release by splenocytes; in the absence of stimulation, splenocytes from none of the groups consistently released detectable levels of cytokines.
Con A-induced splenocyte IL-5 release was significantly and similarly increased in helminth-infected, OVA-challenged, and helminth-infected/OVA-challenged mice, suggesting a nonspecific systemic Th2 bias induced by both the parasites and the atopic inflammation. In contrast, Ag-induced IL-5 release was not increased in the helminth-infected group, was markedly elevated in OVA-challenged mice, and was relatively suppressed (although elevated in comparison with control mice) in the worm/OVA group, demonstrating a clear reduction in Ag-specific Th2 responses following helminth infection (Fig. 4⇓A). Like IL-5, OVA-stimulated IL-4 release was not significantly increased in the helminth-infected group, markedly enhanced (p < 0.0001) in OVA-challenged mice, and this release was significantly suppressed in the helminth-infected/OVA-challenged mice (data not shown).
IFN-γ levels were greater following Con A stimulation than after OVA rechallenge, but the response patterns were similar: IFN-γ release by splenocytes from helminth-infected mice was no different from control mice, and release was suppressed in OVA-challenged mice, regardless of their prior exposure to helminths (Fig. 4⇑B). Therefore, the increased Con A-induced IL-5 in helminth-infected mice is not due to reduced Th1 bias, nor is the reduced OVA-stimulated IL-5 release in helminth-infected/OVA-challenged mice due to a Th1 response.
Finally, we measured IL-10, a cytokine that can be associated with both Th2 and regulatory responses. We found that nonspecific stimulation with Con A increased release of IL-10 in helminth-infected and OVA-challenged mice, and IL-10 release was greatest in helminth-infected/OVA-challenged mice (Fig. 4⇑C). OVA rechallenge, in contrast, induced IL-10 in OVA-challenged, but not in worm-infected mice; IL-10 release was relatively suppressed in the group of helminth-infected/OVA-challenged mice. These data, which parallel the changes in IL-5 release, are most consistent with the explanation that helminth-infected mice demonstrate reduced allergen-specific and enhanced nonspecific Th2 responses. They do not clearly point to the existence of helminth-induced regulatory mechanisms, but any such response could be obscured by the potent allergen-specific and nonspecific Th2 influence of OVA challenge and parasitic infection, respectively.
IL-10 is critical to the protective effects of H. polygyrus
Despite the inconclusive in vitro cytokine data, we chose to conclusively evaluate the importance of IL-10 in the protective effect of the helminth infection by subjecting H. polygyrus-infected and noninfected IL-10−/− mice to the atopic asthma protocol. Unlike wild-type mice, mice lacking the ability to produce IL-10 demonstrated no suppression of airway eosinophilia in response to helminth infection (Fig. 5⇓A), nor any reduction in airway hyperreactivity (data not shown). In vitro OVA rechallenge of IL-10−/− splenocytes induced IL-5 release that was not reduced in the helminth-infected group (Fig. 5⇓B); IFN-γ responses resembled those of wild-type BALB mice, with the exception that OVA alone did not suppress levels of this Th1 cytokine (Fig. 5⇓C). Ig responses were similar in wild-type and IL-10−/− mice, with enhancement of IgE in worm-infected mice (data not shown).
H. polygyrus infection induces regulatory mechanisms
The requirement for IL-10 in the protection against atopic asthma provided by enteric parasitic infection increases the likelihood that the helminths act through induction of a population of regulatory cells. First asking whether increased regulatory T cells could be identified in worm-infected mice, we evaluated cells isolated from spleen, lung parenchyma, and thoracic lymph nodes for CD4+CD25+ Foxp3+ cells. We found that expression of Foxp3 by CD4+ cells was increased (Fig. 6⇓A), as was the percentage of CD4+CD25+ cells that expressed Foxp3 (Fig. 6⇓B), among cells isolated from thoracic lymph nodes: this pattern was not seen in cells isolated from the lung or from the spleen.
Protection against passive atopic transfer (as opposed to a response that only inhibited sensitization) would functionally support a role for regulatory cells; thus, we examined the effect of prior infection with H. polygyrus on airway eosinophilia resulting from adoptive transfer of OVA-sensitized CD4+ lymphocytes, followed by OVA inhalation (Fig. 1⇑B). We found that mice infected with H. polygyrus develop significantly less airway eosinophilia than noninfected mice (Fig. 7⇓A), suggesting induction of active or regulatory response by the parasites. In parallel with the number of eosinophils, OVA-induced IL-5 in BAL fluid was also significantly suppressed in the infected mice (uninfected mice, 89.5 ± 9.6 pg/ml; helminth-infected mice, 36.6 ± 6.5 pg/ml, p < 0.01).
We next examined whether the protective effect engendered by parasite infection could be transferred. For these studies, splenocytes were isolated from each of four groups of mice (control, worm, OVA, and worm/OVA) and transferred to OVA-sensitized recipient mice 1 day before the first of two OVA inhalation challenges (Fig. 1⇑C). We found that transfer of cells from either helminth-infected (worm) mice or OVA-challenged (OVA) mice significantly increased OVA-induced eosinophilia in the recipient mice, but that transfer of splenocytes from helminth-infected/OVA-exposed (worm/OVA) mice did not induce this response (Fig. 7⇑B), suggesting induction of a population of cells that requires exposure to Ag as well as the stimulation of helminth infection for optimal benefit.
Recent epidemiological studies support the hygiene hypothesis, that children raised in extremely clean environments lack protective anti-atopic mechanisms, resulting in increased susceptibility to atopic disorders (5). Based on the immunological framework of Th1 and Th2 balance (9), an increased prevalence of atopic disease could be due to unopposed Th2 responses from reduced exposure to (microbial-induced) Th1 responses. However, this explanation does not account for the observations that even Th2-skewing organisms such as helminths appear to protect against asthma and atopy (13). Although experimental models of Th1-skewing infections, such as mycobacteria, have demonstrated suppression of Ag-induced eosinophilic airway inflammation in murine models of asthma (28, 29), few studies have examined the protective effect of helminths in allergic inflammation (20, 30). Wohlleben et al. (20) used the hookworm Nippostrongylus brasiliensis, which traverses the lung during its life cycle, as an immune stimulus; in contrast, the effects of H. polygyrus do not require its presence in the lung. Quite recently, Wilson et al. (30) reported that H. polygyrus is effective in suppressing atopic inflammation in a murine model of asthma; in this study, we confirm and extend their findings.
In this study, we demonstrate that intestinal helminths protect against atopic responses in a murine model of asthma. We first showed that prior helminth infection suppressed OVA-induced airway eosinophilic inflammation, bronchial hyperreactivity, and in vitro OVA-induced IL-5 release from splenocytes. This was not associated with a Th1-type immune response: indeed, both total and Ag-specific IgE levels were enhanced, and IgG2a levels suppressed, in the worm-infected mice. Stimulation of splenocytes with either mitogen or Ag showed that cells from OVA-challenged mice have a Th2-skewed response; OVA-stimulated (but not Con A-stimulated) IL-5 release is suppressed in the helminth-infected OVA-challenged mice. Induction of IL-10 by Con A is enhanced, and by OVA is suppressed, in the helminth-exposed mice. We did not find a convincing elevation of IL-10 (nor of TGF-β) in the BAL fluid of infected C57BL6 or BALB/c mice. Nevertheless, the protective effects of H. polygyrus infection are not seen in the absence of IL-10; this is in contrast to the findings of Wilson et al. (30), who reported no significant reduction in the anti-inflammatory response to H. polygyrus following administration of anti-IL-10R Abs. This discrepancy could be due to an incomplete ablation of IL-10 responses using an Ab, in comparison with the complete abrogation of the cytokine in the knockout mouse; however, the ability of cells transferred from H. polygyrus-infected IL-10−/− mice in that study suggests that other pathways, such as TGF-β responses, may also be induced.
IL-10 plays a complex role in the modulation of inflammation and airway responses in asthma. In regard to eosinophilia, by reducing expression of IL-5, IL-10 generally functions to suppress inflammation. Nevertheless, reports in the literature do not consistently find increases in Ag-induced airway eosinophilia in the absence of IL-10; some find reduced levels of eosinophils (31, 32), whereas others report modest to significant increases in this parameter of atopic inflammation (33, 34, 35). In part, this may be due to modulation of the Th1 arm, as IL-10 also suppresses expression of IFN-γ (36, 37); indeed, we have shown marked enhancement of the Th1 response to the TLR9 agonist, CpG DNA, in the absence of IL-10 (38). But mice genetically deficient in IL-10 do not necessarily have an overactive Th1 response in the absence of Th1 inducers. In this study, IFN-γ release by splenocytes of mice genetically deficient in IL-10 is similar to that from splenocytes of wild-type BALB/c mice.
The effect of IL-10 on airway hyperreactivity seems independent of effects on inflammation. Although some investigators have reported that airway hyperreactivity is not seen in the absence of IL-10 (31, 39), others have found similar (34) or only moderately reduced (40) levels of hyperreactivity in IL-10−/−, as well as elevated baseline airway resistance (33) compared with wild-type mice. We have found (using IL-10−/− mice on the BALB/c background) that these mice develop airway hyperreactivity in conjunction with airway inflammation; levels of both responses are modestly reduced compared with wild-type mice.
Regulatory mechanisms are suggested by the ability to transfer protection and the effect of helminth infection on passively transferred sensitization. CD4+CD25+Foxp3+ regulatory cells were increased within the population of CD4+ (and CD4+CD25+) cells within the thoracic lymph nodes of H. polygyrus-infected, OVA-challenged mice. These may be the cellular mediators for the protective effect of helminth infection, although other cell populations (including CD4− cells) were found by Wilson and Maizels (41) to offer some protection.
The protective responses engendered by H. polygyrus in BALB/c mice were not strain specific; we found (data not shown) that C57BL/6 mice were equally protected against atopic airway inflammation. We further examined whether CpG-rich DNA potentially released from worms was involved in the immunomodulation, and saw that TLR9−/− mice (on a C57BL/6 background) were equally protected against eosinophilic airway inflammation as their parent strain (data not shown).
As previously reported, we found that colonization with H. polygyrus enhances total IgE (42). It has been suggested (the IgE blocking hypothesis) that elevated polyclonal IgE saturates the FcεR (IgE receptor), which may prevent the binding of Ag-specific IgE to mast cells and other cells central to atopic inflammation, resulting in reduced atopic symptoms. In our current study, we found that IL-10−/− mice developed equally high levels of IgE as was seen in wild-type mice, suggesting that induction of polyclonal IgE is not likely to be an important mechanism of helminth-induced protection in this model; moreover, recent clinical studies have cast doubt on the IgE blocking hypothesis (15, 43, 44, 45). In addition to increasing IgE, helminth colonization suppressed Th1-associated IgG2a; similar patterns were seen in OVA-specific and in total Ig levels. These effects support a systemic Th2-type (and relative suppression of Th1-type) skewing of the immune system by worm infection. Thus, regulation of Ig responses did not parallel regulation of airway eosinophilia. Supporting our conclusion that induction of Th1 responses is not involved in helminth-provided anti-atopic responses, mice deficient in IFN-γ and IL-12 (IFN-γ−/−IL-12−/− on a C57BL/6 background (46)) demonstrated equal protection and similar immune responses (data not shown).
It has been shown that heat-killed Mycobacterium vaccae suppress Ag-induced eosinophilic airway inflammation by induction of Ag-specific CD4+CD45RBlow regulatory T cells (47) that depend on IL-10 and TGF-β. Although, in our study, the protective effects of colonization with H. polygyrus required the ability to generate IL-10, we did not find Ag-specific regulatory responses; IL-10 release by OVA-restimulated splenocytes was actually somewhat suppressed in helminth-infected mice. However, in these studies, all of the mice were first sensitized to OVA; the suppression of IL-10 release in helminth-infected mice may represent reductions in Ag-responsive Th2 cells. Indeed, the induction of IL-10 by mitogen stimulation could either reflect a larger non-Ag-specific Th2 population, or alternatively, release of cytokine from active regulatory cells. A number of reports have demonstrated the induction of T cells with non-Ag-specific regulatory properties. These include so-called type 1 T regulatory cells that may release both IL-10 and IL-5, but whose effects are IL-10 dependent, as we have seen in these studies (48, 49, 50).
Our findings that infection with a helminth that does not traverse the lung during its life cycle can prevent atopic airway inflammation are consistent with the hygiene hypothesis. This study bolsters, using an animal model, the epidemiologic observations that Th2-skewing helminth infections are associated with a reduced prevalence of clinically significant atopic disorders. In this model, Th1-type responses do not play a role, supporting the growing consensus that such responses are not central to the protection against atopic inflammation associated with microbes and their products; induction of a regulatory cell population is more likely the responsible mechanism. We have previously reported that the anti-asthmatic protective effects of CpG oligodeoxynucleotides, which mimic exposure to microbial DNA, induce, but are not dependent on Th1 responses (46). In that model system as well, IL-10 plays an important role in suppressing Th2-skewed responses (38). Increasing evidence supports the speculation that antiallergic effects of both Th1- and Th2-skewing microbial stimuli may be due to a common cellular mechanism, although the proximal innate immune receptors may differ. Clarification of these common mechanisms remains an important line of research.
Thanks to Joseph Zabner for critical review of this manuscript.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This work was supported by the following grants: HL59324, HL079447, HL079448, and ES05605.
↵2 Current address: Division of Gastroenterology, Tufts-New England Medical Center, Boston, MA 02111.
↵3 Address correspondence and reprint requests to Dr. Joel N. Kline, C33GH University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, IA 52242. E-mail address:
↵4 Abbreviations used in this paper: L3, third-stage larva; BAL, bronchoalveolar lavage.
- Received November 10, 2005.
- Accepted May 10, 2006.
- Copyright © 2006 by The American Association of Immunologists