The Lung Is an Important Site for Priming CD4 T-Cell-Mediated Protective Immunity against Gastrointestinal Helminth Parasites

Mice.The G4/IL-4 (on a BALB/c background), STAT6^−/−, G4/G4 (IL-4^−/− in a BALB/c background), IL-5^−/− B6Aao (MHC [major histocompatibility complex] class II^−/−), BALB/c, μMT (B cell^−/−), CD1^−/−, and C57BL/6J mice used in these experiments were bred by the Biomedical Research Unit, Malaghan Institute of Medical Research, Wellington, New Zealand. Six 10-week-old age- and sex-matched mice were used in all of the experiments. All of the experimental procedures described in this report were approved by the Victoria University Animal Ethics Committee and carried out in accordance with the guidelines of the Victoria University of Wellington, Wellington, New Zealand.

Maintenance, isolation, and use of the L3, L4, and L5 stages of N. brasiliensis. N. brasiliensis was originally sourced from Lindsay Dent (University of Adelaide) and has been maintained by monthly passage through Lewis rats for 15 years with additional selective passage in immunodeficient and immunocompetent mice at previous times in its life history to improve parasitism in the mouse host. iL3 worms were prepared from 2-week rat fecal cultures, with careful preparation of 100% viable iL3 worms being essential for consistent infection studies with mice (4). Viable motile fourth-stage larval forms of N. brasiliensis (L4 worms) were isolated by migration from in vitro cultures of 2-day postinfection lung tissues, followed by extensive washing in phosphate-buffered saline (PBS) (4). The homogeneity and viability of L4 worms were confirmed by microscopic analysis. Lung infection of mice with L4 worms was achieved by the intranasal (i.n.) administration of ∼200 freshly harvested L4 worms in a 50-μl drop to the noses of mice which were lightly anesthetized with xylazine and ketamine for the duration of the inoculation procedure. Regular breathing of the mice causes the drops containing L4 worms to be inspired, with the viable L4 worms being detected in lung tissues following this procedure. L5 adult worms were prepared by migration from the intestines of infected mice, washed extensively in PBS, and counted, and their viability was determined by microscopy (4). Infection of mice with L5 worms was performed by gavage of ∼300 freshly harvested L5 adult worms. The anthelminthic pyrantel embonate was administered where specified by oral gavage daily at 1 mg/mouse. Subcutaneous (s.c.) infections were performed by inoculating mice in the scruff of the neck with 600 dead or live iL3 worms in a volume of 200 μl (4). For intradermal (i.d.) infections, mice were anesthetized using xylazine and ketamine (Phoenix, New Zealand) and 30 μl containing 600 dead or live iL3 worms was injected into the ear pinnae using a BD Ultrafine 29-gauge needle and syringe (Becton Dickinson, New South Wales, Australia).

CFSE labeling of live parasites.iL3 worms were washed several times in PBS, incubated at room temperature for 8 min in 2.5 mM carboxyl fluorescein succinimidyl ester (CFSE), and then washed in PBS three times before infection of the host.

Preparation of dead iL3 worms.iL3 worms were harvested from fecal cultures and washed five times in sterile PBS. Larvae were made up to a concentration of 600 L3 worms per 30 μl in PBS and put through three freeze-thaw cycles. Larvae were then checked for loss of viability by microscopic analysis.

In vitro skin penetration assay.The abdomen skin was shaved, and 2 cm² was harvested from naïve or previously infected mice. Skin was placed on 15-ml tubes containing PBS. Two hundred L3 worms were added to each tube and incubated overnight at 37°C. Larvae penetrating the skin accumulated at the bottom of the tube. Skin was then incubated in PBS to cause the larvae remaining on the skin to migrate out. This method is an adaptation of that of Brown et al. (3).

Imaging of larvae in ear tissue.CFSE-labeled L3 worms were injected i.d. into the ear pinnae. The ears were excised, and the dorsal and ventral sheets were separated. Images were captured using an Olympus BX51 microscope, an Olympus DP70 camera, and AnalySIS LS software.

Viable worm recovery assay.Tissue (lung or gut) was diced, placed on cheesecloth, and suspended in a 50-ml centrifuge tube containing PBS at 37°C for at least 2 h. For recovery of worms from inoculated ears, the dorsal and ventral sheets were separated, placed on cheesecloth, and suspended in PBS at 37°C. Viable worms migrated out and accumulated at the bottom of the tube before being counted on a gridded counting plate.

Cell preparation and flow cytometric analysis of lymphocytes.Mediastinal and mesenteric lymph nodes were isolated and pressed through a 70-μm cell strainer to make a single-cell suspension. Isolated cells were then resuspended in fluorescence-activated cell sorting buffer, and Fc receptors were blocked (2.4G2) before staining with monoclonal antibodies against CD4 (GK1.5) and CD3 (2C11) prior to collection on a FACScalibur flow cytometer (BD). Analysis was performed with Flowjo software (Treestar).

Lung histopathology.Lung tissues were harvested from naïve and N. brasiliensis-infected mice into 10% formalin before embedding in paraffin. Mucus-containing goblet cells were detected by staining of 4-μm-thick slice specimens with periodic acid-Schiff (PAS) and alcian blue. All bronchioles with a diameter of 200 to 600 μm were counted in both lungs, with the number of PAS-positive cells per bronchiole being reported. Further sections were stained with hematoxylin and eosin (H&E) to assess the inflammatory infiltrate.

Statistical analysis.Statistical analysis was performed with Prism software using two-tailed Mann-Whitney U testing.

Kinetics of larval elimination and Th2-type immunity development during primary and secondary N. brasiliensis infections.To determine if the induction of Th2-type immunity directly correlates with the worm burden, we used the viable worm recovery assay to compare the kinetics of larval recovery with Th2 development at different tissue sites of infected, naïve, or reinfected G4/IL-4 mice. iL3 worms were inoculated by the i.d. (Fig. 1 A) or s.c. (Fig. 1C) route to ensure a more uniform infection than experimentally possible by natural penetration of the skin by iL3 worms. Indeed, the kinetics of infection and fecundity (unpublished observation, data not shown) of adult L5 worms did not appear to be affected when different injection routes were used (Fig. 1C, s.c. exposure; see Fig. S1A in the supplemental material for i.n. exposure) and a linear relationship between the number of L3 worms inoculated and the number of viable worms which could be recovered from the lungs was achieved in the dose range used in these experiments (see Fig. S2 in the supplemental material for s.c. inoculation). It is important to note that the degree of worm recovery from either lung or gut tissue never exceeded 50 to 60% of the input inoculum, even in completely immunodeficient mice, indicating that there is a natural attrition of worm numbers, probably due to nonhost factors (see Fig. S2 in the supplemental material; data not shown). Our attempts to enumerate worm migration and trapping by histological analysis proved problematic due to the relatively long, hollow, tubular shape of the parasite. Therefore, we developed a CFSE staining (Fig. 1B) and viable worm recovery protocol adapted from techniques of Camberis et al. (4) for the accurate quantification of viable versus trapped worms in tissues. In naïve mice, the majority (>80%) of the L3 worms injected into the ear pinnae rapidly disappeared (5 to 60 min) from the i.d. injection site during a primary infection (Fig. 1A). Worms were observed in the lungs from 24 to 72 h after s.c. infection (kinetics were identical after i.d. infection, data not shown) and then accumulated in the gut, with the worm burden peaking at day 6 postinfection (Fig. 1C). When the activation of GFP⁺ Th2 CD4 T cells in the draining lung and gut lymph nodes was followed during the same primary infection, the numbers of GFP⁺ Th2 cells per lymph node peaked at day 9 postinfection, after 90% of the L5 worms had been expelled from the intestine of the host (Fig. 1E). To follow the fate of worms during reinfection, previously s.c. infected mice were rested for a total of 30 days postinfection to allow the cellular inflammatory response, indicated by the number of CD4⁺ GFP⁺ cells in the draining lymph nodes of the lungs (see Fig. S3B in the supplemental material), to subside and then reinfected s.c. with N. brasiliensis L3 worms. Surprisingly, despite an increase in cellularity of the previously infected ear tissue (data not shown), the iL3 worms were not trapped in the skin as we expected and were able to migrate from the skin site with a speed and efficiency equal to those of worms infecting a naïve mouse, with 80% of the larvae gone by 60 min (Fig. 1A). However, when the lungs of s.c. reinfected mice were analyzed for worm burdens (Fig. 1D), there was a dramatic 10-fold reduction in the number of recoverable L3/L4 worms throughout the 24- to 72-h postinfection period compared to the worm burdens in lungs of identically s.c. infected naïve mice (Fig. 1C). Similarly, in the guts of s.c. reinfected mice, a dramatic 10-fold reduction in the L5 worm burden was observed (Fig. 1D). The lack of evidence for trapping of worms in the skin of reinfected mice was confirmed by the finding that identical skin, lung, and gut worm recoveries were obtained in reinfected mice, whether the skin reinfection site was the same as, or distant from, the site of the primary skin infection (data not shown). Of the most significance was the finding that the reduction in worm burden seen in s.c. reinfected mice correlated with the rapid induction of a CD4 GFP⁺ T-cell response in the lung draining mediastinal lymph nodes that was strikingly faster (detectable at 48 h) and 20-fold greater than the CD4 GFP⁺ T-cell response seen in the lymph nodes of s.c. infected naïve mice (Fig. 1E and F). Similarly, faster and increased CD4 GFP⁺ responses were detected in the mesenteric lymph nodes of s.c. reinfected mice, although the duration of the CD4 GFP⁺ response was greatly reduced. This was most probably due to the significant decrease in L4 and L5 worms making it to the guts of the reinfected mice compared to those of naïve mice.

• Open in new tab
• Download powerpoint

FIG. 1.

Reduction in worm burden correlates with induction of CD4 GFP⁺ Th2 immune response in mice reinfected with N. brasiliensis. (A) Number of larvae migrating from site of skin infection. G4/IL-4 mice (n = 3 per time point) were infected i.d. in the ear pinna with 600 L3 N. brasiliensis worms (open squares) or left uninfected (closed squares), rested, and challenged by i.d. reinfection with 600 L3 N. brasiliensis worms. At the time points indicated, viable larvae that migrated out of excised ears were enumerated. Values are expressed as percent burden ± the standard error of the mean (SEM) and are representative of three experiments. (B) Representative photomicrograph of a mouse ear 5 min after the injection of CFSE-labeled L3 N. brasiliensis worms. Inset, differential interference contrast versus fluorescence microscopy of the same worms. The worms pictured are 0.5 mm in length. (C) Parasite burden during s.c. primary N. brasiliensis infection. G4/IL-4 mice (n = 4) were primed with 600 L3 N. brasiliensis worms s.c. Lung (closed squares) and gut (open squares) tissues were harvested, and viable worms that migrated out were enumerated. The values shown represent the mean number of worms recovered ± the SEM and are representative of three independent experiments. (D) Parasite burden during N. brasiliensis reinfection. G4/IL-4 (n = 4) mice were primed with 600 L3 N. brasiliensis worms s.c., rested, and reinfected s.c. with 600 L3 N. brasiliensis worms. Lung (closed squares) and gut (open squares) tissues were harvested, and viable worms that migrated out were enumerated. The values shown represent the mean number of worms recovered ± the SEM and are representative of three independent experiments. (E) GFP⁺ CD4 T-cell responses during primary s.c. N. brasiliensis infection. Mediastinal (closed squares) and mesenteric (open squares) lymph nodes from s.c. N. brasiliensis-infected animals (n = 4) were harvested and surface stained for CD3 and CD4. The values shown represent the mean number of cells ± the SEM and are representative of at least three independent experiments. (F) GFP⁺ CD4 T-cell responses during N. brasiliensis reinfection. Mediastinal (closed squares) and mesenteric (open squares) lymph nodes from s.c. N. brasiliensis-infected animals (n = 4) were harvested and surface stained for CD3 and CD4. The values shown represent the mean number of cells ± the SEM and are representative of at least three independent experiments.

Immune skin does not protect against N. brasiliensis L3 worm invasion.The reduction in viable worm recovery from the lungs and guts of s.c. reinfected mice implied that adaptive immune-mediated mechanisms either reduced the viability of the L3, L4, and L5 worms or blocked their migration through the host to reach adulthood in the gut. Previous studies have already demonstrated that L3 worms can be trapped at the initial inoculation site and fail to migrate to the lung, especially in situations where there is an abundance of eosinophils accumulating from previous infections or aberrant IL-5 cytokine expression (6, 17). To determine whether skin of previously s.c. infected mice prevented the migration of L3 worms either through tissue remodeling or chronic accumulation of residual immune cells, we adapted an in vitro assay to follow the migration of L3 worms through skin (Fig. 2 A). First we examined the ability of L3 worms to migrate through skin that has been isolated from either naïve or previously s.c. infected/immune mice. Worms were placed on either naïve or immune skin, a temperature gradient was applied, and the worms that migrated through the skin were enumerated. No difference was observed in the ability of worms to migrate through skin from previously s.c. infected, hence immune, animals or skin from naïve mice (Fig. 2B). In a separate experiment, it was observed that infection with living worms was necessary for the development of protection, as mice vaccinated i.d. with dead L3 worms were not protected from s.c. reinfection (Fig. 2C). A time course analysis of the duration of the protective effect in the lung and gut established that the reduction in worm burdens in the lung began to wane after 180 days but the ability to reduce the burden of adult worms in the gut was maintained essentially for the life of the mouse (see Fig. S3A in the supplemental material). Taken together with the results in Fig. 1, these data indicate that the reduction of lung and gut worm burdens in reinfected mice is not due to the worms being trapped in the skin but rather to largely undetermined tissue immune-mediated events leading to the failure of adult worms to establish an infection in the GI tract.

• Open in new tab
• Download powerpoint

FIG. 2.

The skin is not the site of protection from reinfection. (A) Schematic representation of in vitro skin migration assay. Skin harvested from naïve or previously s.c. infected mice (n = 4 per group) was suspended over PBS. iL3 worms were placed on the skin samples, and those that migrated through were enumerated. (B) L3 worm skin migration efficiency in vitro. L3 N. brasiliensis worms able to migrate through naïve or immune skin (as described for panel A) were enumerated. The values shown represent the mean number of larvae that migrated ± the SEM. (C) Priming with dead worms compromises protection. G4/IL-4 mice (n = 3 per group) were inoculated i.d. in the ear pinna with 600 dead or live L3 N. brasiliensis worms, rested, and challenged s.c. with 600 live L3 N. brasiliensis worms. Lungs were excised on day 2 postinfection, and the viable worms present were enumerated using the migration assay described in Materials and Methods. Data show the mean number of worms that migrated out ± the SEM. All experiments are representative of three experiments.

Priming of CD4 T cells in the lung is necessary for protective immunity against infective N. brasiliensis L3 worms.Having established that primary s.c. infection with N. brasiliensis generated long-lived adaptive responses which could significantly reduce the burden of viable of L3/L4 worms that could establish an infection in reinfected mice, we wished to determine in which tissues and with which immune mechanisms the protective immunity was induced and active. Therefore, we developed tissue-specific (skin, lung, or gut) priming models using distinct phases of the N. brasiliensis life cycle (illustrated in Fig. 3) to identify the site and source of protective immunity against reinfection. It was already apparent that skin-localized priming with dead N. brasiliensis was insufficient to confer protective immunity (Fig. 2C). To determine the role of the circulation, lung, and gut in the protective response, anthelminthic treatment was combined with inoculation with the respective skin, lung, and gut tissue stages of N. brasiliensis as follows. In group A, skin and lung (S/L) priming was achieved by s.c. infection with L3 worms and infection of the GI tract was prevented by anthelminthic treatment from day 2 to day 9 of infection. In group B, lung-only (L) priming was achieved by i.n. administration of L4 worms and infection of the GI tract was prevented by anthelminthic treatment from day 2 to day 9 of infection. In group C, gut-only (G) priming was achieved by gavage of adult L5 worms. In group D, skin, lung, and gut (S/L/G) priming was achieved by conventional s.c. infection with L3 worms. It is important to note that L4 worms introduced i.n. into non-drug-treated mice were able to complete their life cycle to the L5 stage and produce eggs (unpublished observation, data not shown), indicating that they were not affected by the extraction procedure (see Fig. S1 in the supplemental material). Also, adult L5 worms introduced into BALB/c mice by gavage were cleared with normal kinetics (expulsion day 1 to day 5 after gavage) while those introduced into class II MHC^−/− or STAT6^−/− mice by gavage failed to be cleared, indicating that the L5 worms were not affected by the extraction and gavage procedures (see Fig. S1B in the supplemental material).

• Open in new tab
• Download powerpoint

FIG. 3.

Schematic representation of the truncated infection strategy used for selective infection of lung and gut tissues. (A) Skin and lung priming. Mice were infected with ∼600 L3 N. brasiliensis worms s.c. and given an anthelminthic daily by gavage from day 2 to day 9 to prevent the establishment of L4/L5 worms in the gut and prevent the induction of a gut-localized immune response to N. brasiliensis. Mice were s.c. reinfected with 600 L3 N. brasiliensis worms at day 30 postinfection. (B) Lung priming. Mice were i.n. administered ∼200 L4 N. brasiliensis worms and given an anthelminthic daily by gavage from day 2 to day 9 to prevent the establishment of L4/L5 worms in the gut and prevent the induction of a gut-localized immune response to N. brasiliensis. Mice were s.c. reinfected with 600 L3 N. brasiliensis worms on day 30 postinfection. (C) Gut priming. Mice were infected with ∼300 L5 N. brasiliensis worms by gavage. Mice were s.c. challenged with 600 L3 N. brasiliensis worms on day 30 postinfection. (D) Skin, lung, and gut priming. Mice were infected with ∼600 L3 N. brasiliensis worms s.c. Mice were s.c. reinfected with 600 L3 N. brasiliensis worms on day 30 postinfection. (E) No-priming (naive control) naïve mice were s.c. infected with ∼600 L3 N. brasiliensis worms at the same time as the groups shown in panels A, B, C, and D were s.c. reinfected to enable comparison of worm burdens and calculation of the level of protection.

Once the tissue-specific priming infections had cleared and cellular immune responses had dissipated (days 25 to 30 postinfection; see Fig. S3B in the supplemental material), all of the test groups and a group of naïve mice (group E) were challenged by s.c. reinfection using 600 iL3 worms and the worm burdens, cellular responses, and protective immune responses were compared (Fig. 4 A). One limitation of this experimental model is that it cannot determine the specific priming and attrition of larvae after migration from the skin and prior to arrival in the lung. Evidence in Fig. 1C and D shows that the same number of worms arrived in the lungs of naïve and protected mice at 24 h after inoculation, but there was no additional accumulation of viable worms after this point in previously s.c. infected mice (Fig. 1D) compared to naïve mice (Fig. 1C). This suggests that protective mechanisms are initiated after 24 h within the lung; however, this experiment does not exclude the possibility that protective mechanisms could exist before entry into the lungs.

• Open in new tab
• Download powerpoint

FIG. 4.

Effect of truncated skin, lung, and gut primary N. brasiliensis infection on the development of immunity to reinfection. G4/IL-4 mice (n = 4) were infected via the skin and lungs, the lungs only, the gut only, or whole infection. The anthelminthic was given from day 2 to day 9 to the skin-and-lung infection and lung infection only groups. Following the 30-day rest period postinfection, mice were s.c. infected with 600 N. brasiliensis L3 worms and on day 2 their lungs were excised and the worms that migrated out were enumerated. The values shown represent the mean number of larvae present in the lung ± the SEM. The data shown are representative of three separate experiments. (B) Worm burdens in the gut following priming with distinct phases of the N. brasiliensis life cycle. G4/IL-4 mice (n = 4) were infected via the skin and lungs, the lungs only, the gut only, or whole infection, and an anthelminthic was given from day 2 to day 9 to the skin-and-lung and lung-only groups. Following a rest period, mice were s.c. infected with 600 N. brasiliensis L3 worms and on day 6, their guts were excised and the worms that migrated out were enumerated. The values shown represent the mean number of larvae present in the gut ± the SEM. The data shown are representative of three separate experiments. (C) GFP⁺ CD4 T-cell responses in mediastinal lymph nodes during distinct phases of the N. brasiliensis life cycle. Mediastinal lymph nodes from animals (n = 4) infected via the skin and lungs, the lungs only, the gut only, or whole infection were harvested on day 2 and surface stained for CD3 and CD4. The values shown represent mean cell numbers ± the SEM and are representative of three independent experiments. (D) GFP⁺ CD4 T-cell responses in mesenteric lymph nodes during distinct phases of the N. brasiliensis life cycle. Mesenteric lymph nodes from animals (n = 4/group) infected via the skin and lungs, the lungs only, the gut only, or whole infection were harvested on day 6 and surface stained for CD3 and CD4. The values shown represent mean cell numbers ± the SEM and are representative of three independent experiments.

Mice whose primary infection was restricted to the skin and lungs (S/L and L) significantly reduced the number of recoverable L4 worms in the lungs compared to the high numbers found in naive s.c. infected mice (Naïve). In contrast, mice whose primary infection with N. brasiliensis was restricted to the gut using L5 worms (G) introduced by gavage failed to significantly reduce their lung worm burdens when s.c. reinfected with iL3 worms. Analysis of the lymph nodes draining the tissues selectively infected with N. brasiliensis (s.c., i.n., or by gavage) revealed that the presence of more than 10⁴ CD4 GFP⁺ cells per mediastinal lymph node, which occurred in mice primed in the skin and lung, the lung only, or the skin, lung, and gut (S/L, L, and S/L/G), correlated with a significant reduction in the worm burden (Fig. 4C). By comparison, significantly lower numbers of CD4 GFP⁺ cells were found in the mediastinal lymph nodes of gut-primed (G) and naive mice (Naive).

In contrast to the apparent need for lung-specific priming by N. brasiliensis to get reductions in the lung worm burden when s.c. reinfected, all of the mice that had received a skin-, lung-, or gut-only primary infection (s.c., i.n., or by gavage, respectively) significantly reduced the numbers of adult L5 worms in the gut at day 6 following reinfection with iL3 worms (Fig. 4B). This protective effect correlated with significantly higher numbers of CD4 GFP⁺ T cells in the mesenteric lymph nodes (Fig. 4D) of all s.c. reinfected mice. Importantly, lung-specific priming with N. brasiliensis was sufficient to protect the mouse GI tract from challenge infection with adult L5 worms (see Fig. S4 in the supplemental material). The protection induced by lung infection appears to be restricted to s.c. or i.n. infection with live worms, as priming with dead L4 worms by either i.d. injection into the skin (see Fig. S5B in the supplemental material) or i.n. administration to the lung did not confer protection (see Fig. S5A in the supplemental material).

In conclusion, our finding that the s.c. reinfected mouse lung worm burden was only reduced in mice whose lungs were exposed to N. brasiliensis indicates that lung-only infection is sufficient to generate protective immunity against tissue-migrating parasites. The important influence of the lung as a site for the generation of protective immunological responses was emphasized in the study demonstrating that immune protection in the gut could be achieved when the primary (s.c. or i.n.) infection only occurred in the lung (see Fig. S4 in the supplemental material). The ability of priming immune responses in the gut (by gavage) to generate effective protective immunity against gut reinfection with L5 worms was consistent with previous studies.

Increased airway inflammatory responses correlate with reduced viable worm recovery in reinfected mice.We used histological analysis to follow the fate of L3 worms migrating to the airways of reinfected mice. Although the detection and reliable enumeration of nonviable worms in lung tissue proved problematic, the inflammatory response stimulated by the worms could be used as a surrogate for their presence in tissues. Lung tissue was taken from mice during the peak (day 9) of primary s.c. infection, following resolution (day 29) of the primary s.c. infection, or during the peak (day 6) of s.c. reinfection and prepared for histological analysis (Fig. 5). Examination of H&E-stained lung sections revealed marked and diffuse leukocyte infiltration of the lung tissue during primary N. brasiliensis infection (Fig. 5B and F). This infiltration appeared to have largely resolved by day 29 after s.c. infection (Fig. 5C and G). However, upon s.c. reinfection, extensive inflammation and leukocyte infiltration were observed as early as day 6 after s.c. reinfection, with the degree of inflammation being significantly higher than that observed on either day 9 or 29 after the primary s.c. infection (Fig. 5D and H). In addition to inflammatory markers, the alveolar structure was observed to be significantly disrupted in primary s.c. infected mice, as previously reported (19). The damage to the alveolar structure persisted even after the leukocyte infiltration had resolved (Fig. 5C and G) and was clearly evident in s.c. reinfected animals (Fig. 5D and H). Of particular note was that the airways of s.c. reinfected mice contained prominent inflammatory cell foci which surrounded identifiable worm structures. A high-power image of lung tissue from s.c. reinfected mice shows clustering of inflammatory cells around what appears to be a cross-section of an N. brasiliensis worm (Fig. 5I, indicated by arrow) at day 6 after s.c. reinfection, which was not observed in primary infected mice.

• Open in new tab
• Download powerpoint

FIG. 5.

Histological analysis of the lungs of N. brasiliensis-infected mice. BALB/c mice (n = 3/group) were infected s.c. with 600 N. brasiliensis worms and rested for 30 days before being reinfected s.c. with 600 iL3 N. brasiliensis worms. Lung tissues were harvested and fixed at the indicated time points before being separated into 4-μm sections and stained with H&E to visualize inflammatory cells. (A and E) Lung tissue from naïve mice at ×40 and ×200 magnifications, respectively. (B and F) Lung tissue from a primary s.c. N. brasiliensis-infected animal at day 9 postinfection at ×40 and ×200 magnifications, respectively. (C and G) Lung tissue from an N. brasiliensis-infected animal at day 29 after s.c. infection (prior to reinfection) at ×40 and ×200 magnifications, respectively. (D, H, and I) Lung tissue taken from mice s.c. reinfected with N. brasiliensis at day 6 after s.c. reinfection at ×40, ×200, and ×1,000 magnifications, respectively. The arrow in panel I indicates an N. brasiliensis worm.

An assessment of mucus production within lung tissues during s.c. N. brasiliensis infection using PAS staining (Fig. 6A) revealed a profound increase in PAS⁺ mucus-producing cells in the bronchioles of s.c. reinfected mice compared to other times during infection (Fig. 6A and E). Although significant mucus production was detected during primary N. brasiliensis infection (Fig. 6A and C), this resolved prior to reinfection (Fig. 6A and D). The greatly increased level of mucus production detected in the lung during reinfection correlates with the observed reduced worm recovery after reinfection with N. brasiliensis worms.

• Open in new tab
• Download powerpoint

FIG. 6.

Mucus production in the airways of N. brasiliensis-infected mice. BALB/c mice (n = 3/group) were infected s.c. with 600 worms and rested for 30 days before being reinfected s.c. with 600 iL3 N. brasiliensis worms. Lung tissues were harvested and fixed at the indicated time points before being separated into 4-μm sections and stained with PAS and alcian blue to visualize mucus-secreting goblet cells. (A) The number of mucus producing cells per bronchiole was determined by counting the PAS-positive cells in all bronchioles 200 to 600 μm in size per lung section. Data represent the mean number of mucus-producing cells per bronchiole ± the SEM. (B) Representative bronchiole from a naive animal, ×400 magnification. (C) Representative bronchiole from an s.c. N. brasiliensis-infected animal at day 9 postinfection, ×400 magnification. (D) Representative bronchiole from an N. brasiliensis-infected animal at day 29 after s.c. infection (prior to reinfection [PTR]), ×400 magnification. (E) Representative bronchiole from an s.c. N. brasiliensis-reinfected animal at day 6 after reinfection, ×400 magnification.

Taken together, these results indicate that L3 worms are able to migrate to the lungs of previously s.c. infected mice and stimulate profound airway inflammatory responses. The detection of apparently nonviable worm structures within intense inflammatory foci in the airways may be evidence of worm killing in situ, indicating a potential role for lung inflammatory responses in protective immunity to migrating worms.

The roles of CD4 T cells, antibodies, and cytokines in lung protective immunity against infective N. brasiliensis L3 worms.To further characterize the key immune mechanisms that act to reduce the worm burdens in reinfected mice, we tested mouse strains genetically deficient in cytokines, signaling pathways, or receptors in our N. brasiliensis model of protection (Table 1). Surprisingly, given the emphatic role established for eosinophils in models of N. brasiliensis skin infection by Dent and colleagues (6, 17), we found that IL-5-deficient (−/−) mice exhibited slightly reduced protection in the lung stage from s.c. challenge infection with N. brasiliensis. However, MHC class II^−/− mice completely lacking in CD4 T cells exhibited a significant loss of protection in both the lungs (Table 1) and the gut (see Fig. S1B in the supplemental material), with worm burdens in these tissues being equivalent to those found at the peak of s.c. infection of naïve mice for extended periods of time (data not shown). This finding indicates an important role for CD4 T cells in generating protective responses at these sites. The requirement for an intact STAT6 signaling pathway has previously been shown to be important for the generation and survival of Th2 memory cells in peripheral tissues such as the lung (18, 29); indeed, protection was significantly diminished in STAT6^−/− mice. Similarly, the significantly impaired ability of IL-4^−/− mice to reduce the numbers of viable worms in the lungs after s.c. reinfection is in contradiction to studies of gut immunity where IL-4 appears to be less important than IL-13 for the expulsion of adult parasites (28). Importantly, mu^−/− mice, which lack B cells and cannot produce IgG and IgE antibodies specific for N. brasiliensis, exhibited only a minor reduction in the level of protection against reinfecting worms in the lung. CD1-restricted CD4⁺ NKT cells did not play a role in this form of protective immunity. The Toll-like receptors (TLR2 and TLR4) have previously been implicated in Th2 immunity (1, 13, 26); however, we found that deficiency in either of these receptors did not impair the ability of these mouse strains to mount a protective immune response in the lung upon s.c. reinfection with N. brasiliensis (data not shown). Taken together, the results in Table 1 highlight the importance in the early phase of lung immunity against migrating parasites of the CD4 T-cell-mediated and IL-4 and STAT6 signaling-dependent immune response pathways. This would most likely involve IL-4- and IL-13-mediated inflammatory processes such as increased goblet cell hyperplasia and mucus production and the development of long-lived alternatively activated macrophages that produce the appropriate reactive molecules and products toxic to helminth larvae (22, 24). In conclusion, we identify an IL-4/STAT6-dependent protective mechanism in the lung that can significantly affect the viability of migrating L3/L4 worms that is distinct from the protective immune mechanisms mediating the expulsion of adult N. brasiliensis worms from the gut.