Signal Transducer and Activator of Transcription Factor 6 Signaling Contributes to Control Host Lung Pathology but Favors Susceptibility against Toxocara canis Infection

Using STAT6−/− BALB/c mice, we have analyzed the role of STAT6-induced Th2 response in determining the outcome of experimental toxocariasis caused by embryonated eggs of the helminth parasite Toxocara canis. Following T. canis infection wild-type BALB/c mice developed a strong Th2-like response, produced high levels of IgG1, IgE, and IL-4, recruited alternatively activated macrophages, and displayed a moderate pathology in the lungs; however, they harbored heavy parasite loads in different tissues. In contrast, similarly infected STAT6−/− BALB/c mice mounted a weak Th2-like response, did not recruit alternatively activated macrophages, displayed a severe pathology in the lungs, but efficiently controlled T. canis infection. These findings demonstrate that Th2-like response induced via STAT6-mediated signaling pathway mediates susceptibility to larval stage of T. canis. Furthermore, they also indicate that unlike most gastrointestinal helminths, immunity against larvae of T. canis is not mediated by a Th2-dominant response.


Introduction
Toxocariasis is a helminth infection considered as a zoonosis and is caused by the larvae of Toxocara canis; this parasite affects many paratenic hosts including humans. is infection is world-wide distributed mainly because its transmission is associated with domestic dogs. is disease is starting to be considered as a public health problem in Latin America and Asia [1,2] but has been extended in the last few years to developed countries as shown by increased numbers of cases-report published more frequently [3][4][5]. Toxocariasis in humans results from accidental ingestion of Toxocaraembryonated eggs from excreta in the environment or by direct dog-to-person contact [6,7]. Although larvae in muscle may be relatively symptom less, those in brain may cause serious neurological disorders [8]; these symptoms are believed to result from in�ammation following degeneration of the parasite [9].
In the experimental model of murine toxocariasis, infection of inbred mice with T. canis induces a strong 2-like response similar to that observed following infection with other helminthes such as Nippostrongylus brasiliensis and Trichuris muris [10][11][12]. While widely accepted, 2-type response in clearing helminth infections has limitations [11,[13][14][15][16][17], its role in mediating protection against toxocariasis in paratenic hosts is not clear, mainly because reinfection in toxocariasis favors parasite survival [18]. Recent advances in the immunobiology of T. canis indicate that regulatory mechanisms are raised aer infection; mainly T regulatory cells have been involved in limiting pathologic damage by the in�ammatory response [19]; however, there are few data regarding on mechanism of protection or susceptibility against such parasite.
Previous studies have found that some extraintestinal larvae from other helminthes such as Taenia crassiceps and Trichinella spiralis are apparently eliminated in infected mice by a 1-mediated in�ammatory response during early phase of infection [20][21][22]. Furthermore, both studies found that STAT6 was involved during early phase of infection in rendering them more susceptible to cysticercosis and trichinellosis, respectively. ese �ndings suggest that while 2-type response may be involved in mediating resistance on gastrointestinal helminth infections, this pathway may be involved in susceptibility to the larval stages of such parasites. Numerous studies using STAT6 −/− mice have shown that STAT6-mediated IL-4/IL-13 signaling pathway is critical for 2 differentiation [23]. For example, STAT6 −/− mice fail to mount a signi�cant 2 response and cannot control worm burdens following infection with gastrointestinal helminth parasites [17,24]. Conversely, STAT6 −/− mice develop a 1 like response and control infections caused by intracellular protozoan parasites such as Leishmania mexicana and Trypanosoma cruzi [25,26] indicating that STAT6-mediated signaling pathway inhibits development of protective immunity by inhibiting a 1 development.
e purpose of this study was to determine the role of a 2-type response induced via STAT6-mediated signaling in the outcome of experimental murine toxocariasis caused by the L2 of the nematode T. canis. To approach this question, we compared the course of T. canis infection in STAT6 −/− BALB/c mice (STAT6 −/− ) with that in the wild-type BALB/c (STAT6 +/+ ) mice. In addition, we analyzed both the antibody and cytokine pro�les in sera, as well as the phenotype of lung macrophages. Our data demonstrate that 2-type response induced via STAT6-signaling pathway mediates susceptibility in toxocariasis.

Mice.
Six-8-week-old male STAT6 −/− and STAT6 +/+ mice in a genetic BALB/c background were originally purchased from e Jackson Laboratory Animal Resources Center (Bar Harbor, Maine, USA) and were maintained in a pathogen-free environment at the FES-Iztacala, U.N.A.M. animal facility in accordance with Institutional and National guidelines.

Isolation of Larvae and Eggs and Infection Protocol. Adult
T. canis females' worms were isolated from the intestine of naturally infected puppies (<3 months). Isolation and embryonation of eggs were performed as follows: female worms were dissected and from the uterus eggs were isolated and putted into distilled water; then the mixture was centrifuged two times for 10 min at 2,000 ×g in a solution of NaHCl at 1%. Aer removal of the supernatant, the sediment was two times washed in distilled water and placed into the solution of formalin at 1% in tissue �asks at 28 ∘ C for 1 month with gentle daily agitation until the end of embryonation which was controlled under the microscope.

Infection.
Five hundred larvated eggs were intragastrically administered with a Foley tube to both STAT6 −/− and STAT6 +/+ mice. Infected mice were sacri�ced at days 5, 14, and 60 postinfection, and the parasites harvested from different tissues (lung, liver, brain, and muscle) were enumerated as described previously [27].
For histological evaluation of different tissues, animals were euthanized at indicated days. e liver, lung, brain, and muscle were removed and �xed in 4% formalin. Tissue samples were embedded in paraffin, and 5 m sections were cut on a microtome and stained with hematoxylin and eosin for histological examination.

Cytokine
Measurements. e IL-4, IL-12, and IFNlevels were quanti�ed in mouse serum at the indicated point times. Antibody pairs were used according to the manufacturer's instruction (Peprotech México, México, DF).

2.�. Toxocara-�peci�c �nti�ody Level and Total IgE.
Peripheral blood was collected at the indicated time points aer Toxocara infection from tail snips. e blood was centrifuged at 2500 rpm for 10 min, and serum was collected and tested for Toxocara-speci�c IgG1 and IgG2a in antigen-coated plates (1 g/mL). Aer an overnight incubation at 4 ∘ C, the plates were washed with PBS supplemented with 0.05% Tween 20 (Sigma, St. Louis, MO, USA) and blocked with PBS supplemented with 1% BSA (US Biological, Swampscott, MA, USA). Serial dilutions (starting from 1 : 100) of the serum samples were added to the plates. e bound antibodies were detected following incubation with HRP-conjugated rat antimouse IgG1 or IgG2a (Zymed, San Francisco, CA, USA). e reactions were developed with ABTS solution (Zymed) and read on a microplate reader at 405 nm (Multiskan Ascent, ermo Labsystems). Results are expressed as the maximal sera dilution (endpoint titer) where OD was detected. Total IgE production was detected by Opt-ELISA from Biolegend.

Flow Cytometry.
It has been previously shown that macrophages recruited by helminth parasites to the site of infection express alternatively activated and suppressive markers, such as mannose receptor (MR), PD1 ligand 1 (PD-L1), and PD-L2. To determine whether T. canis-infected mice recruit such population, �ow cytometry was performed on lung exudates cells from T. canis-infected at different times aer infection. Brie�y, 5, 14, and 60 days aer infection, lungs lavages were aseptically obtained, and 1 × 10 6 cells were incubated with anti-CD16 and anti-CD32 (Biolegend, San Diego, CA, USA) to block nonspeci�c antibody binding. e cells were then stained with APC-conjugated anti-F4/80, FITC-conjugated anti-MR, PE-conjugated anti-PDL1, and PE-conjugated anti-Gr1 (all from Biolegend) and incubated for 30 min. at 4 ∘ C in FACS staining buffer (1% FBS, 0.5% sodium azide in PBS). e cells were analyzed using a FACSCalibur and Cell Quest soware (Becton Dickinson).

RT-PCR.
RNA was extracted from isolated spleen cells aer different day's postinfection using the TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) and the isopropanolchloroform technique. e RNA was quanti�ed, and 5 g of RNA was reverse transcribed using the Superscript II First Strand Synthesis Kit (Invitrogen). PCR reactions containing 5x PCR Buffer blue, 10 mM dNTP, 40 nM each forward and reverse primers, 1 U Taq DNA polymerase (Sacace Biotechnologies, Italy), and 2 L of the cDNA were prepared in a 25 L �nal volume. e PCR conditions were elsewhere described [28]. Brie�y, consisted of an initial denaturation step at 95 ∘ C for 5 min; 35 cycles of 95 ∘ C for 40 s, the indicated melting temperature for 50 s and 72 ∘ C for 40 s; a �nal extension step at 72 ∘ C for 4 min in a thermal cycler (Corbett Research, Australia). e ampli�ed products were mixed with loading buffer containing SYBR green and observed in a 1.5% agarose gel with the Fuji�lm FLA 5000 scanner (Fuji, �apan) using the image reader V2.1 soware to capture the images.

Statistical Analysis.
Comparisons between wild-type (STAT6 +/+ ) and STAT6 −/− groups considered in this work were made using student's unpaired t test. was considered signi�cant. e statistical signi�cance of the sera titers were determined by nonparametric tests using Mann-Whitney U-Wilcoxon Rank.

Results and Discussion
It is largely accepted that the 2-like response induced via STAT6-mediated signaling pathway (through IL-4/IL-13 receptors) plays a critical role in mediating protective immunity against most helminthes [12,17,28]. For example, STAT6-mediated signaling has been shown to promote protective immunity against gastrointestinal helminthes such as Trichinella spiralis, N. brasiliensis, and Hymenolepis diminuta [12,17,29,30]. However, the role for many molecules associated with the immune response, including STAT6, during infection with Toxocara canis is unknown. Here we analyzed the potential role of STAT6 in modulating immunity against this nematode parasite. One of the �rst organs that T. canis larvae reach early aer infection is the lungs, in the present study we detected a signi�cant greater number of larvae in STAT6 +/+ compared to STAT6 −/− mice at day 5 aer infection, but later both groups displayed comparable parasite burdens at 14 and 60 days postinfection (Figure 1(a)). In contrast, in the liver as early as 5 days pi STAT6 −/− displayed lower parasite burdens, and this was more evident aer 60 days pi (Figure 1(b)). In a similar way the parasite burdens in the brain were statistically different 60 days pi (Figure 1(c)). Interestingly, as infection progressed, the number of larvae in the muscles increased signi�cantly in STAT6 +/+ mice as compared to STAT6 −/− mice that successfully reduced the number of parasites by day 60 postinfection (Figure 1(d); * ). Unexpectedly, the lungs from STAT6 −/− mice displayed a greater macroscopic damage as we observed an increased number of hemorrhagic spots in such organs (Figure 2(a)) although that numbers of parasites detected in both groups were closely similar at 14 days pi or even lower in STAT6 −/− mice at 5 days pi. It is known that T. canis infection promotes the recruitment of leukocytes to the lungs generating an acute in�ammatory response; here we observed a greater in�ammatory in�ltration in the lungs of STAT6 −/− mice as early as 5 dpi which was maintained higher in these mice until 60 dpi (Figure 2(b)). Moreover, disruption of alveoli was more frequently observed in STAT6 −/− mice, as well as a dominant polymorphic cell in�ltration ( Figure  2(c)). ese �ndings suggest that STAT6-mediated signaling pathway is involved in both susceptibility and pathogenesis during T. canis infection in susceptible BALB/c mice.
Previous studies have demonstrated that STAT6-mediated signaling pathway prevents development of protective immunity mainly against intracellular parasites by inhibiting 1 development [25,26]. erefore, we measured levels of 1-associated IgG2a as well as 2-associated IgG1 and Total IgE antibodies in STAT6 −/− and STAT6 +/+ mice at different time points following infection with T. canis. Additionally, we also compared the cytokine circulating levels from these same mice. Early in the infection, T. canisinfected STAT6 +/+ and STAT6 −/− mice displayed minimum and comparable levels of T. canis Ag-speci�c 1-associated IgG2a antibodies (Figure 3(a)); these data are in line with those previously reported [27], who also found a low production of this subclass of antibody at early times postinfection. However, by day 60 pi STAT6 −/− mice displayed signi�cantly higher levels of speci�c IgG2a antibodies against Toxocara antigens (Figure 3(a)). On the other hand, clear differences were observed with the 2-associated IgG1 production, where STAT6 +/+ mice displayed signi�cantly higher titers of Toxocara-speci�c IgG1 as compared to STAT6 −/− mice since day 14 aer infection ( Figure 3(b)). Although 2-associated IgE has been shown to play a role in mediating immunity against certain extraintestinal helminthes, we found that T. canis-infected STAT6 −/− mice harbored lower parasite burdens despite producing signi�cantly lower levels of IgE as compared to similarly infected STAT6 +/+ mice, suggesting that IgE may have a limited role in mediating protective immunity against L2 of T. canis (Figure 3(c)). Here it is noteworthy that in spite of a higher 2-associated antibody response in WT mice, these displayed greater susceptibility to T. canis. ese data agree with those recently reported by [18], who found that aer T. canis reinfection in BALB/c mice, reinfected mice displayed signi�cantly higher titers of T. canis-speci�c IgG1; moreover, those antibodies showed a greater avidity for T. canis antigens. However, re-infected mice displayed a major number of larvae in different tissues [18]. Together with our data, such �ndings strongly suggest that a humoral immune response is not protective against L2 T. canis infection. Next we analyzed the cytokine pro�le in sera that STAT6 +/+ and STAT6 −/− mice displayed during toxocariasis. BALB�c mice at 5 dpi produced signi�cantly more IL-4 and IFN-than STAT6 −/− mice (Figures 4(a) and 4(b)), whereas the levels of IL-12 were closely similar between groups (Figure 4(c)), indicating that not a clear 1-type polarization of the immune response was observed in T. canis-infected STAT6 −/− mice. However, a lack of 2-type response was con�rmed in such mice given the low levels of Ig�1 and IgE together with lower systemic levels of IL-4.
To further analyze the immune response, the spleen cells from T. canis-infected STAT6 +/+ and STAT6 −/− mice were obtained for RT-PCR analysis of several markers for alternatively activated macrophages (AAM) and some cytokines. While splenocytes from STAT6 +/+ mice displayed expression of Arginase-1 and Ym-1, both markers for AAM, during early phase of infection, those from STAT6 −/− mice did not overexpress such markers (Figure 5(a)). In contrast STAT6 −/−infected mice displayed expression of iNOS, a marker for classically activated macrophages, at 14 dpi, whereas STAT6 +/+ mice did not express iNOS ( Figure 5(b)). Regarding cytokines we observed a major expression of IL-4 on spleen cells from STAT6 +/+ than spleen cells from STAT6 −/− mice ( Figure  5(a)), whereas a similar level of expression was observed for IFN-( Figure 5). us our data revealed the presence of AAM and IL-4 in the spleens of STAT6 +/+ mice.
In the last few years, a new cell population has been detected in most of the helminth infections, such population is AAM; the role of these cells appears to be divergent [30] as different authors have demonstrated, for example in gastrointestinal infections by Nippostrongylus brasiliensis and Heligmosomoides bakeri the presence of AAM is key for worm expulsion [31], whereas in other helminth infections such as Trichuris muris and Hymenolepis diminuta the presence of AAM was irrelevant [32][33][34]. In contrast, for schistosomiasis and experimental cysticercosis the presence of AAM appears to be crucial; in the �rst case, the absence of AAM leads to pathologic disorders in the liver and the hosts die [35,36], whereas in experimental cysticercosis the presence of AAM leads to susceptibility, given that eliminating AAM with clodronate liposomes helps to clear the infection in otherwise susceptible hosts [37]. According to our knowledge this is the �rst time that the markers for AAM are reported in T. canis infection, but their role is still unknown.
In order to gain knowledge on a possible role for AAM in acute T. canis infection, we analyzed the pro�le of macrophages that reach the lungs as early as 5 days pi (time where striking differences in pathology were observed). Aer 5 dpi lungs were obtained and cut in small pieces, which were further passed through a mesh. Cells were stained for different markers and assayed for cytometry. Stained cells were captured in log, and the region that displayed both high granularity and high size was selected for analysis. As observed in Figure 6(a), lung cells from T. canis-infected STAT6 +/+ mice displayed a greater recruitment of F4/80 + MR + cells compared to both naïve mice or T. canis-infected STAT6 −/− mice; these data suggest a recruitment of AAM, which may participate in tissue repair in the lungs, while in STAT6 −/− mice the increase of F4/80 + MR + cells was gradual, perhaps because they are unable to mount an efficient 2 response that may impair the recruitment of AAM in the �rst days pi, which could be associated with a greatest tissue damage in the lungs (Figure 2(a)). Interestingly, by day 60 pi this population is increased in both groups, but still more damage prevailed in STAT6 −/− mice; this apparent contradiction may be explained just in a time-dependent point of view; it means that AAM early recruited in WT mice had more time to repair the tissue. In contrast, the apparent delay of AAM in reaching the lungs in STAT6 −/− mice may need longer time to accomplish their function. An alternate explanation may be the participation of different cell populations for a rapid tissue repair, thus, our observation that another population of cells F4/80 + -Gr-1 + is also recruited differentially in the lungs of STAT6 +/+ mice may be indicative that more than onecell population is involved in tissue repair in this infection ( Figure 6(b)). Intriguingly, such F4/80 + -Gr-1 + took longer to reach the lungs in STAT6 −/− mice. However, these cells reach similar levels than in wild-type mice until 60 dpi ( Figure 6(b)), if these cells represent a different population of regulatory cells needs further research. On the other hand, the same pattern of recruitment was followed when we analyzed the F4/80 + PDL-1 + population ( Figure 6(c)). All these surface markers have been previously reported associated with macrophages that undergo a distinct activation phenotype in the presence of the 2 cytokines or helminth infections named AAM [38]. Besides, these AAM have an upregulated expression of arginase-1, RELM-, and chitinase-like protein Ym1, among other markers. Moreover, although AAM can exhibit antiparasitic activity [31], their most important function in the context of migrating helminth parasites appears to be associated with tissue-repair responses [39,40]. Taken together these cytometry analyses with those obtained in the macroscopic and microscopic lung analyses, we may associate less tissue damage on T. canis-infected STAT6 +/+ mice with the presence of AAM, and by contrary, we would associate the absence of AAM in T. canis-infected STAT6 −/− mice with greater lesions and increased lung-cell in�ltration. erefore, we hypothesized that AAM may be involved in mediating protection against helminth-induced immunopathology in the lungs during acute toxocariasis.
Several different reports have shown that 2 responses are not de�nitive to kill extraintestinal phases of helminth parasites. For example during muscle infection with T. spiralis, BALB/c mice lacking eosinophils displayed similar larval burdens to those of wild-type BALB/c mice [21]. In line with our results, also T. crassiceps infection is fully controlled in the absence of STAT6 [22], and more interestingly during experimental neurocysticercosis caused by Mesocestoides corti infection STAT6 −/− mice displayed a reduction in the number of brain larvae but an increase in clinic neurological symptoms that were associated with lack of AAM [41]. Together all these �ndings oppose the dogma that 2-type responses play a critical role in the elimination of all kind of helminthes [42]. ese data also suggest that STAT6 pathway may act to limit Toxocara larvae-induced immunopathology at least in the lungs.
In conclusion, STAT6 −/− BALB/c mice mount a null 2like response and efficiently control T. canis infection. In contrast, STAT6 +/+ BALB/c mice develop a predominant 2like response that is associated with high levels of IL-4, IgG1, IgE, and AAM and displayed signi�cantly higher parasite burdens in different tissues but interestingly less associated pathology. e �ndings in our study support the conclusion that STAT6-mediated signaling is critical for the suppression of the immune response that is required for controlling L2 toxocariasis. We postulate that 2 cytokines may have a dual role during toxocariasis, on one hand may contribute to host susceptibility via STAT6 activation and that neither AAM nor IgE are essential or primarily responsible for eliminating T. canis tissue infection, but on the other hand, such response may downregulate the immunopathology induced by T. canis larvae in the lungs. Flores and Tomas Villamar for their excellent care of animals and Biol. Ana F. Chavez Sanchez for histology assistance.