Ultrastructural localization of matrix metalloproteinase-9 in eosinophils from the cerebrospinal fluid of mice with eosinophilic meningitis caused by Angiostrongylus cantonensis

Matrix metalloproteinase-9 (MMP-9) has been implicated in the pathogenesis of eosinophilic meningitis caused by Angiostrongylus cantonensis. In the present study, such meningitis in mice was found to be associated with elevated expression of MMP-9 mRNA, elevated MMP-9 concentrations and enhanced MMP-9 activity in the cerebrospinal fluid (CSF). Immunocytochemistry showed that an anti-MMP-9 antibody reacted with macrophages, neutrophils and eosinophils from the CSF. As eosinophils are generally considered to be effector cells in host defence against A. cantonensis infection, high-resolution immuno-electron microscopy was then used to confirm the localization of MMP-9 in the eosinophils from the CSF. The method used, which was based on immunogold, indicated that the eosinophilic MMP-9 was mostly localized in the 'small' granules in the cytoplasm and along the cell membrane, and not in the crystalloid-containing secretory granules observed. It therefore appears that MMP-9 is synthesised and/or stored in the small granules of the eosinophils, and is released into the subarachnoid space of the host's brain by secretion or cell rupture.     

Eosinophil chemoattracted by eotaxin from cerebrospinal fluid of mice infected with Angiostrongylus cantonensis assayed in a microchamber

When non-permissive hosts are infected with Angiostrongylus cantonensis, the migration of the worms to the brain and their subsequent development manifests as marked eosinophilic pleocytosis. We used microchambers to demonstrate direct eosinophil chemotactic activity by adding a variety of antibodies into cerebrospinal fluid (CSF) of BALB/c mice 21 days post-infection with A. cantonensis. The antibodies were directed to neutralize eotaxin, RANTES (regulated on activation, normal T-cells expressed and secreted), macrophage inflammatory protein (MIP)-1alpha, and platelet-activating factor (PAF), respectively. Eosinophil migration into the polycarbonate membrane covering CSF with anti-eotaxin or anti-MIP-1alpha antibodies was significantly lower than that for antibody-free CSF (Student's t test: p < 0.01, p < 0.05). We also collected CSF from mice 21 days after infection with 10, 20, 30, 40, and 50 third-stage larvae (L3) respectively for dose-dependent testing, and 40 L3 at days 7, 14, and 21 after infection for time-dependent testing. Chemokine production in CSF was affected by A. cantonensis infection intensity and post-infection time. In conclusion, eotaxin and MIP-1alpha released in the CSF of A. cantonensis-infected mice have eosinophil chemotactic activity in this in vitro assay.     

Blood-brain barrier dysfunction occurring in mice infected with Angiostrongylus cantonensis

Several indices were used to assess whether blood-brain barrier (BBB) damage occurs in neurological disorders. Dysfunction of the BBB was surmised to be involved in the pathological changes of eosinophilic meningitis caused by the infection of Angiostrongylus cantonensis. The mean concentration of protein and albumin in the cerebrospinal fluid (CSF) of infected mice gradually increased from days 0 to 18 after infection and then rapidly increased 21 days after infection. The concentrations of protein and albumin in the CSF of infected mice 15 days after infection were all significantly higher than those in uninfected mice (all P-values at least <0.05). Parallel with the increase in protein and albumin in the CSF, infected mice showed a gradual increase in their CSF/serum protein and albumin ratios. The increase became significant at days 21 and 18 after infection, respectively (P<0.01 and P<0.05, respectively). The higher the worm counts in the brain, the higher the CSF/serum albumin ratio was observed in infected mice at day 21 after infection (P<0.001). In addition, the ratios of the CSF/serum albumin were positively correlated with the worm counts in the brain (P<0.001). The total leukocyte and eosinophil counts were also positively correlated with ratios of CSF/serum albumin (P<0.01). The amount of Evans blue in the brain of mice 21 days after infection from peripheral blood via BBB became significantly increased than those in uninfected mice (P<0.001). Thus, the evidence of high concentrations of protein and albumin, high leukocyte counts in CSF, high ratio of CSF/serum protein and albumin, and high permeability of BBB show that dysfunction of the BBB occurred in mice infected with A. cantonensis.     

T-cell-dependent eosinophilia in the cerebrospinal fluid of the mouse infected with Angiostrongylus cantonensis

Eosinophilia of the cerebrospinal fluid (CSF) in permissive (rats) and non-permissive (mice) hosts infected with Angiostrongylus cantonensis, and the possible mechanism of the eosinophilia were studied. In three strains of thymic mice (ICR, ddY and BALB/c), the infection provoked a marked CSF eosinophilia starting at around day 12, reaching a peak level at day 20 and maintaining significantly high levels until day 35. In contrast, in athymic nude mice of BALB/c strain the infection failed to evoke this eosinophilia, suggesting T-cell dependence of murine CSF eosinophilia. Humoral antibodies did not correlate with the induction of eosinophilia. A time-course study of worm recovery in the mouse brains indicated a gradual but consistent reduction in worm burden in accordance with the rapid rise in CSF eosinophil levels. Bone marrow eosinophilia occurred in mice at day 5, which preceded CSF eosinophilia. Jirds, a permissive but less susceptible host, developed a CSF eosinophilia with a peak level at day 17, but which declined rapidly following the peak. Permissive rat hosts developed significant peripheral and bone marrow eosinophilia at day 35 but their CSF eosinophilia was markedly less prominent than that of mice and jirds. These data clearly indicate that there are distinct differences in the mechanism of eosinophilia and eosinophilia-inducing factors between permissive and non-permissive hosts.     

Monoclonal antibody 12D5 inhibits eosinophil infiltration in the brain of Angiostrongylus cantonensis-infected BALB/c mice

Each of BALB/c mice was infected with 50 Angiostrongylus cantonensis larvae. One group of mice received an intraperitoneal injection of 50 μg 12D5 monoclonal antibody (mAb) against a 98 kDa antigen of adult worms at 10 days post-infection (dpi), with a booster injection of 25 μg at 12 dpi. Five mice from each group were sacrificed at 14 dpi for pathological examination and RNA extraction. The infiltration of eosinophils and severity of eosinophilic meningitis were reduced in 12D5 mAb-treated mice compared with the infected mice without 12D5 treatment. The levels of eotaxin mRNA expression in spleen significantly increased and the expression of the Th2-type cytokine IL-5 significantly decreased. However, the expression of IL-4 was not changed. 12D5 mAb can observably enhance the survival rate of infected mice and reduce symptoms of angiostrongyliasis. A. cantonensis infection is a major cause of eosinophilic meningoencephalitis. The results of this study could be helpful for the development of treatment of human angiostrongylosis.     

Ablation of eosinophils with anti-IL-5 antibody enhances the survival of intracranial worms of Angiostrongylus cantonensis in the mouse

Effects of depressed eosinophilia on the development of Angiostrongylus cantonensis in the mouse were studied using monoclonal rat anti-mouse-interleukin-5 antibody (anti-IL-5). The administration of anti-IL-5 strongly depressed peripheral, cerebrospinal fluid (CSF) and medullary eosinophilic responses in mice infected with A. cantonensis, when compared with groups treated with phosphate-buffered saline solution (PBS) alone or isotype-matched rat IgG. There was no significant difference in A. cantonensis antigen specific IgG and IgE antibody responses between rat IgG treated and anti-IL-5 treated mice. Intracranial worm recovery in anti-IL-5 treated mice was consistently high throughout the course of the study and some worms migrated from the brain to the lungs. By contrast, almost all the intracranial worms in the mouse groups treated with PBS alone or rat IgG died before day 32. These data clearly indicate that IL-5 is essential for eosinophil responses in A. cantonensis infected mice and also that eosinophils serve as a potential effector cell in the killing of the intracranial worms in mice.     

Trypanosoma cruzi infection in mice enhances the membrane expression of low-affinity Fc receptors for IgG and the release of their soluble forms

The membrane expression of low-affinity Fc receptors for IgG (FcγRII/III) on cells and the number of FcγRII/III(+) cells were studied by flow cytometry, using the 2-4G2 MoAb, in mice infected by Trypanosoma cruzi. Cells from spleen, mesenteric lymph nodes and peritoneum were collected on days 10, 20, 30 and 40 post infection (p.i.). The in vivo serum level of soluble FcγRII/III, as well as its in vitro release by cells from infected mice were studied. Parasitaemia and IgG1, IgG2a and IgG2b T. cruzi-specific antibody titres were also recorded. Both the expression of FcγR on cell membrane and the absolute number of FcγR(+) cells increased in spleen and in mesenteric lymph nodes, but not in peritoneum. The modifications in spleen occurred in the early and late parasitaemic phase of infection, i.e., before and after detection of T. cruzi-specific antibodies (from day 10 to 40 p.i.). In mesenteric lymph nodes, the variations were observed only in the early acute infection, when antibodies were not yet detectable at significant levels (on days 10 and 20 p.i.). Higher levels of soluble FcγR were detected in sera and in culture supernatants of spleen and lymph node cells from day 20 to 40 p.i. These results show that T. cruzi infection in mice upregulates the expression and the release of FcγRII/III, in the acute phase of infection, before as well as after the rise of antibody response.     

Eosinophilic meningitis associated with angiostrongyliasis: clinical features, laboratory investigations and specific diagnostic IgG and IgG subclass antibodies in cerebrospinal fluid

Eosinophilic meningitis (EOM) associated angiostrongyliasis mostly induced by the nematode Angiostrongylus cantonensis, is a common disease with worldwide prevalence. Heavy infections can lead to chronic disabling disease and even death. This study was conducted to shed light on the overall specific IgG antibody response as well as the specific IgG antibody subclass responses in cerebrospinal fluid (CSF) of patients with EOM. Fifteen patients with EOM associated with angiostrongyliasis were included in the study. Sera were screened by immunoblotting for the presence of IgG antibody to the 29 kDaA. cantonensis antigenic polypeptide. CSF was examined by ELISA for the presence of specific IgG and IgG subclass antibodies. Patients presented with headache (100%), neck stiffness (20%), fever (40%), nausea (87%), vomiting (73%), paresthesia (7%), and muscle weakness (7%). Seven of 15 (47%) patients showed peripheral blood eosinophilia and all patients presented with eosinophils in CSF. A sensitivity of 80 % was obtained by combining the diagnostic values of immunoblotting in sera and IgG and IgG subclasses-based ELISA in CSF. The combination of a history of eating raw or semi-cooked infected foods, clinical features, complete blood count, differential cell counts, CSF profiles, and serum and CSF antibodies to A. cantonensis can be used to increase the sensitivity for the diagnosis of human angiostrongyliasis.     

Intraperitoneal murine alveolar hydatidosis: relationship between the size of the larval cyst mass, immigrant inflammatory cells, splenomegaly and thymus involution

Both phagocytic and nonphagocytic inflammatory cells infiltrate the peritoneal cavity of mice infected intraperitoneally with Echinococcus multilocularis cysts. A longitudinal study on the kinetics of peritoneal leukocytosis at 3 days, 1, 2, 3, 4, 6, 8, 10 and 14 weeks postinfection revealed that the restrictive and progressive growth phases of the alveolar hydatid cyst correspond sharply with the increasing and decreasing levels of peritoneal cells, respectively. The restrictive phase is characterized by the progressive peritoneal accumulation of lymphocytes, monocytoid cells and eosinophils. Between 6 and 14 weeks p.i., the alveolar cyst increased in weight 30 fold. This phase was associated with peritoneal neutrophilia, splenomegaly, involution of the thymus and a significant decline in the extravasated lymphocytes, monocytoid cells and eosinophils. These results in conjunction with our previous studies indicate that host's hydatid immunosurveillance is compromised as a result of profound immunopathologic disorders during the progressive growth phase of the alveolar cyst. In order to understand the prolonged survival of alveolar cyst, further investigation of inflammatory cell-cyst interactions is indicated.     

广州管圆线虫在长爪沙鼠体内的发育及其引起的病理损害

目的观察广州管圆线虫在长爪沙鼠体内的发育及其引起的病理损害。方法以广州管圆线虫第Ⅲ期幼虫经腹腔注射感染成年长爪沙鼠（10条幼虫／鼠）,感染后不同时间检查脑、心肺组织内的虫体,肉眼观察和病理切片检查脑、肺组织,于感染后第46d采用直接涂片法检查粪便。结果从感染的长爪沙鼠脑和心肺组织共检获虫体127条（♀62／♂65）,平均虫数为（3．63±1．46）／鼠,感染后第23－30d,检获虫体数最多;虫体在感染后30d内主要分布在脑,30d后主要分布在肺;幸存的长爪沙鼠大多数广州管圆线虫能在其肺中发育至成虫并产卵,同时在部分沙鼠粪便中检查到Ⅰ期幼虫;病理结果显示蛛网膜下腔可见虫体,肺内有虫卵,广州管圆线虫从脑移行至肺导致嗜酸性粒细胞增多性脑膜炎和肉芽肿性肺炎。结论广州管圆线虫可在长爪沙鼠体内发育成熟并造成脑和肺组织的损害。     

广州管圆线虫感染小鼠血清细胞因子及IgG亚类的动态观察

目的观察小鼠感染广州管圆线虫后机体免疫的动态变化。方法分别采集感染前、感染后第1、3、7、18d的小鼠血清,用ELISA方法检测血清中细胞因子IL-2、IL-4以及特异性免疫球蛋白G（IgG）亚类水平。结果广州管圆线虫感染的小鼠血清中IL-2的水平较感染前呈逐渐下降趋势,IL-4的水平与感染前小鼠相比先下降后呈升高趋势。抗体IgG1的水平与感染前小鼠相比明显升高,IgG2a的水平与感染前小鼠相比未见明显变化。结论小鼠Th1型免疫应答较弱,Th2型免疫应答增强。表明小鼠感染广州管圆线虫后机体细胞免疫较弱,体液免疫较强。     

BALB/c小鼠感染广州管圆线虫后脑内的动态变化

目的观察广州管圆线虫感染BALB/c小鼠后其虫数在小鼠脑内的动态变化,探讨其感染后对人体可能的致病机制。方法以广州管圆线虫Ⅲ期幼虫感染实验BALB/c小鼠,按感染的不同时间分批处死,镜下及肉眼观察脑内虫体数量及分布情况。结果在BALB/c小鼠大、小脑内,广州管圆线虫的分布符合寄生虫感染宿主后先增后减的一般基本变化规律,虫体寄生的部位因感染虫体而受损。引起症状的时间变化与虫体在脑内变化规律基本相符。结论小鼠感染广州管圆线虫以后,出现共济失调,抽搐等症状的时间与虫体在小鼠脑内的变化规律基本相符。为临床研究该病的相关临床症状及治疗提供了一定的实验依据。     

Resolution of Cryptosporidia! infection in mice correlates with parasite-specific lymphocyte proliferation associated with both Th1 and Th2 cytokine secretion

This study was designed to investigate and characterize T-cell responses which lead to elimination of a primary infection of Cryptosporidium muris in BALBjc mice. The proliferative response of spleen cells to parasite antigen was measured by uptake of ³H-thymidine and, in parallel, supernatants were removed from cells to measure levels oflFN-γ, TNF, IL-2 and IL-4 by ELISA. Oocyst excretion in faeces was first detected on day 10 post infection (p.i.); the level of shedding subsequently increased until day 14 and then declined until no oocysts were detected by day 25. The proliferative response of spleen cells from infected animals was similar to control levels up to day 14p.i. but increased significantly on day 21 and was even greater on day 26. IFN-γ and IL-2 were detected initially on day 14 p.i. and significantly higher concentrations were found on days 21 and 26. IL-4 secretion was also detected, but not until day 21 p.i., and production of TNF was not found at any time. Depletion of T-cells or CD4 + cells from spleen cells cultured with antigen resulted in a significant decrease in the levels of cytokine detected. These results indicated, therefore, that in BALB/c mice there was a correlation between the development of immunity to C muris infection and both a parasite antigen-specific proliferative response and T_hl and T_h,2 cytokine production by spleen cells     

Influence of nalcrom (sodium cromoglycate) on the course of the intestinal phase of trichinellosis in mice]

The influence of Nalcrom (sodium cromoglycate) on the course of the intestinal phase of trichinellosis in mice was investigated. The animals infected with 200 Trichinella spiralis larvae were treated with Nalcrom between 7-20 or 3-20 days after infection (d.a.i.). The drug was administered in two doses: 0.6 or 1.7 mg/mouse/day. In the all groups of animals received Nalcrom higher number of mast cells and eosinophils than in the control groups was observed. These results are the opposite of those obtained with Nalcrom in ulcerative colitis in man.     

丙烯硫脲对日本血吸虫感染小鼠肝脏病变的影响

目的　观察丙烯硫脲对日本血吸虫感染小鼠肝脏病变的影响。方法　小鼠感染日本血吸虫尾蚴 2 2d后 ,隔日一次按 30 0mg/kg腹腔注射酚酶抑制剂丙烯硫脲 ,至感染后 4 2d剖杀动物 ,观察日本血吸虫感染小鼠肝脏的病理变化。结果　与感染对照组相比 ,实验组小鼠肝脏表现为散在分布的炎性细胞浸润灶 ,中心未见虫卵 ,仅见一些颗粒状物质。其平均直径和面积均较感染对照组的典型虫卵肉芽肿显著减小 (P <0 .0 1)。结论　感染小鼠腹腔注射丙烯硫脲后肝脏内未见虫卵肉芽肿的形成。     

Toxocara canis-induced murine pulmonary inflammation: analysis of cells and proteins in lung tissue and bronchoalveolar lavage fluid

The pulmonary immuno-inflammatory reaction and its effect on microvascular integrity was studied in Toxocara canis infected BALB/c mice. The investigation aimed to compare changes in lung histology and composition of bronchoalveolar lavage fluid (BALF) caused by T. canis infection with those described to occur in allergic asthma. Groups of (non)-infected mice (1000 ova) were investigated until 90 days post infection (p.i.). Migration of the larvae through the lungs was followed by a rapidly progressing multifocal interstitial and alveolar inflammation. Eosinophils and lymphocytes formed perivascular and partially peribronchial mixed cellular infiltrates. Lymphocytes with plasma cell morphology staining intracellularly for either α, ɛ or γ immunoglobulins were demonstrated. BALF, collected from mice infected with either 250, 500 or 1000 ova was analysed at 14 and 28 days p.i. A dose-related increase in cell numbers and in albumin and IgA concentration was observed. IgE increase was independent of the infective dose. Peak values were measured at 14 days p.i. Albumin increase in lung homogenate was highest at 28 days p.i. 30% of the lymphocytes consisted of T cells carrying Thy-1.2 and L₃T₄ surface antigens. It is concluded that T. canis-induced pulmonary inflammation affects the permeability of the microvasculature. This is expressed by interstitial oedema and plasma exudation in the airway lumen. Both phenomena occur also in allergic asthma. It is suggested that increased permeability of the microvasculature is mediated by T cells and eosinophils.     

Ultrastructural evidence for eosinophil-mediated destruction of Angiostrongylus cantonensis transferred into the pulmonary artery of non-permissive hosts

Ultrastructural and cytochemical analyses were carried out on cellular reactions to the young adult worms of Angiostrongylus cantonensis surgically transferred into the pulmonary arteries of permissive (rat) and non-permissive (rabbit and guinea-pig) hosts. In permissive hosts, no appreciable cellular reactions could be found around worms throughout the course of the observations. By contrast, the infiltration of neutrophils along with eosinophils was observed around worms in non-permissive hosts even at early stages (days 2 to 4). At day 7 and later, the prominent degranulation (solubilization of the whole granule or the matrix alone with preserved crystalloid, tubulovesicular structure formation, and vacuole formation containing lysosomal contents, etc.) of eosinophils, and subsequent release of the lysosomal contents on to the worm surface were noted. Discharge of large amounts of peroxidase on to the worm surface was also demonstrated. The worms were thus damaged and their cuticular fragments were frequently found removed. In addition to this, degenerative changes, such as lipid-droplet and vacuole formations, were detectable in the hypodermis, somatic musculature and intestine of the parasites transferred into the non-permissive hosts, as early as day 4 after transfer. These data suggest that eosinophils would serve as a potential effector cell for killing of pulmonary arterial A. cantonensis in non-permissive hosts.     

Host immune reactions and worm kinetics during the expulsion of Ascaris suum in pigs

Pigs single inoculated with Ascaris suum eggs expel the majority of larvae between days 14 and 21 post inoculation (p.i.), but the role of the immune system in expulsion is unclear. To investigate the dynamics of immune responses before, during and after the expulsion of A. suum larvae, pigs inoculated with 10 000 A. suum eggs were sequentially necropsied. Ascaris suum gradually moved distally from days 10-14 p.i. and only a few larvae were left by day 21 p.i. Pronounced increases in mucosal A. suum-specific IgA antibody secreting cells (ASCs) were already found by day 10 p.i. especially in the proximal jejunum, while only small increases in parasite-specific IgM ASCs were observed by day 21 p.i. in both proximal and distal jejunum. No mucosal IgG ASC responses could be detected. Increases in systemic A. suum-specific IgG1, IgM and to a lesser extent IgA antibodies were observed, while IgG2 remained almost unchanged. The levels of eosinophils and mast cells in the small intestinal mucosa did not change throughout infection. The results demonstrate that both systemic and mucosal A. suum-specific effector mechanisms are strongly stimulated in A. suum single infections and indicate that mucosal IgA may be an important mediator in the expulsion of A. suum.     

Comparative efficacies of albendazole and the Chinese herbal medicine long-dan-xie-gan-tan, used alone or in combination, in the treatment of experimental eosinophilic meningitis induced by Angiostrongylus cantonensis

Angiostrongylus cantonensis, the rat lungworm, is the principal cause of human eosinophilic meningitis or meningoencephalitis world-wide. In the present study, the efficacies of early-stage treatment with the Chinese herbal medicine long-dan-xie-gan-tan (LDXGT) and albendazole, used alone or in combination, were evaluated in BALB/c mice with A. cantonensis-induced dysfunction of the blood-central-nervous-system barrier and eosinophilic meningo-encephalitis. Indicators of the therapeutic effect included worm recovery, histopathological scores for the meningitis, assays of tissue-type plasminogen activator (PA), urokinase-type PA and matrix metalloproteinase-9 (MMP-9) in the brain, the ratio between albumin concentrations in the cerebrospinal fluid (CSF) and serum, and counts of eosinophils in the CSF. Combined treatment with albendazole and LDXGT gave better results than monotherapy based on either drug, apparently inhibiting eosinophilic meningitis via antagonists of the PA/MMP-9 system. LDXGT may have a therapeutic role in reducing inflammatory reaction in the subarachnoid space. Monotherapy with such an anti-inflammatory drug may relieve the symptoms of mild infection and the host's immune responses to A. cantonensis larvae. In severe infection, however, co-therapy with an anthelmintic (to kill the larvae) and an anti-inflammatory agent (to provide symptomatic relief) is probably a better approach. The therapeutic strategy should be tailored to the severity of the illness and the numbers of eosinophils in the CSF.     

Immune-mediated damage is not essential for the expulsion of Nippostrongylus brasiliensis adult worms from the small intestine of mice

Nippostrongylus brasiliensis adult worms are expelled from the rat small intestine during a primary infection by two steps. First, host immune responses cause damage to the worms, and then a nonspecific inflammatory response initiates expulsion. We have tested the two-step expulsion hypothesis in mice infected with N. brasiliensis. After a primary infection in C57BL/6 mice, adult worms started to lay eggs on day 5 postinfection (p.i.) and were expelled around day 9-10 p.i. According to the rat system, 5 day- and 8-day-old worms were assumed to be 'normal' and 'damaged', respectively. When 5 day- and 8 day-old worms obtained from C57BL/6 mice were transferred surgically into the small intestine of naive C57BL/6 mice, both 5 day- and 8 day-old worms were almost simultaneously expelled by day 6 postworm implantation (p.w.i.). In contrast, when 5 day- and 8 day-old worms of mouse origin were implanted into naive Wistar rats, 8 day-old worms were expelled by day 5 p.w.i., while 5 day-old worms were expelled by day 8 p.w.i. Similar results were obtained when BALB/c mice were used. Therefore, mice can expel N. brasiliensis adult worms as rapidly as rats expel 'damaged' worms, regardless of the status of the worms ('normal' or 'damaged'). Stat6-deficient mice were unable to expel implanted 5 day-old worms up to day 10 p.w.i., suggesting that cellular mechanisms depending on Stat6-signalling system are necessary for the expulsion. When N. brasiliensis adult worms obtained from Stat6-deficient mice 5 and 15 days after a primary infection were implanted into Wistar rats, the former established in the recipient rats for approximately 1 week and were then expelled by day 10 p.w.i., whereas the latter were expelled by day 4 p.w.i. These results suggest that immune-mediated damage of N. brasiliensis adult worms (first step) is not a prerequisite for expulsion from the small intestine of mice, although adult worms are actually damaged by Stat6-independent immune mechanisms.