Th2 CD4+ T Cells Are Necessary and Sufficient for Schistosoma-Pulmonary Hypertension.

Background Inflammation underlies many forms of pulmonary hypertension (PH), including that resulting from Schistosoma infection, a major cause of PH worldwide. Schistosomiasis-associated PH is proximately triggered by embolization of parasite eggs into the lungs, resulting in localized type 2 inflammation. However, the role of CD4+ T cells in this disease is not well defined. Methods and Results We used a mouse model of schistosomiasis-associated PH, induced by intraperitoneal egg sensitization followed by intravenous egg challenge, with outcomes including right ventricle systolic pressure measured by cardiac catheterization, and cell density and phenotype assessed by flow cytometry. We identified that embolization of Schistosoma eggs into lungs of egg-sensitized mice increased the perivascular density of T-helper 2 (Th2) CD4+ T cells by recruitment of cells from the circulation and triggered type 2 inflammation. Parabiosis confirmed that egg embolization is required for localized type 2 immunity. We found Th2 CD4+ T cells were necessary for Schistosoma-induced PH, given that deletion of CD4+ T cells or inhibiting their Th2 function protected against type 2 inflammation and PH following Schistosoma exposure. We also observed that adoptive transfer of Schistosoma-sensitized CD4+ Th2 cells was sufficient to drive type 2 inflammation and PH. Conclusions Th2 CD4+ T cells are a necessary and sufficient component for the type 2 inflammation-induced PH following Schistosoma exposure.

Clinical Perspective

What Is New?

Inflammation contributes to the pathogenesis of many forms of pulmonary hypertension (PH), but how it does this is unknown.

Schistosomiasis is a major cause of PH worldwide.

We found that embolization of Schistosoma mansoni eggs is critical to experimental schistosomiasis‐triggered PH, and also found that T‐helper 2 CD4 T cells are necessary and sufficient for the type 2 inflammation and PH that is triggered by experimental Schistosoma exposure in mice.

What Are the Clinical Implications?

Targeting T cells may be a future way to prevent or treat inflammatory forms of PH, such as that induced by schistosomiasis.

Introduction

Inflammation can trigger the development of pulmonary vascular disease in humans and experimental models, resulting in pulmonary hypertension (PH). Furthermore, there is a significant expansion of B and T cells in the pulmonary vascular adventitia in PH.1, 2 These lymphocytes drive adaptive immunity, but earlier studies in this field have primarily used triggers of innate immunity, such as hypoxia, leaving the roles of these cells unclear.

Schistosomiasis, resulting from infection with parasitic helminth Schistosoma, has a worldwide prevalence of >200 million.3 Schistosomiasis is a major cause of World Health Organization Group 1 pulmonary arterial hypertension (PAH) worldwide, affecting around 6% of those chronically and recurrently infected.4, 5 Despite the availability of effective antihelminthic treatment, numerous social, economic, and political barriers have prevented widespread treatment of schistosomiasis, leading to its characterization as a “neglected tropical disease.” Once significant schistosomiasis‐associated PAH has developed, it is clinically no longer responsive to antihelminthic therapy; vasodilators can improve symptoms and prolong life, but the condition remains fatal.

Schistosomiasis‐associated PAH is thought to be triggered by embolization of Schistosoma mansoni eggs into the lungs (the Schistosoma species particularly associated with PAH), resulting in inflammation and associated severe vascular remodeling. The type 2 immune response triggered by Schistosoma egg antigens is characterized by T cells, macrophages, eosinophils, and basophils, and the release of inflammatory cytokines, including interleukin (IL)‐4, ‐5, ‐10, and ‐13.6, 7 The pulmonary pathology includes peri‐egg granulomas and vascular remodeling with perivascular infiltrates and thickening of the vessel walls. We and others have previously shown that adaptive immunity and type 2 inflammation is required for the pulmonary phenotype.8, 9, 10 These cytokines trigger a cascade of inflammation, including the recruitment of Ly6C+ monocytes, into the lung interstitium, where these cells express thrombospondin‐1 resulting in local activation of transforming growth factor beta, which drives the vascular disease.11

In this study, we hypothesized that T‐helper 2 (Th2) CD4+ T cells are both necessary and sufficient to drive Schistosoma‐induced type 2 inflammation and PH. We found that embolization of Schistosoma eggs in egg‐sensitized mice induces the recruitment and proliferation of CD4+ T cells in the lungs, and blockade of either T cells or, more specifically, their Th2 phenotype inhibited the pulmonary vascular pathology. In addition, adoptive transfer of primed CD4+ Th2 cells from egg‐sensitized mice was sufficient to convey the adaptive immunity which drives the PH phenotype. These results reveal that egg‐antigen–specific Th2 CD4+ effector T cells are the critical immunological driver of Schistosoma‐PH.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Animals

C57Bl6/J wild‐type and Rag1 −/− mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Il4 Il13 mice were obtained from Dr Thomas Wynn (NIAID, NIH). Il4 Il13 Rag1 −/− mice were generated by crossing these 2 lines. The mice used were female and between 6 and 8 weeks of age at the start of the experiment. All animals were housed under specific pathogen‐free conditions in an American Association for the Accreditation of Laboratory Animal Care–approved facility of University of Colorado (Aurora, CO). All animal studies were approved by the University of Colorado Institutional Animal Care and Use Committee. The procedures followed were in accord with institutional guidelines.

Schistosoma‐Induced PH Model

Live S. mansoni eggs were obtained by homogenizing and purifying livers of Swiss–Webster mice preinfected with S. mansoni cercariae, from the Biomedical Research Institute (Rockville, MD). Similar to earlier publications,8, 11, 12 experimental mice were intraperitoneally sensitized to 240 S. mansoni eggs/g body weight and then 2 weeks later intravenously challenged with 175 S. mansoni eggs/g body weight. Control mice were unexposed to S. mansoni eggs.

Shared Circulatory System (Parabiosis)

Parabiosis was performed using published techniques.13 In brief, mice were anesthetized with isoflurane, carprofen, and buprenorphine, incisions made on opposing flanks, and skin flaps sutured together. Postoperatively, carprofen and buprenorphine were administered for pain and sulfamethoxazole‐trimethoprim antibiotic administered in rodent chow for 10 days. Two weeks after surgery, to allow time for recovery, mice were used for subsequent experiments.

Adoptive Transfer of CD4 Cells or Splenocytes

Some of the mice were used as donors for CD4+ T cells or splenocytes. These mice were euthanized and spleens removed, homogenized, and CD4+ cells purified by negative selection column (BD Biosciences, San Jose, CA). CD4 cells (3–8×106) were administered by intraperitoneal injection to each recipient. Recipients of splenocytes received 3.8×107 cells from homogenized donor spleens by intraperitoneal injection.

Assessment of PH

At the conclusion of the experiments, mice underwent terminal right heart catheterization and tissue harvest, as described previously.8, 11, 12 Briefly, mice were sedated with intraperitoneal ketamine‐xylazine and a tracheostomy placed and mechanical ventilation initiated at 6 cc/kg. Sharp dissection was used to open the abdomen and diaphragm, and a 1‐Fr pressure‐volume catheter (PVR‐1035; Millar ADInstruments, Houston, TX) was placed directly into the right ventricular and then left ventricular chambers through the free walls. Lungs were then flushed with PBS, the right bronchus sutured, and the left lung inflated with 1% low‐melt agarose for formalin fixation and paraffin embedding for histology, and the right lung divided for snap freezing for protein quantification or placed in RNAlater (Life Technologies, Carlsbad, CA) for RNA quantification.

Flow Cytometry

Three days after intravenous egg augmentation, mouse lungs were flushed with PBS and digested for flow cytometry analysis as previously reported.11 Briefly, lungs were digested with Liberase (Roche, Waiblingen, Germany) and dissolved in RPMI medium (Mediatech, Corning, NY); tissue was disrupted by passing it 5 times through a 16‐Gg needle followed by 5 times through an 18‐Gg needle. Cells were then filtered using a 100‐μm cell strainer (Thermo Fisher Scientific, Waltham, MA) and centrifuged for 5 minutes at 300g. Red blood cells were lysed with ACK lysis buffer (Gibco, Grand Island, NY), and cells were resuspended in RPMI, filtered again, centrifuged, and resuspended into flow wash buffer (5% BSA in PBS with EDTA). Blocking of nonspecific Fc‐receptormediated antibody binding was performed (CD16/CD32; BD Biosciences), and cells were stained (extracellularly for all antibodies except IL‐4) using the antibodies listed in Table S1. Data were acquired using a BD Biosciences LSRII flow cytometer with a BD Biosciences Facs DIVA software. Compensation was calculated using DIVA software based on compensation controls for each fluorochrome used in the experiment. The raw data were then analyzed using FlowJo (version 7.6; BD Biosciences). RT‐PCR for IL‐4 and IL‐13 mRNA was performed on sorted cells using primers from Applied Biosystems (Foster City, CA), with the 2−ΔCt method, using β‐actin and glyceraldehyde‐3‐phosphate dehydrogenase as reference genes.

Protein Quantification

Samples of frozen right lung tissue were macerated and sonicated in RIPA buffer containing antiproteases. Total protein concentration was determined by Bradford assay, and IL‐4 and IL‐13 protein concentrations were quantified by ELISA in mouse lung lysates using the kits listed in Table S2.

Histopathology Assessment

Immunostaining for CD3 and α‐smooth muscle actin (α‐SMA) was performed on formalin fixation and paraffin embedding mouse lung tissue, using the reagents listed in Table S3. For quantification of perivascular CD3+ density, slides were scanned using an Aperio VERSA slide scanner (Leica Biosystems, Buffalo Grove, IL), the number of CD3+ T cells in the perivascular region was counted, and the cross‐sectional area of the adventitial space (defined as the space between the outside of the medial layer and inside of the adjacent alveoli) and vessel radius (calculated from the measured cross‐sectional area encompassed by the outside of the medial layer) was determined for 10 to 21 vessels per specimen. For quantification of fractional media thickness, images of 10 to 12 vessels of each specimen from the α‐SMA–stained slides were acquired using a Nikon Eclipse E800 microscope (Nikon, Melville, NY) with a black and white charge‐coupled device camera (Photometrics, Tuscon, AZ). External and internal perimeters of the media were identified using image‐processing software (Image Pro Plus v4.5.1; Media Cybernetics, Bethesda, MD), the radii of each layer calculated (r=sqrt(A/pi)), and the fractional thickness calculated from the difference between the 2 radii divided by the external radius. Peri‐egg granuloma volumes were measured using a stereological method, termed the optical rotator, with formalin fixation and paraffin embedding tissue was stained with hematoxylin and eosin, 8 to 10 images of granulomas with single visible egg acquired for each sample, and the rotator method for object volume estimation applied using the egg as the central point (Image Pro Plus v4.5.1; Media Cybernetics).8, 14

Egg Burden Quantification

The number of residual S. mansoni eggs in mouse lung tissue was determined by digesting 20 to 30 mg of frozen tissue in 4% KOH for 18 hours at 33°C and counting the number of eggs in the digest.15

Statistical Analysis

ProStat (Poly Software International, Pearl River, NY) and SigmaPlot (Systat Software, San Jose, CA) were used to perform statistical analyses and graphs. Differences between 2 groups were assessed by t test; for ≥3 groups, differences were assessed by ANOVA followed by Tukey's post hoc test. Non‐normally distributed data or groups with samples <5 were analyzed by nonparametric analysis. P values <0.05 were considered statistically significant.

Results

There Is an Increased Number of Th2 CD4+ T Cells in Lungs of Mice Exposed to Schistosoma

We used an established experimental murine model of S. mansoni–induced PH, which is triggered by intraperitoneal egg injection (sensitization) followed by intravenous embolization of egg antigens into the pulmonary circulation 2 weeks later (challenge; Figure 1A).8, 10, 11 This model mimics the natural form of infection with S. mansoni, a helminth parasite that migrates to the portal vasculature and lays hundreds of eggs per day, which promote development of a systemic type 2 adaptive immune response to the egg antigens. Some patients with chronic and severe hepatosplenic disease develop portal hypertension, resulting in opening of portocaval shunts that facilitate the embolization of Schistosoma eggs from the portal venous system to the pulmonary circulation. This experimental model thus focuses on the immunological trigger of intrapulmonary vascular disease, while avoiding the confounding role of portal hypertension that could also contribute to PH development. We have shown previously that mice must be sensitized to parasite eggs to develop PH after egg embolization, suggesting a role for egg‐antigen–specific adaptive immune cells in the pathogenesis of PH.8

Figure 1

Intrapulmonary Th2 CD4+ T cells are increased following Schistosoma exposure. A, Experiment outline. B, Representative sorting of CD3+ CD4+ singlets from whole lung digest from unexposed or Schistosoma‐exposed mice and the absolute number of CD4+ cells per mouse lung. C, Intracellular IL‐4 staining of sorted cells (not further stimulated; representative of 2 independent experiments). D, Representative immunostaining and quantification of the CD3+ T‐cell profile density in vessel adventitia (asterisk, vessel lumen; arrows, representative positive CD3 cells; scale bars, 100 μm). Mean±SD plotted; t test, *P<0.05; **P<0.01; ***P<0.005. FSC indicates forward scatter; IL‐4, interleukin‐4; SSC, side scatter; WT, wild type.

Flow cytometry of cell‐dispersed murine lungs showed that the absolute number of CD4+ T cells increased in Schistosoma‐exposed mice (Figure 1B and Figure S1), with a large increase in the percent of CD4+ T cells expressing intracellular IL‐4, consistent with a Th2 phenotype (Figure 1C). To confirm the Th2 phenotype of recruited cells, we sorted CD3+CD4+ cells and observed that both IL‐4 and IL‐13 mRNA were measurable in cells from Schistosoma‐exposed mice, but neither was detectable in cells from unexposed mice (Figure S2).

By immunostaining, we quantified CD3+ T cells in the perivascular adventitia specifically, a position most likely to interact with the vasculature. We determined that density was significantly higher in Schistosoma‐exposed wild‐type mice as compared with unexposed mice, around both small and large pulmonary vessels (Figure 1D and Figure S3). This observation in experimentally treated animals parallels the reported increase in perivascular T cells in lung tissue of patients who died of schistosomiasis‐associated PAH by Mauad et al.16

There could also be an increase in circulating CD4 T cells, potentially preceding the observed increase in lung tissue, and we thus assessed the time course of CD4 T‐cell density in the blood and in the lung tissue. We observed that there was evidence of an increased density of CD4 T cells in both the lung and in the blood, starting 3 days after intravenous challenge, but did not find that there was an earlier change in 1 compartment compared with the other (Figure S4).

Eosinophils and basophils are also present in Schistosoma‐exposed mouse lungs and could contribute to the PH pathology. Flow cytometry of lung digest samples for these cells revealed a 20‐fold increase in the number of eosinophils and a 10‐fold increase in the number of basophils, both peaking 3 days after intravenous Schistosoma egg challenge (Figures S1 and S5).

Schistosoma Egg Embolization Into the Lungs Triggers CD4+ T‐Cell Recruitment From the Circulation and Th2 Phenotype

To test whether Schistosoma egg embolization into the pulmonary vasculature drives localized T‐cell activation and type 2 immunity, we performed parabiosis surgery, linking the circulatory systems of 2 mice by creation of anastomoses in the subcutaneous tissue in the flanks. We administered intraperitoneal egg sensitization to both, but then intravenous egg challenge to only 1 parabiont (Figure 2A): Schistosoma eggs have a short axis diameter of ≈50 μm, preventing further migration distal to the pulmonary vasculature, such as to the noninjected partner. This revealed, interestingly, a similar density of CD4+ T cells in lungs of both parabionts (Figure 2B), which was as high as the density in an individually challenged mouse (mean of 4×105; Figure 1B). These data indicate that intravenous antigen challenge can induce a lung‐specific homing mechanism, such as has been reported for inhaled antigens.17 We observed there was a higher number of proliferating (Ki67+) CD4+ T cells in lungs of the parabiont that received the intravenous eggs as compared with the unchallenged parabiont: Here, Ki67+ as a marker of proliferation indicates CD4 T‐cell activation (Figure 2C). We also observed more IL‐4 expression in lung tissue and a trend toward increased IL‐13 levels in the parabiont that received the intravenous eggs as opposed to the unchallenged parabiont (Figure 2D and 2E), results further consistent with egg embolization triggering localized type 2 immunity.

Figure 2

Intravenous eggs are necessary for CD4+ T‐cell activation and type 2 immunity. A, Design of the experiment. B, Absolute number (Abs. nos.) of CD3+ CD4+ cells in lungs of Schistosoma‐exposed and unexposed parabionts (representative of 3 parabiont pairs per experimental group). C, Fraction of CD3+ CD4+ cells, which are Ki67+ in the lungs of each parabiont. D, IL‐4 and E, IL‐13 protein concentrations (N=4/group; however, the IL‐13 was undetected in 2 and 1 samples from IP only and IP/IV groups, respectively) in whole lung lysates of each parabiont (N=number of animals/group; mean±SD plotted; t test for [B and C], rank sum test for all [D]; *P<0.05; ***P<0.005). IL‐4 indicates interleukin‐4; IL‐13, interleukin‐13; IP, intraperitoneal sensitization alone; IP/IV, intraperitoneal sensitization and intravenous challenge with Schistosoma mansoni eggs; NS, nonsignificant.

Deficiency of CD4+ T Cells Protects Against Schistosoma‐PH

To understand the functional role of CD4+ T cells in Schistosoma‐PH, we studied the phenotype of mice with an absence of functional T and B cells using recombinase 1–deficient (hereafter designated Rag −/−) mice, which prevents T‐ and B‐cell receptor maturation.18 In wild‐type mice, Schistosoma egg embolization following earlier egg sensitization results in a significant increase in right ventricular systolic pressure (RVSP), vascular media thickness, and IL‐4 and IL‐13 cytokine concentrations in lung tissue (Figure 3). However, upon Schistosoma egg embolization, previously egg‐sensitized Rag −/− mice had no increase in RVSP, change in vascular media thickness, or a change in concentration of pulmonary IL‐4 and IL‐13 as compared with unexposed Rag −/− mice. The volume of the peri‐egg granulomas—which consist of macrophages, fibroblasts, eosinophils, and T cells—can be used as another readout of total type 2 inflammation in lung.8, 19 Similar to the lower IL‐4/IL‐13 cytokine levels, we observed that granulomas were also much smaller in Schistosoma‐challenged Rag −/− mice compared with wild‐type mice challenged with Schistosoma.

Figure 3

Compared with wild‐type (WT) mice, Rag −/− mice, which lack B and T cells, do not develop PH or type 2 inflammation following Schistosoma exposure. A, Experimental outline. B, Right ventricular systolic pressure (RVSP); (C) media thickness (asterisks, vessel lumen; scale bars, 50 μm); (D) IL‐4 and (E) IL‐13 protein concentrations; (F) estimated granuloma volumes, as measured by stereology; (G) number of Schistosoma mansoni eggs recovered from lung digests in whole lung lysates of unexposed or Schistosoma‐exposed WT or Rag −/− mice (mean±SD plotted; t test for [B, C, F, and G], rank‐sum test for [D and E]; *P<0.05; ***P<0.005; ****P<0.001). DAPI indicates 4′,6‐diamidino‐2‐phenylindole; IL‐4, interleukin‐4; IL‐13, interleukin‐13; IP/IV, intraperitoneal/intravenous S. mansoni eggs; NS, nonsignificant; PH, pulmonary hypertension; RV, right ventricular; α‐SMA, alpha smooth muscle actin.

We observed no significant differences in left ventricular systolic pressure, right or left ventricular diastolic pressures, heart rate, or body weight between Schistosoma‐exposed and unexposed Rag −/− mice (Table S4). Of note, the otherwise unchallenged Rag −/− mice had a higher RVSP than unchallenged wild‐type mice, suggesting that deficiency of B and T cells can impair normal lung vascular development, but there was no evidence of a baseline change in pulmonary vessel media thickness or type 2 immune activation.

Reconstitution of Rag −/− Mice With CD4+ T Cells Restores Schistosoma‐PH

The Rag −/− phenotype could be confounded by B‐cell deficiency: B cells can present antigen to effector and memory T cells and can modulate Th2 immunity in schistosomiasis.20 To confirm that CD4+ T cells specifically are the subset of lymphocytes necessary for the Schistosoma‐PH phenotype, we adoptively transferred CD4+ T cells, isolated by negative selection from the spleens of naïve wild‐type mice, into Rag mice and then sensitized and challenged them with Schistosoma eggs (Figure 4A). Adoptive transfer of wild‐type CD4+ T cells was sufficient to induce PH as evidenced by higher RVSP and increased media thickness, as compared with CD4+ T‐cell–reconstituted Rag −/− mice that remained unchallenged (Figure 4B and 4C). These gain‐of‐function mice also had significant increases in the concentration of IL‐4 and IL‐13 following Schistosoma challenge (Figure 4D and 4E). There was also a modest trend (P=0.1, with correction for multiple comparisons) toward an increase in the estimated volume of the peri‐egg granulomas between Schistosoma‐exposed Rag −/− and Rag −/− mice reconstituted with wild‐type CD4+ T cells (Figure S6). We considered the possibility that the immune system impairment in Rag −/− mice may block the clearance of schistosome eggs, which could confound the PH phenotype. However, we observed no difference in the residual egg burden in lungs measured 1 week after intravenous challenge. There were no significant differences in systemic pressures, heart rate, or body weight between Schistosoma‐exposed and unexposed Rag −/− mice reconstituted with wild‐type CD4+ T cells (Table S5).

Figure 4

Adoptive transfer of wild‐type CD4+ T cells into Rag −/− mice restores Schistosoma‐induced PH and type 2 immunity. A, Experimental outline of wild‐type CD4+ T‐cell reconstitution into Rag −/− mice. B, Right ventricular systolic pressure (RVSP); (C) media thickness; (D) IL‐4 and (E) IL‐13 protein concentrations in whole lung lysates in unexposed and Schistosoma‐exposed Rag −/− mice reconstituted with wild‐type CD4+ T cells (mean±SD plotted; t test for all panels: *P<0.05; **P<0.01; ***P<0.005; asterisks, vessel lumen; scale bars, 50 μm). DAPI indicates 4′,6‐diamidino‐2‐phenylindole; IL‐4, interleukin‐4; IL‐13, interleukin‐13; IP/IV, intraperitoneal/intravenous Schistosoma mansoni eggs; PH, pulmonary hypertension; RV, right ventricular; α‐SMA, alpha smooth muscle actin.

Egg Antigen‐Specific Th2 Cells Are Required for Schistosoma‐PH

IL‐4/IL‐13–deficient mice do not develop Schistosoma‐PH,10 suggesting that cytokine production by CD4+ T cells may be required to drive type 2 inflammation and PH. To determine the role of Th2 cells in Schistosoma‐PH, we reconstituted Rag −/− mice with CD4+ T cells harvested from Il4 −/− Il13 −/− donors. We observed that these mice were protected from PH following Schistosoma sensitization and intravenous challenge, as evidenced by lower RVSP, less vascular remodeling, and less intrapulmonary IL‐4 and IL‐13, as compared with Schistosoma‐exposed Rag −/− mice reconstituted with wild‐type CD4+ T cells (Figure 5). Furthermore, the Schistosoma‐exposed Rag −/− mice reconstituted with Il4 −/− Il13 −/− CD4+ T cells had significantly smaller peri‐egg granulomas compared with recipients of wild‐type CD4+ T cells, consistent with reduced type 2 inflammation. There were no significant differences in systemic pressures, heart rate, or body weight between Schistosoma‐exposed Rag −/− mice reconstituted with wild‐type or Il4 −/− Il13 −/− CD4+ T cells (Table S6).

Figure 5

Adoptive transfer of Th2‐deficient CD4+ T cells into Rag −/− mice blocks Schistosoma‐induced PH and type 2 immunity. A, Right ventricular systolic pressure (RVSP); (B) media thickness; (C) IL‐4 and (D) IL‐13 protein concentrations (in Rag −/− recipients of dKO BM, IL‐13 levels were undetectable in 3 of 4 samples); (E) estimated granuloma volumes, as measured by stereology; (F) number of Schistosoma mansoni eggs recovered from lung digests in whole lung lysates in unexposed and Schistosoma‐exposed Rag −/− mice reconstituted with Il4 +/+ Il13 +/+ (WT) or Il4 −/− Il13 −/− (dKO) CD4 T cells mean±SD plotted; rank‐sum test for all panels, *P<0.05; **P<0.01; asterisks, vessel lumen; scale bars, 50 μm. BM indicates bone marrow; DAPI, 4′,6‐diamidino‐2‐phenylindole; dKO, double knockout; IL‐4, interleukin‐4; IL‐13, interleukin‐13; IP/IV, intraperitoneal/intravenous Schistosoma mansoni eggs; NS, nonsignificant; PH, pulmonary hypertension; α‐SMA, alpha smooth muscle actin; WT, wild type.

Primed CD4+ T Cells Are Sufficient to Drive Schistosoma‐PH in Mice Lacking an Adaptive Immune System

Adaptive immunity is required for the Schistosoma‐PH phenotype, given that a single intravenous challenge of S. mansoni eggs alone is inadequate to trigger PH.10 We suspected that the key cells which need to be activated in this adaptive immune response are the CD4+ T cells. To test this hypothesis, we first sensitized wild‐type mice with Schistosoma eggs, and then transferred CD4+ T cells isolated from the spleens of these primed mice into Rag −/− mice, followed by intravenous challenge with Schistosoma eggs alone (without further sensitization). The control animals for this experiment were Rag −/− mice reconstituted with CD4+ T cells from unsensitized donors and challenged with Schistosoma eggs. Interestingly, we observed significantly higher RVSP and vascular remodeling in the challenged Rag −/− recipients of primed CD4+ T cells, as compared with challenged recipients of unprimed CD4+ T cells, accompanied by significantly higher IL‐4 and IL‐13 concentrations in lung (Figure 6). There was also a trend toward increased granuloma volumes in the Rag −/− recipients of primed CD4+ T cells, as compared with recipients of unprimed CD4+ T cells (P=0.079). These results suggest that egg‐antigen–specific CD4+ T cells are the only component of the adaptive immune response required for Schistosoma‐PH. There were no significant differences in systemic pressures, heart rate, or body weight between Rag −/− recipients of Schistosoma‐sensitized or unsensitized wild‐type CD4+ T cells followed by Schistosoma challenge (Table S7).

Figure 6

Reconstitution of Rag mice with CD4+ T cells from egg‐sensitized mice is sufficient to drive type 2 inflammation and PH following Schistosoma egg challenge alone. A, Experimental outline. B, Right ventricular systolic pressure (RVSP); (C) media thickness; (D) IL‐4 and (E) IL‐13 protein concentration; (F) estimated granuloma volumes, as measured by stereology; (G) number of Schistosoma mansoni eggs recovered from lung digests in whole lung lysates in intravenously challenged Rag −/− mice reconstituted with unprimed or primed CD4+ T cells mean±SD plotted; t test, *P<0.05; ****P<0.001; asterisks, vessel lumen; scale bars, 50 μm. DAPI indicates 4′,6‐diamidino‐2‐phenylindole; IL‐4, interleukin‐4; IL‐13, interleukin‐13; IP/IV, intraperitoneal/intravenous S. mansoni eggs; NS, nonsignificant; PH, pulmonary hypertension; RV, right ventricular; α‐SMA, alpha smooth muscle actin.

IL‐4/IL‐13 Expression by Parenchymal Cells Is Also Required for Schistosoma‐PH

Type 2 immunity by cells other than CD4 T cells may also contribute to Schistosoma‐induced PH. We crossed B‐/T‐cell–deficient mice with IL‐4/IL‐13–deficient mice to generate mice which lack B cells, T cells, and expression of IL‐4 and IL‐13 by all other cells (Il4 −/− Il13 −/− Rag −/−). When sensitized and challenged with Schistosoma eggs, these mice also did not develop pulmonary hypertension (Figure 7A). We then reconstituted the mice with CD4+ T cells from wild‐type donors and observed that these mice had only a nonsignificant trend toward mildly increased RVSP, with a mean increase of only 2 mm Hg compared with nonreconstituted mice. Similarly, there were nonsignificant differences in the Fulton index, vascular remodeling, and granuloma volumes following Schistosoma sensitization and challenge (Figure 7A through 7D). There were also no significant changes in IL‐4 or IL‐13 whole lung expression (Figure 7E and 7F). Furthermore, we found that reconstituting these mice with splenocytes from a wild‐type mouse (which includes CD4 T cells as well as many additional cells, including ILC2s, eosinophils, basophils, and others) did not alter the PH or type 2 immunity phenotype compared with CD4 T cells alone (Figure 7). These results indicate that there is an additional requirement for IL‐4/IL‐13 expression by cells other than CD4 T cells to trigger the type 2 immunity and PH phenotype in response to Schistosoma exposure, and that these additional cells likely have an extravascular, parenchymal location.

Figure 7

PH phenotype of Il4−/−Il13−/−Rag−/− mice exposed to IP/IV Schistosoma eggs is not rescued by reconstitution with either CD4 T cells or splenocytes. A, Right ventricular systolic pressure (RVSP); (B) Fulton index; (C) media thickness; (D) granuloma volumes; and (E) IL‐4 and (F) IL‐13 protein concentration in whole lung lysates in Schistosoma‐sensitized and intravenously challenged Il4−/−Il13−/−Rag−/− mice reconstituted with no cells, CD4+ T cells, or splenocytes from a wild‐type mouse mean±SD plotted; ANOVA P values: 0.07, 0.16, 0.11, 0.11, 0.81, and 0.33, respectively; t‐test values are shown in all panels; IL‐13 was undetectable in 4 of 4, 1 of 4, and 2 of 5 samples in the no cells, CD4 T cells, and splenocytes groups, respectively. IL‐4 indicates interleukin‐4; IL‐13, interleukin‐13; IP/IV, intraperitoneal/intravenous S. mansoni eggs; LV, left ventricular; RV, right ventricular.

Discussion

We observed Th2 CD4+ T cells recruited from the circulation of Schistosoma‐egg–sensitized mice are both necessary and sufficient for orchestrating the type 2 inflammation and experimental PH triggered by the pulmonary embolization of S. mansoni eggs. The significance of these data is reinforced by previous findings of type 2 inflammation in the lungs and peripheral blood of patients with schistosomiasis‐associated PAH8, 21 and in lungs and peripheral blood of mice exposed to Schistosoma.8, 22 Furthermore, an increased density of perivascular CD4+ T cells has been previously reported in human schistosomiasis‐associated PAH,16 which closely parallels the observations here in Schistosoma‐PH mice. Perivascular CD4 T cells have also been observed in idiopathic PAH,1, 2 but the role for these cells in the pathogenesis of this particular disease etiology is unclear. The clear role of antigen‐specific adaptive immunity in Schistosoma‐PH facilitates mechanistic analysis of the connection between antigen and vascular remodeling, in contrast to the use of hypoxia or nonspecific inflammation following monocrotaline treatment or vascular endothelial growth factor receptor 2 blockade commonly used in other PH animal models.

Previous studies have implicated other CD4+ T‐cell phenotypes in other forms of experimental PH. In athymic, T‐cell–deficient rats, vascular endothelial growth factor receptor 2 blockade induces a severe PH phenotype, whereas restoring FoxP3+ regulatory T cells is protective.23 Mice challenged with inhaled ovalbumin can develop type 2 immunity and some degree of PH, and the PH phenotype is inhibited by anti‐CD4 antibody administration, or blocking IL‐4 or IL‐13 cytokines.24 In contrast, Th17 T cells contribute to the development of hypoxia‐induced PH.25, 26 Interestingly, mice challenged with a combination of hypoxia and inhaled ovalbumin develop PH, which is mediated predominantly by type 2 immunity.27 In this model, the PH phenotype is inhibited by deletion of the Th2 promoting receptor, CRTH2, with the phenotype restored by supplementing the mice with wildtype CD4 T cells, and reversed by IL‐4/IL‐13 dual neutralization.27 An open question is whether inflammation contributes to the ongoing propagation of PH after the disease has been initiated. The potential for targeted immune blockade to reverse established PH needs to be tested in future studies: The clinical relevance is the observed general absence of clinical benefit of broad immunosuppressant therapy in PH patients, except in particularly inflammatory conditions such as lupus‐associated disease.28

Our use of Rag −/− mice could be confounded by a phenotype resulting from B‐cell deficiency: B cells can present antigen to effector and memory T cells and are reported to modulate Th2 immunity in schistosomiasis.29 CD8+ T cells are also depleted in Rag −/− mice and could contribute to the PH phenotype. However, the regained Schistosoma‐PH phenotype in Rag −/− mice following adoptive transfer with CD4+ T cells alone suggests that B cells and CD8 T cells are not required for Schistosoma‐PH. We found that Rag −/− mice at baseline have PH, potentially attributed to abnormal perinatal lung development with this germline absence of mature T and B cells. It has been reported that primary immunodeficiency in humans can be associated with the spontaneous occurrence of PH.30, 31

We investigated a potential role for other non‐B/T cells in modulating type 2 immunity in Schistosoma‐PH. We found that expression of IL‐4/IL‐13 by parenchymal cells was required for the proximate activation of type 2 immunity (IL‐4 expression in particular) and PH: We suspect that these are dendritic cells which guide the activation of CD4 T cells, also called “signal 3.”32 Consistent with these data, we previously observed that IL‐4 knockout mice had less right ventricular hypertrophy and smaller granuloma volumes compared with wild‐type mice following Schistosoma challenge.8 Our data do not exclude the possibility that other non‐B/T cells, such as ILC2s, eosinophils, or basophils, could amplify the type 2 immunity or drive PH downstream of Th2 CD4+ T cells, and indeed we found an increase in density of eosinophils and basophils in Schistosoma‐exposed mouse lungs. However, in models of liver fibrosis resulting from Schistosoma infection, both eosinophils and basophils have been shown to be dispensable to granuloma formation and fibrosis.33, 34 We suspect that circulating monocytes and interstitial macrophages are critical targets of Th2 signaling, because we recently reported that recruitment of CCR2+Ly6c+ monocytes to the vascular adventitia is a necessary step for transforming growth factor beta activation and the PH phenotype in Schistosoma‐exposed mice.11

We observed evidence of lung‐specific homing of activated CD4 T cells, given that a parabiosis experiment in which only 1 parabiont received intravenous egg challenge resulted in a comparable increase in density of CD4+ T cells in both parabionts. Lung‐specific CD4+ T‐cell homing has been previously described in mycobacterium infection35, 36 and in response to inhaled ovalbumin or influenza infection by a C‐C chemokine receptor type 4 (CCR4)‐dependent mechanism.17

A highly relevant modulation of CD4 T cells in Schistosoma‐PAH will occur in the estimated 6M individuals worldwide who are simultaneously infected with human immunodeficiency virus.37 Given that both human immunodeficiency virus and Schistosoma are triggers of PAH, it is possible that dually infected individuals will have a higher prevalence of PAH, and in line with this hypothesis is the observation that there is a higher prevalence of liver disease complicating schistosomiasis in dually infected individuals, as compared with those infected with Schistosoma alone.38 Alternatively, it is possible that the immune suppression resulting from human immunodeficiency virus infection could block the type 2 immunity required for Schistosoma‐induced PH as shown here.

In summary, we found Th2 CD4+ T cells are necessary and sufficient for the type 2 immunity that drives PH following Schistosoma exposure. Depleting CD4+ T cells, blocking the Th2 phenotype, or switching the cells to a different phenotype could be a potential therapeutic target in inflammation‐triggered forms of PAH, including that following Schistosoma infection.

Sources of Funding

Grant funding was provided by the American Heart Association Grants 17POST33670045 and 19CDA34730030 (Kumar); NIH Grants P01HL014985 (Tuder and Graham), R03HL133306 (Graham), and R01HL135872 (Graham). Ministerio de Economia y Competitividad (SAF2016‐77222‐R to Cogolludo), with funds cofinanced by ERDF (FEDER) Funds from the European Commission, “A way of making Europe,” and the Cardiovascular Medical Research and Education Fund (Cogolludo and Butrous).

Disclosures

None.

Table S1. Flow Cytometry Reagents

Table S2. Protein Quantification Reagents

Table S3. Immunostaining Reagents

Table S4. Left Ventricular Systolic Pressure (LVSP), Right and Left Ventricular Diastolic Pressures (RVDP/LVDP), Heart Rate (HR), and Body Weight (Bd. wt.) of Schistosoma‐Exposed (IP/IV Eggs) and Unexposed Rag −/− Mice Mean±SD. N=6 to 8 Mice Per Group

Table S5. Left Ventricular Systolic Pressure (LVSP), Right and Left Ventricular Diastolic Pressures (RVDP/LVDP), Heart Rate (HR), and Body Weight (Bd. wt.) of Schistosoma‐Exposed (IP/IV) and Unexposed Rag −/− Mice Reconstituted With Wild‐type CD4+ T Cells Mean±SD. N=3 to 13 Mice Per Group

Table S6. Left Ventricular Systolic Pressure (LVSP), Right and Left Ventricular Diastolic Pressures (RVDP/LVDP), Heart Rate (HR), and Body Weight (Bd. wt.) of Schistosoma‐Sensitized and Unsensitized Wild‐Type CD4+ T Cells Reconstituted Into Rag −/− Mice. Mean±SD. N=8 Mice Per Group

Table S7. Left Ventricular Systolic Pressure (LVSP), Right and Left Ventricular Diastolic Pressures (RVDP/LVDP), Heart Rate (HR), and Body Weight (Bd. wt.) of Schistosoma‐Exposed Rag −/− Mice Reconstituted With Il4 +/+ Il13 +/+CD4+ T Cells and Il4 −/− Il13 −/− CD4+ T Cells

Figure S1. Flow gating strategies for CD4 T cells, eosinophils, and basophils.

Figure S2. IL‐4 and IL‐13 mRNA expression in CD4+ T cells from control and Schistosoma‐exposed mice.

Figure S3. Representative images of adventitial space.

Figure S4. Time course of CD4 T cells in lung and blood.

Figure S5. Time course of eosinophils and basophils in lung.

Figure S6. Granuloma volumes and residual egg density in Schistosoma‐exposed wild‐type, Rag −/−, and Rag −/− mice reconstituted with wild‐type CD4+ T cells.

Click here for additional data file.