American Journal of Respiratory and Critical Care Medicine

Rationale: Pulmonary arterial hypertension (PAH) and pulmonary veno-occlusive disease (PVOD) both display occlusive remodeling of the pulmonary vasculature responsible for increased pulmonary vascular resistances. Cytotoxic T (CTL), natural killer (NK), and natural killer T (NKT) cells play a critical role in vascular remodeling in different physiological and pathological conditions. Granulysin (GNLY) represents a powerful effector protein for all these subpopulations.

Objectives: To analyze the cytolytic compartment of inflammatory cells in patients with PAH and PVOD.

Methods: The overall functional status of the cytolytic compartment was studied through epigenetic analysis of the GNLY gene in explanted lungs and in peripheral blood mononuclear cells. Flow cytometry technology allowed analysis of specific circulating cytolytic cells and GNLY contents. A GNLY-specific ELISA allowed measurement of GNLY serum concentrations.

Measurements and Main Results: A decrease in GNLY demethylation in the gDNA extracted from peripheral blood mononuclear cells and explanted lungs was found specifically in PVOD but not in PAH. This was associated with a decrease in populations and subpopulations of CTL and NKT and an increase of NK populations. Despite the reduced granulysin-containing cells in patients with PVOD, GNLY serum levels were higher, suggesting these cells were wasting their content. Furthermore, the increase of GNLY concentration in the serum of PVOD was significantly higher than in patients with PAH.

Conclusions: PVOD is characterized by alterations of circulating cytotoxic cell subpopulations and by epigenetic dysregulation within the GNLY gene. Our findings may be helpful in the quest to develop needed diagnostic tools, including flow cytometry analyses, to screen for suspected PVOD in patients with pulmonary hypertension.

Scientific Knowledge on the Subject

Pulmonary veno-occlusive disease (PVOD) shares many similarities with idiopathic or heritable pulmonary arterial hypertension (PAH), from risk factors to clinical or hemodynamic presentation, which can easily lead to misdiagnosis between these two conditions. The need to establish a correct and early diagnosis of PVOD is justified by the worse prognosis of these patients and by their risk of developing severe pulmonary edema with specific PAH therapy. We hypothesized that the cytolytic fraction/subpopulation of inflammatory cells are differently regulated in patients with PAH and PVOD.

What This Study Adds to the Field

PVOD is characterized by epigenetic regulation of the granulysin (GNLY) gene in explanted lungs and in peripheral blood mononuclear cells, as well as by alterations of circulating cytotoxic cell subpopulations and their GNLY contents. Our results suggest that differences in immune regulation of these cells may contribute to the pathophysiology of PVOD and PAH. The fact that granulysin and cytolytic cells are dysregulated may allow future development of biological diagnostic tools to aid the difficult diagnosis of patients with PVOD.

Pulmonary arterial hypertension (PAH) belongs to a heterogeneous group of progressive diseases characterized by an increase in resting mean pulmonary arterial pressure above 25 mm Hg (1, 2). PAH can be idiopathic, heritable, or associated with different conditions or drug exposure (3). Pulmonary veno-occlusive disease (PVOD) is a rare form of pulmonary hypertension (PH) belonging to group 1′ of the Dana Point classification; it is defined by a predominant involvement of small pulmonary veins and characterized by poor prognosis, poor response to specific PAH therapy, and the possibility of developing severe pulmonary edema with these therapies (4, 5).

Increasing evidence suggests that inflammatory mechanisms play a role in PAH pathogenesis, including lymphocyte infiltration (613). Approximately 10% of leukocytes in the naive lung are natural killer (NK) cells (14, 15). Lung homeostasis of these cells is important. By way of example, there is a profound deficiency in lung NK cells in a severe model of bleomycin-induced lung fibrosis in mice (16). CD8+ cytotoxic T (CTL), NK, and natural killer T (NKT) cells are classically known to be the cytolytic arm of the cellular immune response. NK cells can be divided into several subsets according to function and phenotype. CD56dimCD16+, highly differentiated NK cells, show strong cell-mediated cytotoxicity as well as antibody-dependent cell-mediated cytotoxicity (17), whereas CD56highCD16, more immature NK cells, are potent producers of cytokines (17). Further stratification of NK cells according to CD8α/α can be of significance, as the CD56+CD8+ subset was shown to have a stronger cytotoxic function compared with its CD8 counterpart (18). NKT cells represent a subpopulation of T cells (expression of CD3 and variable expression of CD8 and CD4) that possesses properties of NK cells (expression of CD56) (19). Last, CTLs have the capacity to directly recognize and kill antigen-expressing cells (20). Granulysin (GNLY) is present in cytolytic granules of CTL, NK, and NKT cells and is broadly cytolytic against tumors and microbes. GNLY is implicated in a myriad of diseases, including infection, cancer, transplantation, autoimmunity, and skin and reproductive diseases, and may give rise to new diagnostic and therapeutic options in a wide variety of diseases (21, 22).

CTL, NK, and NKT populations may play a role in vascular remodeling (e.g., in remodeling of fetal blood supply during pregnancy [23] or in arteriosclerosis [24]). This led us to hypothesize that pulmonary vascular remodeling responsible for PAH or PVOD might be linked to alterations of the circulating and pulmonary cytolytic compartment of the inflammatory cells. We aimed to analyze the CTL, NK, and NKT populations in patients with idiopathic or heritable PAH and PVOD. We measured the prevalence of these cells in circulatory and pulmonary compartments and assessed their function through their capacity to produce GNLY.

Collections of Lung Specimens and Blood Samples

Patients were older than 18 years, and patients with PAH had a diagnosis confirmed by right-heart catheterization. The diagnosis of PVOD was considered highly probable if patients fulfilled the following criteria: precapillary pulmonary hypertension, presence of two or more radiological PVOD-like abnormalities on high-resolution computed tomography scan of the chest (comprising lymph node enlargement, centrilobular ground-glass opacities, and septal lines), low carbon monoxide diffusing capacity (DlCO), or occult alveolar hemorrhage. Diagnosis of PVOD was considered as confirmed when histological proof was available or when patients with signs of highly probable disease developed pulmonary edema after initiation of specific PAH therapy. For patients undergoing lung transplantation, diagnostic features of PVOD included the presence of occluding venous and venular fibrosis in a large proportion of postcapillary vessels (>50%) and the presence of multiple capillary hemangiomatosis-like foci, together with indirect signs, such as accumulation of intraalveolar siderin-laden macrophages and absence of plexiform lesions (Figures 1A–1C). In contrast, precapillary PAH (idiopathic or heritable) was histologically confirmed when plexiform lesions were present and in the absence of capillary hemangiomatosis-like foci. Only minor venous remodeling was accepted in PAH cases (Figures 1D–1F).

Blood samples were collected in patients with PAH and PVOD during follow-up and in control subjects. Characteristics at diagnosis and follow-up were stored in the Registry of the French Network of Pulmonary Hypertension set up in agreement with French bioethics laws (French Commission Nationale de l’Informatique et des Libertés) (25). Patient characteristics are described in Tables E1 to E4 of the online supplement.

Lung specimens including idiopathic or heritable PAH and confirmed PVOD were collected at the time of lung transplantation, and control specimens were obtained during lobectomy or pneumonectomy for localized lung cancer. Tissues were snap-frozen as previously described (26). Patients studied were part of the French Network on Pulmonary Hypertension, a program approved by our institutional Ethics Committee, and had given written informed consent (Protocol N°CO-08-003, ID RCB: 2008-A00485-50, approved on June 18, 2008). Patients with PVOD and idiopathic and heritable PAH tested for mutations in PAH-predisposing genes, underwent genetic counseling, and signed written informed consent.

The results obtained from patients with idiopathic and heritable PAH during this study proved to be statistically indistinguishable. We therefore considered both groups as a single PAH group in data analysis.

Epigenetic Assays for Immune Cell Monitoring In Lung Tissue and Blood of Patients with PAH and PVOD

Epigenetic assays were performed by Epiontis (Berlin, Germany). After bisulfite conversion, genomic DNA was subjected to quantitative real-time PCR analysis for the determination of numbers and ratios of regulatory T cells (Treg) in the samples, according to Wieczorek and colleagues (27). In brief, stable expression of FOXP3 in naturally occurring Treg requires epigenetic (i.e., DNA methylation–based) regulation. Demethylation at a highly conserved region within the human FOXP3 gene (Treg-specific demethylated region [TSDR]) is restricted to Treg. In addition to the high specificity for Treg, FOXP3 TSDR demethylation occurred only in natural Treg but not in recently activated effector T cells transiently expressing FOXP3. It is associated with stable FOXP3 expression on Treg expansion in vitro. Hence, epigenetic modifications in the FOXP3 TSDR serve as a valuable marker for the identification of T cells with a stable Treg phenotype. We recently showed by flow cytometry that their number was not affected in PAH, although their function was impaired (28). Thus, Treg quantification was used as a control of the assay. The same strategy and method was used for the GNLY assay. Demethylation within the human GNLY gene is restricted to GNLY-expressing cytolytic cells. The assay consists of two PCR systems: one system is specific for the demethylated version and the other for the methylated version of the granulysin gene. The percentage of granulysin-positive cells in a sample was calculated as follows: % granulysin-positive cells = [demethylated GNLY gene copies]/[ demethylated + methylated GNLY gene copies] × 100. In this equation, a factor of 100 is used to translate the result to percent.

Flow Cytometry Assays for Circulating Immune Cell Monitoring

Peripheral blood mononuclear cells (PBMC) from patients with PAH and PVOD and control subjects (matched for age and sex) were stained for surface antigens with anti-CD8 VioBlue (BW135/80 mAb, Miltenyi, Paris, France), anti-CD3 AF488 (UCHT1 mAb, BD Pharmingen, Le Pont-De-Claix, France), anti-CD16 PerCP (3G8 mAb, Biolegend, San Diego, CA), and anti-CD56 AF674 (HCD56 mAb, Biolegend). Then cells were fixed and permeabilized with fixation and permeabilization wash buffer from Biolegend for intracellular staining with anti-granulyzin PE (DH2 mAb, Biolegend). Flow cytometry was performed on a MACSQuant analyzer (Miltenyi). Gating conditions were set and normalized against isotype- and fluorophore-matched nonimmune IgGs. The data were analyzed using FlowJo software program (Tree Star, Inc, Ashland, OR). CTL, NK, and NKT populations and their GNLY content were analyzed as shown in Figure 2.

Granulysin-Specific ELISA

Serum concentration of GNLY was measured using ELISA developed according to Ogawa and colleagues (29) (see supplemental methods).

Statistical Evaluation

Quantitative variables are presented as means ± SEM unless otherwise stated. We used Student t tests or Mann-Whitney tests to compare groups, depending on the normal or nonnormal nature of the distribution. We used the Kruskal-Wallis test to compare more than two groups. Receiver operator curve (ROC) analyses were performed with the Graphpad Prism 5 Software (GraphPad Software, San Diego, CA). P values less than 0.05 were considered to indicate statistical significance.

GNLY Gene Is Hypermethylated in gDNA from Lung and Blood of Patients with PVOD Compared with Patients with PAH

To analyze the potential alterations of GNLY expression in the lung and blood of patients with PAH and PVOD, we chose a gene methylation approach. This epigenetic assay advantageously uses DNA as analyte, a stable substrate with predetermined copy numbers per cell. No detection thresholds arbitrarily segregate positive from negative cells. The percentage of GNLY demethylation was significantly and specifically decreased in the gDNA extracted from lung of patients with PVOD (0.87 ± 0.16%, n = 11) as compared with patients with PAH (2.19 ± 0.25%, n = 14, P < 0.01) and control subjects (1.77 ± 0.16%, n = 12, P < 0.01). The percentage of GNLY demethylation in the blood gDNA was significantly decreased in patients with PVOD (3.76 ± 0.45%, n = 10) as compared with control subjects (6.38 ± 0.62%, n = 9, P < 0.05). Demethylation in the blood gDNA of patients with PAH (4.86 ± 1.08%, n = 10) was not different from patients with PVOD or control subjects (Figure 3). As a control gene we analyzed FOXP3 demethylation, which was unchanged between control subjects and subgroups of PAH (data not shown), and expected from previous studies in patients with PAH (28).

Alterations of Circulating CTL, NK, and NKT Subpopulations, Together with Decrease in GNLY Contents, Occur Specifically in Patients with PVOD

We analyzed the proportions of CTL, NK, and NKT subpopulations by flow cytometry, together with GNLY intracellular contents in peripheral blood lymphocytes from patients with PAH and PVOD compared with control subjects.

As we have previously shown (11), we found a decrease in the overall circulating T lymphocyte compartment (CD3+CD56 cells) in patients with PAH (n = 18) as compared with control subjects (n = 9). Interestingly, we also observed a decrease of circulating T lymphocytes in patients with PVOD (n = 8) associated with alterations of circulating cytotoxic cell subpopulations (Figure 4A, Table 1). A decrease of CTL and NKT populations, together with an increase of the CD56dimCD16+ highly differentiated NK population, were specifically observed in patients with PVOD (Figures 4B, 4D, 4F, and Table 1). Moreover, the proportions of GLNY-containing cells in these circulating cytotoxic populations were significantly decreased in patients with PVOD (Figures 4C, 4E, 4G, and Table 1). The immature CD56brightCD16-NK population and its GNLY content were unchanged in patients compared with control subjects (Figures 4H, 4I, and Table 1). Variations in CD8, CD8dim, and CD8+ (Figure 2) subpopulations and GLNY contents in these populations mirrored exactly the alterations of their mother NK and NKT populations (data not shown). ROC analysis suggested that granulysin expression in CTL measured by flow cytometry (Figure 4C) may be predictive of PVOD. ROC analysis showed that a cut-off of 1.01% allows discrimination between patients with PVOD and PAH with a sensitivity of 87.5% and a specificity of 77.8% (area under the ROC curve 0.85 and likelihood ratio 3.94) (see Figure E1 in the online data supplement).

TABLE 1. Flow cytometry analysis of the circulating cytolytic cells in blood samples from control subjects and patients with pulmonary veno-occlusive disease and pulmonary arterial hypertension

Control (n = 9)PVOD (n = 8)PAH (n = 18)
CD3+CD56 LT (% L)66 ± 2.648 ± 4.9*,57 ± 2.5*,
CD3+CD56CD8+ CTL (% L)20 ± 1.412 ± 2.3*,21 ± 2.5,§
CTL GNLY+ (% CTL)2 ± 0.60.6 ± 0.1*,3 ± 0.7,§
CD3+CD56+ NKT (% L)4 ± 0.91 ± 0.3*,6 ± 1.4,§
NKT GNLY+ (% NKT)21 ± 4.65 ± 1.7*,19 ± 4.2,§
CD3CD56dimCD16+ NK (% L)10 ± 1.724 ± 4.0*,11 ± 1.5,§
NK CD56dim GNLY+ (% NK CD56dim)54 ± 4.033 ± 6.9*,47 ± 3.9
CD3CD56brightCD16 NK (% L)0.47 ± 0.050.47 ± 0.080.46 ± 0.05
NK CD56bright GNLY+ (% NK CD56bright)58 ± 3.646 ± 7.256 ± 4.1

Definition of abbreviations: CTL = cytotoxic T lymphocytes; GNLY = granulysin; L = lymphocytes; LT = T lymphocytes; NK = natural killer cells; NKT = natural killer T cells; PAH = pulmonary arterial hypertension; PVOD = pulmonary veno-occlusive disease.

*Compared to control.

P < 0.01.

P < 0.05.

§Compared to PVOD.

Increased Serum GNLY Concentration in Patients with PVOD and PAH

After analysis of GNLY epigenetic status at the nucleic level and its intracellular protein level, we analyzed the free extracellular protein level in the sera of patients with PAH and PVOD compared with control subjects. Serum GNLY concentrations were significantly increased in patients with PVOD (5.2 ± 2.0 ng/ml, n = 15, P < 0.0001) and PAH (3.7 ± 0.25 ng/ml, n = 38, P < 0.01) as compared with control subjects (2.6 ± 0.16 ng/ml, n = 18). Patients with PVOD had higher circulating levels of GNLY than patients with PAH (P < 0.01) (Figure 5). There was no significant association between serum GNLY concentration and clinical parameters: body mass index, New York Heart Association classification, 6-minute walk distance, mean pulmonary artery pressure, cardiac index, cardiac output, pulmonary capillary pressure, pulmonary vascular resistance, specific PAH therapies (epoprostenol, endothelin receptors antagonists, or phosphodiesterase type 5 inhibitors), or tobacco exposure (data not shown).

The main finding of this study is that epigenetic regulation of the GNLY gene in explanted lungs and in PBMC, as well as alterations of circulating cytotoxic subpopulations and their GNLY contents, may allow discrimination between patients with PVOD and PAH. A decrease in GNLY demethylation in the gDNA extracted from blood cells and explanted lungs was found specifically in PVOD but not in idiopathic or heritable PAH. This alteration of GNLY demethylation in patients with PVOD was associated with a decrease in populations and subpopulations of CTL and NKT cells, with a possibly counterbalancing increase of NK populations. Moreover, all circulating cytotoxic populations without exception displayed a waste of their protein GLNY content detected by flow cytometry in patients with PVOD, highlighting a loss of cytolytic function of these cells. Finally, there was an increase of GNLY concentration in the serum of patients with PVOD and PAH, the latter with a significantly lower level as compared with patients with PVOD. This observation could be due to an increased degranulation of GNLY and maybe to an intracellular content release after cell death. Further investigations are needed to address this issue.

PVOD shares many similarities with idiopathic or heritable PAH, from risk factors to clinical or hemodynamic presentation, which can easily lead to misdiagnoses between these two conditions (4, 5). Definitive differentiation can only be achieved by histology, wherein lungs of patients with PVOD display histological hallmarks: a widespread fibrous intimal thickening that predominantly involves the pulmonary veins and venules associated with focal multiplication of alveolar capillaries. In contrast, idiopathic PAH is characterized by a major remodeling of small precapillary pulmonary arteries and common frequent plexiform (30). Histological proof is required for a definitive diagnosis of PVOD, but because surgical lung biopsy is associated with a significant mortality risk in these patients, the procedure is not recommended. Distinguishing PVOD from PAH on clinical grounds or hemodynamic evaluation is generally more than challenging and should be undertaken in specialized centers. However, the importance of establishing a correct and early diagnosis of PVOD is justified by the worse prognosis of these patients and by their risk of developing severe pulmonary edema with specific PAH therapy (4). Noninvasive tests may be helpful, and the probable diagnosis is usually based on an integrated assessment that comprises high-resolution computed tomography scanning, pulmonary function testing, and bronchoalveolar lavage (4). Currently, no blood tests are included in this noninvasive approach to screening patients with PVOD. The results of our study show specific alterations in the epigenetic regulation of the GNLY gene and in circulating cytotoxic subpopulations of patients with PVOD and thus may offer a new perspective on the development of biological diagnostic tools for this rare form of pulmonary hypertension (e.g., GNLY content in CTL measured by flow cytometry or serum concentration levels of the GNLY protein).

The impaired number and GLNY-mediated functions of the CTL, NK, and NKT cells in the setting of PVOD cells may suggest exhaustion of these cells, possibly as a result of overstimulation by chronic exposure to inflammatory mediators (31). This hypothesis is consistent with the observation of Raychaudhuri and colleagues (32) showing that circulating monocytes in PAH have decreased activation and are hyporesponsive to lipopolysaccharide stimulation, despite obvious signs of the proinflammatory background in PAH (8, 33). We have recently shown that Treg function is inhibited in a leptin-dependent manner in idiopathic PAH and scleroderma-associated PAH (28) and that increased levels of circulating leptin are endothelial cell (EC) derived. Treg are T lymphocytes known to dampen autoreactive responses and are able to delay the onset and progression of autoimmune disorders. Interestingly, Wrann and colleagues (34) showed that long-term leptin exposure inhibits important NK cell functions, including cytotoxicity, production of IFN-γ, and cell proliferation. This suggests a possible role for leptin in PAH and especially in PVOD pathogenesis, via the modulation of Treg and NK cell functions to promote possible autoimmune signaling. Circulating autoantibodies (Ab), in particular anti-EC and anti-fibroblast Ab, have been reported in 10 to 40% of patients with idiopathic PAH (35, 36), suggesting a possible role for autoimmunity in the pathogenesis of pulmonary vascular lesions. Besides, invariant NKT (iNKT) cells inhibit autoreactive B cells in a contact- and CD1d-dependent manner (37). In rheumatoid arthritis, iNKT cells suppress IgG anti-DNA Ab and rheumatoid factor production and reduce IL-10–secreting B cells in a contact-dependent manner (37). van der Vliet and colleagues (38) recently showed that circulating iNKT cell numbers are decreased in a broad variety of disorders with immune-mediated pathogenesis, affecting the skin, bowel, central nervous system, and joints, regardless of disease duration or activity, and that defective NKT cell populations are associated with autoreactive tissue damage rather than with the propensity to develop autoimmune disease. This may link the loss of function of the NKT compartment of patients with PVOD to a possible autoreactive vascular damage and remodeling in the setting of autoimmune mechanisms against vascular wall component in PAH and especially in PVOD. This is especially relevant in the context of systemic sclerosis, a heterogeneous connective tissue disease characterized by dysfunction of the endothelium, dysregulation of fibroblasts resulting in excessive production of collagen, and abnormalities of the immune system. Prospective studies have shown that pulmonary vascular involvement complicates systemic sclerosis with a prevalence of 8 to 12% and is the major cause of death in these patients (39). A study of post mortem lung histology of patients with connective tissue disease–associated PAH (40) demonstrated that 75% had PVOD-like involvement of the postcapillary vascular bed (40). Indeed, in a larger series, we recently showed that more than 60% of patients with systemic sclerosis–associated pulmonary hypertension had radiological signs of PVOD (41). Moreover, in this series, radiological signs of PVOD were associated with occurrence of pulmonary edema after initiation of specific PAH therapy and with more rapid progression to death (41). All this evidence confirms the high prevalence of pulmonary venous involvement in the context of autoimmune disorders, and recently PVOD has even been considered as “the bête noire of pulmonary hypertension in connective tissue diseases” (42). Thus, our findings on dysregulation of granulysin and cytolytic cells suggest that immune dysregulation may participate in the pathophysiology of PVOD even in the absence of known connective tissue disease.

GNLY expression has been widely correlated with a good outcome in a variety of cancers. Low intracellular expression of GNLY in NK cells, but not perforin (a well known cytolytic protein), correlates with progression of cancer (43). Flow cytometric analysis showed high levels of both perforin and GNLY in the NK cells in normal healthy control subjects. Tumor-free patients expressed GNLY at levels similar to control subjects, whereas cancer patients exhibited significantly decreased GNLY levels. In contrast, perforin levels were similar in all groups. This study suggests that impaired expression of GNLY in circulating NK cells correlates with tumor progression (21, 43). Examination of NK cell functions showed that supernatants from ECs isolated from normal lungs were potent activators of NK cells, as indicated by their secretion of tumor necrosis factor-α and IFN-γ, whereas ECs isolated from tumor-bearing lungs had a significantly diminished capacity to activate NK cells (44). Indeed, tumors skew ECs to disrupt NK-cell, T-cell, and macrophage functions (45). This suggests that dysfunctional pulmonary ECs may interfere with the function of the inflammatory cytolytic cells of patients with PVOD in a similar fashion as occurs in cancer. Indeed, the pathogenic significance of pulmonary capillary hemangiomatosis (PCH)-like foci in PVOD has been subject of diverse speculation. When considering publications on PCH, it had been primarily speculated that the foci of proliferating capillaries corresponded to some form of angiomatous growth with signs of atypia (46) or to a nodular, locally aggressive but benign or low-grade vascular neoplasm of the lung (47, 48). Also, a recent histological analysis of lungs from a patient with PCH revealed an immunohistochemical phenotype of capillary proliferation, comparable to plexiform lesions with increased expression of markers associated with cellular proliferation and angiogenesis, such as vascular endothelial growth factor and MiB-1 (49). However, a large retrospective analysis by Lantuéjoul and colleagues from 2007 has suggested that a majority of PCH cases probably represent a secondary angioproliferative process caused by postcapillary obstruction (e.g., PVOD) (50).

Last, our detection of an increase in serum GNLY concentration in patients with PAH indicates that some cytotoxic cell-related mechanisms also occur in idiopathic and heritable PAH but at a significantly lower level than in PVOD. This is in accordance with a recent study from Ormiston and colleagues (51) showing an expansion of the functionally defective CD56/CD16+ NK subset that was not observed in patients with chronic thromboembolic pulmonary hypertension.

In conclusion, we have demonstrated that PVOD is characterized by alterations of circulating cytotoxic subpopulations and by epigenetic regulation of the GNLY, suggesting a key role for this gene. Patients with PVOD have a poor prognosis and are susceptible to the development pulmonary edema with specific PAH therapy. Hitherto, the diagnosis of “suspected” PVOD is based on a noninvasive approach but remains a clinical challenge. Our findings may be helpful to develop simple biological diagnostic tools to screen for PVOD. Prospective studies are needed to evaluate the potential interest of this measurement in the diagnosis of patients with PVOD. Indeed, our study uncovers a new pathway for future investigations on the cross-talk between inflammatory cytotoxic cells and pulmonary ECs in pulmonary hypertension. GNLY-expressing cells may represent a potential target for innovative PAH therapy.

The authors thank Pr Phillipe Dartevelle (Marie Lannelongue Hospital, Université Paris-Sud 11, France) for facilitating access to explanted lungs from patients with PAH and PVOD and lung control biopsies, and Dr Laura Price (Royal Brompton Hospital, United Kingdom), who provided assistance with English language editing.

1. Galie N, Hoeper MM, Humbert M, Torbicki A, Vachiery JL, Barbera JA, Beghetti M, Corris P, Gaine S, Gibbs JS, et al.. Guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Respir J 2009;34:12191263.
2. Humbert M, Sitbon O, Simonneau G. Treatment of pulmonary arterial hypertension. N Engl J Med 2004;351:14251436.
3. Simonneau G, Robbins IM, Beghetti M, Channick RN, Delcroix M, Denton CP, Elliott CG, Gaine SP, Gladwin MT, Jing ZC, et al.. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol 2009;54:S43S54.
4. Montani D, Price LC, Dorfmuller P, Achouh L, Jais X, Yaici A, Sitbon O, Musset D, Simonneau G, Humbert M. Pulmonary veno-occlusive disease. Eur Respir J 2009;33:189200.
5. Montani D, Achouh L, Dorfmuller P, Le Pavec J, Sztrymf B, Tcherakian C, Rabiller A, Haque R, Sitbon O, Jais X, et al.. Pulmonary veno-occlusive disease: clinical, functional, radiologic, and hemodynamic characteristics and outcome of 24 cases confirmed by histology. Medicine (Baltimore) 2008;87:220233.
6. Balabanian K, Foussat A, Dorfmuller P, Durand-Gasselin I, Capel F, Bouchet-Delbos L, Portier A, Marfaing-Koka A, Krzysiek R, Rimaniol AC, et al.. CX(3)C chemokine fractalkine in pulmonary arterial hypertension. Am J Respir Crit Care Med 2002;165:14191425.
7. Dorfmuller P, Perros F, Balabanian K, Humbert M. Inflammation in pulmonary arterial hypertension. Eur Respir J 2003;22:358363.
8. Humbert M, Monti G, Brenot F, Sitbon O, Portier A, Grangeot-Keros L, Duroux P, Galanaud P, Simonneau G, Emilie D. Increased interleukin-1 and interleukin-6 serum concentrations in severe primary pulmonary hypertension. Am J Respir Crit Care Med 1995;151:16281631.
9. Jais X, Launay D, Yaici A, Le Pavec J, Tcherakian C, Sitbon O, Simonneau G, Humbert M. Immunosuppressive therapy in lupus- and mixed connective tissue disease-associated pulmonary arterial hypertension: a retrospective analysis of twenty-three cases. Arthritis Rheum 2008;58:521531.
10. Jouve P, Humbert M, Chauveheid MP, Jais X, Papo T. Poems syndrome-related pulmonary hypertension is steroid-responsive. Respir Med 2007;101:353355.
11. Perros F, Dorfmuller P, Montani D, Hammad H, Waelput W, Girerd B, Raymond N, Mercier O, Mussot S, Cohen-Kaminsky S, et al.. Pulmonary lymphoid neogenesis in idiopathic pulmonary arterial hypertension. Am J Respir Crit Care Med 2012;185:311321.
12. Perros F, Dorfmuller P, Souza R, Durand-Gasselin I, Godot V, Capel F, Adnot S, Eddahibi S, Mazmanian M, Fadel E, et al.. Fractalkine-induced smooth muscle cell proliferation in pulmonary hypertension. Eur Respir J 2007;29:937943.
13. Sanchez O, Marcos E, Perros F, Fadel E, Tu L, Humbert M, Dartevelle P, Simonneau G, Adnot S, Eddahibi S. Role of endothelium-derived cc chemokine ligand 2 in idiopathic pulmonary arterial hypertension. Am J Respir Crit Care Med 2007;176:10411047.
14. Klemm A, Tschernig T, Krug N, Pabst R. Lymphocyte subsets in distinct lung compartments show a different ability to produce interferon-gamma (IFN-gamma) during a pulmonary immune response. Clin Exp Immunol 1998;113:252257.
15. Davis GS, Pfeiffer LM, Hemenway DR. Interferon-gamma production by specific lung lymphocyte phenotypes in silicosis in mice. Am J Respir Cell Mol Biol 2000;22:491501.
16. Jiang D, Liang J, Hodge J, Lu B, Zhu Z, Yu S, Fan J, Gao Y, Yin Z, Homer R, et al.. Regulation of pulmonary fibrosis by chemokine receptor CXCR3. J Clin Invest 2004;114:291299.
17. Deniz G, Akdis M, Aktas E, Blaser K, Akdis CA. Human NK1 and NK2 subsets determined by purification of IFN-gamma-secreting and IFN-gamma-nonsecreting NK cells. Eur J Immunol 2002;32:879884.
18. Addison EG, North J, Bakhsh I, Marden C, Haq S, Al-Sarraj S, Malayeri R, Wickremasinghe RG, Davies JK, Lowdell MW. Ligation of CD8alpha on human natural killer cells prevents activation-induced apoptosis and enhances cytolytic activity. Immunology 2005;116:354361.
19. Berzins SP, Smyth MJ, Baxter AG. Presumed guilty: natural killer T cell defects and human disease. Nat Rev Immunol 2011;11:131142.
20. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 2012;12:252264.
21. Krensky AM, Clayberger C. Biology and clinical relevance of granulysin. Tissue Antigens 2009;73:193198.
22. Gansert JL, Kiessler V, Engele M, Wittke F, Rollinghoff M, Krensky AM, Porcelli SA, Modlin RL, Stenger S. Human NKT cells express granulysin and exhibit antimycobacterial activity. J Immunol 2003;170:31543161.
23. Moffett-King A. Natural killer cells and pregnancy. Nat Rev Immunol 2002;2:656663.
24. Tupin E, Nicoletti A, Elhage R, Rudling M, Ljunggren HG, Hansson GK, Berne GP. CD1D-dependent activation of NKT cells aggravates atherosclerosis. J Exp Med 2004;199:417422.
25. Humbert M, Sitbon O, Chaouat A, Bertocchi M, Habib G, Gressin V, Yaici A, Weitzenblum E, Cordier JF, Chabot F, et al.. Pulmonary arterial hypertension in France: results from a national registry. Am J Respir Crit Care Med 2006;173:10231030.
26. Perros F, Montani D, Dorfmuller P, Durand-Gasselin I, Tcherakian C, Le Pavec J, Mazmanian M, Fadel E, Mussot S, Mercier O, et al.. Platelet-derived growth factor expression and function in idiopathic pulmonary arterial hypertension. Am J Respir Crit Care Med 2008;178:8188.
27. Wieczorek G, Asemissen A, Model F, Turbachova I, Floess S, Liebenberg V, Baron U, Stauch D, Kotsch K, Pratschke J, et al.. Quantitative DNA methylation analysis of FOXP3 as a new method for counting regulatory T cells in peripheral blood and solid tissue. Cancer Res 2009;69:599608.
28. Huertas A, Tu L, Gambaryan N, Girerd B, Perros F, Montani D, Fabre D, Fadel E, Eddahibi S, Cohen-Kaminsky S, et al.. Leptin and regulatory T lymphocytes in idiopathic pulmonary arterial hypertension. Eur Respir J 2012;40:895904.
29. Ogawa K, Takamori Y, Suzuki K, Nagasawa M, Takano S, Kasahara Y, Nakamura Y, Kondo S, Sugamura K, Nakamura M, et al.. Granulysin in human serum as a marker of cell-mediated immunity. Eur J Immunol 2003;33:19251933.
30. Pietra GG, Capron F, Stewart S, Leone O, Humbert M, Robbins IM, Reid LM, Tuder RM. Pathologic assessment of vasculopathies in pulmonary hypertension. J Am Coll Cardiol 2004;43:25S32S.
31. Oppenheim DE, Roberts SJ, Clarke SL, Filler R, Lewis JM, Tigelaar RE, Girardi M, Hayday AC. Sustained localized expression of ligand for the activating NKG2D receptor impairs natural cytotoxicity in vivo and reduces tumor immunosurveillance. Nat Immunol 2005;6:928937.
32. Raychaudhuri B, Bonfield TL, Malur A, Hague K, Kavuru MS, Arroliga AC, Thomassen MJ. Circulating monocytes from patients with primary pulmonary hypertension are hyporesponsive. Clin Immunol 2002;104:191198.
33. Soon E, Holmes AM, Treacy CM, Doughty NJ, Southgate L, Machado RD, Trembath RC, Jennings S, Barker L, Nicklin P, et al.. Elevated levels of inflammatory cytokines predict survival in idiopathic and familial pulmonary arterial hypertension. Circulation 2010;122:920927.
34. Wrann CD, Laue T, Hubner L, Kuhlmann S, Jacobs R, Goudeva L, Nave H. Short-term and long-term leptin exposure differentially affect human natural killer cell immune functions. Am J Physiol Endocrinol Metab 2012;302:E108E116.
35. Tamby MC, Chanseaud Y, Humbert M, Fermanian J, Guilpain P, Garcia-de-la-Pena-Lefebvre P, Brunet S, Servettaz A, Weill B, Simonneau G, et al.. Anti-endothelial cell antibodies in idiopathic and systemic sclerosis associated pulmonary arterial hypertension. Thorax 2005;60:765772.
36. Terrier B, Tamby MC, Camoin L, Guilpain P, Broussard C, Bussone G, Yaici A, Hotellier F, Simonneau G, Guillevin L, et al.. Identification of target antigens of antifibroblast antibodies in pulmonary arterial hypertension. Am J Respir Crit Care Med 2008;177:11281134.
37. Yang JQ, Wen X, Kim PJ, Singh RR. Invariant NKT cells inhibit autoreactive B cells in a contact- and CD1D-dependent manner. J Immunol 2011;186:15121520.
38. van der Vliet HJ, von Blomberg BM, Nishi N, Reijm M, Voskuyl AE, van Bodegraven AA, Polman CH, Rustemeyer T, Lips P, van den Eertwegh AJ, et al.. Circulating V(alpha24+) Vbeta11+ NKT cell numbers are decreased in a wide variety of diseases that are characterized by autoreactive tissue damage. Clin Immunol 2001;100:144148.
39. Steen VD, Medsger TA. Changes in causes of death in systemic sclerosis, 1972–2002. Ann Rheum Dis 2007;66:940944.
40. Dorfmüller P, Humbert M, Perros F, Sanchez O, Simonneau G, Müller KM, Capron F. Fibrous remodeling of the pulmonary venous system in pulmonary arterial hypertension associated with connective tissue diseases. Hum Pathol 2007;38:893902.
41. Gunther S, Jais X, Maitre S, Berezne A, Dorfmuller P, Seferian A, Savale L, Mercier O, Fadel E, Sitbon O, et al.. Computed tomography findings of pulmonary veno-occlusive disease in scleroderma patients presenting with precapillary pulmonary hypertension. Arthritis Rheum 2012;64:29953005.
42. O’Callaghan DS, Dorfmuller P, Jais X, Mouthon L, Sitbon O, Simonneau G, Humbert M, Montani D. Pulmonary veno-occlusive disease: the bête noire of pulmonary hypertension in connective tissue diseases? Presse Med 2011;40:e65e78.
43. Kishi A, Takamori Y, Ogawa K, Takano S, Tomita S, Tanigawa M, Niman M, Kishida T, Fujita S. Differential expression of granulysin and perforin by NK cells in cancer patients and correlation of impaired granulysin expression with progression of cancer. Cancer Immunol Immunother 2002;50:604614.
44. Mulligan JK, Young MR. Tumors induce the formation of suppressor endothelial cells in vivo. Cancer Immunol Immunother 2010;59:267277.
45. Mulligan JK, Lathers DM, Young MR. Tumors skew endothelial cells to disrupt NK cell, T-cell and macrophage functions. Cancer Immunol Immunother 2008;57:951961.
46. Wagenvoort CA, Beetstra A, Spijker J. Capillary haemangiomatosis of the lungs. Histopathology 1978;2:401406.
47. Eltorky MA, Headley AS, Winer-Muram H, Garrett HE, Griffin JP. Pulmonary capillary hemangiomatosis: a clinicopathologic review. Ann Thorac Surg 1994;57:772776.
48. Tron V, Magee F, Wright JL, Colby T, Churg A. Pulmonary capillary hemangiomatosis. Hum Pathol 1986;17:11441150.
49. Sullivan A, Chmura K, Cool CD, Keith R, Schwartz GG, Chan ED. Pulmonary capillary hemangiomatosis: an immunohistochemical analysis of vascular remodeling. Eur J Med Res 2006;11:187193.
50. Lantuéjoul S, Sheppard MN, Corrin B, Burke MM, Nicholson AG. Pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis: a clinicopathologic study of 35 cases. Am J Surg Pathol 2006;30:850857.
51. Ormiston ML, Chang C, Long LL, Soon E, Jones D, Machado R, Treacy C, Toshner MR, Campbell K, Riding A, et al.. Impaired natural killer cell phenotype and function in idiopathic and heritable pulmonary arterial hypertension. Circulation 2012;126:10991109.
Correspondence and requests for reprints should be addressed to Frédéric Perros, Ph.D., INSERM U999, Centre Chirurgical Marie Lannelongue, 133, Avenue de la Resistance, F-92350 Le Plessis Robinson, France. E-mail:

Supported by Fondation pour la Recherche Médicale (FRM), team FRM 2010, grant DEQ20100318257 (F.P. and the team from the INSERM U999 unit; and the Association HTAPFrance (P.D. and D.M.).

Author Contributions: Conception and design: F.P., S.C.-K., G.S., M.H., P.D., and D.M. Acquisition of data: F.P., N.G., B.G., N.R., I.K., O.M., and E.F. Analysis and interpretation: F.P., S.C.-K., N.G., B.G., P.D., and D.M. Drafting the manuscript for important intellectual content: F.P., S.C.-K., N.G., B.G., A.H., G.S., M.H., P.D., and D.M.

This article has an online supplement, which is accessible from this issue’s table of contents at

Originally Published in Press as DOI: 10.1164/rccm.201208-1364OC on December 6, 2012

Author disclosures are available with the text of this article at


No related items
American Journal of Respiratory and Critical Care Medicine

Click to see any corrections or updates and to confirm this is the authentic version of record