The presence of a matrix-derived neutrophil chemoattractant in bronchiolitis obliterans syndrome after lung transplantation

Abstract

Lung transplantation is a therapeutic modality frequently used in end-stage lung disease. Unfortunately, lung transplant recipients have poor clinical outcomes, often due to the development of bronchiolitis obliterans syndrome (BOS). This process is often characterized by the pathologic findings of obliterative bronchiolitis: neutrophil influx and extracellular matrix remodeling leading to luminal obstruction and airway inflammation. The molecular mechanisms underlying BOS are poorly understood and disease-specific biomarkers are lacking. We report that in addition to increased levels of IL-8, the level of the neutrophil chemoattractant proline-glycine-proline (PGP) is elevated in BOS patient bronchoalveolar lavage (BAL) fluid. The enzymes responsible for generating PGP, matrix metalloproteases 8 and -9 and prolyl endopeptidase, are also elevated in these samples. Together, IL-8 and PGP account for most of the neutrophil chemoattractant capacity seen in BOS BAL fluid. Using specific neutralizing Abs to both IL-8 and PGP, we demonstrate that PGP is a prominent neutrophil chemoattractant found in BAL fluid from individuals at the time of diagnosis of BOS. These findings highlight the influence of a matrix-derived neutrophil chemoattractant in posttransplantation BOS and provide opportunities for the development of unique diagnostics and therapeutics to potentially improve disease outcomes.

Figure 1

Figure 1A: Zymogram of representative patient BAL. Chronic rejection samples are paired at three months prior to (a) and at time of diagnosis (b) per individual. The majority of the chronic rejection samples collected at the time of diagnosis displayed increased gelatinolytic activity compared to their prior matched samples. None of the non-rejection or acute rejection samples displayed any detectable gelatinolytic activity. Figure 1B: Fold difference of MMP-8 activity relative to non-rejection transplant controls. Chronic rejection samples collected at the time of diagnosis had elevated levels of MMP-8 activity compared to controls and significantly higher quantities of the zymogen form of the enzyme compared to all patient populations (* p<0.05). Figure 1C: Relative activity of MMP-9 in BAL fluid from patient populations. The activity of both the active (* p< 0.01) and inducible († p<0.01) forms of MMP-9 are elevated in chronic rejection BAL at the time of diagnosis compared to non-rejection transplant controls.

Figure 1

Figure 1A: Zymogram of representative patient BAL. Chronic rejection samples are paired at three months prior to (a) and at time of diagnosis (b) per individual. The majority of the chronic rejection samples collected at the time of diagnosis displayed increased gelatinolytic activity compared to their prior matched samples. None of the non-rejection or acute rejection samples displayed any detectable gelatinolytic activity. Figure 1B: Fold difference of MMP-8 activity relative to non-rejection transplant controls. Chronic rejection samples collected at the time of diagnosis had elevated levels of MMP-8 activity compared to controls and significantly higher quantities of the zymogen form of the enzyme compared to all patient populations (* p<0.05). Figure 1C: Relative activity of MMP-9 in BAL fluid from patient populations. The activity of both the active (* p< 0.01) and inducible († p<0.01) forms of MMP-9 are elevated in chronic rejection BAL at the time of diagnosis compared to non-rejection transplant controls.

Figure 1

Figure 1A: Zymogram of representative patient BAL. Chronic rejection samples are paired at three months prior to (a) and at time of diagnosis (b) per individual. The majority of the chronic rejection samples collected at the time of diagnosis displayed increased gelatinolytic activity compared to their prior matched samples. None of the non-rejection or acute rejection samples displayed any detectable gelatinolytic activity. Figure 1B: Fold difference of MMP-8 activity relative to non-rejection transplant controls. Chronic rejection samples collected at the time of diagnosis had elevated levels of MMP-8 activity compared to controls and significantly higher quantities of the zymogen form of the enzyme compared to all patient populations (* p<0.05). Figure 1C: Relative activity of MMP-9 in BAL fluid from patient populations. The activity of both the active (* p< 0.01) and inducible († p<0.01) forms of MMP-9 are elevated in chronic rejection BAL at the time of diagnosis compared to non-rejection transplant controls.

Figure 2

Figure 2A: PE detection in pooled BAL fluid from study populations. BAL fluid was probed using a specific polyclonal antibody for prolyl endopeptidase (PE). Only the chronic rejection samples collected at the time of diagnosis detected PE (~80 kDa). Figure 2B: PE activity in pooled patient BAL fluid samples relative to transplant controls. Chronic rejection patients at the time of diagnosis demonstrate significantly elevated PE activity compared to non-rejection controls (* p< 0.05), and matched samples collected three months prior to diagnosis († p< 0.01).

Figure 2

Figure 2A: PE detection in pooled BAL fluid from study populations. BAL fluid was probed using a specific polyclonal antibody for prolyl endopeptidase (PE). Only the chronic rejection samples collected at the time of diagnosis detected PE (~80 kDa). Figure 2B: PE activity in pooled patient BAL fluid samples relative to transplant controls. Chronic rejection patients at the time of diagnosis demonstrate significantly elevated PE activity compared to non-rejection controls (* p< 0.05), and matched samples collected three months prior to diagnosis († p< 0.01).

Figure 3

Figure 3A: Myeloperoxidase activity in pooled BAL fluid samples relative to transplant controls. Chronic rejection samples from the time of diagnosis are significantly elevated compared to both non-rejection transplant controls (* p< 0.05), and matched samples collected three months prior to diagnosis († p< 0.05). Figure 3B: IL-8 levels in BAL fluid from lung transplant populations. IL-8 levels were measured in all groups using a commercially available kit as previously described. Chronic rejection BAL collected at the time of diagnosis demonstrated a six-fold increase in IL-8 levels compared to the individual’s prior matched samples († p<0.05) and non-rejection transplant controls (* p<0.05). Figure 3C: PGP levels in BAL fluid from lung transplant populations. Chronic rejection samples demonstrated a >25 fold increase in PE activity compared to the individual’s prior matched samples († p<0.05) and non-rejection transplant controls (* p< 0.05).

Figure 3

Figure 3A: Myeloperoxidase activity in pooled BAL fluid samples relative to transplant controls. Chronic rejection samples from the time of diagnosis are significantly elevated compared to both non-rejection transplant controls (* p< 0.05), and matched samples collected three months prior to diagnosis († p< 0.05). Figure 3B: IL-8 levels in BAL fluid from lung transplant populations. IL-8 levels were measured in all groups using a commercially available kit as previously described. Chronic rejection BAL collected at the time of diagnosis demonstrated a six-fold increase in IL-8 levels compared to the individual’s prior matched samples († p<0.05) and non-rejection transplant controls (* p<0.05). Figure 3C: PGP levels in BAL fluid from lung transplant populations. Chronic rejection samples demonstrated a >25 fold increase in PE activity compared to the individual’s prior matched samples († p<0.05) and non-rejection transplant controls (* p< 0.05).

Figure 3

Figure 3A: Myeloperoxidase activity in pooled BAL fluid samples relative to transplant controls. Chronic rejection samples from the time of diagnosis are significantly elevated compared to both non-rejection transplant controls (* p< 0.05), and matched samples collected three months prior to diagnosis († p< 0.05). Figure 3B: IL-8 levels in BAL fluid from lung transplant populations. IL-8 levels were measured in all groups using a commercially available kit as previously described. Chronic rejection BAL collected at the time of diagnosis demonstrated a six-fold increase in IL-8 levels compared to the individual’s prior matched samples († p<0.05) and non-rejection transplant controls (* p<0.05). Figure 3C: PGP levels in BAL fluid from lung transplant populations. Chronic rejection samples demonstrated a >25 fold increase in PE activity compared to the individual’s prior matched samples († p<0.05) and non-rejection transplant controls (* p< 0.05).

Figure 4

Figure 4A: Correlation of MMP-9 activity and PGP levels. The PGP levels and MMP-9 activity in chronic rejection samples taken at the time of diagnosis displayed a strong correlation r value of 0.92 (p<0.05). Figure 4B: Correlation of PE activity and PGP levels. The PGP levels and PE activity in chronic rejection samples taken at the time of diagnosis displayed a strong correlation r value of 0.88 (p< 0.05).

Figure 5. Neutrophil chemotaxis of BAL fluid from transplant populations

Chronic rejection BAL collected three months prior to and at the time of diagnosis demonstrated elevated chemotaxis indices compared to non-rejection transplant controls. Samples collected at the time of diagnosis demonstrated significantly elevated neutrophil chemotaxis compared to control transplant populations (p< 0.05) and matched samples prior to chronic rejection diagnosis (p< 0.05).

Figure 6. Relative IL-8 and PGP Chemotaxis in Chronic Rejection BAL fluid

The samples collected at the time of diagnosis demonstrated a shift in the chemokine profile from samples collected three months prior to chronic rejection. IL-8 neutralizing antibody (1:1000) created a 76.5% and 28.1% inhibition of chronic rejection (three months prior) and chronic rejection (at diagnosis) induced PMN chemotaxis, respectively. In contrast, the inhibition by the PGP neutralizing antibody (1:1000) was 29.2% (in the three months prior samples) and 58.4% (in the samples at time of diagnosis). Together, the antibodies were able to almost completely ablate the chemotactic capacity of the BAL fluid collected three months prior to (93.1%) and at the time of diagnosis (91.4%) of chronic rejection.

Figure 7. Model of ongoing neutrophilic inflammation in chronic rejection

In early chronic rejection there is mild damage to the airway and the epithelial cells produce IL-8 which causes PMN chemotaxis. The increased neutrophil burden increases levels of MMPs which further damage the tissue exposing the surrounding collagen. With the collagen breakdown by MMPs and PE, PGP is released causing further PMN influx and a worsening of the fibrosis and matrix turnover already occurring. The dysregulation of this system is most likely a major factor in the morbidity and mortality due to chronic rejection in lung transplant patients.