Tumour necrosis factor-alpha processing in interstitial lung disease: a potential role for exogenous proteinase-3

Abstract

Tumour necrosis factor (TNF) blockade has become an important immunomodulatory therapy, particularly in patients refractory to conventional immunosuppression, but responses can be unpredictable. Understanding the complex biology of TNF processing may be key to predicting such responses and reduce unwanted side effects. TNF bioavailability is regulated partly by TNF-alpha converting enzyme (TACE) cleavage; however, it can also be cleaved by proteinase-3 (PR-3). We have demonstrated this mechanism previously in healthy human alveolar macrophages (AM), leading us to hypothesize that PR-3-mediated TNF processing may be an important mechanism in inflammatory lung disease. Furthermore, this may be more apparent in diseases with a neutrophil component typical of usual interstitial pneumonia (UIP), compared with sarcoidosis, where lymphocytes predominate. We isolated AM from patients with UIP and sarcoidosis and healthy subjects. We found increased TACE expression on AM in sarcoidosis. In contrast, TACE was not increased in UIP; we found increased cleavage of glutathione S-transferase-proTNF) substrate, relative to both sarcoidosis and healthy controls. Furthermore, cleavage was subject to inhibition by serine protease inhibitor, rather than a TACE inhibitor BB-3103. Cleavage was proportional to the number of neutrophils isolated from bronchoalveolar lavage, whereas there was an inverse relationship between neutrophils and BB-3103 inhibition. There was also increased PR-3 on the AM surface in UIP relative to healthy controls. These data provide evidence for PR-3-mediated cleavage in UIP, which may have implications for future therapeutic targeting of TACE.

Fig. 1

Freshly isolated alveolar macrophages (AM) were incubated with specific phycoerythrin (PE)-labelled anti-tumour necrosis factor (TNF)-α converting enzyme (TACE) antibody and acquired on a flow cytometer (a). Data are expressed relative to PE-labelled isotype control. *P < 0·01 sarcoidosis (n = 15) versus healthy controls (n = 15), P < 0·05 sarcoidosis versus usual interstitial pneumonia (UIP) (n = 20), Tukey's multiple comparison tests (mct). AM were incubated with glutathione S-transferase (GST)-proTNF substrate for 2 h. The cleaved substrate was then detected by Western blotting using a specific GST antibody. *P < 0·05 sarcoidosis (n = 15) versus normal controls (n = 15), §P < 0·001 UIP (n = 20) versus healthy controls, P < 0·05 UIP versus sarcoidosis Tukey's mct (b). (a,b) P = 0·005, one-way analysis of variance.

Fig. 2

Sarcoidosis alveolar macrophages (AM) (n = 15) (a) and usual interstitial pneumonia (UIP) AM (n = 20) (b) were incubated with glutathione S-transferase (GST)-proTNF) substrate for 2 h. The cleaved substrate was then detected by Western blotting using a specific GST antibody. Cleavage of GST-proTNF by AM (n = 50) is associated with an increased number of neutrophils in the bronchoalveolar lavage fluid (BALF). P = 0·001, Pearson's correlation (c). The ability of BB-3103 to inhibit GST-proTNF cleavage is related inversely to the percentage number of alveolar neutrophils present in the BALF. AM (n = 50) were preincubated for 30 min with 10 µM BB-3103 before addition of GST-proTNF substrate for 2 h. P = 0·006, Pearson's correlation (d).

Fig. 3

Alveolar macrophages (AM) were cultured in the presence or absence of 10 µM BB-3103 and 1 mM Pefabloc for 30 min before addition of glutathione S-transferase (GST)-proTNF) substrate for a further 2 h. Data are expressed as a percentage of medium alone. Grey bars, medium alone; diagonal bars, 10 µM BB-3103; horizontal bars, Pefabloc; hatched bars, Pefabloc + BB-3103. **P < 0·01 sarcoidosis BB-3103 versus sarcoidosis medium alone; §P < 0·01 usual interstitial pneumonia (UIP) Pefabloc versus UIP medium alone, #P < 0·01 UIP Pefabloc + BB-3103 versus UIP BB-3103 alone, Student's paired t-test (a). In vitro cleavage of GST-proTNF by AM from a subject with UIP. Lane 1, no inhibitors; lane 2, Pefabloc; lane 3, BB-3103; lane 4, BB-3103 + Pefabloc; lane 5, assay in the absence of AM (b). Soluble TNF release by AM from healthy subjects (c), subjects with sarcoidosis (d) and UIP subjects (e) treated with BB-3103 (BB) or Pefabloc (PF), *P < 0·05, **P = 0·01, ***P < 0·001 versus medium alone.