Procollagen III N-terminal propeptide and desmosine are released by matrix destruction in pulmonary tuberculosis

Abstract

Background: Tuberculosis is transmitted by patients with pulmonary disease. Matrix metalloproteinases (MMPs) drive lung destruction in tuberculosis but the resulting matrix degradation products (MDPs) have not been studied. We investigate the hypothesis that MMP activity generates matrix turnover products as correlates of lung pathology.

Methods: Induced sputum and plasma were collected prospectively from human immunodeficiency virus (HIV) positive and negative patients with pulmonary tuberculosis and controls. Concentrations of MDPs and MMPs were analyzed by ELISA and Luminex array in 2 patient cohorts.

Results: Procollagen III N-terminal propeptide (PIIINP) was 3.8-fold higher in induced sputum of HIV-uninfected tuberculosis patients compared to controls and desmosine, released during elastin degradation, was 2.4-fold higher. PIIINP was elevated in plasma of tuberculosis patients. Plasma PIIINP correlated with induced sputum MMP-1 concentrations and radiological scores, demonstrating that circulating MDPs reflect lung destruction. In a second patient cohort of mixed HIV seroprevalence, plasma PIIINP concentration was increased 3.0-fold above controls (P < .001). Plasma matrix metalloproteinase-8 concentrations were also higher in tuberculosis patients (P = .001). Receiver operating characteristic analysis utilizing these 2 variables demonstrated an area under the curve of 0.832 (P < .001).

Conclusions: In pulmonary tuberculosis, MMP-driven immunopathology generates matrix degradation products.

Keywords: extracellular matrix; immunopathology; lung; matrix metalloproteinase; mycobacteria.

Figure 1.

Procollagen III N-terminal propeptide (PIIINP) concentrations are elevated in patients with tuberculosis (TB). (A) PIIINP was measured by ELISA in induced sputum samples from human immunodeficiency virus (HIV) negative patients with pulmonary TB and uninfected controls. (B) Sputum PIIINP concentrations correlated with sputum MMP-1 concentrations measured by luminex array, on Spearman analysis.

Figure 2.

Desmosine is elevated in the induced sputum of patients with pulmonary tuberculosis (TB) and correlates with MMP and cytokine concentrations. (A) Desmosine concentration was analyzed by ELISA in induced sputum from human immunodeficiency virus (HIV) negative patients with pulmonary tuberculosis. Mean desmosine concentrations were 2.4-fold higher in tuberculosis. (B–D) MMPs in induced sputum were measured by luminex multiplex array. MMP-2, -8 and -9 correlated most closely with desmosine concentrations (E and F). Cytokine concentrations were measured by luminex array and IL-6 and TNF-α associated most closely with sputum desmosine concentration.

Figure 3.

Procollagen III N-terminal propeptide (PIIINP) concentrations are elevated in the plasma of patients with tuberculosis (TB). (A) PIIINP concentration was measured by ELISA in plasma from patients of mixed human immunodeficiency virus (HIV) seroprevalence with a CD4 cell count of >200. PIIINP was significantly elevated in patients with pulmonary tuberculosis compared to controls. (B) Plasma Total C-terminal telopeptides of type I collagen (CTX-I) was no different between patients with tuberculosis and controls. (C) Plasma PIIINP correlated with induced sputum MMP-1 on Spearman's analysis. (D) Plasma PIIINP correlated with chest X-ray (CXR) infiltration scored on a scale of 1–10.

Figure 4.

Plasma PIIINP is elevated in a second cohort of patients with tuberculosis (TB) of mixed human immunodeficiency virus (HIV) seroprevalence and correlates with tissue damage on chest radiographs. PIIINP concentrations were analyzed in a second cohort recruited in Durban, South Africa, of mixed HIV seroprevalence. (A) Plasma PIIINP was significantly increased in patients with tuberculosis. (B) Plasma PIIICP concentration showed a trend towards increase in tuberculosis that did not achieve statistical significance. (C) Plasma PIIINP concentration was elevated in patients with an abnormal CXR compared to those with a normal radiograph. (D) Patients with cavitary pulmonary tuberculosis assessed by chest radiography had higher plasma PIIINP concentrations than those without cavities.

Figure 5.

Plasma MMP-7 and -8 are elevated, while MMP-2 is suppressed, in patients with tuberculosis (TB). Plasma MMP and extracellular matrix metalloproteinase inducer (EMMPRIN) concentrations were analyzed by luminex array. In patients with pulmonary tuberculosis, plasma MMP-2 concentrations were suppressed, while MMP-7 and MMP-8 were increased in tuberculosis. No other MMPs analyzed differed significantly.

Figure 6.

Receiver Operating Characteristic analysis of matrix-turnover products in plasma from patients with pulmonary tuberculosis (TB) compared to controls. (A) Variables included in the initial model were plasma PIIINP, PIIICP, MMP-2, MMP-7 and MMP-8, and the patients' BMI. (B) Variables in the final model were plasma PIIINP and MMP-8. Matrix turnover products may be incorporated into a multi-analyte panel to identify patients with pulmonary tuberculosis. Abbreviations: AUC, area under the curve; BMI, body mass index.