Clearing the airways: Distinguishing smoking-related diseases by NMR

Airway diseases

One of the many consequences of smoking tobacco is the risk of developing airway respiratory diseases but clinicians can find it difficult to diagnose them accurately and to distinguish one from the other. The symptoms can be similar and they can be masked by the effects of smoking itself.

A team of scientists based in Italy has taken a novel approach to this diagnostic dilemma by employing NMR spectroscopy to study the metabolome of exhaled breath condensate (EBC). Andrea Motta from the Institute of Biomolecular Chemistry, National Research Council, Naples, and colleagues from other Italian organisations knew from previous experience with EBC that the technique can give results within 10-15 minutes after minimal sample preparation. When working with EBC, NMR spectroscopy can also detect contamination by external influences and by saliva, which contains characteristic carbohydrates.

They tested the procedure by comparing the metabolic profiles of current smokers without any indications of respiratory disease to those of smokers with chronic obstructive pulmonary disease (COPD) and pulmonary Langerhans cell histiocytosis (PLCH).

COPD can take hold early in the smoking career, causing irreversible damage to the lungs. Giving up smoking can slow down the progression of the disease but it cannot be cured. In reality, COPD is a blanket term for several lung diseases, including chronic bronchitis, emphysema and chronic obstructive airways disease.

PLCH is far less common than COPD but is closely associated with smoking since 90% of the cases are smokers. PLCH is characterised by uncontrolled proliferation of Langerhans cells in clusters in the lungs to form nodules. Patients can recover completely if they stop smoking permanently but that is not guaranteed.

Exhaled breath condensate

The test panel consisted of 15 current smokers with COPD, 15 patients with PLCH and 20 “healthy” smokers. They all donated EBC by breathing for 15 minutes into a collection device which condensed the breath and also trapped any saliva to prevent its interference. The samples were sealed into glass vials, frozen in liquid nitrogen and stored at -80°C to quench any continuing metabolism and maintain the concentrations of the metabolites.  

After defrosting, the samples were treated with a solution of a deuterated compound in deuterated water to act as a frequency lock and chemical shift reference. One- and two-dimensional NMR spectra were collected and the data were processed by statistical analysis using the projection to latent structures discriminant analysis (PLS-DA), which was more discerning than principal components analysis.

Metabolite biomarkers for each disease

The PLS-DA analysis of the spectra successfully classified the samples. The people with COPD were clearly distinguished from those with PLCH and they were both also distinguished from the smokers without pulmonary disease.

It took only nine metabolites to discriminate between the three conditions. They were common to all of the subjects but their concentrations differed. Acetate, methanol, ethanol, propionate and 2-propanol increased from healthy smokers to COPD and this group decreased slightly from COPD to PLCH, apart from methanol which increased slightly. Conversely, formate, 1-methylimidazole, acetoin and isobutyrate were lower in COPD and more abundant in PLCH.

More specifically, COPD and PLCH were distinguished from healthy cases by high acetate and low 1-methylimidazole levels. When comparing COPD with PLCH, COPD had high 2-propanol and low isobutyrate whereas the converse was observed for PLCH, with high isobutyrate and low 2-isopropanol.

These findings confirm that chronic smoking does not produce a common metabolic state for the two diseases. COPD and PLCH generate different metabolic profiles which override the general profile induced by tobacco exposure and this can be detected by NMR spectroscopy.

Although the results must be considered preliminary due to the small number of participants, there is no reason to assume that the technique cannot be extended to other diseases of the airways, presenting a novel, non-invasive diagnostic approach.