Metabolic crisis after traumatic brain injury is associated with a novel microdialysis proteome

Abstract

Background: To examine if the metabolic distress after traumatic brain injury (TBI) is associated with a unique proteome.

Methods: Patients with severe TBI prospectively underwent cerebral microdialysis for the initial 96 h after injury. Hourly sampling of metabolism was performed and patients were categorized as having normal or abnormal metabolism as evidenced by the lactate/pyruvate ratio (LPR) threshold of 40. The microdialysate was frozen for proteomic batch processing retrospectively. We employed two different routes of proteomic techniques utilizing mass spectrometry (MS) and categorized as diagnostic and biomarker identification approaches. The diagnostic approach was aimed at finding a signature of MS peaks which can differentiate these two groups. We did this by enriching for intact peptides followed by MALDI-MS analysis. For the biomarker identification approach, we applied classical bottom-up (trypsin digestion followed by LC-MS/MS) proteomic methodologies.

Results: Five patients were studied, 3 of whom had abnormal metabolism and 2 who had normal metabolism. By comparison, the abnormal group had higher LPR (1609 +/- 3691 vs. 15.5 +/- 6.8, P < 0.001), higher glutamate (157 +/- 84 vs. 1.8 +/- 1.4 microM, P < 0.001), and lower glucose (0.27 +/- 0.35 vs. 1.8 +/- 1.1 mmol/l, P < 0.001). The abnormal group demonstrated 13 unique proteins as compared with the normal group in the microdialysate. These proteins consisted of cytoarchitectural proteins, as well as blood breakdown proteins, and a few mitochondrial proteins. A unique as yet to be characterized peptide was found at m/z (mass/charge) 4733.5, which may represent a novel biomarker of metabolic distress.

Conclusion: Metabolic distress after TBI is associated with a differential proteome that indicates cellular destruction during the acute phase of illness. This suggests that metabolic distress has immediate cellular consequences after TBI.

Fig. 1

The computerized tomographic (CT) and magnetic resonance imaging of each patient. The three patients with elevated LPR are labelled as abnormal and the two with normal LPR are labelled normal. The arrow shows the locations of the microdialysis and the ventriculostomy (EVD). Each pair of images is labelled by patient

Fig. 2

The time plot of hourly microdialysis lactate/pyruvate ratio for all five patients. The abnormal patients are solid and normal patients are open circles. Each patient is represented by a continuous stream of data points. The LPR values have been log transformed to be able to plot the time series

Fig. 3

a, b The gel view of the mass spectra of abnormal and normal patients. The window shown is from m/z 4500–5000. The hyperdense band at m/z 4733.5 is seen in each of the abnormal patients and in none of the normal patients

Fig. 4

A Venn diagram showing the number of unique proteins for each subject and those proteins that were common among the abnormal and normal subgroups. The intersection shows the number of proteins that were in common

Fig. 5

An example of deconvoluted mass spectra in the abnormal patients. The inset shows the list of peptides that matched to hemoglobin beta. Many more peaks are visible as compared with the normal subject in Fig. 6

Fig. 6

An example of deconvoluted mass spectra in the normal patients

Fig. 7

The phosphopeptides at m/z 1545 and 1616 aligned from all abnormal and normal samples

Fig. 8

Phosphopeptides at m/z 1616 and 1545 matched to fibrinopeptide A of fibrinogen alpha chain precursor protein. The difference between these two peptides is the alanine residue shown in box