Comparison of mass spectra between amyotrophic lateral sclerosis (ALS) and controls using SAX2 chip. (Top panel) Representative spectral peaks of ALS and control subjects on SAX2 Proteinchips. The mass-to-charge ratio in kilodaltons (2–20 kDa) is shown on the x-axis, and the relative abundances (peak intensities) on the y-axis. (Bottom panel) Magnification (indicated by the arrows) of the spectra in the range of 3–6 kDa. Numbers denote spectral peaks with significant differences (p < 0.05) in relative abundance between control and ALS subjects as determined by Mann–Whitney statistical analysis. Numbers 1–5 correspond to spectral peaks 3.92, 4.12, 4.47, 4.80 and 5.87 kDa.

Univariate analysis of analytes from SAX2 and Zn-IMAC30. SAX2 (a) and Zn-IMAC30 (b) datasets exhibited several peak intensities with statistical significance (p ≤ 0.01) between amyotrophic lateral sclerosis (ALS) (stippled bars) and control (solid black bars) subjects. Mass in kilodaltons (kDa) is shown on the x-axis, and the mean peak intensity values on the y-axis. Error bars correspond to standard error of mean (SEM). All analyte signals appeared in greater than 30% of subjects. The non-parametric Mann–Whitney test was used for statistical analysis.

Validation of the 13.78 (transthyretin) and 13.38 (cystatin C) kDa protein peaks by western blot.(a) Immunoblot of transthyretin (TTR) protein in control and amyotrophic lateral sclerosis (ALS) CSF. Fifty micrograms of control (lanes 1–4) or ALS (lanes 5–8) CSF probed for TTR. Purified human TTR (50 ng) was used as a positive control (lane ‘+’). The arrows on the right indicate the monomeric and homotetrameric forms of TTR protein. (b) Immunoblot of cystatin C protein in control and ALS CSF. Twenty-five micrograms of total CSF protein from control (lanes 1–4) or ALS (lanes 5–8) subjects probed for cystatin C. Purified human cystatin C (10 ng) was used as a positive control (lane ‘+’). Molecular weight markers indicated in kDa are listed to the left of the positive control lanes in (a) and (b). (c) Densitometric quantification of TTR and cystatin C proteins from panels (a) and (b) using NIH 1.58 software. The relative abundance was expressed per amount of CSF protein loaded per gel lane and the height of the bars corresponds to the average value for each protein in control and ALS subjects ± standard error mean (SEM). Single factor anova was performed with a confidence interval of 95% to compare the protein levels of controls and ALS. The p-values for TTR and cystatin C were 0.03 and 0.04, respectively. Asterisks indicate the statistical significance.

Transthyretin (TTR) localization in lumbar spinal cord motor neurons. TTR immunoreactivity is reduced in amyotrophic lateral sclerosis (ALS) spinal motor neurons. Human lumbar spinal cord tissues from 16 ALS and eight age-matched non-neurologic disease controls were immunostained for TTR. TTR was visualized using 3-amino-9-ethylcarbazole (AEC) as chromagen (reddish brown color) and counterstained with haematoxylin. Magnification for all panels is × 100 with the insets at × 200. (a and b) Representative panels from two separate control subjects. Lumbar spinal cord motor neurons contain TTR within punctate structures in the cytoplasm. (c and d) ALS subjects exhibit diminished numbers of motor neurons and reduced TTR immunoreactivity within remaining motor neurons. Each panel represents distinct ALS subjects.