Infrared microspectroscopy: a multiple-screening platform for investigating single-cell biochemical perturbations upon prion infection

Abstract

Prion diseases are a group of fatal neurodegenerative disorders characterized by the accumulation of prions in the central nervous system. The pathogenic prion (PrP(Sc)) possesses the capability to convert the host-encoded cellular isoform of the prion protein, PrP(C), into nascent PrP(Sc). The present work aims at providing novel insight into cellular response upon prion infection evidenced by synchrotron radiation infrared microspectroscopy (SR-IRMS). This non-invasive, label-free analytical technique was employed to investigate the biochemical perturbations undergone by prion infected mouse hypothalamic GT1-1 cells at the cellular and subcellular level. A decrement in total cellular protein content upon prion infection was identified by infrared (IR) whole-cell spectra and validated by bicinchoninic acid assay and single-cell volume analysis by atomic force microscopy (AFM). Hierarchical cluster analysis (HCA) of IR data discriminated between infected and uninfected cells and allowed to deduce an increment of lysosomal bodies within the cytoplasm of infected GT1-1 cells, a hypothesis further confirmed by SR-IRMS at subcellular spatial resolution and fluorescent microscopy. The purpose of this work, therefore, consists of proposing IRMS as a powerful multiscreening platform, drawing on the synergy with conventional biological assays and microscopy techniques in order to increase the accuracy of investigations performed at the single-cell level.

Keywords: PrPSc; Prion; atomic force microscopy; chemical mapping; infrared microspectroscopy; synchrotron radiation.

Figure 1

Characteristic IR spectra of normal and scrapie GT1-1 cells (black and gray line, respectively). The spectral regions more representative for lipids, proteins, nucleic acids, and carbohydrates are shown. The origin of the most important absorption bands is reported in Table 1.

Figure 2

Prion infection in GT1-1 cells. (a) RML infected, ScGT1-1, and noninfected GT1-1 cells were tested for levels of PrP^Sc by Western blot. It can be noticed how in GT1-1 cells PrP is fully digested after PK treatment. (b) Cluster analysis (Euclidean distances; Ward’s algorithm) of first derivatives of spectra extended over the spectral range 1800−1150 cm⁻¹. Only spectra within one SD of the mean for the intensity criterion detailed in the Methods section 3.3 were considered. A clear classification of cellular spectra ascribable to prion infection can be inferred by the shown dendrogram.

Figure 3

(a) Average cell volumes of GT1-1 and ScGT1-1 as calculated by AFM. Healthy and scrapie cells display comparable average volumes [Avg(VGT1) = 307 μm³, Avg(VScGT1) = 286 μm³]. (b) 3D reconstruction of a representative GT1-1 cell fixed on Si₃N₄ membrane (45 μm × 35 μm scanned area). (c) 3D reconstruction of a representative ScGT1-1 cell fixed on Si₃N₄ membrane (40 μm × 40 μm scanned area). It can be noticed how GT1-1 and ScGT1-1 cells have similar pyramidal shapes in which the nuclear region represents the apex [(Min−Max) = (0−2.2 μm)].

Figure 4

(a) Vector normalized first derivatives of the spectra of GT1-1 (black line) and ScGT1-1 (gray line) in the 1800−1150 cm⁻¹ range; line thickness is proportional to one SD, comparable for infected and noninfected cells. Peak maxima in the original spectra cross the zero line in first derivatives. A, B, and C regions are defined by dashed line boxes. The arrowheads point out the spectral intervals showing major differences among spectra: a, 1710−1680 cm⁻¹; a′, 1610−1580 cm⁻¹; b, 1425−1325 cm⁻¹ ; c, 1290−1180 cm⁻¹. (b) Upper panel: average absorbance spectra of GT1-1 (black line with thickness proportional to one SD) and ScGT1-1 (gray line with thickness proportional to one SD) in the 1760−1480 cm⁻¹ spectral range. Lower panel: second derivatives of average spectra for GT1-1 and ScGT1. (c, d) Average absorbance spectra of GT1-1 (black line with thickness proportional to SD) and ScGT1-1 (gray line with thickness proportional to SD) in the 1425−1325 and 1290−1180 cm⁻¹ ranges, respectively. Plotted absorbance spectra are baseline corrected and normalized (minimum absorbance unit 0, maximum 2).

Figure 5

(a−c) Average cellular content of proteins, lipids, and phospholipids normalized on the total cellular biomass for GT1-1 and ScGT1-1 cells as estimated by semiquantitative analysis and detailed in the Methods section 3.3. The following values were obtained: proteins_GT1 = 0.190 ± 0.003 au, proteins_ScGT1 = 0.191 ± 0.002 au; lipids_GT1 = 0.043 ± 0.001 au, lipids_ScGT1 = 0.040 ± 0.001 au; phospholipids_GT1 = 0.0016 ± 0.0001 au, phospholipids_ScGT1 = 0.0012 ± 0.0001 au. For testing significant differences, unpaired Student’s t test was performed: **P < 0.01, n = 67 and *P < 0.05, n = 67. (d) Average total protein content as calculated by band integration in the spectral interval 1700−1480 cm⁻¹: proteins_GT1 = 3.814 ± 0.131 au, proteins_ScGT1 = 2.78 ± 0.047 au; significant for **P < 0.01, n = 67. (e) Protein content per cell as estimated by BCA: 1.47 × 10⁻¹⁰ ± 0.22 × 10⁻¹⁰ gr/GT1-1 cell and 1.07 × 10⁻¹⁰ ± 0.09 × 10⁻¹⁰ gr/ScGT1-1 cell; significant for *P < 0.05, n = 3. Black bars: GT1-1. Gray bars: ScGT1-1.

Figure 6

(a) Cell optical image (30 × 57 μm; scale bar 5 μm) of a single GT1-1 cell and related chemical maps, collected setting knife-edge apertures to 6 μm (oversampling 1:2), of total protein content [P, (Min−Max) = (0−6 au)], acyl chains of lipids [L, (Min−Max) = (0−2.5 au)], carbonyl ester of phospholipids [Ph, (Min−Max) = (0.01−0.08 au)]. (b) Cell optical image (30 × 40 μm; scale bar 5 μm) of a single GT1-1 cell and related chemical maps, collected setting knife-edge apertures to 6 μm (oversampling 1:2), of total protein content [P, (Min−Max) = (0−6 au)], acyl chains of lipids [L, (Min−Max) = (0−2.5 au)], carbonyl ester of phospholipids [Ph, (Min−Max) = (0.01−0.08 au)]. (c) Cell optical image (30 × 57 μm; scale bar 5 μm) of a single ScGT1-1 cell and related chemical maps, collected setting knife-edge apertures to 6 μm (oversampling 1:2), of total protein content [P, (Min−Max) = (0−4 au)], acyl chains of lipids [L, (Min−Max) = (0−1 au)], carbonyl ester of phospholipids [Ph, (Min−Max) = (0.01−0.04 au)]. (d) Cell optical image (23 × 66 μm; scale bar 5 μm) of a single ScGT1-1 cell and related chemical maps, collected setting knife-edge apertures to 6 μm (oversampling 1:2), of total protein content [P, (Min−Max) = (0−4 au)], acyl chains of lipids [L, (Min−Max) = (0−1 au)], carbonyl ester of phospholipids [Ph, (Min−Max) = (0.01−0.04 au)].

Figure 7

Lysosome detection and quantification. ScGT1-1 and GT1-1 cells were treated with Lysotracker and fixed as described in the Methods section 3.4. Lysosomes were stained in red, while nuclei were counterstained in blue with DAPI; merged images are shown on the right. ScGT1-1 cells exhibit bigger and more numerous lysosomes compared to GT1-1 cells. Images are representative of three independent experiments of staining. Scale bar, 20 μm. The fluorescence signal was quantified using ImageJ software, and the results are plotted in the adjacent histogram (black bar, GT1-1; gray bar, ScGT1-1). Statistics were performed using Student’s t test on a set of three independent experiments. *P < 0.05.