Technical advance: Caspase-1 activation and IL-1β release correlate with the degree of lysosome damage, as illustrated by a novel imaging method to quantify phagolysosome damage

Abstract

In addition to the lysosome's important roles in digestion, antigen processing, and microbial destruction, lysosome damage in macrophages can trigger cell death and release of the inflammatory cytokine IL-1β. To examine the relationship among endocytosis, lysosome damage, and subsequent events, such as caspase-1 activation and IL-1β secretion, we developed a method for measuring lysosome disruption inside individual living cells, which quantifies release of Fdx from lysosomes. Unperturbed, cultured BMM exhibited low levels of lysosome damage, which were not increased by stimulation of macropinocytosis. Lysosome damage following phagocytosis differed with different types of ingested particles, with negligible damage after ingestion of sRBC ghosts, intermediate damage by polystyrene (PS) beads, and high levels of damage by ground silica. Pretreatment with LPS decreased the amount of lysosome damage following phagocytosis of PS beads, silica microspheres, or ground silica. Activation of caspase-1 and subsequent release of IL-1β were proportional to lysosome damage following phagocytosis. The low level of damage following PS bead phagocytosis was insufficient to activate caspase-1 in LPS-activated macrophages. These studies indicate that lysosome damage following phagocytosis is dependent on particle composition and dose and that caspase-1 activation and IL-1β secretion correlate with the extent of lysosome damage.

Figure 1.. Detection of Fdx released from lysosomes.

RAW macrophage lysosomes were loaded overnight with Fdx and then chased for several hours in fresh medium. Cells were left untreated (A) or incubated with LLME for 30 min (B), and then epi-fluorescence images were collected. All pH images are color-coded such that warm colors (red, pink, and yellow) represent Fdx in acidic compartments, and cool colors (cyan, dark cyan, blue, and dark blue) represent Fdx in pH neutral compartments. Ratios (ex. 492 nm:ex. 435 nm) were converted to pH using calibration curves generated by measuring ex. 492 nm:ex. 435 ratios at fixed pHs (C). The brightness of the Fdx ex. 435 nm and ex. 492 nm images is scaled identically in all micrograph sets. (D) The percentage of Fdx released from the lysosomes is plotted as mean ± sem of ≥70 cells (P<0.0001). BMM were loaded and imaged as above. (E) Epi-fluorescence images of otherwise unperturbed macrophages were then collected. Images of unperturbed cells depicting very low (i, 7.7%), slight (ii, 22.4%), and moderate (iii, 31.6%) lysosome release. (F) Lysosome release from all of the negative control images acquired for G and see Fig. 2. (G) Typical images of BMM following M-CSF-induced macropinocytosis. Original scale bars represent 10 μm. (H) Lysosome release from all of the cells in each group was calculated and the averages plotted. Error bars are sem. Data are pooled averages of 3 experiments; n ≥ 200 cells total; P < 0.0001.

Figure 2.. Lysosome damage following phagocytosis.

BMM lysosomes were labeled with Fdx. Cells were fed (A) no particles, (B) IgG-opsonized sRBC ghosts (labeled with Texas-Red dextran; sRBC 3.3×10⁶ sRBC/mL), (C) PS beads (2×10⁵ PS beads/mL), or (D) ground silica (170 μg/mL) and imaged 1 h later. These particle doses were determined empirically to yield images with 1–5 particles/cell. Texas-Red images were used to identify cells containing sRBC ghosts. Original scale bar represents 10 μm. (E) The average percent of lysosome release was calculated for each condition. Data represent mean ± sem for 3 separate experiments with at least 80 cells examined in each condition. All conditions are significantly different (P<0.0005) from each other, except for no particle (NP) versus sRBC ghost. (F) The data from E, replotted as histograms, with percent of Fdx released (x-axis) compared with the number of cells containing a given level of Fdx release (y-axis).

Figure 3.. LPS treatment decreases lysosome release following phagocytosis.

BMM were labeled with Fdx and chased in unlabeled medium with (open bars) or without (closed bars) 100 ng/mL LPS. Cells were fed particles and imaged, and the average percent of lysosome release was calculated for each condition. y-Axis for all graphs plot the average percent of Fdx released. (A) LPS-stimulated or naïve BMM were fed 200 μg/mL ground silica, and lysosome damage was measured. (B) The number of PS beads/cell was counted and indicated for each PS bead dose. (A and B) Bars are pooled averages from 3 experiments with at least 170 cells/condition (means±sem). (C) Cells were fed acid-washed SMS coated with nothing (upper left), PLL (upper right), BSA (lower left), or BSA and anti-BSA IgG (lower right). Cells were imaged, and the number of phagocytosed SMS was recorded for each cell. Cells were grouped by number of SMS phagocytosed, and mean ± sem was plotted for each group. Statistical comparisons in A and B are as indicated with brackets and in C are between unstimulated and LPS-prestimulated groups within each SMS type and count. P values: NSD, Not significantly different; *P < 0.05; **P < 0.0001.

Figure 4.. Caspase-1 activation correlates with lysosome damage.

BMM were treated overnight with or without LPS and then fed ground silica at 57, 170, or 500 μg/mL or were left unfed (No particles). (A and B) Cells were then stained for an additional 1 h using the caspase-1 FLICA reagent; propidium iodide was added for the last 10 min to detect cells with permeabilized plasma membranes. Active caspase-1 (FLICA) and propidium iodide were imaged and quantified. (A) Sample images of LPS-pretreated cells fed 170 μg/mL ground silica. The cell on the right shows high levels of caspase-1 activation detected by FLICA, nuclear propidium iodide staining, and a phase-dark nucleus, and the cell on the left contains silica but is negative for FLICA, propidium iodide, and nuclear condensation. Original scale bar represents 10 μm. (B) Data are displayed as histograms with an integrated FLICA signal for individual cells plotted on the horizontal axis. (C) BMM lysosomes were labeled with Fdx in LPS-containing medium. Macrophages were fed silica as above and imaged 1 h later. Data are the pooled results of three independent experiments (n≥400 total cells/condition). (D) BMM treated with LPS were fed nothing (No particles), PS beads, ground silica, or sRBC ghosts for 1 h and then stained for active caspase-1. Histograms show integrated caspase-1 FLICA signal for individual cells plotted on the horizontal axis. Data are pooled results of three independent experiments with at least 400 total cells examined per condition. (E) Data from LPS-prestimulated BMM are replotted as histograms for comparison with D.

Figure 5.. Secretion of IL-1β following particle phagocytosis is proportional to lysosome damage.

BMM pretreated with LPS were fed opsonized sRBC ghosts, PS beads, or silica. After 4 h, culture supernatants were harvested and analyzed for IL-1β content. Data are averaged results from 3 independent experiments.