Phagocytosis of Borrelia burgdorferi, the Lyme disease spirochete, potentiates innate immune activation and induces apoptosis in human monocytes

Abstract

We have previously demonstrated that phagocytosed Borrelia burgdorferi induces activation programs in human peripheral blood mononuclear cells that differ qualitatively and quantitatively from those evoked by equivalent lipoprotein-rich lysates. Here we report that ingested B. burgdorferi induces significantly greater transcription of proinflammatory cytokine genes than do lysates and that live B. burgdorferi, but not B. burgdorferi lysate, is avidly internalized by monocytes, where the bacteria are completely degraded within phagolysosomes. In the course of these experiments, we discovered that live B. burgdorferi also induced a dose-dependent decrease in monocytes but not a decrease in dendritic cells or T cells and that the monocyte population displayed morphological and biochemical hallmarks of apoptosis. Particularly noteworthy was the finding that apoptotic changes occurred predominantly in monocytes that had internalized spirochetes. Abrogation of phagocytosis with cytochalasin D prevented the death response. Heat-killed B. burgdorferi, which was internalized as well as live organisms, induced a similar degree of apoptosis of monocytes but markedly less cytokine production. Surprisingly, opsonophagocytosis of Treponema pallidum did not elicit a discernible cell death response. Our combined results demonstrate that B. burgdorferi confined to phagolysosomes is a potent inducer of cytosolic signals that result in (i) production of NF-kappaB-dependent cytokines, (ii) assembly of the inflammasome and activation of caspase-1, and (iii) induction of programmed cell death. We propose that inflammation and apoptosis represent mutually reinforcing components of the immunologic arsenal that the host mobilizes to defend itself against infection with Lyme disease spirochetes.

FIG. 1.

Phagocytosis of live B. burgdorferi results in enhanced transcription of proinflammatory cytokine genes. (A) PBMCs from five healthy volunteers were incubated for 4 h with live or lysed B. burgdorferi at an MOI of 10 or with polystyrene microspheres at a bead-to-cell ratio of 10. RNAs then were isolated for measurement of cytokine gene transcripts by qRT-PCR. The y axes show the fold increases in GAPDH-normalized values for TNF-α, pro-IL-1β, IL-6, and IFN-γ relative to suspensions incubated without bacteria; the bars indicate the means, and the error bars indicate the standard errors of the means. An asterisk indicates that the P value is <0.05 (for comparisons of live B. burgdorferi and lysates). (B) Monocytes in PBMCs incubated for 8 h with live or lysed Bb-GFP at the indicated MOIs were analyzed for GFP fluorescence by flow cytometry. Note that all monocytes, including those incubated with live spirochetes at an MOI of 100, remained strongly CD14⁺. The percentage of GFP-positive cells is indicated in each cytogram. The cytograms are representative of three independent experiments. SDS-PAGE (inset) confirmed that the incubation mixtures contained equivalent amounts of live or lysed B. burgdorferi.

FIG. 2.

Live B. burgdorferi, but not lysed B. burgdorferi, induces monocyte loss and shrinkage in PBMCs. (A) Enumeration of monocytes in PBMCs following 8 h of incubation with graded doses of live or lysed B. burgdorferi. The bars indicate the averages and the error bars indicate the standard errors of the means of six independent experiments. An asterisk indicates a value that is statistically significantly different (P ≤ 0.05) than the value for unstimulated (Uns) cultures. (B) FSC/SSC profiles of monocytes, DCs, and T cells from PBMCs incubated with live or lysed B. burgdorferi at graded MOIs. The percentages of each immune cell type in the PBMC suspensions are indicated in the density plots. Monocytes with altered SSC/FSC properties are indicated by arrows. The cytograms for the monocytes are representative of six independent experiments, while those for DCs and T cells are representative of three independent experiments. The cytograms were obtained using PBMCs from the same individual.

FIG. 3.

Live B. burgdorferi induces DNA fragmentation predominantly in monocytes containing spirochetes. (A) PBMCs were incubated for 4 h with 1 μM staurosporine (STS), with fluorescent carboxylate beads (bead-to-cell ratio, 10), or with live or lysed B. burgdorferi (Bb) (MOI, 10 or 100) and then assayed for DNA fragmentation by determination of TUNEL activity. Monocytes and lymphocytes were identified via their distinctive FSC/SSC properties. The monocyte graphs show the means and standard errors of the means for four independent experiments; the lymphocyte histogram is representative of the histograms obtained in four experiments. (B) PBMCs incubated with live or lysed Bb-GFP were assayed as described above and analyzed by flow cytometry. Monocytes stimulated with staurosporine became TUNEL positive but remained GFP negative; cells incubated with inert fluorescent beads became GFP positive without evidence of DNA disruption. The percentage of TUNEL-positive cells is indicated in each cytogram. The density plots are representative of four independent experiments.

FIG. 4.

Internalization of live spirochetes induces morphological changes consistent with programmed cell death. PBMCs incubated for 6 h with Bb-GFP at an MOI of 10 or with staurosporine (1 μM) were stained with DAPI and examined for microscopic evidence of programmed cell death. (A) Phase-contrast microscopy (top row) revealed blebbing (arrows) in cells stimulated with staurosporine or B. burgdorferi. With fluorescence microscopy (bottom row), the same cells exhibited nuclear fragmentation and condensation as determined by DAPI staining (arrows). GFP can be seen in the micrograph showing an apoptotic cell incubated with Bb-GFP. (B) Localization of GFP in cells demonstrating nuclear disruption. (C) Costaining of cells with LysoTracker Red confirmed that the GFP was confined to lysosomal compartments.

FIG. 5.

Inhibition of phagocytosis prevents loss of monocytes. Cell suspensions were preincubated for 1 h without or with 10 μg/ml cytochalasin D (CD) (Sigma) prior to the addition of spirochetes. The numbers of monocytes remaining after a subsequent 8 h of incubation were determined by flow cytometry. The bars indicate the means and the error bars indicate the standard errors of the means for three independent experiments. An asterisk indicates that there was a statistically significant difference. Uns, unstimulated; NS, not significant.

FIG. 6.

Heat-killed B. burgdorferi and viable B. burgdorferi induce comparable levels of apoptosis following internalization by monocytes. (A) Fluorescence microscopy and flow cytometry of live and heat-killed (H.Killed) Bb-GFP. The density plots from one of two separate analyses are shown. Staining with the cell-permeant nucleic acid dye SYTO59 was used to confirm that GFP fluorescence was associated with organisms. (B) Internalization of live and heat-killed Bb-GFP by monocytes as determined by flow cytometry. The bars indicate the means and the error bars indicate the standard deviations for three independent experiments. None of the differences was statistically significant. Uns, unstimulated. (C) Internalization of heat-killed B. burgdorferi (HK B.b) by monocytes induces changes in FSC/SSC properties consistent with cell shrinkage. The percentages in the top panels indicate the percentages of GFP-positive monocytes, while the percentages in the lower panel indicate the proportions of monocytes remaining in PBMC suspensions. The arrows indicate monocytes with altered FSC/SSC properties. The cytograms are representative of three independent experiments. (D) Live and heat-killed B. burgdorferi induce comparable losses of monocytes from PBMCs. The bars indicate the means and the error bars indicate the standard deviations for three independent experiments. (E) Correlation of internalization with induction of DNA fragmentation, as determined by the TUNEL assay. The percentage of TUNEL-positive monocytes is indicated in each cytogram. Representative results from two independent experiments are shown. Live Bb, live B. burgdorferi; H.K Bb, heat-killed B. burgdorferi.

FIG. 7.

Heat-killed B. burgdorferi induces a significantly diminished inflammatory cytokine response compared to live B. burgdorferi. TNF-α (A) and IL-1β (B) levels were measured in culture supernatants following 8 h of incubation of PBMCs with live, heat-killed (H. Killed), or lysed spirochetes at the indicated MOIs. One asterisk indicates a P value of ≤0.05 for a comparison between live B. burgdorferi and heat-killed B. burgdorferi at equivalent MOIs; two asterisks indicate a P value of ≤0.05 for a comparison between heat-killed B. burgdorferi and lysed B. burgdorferi at equivalent MOIs. The bars indicate the means and the error bars indicate the standard deviations for four independent experiments. (C) IFN-γ production by NK cells as measured by intracellular cytokine staining in two independent experiments. Uns, unstimulated; NS, not significant.

FIG. 8.

Opsonophagocytosis of T. pallidum does not induce programmed cell death in monocytes. PBMCs were incubated for 8 h with live T. pallidum at an MOI of 1, 10, or 100 in the presence or absence of 10% heat-inactivated human syphilitic serum. Changes in monocyte (A) numbers and (B) FSC/SSC properties were assessed by flow cytometry after an 8-h incubation period. The percentages of monocytes remaining in PBMC suspensions are indicated in the cytograms in panel B. In panel A the bars indicate the means and the error bars indicate the standard errors of the means from five independent experiments. No statistically significant differences were observed when opsonized T. pallidum was compared with unstimulated controls. (C) Percent recovery of T. pallidum (in the absence or presence of human syphilitic serum) and B. burgdorferi following 8 h of incubation with PBMCs. The values for T. pallidum and B. burgdorferi are the means of two and three experiments, respectively. Tp, T. pallidum; Bb, B. burgdorferi; Uns, unstimulated; HSS, human syphilitic serum.