BTK inhibitor-induced defects in human neutrophil effector activity against Aspergillus fumigatus are restored by TNF-α

Abstract

Inhibition of Bruton's tyrosine kinase (BTK) through covalent modifications of its active site (e.g., ibrutinib [IBT]) is a preferred treatment for multiple B cell malignancies. However, IBT-treated patients are more susceptible to invasive fungal infections, although the mechanism is poorly understood. Neutrophils are the primary line of defense against these infections; therefore, we examined the effect of IBT on primary human neutrophil effector activity against Aspergillus fumigatus. IBT significantly impaired the ability of neutrophils to kill A. fumigatus and potently inhibited reactive oxygen species (ROS) production, chemotaxis, and phagocytosis. Importantly, exogenous TNF-α fully compensated for defects imposed by IBT and newer-generation BTK inhibitors and restored the ability of neutrophils to contain A. fumigatus hyphal growth. Blocking TNF-α did not affect ROS production in healthy neutrophils but prevented exogenous TNF-α from rescuing the phenotype of IBT-treated neutrophils. The restorative capacity of TNF-α was independent of transcription. Moreover, the addition of TNF-α immediately rescued ROS production in IBT-treated neutrophils, indicating that TNF-α worked through a BTK-independent signaling pathway. Finally, TNF-α restored effector activity of primary neutrophils from patients on IBT therapy. Altogether, our data indicate that TNF-α rescued the antifungal immunity block imposed by inhibition of BTK in primary human neutrophils.

Keywords: Cancer; Fungal infections; Immunology; Infectious disease; Neutrophils.

Figure 1. IBT inhibition dampened human neutrophil effector activity against A. fumigatus.

(A) Metabolic activity of A. fumigatus B5233 strain measured using resazurin. Human neutrophils were pretreated for 4 hours (4h) with IBT and stimulated with A. fumigatus (MOI:0.25) for 5h. Data are shown as mean ± SD, n = 3; data are representative of at least 3 independent experiments. (B) Percentages of growth inhibition derived from A using linear regression analysis in a Gompertz fit. Data are shown as 95% CI, n = 3. Ordinary 1-way ANOVA and Tukey’s multiple-comparison test with a single pooled variance demonstrated a P < 0.0001 for all IBT treatments versus DMSO alone. (C and D) Human neutrophils were treated for 4h with IBT or DMSO and then stimulated with 1 mg/mL A. fumigatus B5233 strain heat-killed hyphal elements (C) or 1 μg/mL PMA (D). ROS production was measured by chemiluminescence using lucigenin. Data are shown as mean ± SD, n = 3. (E) Human neutrophils were treated with IBT or DMSO for 4h and incubated with Af488-labeled A. fumigatus B5233 strain (conidia⁺) swollen spores (MOI: 10). A subset of neutrophils was pretreated with 20 μM of cytochalasin D (Cyto D). The displayed percentage of phagocytic neutrophils (CD45-AF700⁺CD66b-APC⁺conidia-AF488⁺) was estimated based on the total number of viable neutrophils (CD45-AF700⁺CD66b-APC⁺). A minimum of 10,000 viable CD66b-APC⁺ events were recorded. (F–H) Human neutrophils were treated with IBT or DMSO for 4h before coincubation with A. fumigatus B5233 strain. Representative microscopy panels from the swarming assay showing neutrophil swarm formations 200 minutes (min) after coincubation, white circles depict areas seeded with A. fumigatus (F). Area of human neutrophil swarm 200 min after coincubation with A. fumigatus seeded spores (G). Area of fungal growth per cluster on swarming array slides after 16h, normalized to A. fumigatus growth without neutrophils (H). Data are shown as mean ± SD, n = 8. Ordinary 1-way ANOVA and Tukey’s multiple-comparison test with a single pooled variance. *P < 0.05; ***P < 0.001; ****P < 0.0001. For all panels, data are representative of at least 3 independent experiments.

Figure 2. IBT induced downstream upregulation of the TNF-α pathway in human neutrophils.

(A) Volcano plot for DEGs in neutrophils treated with 0.3 μM IBT versus DMSO (4.5h, unstimulated). DEGs based on log₂ fold change and P_adj < 0.05. FDRs were calculated using the Benjamini-Yekutieli method with 3 biological replicates per condition. Red and blue dots represent upregulated and downregulated genes, respectively. (B) TNF-α KEGG pathway was created for all probed genes for IBT-treated neutrophils versus DMSO. Genes in white boxes are genes not included in the nCounter panel. Numbers in circles represent pathways: (1) MAPK signaling pathway; (2) ubiquitin-mediated proteolysis; (3) NF-κB signaling pathway; and (4) PI3K/Akt signaling pathway.

Figure 3. TNF-α rescued IBT-induced immune defects in neutrophils against A. fumigatus.

Human neutrophils were treated with 0.03 μM IBT, 0.3 μM IBT, or DMSO for 30 min followed by a 4h incubation with TNF-α and coincubated with A. fumigatus B5233 strain for all figure panels. For all panels, data are representative of at least 3 independent experiments. (A) Neutrophils were incubated with A. fumigatus (MOI:0.25) for 5h, and metabolic activity was measured by resazurin assay. Data calculated through time course study (see raw data in the Supporting Data Values file) and panel represent the output from linear regression analysis using Gompertz fit with percentages of growth inhibition of A. fumigatus by neutrophils in reference to IBT-treated neutrophils. Data are shown as 95% CI, n = 3. Ordinary 1-way ANOVA and Tukey’s multiple-comparison test with a single pooled variance demonstrated a P < 0.001 for all TNF-α treatments versus IBT alone. (B) Neutrophils were stimulated with 1 mg/mL A. fumigatus heat-killed hyphae. ROS production was measured by chemiluminescence using lucigenin. Data are shown as mean ± SD, n = 3. (C) Microscopy panels showing neutrophils swarm formations 200 min after coincubation. (D) Area of neutrophil swarm after 200 min. (E) Area of fungal growth normalized to the growth of A. fumigatus without neutrophils after 16h. Data are shown as mean ± SD, n = 24. Ordinary 1-way ANOVA and Tukey’s multiple-comparison test with a single pooled variance. **P < 0.01; ****P < 0.0001. (F) Neutrophils were coincubated with AF488-labeled A. fumigatus swollen spores (MOI: 10). The displayed percentage of phagocytic neutrophils (CD45-AF700⁺CD66b-APC⁺conidia-AF488⁺) was estimated based on the total number of viable neutrophils (CD45-AF700⁺CD66b-APC⁺). At minimum, 10,000 viable CD66b-APC⁺ events were recorded. (G) Heatmap for DEG based on log₂ fold change (1.5 < log₂ fold change < –1.5) and a P_adj < 0.05. FDR was calculated using the Benjamini-Yekutieli method with 3 biological replicates per condition. RNA from neutrophils coincubated for 5h with A. fumigatus B5233 strain (MOI: 2.5).

Figure 4. TNF-α restored defects caused by multiple BTK inhibitors on neutrophil immune activity against A. fumigatus.

Human neutrophils were treated with ABT, ZBT, or DMSO for 30 min followed by a 4h incubation with TNF-α and coincubated with A. fumigatus B5233 strain for all figure panels. For all panels, data are representative of at least 3 independent experiments. (A) Neutrophils were incubated with A. fumigatus (MOI: 0.25) for 5h, and metabolic activity was measured using a resazurin assay. Data calculated through time course study (see raw data in the Supporting Data Values file) and panel represent the output from linear regression analysis using Gompertz fit with percentages of growth inhibition of A. fumigatus by neutrophils in reference to neutrophils treated with the respective BTK inhibitor. Data are shown as 95% CI, n = 3. Ordinary 1-way ANOVA and Tukey’s multiple-comparison test with a single pooled variance demonstrated a P < 0.001 for all IBT treatments versus BTK inhibitor (ABT or ZBT) alone. (B) Neutrophils were incubated with 1 mg/mL A. fumigatus heat-killed hyphae. ROS production was measured by chemiluminescence using lucigenin. Data are shown as mean ± SD, n = 3. (C) Neutrophils treated with ABT (left 2 panels) or ZBT (right 2 panels) were coincubated with labeled A. fumigatus swollen spores (MOI: 10). The displayed percentage of phagocytic neutrophils (CD45-AF700⁺CD66b-APC⁺conidia-AF488⁺) was estimated based on the total number of viable neutrophils (CD45-AF700⁺CD66b-APC⁺). At minimum, 10,000 viable CD66b-APC⁺ events were recorded. (D–G) Swarming assay was measured by confocal microscopy view of A. fumigatus conidia spots after 200 min. Area of neutrophil swarm after 200 min for neutrophils treated with ABT (D) or ZBT (E). Area of fungal growth per cluster on swarming array slides normalized to the growth of A. fumigatus without neutrophils after 16h, for neutrophils treated with ABT (F) or ZBT (G). Treatment controls correspond to the same swarming array experiment (D–G). Data are shown as mean ± SD, n = 24. Ordinary 1-way ANOVA and Tukey’s multiple-comparison test with a single pooled variance. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 5. Restorative activity of exogenous TNF-α signals independent of transcription.

(A) Neutrophils were incubated with A. fumigatus B5233 strain (MOI: 2.5) for 5h, and metabolic activity was estimated by fluorescence. Data calculated through time course study (see raw data in the Supporting Data Values file) and panel represent the output from linear regression analysis using Gompertz fit with data shown as 95% CI, n = 3. Ordinary 1-way ANOVA and Tukey’s multiple-comparison test with a single pooled variance demonstrated a P < 0.001 for TNF-α alone, IFM alone, and in combination with IBT treatments versus IBT alone and P = 0.0004 for IBT + TNF-α versus IBT alone. (B) ROS production in IBT-treated neutrophils incubated with 25 μg/mL IFM in the presence of exogenous TNF-α and coincubated with 1 mg/mL A. fumigatus heat-killed hyphae. Data are shown as mean ± SD, n = 3; data are representative from at least 3 independent experiments. (C) Neutrophils were treated with 0.3 μM IBT for 30 min followed by 5 ng/mL TNF-α for the time indicated. To better visualize the starting point of ROS production (black dotted line, 20 min), only the trend but not the time points are shown. (D) Neutrophils were treated with DMSO or 0.3 μM IBT for 30 min followed by 1 μg/mL actD for 15 min and by 5 ng/mL TNF-α for 1h. ROS production was measured after stimulation with 1 mg/mL A. fumigatus heat-killed hyphae. The black dotted line represents the starting point of ROS production (15 min) upon stimulation with A. fumigatus for treatments containing actD.

Figure 6. TNF-α compensated for immune defects against A. fumigatus in neutrophils from IBT-treated patients.

Human neutrophils from IBT-treated patients or healthy donors were incubated for 4h with TNF-α and coincubated with A. fumigatus B5233 strain for all figure panels. (A) Neutrophils were incubated with A. fumigatus (MOI: 0.25) for 5h, and metabolic activity was estimated by resazurin-based assay. Data are shown as the percentage of A. fumigatus killing efficiency corresponding to neutrophils from each IBT-treated patient. Data are shown as mean ± SD, n = 5. (B) Neutrophils were incubated with 1 mg/mL A. fumigatus heat-killed hyphae. ROS production was measured by chemiluminescence using lucigenin. Data represent normalized ROS production from IBT-patient neutrophils to ROS production from healthy donors, per patient. Data are shown as mean ± SD, n = 4. (C) Neutrophils were coincubated with labeled A. fumigatus swollen spores (MOI: 10). The displayed percentage of phagocytic neutrophils (CD45-AF700⁺CD66b-APC⁺conidia-AF488⁺) was estimated based on the total number of viable neutrophils (CD45-AF700⁺CD66b-APC⁺). At minimum, 10,000 viable CD66b-APC⁺ events were recorded. Data represent the percentage of phagocytic neutrophils for neutrophils from each IBT-treated patient. Because of limits placed on peripheral blood draws for these patients, not all assays were performed on the 5 patients. Data are shown as mean ± SD, n = 3. **P < 0.01; ***P < 0.001.