Figure 1.
Role of autophagy for the training of monocytes.
(a) Transcriptome profiling and pathway analysis of β-glucan training of monocytes compared to LPS stimulation. Factorial design analysis was performed on genes in each K-means cluster to assess significance of response differences elicited by LPS and β-glucan (Benjamini-Hochberg (BH)-adjusted p<0.05). The signal∶noise ratio is shown as heatmaps. Functional enrichment (or molecular concept) map was generated for genes exhibiting significantly weaker LPS response relative to β-glucan response. This map summarizes the extent of mutual overlap between gene sets and identifies a cluster of strongly connected gene sets that are enriched among genes showing stronger β-glucan response. Only enriched gene sets in the significant range with gene set enrichment score (−Log10(p)>1.3; p<0.05) are shown. Nodes denote enriched gene sets or “annotation terms/categories”, assembled from (K) KEGG pathways, (G) Gene Ontology, (P) Panther pathways, (R) Reactome. Node size corresponds to the number of gene members in each gene set. Node color denotes the gene set enrichment score. Please refer to graphical legend (boxed) in figure. The extent of mutually overlapping genes between gene sets is represented by thickness and color intensity of edges connecting nodes. The overlap score is the average of the Jaccard and Overlap coefficients. Strongly connected network components were identified using Tarjan's algorithm. Important ubiquitin-related processes in map are highlighted. (b) Diagram showing the course of the in vitro preincubation experiment. (c–f) BCG (c–d) or β-glucan (e–f) training in vitro in the presence or absence of 3MA using freshly isolated human monocytes and different stimuli for restimulation (LPS, B. burgdorferi). *P<0.05, **P<0.01.
Figure 2.
Polymorphisms in ATG2B or ATG5 diminish the training capacity of human monocytes.
(a–i) Blood was collected from volunteers and genotyped for ATG2B rs3759601 (a–f) and ATG5 rs2245214 (g–i). Human monocytes were trained with BCG for 24 h, washed and incubated in RPMI (10% human serum) for 6 d, after which they were restimulated for 24 h with a second stimulus (LPS, Bb, or C. albicans). Proinflammatory cytokine production (IL-6 and TNF-α) was assessed by ELISA in the supernatants. (j–k) PBMCs isolated from volunteers carrying different genotypes for SNPs rs3759601 or rs2245214 were stimulated for 24 h with LPS or B. burgdorferi. IL-6 was measured in the supernatants by ELISA. (l) Human monocytes carrying different genotypes for SNP rs3759601 were trained with BCG for 4 h. Expression of ATG2B was assessed by qPCR *P<0.05, **P<0.01.
Figure 3.
Autophagy affected by SNP in ATG2B.
(a–b) Monocytes genotyped for ATG2B rs3759601 were seeded on coverslips, and stimulated with BCG. After 1 hour of stimulation, cells were fixed and stained with an antibody against LC3. Slides were analyzed by confocal microscopy. Data are representative for 3 experiments.
Figure 4.
SNP in ATG2B affects the efficacy of in vivo BCG-induced trained immunity.
(a–b) Monocytes isolated before and 3 months after vaccination of 16 naïve (nonexposed) volunteers were stimulated in vitro with B. burgdorferi. Proinflammatory cytokine production (IL-1β [a], TNF-α [b]) was assessed by ELISA in the supernatants. (c–d) Kaplan-Meier curves for recurrence-free (c) and progression-free (d) survival according to rs3759601 SNP genotype of 192 patients suffering from non-muscle invasive bladder cancer treated with ≥6 intravesical instillations of BCG. Each drop in a probability curve indicates one or more events in that group. Vertical lines indicate censored patients, i.e. those who reached the end of their follow-up without experiencing the event. Total number of patients and number of events (between brackets) per genotype category are indicated next to the corresponding curve. Numbers of patients at risk at selected time points for each genotype category are given below the plots. (e–g) Monocytes of bladder cancer patients isolated before and after 6 intravesical BCG instillations as initial treatment were stimulated in vitro with LPS. Proinflammatory cytokine production (IL-1β [e], IL-6 [f], TNF-α [g]) was assessed by ELISA in the supernatants *P<0.05, **P<0.01.
Figure 5.
Impairment of autophagy decreases trimethylation at H3K4 in human monocytes.
ChIP analysis of the enrichment of H3K4me3 at the promoter of (A) TNF-α and (B) IL-6 in human monocytes isolated from volunteers carrying the major variant (GG) or minor variant (CC) alleles for ATG2B after training with BCG. ChIP analysis of the enrichment of H3K4me3 at the promoter of (C) TNF-α and (D) IL-6 in human monocytes trained with BCG in the presence or absence of 3MA *p<0.05, **p<0.01.