Knockdown of a regulatory barcode shifts macrophage polarization destination from M1 to M2 and increases pathogen burden upon S. aureus infection
- IISc Mathematics Initiative, Indian Institute of Science, India
- Department of Microbiology and Cell Biology, Indian Institute of Science, India
- Department of Biochemistry, Indian Institute of Science, India
- Centre for Biosystems Science and Engineering, Indian Institute of Science, India
Figures
Figure 1 with 3 supplements
A clustergram showing K = 13 for samples spanning the macrophage polarization spectrum and RF barcodes characteristic of each polarization state.
The clustering pattern was obtained by using M3C, a widely used agnostic clustering method based on Monte Carlo referencing that uses stability selection to estimate the number of clusters. (A) Cluster–member associations in the consensus matrix for K = 13. Clusters with a single member are starred. Clusters 2 and 12 are singletons. (B) A heatmap showing pathways commonly enriched in inflammation and other immunological processes. The pathways marked in red are upregulated, while those marked in blue are downregulated. The metric chosen was the negative logarithm of the p-value for each associated pathway. (C) RF barcodes which are characteristic of the 13 different clusters. Each barcode is a heatmap of the differential expression profile of the identified transcription factors (as compared to the unstimulated Mo state) in each cluster. The length of the barcode (which is the number of RFs in each cluster) is indicated within parentheses. Barcodes of members within a cluster resemble each other extensively and are shown together.
Figure 1—figure supplement 1
Schematic diagram explaining the ResponseNet and EpiTracer algorithms.
Figure 1—figure supplement 2
M3C-based identification of the optimal number of clusters for the entire transcriptome.
Clustering pattern based on the differential profile of the entire transcriptome (n=12,164) showed thirteen different clusters spanning the macrophage polarization spectrum based on the (A) RCSI (spiked at K=13) with (B) significant p-value.
Figure 1—figure supplement 3
Clustering based on RF-set1 vs. entire transcriptome.
Entanglement score ranges from 0 to 1. Lower the score, better the alignment.
Figure 2 with 2 supplements
Switching factors that guide M1 to M2 destination.
(A) The regulatory core for the M1 cluster (C11). (B) The regulatory core for the M2 cluster (C3). The core networks contain the top perturbed regulatory interactions common across C11 members (stimulants: TPP, sLPS, and sLPS + IC) and C3 members (stimulants: IL4 and IL13), respectively. (C) A heatmap of the genes in the barcode that exhibits a reversal in the fold change pattern between M1 (C11) and M2 (C3). RFs that are selected for experimental validation are starred.
Figure 2—figure supplement 1
Switching factors that guide M2 to M1 destination.
A heatmap of the genes in the barcode that exhibits a reversal in the fold change pattern between M2 (C3) and M1 (C11).
Figure 2—figure supplement 2
Principal Component Analysis of the thirteen clusters.
(A) PCA plot segregating the 13 clusters based on the fold changes of hallmark genes of inflammation derived from MSigDB (Liberzon et al., 2011) . (B) PCA plot segregating the 13 clusters based on the fold changes of the NCB gene panel.
Figure 3 with 4 supplements
siRNA knockdown of NFE2L2, CEBPB, and BCL3 in LPS infected THP-1-derived macrophage cell line attenuates macrophages polarizing to C11 state (M1).
(A) Schema describing the experimental design for testing the role of the 3-gene set NCB, predicted to be a switching factor combination between M1 and M2 destinations. The LPS and IL4 treatment conditions are included to verify the experimental setup. The non-targeting siRNA is used as a negative control while siSTAT4 is used as a positive control. siNCB is a knock-down of the 3-gene set (NCB) which is evaluated in this study. (B) Gene expression changes of selected RFs and known markers of M1 (IL1B) and M2 (IL10) in THP1 cell lines upon LPS and IL4 treatment (for 72 hr), respectively. Color coding scheme is similar to that in (A). THP1 monocytes were treated with PMA (20 ng/ml) for 16 hr. Post 24 hr, LPS (100 ng/ml) and IL-4 (1000 IU) treatment was given for 72 hr. Two-way ANOVA (ns: p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001; N = 3 biological replicates) is used for estimating significance in gene expression fold change normalized over untreated cells. (C) Gene expression changes in the M1 markers upon siRNA knockdown of STAT4 (positive control). THP1 monocytes were treated with PMA (20 ng/ml) for 16 hr. STAT4-specific siRNA was transfected. Post 8 hr, LPS (100 ng/ml) treatment was given for 72 hr. Gene expression changes in M1 markers were quantified (two-way ANOVA; ns: p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001; N = 3 biological replicates; NT: non-targeting siRNA). (D) Gene expression changes in the M1 and M2 markers upon siRNA knockdown of the NCB set. THP1 monocytes were treated with PMA (20 ng/ml) for 16 hr. NCB (NFE2L2, CEBPB, and BCL3) specific siRNA was transfected. Post 8 hr, LPS (100 ng/ml) treatment was given for 72 hr. Gene expression changes of M1 and M2 markers were quantified (two-way ANOVA; ns: p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001; N = 3 biological replicates; NT: non-targeting siRNA).
Figure 3—figure supplement 1
Validation of STAT4 siRNA.
THP1 monocytes were treated with PMA (20ng/ml) for 16h. STAT4 specific siRNA was transfected. Post 8hr, LPS (100ng/ml) treatment was given for 72hr. Transcript levels of STAT4 were assessed. {Two-way ANOVA,P > 0.05 :ns, P < 0.05 :*, P<0.01 :**, P<0.001 :***, N=3, NT: non-targeting siRNA}.
Figure 3—figure supplement 2
Validation of NCB siRNA.
THP1 monocytes were treated with PMA (20ng/ml) for 16h. NCB (NFE2L2, CEBPB , BCL3) specific siRNA was transfected. Post 8hr, LPS (100ng/ml) treatment was given for 72hr. Transcript levels of STAT4 were assessed. {Two-way ANOVA,P >0.05 :ns, P < 0.05 :*, P<0.01 :**, P<0.001 :***, N=3 biological replicates, NT: non-targeting siRNA}.
Figure 3—figure supplement 3
Validation of CEBPB, NFE2L2, and BCL3 siRNA.
(A) Validation of CEBPB siRNA. Gene expression changes of selected known markers of M1 (CXCL2, IL1B, iNOS and SOCS3) and M2 (ARG1, IL10, MRC1 and TGFB) upon siRNA knockdown of CEBPB. Color coding scheme is similar to that in Figure 2A. (B) Validation of NFE2L2 siRNA. Gene expression changes of known M1 and M2 markers upon siRNA knockdown of NFE2L2. (C) Validation of BCL3 siRNA. Gene expression changes of known M1 and M2 markers upon siRNA knockdown of BCL3. {Two-way ANOVA,P > 0.05 :ns, P < 0.05 :*, P<0.01 :**, P<0.001 :***, N=3 biological replicates, NT: non-targeting siRNA}.
Figure 3—figure supplement 4
Gene expression change of the NCB gene panel during stimulation of M0 cells with M1 and M2 associated stimulants respectively.
(A) Gene expression change of NFE2L2. (B) Gene expression change of CEBPB. (C) Gene expression change of BCL3.
(A) Gene expression change of NFE2L2. (B) Gene expression change of CEBPB. 967 (C) Gene expression change of BCL3.
Figure 4 with 4 supplements
siRNA knockdown of NCB (NFE2L2, CEBPB, and BCL3) in S. aureus infected macrophage increases infectivity.
(A) Schema describing the experimental design. (B) Effect of the NCB set knockdown in the S. aureus infection model. siRNA knockdown of NFE2L2, CEBPB, and BCL3 in THP-1 cells infected with S. aureus leads to a significant decrease in the gene expression of the M1 markers (CXCL2, IL1B, iNOS, and SOCS3) and an increase in the expression of the M2 markers (TGFβ and IL10). Color coding is based on the schema shown in (A). THP1 monocytes were treated with PMA (20 ng/ml) for 16 hr. NFE2L2-, CEBPB-, and BCL3-specific siRNAs were transfected. Post 24 hr, S. aureus infection at 1:10 MOI was given for 24 hr. Gene expression levels of M1 and M2 markers were assessed (two-way ANOVA; ns: p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001; N = 3 biological replicates; NT: non-targeting siRNA). (C) In vitro S. aureus CFU assay. THP-1 cells were differentiated to macrophages with PMA for 16 hr. Post 24 hr, siRNA specific for NFE2L2, BCL3, and CEBPB were transfected. S. aureus infection (MOI 10) was given for 2 hr prior to 2 hr gentamicin treatment. CFU assay was performed at 0 and 24 hr (two-way ANOVA; ns: p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001; N = 3 biological replicates; NCB: NFE2L2, CEBPB, and BCL3). CFU: colony-forming units.
Figure 4—figure supplement 1
S. aureus infected macrophages polarize to M1.
Heatmap showing the differential transcriptome profile of 16 TFs in C11 barcode (M1) and M1 markers (IL1B , SOCS3, CXCL2) in S. aureus infected peripheral macrophages at 8hr, 24 hr and 48 hr.
Figure 4—figure supplement 2
Gene expression changes in M1 and M2 markers upon S. aureus infection at different time points.
THP1 monocytes were treated with PMA (20ng/ml) for 16hr. Post 24hr infection with S. aureus at 1:10MOI was given for mentioned time duration. Transcript levels of (A) M1 markers and (B) M2 markers were assessed. {Two way ANOVA, P > 0.05 :ns, P < 0.05 :*, P<0.01 :**, P<0.001 :***, N=3 biological replicates}.
Figure 4—figure supplement 3
Gene expression changes in STAT4 and NCB-set upon S. aureus infection at different time points.
THP1 monocytes were treated with PMA (20ng/ml) for 16hr. Post 24hr, infection with S. aureus at 1:10MOI was given for mentioned time duration. Transcript levels of STAT4, NFE2L2, BCL3 adn CEBPB were assessed. {Two way ANOVA,P > 0.05 :ns, P < 0.05 :*, P<0.01 :**, P<0.001 :***, N=3 biological replicates}.
Figure 4—figure supplement 4
NCB (NFE2L2, CEBPB, and BCL3) siRNA validation in S. aureus data.
THP1 monocytes were treated with PMA (20ng/ml) for 16hr. NFE2L2, CEBPB and BCL3 specific siRNAs were transfected. Post 24hr, S. aureus infection at 1:10 MOI was given for 24hr. Transcript levels of mentioned transcription factors were assessed. {Two-way ANOVA, **, P<0.001 :***, N=3, NT: non-targeting siRNA}.
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Strain, strain background (Staphylococcus aureus Cowan 1) | Staphylococcus aureus Cowan 1 | MTCC, IMTECH, Chandigarh | MTCC 902 | |
| Cell line (Homo sapiens) | THP-1 | NCCS, Pune | RRID:CVCL_0006 | |
| Chemical compound, drug | Phosphate-buffered saline (PBS), pH 7.4 | Gibco | Cat# 10010023 | |
| Chemical compound, drug | Fetal bovine serum (FBS) | Gibco, Thermo Fisher Scientific | Cat# 10270106 | |
| Chemical compound, drug | RPMI 1640 Medium, powder | Gibco, Thermo Fisher Scientific | Cat# 31800022 | |
| Chemical compound, drug | Phorbol 12-myristate 13-acetate (PMA) | Sigma-Aldrich | Cat# P8139 | |
| Chemical compound, drug | Polyethyleneimine | Sigma-Aldrich | Cat# 40872-7 | |
| Chemical compound, drug | Opti-MEM Reduced Serum Medium | Gibco | Cat# 31985070 | |
| Chemical compound, drug | TRI Reagent | Sigma-Aldrich | Cat# T9424 | |
| Chemical compound, drug | Lipopolysaccharide | Sigma-Aldrich | Cat# 8630 | |
| Chemical compound, drug | IL-4 | R&D Systems | Cat# 204-IL | |
| Chemical compound, drug | Gentamicin | HIMEDIA | Cat# SD195 | |
| Chemical compound, drug | Triton X-100 | Sigma-Aldrich | Cat# T8787 | |
| Chemical compound, drug | Luria Broth | HIMEDIA | Cat# G575 | |
| Chemical compound, drug | Agar powder, Bacteriological grade | HIMEDIA | Cat# 9002-18-0 | |
| Sequence-based reagent | Primers for qRT-PCR | See Supplementary file 1 | See Supplementary file 1 | |
| Sequence-based reagent | siRNA Negative control | Eurogentec | See Supplementary file 1 | |
| Sequence-based reagent | siGENOME Human NFE2L2 | Dharmacon | Cat# M-003755-02-0005 | |
| Sequence-based reagent | siGENOME Human BCL3 | Dharmacon | Cat# M-003874-02-0005 | |
| Sequence-based reagent | siGENOME Human CEBPB | Dharmacon | Cat# M-006423-03-0005 | |
| Sequence-based reagent | siGENOME Human STAT4 | Dharmacon | Cat# M-011784-01-0005 | |
| Software, algorithm | GraphPad Prism 10.0 software | GraphPad | RRID:SCR_002798 | |
| Software, algorithm | Microsoft Excel | Microsoft | RRID:SCR_016137 | |
| Software, algorithm | Microsoft PowerPoint | Microsoft | RRID:SCR_023631 |
Additional files
-
Supplementary file 1
In vitro stimulant data.
- https://cdn.elifesciences.org/articles/91547/elife-91547-supp1-v1.docx
-
Supplementary file 2
The list of primers used for the qRT-PCR.
- https://cdn.elifesciences.org/articles/91547/elife-91547-supp2-v1.docx
-
Supplementary file 3
The list of clusters, associated members, network details of regulatory cores, and the number of RFs in the barcodes.
- https://cdn.elifesciences.org/articles/91547/elife-91547-supp3-v1.docx
-
Supplementary file 4
Data files.
Data file S1. The human gene regulatory network. Data file S2: Twenty-eight different subnetworks capturing highest ranked regulatory perturbations. Data file S3: RF-set1 and RF-set2.
- https://cdn.elifesciences.org/articles/91547/elife-91547-supp4-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/91547/elife-91547-mdarchecklist1-v1.docx
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Knockdown of a regulatory barcode shifts macrophage polarization destination from M1 to M2 and increases pathogen burden upon S. aureus infection
eLife 12:RP91547.
https://doi.org/10.7554/eLife.91547.3
