Unveiling the signaling network of FLT3-ITD AML improves drug sensitivity prediction
- Cellular and Molecular Biology, Department of Biology, University of Rome Tor Vergata, Italy
- Department of Biology, University of Rome Tor Vergata, Italy
- Sezione di Ematologia, Dipartimento di Scienze Radiologiche ed Ematologiche, Università Cattolica del Sacro Cuore, Italy
- Dipartimento di Diagnostica per Immagini, Radioterapia Oncologica ed Ematologia, Fondazione Policlinico A. Gemelli IRCCS, Italy
- Health Campus for Inflammation, Immunity and Infection (GCI3), Otto-von-Guericke University of Magdeburg, Germany
- Department of Hematology and Oncology, Otto-von-Guericke University of Magdeburg, Germany
- Institute of Molecular and Clinical Immunology, Otto-von-Guericke University of Magdeburg, Germany
- Department of Biology, Fondazione Human Technopole, Italy
- Telethon Institute of Genetics and Medicine (TIGEM), Italy
Figures
Figure 1
Summary of the experimental strategy.
(A) Manual curation of FLT3-internal tandem duplication (ITD)-specific prior knowledge network (PKN). (B) Multiparametric analysis of signaling perturbations in tyrosine kinase inhibitor …
Figure 2 with 3 supplements
Generation of the training dataset.
(A) Schematic representation of the FLT3-internal tandem duplication (ITD) prior knowledge network (PKN) manual curation, integration of data-driven edges, and manual integration of RTKs pathways …
Figure 2—figure supplement 1
FLT3-internal tandem duplication (ITD) manually curated prior knowledge network (PKN).
Network representation of the PKN, each node represents a protein, a cytokine (green) or a small molecule inhibitor (yellow). The proteins are colored following the CNO pipeline graphics: target …
Figure 2—figure supplement 2
Global overview of multiparametric data.
(A–B) Heatmap showing the Pearson correlation coefficients between the different biological replicates in FLT3 ITD-JMD cells (A) and FLT3 ITD-TKD cells (B). (C) Heatmap of the complete dataset …
Figure 2—figure supplement 3
Normalization of analytes activity through Hill curves.
Experimental data were normalized and scaled from 0 to 1 using analyte-specific Hill functions. Raw data are reported as triangles, normalized data, and squares. FLT3 ITD-JMD (A) and -TKD (B) plots …
Figure 3 with 2 supplements
Optimized Boolean models recapitulate the different sensitivity of FLT3ITD-JMD and FLT3ITD-TKD cells to tyrosine kinase inhibitor (TKI).
(A) Color-coded representations of the experimental activity modulation (T90–T0) of sentinel proteins used to train the two Boolean models (upper panel) and the average prediction of protein …
Figure 3—figure supplement 1
Overview of the optimized Boolean models.
(A) Heatmaps represent the absolute value of the difference between model simulation results to normalized experimental data in a color-coded from light yellow to red. Simulation results of the …
Figure 3—figure supplement 2
Validation of FLT3-internal tandem duplication (ITD)-specific models on real-world independent datasets.
(A) Comparison of our models’ steady states upon FLT3 inhibition with data from Massacci et al., 2023, paper. The agreement between the activation status of nodes in the FLT3ITD-JMD (upper table) …
Figure 4 with 1 supplement
In silico simulation of the FLT3ITD-TKD logic model allows the prediction of novel combinatorial treatment reverting tyrosine kinase inhibitor (TKI) resistance.
(A) Cartoon of the in silico simulation conditions. (B–C) Bar plot showing the in silico simulation of proliferation activation (B) and apoptosis inhibition (C) levels in untreated and FLT3i …
-
Figure 4—source data 1
Original files for the western blot analysis in Figure 4G of phospho-JNK(T183/Y185), JNK, phospho-CDK1 (Y15), phospho-CDK1 (T161), CDK1, phospho-CDK2(T160), CDK2, CyclinB1, CycinE2, and Vinculin.
- https://cdn.elifesciences.org/articles/90532/elife-90532-fig4-data1-v1.zip
-
Figure 4—source data 2
PDF containing Figure 4G and original scans for the western blot analysis of phospho-JNK(T183/Y185), JNK, phospho-CDK1 (Y15), phospho-CDK1 (T161), CDK1, phospho-CDK2(T160), CDK2, CyclinB1, CycinE2, and Vinculin, with highlighted band.
- https://cdn.elifesciences.org/articles/90532/elife-90532-fig4-data2-v1.pdf
Figure 4—figure supplement 1
Quantification of western blot analysis of cell cycle proteins.
(A–B) Bar plot showing the absorbance values at 595 nm normalized on control conditions for FLT3 ITD-JMD (yellow) and -TKD (blue) cells treated with Midostaurin and/or UO126 (MEK inhibitor) (A) and …
Figure 5 with 2 supplements
FLT3-internal tandem duplication (ITD) patient-specific Boolean models.
(A) Schematic representation of our computational approach to obtain personalized logic models. (B) Hierarchical clustering of patients according to their clinical characteristics (response to …
Figure 5—figure supplement 1
Overview of patient-specific Boolean models.
(A) Heatmap showing the mutational profile (columns) of each patient (rows). Acute myeloid leukemia (AML) driver genes are highlighted with black squares. White, blue, and red squares represent …
Figure 5—figure supplement 2
Personalized Boolean models.
(A) High-confidence Boolean models of representative patients (JMD1, JMD8, JMD-TKD1, TKD2). Nodes are color-coded according to their activity upon in silico simulation of FLT3 inhibition. (B) …
Author response image 1
Primary samples from AML patients with the FLT3ITD-TKD mutation (n=2, yellow bars) or the FLT3ITD-JMD/TKD mutation (n=3, blue bars) were exposed to Midostaurin (100nM, PKC412), and JNK inhibitor (10µM, SP600125) for 48 hours, or combinations thereof.
The specific cell death of gated AML blasts was calculated to account for treatment-unrelated spontaneous cell death. The bars on the graph represent the mean values with standard errors.
Author response image 2
Partial Figure of normalization of analytes activity through Hill curves.
Experimental data were normalized and scaled from 0 to 1 using analyte-specific Hill functions. Raw data are reported as triangles, normalized data and squares. Partial Figure representing three …
Author response image 3
Author response image 4
Author response image 5
Tables
Table 1
Small molecule inhibitors and stimuli for the multiparametric analysis.
| Inhibitors | Target | Usage | Time | Stimuli | Usage | Time |
|---|---|---|---|---|---|---|
| Midostaurin | FLT3 | 100 nM | 90 min | IGF1 | 100 ng/ml | 10 min |
| SB203580 | p38 | 10 μM | 90 min | TNFα | 10 ng/ml | 10 min |
| SP600125 | JNK | 20 μM | 90 min | |||
| Wortmannin | PI3K | 50 nM | 90 min | |||
| Rapamycin | mTOR | 100 nM | 90 min | |||
| UO126 | MEK1/2 | 15 μM | 90 min | |||
| LY2090314 | GSK3 | 20 nM | 90 min |
Table 2
xMAP analytes.
| Analytes | Cat. no. | Phosphosite measured | Activity annotation |
|---|---|---|---|
| CREB1 | 42-680MAG | Ser133 | 1 |
| ERK1/2 | 42-680MAG | Thr185/Tyr187 | 1 |
| JNK | 42-680MAG | Thr183/Tyr 185 | 1 |
| p38 | 42-680MAG | Thr180/Tyr182 | 1 |
| STAT3 | 42-680MAG | Ser727 | 1 |
| STAT5 | 42-680MAG | Tyr694/699 | 1 |
| p70S6K | 42-611MAG | Thr412 | 1 |
| RPS6 | 42-611MAG | Ser235/236 | 1 |
| MTOR | 42-611MAG | Ser2448 | 1 |
| IGF1R | 42-611MAG | Tyr1135/Tyr1136 | 1 |
| PTEN | 42-611MAG | Ser380 | –1 |
| TSC2 | 42-611MAG | Ser939 | –1 |
| GSK3A | 42-611MAG | Ser21 | –1 |
| GSK3B | 42-611MAG | Ser9 | –1 |
| β-Tubulin | 46-413MAG | Total protein | Loading control |
Additional files
-
Supplementary file 1
FLT3-internal tandem duplication (ITD) prior knowledge network (PKN), table downloaded form SIGNOR, data-driven edges integration, PKN in .sif format; regulators of phenotypes annotated by ProxPath resource.
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp1-v1.xlsx
-
Supplementary file 2
Experimental design of the multiparametric experiment of FLT3 ITD-JMD and ITD-TKD BaF3 cell lines; treatments and analytes measured; activity readout annotation.
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp2-v1.xlsx
-
Supplementary file 3
Cue-sentinel-response multiparametric dataset raw data and statistics (MILLIPLEX kit: 9plex_Cat.No.48-680MAG).
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp3-v1.xlsx
-
Supplementary file 4
Cue-sentinel-response multiparametric dataset, raw data, and statistics (MILLIPLEX kit: 11plex_Cat.No.48-611MAG).
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp4-v1.xlsx
-
Supplementary file 5
Complete cue-sentinel-response multiparametric dataset used for Boolean models building in MIDAS format, raw and normalized data.
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp5-v1.xlsx
-
Supplementary file 6
Data used for the validation of FLT3-internal tandem duplications (ITDs) Boolean models using independent resources.
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp6-v1.xlsx
-
Supplementary file 7
Panel of 262 mutations relevant to hematological malignancies analyzed in de novo acute myeloid leukemia (AML) cohort of 14 patients.
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp7-v1.xlsx
-
Supplementary file 8
Results of the getITD output for the classification of the de novo acute myeloid leukemia (AML) cohort of 14 patients.
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp8-v1.xlsx
-
Supplementary file 9
Clinical data, mutation profile, and RNAseq results of the acute myeloid leukemia (AML) patients’ cohort.
- https://cdn.elifesciences.org/articles/90532/elife-90532-supp9-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/90532/elife-90532-mdarchecklist1-v1.pdf
