Fig. 5: DPP4 is the most strongly differentially expressed gene between P and NP cells.
From: Image-seq: spatially resolved single-cell sequencing guided by in situ and in vivo imaging
a, Differentially expressed genes between P and NP cells calculated using DESeq2 (ref. 60), which determines statistical significance using the Wald test and uses an interpretation of the Benjamini–Hochberg method to control the false discovery rate. The vertical dashed lines show the cut-off for gene filtering (log2FoldChange 2 and −2) and the horizontal dashed line signifies an adjusted P value of 0.01. b, Percentage of leukemia cells expressing DPP4 during disease progression, as analyzed using cell surface staining and flow cytometry. DPP4+ cells are defined as DPP4high and DPP4int cells according to the gating strategy in Extended Data Fig. 8c (day 4 n = 2, day 3 and 4 weeks n = 3, days 9–13 and 2 weeks n = 3, days 5–7 n = 5 biological replicates). Error bars denote standard deviation. Data were collected from three independent experiments. c, DPP4 antibody labeling detected using intravital microscopy after i.v. injection of DPP4-AF568 antibody. Ratio of red (that is, DPP4-AF568) to green (that is, AML-GFP) fluorescence is plotted for P (5 biological replicates) and NP cells (25 biological replicates), as well as control P (6 biological replicates) and NP (14 biological replicates) cells labeled with isotype antibody (IT). Statistical significance was determined using a two-tailed Mann–Whitney test. Data were collected from three independent mice. d, Examples of intravital multiphoton microscopy images used for the analysis in c. Red (DPP4-AF568), green (HA9M1-GFP), gray (bone SHG signal) channels and merged images are shown at a single z plane. Images of P cells have red borders; images of NP cells have green borders. Scale bars, 50 μm. Imaging of DPP4 antibody staining: 25 biological replicates with NP cells and 5 biological replicates with P cells (two independent mice). For b and d the definition of P/NP/IM is based on cell number at day 3. e,f, Cell cycle analysis on day 7 after transplantation for DPP4+ and DPP4− AML cells by Ki-67 staining in HA9M1 cells (e) and an MLL-AF9 leukemia model (f) (HA9M1, n = 8; MLL-AF9, n = 3 biological replicates). Statistical significance was determined by unpaired, two-sided Student’s t-test, and the error bars denote s.e.m. Data were collected from two independent experiments. g, Overlap of the top 300 DPP4-correlated genes in the Oregon Health and Science University (OHSU)48 and Firehose49 datasets plotted as a Venn diagram. h, Top GO terms for overlapping genes from g. The statistical analysis was done using a hypergeometric test. HA9M1, Hoxa9/Meis1-Ubiquitin-c-GFP.
