Deciphering intratumoral heterogeneity using integrated clonal tracking and single-cell transcriptome analyses

Cellular heterogeneity is a major cause of treatment resistance in cancer. Despite recent advances in single-cell genomic and transcriptomic sequencing, it remains difficult to relate measured molecular profiles to the cellular activities underlying cancer. Here, we present an integrated experimental system that connects single cell gene expression to heterogeneous cancer cell growth, metastasis, and treatment response. Our system integrates single cell transcriptome profiling with DNA barcode based clonal tracking in patient-derived xenograft models. We show that leukemia cells exhibiting unique gene expression respond to different chemotherapies in distinct but consistent manners across multiple mice. In addition, we uncover a form of leukemia expansion that is spatially confined to the bone marrow of single anatomical sites and driven by cells with distinct gene expression. Our integrated experimental system can interrogate the molecular and cellular basis of the intratumoral heterogeneity underlying disease progression and treatment resistance.

Clonal abundance across different tissues and organs of secondary recipient female mice that were xenografted with the ALL04 sample. Significant extramedullary expansion in the ovaries was observed in all of these mice. Each color represents one distinct genetic barcode corresponding to a leukemia clone. Notably, the examined tissues were saturated with human cells. Spatially confined clonal expansion in the bone marrow was undetectable in these secondary recipient mice, in contrast to the primary recipient mice ( Fig. 2c and Supplementary Fig. 4d). Other recipient mice receiving the ALL04 sample, including the primary recipients, were male. Patient ALL04 was also a male (Supplementary Table 1). Fig. 6 Clonal distribution across different tissues and organs in primary recipient mice. Each color represents one distinct genetic barcode corresponding to a leukemia clone. Red arrows highlight the bone marrow that exhibited different clonal compositions. Shown are all experimental mice, in addition to the one in Fig. 2c, that received the following patient samples: a ALL20. b ALL06. c the naïve stage of patient JFK93. d ALL04. e the relapsed stage of patient JFK88. f the naïve stage of patient JFK88. Supplementary Fig. 11 Verifying CRISPR/Cas9 gene knockout by T7E1 assays. Shown are representative gel images of T7E1-treated PCR products amplified from the sgRNA target sites. PCR templates were derived from REH and KOPN-8 cells that were transduced with Cas9 and pooled sgRNAs targeting negative control luciferase (LUC) gene or S100A16 gene. G1-G5, sgRNAs 1-5. Ladder = 1000bp, 900bp, 800bp, 700bp, 600bp, 500bp, 400bp, 300bp, 200bp, 100bp. Two independent experiments were performed with each experiment using a different cell line. The images depict both results.

Supplementary
Supplemental Fig. 12. Migration of one out of three B-ALL cell lines is altered by the knockout of S100A16 gene. S100A16 and a negative control luciferase (LUC) gene were knocked out using CRISPR/Cas9 technology in human B-ALL cell lines (REH, . Migration of the B-ALL cells was analyzed after 12 or 24 hours of incubation (24 hours for REH and KOPN8, 12 hours for NALM6). Three independent experiments (n=12 for REH and KOPN8, n=8 for NALM6). Data shown as mean ± SEM. **** P value = 2.85x10 -6 , by two-sided t-test without adjustment.
Supplementary Fig. 13 Growth of human B-ALL cells is not altered by the knockout of S100A16 gene. S100A16 and a negative control luciferase (LUC) gene were knocked out using CRISPR/Cas9 technology in human B-ALL cell lines (REH, KOPN-8 and NALM6). The MTS assay was used to quantify viable cells at various time points during the cell culture in complete media. Data from one representative experiment out of three independent experiments are shown. Cell growth was measured in replicates (n=8) at each timepoint. Data shown as mean ± SEM. Supplementary Fig. 21 ALL04 clones that responded significantly differently to different chemo treatments. Each plot represents data from one clone, and each marker represents data from one tissue and one mouse. Data from 25 mice were used in each graph. MNC -mononuclear cells; *** P < 0.001; ** P < 0.01; * P < 0.05. P values were calculated by two-sided Kruskal Wallace test with Bonferroni correction. treatments. Each plot represents data from one clone, and each marker represents data from one tissue and one mouse. Data from 23 mice were used in each graph. MNC -mononuclear cells; *** P < 0.001; ** P < 0.01; * P < 0.05. P values were calculated by two-sided Kruskal Wallace test with Bonferroni correction.
Supplementary Fig. 23 ALL20 clones that responded significantly differently to different chemo treatments. Each plot represents data from one clone, and each marker represents data from one tissue and one mouse. Data from 50 mice and 49 mice were used for Clone 0 and Clone 2 respectively, and data from 27 mice were used in the remaining graphs, as some clones did not engraft in every mouse. MNC -mononuclear cells; *** P < 0.001; ** P < 0.01; * P < 0.05. P values were calculated by two-sided Kruskal Wallace test with Bonferroni correction.
Supplementary Fig. 24 ALL06 clones that responded significantly better to combination therapy than to intensive therapy. Each marker represents data from one tissue and one mouse. Data from 23 mice were used in each graph . ** P < 0.01; * P < 0.05. P values were calculated by two-sided Kruskal Wallace test with Bonferroni correction.
Supplementary Fig. 25 Comparing the gene expression of ALL04 clones that responded significantly better to combination therapy than to intensive therapy with all other ALL04 clones (Fig. 5c). Shown are False Positive Scores (FPS) and P values of differential expression analysis. scRNA-seq data of 41 cells from the selected clones and 958 cells from the rest of the clones were used in the analysis. P values were calculated by the one-sided Mann Whitney U-test and adjusted using both the experimental data and the scramble data, see the Methods section for detail. Each dot represents a gene. Dashed lines show 0.05 for FPS and P values.
Supplementary Fig. 26 Comparing the gene expression of ALL04 clones that responded better to maintenance therapy than to intensive therapy with all other ALL04 clones (Fig. 5d). Shown are False Positive Scores (FPS) and P values of differential expression analysis. scRNA-seq data of 97 cells from the selected clones and 902 cells from the rest of the clones were used in the analysis. P values were calculated by the one-sided Mann Whitney U-test and adjusted using both the experimental data and the scramble data, see the Methods section for detail. Each dot represents a gene. Dashed lines show 0.05 for FPS and P values.
Supplementary Fig. 27 Comparing the gene expression of ALL20 clones that responded better to intensive therapy than to maintenance therapy with all other ALL20 clones (Fig. 5e). Shown are False Positive Scores (FPS) and P values of differential expression analysis. scRNA-seq data of 63 cells from the selected clones and 1038 cells from the rest of the clones were used in the analysis. Each dot represents a gene. P values were calculated by the one-sided Mann Whitney U-test and adjusted using both the experimental data and the scramble data, see the Methods section for detail. Dashed lines show 0.05 for FPS and P values.
Supplementary Fig. 28 In vitro culture of B-ALL cells. Equal numbers of leukemia cells acquired at different disease stages, naïve (N) and relapsed (R), from two patients (JFK88 and JFK93) were cultured in vitro during the lentiviral transduction. a 24-hours after plating. b 48-hours after plating right before transplantation. White bar represents a 10 μm scale bar. Representative images are shown from an experiment with five replicates per sample. b a Supplementary Fig. 29 Engraftment of human B-ALL cells. The engraftment rate was calculated based on the number of donor cells and the number of clones detected in the peripheral blood, bone marrow, spleen and enlarged extramedullary organs, adjusted for GFP%. Each dot represents data from one mouse.
Supplementary Fig. 30 Transduction of human B-ALL cells. Each line represents data from one recipient mouse. An unpaired two-tailed t-test was performed to measure the statistical significance for JFK88 at day 81 post transplantation. ***P value = 0.000015.