Correlating Chemical Sensitivity with Low Level Activation of Mechanotransduction Pathways in Hematologic Malignancies

Abstract

Large-scale screening has revealed that human hematopoietic cancer cell lines are generally more sensitive to various classes of drugs than cell lines established from solid tumors. A detailed examination of data in the Cancer Therapeutics Response Portal (http://portals.broadinstitute.org/ctrp/) suggests that this enhanced sensitivity is due to lower basal levels of activation of TAZ-TEAD mechanotransduction pathways in hematopoietic versus non-hematopoietic cells. Translation inhibitors such as omacetaxine mepesuccinate (homoharringtonine) fall into this category of hematopoietic-selective compounds. Moreover, additional molecular determinants of sensitivity suggest that homoharringtonine might show therapeutic efficacy in certain patients with advanced hematologic malignancies despite activation of these pathways.

Keywords: Omacetaxine mepesuccinate (homoharringtonine); RPL3; TAZ-TEAD; TNFRSF12A.

Fig. 1. Human CCLs with increased RPL3 expression exhibit enhanced sensitivity to the translation inhibitor omacetaxine mepesuccinate (homoharringtonine)

A. Expression-sensitivity correlations for homoharringtonine and RPL3 expression for hematopoietic and non-hematopoietic CCLs in the CTRP. Hematopoietic CCLs are highlighted in red. B. Expression-sensitivity correlations for homoharringtonine and RPL3 expression for non-hematopoietic CCLs in the CTRP. C. Expression-sensitivity correlations for homoharringtonine and RPL3 expression for hematopoietic CCLs in the CTRP. Expression-sensitivity correlations were determined using the CellMinerCDB web application. AUC, area under the concentration-response curve; r, Pearson expression-sensitivity correlation coefficient.

Fig. 2. Human CCLs with low level TNFRSF12A expression exhibit enhanced sensitivity to the translation inhibitors omacetaxine mepesuccinate (homoharringtonine) and CR-1-31B

A. Expression-sensitivity correlations for homoharringtonine and TNFRSF12A expression for hematopoietic and non-hematopoietic CCLs in the CTRP. Hematopoietic CCLs are highlighted in red. B. Expression-sensitivity correlations for CR-1-31B and TNFRSF12A expression for hematopoietic and non-hematopoietic CCLs in the CTRP. Hematopoietic CCLs are highlighted in red. Expression-sensitivity correlations were determined using the CellMinerCDB web application. AUC, area under the concentration-response curve; r, Pearson expression-sensitivity correlation coefficient.

Fig. 3. Relative expression levels of TNFRSF12A, ITGB5 and CRIM1 in hematopoietic and non-hematopoietic CCLs in the CCLE

Box-and-whisker plots show the transcript distribution levels for each lineage, ordered by the median expression level (line), the inter-quartile range (box) and up to 1.5× the inter-quartile range (bars). Sample numbers (n) are indicated in parentheses. Note that some non-hematopoietic CCLs that were observed to be sensitive to many of the small molecule perturbagens to which hematopoietic CCLs are sensitive (e.g., CCLs from the neuroblastoma lineage), also generally express lower levels of these transcripts. RMA, robust-multi-array-average values.

Fig. 4. Box-and-whisker plots of 18,543 correlation coefficients of transcript levels to omacetaxine mepesuccinate (homoharringtonine; left) and CR-1-31B (right) sensitivity

Expression-sensitivity correlations were determined across all CCLs (any primary site/subtype and any growth mode) using the CTRP resource. Outlier points represent 144 transcripts for homoharringtonine (z score > 9) and 138 transcripts for CR-1-31B (z score > 11). TNFRSF12A is highlighted as are YAP1, WWTR1/TAZ, TEAD1, TJP1, ITGB5 and CRIM1. r, Pearson expression-sensitivity correlation coefficient. z, z(r)-scored Pearson expression-sensitivity correlation coefficient. See Table S1 for the complete lists of outlier transcripts.