Selective Inhibition of a Panel of Cancer Stem Cell Lines by Novel Glycosaminoglycan Mimetics

Abstract

PurposeTo characterize the anti‐cancer stem cell (CSC) properties of non‐saccharide glycosaminoglycan mimetics (NSGMs) and their lipid‐modified analogs in a panel of colon cancer cells.MethodFour novel NSGMs were screened for their ability to inhibit the formation of 1° spheroids in 12 intestinal tumor cells belonging to one of four consensus molecular subtypes (CMS) for intestinal cancers and were chosen to be representative of common genetic variants in colon cancer. Spheroid were grown by suspending monolayer cancer cells in stem cell media in low‐adhesion plates and the number of spheroids was measured with light microscopy. Spheroid forming cell frequency (SFF) was determined in the 1° spheroids by plating them at varying concentration 1→124 cells/well and recording 2° spheroids (present or absent) in each well using http://bioinf.wehi.edu.au/software/elda web tool. Dose‐response curves were generated by examining 1° spheroid formation following treatment with increasing concentration of NSGMs (0→5000 μM). The IC50 values for each of the NSGMs were determined using the standard sigmoidal regression equation. Inhibition of monolayers of corresponding cancer cells by NSGMs was assessed through a colorimetric MTT assay.ResultsThe MTT assay in monolayer conditions revealed no or minimal cytotoxicity with all the NSGMs at concentrations >300μM, suggesting selective targeting of CSCs. Screening for primary spheroid inhibition revealed three distinct responses concerning the lead NSGM G2.2. G2.2 showed a mean IC50 of 225 μM across all cell lines (range 134 – 252 μM). Intriguingly, CMS1, CMS2, and neuroendocrine cells were either moderately resistant or highly resistant. On the other hand, none of the CMS3 and CMS4 cell lines were highly resistant, and a majority (4/8) were sensitive to G2.2. Intriguingly, we observed a significant inverse correlation (r = −0.56, p= 0.06) between SFF and IC50 for G2.2, suggesting that the level of pre‐existing CSCs could determine G2.2's efficacy. Interestingly, the lipid‐modified analogs showed higher potency in comparison with the parent molecule G2.2. Of the three lipid‐modified analogs, G8C was the most potent inhibitor of CSC's (with a mean IC50 of 11 μM) against intestinal cancer spheroids. In fact, G8C displayed sub‐micromolar inhibition (IC50 = 0.73 μM) against the most sensitive colorectal cell line HT‐29 (CMS3). Except for one case, the resistant cell lines turned sensitive or less resistant to treatment with the three lipid analogs.ConclusionsG2.2 (and related NSGMs) selectively inhibit CSCs and its efficacy strongly correlates with the levels of pre‐existing CSCs and/or CMS subtypes 3 (KRAs activation) and 4 (TGF‐β pathway activation) providing valuable clues about their mechanism. Lipid‐modified NSGMs represent potent and selective anti‐CSCs agents with paradigm‐shifting implications for colorectal cancer treatment.Support or Funding InformationGrant support from the following sources:1P01 HL1071521R25 HL1286391R01 HL090586This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.