Targeted disruption of β-arrestin 2-mediated signaling pathways by aptamer chimeras leads to inhibition of leukemic cell growth

Abstract

β-arrestins, ubiquitous cellular scaffolding proteins that act as signaling mediators of numerous critical cellular pathways, are attractive therapeutic targets because they promote tumorigenesis in several tumor models. However, targeting scaffolding proteins with traditional small molecule drugs has been challenging. Inhibition of β-arrestin 2 with a novel aptamer impedes multiple oncogenic signaling pathways simultaneously. Additionally, delivery of the β-arrestin 2-targeting aptamer into leukemia cells through coupling to a recently described cancer cell-specific delivery aptamer, inhibits multiple β-arrestin-mediated signaling pathways known to be required for chronic myelogenous leukemia (CML) disease progression, and impairs tumorigenic growth in CML patient samples. The ability to target scaffolding proteins such as β-arrestin 2 with RNA aptamers may prove beneficial as a therapeutic strategy.

Highlights: An RNA aptamer inhibits β-arrestin 2 activity.Inhibiting β-arrestin 2 impedes multiple tumorigenic pathways simultaneously.The therapeutic aptamer is delivered to cancer cells using a cell-specific DNA aptamer.Targeting β-arrestin 2 inhibits tumor progression in CML models and patient samples.

Figure 1. Schematic of signaling pathways inhibited by β-arrestin 2 aptamer binding.

The β-arrestin 2 aptamer binds β-arrestin 2 and inhibits its interactions with proteins in both the Hh/Smo and Wnt/Fz signaling pathways. This significantly decreases the transcription of genes controlled by these pathways, preventing diseased cells from replicating. For a detailed review of these pathways see .

Figure 2. Development of the β-arrestin 2 aptamer.

(A) SELEX was performed on a random library of 10¹⁴ RNA oligonucleotides against β-arrestin 2. Each round, the pool of aptamers that had affinity for β-arrestin 2 was enriched, and the binding affinity of each round as measured by its dissociation constant (Kd) increased. Aptamers from round 9 and 12 were cloned and analyzed. (B) The three aptamers that bound β-arrestin 2 with the highest affinity were β-arr2A1 (Kd = 19.83 nM, Bmax = 86.04), β-arr2A2 (Kd = 4.13 nM, Bmax = 97.38), and β-arr2A3 (Kd = 22.03 nM, Bmax = 83.61). (C) Those aptamers showed selectivity for β-arrestin 2, as they were very poor binders of β-arrestin 1 in comparison, β-arr2A1 (Kd = 965.5 nM, Bmax = 53.99), β-arr2A2 (Kd = 2159.0 nM, Bmax = 97.31), and β-arr2A3 (Kd = 748.5 nM, Bmax = 22.58). (D) In vitro interactions between purified β-arrestin 2 and its cytoplasmic binding partner Erk were measured in the presence or absence of β-arrestin 2 binding aptamers and then precipitated with S-tag beads. Coimmunoprecipitation of recombinant Erk was visualized by western blot and then quantified and compared to a control reaction. β-arr2A2, and β-arr2A3 significantly reduced the interaction between β-arrestin 2 and Erk. Representative blot image shown.

Figure 3. Construction and delivery of an internalizing β-arrestin 2 aptamer chimera.

(A) Comparative western analysis was performed on the nucleolin protein levels of different cellular fractions in two different cell types. K562 cells, a CML cell-line, had approximately 30x more membrane-associated nucleolin than lymphoblastoid cells, which are non-cancerous human B cells. Representative blot image shown. (B) The nucleolin aptamer, which requires membrane-associated nucleolin for cell binding and internalization, bound and internalized into K562 cells as analyzed by flow cytometry. (C) Aptamer chimeras were generated with the nucleolin aptamer and the β-arr2A3 aptamer. The nucleolin aptamer acted as a cell-specific delivery agent, and was joined to the β-arr2A3 aptamer by complementary base pair annealing. (D) Biotinylated β-arr2A3 was hybridized to nucleolin aptamer and then allowed to internalize into cells for 24 hours. Cells were then lysed and subjected to biotin pull-downs by way of streptavidin beads. Coimmunoprecipitated β-arrestin 2 was visualized by western blot. (E) Quantification of β-arrestin 2 signals from (D), representative results of 3 independent experiments are shown. (F) Nucleolin-β-arr2 was applied to K562 cells and allowed to internalize for 6 hours. Cells were lysed, and lysates were immunoprecipitated using an anti-β-arrestin 2 antibody. Reactions were then subjected to northern blot analysis and were probed for the presence of the β-arr2A3 aptamer. (### = Control IgG antibody).

Figure 4. β-arrestin 2 targeting aptamer chimera interrupts multiple signaling pathways in K562 cells.

(A–D) K562 cells were treated with 200 nM of indicated aptamer chimeras or vehicle control for 72 hours. Cells were then harvested, lysed and subjected to western blot analysis. β-arrestin 2, Gli, and β-catenin were visualized along with Tubulin as a loading control. n = 5, *p<0.05 using one-way ANOVA with Bonferroni correction. (A) Quantification of β-arrestin 2 protein levels. SA3 = β-arrestin siRNA (B) Quantification of activated Gli protein levels. (C) Quantification of activated β-catenin protein levels. (D) Representative western blot from these experiments. (E) K562 cells were treated with 200 nM of nucleolin-β-arr2A3 or a control chimera and incubated for 72 hours. Total RNA was then purified from these cells and subjected to RT-PCR analysis as described in materials and methods. Left-panel – Downstream targets of the Wnt/Fz signaling axis. Right panel – Downstream targets of the Hh/Smo signaling axis. *p<0.05 using one-way ANOVA with Bonferroni correction.

Figure 5. Inhibition of leukemic cell growth via β-arrestin 2 inhibition.

(A–B) K562 cells (1000 cells per well) were plated in triplicate in methylcellulose and treated with 40 nM of the indicated aptamer chimera. Cells were incubated for 14 days, and then colonies were counted. n = 5 (C) K562 cells were treated with 400 nM, 40 nM or 4 nM of the indicated aptamer constructs. Cells were plated in duplicate and incubated for 14 days, and colonies were counted. n = 4. (D) Mice were infected with the BCR-ABL transgene and allowed to develop CML as described . Leukemic cells were purified from mouse spleens and treated with the indicated aptamer chimeras at 40 nM. Cells were plated triplicate in methylcellulose and incubated for 14 days before colony forming units were counted. n = 5 animals. (E) Blood samples from human patients were collected and enriched for CD34+ leukemia cells. Cells were treated with the indicated aptamer chimeras at 400 nM and plated in triplicate. After a14 day incubation, colonies were counted. n = 4 patients. All p-values generated using one-way ANOVA with Bonferroni correction.