Aptamers selected against the unglycosylated EGFRvIII ectodomain and delivered intracellularly reduce membrane-bound EGFRvIII and induce apoptosis

Abstract

Epidermal growth factor receptor variant III (EGFRvIII) is a glycoprotein uniquely expressed in glioblastoma, but not in normal brain tissues. To develop targeted therapies for brain tumors, we selected RNA aptamers against the histidine-tagged EGFRvIII ectodomain, using an Escherichia coli system for protein expression and purification. Representative aptamer E21 has a dissociation constant (Kd) of 33x10(-9) m, and exhibits high affinity and specificity for EGFRvIII in ELISA and surface plasmon resonance assays. However, selected aptamers cannot bind the same protein expressed from eukaryotic cells because glycosylation, a post-translational modification present only in eukaryotic systems, significantly alters the structure of the target protein. By transfecting EGFRvIII aptamers into cells, we find that membrane-bound, glycosylated EGFRvIII is reduced and the percentage of cells undergoing apoptosis is increased. We postulate that transfected aptamers can interact with newly synthesized EGFRvIII, disrupt proper glycosylation, and reduce the amount of mature EGFRvIII reaching the cell surface. Our work establishes the feasibility of disrupting protein post-translational modifications in situ with aptamers. This finding is useful for elucidating the function of proteins of interest with various modifications, as well as dissecting signal transduction pathways.

Figure 1. Bead selection for EGFRvIII ectodomain significantly increased RNA pool affinity

The affinity of the initial (R0) and final (R12) rounds RNA for EGFRvIII were compared in a double filter binding assay. Representative clones E17 and E21 for EGFRvIII are included, along with the pools.

Figure 2. Aptamers bind their target proteins with high affinity

(A) ELISA assay showing in vitro aptamer-protein interaction. (B, C) The interaction between protein EGFRvIII ectodomain and aptamer E21 is confirmed by surface plasmon resonance (Biacore 3000). For E21, the determined constants are k_a = 1×10⁴ m⁻¹ s⁻¹; k_d = 4×10⁻³ s⁻¹.

Figure 3. EGFRvIII aptamers specifically bind unglycosylated EGFRvIII

(A) Western blot showing the different molecular masses of EGFRvIII from E. coli (E), baculovirus (B), and two deglycosylated EGFRvIII. 1: chemical deglycosylation with trifluoromethanesulfonic acid, 2: enzymatic deglycosylation with PNGase F. Left panel: a light exposure to show EGFRvIII expressed from E. coli. (B) The binding assay revealed that aptamer E21 binds only E. coli-expressed EGFRvIII and not baculovirus-expressed, glycosylated EGFRvIII.

Figure 4. EGFRvIII aptamers reduce membrane EGFRvIII after transient transfection

(A) Representative Western blot to show that membrane EGFRvIII expression is reduced in response to aptamer transfection. The same blot was stripped, and reprobed with anti-GAPDH antibody to monitor loading. (B) Percent densitometric values of membrane EGFRvIII protein levels demonstrated that membrane EGFRvIII levels were significantly decreased in NR6M cells transfected with aptamer E17 or E21. *p<0.05 vs. library; **p<0.01 vs. library; n=3 experiments.

Figure 5. EGFRvIII aptamers induce cell apoptosis after transient transfection

(A) Hoechst 33342 staining of apoptotic nuclei. Condensed, bright nuclei exhibiting features of apoptosis are indicated by arrows. (B) Percentage of apoptotic nuclei. Apoptotic nuclei were counted from five different areas and averaged. *p<0.05 vs. library control. n=3 independent experiments.