Fig 1.
APP contains a putative G-quadruplex.
(A) Schematic of the APP mRNA that contains the putative G-quadruplex sequence in the coding region at position 957 (as identified by Westmark et al. 2007) and the G-quadruplex sequence in the 3’UTR (discussed in this paper) at position 3008. (B) G-quadruplex consensus sequence and comparison of the G-quadruplex sequence in the 3’UTR of human APP with the 3’UTR APP for other species.
Fig 2.
CD potassium ion titration of the APP 3’UTR G-quadruplex shows that it forms in vitro and is in parallel topology with 3-state folding.
(A) CD spectra collected as a function of K+ ion concentration. K+-mediated G-quadruplex folding is performed at 2.5 μM RNA under 10 mM lithium cacodylate (LiCac) (pH 7.0), with K+ ion concentration ranged from 0 to 1 M. The positive peak at ~260 nm and negative peak at ~240 nm are CD signatures for parallel topology of G-quadruplex. (B) CD signal (ellipticity monitored at 262 nm) as a function of K+ ion concentration from panel A shows clear three-state transitions in G-quadruplex folding. The fitting was performed using Eq 1 (see Material and Methods). At physiological K+ ion concentration (~150 mM), the G-quadruplex is fully folded. The K+1/2 and Hill coefficients (n) are provided in the plot. (C) CD titration and comparison of APP 3’UTR wild-type and mutant G-quadruplex sequence. 2.5 μM RNAs were used under 10 mM LiCac (pH 7.0) and physiological 150 mM or 0 mM K+ ion concentration The GGGG to AAAA substitution in the mutant disfavor G-quadruplex formation as evident by the reduction in CD characteristic signals for G-quadruplex (compare blue and red), and yield similar CD signal to the wild type sequence at 0 mM K+ ion concentration (green).
Fig 3.
G-quadruplex regulation of Luciferase gene expression.
(A) mRNA levels as assessed by qPCR and represented ΔΔCT values represents the normalization of the ΔCT of firefly luciferase to the ΔCT renilla luciferase then normalized to the wild type. (B) Quantification of Dual Luciferase Assay comparing the wild type G-quadruplex sequence to the mutant G-quadruplex sequence (G-quad Mut). Empty Firefly Luciferase plasmid was used as a control which does not contain the 3’UTR of APP. Firefly Luciferase values were normalized to Renilla Luciferase.
Fig 4.
G-quadruplex regulation of APP gene expression.
(A) Western blot analysis of cells transfected with reporter constructs containing APP 695 coding sequence with wild type (G-quad WT) or mutant (G-quad Mut) sequence. Mock Transfection was used to confirm that these constructs were over-expressed in HeLa cells. Antibody C1/6.1 recognizes both full-length APP695 and CTF. β-Actin was used as a loading control. (B-C) Quantification of Western blots as in 4A with (B) APP levels normalized to β-Actin and (C) Endogenous CTF intensity values were subtracted from total CTF intensity values to obtain exogenous CTF values which were then normalized to β-Actin values. (D) Aβ ELISA quantification of total Aβ levels from conditioned medium of cells transfected with APP constructs containing the wild type or mutant 3’UTR G-quadruplex.
Fig 5.
G-quadruplex regulation of APP translation.
(A) Western blot analysis of cells transfected with reporter constructs containing APP 695 coding sequence containing a C-terminal myc tag with either wild type (WT G-quad) or mutant (Mut G-quad) G-quadruplex sequence. Mock transfection was used to confirm that these constructs were over-expressed. Antibody 9B11 (anti-Myc) was used to detect APP-Myc. (B) Schematic representation of methods taken to identify newly synthesized APP (See methods). (C) Western blot analysis for immunoprecipitation demonstrating the successful pull down of Myc-tagged APP constructs (IP: using 9B11 anti myc mAb, IB: C1/6.1 anti APP). (D) Western blot analysis of total APP (top panel, IB: C1/6.1 anti APP) and newly synthesized APP (bottom panel, IB: Streptavidin). Statistical analysis was performed by normalizing the newly synthesized APP/Total APP (Streptavidin/C1/6.1) using a ratio paired t-test using Prism 6.0g for Mac.