The design and optimization of RNA trans-splicing molecules for skin cancer therapy

Abstract

Targeting tumor marker genes by RNA trans-splicing is a promising means to induce tumor cell-specific death. Using a screening system we designed RNA trans-splicing molecules (RTM) specifically binding the pre-mRNA of SLCO1B3, a marker gene in epidermolysis bullosa associated squamous cell carcinoma (EB-SCC). Specific trans-splicing, results in the fusion of the endogenous target mRNA of SLCO1B3 and the coding sequence of the suicide gene, provided by the RTM. SLCO1B3-specific RTMs containing HSV-tk were analyzed regarding their trans-splicing potential in a heterologous context using a SLCO1B3 expressing minigene (SLCO1B3-MG). Expression of the chimeric SLCO1B3-tk was detected by semi-quantitative RT-PCR and Western blot analysis. Cell viability and apoptosis assays confirmed that the RTMs induced suicide gene-mediated apoptosis in SLCO1B3-MG expressing cells. The lead RTM also showed its potential to facilitate a trans-splicing reaction into the endogenous SLCO1B3 pre-mRNA in EB-SCC cells resulting in tk-mediated apoptosis. We assume that the pre-selection of RTMs by our inducible cell-death system accelerates the design of optimal RTMs capable to induce tumor specific cell death in skin cancer cells.

Keywords: 3-(4,5-Dimethylthiazol-2-yl)2,5-diphenyltetrazolium bromide; BD; BP; Cancer gene therapy; DT-A; EB; EB-SCC; Epidermolysis bullosa; GAPDH; GCV; GCVTP; HSV; Herpes simplex virus – thymidine kinase (HSV-tk); IRES; MG; MMP9; MTT; PPT; RDEB; RNA trans-splicing; RNA trans-splicing molecule; RTM; RTM mutated 1, RTM mutated 2; RTM original; RTMm1, RTMm2; RTMmut; RTMorg; SCC; SLCO1B3; SMaRT; SS; STS; SqRT-PCR; Squamous cell carcinoma; TUNEL; binding domain; branch point; diphtheria toxin subunit A; epidermolysis bullosa; epidermolysis bullosa-associated squamous cell carcinoma; ganciclovir; ganciclovir triphosphate; glyceraldehyde 3-phosphate dehydrogenase; herpes simplex virus; internal ribosomal entry site; matrix metalloproteinase 9; minigene; poly pyrimidine tract; recessive dystrophic epidermolysis bullosa; segmental trans-splicing; semi-quantitative real time PCR; solute carrier organic anion transporter family member 1B3; splice site; spliceosome-mediated RNA trans-splicing; splicing deficient RTM; squamous cell carcinoma; terminal deoxynucleotidyl transferase dUTP nick end labeling; thymidine kinase; tk; β-subunit of human gonadotropin gene 6; βhCG6.

Figure 1

Evaluation of a specific binding domain (BD) for efficient trans‐splicing. (A) Schematic diagram of the screening procedure used for evaluation of BDs. (B) Binding position of RTM_1 within the target intron 3 of SLCO1B3 (nt1012‐nt1118). (C) Analysis of AcGFP expression by flow cytometry in HEK293AD cells transfected with the target molecule, RTM_1, or both the target molecule + RTM_1. Only cells co‐transfected with the target molecule and RTM_1 showed green fluorescence signal, resulting from fusion of both parts of the split AcGFP gene by RNA trans‐splicing. BD: Binding Domain, BP: Branch Point, PPT: Poly Pyrimidine Tract, 3′ ss: 3′ splice site.

Figure 2

(A) Constructs used in the experimental procedure. (B) Schematic diagram of the double‐transfection system used for RTM evaluation. Correct trans‐splicing of the RTM to SLCO1B3‐MG results in a fusion mRNA consisting of exon 3 of SLCO1B3 and tk. BD: Binding Domain, BP: Branch Point, PPT: Poly Pyrimidine Tract, 3′ ss: 3′ splice site, A: polyadenylation signal.

Figure 3

Detection of accurate trans‐splicing on the mRNA and protein level. (A) SqRT‐PCR analysis of HEK293AD cells double‐transfected with SLCO1B3‐MG and either the original RTM (RTMorg), or the mutated RTMs (RTMm1 and RTMm2). The correct PCR product of the fusion mRNA of SLCO1B3 and HSV‐tk (205 bp) is shown. GAPDH was used as housekeeping gene. All PCR products were verified by direct sequencing, demonstrating correct trans‐splicing between SLCO1B3‐MG and the RTMs on the mRNA level. A representative figure shows the sequence of the SLCO1B3‐HSV‐tk junction (right panel). (B) Sequence optimization leads to reduced background expression of RTMs as shown by Western blot analysis: Upper panel: α‐actinin staining (100 kDa) was used as a loading control. Lower panel: Western blot analysis demonstrated the expression of the SLCO1B3‐fusion protein (40 kDa) in all cell populations double‐transfected with the SLCO1B3‐MG (MG) and each RTM [lane 4, 5, 6]. Sequence optimized RTMm1 and RTMm2 showed greatly reduced background expression of HSV‐tk (37 kDa) in both single [lane 1, 2] and double [lane 4, 5] transfected HEK293 cells as compared to the RTMorg [lane 3, 6] (marked by asterisk). Cells transfected with the MG alone [lane 7], either the fusion [lane 9] or HSV‐tk [lane 10] construct served as negative and positive controls, respectively. Lane 8: Precision Plus Protein WesternC Standards (Biorad).

Figure 4

Biological activity of the fusion protein SLCO1B3 and HSV‐tk resulting from accurate trans‐splicing. (A) MTT assay was performed to measure reduction in cell viability. 100 ng of each RTM as well as different concentrations of SLCO1B3‐MG (100 ng, 500 ng, 1000 ng) were transfected into HEK293AD cells and the cells were treated with 100 μM GCV for 72 h. HEK293AD cells transfected with SLCO1B3‐MG alone, the SLCO1B3‐tk fusion or HSV‐tk served as negative and positive controls, respectively. The mean ± SD of two independent experiments is shown. Six replicates of each sample were measured in every experiment. The percentage of cell viability was calculated by the equation: (OD GCV‐treated/OD GCV‐untreated) × 100/negative control (OD GCV‐treated/OD GCV‐untreated). (B) SqRT‐PCR analysis of HEK293AD cells co‐transfected with RTMm2 (100 ng) and different amounts of SLCO1B3‐MG (100 ng, 500 ng, 1000 ng). Both, the transcripts of SLCO1B3‐MG and the fusion mRNA of SLCO1B3 and HSV‐tk were amplified. GAPDH was used as housekeeping gene. Data were calculated relative to the transcript level in HEK293AD cells transfected with 100 ng MG and 100 ng RTMm2.

Figure 5

Detection of cell death after RNA trans‐splicing. (A) In situ detection of cell death by TUNEL assay of HEK293AD cells treated with RTMm2 and RTMm2+MG was assessed by flow cytometric analysis. The SLCO1B3‐tk fusion served as positive control. Overlay of TMR red positive cells is shown. The black line represents the GCV‐untreated and the red line the GCV‐treated cell populations. (B) The reduction in cell number of transfected and GCV‐treated (100 μM) HEK293 cells (E–H) in comparison to GCV‐untreated cells (A–D) was visualized by light microscopy analysis. A decrease in cell viability of SLCO1B3‐tk and RTMm2+SLCO1B3‐MG transfected cells 72 h after addition of GCV was detected. TMR red staining (TUNEL assay) and nuclear counterstaining with DAPI of RTMm2 and SLCO1B3‐MG co‐transfected HEK293 cells revealed a significant higher rate of apoptotic cells after GCV treatment (J) in comparison to the level seen in the GCV‐untreated cell population (I).

Figure 6

Trans‐splicing into endogenous SLCO1B3 pre‐mRNA. (A) SqRT‐PCR confirmed a correct trans‐splicing product (245 bp) only in RTMm2 transduced EB‐SCC cells. The housekeeping gene GAPDH served as control for quality and quantity. PCR products were verified by direct sequencing (data not shown). (B) mRNA level of endogenous trans‐splicing compared to cis‐splicing within the SLCO1B3 gene in RTMm2 transduced EB‐SCC cells. Specific PCR products were amplified by sqRT‐PCR and trans‐splicing efficiency (total SLCO1B3‐tk transcripts) was calculated relative to the cis‐splicing level (total SLCO1B3 transcripts = 100%). GAPDH was used as housekeeping gene (C) Detection of SLCO1B3‐tk fusion protein (marked by asterisk) by immunoprecipitation (using anti‐FLAG antibody) in combination with Western blot analysis (using anti‐HSV‐tk antibody) in cells containing transduced RTMm2 or the fusion ctrl. In contrast, no specific protein was isolated from RTMmut transduced cells. Fusion proteins were purified using protein G sepharose (Fusion Ctrl, RTMm2 and RTMmut) or FLAG affinity gel (RTMm2#) and visualized with anti‐HSV‐tk antibody by Western blot analysis. (D) Expression of SLCO1B3‐tk fusion protein resulted in reduced cell survival as detected by MTT assay. The mean of at least 4 different experiments ± SD is shown. *P < 0.05 Student's t test, unpaired, two‐tailed.