A conditionally replicative adenovirus vector containing the synNotch receptor gene for the treatment of muscle-invasive bladder cancer

Abstract

Muscle-invasive bladder cancer (MIBC), a highly heterogeneous disease, shows genomic instability and a high mutation rate, making it difficult to treat. Recent studies revealed that cancer stem cells (CSCs) play a critical role in MIBC frequent recurrence and high morbidity. Previous research has shown that Cyclooxygenases-2 (COX-2) is particularly highly expressed in bladder cancer cells. In recent years, the development of oncolytic adenoviruses and their use in clinical trials have gained increased attention. In this study, we composed a conditionally replicative adenovirus vector (CRAd-synNotch) that carries the COX-2 promotor driving adenoviral E1 gene, the synNotch receptor therapeutic gene, and the Ad5/35 fiber gene. Activation of the COX-2 promoter gene causes replication only within COX-2 expressing cancer cells, thereby leading to tumor oncolysis. Also, CD44 and HIF signals contribute to cancer stemness and maintaining CSCs in bladder cancer, and the transduced synNotch receptor inhibits both CD44 and HIF signals simultaneously. We performed an in vivo study using a mouse xenograft model of T24 human MIBC cells and confirmed the significant antitumor activity of CRAd-synNotch. Our findings in this study warrant the further development of CRAd-synNotch for treating patients with MIBC.

Fig. 1. Construction of ADX730 and CRAd-synNotch.

Construction of recombinant replication-deficient adenoviral vector. A synthetic fusion gene encoding a synNotch receptor (a) was introduced into the ΔE1 region of the adenovirus type 5 vector to construct a recombinant replication-deficient adenoviral vector, ADX730 (b). Construction of conditionally replicative adenovirus vector. (CRAd-synNotch) carries the COX-2 promotor gene upstream of the E1 region of the adenovirus vector genome, the synNotch receptor gene in the E3 region, and the Ad5/35 fiber gene downstream of the E3 region (c).

Fig. 2. Expression of CD46, CAR, CD44 and induction of hyaluronan synthases (HAS1, HAS2, and HAS3) and COX-2 mRNA in KK47, 5637 and T24 cells.

a The expression of CD46, CAR, and CD44 on the surface of T24 cells was evaluated by flow cytometry. b The expression of CD46, CAR, and CD44 on the surface of BT474 cells was evaluated by flow cytometry. c The expression of CD46, CAR, and CD44 on the surface of KK47 cells was evaluated by flow cytometry. d The expression of CD46, CAR, and CD44 on the surface of 5637 cells was evaluated by flow cytometry. e The mRNA expressions of HAS1, HAS2 and HAS3 in KK47, 5637 and T24 cells were measured by real-time RT-PCR. f The mRNA expressions of COX-2 in KK47, 5637 and T24 cells were measured by real-time RT-PCR. All mRNA levels were standardized by the expression levels of the control gene TATA-binding protein (TBP) and analyzed using the ΔΔCt method. All these results were expressed as the mean ± SE for three separate experiments (*p < 0.05, **p < 0.01).

Fig. 3. CRAd-synNotch replicates dependently on COX-2.

The copy numbers of wild-type adenovirus (wtAd) (a), ADX730 (b), CRAd-GFP(c) and CRAd-synNotch (d) infected into BT474, KK47, 5637 and T24 cells were evaluated using absolute quantification by RT-PCR. All these results were expressed as the mean ± SE for three separate experiments (*p < 0.05, **p < 0.01). The standard curve was generated using the pAd1128 plasmid from the AdenoQuick 2.0 Kit (O.D.260Inc).

Fig. 4. Both CRAd-synNotch and ADX730 transduce synNotch receptor fusion proteins in T24 cells and inhibits cell growth in T24 cells.

The T24 cells were infected with ADX730 (a) at 100 MOI and CRAd-synNotch (c) at 50 MOI, and cultured for 24, 48, and 72 h. The expression of the synNotch receptor was analyzed by Western blotting using an anti-CD44 antibody. The T24 cells were infected with ADX730 (b) at 100 MOI and CRAd-synNotch (d) at 50 MOI, and cultured for 24, 48, and 72 h. The expression of the synNotch receptor was analyzed by Western blotting using a HIF-3α4 antibody. e The gene expression of HIF-3α4 in T24 cells was measured by real-time RT-PCR after Ad-lacZ and ADX730 infections. f The gene expression of HIF-3α4 in T24 cells was measured by real-time RT-PCR after CRAd-GFP and CRAd-synNotch infections. T24 (g) and BT474 (h) cells were treated with 50 MOIs of ADX730, CRAd-GFP, and CRAd-synNotch and incubated at 37°C for 72 h under oxygen concentrations of 2% (Hypoxia) or 21% (Normoxia). The absorbance was measured at a wavelength of 492 nm. All mRNA levels were standardized by the expression levels of the control gene TATA-binding protein (TBP) and analyzed using the ΔΔCt method. All these results were expressed as the mean ± SE for three separate experiments (**p < 0.01).

Fig. 5. Gene expression of SOX-2, Cortactin, OCT4, Nanog, c-Myc, Cyclin-d1, VEGF, PHD3, Bcl-xL and CyclinG2 in T24 cells infected with ADX730 and CRAd-synNotch.

The mRNA expression of SOX-2 (a) Cortactin (b) OCT4 (c) Nanog (d) c-Myc (e) Cyclin-d1 (f) VEGF (g) PHD3 (h) Bcl-xL (i) and CyclinG2 (j) levels in T24 cells infected with Ad-LacZ and ADX730 at 100 MOI were measured by real-time RT-PCR. The mRNA expression of SOX-2 (k) Cortactin (l) OCT4 (m) Nanog (n) c-Myc (o) Cyclin-d1 (p) VEGF (q) PHD3 (r) GLUT1 (s) and CyclinG2 (t) levels in T24 cells infected with CRAd-GFP and CRAd-synNotch at 50 MOI were measured by real-time RT-PCR. After viral infection, Cells were cultured under under oxygen concentrations of 2% or 21%. All mRNA levels were standardized by the expression levels of the control gene TATA-binding protein (TBP) and analyzed using the ΔΔCt method. All these results were expressed as the mean ± SE for three separate experiments (*p < 0.05, **p < 0.01).

Fig. 6. Immunohistochemical staining for CD44 and antitumor effect of CRAd-synNotch in mice with T24 tumors.

a Immunohistochemical staining image after subcutaneously injecting T24 cells into nude mice and infecting them with PBS, ADX730, CRAd-GFP, and CRAd-synNotch; HPF: High Power Field, Original magnification: x400; LPF: Low Power Field, Original magnification: x80 (n = 3 per group, average ±SE bars, **p < 0.01). b Quantitative analysis of (a) using Visiopharm Oncotopix Discovery software. DAB: 3,3’-diaminobenzidine. c Tumor growth curve after subcutaneously injecting T24 cells into nude mice and infecting them with PBS, ADX730, CRAd-GFP, and CRAd-synNotch (n = 5 per group, average ±SE bars, **p < 0.01). d Survival in the T24 nude mouse model (n = 5 per group).