[Experimental study on small interfering RNA silencing expression of tumor necrosis factor alpha and inhibiting...

To investigate the possibility of gene therapy of osteolysis around artificial joint prosthesis by constructing the recombinant adenovirus which can silence tumor necrosis factor alpha (TNF-alpha).

The primer of small interfering RNA (siRNA) coding sequence of silent TNF-alpha was designed and amplified, and then RAPAD adenovirus packaging system was used to load the sequence to adenovirus, and the recombinant adenovirus Ad5-TNF-a-siRNA-CMVeGFP which lacked both E1 and E3 regions was constructed. Then 64 female BABL/C mice (weighing, 20-25 g) were randomly divided into 4 groups (n=16): blank control (group A), positive control (group B), simple adenovirus (group C), and treatment group (group D). The prosthetic-model was established in group A, and the prosthetic-loosening-model in groups B, C, and D. At 2 weeks after modeling, PBS solution was injected first, and then the same solution was injected 24 hours later in group A; titanium particle solution was injected, and then PBS solution, Ad5 E1-CMVeGFP (1 x 10(9) PFU/mL), and Ad5-TNF-alpha-siRNA-CMVeGFP (1 x 10(9) PFU/mL) were injected, respectively in groups B, C, and D 24 hours later, every 2 weeks over a 10-week period. The general condition of mice was observed after operation. The tissues were harvested for histological observation, and the expression of TNF-a was detected by Western blot at 12 weeks after operation.

The positive clones were achieved by enzyme digestion and confirmed by DNA sequencing after loading the target genes into adenovirus vector, and then HEK293 cells were successfully transfected by recombinant adenovirus Ad5-TNF-alpha-siRNA-CMVeGFP. All mice survived to the completion of the experiment. Histological observation showed that there were few inflammatory cells and osteoclasts in group A, with a good bone formation; there were a large number of inflammatory cells and osteoclasts in groups B and C, with obvious bone destruction; inflammatory cells and osteoclasts in group D was less than those in groups B and C, with no obvious bone destruction. Significant difference was found in the limiting membrane thickness and the number of osteoclasts (group A < group D < group B < group C, P < 0.05). Western blot showed that the TNF-a expression levels were 0.235 +/- 0.022, 0.561 +/- 0.031, 0.731 +/- 0.037, and 0.329 +/- 0.025 in groups A, B, C, and D respectively, showing significant difference among 4 groups (P < 0.05).

The recombinant adenovirus for silencing TNF-alpha is successfully constructed, which can effectively inhibit osteolysis by silencing TNF-alpha expression in the tissues around prosthesis in mice.