Our serial in vitro, ex vivo and in vivo studies used naked siRNA to minimize off-target effects, which were often induced by siRNA vehicles. The recognition of the naked caspase-3 siRNA duplexes as a foreign entity by the innate immune system acts as a trigger for the rapid release of inflammatory mediators as part of the natural defense of hosts against pathogens. The naked siRNA, however, has very short half-life in serum. In the caspase-3 siRNA preserved 48-h auto-transplanted kidneys, the expression of TLR3 and TLR7, their adaptors MyD88 and TRIF, as well as TLR-independent PKR, down-stream mediators NF-κB and c-Jun, IFN-α, β and γ, cytokines TNF-α, IL-1β and IL-6 were all up-regulated. These changes might be more likely ascribed to systemic complementary responses, although the persistent action initiated by caspase-3 siRNA cannot be completely ruled out.
Due to effectiveness of caspase-3 siRNA in kidney preservation, but easy degenerating property of naked siRNA, more caspase-3 proteins were synthesized and cleaved into active subunits in the 48-h auto-transplanted kidneys. In addition, the cellular stress upon reperfusion without systemic supplement of caspase-3 siRNA also could contribute to the enhanced caspase-3 expression. Although all possible aspects involved in unprotected auto-transplant kidney were well discussed in our previous publication , the activation of innate immunity and its possible pathways, as well as amplified inflammatory reactions, were further demonstrated in this study. These orchestrated cascade changes might be mainly initiated by the systemic feedback due to the low level of caspase-3 mRNA and protein in the caspase-3 siRNA preserved kidneys. The end results were not only more severe inflammation mellitus, but also more apoptotic cell death and renal tissue damage.
It has been shown that, TLR3, apart from TLR2 and TLR4, induces cellular apoptosis via the PI3K-Akt signaling pathway . Endosomal TLR3 signaling occurs through the TRIF adaptor leading to IFN regulatory factor 3 activation and IFN-β expression. The TLR3 signaling also activates down-stream mediators NF-κB and c-Jun, which cause the production of inflammatory cytokines. TLR3 in the surface of endothelial cells could be also activated by exogenous siRNA, at least 21 nt in length, in vivo, leading to the production of cytokines such as IFN-γ and IL-12 . The up-regulated TLR3 and TRIF were observed in the caspase-3 siRNA persevered auto-transplant kidneys, which indicate the activation of TLR3 signaling. However, it was very difficult to trace the location of caspase-3 siRNA and prove whether porcine TLR3 was directly activated by the caspase-3 siRNA in this study. It also needs to be taken into account that there is a great homology between porcine and human, whereas human TLR3 is not activated by siRNA due to the structure difference in the activation point .
In addition, the production of IFN-α and inflammatory cytokines could be triggered by synthetic siRNAs and short hairpin RNAs dose-dependently via TLR7 in immune cells including plasmacytoid dendritic cells (DCs) and myeloid DCs [3, 13, 14]. TLR7 is constitutively expressed in these cells . Interestingly, TLR7 preconditioning mediates neuroprotection against ischemic injury and the mechanism involved is unique in contrast to other TLR preconditioning ligands, which dependents on type I IFN receptor, but is independent of TNF . Consistently, the increased TLR7 was demonstrated in the caspase-3 siRNA preserved auto-transplant kidneys in this study, accompanied by MyD88 activation, which led to the activation and nuclear translocation of NF-κB and c-Jun. As a result, the production of IFN-α, IFN-β and IFN-γ, as well as inflammatory cytokines TNF-α, IL-1β and IL-6, were remarkably increased and auto-transplant kidney structure was more severely damaged [6, 17–20]. These cascade responses more liked were due to a systemic feedback post 48-h transplantation, rather than persistent actions initiated by caspase-3 siRNA in 24-h cold preservation.
HMGB1 secretion and inflammation promote each other [21, 22]. Both HMGB1 and infiltrated myeloperoxidase (MPO)+ neutrophils were significantly higher in siRNA preserved auto-transplant kidneys , which indicate MPO+ cells might be recruited by highly expressed HMGB1 protein locally. Combined with our previous data , the positive correlations between HMGB1 protein and MPO+ cells, inflammatory mediators, apoptosis or innate immune activation were revealed, which indicated HMGB1 might promote inflammation and apoptosis. A recent study also revealed that fewer neutrophils infiltrated in the myocardium of tlr4 (HMGB1 receptor) mutant mice was observed after myocardial IR, and tlr4 deficiency markedly decreased IRI with inhibited HMGB1, TNF-α, and IL-8 . Furthermore, HMGB1 promotes neutrophil extracellular trap formation and may contribute to the severity of neutrophil-associated inflammatory conditions .
Interestingly, the close correlations were observed, between HMGB1 and caspase-3 protein, positive with 17 kD active subunit, but negative with 32 kD precursor. As we explained in the previous study , the feedback regulation induced by short-term caspase-3 silence caused more caspase-3 precursor to be cleaved into active subunits. Hence, more cells in the siRNA preserved auto-transplant kidneys underwent to apoptosis followed by HMGB1 secretion. With the siRNA treatment, the increased HMGB1 recruited more neutrophils, and produced more inflammatory cytokines, which then amplified inflammation and aggravated tissue lesion. In addition, HMGB1 expression was strong correlated with TLR3 expression, not TLR7 or PKR. TRIF, an adapter protein in TLR3 signal transduction, not MyD88, also correlated positively with serum creatinine. Therefore, it suggests that TLR3 activation together with increased TRIF might take main responsibility for renal tissue damage.
Recently, certain modulations of siRNAs have been developed to maintain its efficacy, but reduce unpleasant side effects. A novel chemical modified 21-mer epithelial sodium channel siRNA with 85% knockdown efficiency showed no evidence for potential to stimulate TLR3, TLR7 or TLR8 . In addition, 8-Alkoxyadenosine phosphoramidites was incorporated into the guide strand of caspase-2 siRNA at different positions, in which single modifications at positions 6 and 10 were effective in blocking siRNA binding to the PKR and might reduce sequence-independent side effects . The bifunctional siRNAs with both gene silencing and innate immune activation properties, however, may represent a new potential strategy for treatment of virus infection. A chemically synthesized HBVx siRNA not only inhibited HBVx mRNA expression, but also increased expression of PKR leading to the higher production of type I IFN .
It has been also noted that using a multi-targeted siRNA cocktail is a better approach than simply increasing the dose of the best performed single siRNA. The combinational siRNA strategy applying lower concentration of each siRNA may reduce the off-target effects without sacrificing silencing potency . Such strategies have been tested in IRI treatment. For instance, the blockade of caspase-3 and caspase-8  or caspase-3 and complement 3  simultaneously in renal IRI were beneficial. A solution with combined siRNAs targeting TNF-α, complement 3 and Fas inhibited heart graft injury with prolonged graft survival . Therefore, knocking down elements involving in TLR pathways together with caspase-3 might be more efficient to ameliorate IRI in transplantation, which is worthy to be further investigated.