TLR7 agonist-induced bronchodilation: key mechanistic questions remain.

TO THE EDITOR: The role of Toll-like receptor-7s (TLR7) in innate immunity against single-stranded RNA viruses is well established. Recently, our group found TLR7 agonists also relax guinea pig and human airway smooth muscle strips via production of nitric oxide (1, 2). Immunostaining identified extensive TLR7 expression on airway nerves, but none on smooth muscle; thus we hypothesized airway nerves are the source of TLR7-induced nitric oxide. Therefore, Larsson et al.’s study (4) evaluating TLR7 agonist-induced bronchodilation was of great interest to us. The authors propose imidazoquinoline agonists bronchodilate via a nonneuronal TLR7independent pathway in airway smooth muscle. Unfortunately, the presented data fail to establish this central conclusion. The authors suggest imiquimod’s mechanism is nonneuronal since neither vagotomy nor tetrodotoxin (a voltage-gated sodium channel antagonist) blocked imiquimod’s effect. However, one may only conclude from these data that imiquimodinduced bronchodilation does not involve a central neuronal reflex nor does it require depolarization of tetrodotoxin-sensitive nerves. Both dorsal root and nodose sensory ganglia contain tetrodotoxin-resistant nerves (3), which the current experiments do not address. Furthermore, while the authors show imiquimod alters calcium homeostasis in cultured airway smooth muscle, they neglected to evaluate whether a similar effect occurs in cultured airway nerves. Indeed, their conclusion that imiquimod does not act on nerves contradicts several recent publications. Liu et al. (6) showed imiquimod induces TLR7-dependent calcium responses in mouse dorsal root ganglia and Lee et al. (5) showed imiquimod influences sensory nerve excitability by inhibiting voltage-gated potassium channels. While these studies evaluated sensory nerves pertaining to itch, it is reasonable to suspect TLR7 agonists also have similar effects on airway sensory and parasympathetic nerves given the widespread expression of TLR7 on airway innervation. The authors also conclude imiquimod’s effect is TLR7independent based on their inability to block bronchodilation with the TLR7 antagonist IRS661. However, the authors fail to provide essential control experiments confirming the antagonist inhibits TLR7 in their model. In our experiments, we found that this same TLR7 antagonist blocked both imiquimodinduced and the structurally unrelated TLR7 agonist poly-Uinduced bronchodilation in guinea pigs (2). It is important to note that several observations by Larsson et al. (2) agree with our previous findings, including that imiquimod acts in an epithelium-independent manner in part through prostaglandins in guinea pigs. In contrast to our findings, they conclude nitric oxide is not involved since the partial inhibition of imiquimod-induced bronchodilation found in N-nitro-L-arginine methyl ester (L-NAME)-treated animals in vivo did not reach statistical significance (P 0.08; Fig. 3A). At this high dose of imiquimod, we also observed a nitric oxide-independent bronchodilatory effect in vivo (albeit with intravenous imiquimod administration) (1), with the predominant effect of nitric oxide at lower doses. Moreover, in human airways, we found imiquimod-induced bronchodilation was wholly blocked by N-monomethyl-L-arginine (L-NMMA) (1), further supporting a role for a nitric oxide-dependent mechanism. The authors suggest our results differ because we contracted tissue “mainly through neuronal stimulation.” On the contrary, we demonstrated imiquimod-induced relaxation was inhibited by L-NMMA in both guinea pig and human airways precontracted with methacholine. Thus, while Larsson et al.’s study provides evidence for imiquimod’s effect on airway smooth muscle, it falls short of establishing a TLR7-independent nonneuronal mechanism for imiquimod-induced bronchodilation.