Preferential conduction block of myelinated axons by nitric oxide

    loading  Checking for direct PDF access through Ovid


Nitric oxide (NO) is a highly reactive molecule with a complex chemistry that is involved in numerous normal and pathological processes. It is produced from L‐arginine by a family of NO synthases: neuronal (nNOS), endothelial (eNOS), and inducible (iNOS). The latter is upregulated in microglia/macrophages and in astrocytes by inflammatory cytokines, resulting in significant increases in NO concentrations. NO and its redox products react with proteins through nitration of tyrosine or S‐nitrosylation of cysteine (Stamler et al., 1992b). It also activates guanylyl cyclase, raising levels of cyclic guanosine monophosphate (cGMP), a second messenger that regulates many cellular functions. There is growing evidence that NO plays a role in the pathology of multiple sclerosis and spinal cord injury, both of which include an inflammatory component (Smith and Lassmann, 2002; Su et al., 2015). In one such role, it has been shown that NO blocks axonal conduction in a reversible manner (Redford et al., 1997; Shrager et al., 1998), and this has raised interest in developing therapeutic measures to prevent it (Smith and Lassmann, 2002). In reading recent literature, we reviewed our earlier work (Shrager et al., 1998) and found a misinterpretation of one experiment. While most of that research was on rat sciatic nerves, we also recorded from vagus nerves and found a large, fast component of the compound action potential (CAP) as well as a smaller, slower component. Both of these were blocked reversibly by NO, and we concluded that both myelinated and unmyelinated axons were therefore susceptible. However, when we reexamined these data we found that the conduction velocity of the smaller component was 5.9 m/s, much too fast to be attributed to unmyelinated axons, which typically conduct at 0.4 to 0.7 m/s at 37 °C. We therefore decided to reexamine the question of the interaction of NO with unmyelinated axons.
While extracellular signals recorded from myelinated fibers are in the range of 0.5 to 5 mV and are readily detected, action potentials from unmyelinated fibers in adult nerves are typically only 0.01 to 0.05 mV and must be extracted from background noise. This can be done with appropriate low‐noise amplifiers and signal averaging. While we recorded from a variety of adult tissues, we also examined developing nerves, in which the signals from unmyelinated (i.e., premyelinated) fibers are more robust. We applied NO through the use of a chemical donor, diethylamine NONOate (DEA NONOate). This compound was particularly useful for these experiments in part because a wide variety of controls and calculations have been done with it. Its half‐life, releasing NO, is 16 min at 22 °C to 25 °C and 2 min at 37 °C, rates which we verified through use of the Griess reagent (Green et al., 1982; Shrager et al., 1998). Correspondingly, its effects were minimal at 25 °C and only blocked when the temperature was raised to 37 °C, showing that the undissociated compound was without effect. Further, solutions were ineffective if allowed to fully dissociate into diethylamine and NO before applying to nerves (Shrager et al., 1998). (The half‐life of NO in aqueous solutions is ∼30 sec [Palmer et al., 1987].) Finally, solving the diffusion equation for a cylinder, and including the half‐life of NO in tissue (∼4 sec), we showed that at concentrations of the order used, DEA NONOate at 37 °C produces levels of NO in nerves that are similar to those that activated macrophages would produce in an inflammatory lesion (∼1 μM) (Shrager et al., 1998). Here we show that in both the central nervous system (CNS) and peripheral nervous system (PNS), in regions that contain both myelinated and unmyelinated axons, conduction in myelinated fibers is preferentially blocked by NO.
    loading  Loading Related Articles