Transient receptor potential (TRP) channels are widely distributed in mam- malian tissue. The TRP family includes at least 20 related cation channels that are responsive to a range of chemical and physical stimuli (Clapham et al. 2001). The functions of TRP channels in sensory neurons include re- sponding to painfully hot temperatures as well as to moderate warming and cooling and, perhaps, to noxious cold. Selectively expressed in many primary nociceptors, TRPV1 mediates the response to painful heat, to extracellular acidosis and to capsaicin, the pungent component in hot peppers (Caterina et al. 2000). TRPV1 is probably a tetramer that forms a nonspeciﬁc cation
channel, depolarizing the cell by allowing inﬂux of Na+ and Ca2+ ions when it opens. The depolarization further activates other voltage-dependent channels,
leading to an increase in excitability (lowering of threshold) for intermedi- ate steady depolarizations or to a membrane that is strongly refractory to stimulation (i.e., a “depolarization block”) for larger depolarizations. Entry of Ca2+ through the open TRPV1 channel can stimulate further increases in intracellular Ca2+, e.g., via Ca2+-induced Ca2+ release from intracellular stores. In addition, intracellular Ca2+ may be released in response to cap- saicin that acts directly on membranes of the endoplasmic reticulum, an or- ganelle that stores calcium and may contain TRPV1 channels (Karai et al.
2004). In distal terminals of skin, muscle, and joints and in central termi- nals in the dorsal horn of the spinal cord, activation of TRPV1 elevates in- tracellular Ca2+, resulting in the release of glutamate and substance P from the primary nociceptor terminals and the subsequent activation of multiple
ionotropic and metabotropic receptors on postsynaptic membranes. Some of these activations may have synergistic interactions, e.g., interactions be- tween NK-1 receptors for substance P and N -methyl-d-aspartate (NMDA)- type glutamate receptors. These and other interactions may induce activa- tion of signal transduction pathways in spinal neurons that have far-reaching acute and chronic actions (Ji and Strichartz 2004), likely important for cen- tral sensitization that contributes strongly to enduring pain from injury and inﬂammation.
Different local anesthetics appear to have different effects on TRPV1. For ex- ample, lidocaine and prilocaine suppress the capsaicin-induced increase in in-
tracellular Ca2+ mediated by recombinant TRPV1 receptors (Hirota et al. 2003), and bupivacaine inhibits the capsaicin-induced current in isolated sensory neurons, although tetracaine actually enhances this current (Komai and Mc- Dowell 2005). The reasons underlying this difference are not known, and the sensitivity of other TRP channels to local anesthetics have not been investi-
gated. Intracellular Ca2+ is elevated by many local anesthetics in several types of cells (Gold et al. 1998b; Johnson et al. 2002), very possibly through their
ability to uncouple mitochondria and thereby release this organelle’s stored Ca2+ into the cytoplasm (Chance et al. 1968, 1969). Consequently, a number of Ca2+-dependent processes are stimulated by local anesthetics, including calmodulin-dependent reactions that turn on kinases and inactivate Ca2+ channels. Thus it is possible that part of the inhibition by local anesthetics of both TRPV1 receptors and other ion channels is due not to a direct action at those sites but through the recruitment of inhibitory pathways activated by such elevated Ca2+.
Local anesthetics appear to have a potentiating effect on the long-lasting, impulse-inhibiting actions of vanilloids. Both in isolated nerves (Shin et al.
1994) and in rat sciatic nerve block in vivo (Kohane et al. 1999), partial reduction of action potentials by local anesthetics is furthered by the addi- tion of capsaicin, possibly through the LA-induced elevation of intracellu- lar Ca2+, which is known to mediate the desensitization of TRPV1 receptors
in sensory neurons (Cholewinski et al. 1993). Capsaicin is currently being used therapeutically for the relief of certain neuropathic pains (Rowbotham et al. 1995, 1996), and a related, high-potency vanilloid, resiniferatoxin, is being developed for analgesia via selective blockade of TRPV1-containing pri- mary nociceptors (Kissin et al. 2002, 2005a, b). Both of these applications involve the coadministration of local anesthetics to block the generation of impulses during initial application. Interactions between local anesthetics and the vanilloid receptor are thereby clinically important as well as scientiﬁcally interesting.
Mechanisms of Spinal and Epidural Anesthesia
The mechanism by which local anesthetics block impulses in peripheral nerves through inhibition of voltage-gated Na+ channel is well established. In contrast, the overall mechanism for spinal and epidural anesthesia is almost certainly more complex than simply the blockade of impulses in nerve roots, involving
pre- and postsynaptic receptors as well as intracellular pathways (Butterworth and Strichartz 1990). Both epidural and spinal (intrathecal) delivery of local anesthetics results in their presence in the spinal cord, as well as in the spinal roots (Gokin and Strichartz 1999), allowing the possibility of altering both synaptic activity and impulse conduction and affecting the responses of cells
contained within the spinal cord, including interneurons, astrocytes and mi- croglia. At an even further level of resolution, local anesthetics can interact with membrane phospholipids and proteins and thereby affect various cellular activities. For example, local anesthetics can affect several subtypes of PKC (Mikawa et al. 1990; Nivarthi et al. 1996), PKA (Gordon et al. 1980), G pro- teins (Hagelüken et al. 1994), and the various receptors that activate them (Hitosugi et al. 1999). As noted above (Sect. 2.4 and 2.5), the effects of LAs on membrane-associated enzymes and second-messenger systems operating in the cytoplasm are extensive. One means of assessing the actions of local anesthetics on integrated responses in spinal cord, such as might occur during a spinal or epidural blockade, is to measure their effects on intracellular signal transduction pathways that are activated by many different inputs and thus report the overall “activation status” of cells in the spinal cord. This approach is described in the following section.
MAP Kinases and Pain Processing in the Spinal Cord
The mitogen-activated protein kinase (MAPK) enzymes are a family of serine- threonine protein kinases that are activated in response to many stimuli and play important roles in cellular signal transduction, both acutely by protein phosphorylation and chronically by affecting gene transcription. MAPKs were originally implicated in the regulation of mitosis, proliferation, differentiation, and survival of mammalian cells. ERKs 1 and 2 are the best-studied members of the MAPK family that transduce extracellular stimuli into intracellular re- sponses. ERK activation in spinal dorsal horn neurons is driven by afferent activity of primary nociceptors (Ji et al. 1999). An increase in excitation in the superﬁcial dorsal horn, caused by transmitters glutamate and substance P (Conn and Pin 1997), stimulates PKA and PKC, leading to the activation of ERK to pERK by its downstream phosphorylation, and subsequently to all
those processes that follow ERK signaling. Elevation of cytoplasmic Ca2+ also appears essential for noxious stimulation-induced ERK activation (Lever et al.
2003). All the above-listed factors have been shown to enhance nociceptive
processing in the spinal cord dorsal horn, one biochemical correlate of so- called “central sensitization” (Ji and Woolf 2001). Bupivacaine has been shown to suppress neuronal ERK activation in spinal cord slices by a variety of sub- stances, including those that act directly on dorsal horn neurons, giving insight into potential mechanisms for spinal anesthesia.
Presynaptic or Postsynaptic Sites for LA Inhibition
Synaptic transmission in the spinal cord may be inhibited directly by LAs through the modiﬁcation of postsynaptic receptors as well as the blockade
of presynaptic calcium channels that must function to stimulate the release of transmitters (see Sect. 2.3, above). Bupivacaine can inhibit the elevation in dorsal horn neurons of pERK by excitatory ionotropic receptor agonists such as capsaicin, S-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/kainate, and NMDA, but with no LA effects on ERK activa- tion by GPCRs such as bradykinin, mGluR 1/5, and NK-1, by immunohis- tochemistry using spinal slices as assayed (Yanagidate and Strichartz 2006). Receptors for capsaicin, acting on the ligand-gated vanilloid-sensitive tran- sient receptor potential channel, TRPV1, are exclusively located at presynap- tic terminals of nociceptive afferents. Thus, bupivacaine almost certainly has
presynaptic actions, blocking Na+ channels, Ca2+ channels, and TRPV1 re- ceptor channels. On the other hand, AMPA and NMDA are classically located on postsynaptic sites, but also can be found on presynaptic endings. Presy-
naptic NMDA autoreceptors appear to facilitate the release of substance P and glutamate in a positive-feedback network (Liu et al. 1997). In contrast, AMPA receptors have an excitatory postsynaptic action, but also a strong
inhibitory presynaptic action on glutamate release from primary afferent terminals in the superﬁcial dorsal horn (Lee et al. 2002). Therefore, the in- hibitory effects of bupivacaine on NMDA-induced pERK might be at both pre- and postsynaptic sites and AMPA-induced pERK must be occurring pre-
dominantly through activation of postsynaptic receptors. Ionotropic gluta- mate receptors are known targets of local anesthetics (Nishizawa et al. 2002; Sugimoto et al. 2003). Other studies also showed the “presynaptic” actions of LAs to inhibit Ca2+ channels in dorsal root ganglion neurons, channels
that are essential for the release of neurotransmitter during depolarization (Rane et al. 1987). Therefore, LAs will act at many locations to alter func- tions during spinal and epidural anesthesia, and any understanding of LA actions in the spinal cord should include their effects on a variety of ion channels, membrane-related enzymes, and intracellular second messenger pathways.
Local anesthetics at the neuraxis often provide analgesia synergistic with the actions of other drugs. Examples of the latter include Ca2+ channel blockers and opiates.