THE past quarter century has witnessed an explosion in the understanding of pain, characterized by the elucidation of anatomic pathways, and the identification of receptors, ion channels, and neurotransmitters that are involved in the transmission of nociceptive information from primary sensory neurons in the dorsal root ganglion (DRG) to the brain.1In parallel, it has become apparent that the emergence of chronic pain, an important and difficult-to-treat condition, involves defined alterations in nociceptive processing in the peripheral nervous system, at the level of the spinal cord, and in the brain. But despite advances in basic and clinical sciences and the concentrated efforts of many academic and pharmaceutical research laboratories, the development of novel effective treatments for chronic pain has been disappointingly slow. In part, the challenge to the development of pain therapeutics is a predictable result of the parsimonious use by the nervous system of a limited repertoire of neurotransmitters, receptors, and ion channels at multiple sites and in many pathways subserving different functions. Therefore, potent small molecules designed to interrupt nociceptive neurotransmission often have “off-target” adverse effects resulting from actions of these molecules in pathways subserving other, non–pain-related functions. In response, several groups have begun to explore the possibility of using gene transfer to achieve analgesic effects, and in this issue of the Journal, Tzabazis et al. 2present the results of a study of gene transfer using a herpes simplex virus (HSV)–based vector that significantly extends the range of gene transfer in the treatment of pain.
The rationale for applying gene transfer techniques to the treatment of pain is based on the presumption that expression of transgene products (usually short-lived potent peptides) in a restricted anatomical distribution may be used to reduce pain perception through modulation of nociceptive neurotransmission at an identified site, with off-target effects limited by the limited anatomic distribution of transgene expression. Transduction of meninges accomplished by intrathecal injection of “naked” plasmid or liposome-encapsulated DNA, or by injection of recombinant viral vectors created from adenovirus or adenoassociated virus to express inhibitory neurotransmitters (e.g. , beta endorphin) or antiinflammatory cytokines (e.g. , interleukin 2 or interleukin 10) has been shown to reduce pain-related behaviors in several animal models of inflammatory and neuropathic pain.3These analgesic effects are presumed to result from modulation of neurotransmission at the synapse between the central afferents of first-order nociceptive neurons (whose cell bodies lie in the DRG) onto second-order neurons located in the dorsal horn of spinal cord. Modulation of nociceptive neurotransmission at that synapse in the spinal cord can also be achieved by peripheral inoculation of vectors created from recombinant HSV, relying on the natural neurotropism of HSV to achieve efficient transport from the periphery to sensory neurons in the DRG.4Reduction of pain related behaviors using HSV-based vectors has been demonstrated in models of chronic inflammatory and neuropathic pain with vectors expressing enkephalin, glutamic acid decarboxylase (to produce gamma amino butyric acid), glial cell–derived neurotrophic factor, interleukin 4, and the truncated soluble tumor necrosis factor receptor.5
In the current report, Tzabazis et al. used a recombinant HSV-based vector encoding an antisense sequence to the calcitonin gene–related peptide (CGRP) gene. They demonstrate that application of the vector to the skin resulted in a significant reduction in CGRP expression in primary afferents in the DRG, producing a significant attenuation in thermal C-fiber hyperalgesia after topical application of capsaicin. CGRP is expressed in 45–70% of lumbar DRG neurons, a majority of which are nociceptors. Although the precise role of CGRP in nociception has not been established, spinal delivery of CGRP antagonists has previously been shown to reduce pain-related behaviors in a variety of models, and in the mouse, strain-related differences in sensitivity to noxious heat correlates with strain-dependent differences in CGRP expression and sensitivity.6In contrast to the transient effects produced by spinal delivery of CGRP antagonists, HSV-mediated knockdown of CGRP expression resulted in an analgesic effect that persisted for 12 weeks. An advantage to gene transfer of an antisense sequence is that no foreign gene products are released from the transduced neurons, although it will be crucial to demonstrate in future studies that HSV-mediated knockdown of gene expression by this antisense expressing vector is limited to CGRP. Nonetheless, the observation by Tzabazis et al. that the HSV vector produced a significant reduction in CGRP gene expression in a majority of nociceptor afferents in the DRG after superficial application to the skin is impressive, and extension of this work to an animal model of chronic pain would serve as an important preclinical step in the development of a treatment for chronic pain.
Gene transfer for the treatment of pain is slowly moving toward the clinic. A trial using intrathecal injection of a plasmid encoding interleukin 10 to treat chronic neuropathic pain has been proposed, and a second trial to establish safety and dose range of a nonreplicating HSV vector encoding enkephalin in patients with pain caused by cancer has received sponsorship. Chronic pain represents an important clinical problem for which there is a substantial unmet need; the current report provides additional hope for the future in this regard.
* Department of Neurology, University of Michigan, and GRECC and Neurology, VA Ann Arbor Healthcare System, Ann Arbor, Michigan. † Department of Neurology, University of Michigan, and Neurology, VA Ann Arbor Healthcare System, Ann Arbor, Michigan. djfink@umich.edu
