- Email: [email protected]
Peptide autoreceptors: does an autoreceptor for substance P exist?
P 12 Weiss, J. M., Morgan, P. H., Lutz, M. W. and Kenakin, T. P. (1996) J. Theor. Biol. 178, 169–182 13 Brown, G. P. and Pasternak, G. W. (1998) J. Pharmacol. Exp. Ther. 286, 376 14 Bouaboula, M. et al. (1997) J. Biol. Chem. 272, 22330–22339 15 Kenakin, T. P. (1984) Br. J. Pharmacol. 81, 131 16 Kenakin, T. P. (1995) Trends Pharmacol. Sci. 16, 256 17 Kenakin, T. P. (1997) Ann. New York Acad. Sci. 812, 116 18 Kjelsberg, M. A., Cotecchia, S., Ostrowski, J., Caron, M. G. and Lefkowitz, R. J. (1992) J. Biol. Chem. 267, 143 19 Spengler, D. et al. (1993) Nature 365, 170–175 20 Rob, S. et al. (1994) EMBO J. 13, 1325–1330 21 Krumins, A. M. and Barber, R. (1997) Mol. Pharmacol. 52, 144 22 Berg, K. A. et al. (1998) Mol. Pharmacol. 54, 94–104 23 Wiens, B. L., Nelson, C. S. and Neve, K. A. (1998) Mol. Pharmacol. 54, 435 24 Van Hooft, J. A. and Vijverberg, H. P. M. (1996) Br. J. Pharmacol. 117, 839–846
R
Marzia Malcangio and Norman G. Bowery The presence of autoreceptors for simple neurotransmitters at synapses in the mammalian nervous system is well established. By contrast, the evidence for such receptors modifying neuropeptide transmission is less obvious. Probably the most well characterized of the neuropeptides is substance P (SP), which appears to play a major role as a primary afferent modulator. This article highlights evidence to support the existence of autoreceptors that might modulate the release of this neuropeptide and which, therefore, could be important in the design of drugs affecting SP function, not only in sensory processing, but also elsewhere in the brain. Throughout the central and peripheral nervous systems the release of transmitters from neurones is controlled, in part, by several ionotropic and metabotropic (G-proteincoupled) presynaptic receptors. These terminal receptors can be either hetero- or auto-receptors depending on activation by transmitters from nearby neurones or from the same neurone, respectively. Autoreceptor activation by an endogenous transmitter can either facilitate or inhibit its further release depending on the nature of the synapse. In some cases, autoreceptors have been the target for novel drugs such as the α2-adreno-autoreceptor antagonists, which have been introduced successfully for the treatment of clinical depression1. Although somatodendritic autoreceptors that modulate the firing rate of
N
C
I
P
L
E
S
25 Keith, D. E. et al. (1996) J. Biol. Chem. 271, 19021–19024 26 Yu, Y. et al. (1997) J. Biol. Chem. 272, 28869–28874 27 Blake, A. D., Bot, G., Freeman, J. C. and Reisine, T. (1997) J. Biol. Chem. 272, 782–790 28 Chidiac, P., Hebert, T. E., Valiquette, M., Dennis, M. and Bouvier, M. (1994) Mol. Pharmacol. 45, 490 29 Jansson, C. C. et al. (1998) Mol. Pharmacol. 53, 963 30 Chakraborty, M., Chatterjee, D., Kellokumpu, S., Rasmussen, H. and Baron, R. (1991) Science 251, 1078–1082 31 Ikezu, T., Okamoto, T., Ogata, E. and Nishimoto, I. (1992) FEBS Lett. 311, 29–32 32 Horne, W. C., Shyu, J-F., Chakraborty, M. and Baron, R. (1994) Trends Endocrinol. Metab. 5, 395–401 33 Kenakin, T. P. and Morgan, P. H. (1989) Mol. Pharmacol. 35, 214–222 34 McLatchie, L. M. et al. (1998) Nature 393, 333–339 35 Sato, M. et al. (1995) J. Biol. Chem. 270, 15269–15276 36 Nanoff, C., Mitteraurer, T., Roka, Hohenegger, M. and Freissmuth, M. (1995) Mol. Pharmacol. 48, 806–817
V
Peptide autoreceptors: does an autoreceptor for substance P exist?
I
I
E
W
P
O
I
N
T
the neurone have been described, autoreceptors are normally assumed to be located on axon terminals where they contribute to the regulation of the chemical signal that originates from nerve endings2. In presynaptic terminals, fast and slow transmitters are stored in vesicles that release their content via exocytosis. Activation of autoreceptors and their second-messenger systems can modulate transmitter release. This could be, for example, by promoting the entry of external Ca21 or mobilizing Ca21 from intracellular stores, thus affecting vesicle trafficking. However, both the storage conditions and the release process for fast transmitters (e.g. glutamate) are distinct from those of slow transmitters (e.g. neuropeptides) and this means that the modulation of their release could imply different mechanisms (for example, modulation of different Ca21 channels). Fast neurotransmitters (with the exception of noradrenaline) are confined to small synaptic vesicles (SSVs), which are close to the active synaptic zone and sensitive to elevations in Ca21 concentration (through voltage-sensitive N-type Ca21 channels) at the site of exocytosis. In contrast, neuropeptides are mostly, but not exclusively, present in large dense core vesicles (LDCVs), which often remain distant from the synaptic zone and undergo exocytosis at structurally non-specialized areas along the plasmalemma. Release of LDCV content is triggered by small increases in Ca21 ion concentration as the ion enters the cytoplasm possibly via L-type Ca21 channels3. Release of neurotransmitters from SSV occurs after vesicle fusion with the plasma membrane and exposure to the extracellular medium. In contrast, peptide release from LDCVs requires alkalinization of the inside of secretory vesicles before exocytosis. This rise in pH is caused by an increase in cytoplasmic Ca21, which facilitates peptide solubilization and release4. In some cases, such as with the undecapeptide substance P (SP) and the amino acid glutamic acid in the central terminals of primary afferent fibres, fast and slow transmitters are co-localized5.
0165-6147/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S0165-6147(99)01388-7
TiPS – October 1999 (Vol. 20)
M. Malcangio, Wellcome RCD Fellow, Neuroscience Research Centre, Guy’s, King’s and St Thomas’ School of Biomedical Sciences, Kings College London, London, UK SE1 7EH. Email: Marzia. [email protected] and N. Bowery, Head of Division and Department of Pharmacology, The Medical School, University of Birmingham, Edgbaston, Birmingham, UK B15 2TT. Email: n.g.bowery@ bham.ac.uk
405
V
I
E
W
P
O
I
N
T
SP Ins(1,4,5)P3
SP
SP PLC G
G PLC PIP2 NK1 receptor
Ca2+
DAG
NK1 receptor
PKC P
Glu
NMDA receptor Ca2+
Glu
N-type Ca2+ channel
Ca2+
L-type channel
K+
trends in Pharmacological Sciences
Fig. 1. Hypothesis for substance P (SP) and glutamate (Glu) interaction at the central and/or peripheral terminals of sensory nociceptive neurones. Glu is stored in small vesicles (orange) whereas SP is in large vesicles (light blue). Stimulation of sensory neurones induces Glu release following Ca21 ion entry via N-type channels at the synaptic active zone. Following Ca21 ion entry via L-type channels and increased Ca21 ion concentration in the bulk cytoplasm, SP is released at ectopic sites outside of the active zone. Once released, Glu facilitates its own release as well as SP release via the activation of NMDA receptors and L-type Ca21 channels. Meanwhile, released SP promotes Glu release following binding to the tachykinin NK1 receptor and protein kinase C (PKC)-induced phosphorylation of NMDA receptor (left-hand side of figure). SP can promote its own release via production of inositol-1,4,5-trisphosphate [Ins(1,4,5)P3] and consequent release of Ca21 ions from internal stores (left-hand side of figure). However, SP could also inhibit its own release (and possibly Glu release) by blocking K1 channels, thus affecting action potential propagation (right-hand side of figure). In this model, the NMDA receptor could be acting as a presynaptic amplifier, whereas the NK1 receptor would be a safety device that limits positive feedback mechanisms to prevent overstimulation. Abbreviations: DAG, diacylglycerol; G, heterotrimeric G protein; PIP2, phosphatidylinositol-4,5-diphosphate; PLC, phospholipase C.
In contrast to fast transmitters, peptides are released following high-frequency bursts of stimulation and, once released, some of them (e.g. neurokinin A) persist in the extracellular space because they are not efficiently rescued by the presynaptic element and are resistant to degradation6. However, this is not true for SP, which is discretely released in the superficial laminae of the dorsal horn, where SP-containing fibres terminate6, following cutaneous noxious stimulation.
SP autoreceptor Morphological evidence for the existence of peptide autoreceptors has been demonstrated but is very limited. The existence of autoreceptors for SP on sensory fibre terminals within the spinal cord and in the periphery could have important consequences because SP is believed to play an important role in nociceptive transmission. However, evidence of their presence is both supporting and conflicting.
Anatomical evidence SP is contained in LDCVs of peripheral (skin and viscera) and central (spinal cord) terminals of a sub-population of unmyelinated sensory neurones, the cell bodies of which are located in the dorsal root ganglia (DRG). There is anatomical evidence for the presence of SP receptors on 406
TiPS – October 1999 (Vol. 20)
peripheral axons and cell bodies of SP-containing fibres but not on spinal cord SP-containing axons. Carlton and colleagues7 have provided evidence for the presence of SP receptors associated with some unmyelinated axons in the rat glabrous skin. Subsequently, a receptor for SP has been identified on DRG neurones, which stains for the peptide8 and expresses mRNA for the tachykinin NK1 receptor9. Other groups have attempted to show the presence of NK1 receptors on central terminals of SP fibres but their results fail to support this hypothesis due to the increased (rather than decreased) number of binding sites for SP in the dorsal horn of neonatally capsaicin-treated rats10,11 or deafferented rats11,12. However, in these studies, any loss of presynaptic sites could have gone undetected by an upregulation of postsynaptic SP sites due to the removal of SP innervation. Furthermore, the number of presynaptic sites might be much lower than the number of postsynaptic sites. Alternatively, other isoforms of the receptor (which are not recognized by available antibodies) might be present at presynaptic sites. Recently, using a new antibody, the presence of NK1 receptor in laminae I and II of the dorsal horn has been shown unequivocally13,14. It seems possible that some of these receptors are autoreceptors on SP-containing terminals that terminate within these two laminae.
Functional evidence Using a pharmacological approach, it has been shown15 that two selective antagonists for the rat NK1 receptor, RP67580 and SR140333, could increase the release of SP from the spinal cord evoked by electrical stimulation of C fibres without changing the peptide basal outflow. This suggests that SP acting on NK1 receptors exerted a negative feedback on its own release via activation of an inhibitory NK1 autoreceptor. Alternatively, SP could have promoted the release of an inhibitory transmitter (such as GABA or enkephalins) from interneurones that would, in turn, inhibit the release of the peptide acting on presynaptic heteroreceptors. However, there is little evidence for NK1 receptor presence on stalked or islet neurones in lamina II of the dorsal horn11. In DRG neurones, anatomical evidence for autoreceptors8,9 is supported by the observation that neurones respond to the application of the peptide by an inward current that results from the opening of nonselective ion channels (Na1 and Ca21). The receptor appears to belong to the metabotropic NK1 subtype9. However, all three tachykinin receptor subtypes (NK1, NK2 and NK3) are known to be expressed by DRG neurones and their activation promotes an increase in intracellular Ca21 concentration16. Taken together, these observations suggest that NK1 receptors can be synthesized in the cell bodies of the umyelinated fibres in the DRG and subsequently anterogradely transported along axons to peripheral – and perhaps central – terminals. Thus, once released, SP can activate the postsynaptic NK1 receptor and depolarize the neuronal membrane, mostly by blocking K1 channels. Long-lasting NK1 receptor-mediated
V block of K1 channels on the presynaptic terminal, resulting in a positive shift in terminal membrane potential, might lead to inactivation of voltage-dependent channels that support the action potential. At the presynaptic site, depolarization, other than by an invading action potential, does not enhance transmission. Primary afferent depolarization (PAD) effectively produces a long-lasting inhibitory effect on the transmission of signals from primary afferents to spinal neurones.
I
E
W
P
O
I
N
T
consequences, for example in the therapy of chronic pain and that associated with migraine. In addition, given that SP receptor antagonists have recently been shown to produce antidepressant activity in man20, it could also have therapeutic consequences in, for example, the treatment of affective disorders. Any increase in selectivity for the postsynaptic receptors would be expected to improve the characteristics of the drug. Selected references
SP–glutamate interaction SP is co-localized with glutamate in some of the central terminals of unmyelinated fibres5. The release of glutamate, which is a fast neurotransmitter, occurs under different conditions to that of SP, a slow neuromodulator, and the release of either one is controlled by heteroreceptor activation. Examples of receptors that contribute to this control mechanism are m-receptors for opioids, GABAB receptors for GABA, both of which are inhibitory, and P2X receptors for ATP, which facilitate release. In addition, there are the NMDA and NK1 receptors, which also modify release. The nature of the interaction between SP and glutamate and their respective pre- and postsynaptic receptors is an important consideration. Both transmitters clearly play a role in pain transmission both within the spinal cord and in the periphery. Even though SP might well inhibit its own release, this peptide has been shown to increase basal and evoked release of glutamate from the spinal cord17. If SP influences its own release, it is possibly doing so via K1 channel block and membrane depolarization following activation of NK1 receptors. It is more likely that the effect on glutamate release is due to some indirect mechanism(s). For example, activation of protein kinase C might cause phosphorylation of N-type Ca21 channels or phosphorylation of the NMDA receptor, or both18. Both of these mechanisms will promote Ca21 entry to increase the release of glutamate (Fig. 1). However, NMDA receptor activation also promotes SP release19 possibly by raising the cytoplasmic concentration of Ca21. The presence of an inhibitory autoreceptor for SP on primary afferent terminals would enable NK1 receptor antagonists to enhance the release of SP (Ref. 15). If, at the same time, glutamate release is occurring unchecked consequent upon a nociceptive event this would then be free to act on the afferent terminal NMDA receptors to increase the release of SP still further. This would tend to oppose the potential analgesic activity of the SP antagonist.
Concluding remarks Autoreceptors for neuropeptides do exist but more evidence is needed to establish their participation in transmission processing. At present, the most-studied system is the SP sensory pathway, but even here there is still very little information about the characteristics of the presynaptic receptors. If it is eventually discovered that this site differs subtly from postsynaptic NK1, NK2 and NK3 receptors, this could have important therapeutic
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
deBoer, T. and Ruigt, G. S. F. (1995) CNS Drugs (Suppl. 1), 29–38 Langer, S. Z. (1994) Trends Pharmacol. Sci. 18, 95–99 Verhage, M. et al. (1991) Neuron 6, 517–524 Han, W., Li, D., Stout, A. K., Takimoto, K. and Levitan, E. S. (1999) J. Neurosci. 19, 900–905 DeBiasi, S. and Rustioni, A. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 7820–7824 Duggan, A. W., Hope P. J., Jarrott, B., Schaible, H-G. and FleetwoodWalker, S. M. (1990) Neuroscience 35, 195–202 Carlton, S. M., Zhou, S. and Coggeshall, R. E. (1996) Brain Res. 734, 103–108 Hu, H-Z., Li, Z-W. and Si, J-Q. (1997) Neuroscience 77, 535–541 Li, H-S. and Zhao, Z-Q. (1998) Eur. J. Neurosci. 10, 1292–1299 Helke, C. J., Charlton, C. G. and Wiley, R. G. (1986) Neuroscience 19, 523–533 Brown, J. L., Liu, H., Maggio, J. E., Vigna, S. R., Mantyh, P. W. and Basbaum, A. I. (1995) J. Comp. Neurol. 356, 327–344 Yashpal, K., Dam, T. and Quirion, R. (1991) Brain Res. 552, 240–247 McLeod, A. L., Krause, J. E., Cuello, A. C. and Ribeiro-Da-Silva, A. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 15775–15780 Yu, X-H. et al. (1999) J. Neurosci.19, 3545–3555 Malcangio, M. and Bowery, N. G. (1994) Br. J. Pharmacol. 113, 635–641 Brechenmacher, C., Larmet, Y., Feltz, P. and Rodeau, J-L. (1998) Neurosci. Lett. 241, 159–162 Kangrga, I. and Randic, M. (1991) J. Neurosci. 10, 2026–2038 Rusin, K. I., Bleakman, D., Chard, P. S., Randic, M. and Miller, R. J. (1993) J. Neurochem. 60, 952–960 Malcangio, M., Fernandes, K. and Tomlinson D. R. (1998) Br. J. Pharmacol. 125, 1625–1626 Kramer, M. S. et al. (1998) Science 281, 1640–1645
Chemical names RP67580: (1-imino-2-[2-methoxyphenyl]ethyl)-7,7diphenyl-4-perhydroisoindolone (3aR,7aR) SR140333: (S)-1-(2-[3-{3,4-dichlorophenyl}-1-{3-isopropoxyphenylacetyl}piperidin-3-yl]ethyl)-4-phenyl1-azoniabicyclo(2.2.2)octane chloride
Acknowledgement Special thanks to Dr Martyn Jones for fruitful discussion while preparing the manuscript.
Letters to the Editor The Editor welcomes correspondence from readers of Trends in Pharmacological Sciences on any articles published. Please state clearly whether or not you wish the letter to be considered for publication. If a letter is considered suitable for publication, it will be sent to the author of the original article for their response; both the letter and reply will be published together.
TiPS – October 1999 (Vol. 20)
407
R
Marzia Malcangio and Norman G. Bowery The presence of autoreceptors for simple neurotransmitters at synapses in the mammalian nervous system is well established. By contrast, the evidence for such receptors modifying neuropeptide transmission is less obvious. Probably the most well characterized of the neuropeptides is substance P (SP), which appears to play a major role as a primary afferent modulator. This article highlights evidence to support the existence of autoreceptors that might modulate the release of this neuropeptide and which, therefore, could be important in the design of drugs affecting SP function, not only in sensory processing, but also elsewhere in the brain. Throughout the central and peripheral nervous systems the release of transmitters from neurones is controlled, in part, by several ionotropic and metabotropic (G-proteincoupled) presynaptic receptors. These terminal receptors can be either hetero- or auto-receptors depending on activation by transmitters from nearby neurones or from the same neurone, respectively. Autoreceptor activation by an endogenous transmitter can either facilitate or inhibit its further release depending on the nature of the synapse. In some cases, autoreceptors have been the target for novel drugs such as the α2-adreno-autoreceptor antagonists, which have been introduced successfully for the treatment of clinical depression1. Although somatodendritic autoreceptors that modulate the firing rate of
N
C
I
P
L
E
S
25 Keith, D. E. et al. (1996) J. Biol. Chem. 271, 19021–19024 26 Yu, Y. et al. (1997) J. Biol. Chem. 272, 28869–28874 27 Blake, A. D., Bot, G., Freeman, J. C. and Reisine, T. (1997) J. Biol. Chem. 272, 782–790 28 Chidiac, P., Hebert, T. E., Valiquette, M., Dennis, M. and Bouvier, M. (1994) Mol. Pharmacol. 45, 490 29 Jansson, C. C. et al. (1998) Mol. Pharmacol. 53, 963 30 Chakraborty, M., Chatterjee, D., Kellokumpu, S., Rasmussen, H. and Baron, R. (1991) Science 251, 1078–1082 31 Ikezu, T., Okamoto, T., Ogata, E. and Nishimoto, I. (1992) FEBS Lett. 311, 29–32 32 Horne, W. C., Shyu, J-F., Chakraborty, M. and Baron, R. (1994) Trends Endocrinol. Metab. 5, 395–401 33 Kenakin, T. P. and Morgan, P. H. (1989) Mol. Pharmacol. 35, 214–222 34 McLatchie, L. M. et al. (1998) Nature 393, 333–339 35 Sato, M. et al. (1995) J. Biol. Chem. 270, 15269–15276 36 Nanoff, C., Mitteraurer, T., Roka, Hohenegger, M. and Freissmuth, M. (1995) Mol. Pharmacol. 48, 806–817
V
Peptide autoreceptors: does an autoreceptor for substance P exist?
I
I
E
W
P
O
I
N
T
the neurone have been described, autoreceptors are normally assumed to be located on axon terminals where they contribute to the regulation of the chemical signal that originates from nerve endings2. In presynaptic terminals, fast and slow transmitters are stored in vesicles that release their content via exocytosis. Activation of autoreceptors and their second-messenger systems can modulate transmitter release. This could be, for example, by promoting the entry of external Ca21 or mobilizing Ca21 from intracellular stores, thus affecting vesicle trafficking. However, both the storage conditions and the release process for fast transmitters (e.g. glutamate) are distinct from those of slow transmitters (e.g. neuropeptides) and this means that the modulation of their release could imply different mechanisms (for example, modulation of different Ca21 channels). Fast neurotransmitters (with the exception of noradrenaline) are confined to small synaptic vesicles (SSVs), which are close to the active synaptic zone and sensitive to elevations in Ca21 concentration (through voltage-sensitive N-type Ca21 channels) at the site of exocytosis. In contrast, neuropeptides are mostly, but not exclusively, present in large dense core vesicles (LDCVs), which often remain distant from the synaptic zone and undergo exocytosis at structurally non-specialized areas along the plasmalemma. Release of LDCV content is triggered by small increases in Ca21 ion concentration as the ion enters the cytoplasm possibly via L-type Ca21 channels3. Release of neurotransmitters from SSV occurs after vesicle fusion with the plasma membrane and exposure to the extracellular medium. In contrast, peptide release from LDCVs requires alkalinization of the inside of secretory vesicles before exocytosis. This rise in pH is caused by an increase in cytoplasmic Ca21, which facilitates peptide solubilization and release4. In some cases, such as with the undecapeptide substance P (SP) and the amino acid glutamic acid in the central terminals of primary afferent fibres, fast and slow transmitters are co-localized5.
0165-6147/99/$ – see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S0165-6147(99)01388-7
TiPS – October 1999 (Vol. 20)
M. Malcangio, Wellcome RCD Fellow, Neuroscience Research Centre, Guy’s, King’s and St Thomas’ School of Biomedical Sciences, Kings College London, London, UK SE1 7EH. Email: Marzia. [email protected] and N. Bowery, Head of Division and Department of Pharmacology, The Medical School, University of Birmingham, Edgbaston, Birmingham, UK B15 2TT. Email: n.g.bowery@ bham.ac.uk
405
V
I
E
W
P
O
I
N
T
SP Ins(1,4,5)P3
SP
SP PLC G
G PLC PIP2 NK1 receptor
Ca2+
DAG
NK1 receptor
PKC P
Glu
NMDA receptor Ca2+
Glu
N-type Ca2+ channel
Ca2+
L-type channel
K+
trends in Pharmacological Sciences
Fig. 1. Hypothesis for substance P (SP) and glutamate (Glu) interaction at the central and/or peripheral terminals of sensory nociceptive neurones. Glu is stored in small vesicles (orange) whereas SP is in large vesicles (light blue). Stimulation of sensory neurones induces Glu release following Ca21 ion entry via N-type channels at the synaptic active zone. Following Ca21 ion entry via L-type channels and increased Ca21 ion concentration in the bulk cytoplasm, SP is released at ectopic sites outside of the active zone. Once released, Glu facilitates its own release as well as SP release via the activation of NMDA receptors and L-type Ca21 channels. Meanwhile, released SP promotes Glu release following binding to the tachykinin NK1 receptor and protein kinase C (PKC)-induced phosphorylation of NMDA receptor (left-hand side of figure). SP can promote its own release via production of inositol-1,4,5-trisphosphate [Ins(1,4,5)P3] and consequent release of Ca21 ions from internal stores (left-hand side of figure). However, SP could also inhibit its own release (and possibly Glu release) by blocking K1 channels, thus affecting action potential propagation (right-hand side of figure). In this model, the NMDA receptor could be acting as a presynaptic amplifier, whereas the NK1 receptor would be a safety device that limits positive feedback mechanisms to prevent overstimulation. Abbreviations: DAG, diacylglycerol; G, heterotrimeric G protein; PIP2, phosphatidylinositol-4,5-diphosphate; PLC, phospholipase C.
In contrast to fast transmitters, peptides are released following high-frequency bursts of stimulation and, once released, some of them (e.g. neurokinin A) persist in the extracellular space because they are not efficiently rescued by the presynaptic element and are resistant to degradation6. However, this is not true for SP, which is discretely released in the superficial laminae of the dorsal horn, where SP-containing fibres terminate6, following cutaneous noxious stimulation.
SP autoreceptor Morphological evidence for the existence of peptide autoreceptors has been demonstrated but is very limited. The existence of autoreceptors for SP on sensory fibre terminals within the spinal cord and in the periphery could have important consequences because SP is believed to play an important role in nociceptive transmission. However, evidence of their presence is both supporting and conflicting.
Anatomical evidence SP is contained in LDCVs of peripheral (skin and viscera) and central (spinal cord) terminals of a sub-population of unmyelinated sensory neurones, the cell bodies of which are located in the dorsal root ganglia (DRG). There is anatomical evidence for the presence of SP receptors on 406
TiPS – October 1999 (Vol. 20)
peripheral axons and cell bodies of SP-containing fibres but not on spinal cord SP-containing axons. Carlton and colleagues7 have provided evidence for the presence of SP receptors associated with some unmyelinated axons in the rat glabrous skin. Subsequently, a receptor for SP has been identified on DRG neurones, which stains for the peptide8 and expresses mRNA for the tachykinin NK1 receptor9. Other groups have attempted to show the presence of NK1 receptors on central terminals of SP fibres but their results fail to support this hypothesis due to the increased (rather than decreased) number of binding sites for SP in the dorsal horn of neonatally capsaicin-treated rats10,11 or deafferented rats11,12. However, in these studies, any loss of presynaptic sites could have gone undetected by an upregulation of postsynaptic SP sites due to the removal of SP innervation. Furthermore, the number of presynaptic sites might be much lower than the number of postsynaptic sites. Alternatively, other isoforms of the receptor (which are not recognized by available antibodies) might be present at presynaptic sites. Recently, using a new antibody, the presence of NK1 receptor in laminae I and II of the dorsal horn has been shown unequivocally13,14. It seems possible that some of these receptors are autoreceptors on SP-containing terminals that terminate within these two laminae.
Functional evidence Using a pharmacological approach, it has been shown15 that two selective antagonists for the rat NK1 receptor, RP67580 and SR140333, could increase the release of SP from the spinal cord evoked by electrical stimulation of C fibres without changing the peptide basal outflow. This suggests that SP acting on NK1 receptors exerted a negative feedback on its own release via activation of an inhibitory NK1 autoreceptor. Alternatively, SP could have promoted the release of an inhibitory transmitter (such as GABA or enkephalins) from interneurones that would, in turn, inhibit the release of the peptide acting on presynaptic heteroreceptors. However, there is little evidence for NK1 receptor presence on stalked or islet neurones in lamina II of the dorsal horn11. In DRG neurones, anatomical evidence for autoreceptors8,9 is supported by the observation that neurones respond to the application of the peptide by an inward current that results from the opening of nonselective ion channels (Na1 and Ca21). The receptor appears to belong to the metabotropic NK1 subtype9. However, all three tachykinin receptor subtypes (NK1, NK2 and NK3) are known to be expressed by DRG neurones and their activation promotes an increase in intracellular Ca21 concentration16. Taken together, these observations suggest that NK1 receptors can be synthesized in the cell bodies of the umyelinated fibres in the DRG and subsequently anterogradely transported along axons to peripheral – and perhaps central – terminals. Thus, once released, SP can activate the postsynaptic NK1 receptor and depolarize the neuronal membrane, mostly by blocking K1 channels. Long-lasting NK1 receptor-mediated
V block of K1 channels on the presynaptic terminal, resulting in a positive shift in terminal membrane potential, might lead to inactivation of voltage-dependent channels that support the action potential. At the presynaptic site, depolarization, other than by an invading action potential, does not enhance transmission. Primary afferent depolarization (PAD) effectively produces a long-lasting inhibitory effect on the transmission of signals from primary afferents to spinal neurones.
I
E
W
P
O
I
N
T
consequences, for example in the therapy of chronic pain and that associated with migraine. In addition, given that SP receptor antagonists have recently been shown to produce antidepressant activity in man20, it could also have therapeutic consequences in, for example, the treatment of affective disorders. Any increase in selectivity for the postsynaptic receptors would be expected to improve the characteristics of the drug. Selected references
SP–glutamate interaction SP is co-localized with glutamate in some of the central terminals of unmyelinated fibres5. The release of glutamate, which is a fast neurotransmitter, occurs under different conditions to that of SP, a slow neuromodulator, and the release of either one is controlled by heteroreceptor activation. Examples of receptors that contribute to this control mechanism are m-receptors for opioids, GABAB receptors for GABA, both of which are inhibitory, and P2X receptors for ATP, which facilitate release. In addition, there are the NMDA and NK1 receptors, which also modify release. The nature of the interaction between SP and glutamate and their respective pre- and postsynaptic receptors is an important consideration. Both transmitters clearly play a role in pain transmission both within the spinal cord and in the periphery. Even though SP might well inhibit its own release, this peptide has been shown to increase basal and evoked release of glutamate from the spinal cord17. If SP influences its own release, it is possibly doing so via K1 channel block and membrane depolarization following activation of NK1 receptors. It is more likely that the effect on glutamate release is due to some indirect mechanism(s). For example, activation of protein kinase C might cause phosphorylation of N-type Ca21 channels or phosphorylation of the NMDA receptor, or both18. Both of these mechanisms will promote Ca21 entry to increase the release of glutamate (Fig. 1). However, NMDA receptor activation also promotes SP release19 possibly by raising the cytoplasmic concentration of Ca21. The presence of an inhibitory autoreceptor for SP on primary afferent terminals would enable NK1 receptor antagonists to enhance the release of SP (Ref. 15). If, at the same time, glutamate release is occurring unchecked consequent upon a nociceptive event this would then be free to act on the afferent terminal NMDA receptors to increase the release of SP still further. This would tend to oppose the potential analgesic activity of the SP antagonist.
Concluding remarks Autoreceptors for neuropeptides do exist but more evidence is needed to establish their participation in transmission processing. At present, the most-studied system is the SP sensory pathway, but even here there is still very little information about the characteristics of the presynaptic receptors. If it is eventually discovered that this site differs subtly from postsynaptic NK1, NK2 and NK3 receptors, this could have important therapeutic
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
deBoer, T. and Ruigt, G. S. F. (1995) CNS Drugs (Suppl. 1), 29–38 Langer, S. Z. (1994) Trends Pharmacol. Sci. 18, 95–99 Verhage, M. et al. (1991) Neuron 6, 517–524 Han, W., Li, D., Stout, A. K., Takimoto, K. and Levitan, E. S. (1999) J. Neurosci. 19, 900–905 DeBiasi, S. and Rustioni, A. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 7820–7824 Duggan, A. W., Hope P. J., Jarrott, B., Schaible, H-G. and FleetwoodWalker, S. M. (1990) Neuroscience 35, 195–202 Carlton, S. M., Zhou, S. and Coggeshall, R. E. (1996) Brain Res. 734, 103–108 Hu, H-Z., Li, Z-W. and Si, J-Q. (1997) Neuroscience 77, 535–541 Li, H-S. and Zhao, Z-Q. (1998) Eur. J. Neurosci. 10, 1292–1299 Helke, C. J., Charlton, C. G. and Wiley, R. G. (1986) Neuroscience 19, 523–533 Brown, J. L., Liu, H., Maggio, J. E., Vigna, S. R., Mantyh, P. W. and Basbaum, A. I. (1995) J. Comp. Neurol. 356, 327–344 Yashpal, K., Dam, T. and Quirion, R. (1991) Brain Res. 552, 240–247 McLeod, A. L., Krause, J. E., Cuello, A. C. and Ribeiro-Da-Silva, A. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 15775–15780 Yu, X-H. et al. (1999) J. Neurosci.19, 3545–3555 Malcangio, M. and Bowery, N. G. (1994) Br. J. Pharmacol. 113, 635–641 Brechenmacher, C., Larmet, Y., Feltz, P. and Rodeau, J-L. (1998) Neurosci. Lett. 241, 159–162 Kangrga, I. and Randic, M. (1991) J. Neurosci. 10, 2026–2038 Rusin, K. I., Bleakman, D., Chard, P. S., Randic, M. and Miller, R. J. (1993) J. Neurochem. 60, 952–960 Malcangio, M., Fernandes, K. and Tomlinson D. R. (1998) Br. J. Pharmacol. 125, 1625–1626 Kramer, M. S. et al. (1998) Science 281, 1640–1645
Chemical names RP67580: (1-imino-2-[2-methoxyphenyl]ethyl)-7,7diphenyl-4-perhydroisoindolone (3aR,7aR) SR140333: (S)-1-(2-[3-{3,4-dichlorophenyl}-1-{3-isopropoxyphenylacetyl}piperidin-3-yl]ethyl)-4-phenyl1-azoniabicyclo(2.2.2)octane chloride
Acknowledgement Special thanks to Dr Martyn Jones for fruitful discussion while preparing the manuscript.
Letters to the Editor The Editor welcomes correspondence from readers of Trends in Pharmacological Sciences on any articles published. Please state clearly whether or not you wish the letter to be considered for publication. If a letter is considered suitable for publication, it will be sent to the author of the original article for their response; both the letter and reply will be published together.
TiPS – October 1999 (Vol. 20)
407