Transmission of Chronic Nociception by Spinal Neurons Expressing the Substance P Receptor

See allHide authors and affiliations

Science  19 Nov 1999:
Vol. 286, Issue 5444, pp. 1558-1561
DOI: 10.1126/science.286.5444.1558


Substance P receptor (SPR)–expressing spinal neurons were ablated with the selective cytotoxin substance P–saporin. Loss of these neurons resulted in a reduction of thermal hyperalgesia and mechanical allodynia associated with persistent neuropathic and inflammatory pain states. This loss appeared to be permanent. Responses to mildly painful stimuli and morphine analgesia were unaffected by this treatment. These results identify a target for treating persistent pain and suggest that the small population of SPR-expressing neurons in the dorsal horn of the spinal cord plays a pivotal role in the generation and maintenance of chronic neuropathic and inflammatory pain.

Chronic pain conditions are caused by ongoing disease states or tissue damage that result in sensitization of primary afferent and spinal cord neurons. This sensitization results in an increased sensitivity to both noxious (hyperalgesia) and non-noxious (allodynia) stimuli that is frequently difficult to treat with current pharmacological or surgical approaches (1).

Spinothalamic (STT) and spinoparabrachial (SPB) neurons are involved in the ascending conduction of acute noxious stimuli. Sensitization of these neurons results in hyperalgesia (2). Although SPR-expressing neurons represent less than 5% of the total neurons in the dorsal horn of the spinal cord (3), the majority of STT and SPB neurons in lamina I of the dorsal horn of the spinal cord express SPR. The majority of lamina I SPR neurons are STT and SPB neurons (4). SPR activation appears to be involved in the excitation and sensitization of STT neurons as well as the development of hyperalgesia (5).

When a conjugate of substance P (SP) and the ribosome-inactivating protein saporin (SAP) was intrathecally ( infused into the spinal cord, the SP-SAP conjugate was specifically concentrated in SPR-expressing neurons. Thirty days after infusion, there was a loss of lamina I spinal cord neurons that express the SPR (5). This treatment attenuated thermal and mechanical hyperalgesia as well as nocifensive behavior produced by capsaicin injection (5).

To determine the concentration-response relation for SP-SAP in ablating SPR-expressing spinal neurons and in blocking capsaicin-induced pain behaviors, we infused 10 μl of 10−7 M, 10−6M, or 5 × 10−6 M SP-SAP 30 days before plantar administration of 10 μg of capsaicin (5). In normal animals, capsaicin results in a profound nocifensive behavior, thermal hyperalgesia, and mechanical allodynia (5). Thirty days after infusion of SP-SAP, there was a distinct concentration-related reduction in these behaviors. The reductions in these pain-related behaviors were all significantly correlated with the concentration-related loss of SPR-expressing lamina I neurons (allr > 0.95; Fig. 1A). On the basis of these results and the lack of any observable side effects at the concentration of 10−6 M, infusion of 10 μl of 10−6 M SP-SAP was used in the remaining experiments.

Figure 1

(A) SP-SAP produces a concentration- related reduction in capsaicin-induced pain behaviors and ablation of SPR-expressing lamina I neurons. Concentration-response relations of SP-SAP (10 μl, on capsaicin-induced nocifensive behavior (red), thermal hyperalgesia (green), mechanical allodynia (blue), and loss of lamina I SPR-positive neurons (open circles) 30 days after SP-SAP treatment.n = 5 per group. (B to D) SP-SAP attenuates pain behaviors in three models of inflammatory pain. In all cases, SP-SAP (10−6 M, 10 μl) was administered 30 days before injection of the inflammatory agent. (B) The behavioral effects of administration of formalin in the rat hind paw. Formalin results in about 90 min of robust paw flinching (6) in saline-treated animals (blue, n = 7). There is a significant reduction in both the number of flinches and the overall length of the flinching behavior in the second phase (20 to 50 min) produced by formalin in SP-SAP–treated animals (red, n = 9). (C) SP-SAP (red,n = 5) attenuates the mechanical allodynia that is present after injection of carrageenan (7) when compared with saline-treated controls (blue, n = 4); h, hours; d, days. (D) SP-SAP (red, n = 6) pretreatment attenuates the mechanical allodynia present after injection of CFA (7) when compared with saline-treated controls (blue,n = 4). There is no significant allodynia in saline-treated animals by 3 days after carrageenan injection or at 7 days after CFA injection. (E and F) SP-SAP attenuates nerve injury (spinal nerve ligation model)–induced allodynia when administered either 30 days before or 7 days after nerve ligation (8). The dashed lines indicate the paw withdrawal threshold (g) in the normal animal. (E) Thirty-day pretreatment with SP-SAP (red, n = 6) reduces the tactile allodynia in nerve-injured animals when compared with saline-treated controls (blue,n = 6). (F) The antiallodynic effect of SP-SAP (red, n = 14) and saline (blue, n = 13) when administered after nerve ligation and development of the persistent pain state. The antiallodynic effect of SP-SAP becomes significant 21 days after infusion. Error bars represent standard error of the mean. Asterisks represent statistical significance from control (P < 0.05). Crosses represent statistical significance from baseline (P < 0.05).

The effect of SP-SAP on three models of inflammatory pain was tested 30 days after administration of 10 μl of 10−6 M SP-SAP (5). Subcutaneous (s.c.) injection of formalin into the hind paw produces a distinct biphasic, nocifensive behavior consisting of an early phase followed by a prolonged second phase lasting nearly 1 hour (6). SP-SAP did not significantly affect the first phase (0 to 10 min); however, in the second phase (20 to 50 min), the paw-flinching behavior was significantly reduced in SP-SAP–treated animals (Fig. 1B). Injection of λ-carrageenan or complete Freund's adjuvant (CFA) into the hind paw produced local inflammation, thermal hyperalgesia, and mechanical allodynia (5), peaking at about 3 hours and 3 days after injection, respectively (7). SP-SAP significantly reduced the thermal hyperalgesia and the mechanical allodynia produced by carrageenan (Fig. 1C) or CFA (Fig. 1D).

The effects of SP-SAP were also assessed in a model of neuropathic pain (8). Tight ligation of the L5 and L6 spinal nerves resulted in long-lasting mechanical allodynia. Thirty-day pretreatment with SP-SAP reduced the mechanical allodynia that is present 7 days after nerve ligation surgery (Fig. 1E). When SP-SAP was administered 7 days after nerve ligation, there was also significant reduction in the mechanical allodynia (Fig. 1F). These findings demonstrate that SP-SAP treatment will inhibit the allodynia associated with nerve injury when administered before or after the development of the neuropathic pain.

To assess long-term effects, we examined measurements of capsaicin-induced pain behaviors and several neuronal and glial cell markers in both spinal cord and dorsal root ganglia at 30, 100, and 200 days after infusion of 10 μl of 10−6 M SP-SAP (Fig. 2 and Table 1) (5). In all cases, saline or 10−6 M SAP did not induce any significant changes in behavior or in the cell populations examined (Table 1). In contrast, SP-SAP produced a significant decrease in the number of SPR-expressing cells in laminae I and III at 30 days and in laminae I, III, IV, and V at 100 and 200 days after infusion (9). A significant change in the total number of lamina I neurons with the NeuN antibody was not detected, presumably because the population of SPR-expressing neurons targeted by SP-SAP makes up such a small percentage of these neurons (3). SP-SAP treatment results in a long-term inhibition of capsaicin-induced hyperalgesia and allodynia with no evidence of loss of this effect over time (Fig. 3). The effects of SP-SAP did not affect morphine analgesia and appeared to be confined to the spinal segments where it was infused. Thus, after lumbar infusion of SP-SAP, the ability of morphine (15 mg/kg, s.c.) to block mechanical allodynia or thermal hyperalgesia after CFA injection into the hind paw was preserved, and formalin-induced pain behaviors in the forepaw (6) were not altered.

Figure 2

Fluorescent confocal images of coronal (A to F) or sagittal (G andH) sections of the lumbar (L4) rat spinal cord 200 days after saline, SAP, or SP-SAP administration. SPR immunoreactivity (orange) is significantly reduced after SP-SAP treatment (E) but not after SAP (C) or saline (A) treatment. In contrast, NeuN labeling shows no significant reduction in the SP-SAP (F), SAP (D), or saline (B) groups. Scale bar (A to F), 200 μm. In the sagittal plane, laminae are indicated by roman numerals, and the loss of SPR-immunoreactive neurons after SP-SAP (H) treatment is apparent in both the superficial and deep laminae when compared with the saline-treated group (G). Scale bar (G and H), 80 μm.

Figure 3

Time course of antihyperalgesic effect of infusion of saline (n = 5, 10, and 6 at 30, 100, and 200 days, respectively), 10−6 M SAP (n = 6, 6, and 6), or 10−6 M SP-SAP (n = 5, 10, and 6). After SP-SAP treatment, there is a reduction in (A) capsaicin-induced nocifensive behavior (expressed as duration, in seconds, over 300-s observation), (B) thermal hyperalgesia (expressed as paw withdrawal latency, in seconds), and (C) mechanical allodynia (expressed as paw withdrawal threshold, in g) that does not diminish with time after injection. Infusion of saline or SAP alone did not produce any changes in these pain behaviors at any of the time points examined. Asterisks represent statistical significance from control (P < 0.05).

Table 1

Cytotoxicity of infused SAP and SP-SAP in the L4 segment of the spinal cord at 30, 100, and 200 days after treatment. There were no statistically significant differences between animals that received saline, animals that received SAP (10−6 M), and normal animals, and therefore all percentages are expressed as percentage of saline-treated controls. The only significant differences observed were in SP-SAP–treated animals (10−6 M), in which there was a reduction in SPR-expressing neurons in laminae I to III (all time points) and laminae IV and V (100 and 200 days). Data are expressed as mean ± standard error of the mean. ND, not determined; CGRP, calcitonin gene-related pepitude; DRG, dorsal root ganglia; GFAP, glial fibrillary acidic protein.

View this table:

These results suggest that SPR-expressing neurons in the dorsal horn of the spinal cord are not the major site of action of morphine and that, after SP-SAP treatment, opiates remain a viable therapy for breakthrough pain. A major reason for the long-term efficacy and apparent lack of side effects after SP-SAP treatment appears to be related to the restricted nature and specificity of SP-SAP action. The actions of infused SP-SAP are confined to the SPR-expressing neurons in the dorsal horn, many of which are STT and SPB neurons (3, 4).

Previous data have suggested that in the peripheral nervous system, neuropathic and inflammatory pain arise from different mechanisms and are conveyed to the spinal cord by distinct groups of primary afferent neurons. Here, SP-SAP treatment reduced the hyperalgesia and allodynia associated with both neuropathic and inflammatory persistent pain states. Whether it is the same SPR-expressing neurons or different subsets of SPR-expressing neurons that convey neuropathic and inflammatory pain is unknown, but recent evidence suggests that a uniting feature of many spinal and forebrain neurons that express the SPR is their involvement in the response to tissue injury and stress (10). Characterization of the gene and protein changes that SPR-expressing spinal neurons undergo during nociception should provide insight into the spinal mechanisms involved in the generation and maintenance of chronic neuropathic and inflammatory pain.

  • * To whom correspondence should be addressed. E-mail: manty001{at}


View Abstract

Navigate This Article