Neuropathic pain results from damage or abnormal function of the peripheral or central nervous system. Although there has been some progress in developing drugs that provide relief from neuropathic pain, current methods of treatment are unsatisfactory. One of our authors (DB) recently observed the neuropathic pain-relieving effects of KRN5500, a spicamycin derivative, on a patient receiving it for chemotherapy. This patient had a severe pain disorder from an IgA monoclonal gammopathy, which has involved all four extremities for more than 26 yr. KRN5500 provided immediate and prolonged relief of his neuropathic pain, but had no effect on nociceptive pain induced by an expanding tumor in his liver. This drug has been used in clinical antitumor trials in the United States and Japan. In our patient, there were minimal side effects from repeated infusions of the drug over a 6 mo-period. During this time, there was no evidence of a KRN5500-induced peripheral neuropathy and his pain remained under control for 6 months until his death.
Based on this observation, we hypothesized that KRN5500 would have similar effects in animal models of neuropathic pain. Such animal models provide methods for systematic evaluation of drugs. Rat models have included chronic constriction injury (CCI) (1) and segmental spinal nerve injury (SSI) (2).
Spicamycin is an antibiotic, antitumor drug that inhibits protein synthesis. It is unknown whether there is an inhibition of axonal transport. Previous work in neuropathic pain models in rats has shown that other antitumor drugs, such as vincristine, prevent nerve injury-induced central sensitization (3,4). In those cases, it was thought that the mechanism by which vincristine abolished neuropathic pain was by blockade of axonal transport, although this effect was transient (4). However, vincristine can produce painful peripheral neuropathy in rats and humans (5–8). Other antineoplastic drugs, including adriamycin, have been used to treat neuropathic pain (9).
Here, we describe the effects of KRN5500 on neuropathic pain produced in rats by using both the Chung and Bennett models (1,2). These are the most widely used models and they have both similarities and differences. We think it is important to test this new experimental drug on two different models to be able to evaluate the extent of the effect of the drug on various types of nerve injuries. Specifically, we tested the hypotheses that KRN5500 would inhibit allodynia after acute administration and as a single administration and would provide prolonged decrease in allodynia. In addition, we tested the effect of this drug on pain thresholds in normal rats and in an acute inflammatory model of pain.
Male Sprague-Dawley rats (n = 52) weighing 150–200 g (Charles River Breeding Laboratories, Wilmington, MA) were used for the experiments. The animals were housed in groups of three in plastic cages with soft bedding under a 12-h light-dark cycle. Food and water were available ad libitum. The animal care committee approved the experimental protocols at the Massachusetts General Hospital. After 1 wk of acclimatization to the laboratory conditions, all animals were tested for baseline mechanical allodynia.
In these experiments, two rodent neuropathic pain models (SSI) and CCI were used. Reports on differences in pain behavior have been noted in each model. We, therefore, chose to test KRN5500 in both of them. The experiments were performed in a single blinded fashion, i.e., the experimenter was not aware of either the drug or the normal saline.
SSI—The Chung Model
One week after acclimatization to the laboratory conditions, baseline measurements were recorded and surgery was performed as previously described by Kim and Chung (2). Rats (n = 16) were anesthetized with halothane in oxygen. Thereafter, they were placed in a prone position, a midline skin incision at L-4 to S-2 was made, and paraspinal muscles were separated from the spinous processes at these levels. The left L-5 and L-6 spinal nerves were identified, each was tightly ligated with 6–0 silk thread, and the wound was closed. After surgery, anesthesia was discontinued, and the animals were returned to their cages with food pellets and water ad libitum (the same environmental conditions as before). The animals recovered from anesthesia within approximately 10 min. They were allowed a 1-wk recovery period before the experiments were performed. Animals were tested for the signs of allodynia 1–2 wk after surgery, and only those animals were included which showed hypersensitivity to pinprick with von Frey filaments (VFF) (one animal was excluded from the study because of lack of development of hypersensitivity to VFF). They were then divided into two groups. Group A (n = 8) was treated with KRN5500 injections and Group B (n = 7) received injections of normal saline (see details below).
CCI—The Bennett Model
We produced a peripheral mononeuropathy as described by Bennett and Xie (1). Under halothane anesthesia, the common sciatic nerve of rats (n = 12) was exposed and loose constrictive ligatures were placed around the sciatic nerve. The wound was then closed and the animals allowed to recover. Approximately 1–2 wk later, animals were tested for the development of mechanical allodynia by using the same methods used for the SSI model (see detailed description below). One animal was excluded from the study because of lack of development of hypersensitivity to VFF.
To produce a nociceptive model, we injected naive animals (n = 20) with 0.1 mL of Complete Freund’s adjuvant (CFA) subcutaneously in the left hindpaw under brief general anesthesia. The animals were then tested for mechanical allodynia 24 h and 48 h after the injection of CFA. After measuring the pain threshold 48 h after the injection of CFA, animals were divided into two groups. Group C (n = 10) was treated with KRN5500 injection and Group D (n = 10) received injection of normal saline.
We used the following formulation of KRN5500: Solution A (with 5 mg KRN5500 potency; 0.05 g of N, N-dimethylacetoamide; 0.4 g of propylenglycol; 0.3 g of polysorbate 80; and appropriate amount of ethanol) was mixed with Solution B (0.1 g of monoethanolamine and appropriate amount of water for injection). Then, the mixture was dissolved in normal saline to a final concentration of 1.66 mg/mL of KRN5500. We injected all animals intraperitoneally (IP) at a dose of 2.5 mg/kg, which is 1.5 mL/kg of the previously mentioned concentration. Treated animals received KRN5500, whereas control animals received an equal volume of normal saline. The study was conducted in a blinded fashion with the researcher unaware of which animal received KRN5500 formulation or normal saline.
To evaluate the effect of KRN5500 on the neuropathic pain model and on the inflammatory model, we tested all animals on mechanical sensitivity for 7 days after drug/normal saline injection. We chose the following time points after injection: 1 h, 2 h, 4 h, 1 day, 3 days, 5 days, and 7 days.
VFF were used to test mechanical allodynia. The animals were placed on a mesh floor and were covered by a transparent plastic box open at the bottom. Calibrated VFF (3.61, 3.84, 4.08, 4.31, 4.56, 4.74, 4.93, and 5.16) were applied to the plantar skin of the left hindpaw with the up-down method and 50% foot withdrawal (paw flinching) threshold was determined. Each VFF was tested by inserting it from below through the mesh floor and applying it to the second, third, and fourth digits of the foot until the filament just bent (10). A trial consisted of four repetitive VFF applications (at a frequency of one per 10–15 s).
The data were presented as mean ± sem. Statistical differences between the groups were determined by the Mann-Whitney U-test. Dunnett’s test after analysis of variance was used to assess the difference between baseline and each time point within the groups. Statistical significance was accepted when P ≤ 0.05.
All animals had a normal high pain threshold (>15 g) and no sign of mechanical allodynia of their left hindpaw, as determined with VFF before neuropathic surgery or before administering CFA. However, within 7–14 days after nerve ligation (both Chung and Bennett models) and 48 h post-CFA intradermal injection, all rats exhibited characteristic pain behavior (mechanical allodynia). Further, there were no statistically significant differences between groups at the baseline before surgery or CFA injection and at 1 to 2 wk after surgery or 48 h after CFA injection (Figures 1–3).
SSI—The Chung Model
All of the rats in this experiment showed a presurgical baseline pain threshold >15 g (15 g was a cutoff) and a significant reduction in this value 10 to 14 days postsurgery. IP injection of a single dose of KRN5500 (2.5 mg/kg) produced a significant increase in the mechanical threshold for paw flinching as early as 1 h postinjection. This value remained increased and thus, was significantly different from the value of the saline group throughout the rest of the experimental period (Figure 1). Rats treated with saline did not show an increase in this value at any time point during the experimental period except at 2 h. This is probably because of the fact that the rats were getting used to the procedure.
CCI—The Bennett Model
KRN5500 induced a significant increase in pain threshold as measured by paw withdrawal (Figure 2) as early as 2 h after treatment, from a baseline value of 1.83 ± 0.2 g to 4.88 ± 1.37 g, P < 0.05) and remained significantly increased for at least 4 h with a trend of decreased response to mechanical allodynia over the time period. Thus, as can be seen from Figure 2, there was a statistically significant difference between the two groups at 2 h (1.44 ± 0.43 g vs 4.88 ± 1.37 g, P < 0.05), 4 h (1.95 ± 0.51 g vs 5.98 ± 1.92 g, P < 0.05) and 7 days (1.55 ± 0.52 g vs 3.86 ± 0.7 g, P < 0.05) posttreatment. Animals treated with saline again did not show any significant change in their pain threshold values except on Day 1 posttreatment.
The intradermal injection of CFA into the plantar surface of the left hindpaw produced a significant paw withdrawal behavior when touched by nonnoxious forces of VFF by 48 h (Figure 3). However, no difference in the pain threshold was seen between the groups that were administered saline and KRN5500 at any time point throughout the experimental period. Nonetheless, this value was significantly increased at 5 days for the saline group (0.58 ± 0.17 g vs 1.66 ± 0.27 g, P < 0.05); and at 1 h and 3 days for the KRN5500-treated group (0.46 ± 0.16 vs 2.25 ± 0.71 g and 2.35 ± 0.88 g, respectively, P < 0.05) (Figure 3). The effect of a single dose of 2.5 mg/kg ip KRN5500 was studied on an additional four rats that neither underwent any surgery nor received CFA. There was no change in their mechanical pain threshold during the entire experimental period (over 7 days).
KRN5500 is a semisynthetic antitumor antibiotic produced by Streptomycesalanosinicus 879-MT3 that shows an antitumor effect on several human tumor xenografts (11). Metabolism of KRN5500 to SAN-Gly is the major mechanism contributing to the cytotoxic potency of KRN5500 (11). Its mode of action appears to be mainly by inhibiting the protein synthesis in P388 cells (11). A single injection of KRN5500 inhibits allodynia in two rat models of neuropathic pain. Both of these animal models of neuropathic pain, CCI and SSI, showed the characteristic sign of mechanical allodynia, which is similar to that seen in patients with neuropathic pain syndrome. There was an early and prolonged relief of mechanical allodynia in animals receiving a single IP dose of KRN5500. A similar alleviation of neuropathic pain symptoms was observed in our patient that received this drug. At present, we do not know the mechanism of action of spicamycin in producing these effects.
KRN5500 had no effect on normal animals. Moreover, our results also indicate that KRN5500 has no effect on animal models of nociceptive pain using CFA. CFA is a common method for inducing acute inflammation in rodents. The inflammation usually peaks at 48 h after intraplantar injection. We administered the drug at 48 h, when the animals showed clear hyperalgesia. We believe that the drug has no effect on acute nociceptive pain, although we have not examined its efficacy in a chronic nociceptive pain model. Interestingly, in our single patient, nociceptive pain, produced by expansion of the liver capsule, was unaffected by this drug, although his neuropathic pain appeared to be preferentially affected and alleviated. Our animal results support our clinical observation, albeit in only a single patient. It is known that some of the antibiotic antitumor drugs, for example, vincristine, have no effects on nociceptive pain. Others, such as cyclophosphamide, inhibit nociceptive pain (12). There are no data on the effects of KRN5500 on other inflammatory indices. Studies are needed to further characterize these issues.
KRN5500 attenuates allodynia in two models of neuropathic pain (Figures 1 and 2). In both the SSI (Chung) and CCI (Bennett) models, KRN5500 produced significant antiallodynic effects. The effects were observed within 2 h after IP injection of the drug. This effect was more profound in the Chung model compared with the Bennett model throughout the experimental period. These differences may indicate that the mechanisms underlying the two models may be different (13). Such differences have been reported in the literature, including the duration of persistent allodynia and differences in the composition of nerve fiber type after damage. It has been postulated with the Chung model that there is abnormal processing of A-β fibers, whereas in the Bennett model, A-δ and C fibers are the major players in the neuropathic pain behavior. Whatever the mechanism of action, this drug has two major effects, including rapidity of onset and prolonged duration of effects (at least in the Chung model) after a single injection. Although the mechanism of action of KRN5500 on neuropathic pain is unknown, our results raise a number of questions regarding possible mechanisms of a drug that produces rapid and prolonged therapeutic effects.
Putative mechanisms for the drug’s therapeutic effect include production of cell death in already damaged dorsal root ganglion neurons (14–16) or by inhibition of protein synthesis (11,16–18). In addition, inhibition of axonal transport as a possible analgesic effect for other antitumor drugs (4,19) has been shown for other antitumor drugs. Because an autoimmune basis for neuropathic pain may be a possibility (20,21), the possibility of KRN5500 for inhibition of this process should also be considered.
In conclusion, KRN5500 has no antinociceptive effects, at least as determined after CFA. However, the drug had a significant effect in alleviating symptoms of mechanical allodynia in both models of neuropathic pain in rats. Currently, the mechanism of action and the long-term effects of KRN5500 on normal sensory nerves is unknown and work to understand these effects is continuing.
1. Bennett GJ, Xie Y-K. A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 1988; 33: 87–107.
2. Kim SH, Chung JM. An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 1992; 50: 355–63.
3. Sotgiu ML, Biella G, Firmi L, Pasqualucci V. Topical axonal transport blocker vincristine prevents nerve injury-induced spinal neuron sensitization in rats. J Neurotrauma 1998; 15: 1077–82.
4. MacFarlane BV, Wright A, Benson HA. Reversible blockade of retrograde axonal transport in the rat sciatic nerve by vincristine. J Pharm Pharmacol 1997; 49: 97–101.
5. Tanner KD, Reichling DB, Levine JD. Nociceptor hyper-responsiveness during vincristine-induced painful peripheral neuropathy in the rat. J Neurosci 1998; 18: 6480–91.
6. Authier N, Coudore F, Eschalier A, Fialip J. Pain related behavior during vincristine-induced neuropathy in rats. Neuroreport 1999; 10: 965–8.
7. Aley KO, Reichling DB, Levine JD. Vincristine hyperalgesia in the rat: a model of painful vincristine neuropathy in humans. Neuroscience 1996; 73: 259–65.
8. Sandler SG, Tobin W, Henderson ES. Vincristine-induced neuropathy: a clinical study of fifty leukemic patients. Neurology 1969; 19: 367–74.
9. Kato S, Otsuki T, Yamamoto T, et al. Retrograde adriamycin sensory ganglionectomy: novel approach for the treatment of intractable pain. Stereotact Funct Neurosurg 1990:54–55:86–9.
10. Chaplan SR, Bach FW, Pogrel JW, et al. Quantitative assessment of allodynia in the rat paw. J Neurosci Methods 1994; 53: 55–63.
11. Kamishohara M, Kawai H, Sakai T, et al. Anti-tumor activity of spicamycin derivative, KRN5500, and its active metabolite in tumor cells. Oncol Res 1994; 6: 383–90.
12. Tanner KD, Levine JD, Topp KS. Microtubule disorientation and axonal swelling in unmyelinated sensory axons during vincristine-induced painful neuropathy in rat. J Comp Neurol 1998; 395: 481–92.
13. Bennett GJ. An animal model of neuropathic pain: a review. Muscle Nerve 1993; 16: 1040–8.
14. Cho ES. Toxic effects of adriamycin on the ganglia of the peripheral nervous system: a neuropathological study. J Neuropathol Exp Neurol 1977; 36: 907–15.
15. Barajon I, Bersami M, Quartu M, et al. Neuropeptides and morphological changes in cisplatin-induced dorsal root ganglion neuropathy. Exp Neurol 1996; 138: 93–104.
16. Gill JS, Windebank AJ. Cisplatin-induced apoptosis in rat dorsal root ganglion neurons is associated with attempted entry into the cell cycle. Clinical Investigation 1998; 101: 2842–50.
17. Benoliel R, Eliav E, Mannes AJ, et al. Actions of intrathecal diphtheria toxin-substance P fusion protein on models of persistent pain. Pain 1999; 79: 243–53.
18. Mantyh PW, Rogers SD, Honore P, et al. Inhibition of hyperalgesia by ablation of lamina I spinal neurons expressing the substance P receptor. Science 1997; 278: 275–9.
19. Yamamoto T, Yaksh TL. Effects of colchicine applied to the peripheral nerve on the thermal hyperalgesia evoked with chronic nerve constriction. Pain 1993; 55: 227–33.
20. Bennett GJ. Does a neuroimmune interaction contribute to the genesis of painful peripheral neuropathies? Proc Natl Acad Sci USA 1999; 96: 7737–8.
21. Gold BG, Yew JY, Zeleny-Pooley M. The immunosuppressant FK506 increases GAP-43 mRNA levels in axotomized sensory neurons. Neurosci Lett 1998; 241: 25–8.