Brain Research, 547 (1991) 335-338 A DONIS 000689939124632B
335
BRES 24632
Characterization of vanilloid receptors in the dorsal horn of pig spinal cord Arpad Szallasi and Peter M. Blumberg Molecular Mechanisms of Tumor Promotion Section, Laboratory of Cellular Carcinogenesis and Tumor Promotion, National Cancer Institute, Bethesda, MD 20892 (U.S.A.)
(Accepted 22 January 1991) Key words: [3H]Resiniferatoxin binding; Vanilloid receptor; Capsaicin; Dorsal horn
Specific [3H]resiniferatoxin binding is thought to represent the postulated vanilloid (capsaicin) receptor. We have previously characterized [~H]resiniferatoxin binding to membranes from rat and pig dorsal root ganglia, which contain the cell bodies of capsaicin-sensitive primary afferent neurons. We now demonstrate specific binding of [3H]resiniferatoxin to particulate preparations from pig dorsal horn, which contains the central nerve endings of the capsaicin-sensitive primary afferent neurons. The Kd was 0.27 _+ 0.03 nM; the Bmax was 370 + 40 fmol/mg protein. Vanilloids of the capsaicin class (capsaicin, piperine, zingerone) and resiniferatoxin class (tinyatoxin, 12-deoxyphorbo113-phenylacetate 20-homovanillate) inhibited binding with affinities consistent with their relative in vivo potencies. Given the interest in vanilloids as potential non-narcotic analgesic agents, this binding assay affords an attractive approach for characterization of structure-activity relations at spinal vanilloid receptors.
Peripheral receptors of capsaicin-sensitive primary afferent neurons include C-polymodal nociceptors, Adelta mechanoheat nociceptors and warm receptors of the skin ~5. Their central nerve endings are located in the substantia gelatinosa of the dorsal horn. This neural system plays a significant role in mediation of pain, nociception and warm sensation and is characterized as being selectively susceptible to the sensory-blocking effect of vanilloids, the best known example of which is capsaicin 2'5'~5. A growing body of evidence indicates that this sensory-blocking effect is a combination of two distinct mechanisms: desensitization of the peripheral receptors 16 and inhibition of release of nociceptive transmitters, such as substance P, somatostatin, and endogenous excitatory amino acids, from the spinal nerve endings 4'9. Recently it has been suggested that antinociception produced by systemically administered capsaicin involves a predominantly central (spinal) rather than a peripheral mechanism of action 1. Whereas structureactivity requirements for peripheral desensitization have been explored in depth ~6, and desensitization by topical capsaicin application has already been shown to have a beneficial effect in the treatment of neuralgic and postsurgical pain 5, little is known about the pharmacology of the spinal vanilloid receptor. There is, however, evidence that peripheral and spinal vanilloid receptors might have different structure-activity requirements: e.g.
olvanil, a capsaicin structural analog that fails to stimulate peripheral nociceptors, has been shown to inhibit the release of sensory neuropeptides in the dorsal horn l. Recently, resiniferatoxin (RTX), a naturally occurring diterpene containing a homovanillic ester, a key structural motif of capsaicin, has been shown to function as an ultrapotent capsaicin analog 6'11'12'18'19. For most of the responses characteristic of capsaicin, R T X is 100-10,000 fold more potent. Specific binding of [3H]RTX by sensory (dorsal root and trigeminal) ganglion membranes provided the first direct evidence for vanilloid receptors and affords a new approach for their pharmacological characterization 14. We now describe a [3H]RTX binding assay using a pig dorsal horn particulate fraction, which, we hope, might facilitate the identification of vanilioids with spinal antinociceptive activity. Pig spinal cords were purchased from Mt. Airy Locker Co. (Mt. Airy, MD), cooled on ice and then frozen on dry ice. Frozen spinal cords were cut into 1 cm pieces and the dorsal horn as well as the ventral horn were dissected with a warm razor blade. In further experiments the dorsal horn was cut into dorsal and ventral halves. Tissues were collected into ice-cold 20 m M Tris-CI buffer (pH 7.4) and disrupted with the aid of a Polytron tissue homogenizer. Particulate fractions were washed twice with the same buffer and then stored at - 7 0 °C. A crude
Correspondence: P.M. Blumberg, Laboratory of Cellular Carcinogenesis, Building 37, Room 3B25, NCI, NIH, Bethesda, MD 20892, U.S.A.
33~ synaptic membrane/microsomal fraction was prepared as follows: the dorsal half of dorsal horn was disrupted at 0 °C in 0.32 M sucrose/20 mM Tris-Cl buffer (pH 7.4) with the aid of a Polytron tissue homogenizer (setting 8, 20 s), and centrifuged at 6000 gmax for 15 rain. The remaining supernatant was then centrifuged for 60 rain at 100,000 gm~ and the resulting pellet resuspended in 20 mM Tris-C1 (pH 7.4). Aliquots were quickly frozen on dry ice and kept at -70 °C until assayed. The binding assays with [3H]RTX were carried out as we described for sensory ganglion membranes TM. Briefly, 18-25 j~g of particulate fraction protein, [3H]RTX and non-radioactive ligands were incubated in 250 ~1 or 2500 /A of 20 mM Tris-Cl (pH 7.4), containing 0.25 mg/ml bovine serum albumin, at 37 °C for l0 min unless otherwise specified. At the end of the incubation the samples were chilled to 0 °C; an aliquot of 50 #1 or 200 /~1 of the assay mixture was removed to determine total radioactivity; and another 150 ~1 or 2000 ~1 of the assay mixture was filtered immediately on Whatman GF/F glass fiber filters presoaked with 10 mg/ml bovine serum albumin in 20 mM Tris-Cl (pH 7.4). The filters were then washed with 50 ml of ice-cold 20 mM Tris-C1 (pH 7.4) containing 0.1 mg/ml bovine serum albumin, and the bound radioactivity was determined by scintillation counting. Non-specific binding was determined in the presence of 100 nM non-radioactive RTX. Binding data from saturation experiments using increasing concentrations of [3H]RTX were analyzed using the collection of computer programs described by McPherson 7. Scatchard and Hill transformations were performed by the Equilibrium Binding Data Analysis (EBDA) program; data were then further analyzed by the curvilinear regression program L1GAND s. RTX binding was also analyzed in the presence of a fixed (100 pM) concentration of [3H]RTX and increasing concentrations of non-radioactive ligands. K~ values were determined by a program fitting a theoretical sigmoidal competition curve to the data. [3H]RTX (37 Ci/mmol) and 12-deoxyphorboi 13phenylacetate 20-homovanillate were synthesized by the Chemical Synthesis and Analysis Laboratory, NCIFCRF, Frederick, MD. Non-radioactive RTX and resiniferonol 9,13,t4-orthophenylacetate was purchased from Chemicals for Cancer Research, Inc. (Chaska, MN) and from Chemsyn Science Laboratories (Lenexa, KS). Capsaicin was from Polysciences (Warrington, PA). Piperine and phorbol 12,13-dibutyrate were purchased from Sigma (St. Louis, MO). Zingerone was from Pflatz and Bauer (Waterbury, CT). The central terminals of capsaicin-sensitive sensory neurons are concentrated in a distinct area of the dorsal horn (Rexed laminae I and II); other areas of the spinal
cord have not been implicated in capsaicm actions:, tl~ agreement with this anatomical localization, we couki detect specific [3H]RTX binding only m the dorsal half ot the dorsal horn, including Rexed laminae 1 and I1~ whereas no specific binding was present either in the ventral half of the dorsal horn or in the ventral horn. Using the total particulate fraction and 300 pM [~H]RTX, we measured specific binding of 12~1 ± 10 fmol!mg protein (mean ± S.E.M., 3 determinations), representing 60% of the total binding. Using the crude synaptic membrane/microsomal fraction the corresponding values were 240 + 10 fmol/mg protein and 75 +- 2% (mean ± S.E.M., 17 determinations), respectively. By increasing the amount of crude synaptic membrane/microsomal fraction protein from 18-25/~g to 40/~g the proportion of specific to total binding became even more favorable: specific binding comprised approximately 90% of the total binding. In time course experiments, an essentially constant level of specific [3H]RTX binding was found between 5 and 60 min incubation at 37 °C; [3H]RTX binding had attained approx. 50% of its final value within 2 min. In addition, similar K d values were measured for 10 rain and 60 min. Based on these results, a standard incubation period of 10 min was employed in the experiments detailed below. In competition experiments using [3H]RTX at 100 pM and increasing concentrations of non-radioactive RTX over the range of 30 pM-30 nM, we obtained a Ki of 0.26 + 0.08 nM: the calculated Bma x w a s 38(I _+ 25 fmol/mg protein (mean + S.E.M., 4 determinations). These, binding assays were carried out in a volume of 250/~1 as described previously for sensory ganglia membranes. In direct binding experiments using increasing concentrations of [3H]RTX under these same assay conditions a large proportion of the total [3H]RTX (approx. 30%) was bound at the low [3H]RTX concentrations. In further experiments, we therefore increased the assay volume to 2.5 ml to achieve a 10-fold receptor dilution. Under these assay conditions no more than 2% of the total added [3H]RTX was bound. [3H]RTX displayed specific, saturable binding with a K a of 0.27 _+ 0.03 nM and a Bma x of 370 + 40 fmo1/mg protein (mean _ S.E.M., 3 determinations): a representative experiment is shown in Fig. 1. The Hill coefficient was 0.95 ± 0.02. The curvilinear analysis of the data confirmed the one-site model suggested by the Scatchard plot and gave final parameter estimates of K 8 = 0.26 ± 0.01 nM and Bmax = 380 + 35 fmol/mg protein, in good agreement with the competition experiments. The pharmacological specificity of [3H]RTX binding was examined for 3 classes of compounds: capsaicin-type vaniUoids, RTX-type vanilloids, and phorbol-related
337 activators of protein kinase C. Capsaicin inhibited binding with a K i of 1.9 + 0.6/~M (range) indicating a 104-fold lower potency than RTX (Fig. 2). This difference in affinity corresponds to the 1000-6000-fold difference in potency for various antinociceptive assays 3'17. Piperine, a less potent, and zingerone, a non-desensitizing but pungent capsaicin analog t6, did not inhibit binding at concentrations up to 100/~M (Table I). The RTX analog tinyatoxin inhibited binding with a g i of 0.8 "+- 0.1 nM (mean + S.E.M., 3 determinations) in excellent agreement with tinyatoxin being 3-5-fold less potent than RTX in its inflammatory 1° and hypothermic effects (unpublished observation) (Fig. 2). 12-Deoxyphorbol 13-phenylacetate 20-homovanillate (HV-dPP), an RTX analog lacking the orthoester group, competed for [3H]RTX binding sites with a K i of 2.2 + 0.5/~M (mean + S.E.M., 3 determinations) in keeping with HV-dPP being 2-4 orders of magnitude less active than RTX in the in vivo assays of inflammation and thermoregulation 13 (Fig. 2). Specific [3H]RTX binding was inhibited neither by resiniferonol 9,13,14-orthophenylacetate, the C-20 deesterified parent compound of RTX, nor by phorbol 12,13-dibutyrate, the typical ligand used for the analysis of binding to protein kinase C (Table I). An advantage of the dorsal horn preparation for analysis of [3H]RTX binding is its physiological relevance. Whereas the spinal vanilloid receptors are thought to play a central role in the antinociceptive activity of vanilloids 5'7, the possible role of binding sites on the perikarya is unclear. Although our limited analysis did not reveal major differences in structure-activity relations for the recep-
IOO
x-~ ~-E t¢- o
80
~
6o
m $
40
2o
10 10
10-8
10 6
10 4
Competing Ligand(M) Fig. 2. Competition curves for inhibition of [3H]RTX binding to crude synaptic membrane/microsomal fraction from the dorsal horn of pig spinal cord by non-radioactive RTX (O), tinyatoxin ( A ) , 12-deoxyphorbol 13-phenylacetate 20-homovanillate ( A ) , and capsaicin (0). Each curve represents a single experiment. Points were determined in triplicate; error bars indicate S.E.M.
tors at the two sites, we found the [3H]RTX binding affinity to be 10-fold greater for the dorsal horn compared to the dorsal root ganglia. This difference in affinity may indicate that a different receptor subclass is present in the central nerve terminals; alternatively, the vanilloid receptors may be modified following transport from the perikarya to the nerve endings. In either case, provided the difference in affinity corresponds to the in vivo situation, the differential binding characteristics may represent one mechanism for selectivity in the site of vanilloid action. At the methodological level, [3H]RTX binding assay utilizing crude synaptic membrane/microsomal fraction obtained from the dorsal half of pig dorsal horn represents a marked improvement compared to our previous binding assay using sensory ganglion membranes. The dorsal horn preparation provides a significantly more favorable proportion of specific to total binding; dorsal
p-
1.0
TABLE I
Pharmacological specificity of [-~H] R TX binding K i (nM) ~- 0.5 "5
5)
m
100
200
300
400
500
[3H] RTX Bound, fmoUmg protein
Fig. 1. Specific binding of [3H]RTX to crude synaptic membrane/ microsomal fraction from the dorsal horn of pig spinal cord. Scatchard plot of specific [3H]RTX binding. The line was fitted using the L I G A N D program. Points represent mean values from a single experiment. Data were normalized to 1 mg membrane protein in a 2 ml assay volume. A n additional two experiments gave similar results.
Resiniferatoxin analogs Resiniferatoxin Tinyatoxin HV-dPP l Capsaicinoids Capsaicin Piperine Zingerone Phorbol esters ROA 2 PDBu 3
0.26 _ 0.08 (n = 4) 0.80 _ 0.10 (n = 3) 2200 _ 460 (n = 3) 1900 _ 600 (range) > 100,000 (n = 2) > 100,000 (n = 2) > 30,000 (n = 2) >100,000 (n = 2)
12-Deoxyphorbo113-phenylacetate 20-homovanillate. z Resiniferonol 9,13,14-orthophenylacetate. 3 Phorbo112,13-dibutyrate.
33~ h o r n is a c o n s i d e r a b l y m o r e a b u n d a n t source of tissue;
c o n v e n i e n t in vitro tool for analysis of vanilloid struc-
and the dissection of the p r e p a r a t i o n
t u r e - a c t i v i t y relations.
is easier. T h e s e
a d v a n t a g e s n o w r e n d e r the [:~H]RTX binding assay a 1 Bettaney, J., Dickenson, A., Dray, A. and Hughes, C., Antinociception induced by the capsaicin analogue, olvanil: peripheral and central sites of action studied in vivo in adult rats and in vitro using neonatal rat, J. Physiol., 424 (1990) 60P. 2 Buck, S.H. and Burks, T.E, The neuropharmacology of capsaicin: review of some recent observations, Pharmacol. Rev.. 38 (1986) 179-226. 3 Campbell, E.A., Dray, A., Perkins, M.N. and Shaw, W., Analgesic and anti-inflammatory activity of resiniferatoxin in the mouse and rat: comparison with capsaicin, J. Physiol., 418 (1989) 149P. 4 Kangrga, I. and Randic, M., Tachykinins and calcitonin generelated peptide enhance release of endogenous glutamate and aspartate from the rat spinal dorsal horn slice, J. Neurosci., 10 (1990) 2026-2038. 5 Maggi, C.A. and Meli, A., The sensory-efferent function of capsaiein-sensitive neurons, Gen. Pharmacol., 19 (1988) 1-43. 6 Maggi, C.A., Patacchini, R., Tramontana, M., Amann, R., Giuliani, S. and Santicioli, P., Similarities and differences in the action of resiniferatoxin and capsaicin on central and peripheral endings of primary sensory neurons, Neuroscience, 37 (1990) 531-539. 7 McPherson, G.A., Analysis of radioligand binding experiments: a collection of computer programs for the IBM PC, J. Pharmacol. Methods, 14 (1985) 213-228. 8 Munson, R.J. and Rodbard, D., LIGAND: a versatile computerized approach for characterization of ligand binding systems, Anal. Biochem., 107 (1980) 220-239. 9 0 h k u b o , T., Shibata, M., Takahashi, H. and Inoki, R., Roles of substance P and somatostatin on transmission of nociceptive
information induced by formalin in spinal cord. J. Pharumcol. Exp. Ther., 252 (1990) 1261-1268. 10 Schmidt, R.J. and Evans, E J., Investigations into the skinirritant properties of resiniferonol ortho esters, Inflammation, 3 (1979) 273-280. 11 Szallasi, A. and Blumberg, P.M., Resinileratoxin, a phorbolrelated diterpene, acts as an ultrapotent analog of capsaicin, the irritant constituent in red pepper. Neuroscience, 30 (1989) 515- 520. 12 Szallasi, A., Job, E and Blumberg, EM., Duration of desensitization and ultrastructurat changes in dorsal root ganglia of rats treated with resiniferatoxin, an ultrapotent capsaicin an~og, Brain Research, 503 (1989) 68-72. 13 Szallasi, A., Sharkey, N.A. and Blumberg, P.M., Structure/ activity analysis of resiniferatoxin analogs, Phytotherapv Res., 3 (1989) 253-257. 14 Szallasi, A. and Blumberg, P.M., Specific binding of resiniferatoxin, an ultrapotent capsaicin analog, by sensory ganglion membranes, Brain Research, 524 (1990) 106-111. 15 Szolcsanyi, J., Sensory receptors and the antinociceptive effects of capsaicin. In R. Hakanson and E Sundler (Eds.), Tachykinin Antagonists, Elsevier, Amsterdam, 1985, pp. 45-53. 16 Szolcsanyi, J., Perspectives of capsaicin-type agents in pain therapy and research. In W.C.V. Parris (Ed~), Comtemporary Issues in Pain Management, Kluwer Academic Publishers, Norwell. in press. 17 Szolcsanyi, J., Szallasi, A., Szallasi, Z., Job, E and Blumberg, P.M., Resiniferatoxin, an ultrapotent neurotoxin of capsaicinsensitive primary afferent neurons, Ann. New York Acad. Sci.. in press.