1014 Annals ofthe Rheumatic Diseases 1992; 51: 1014-1018

REVIEWS Ann Rheum Dis: first published as 10.1136/ard.51.8.1014 on 1 August 1992. Downloaded from Role of in inflammatory arthritis

N E Garrett, P I Mapp, S C Cruwys, B L Kidd, D R Blake

The substance P (SP) has wide (a, j, and y preprotachykinin) owing to alternate ranging effects on many cell types. Nilsson et al RNA splicing, and all three can encode for SP. showed that SP could stimulate connective However, a preprotachykinin mRNA can only tissue cell growth by stimulating DNA synthesis encode for SP, whereas j3 and y preprotachykinin in human skin fibroblasts and arterial smooth mRNA can also encode for and muscle cells in vitro. 1 Substance P has also been other tachykinins.9 implicated in owing to its effects Substance P is widely distributed throughout on cells associated with acute and chronic the central and peripheral nervous systems'0 11 inflammatory responses. Substance P promotes and is synthesised in the cell bodies ofperipheral chemotaxis of human monocytes, and this nerves located in the dorsal root ganglia.'2 After appears to be an SP receptor mediated effect as synthesis SP is distributed to the central and it is blocked by D-amino acid analogues of sp.2 peripheral nerve terminals by a fast axonal Substance P mediates macrophage activation, transport system. 13 The rate of transport to the including enhancement of the oxidative burst periphery is greater than the transport rate to with production of significant amounts of the the central terminals in the spinal cord. 4 superoxide anion, hydrogen peroxide, and In peripheral nerves SP has been localised to release of prostaglandin E2, thromboxane B2, small unmyelinated nerve fibres,'5 described as and leukotriene C4.3 4 A later event induced by afferent C fibres, and small myelinated fibres SP is the release of lysosomal enzymes from (A delta fibres). These fibres are thought to activated macrophages.4 have a nociceptive function-that is, they are Direct injection of SP into human skin only usually stimulated by painful (noxious) produces a wheal and flare reaction as seen with stimuli, with such as SP being histamine, but SP is about 100 times more the central .'6 As well as this potent.5 Substance P induces the specific release afferent nociceptive role, SP also has an efferent of histamine from mast cells,6 which will action, where sensory nerve activation causes

produce vasodilatation, oedema, and an increase antidromic (reversed) transmission of nerve http://ard.bmj.com/ in vascular permeability-that is, SP can pro- impulses (in addition to orthodromic transmis- duce an inflammatory response by a direct or an sion to the spinal cord) in the branches of the indirect effect. It is apparent that SP may have a same peripheral C fibre (fig 2). This is known as role in inflammation and perhaps is involved in an axon reflex, and it has been shown that SP is the pathogenesis of inflammatory diseases such released from sensory nerves after antidromic as rheumatoid arthritis. This review attempts to stimulation.'7 Jansco et al showed that antidro-

tie together past and current ideas about SP and mic stimulation of a peripheral sensory nerve on October 2, 2021 by guest. Protected copyright. inflammatory arthritis. caused vasodilatation, plasma extravasation into the tissue, and increased vascular permeabi- lity,'8 all typical responses seen in acute inflam- Substance P mation. It therefore seems that sensory nerve In 1931 Von Euler and Gaddum investigated stimulation results in an acute response, des- the distribution of acetylcholine in various cribed as , with SP organs of the horse.7 Extracts from certain being a possible mediator of the response. tissues, in particular the intestine and the brain, produced a material which caused a lowering of arterial blood pressure in the atropinised rabbit Substance P and joint innervation and had a stimulatory effect on the tone and For SP to play a part in arthritis there must be rhythm of atropinised rabbit intestine. Von sensory nerve innervation to the joint and a role Euler and Gaddum designated this material as for the nervous system in the pathophysiology preparation 'P', now known as substance P. of the disease. Clinical evidence for a nervous TheInflammationGroup, Substance P is an undecapeptide member of system involvement came initially from ARC Building, The London Hospital the tachykinin family; others include neurokinin observations by Thompson and Bywaters'9 and Medical College, A and , which share the common C Glick,20 who found that in patients whose limbs Whitechapel, terminal sequence Phe-X-Gly-Leu-Met (fig 1). had been paralysed as a result of an upper moto- London El 2AD, There are two distinct genes for mammalian neurone or a lower motoneurone United Kingdom hemiplegia N E Garrett tachykinins: the SP/neurokinin A or prepro- lesion, such as poliomyelitis, the affected limbs P I Mapp tachykinin I gene and the neurokinin B or were spared the symptoms of any subsequent S C Cruwys II gene.8 Substance P is development of rheumatoid arthritis (RA). B L Kidd preprotachykinin D R Blake encoded by messenger RNAs (mRNA) arising Evidence from animal studies by Courtright and A Kuzell showed that sectioning of one sciatic Correspondence to: from SP/neurokinin gene transcription. Professor Blake. Although only one gene, there are three mRNAs nerve seven days before the induction of Role ofsubstance P in inflammatory arthritis 1015

Substance P Arg-Pro-Lys-Pro-Gln-Gln-Phe-Phe-4Gly-Leu-Met-NH2 joints) developed a more severe arthritis than the knee.25 Is there a differential innervation of Ann Rheum Dis: first published as 10.1136/ard.51.8.1014 on 1 August 1992. Downloaded from Neurokinin A His-Lys-Thr-Asp-Ser-Phe-Val-4Gly-Leu-Met-NH2 human joints that reflects the pattern of arthritis in rheumatoid disease? Using neurofilament antiserum, Gronblad et Neurokinin B Asp-Met-His-Asp-Phe-Phe-Val- 31y-Leu-Met-NH2 al showed nerve structures in samples of Figure I Amino acid sequences oftachykininpeptides. normal, osteoarthritic, and rheumatoid synovium.26 The nerves were mainly perivascu- lar in location, perhaps indicating a regulatory Dorsal root ganglion role of the local joint vasculature. When antisera (site of neuropeptide synthesis) to neuropeptides, including SP, were used immunoreactivity was again shown to be mainly periv3scular, but some SP-like immunoreactiv- ity was detected in free stromal nerves-mainly Release in normal and osteoarthritic synovial samples. of substance P As well as a lack of immunoreactivity in the free nerves, the intensity of the staining for neuro- (0\ was also much weaker in rheumatoid Blood vessel synovia than in normal or osteoarthritic synovia. It is suggested that activation of noci- 0 ceptors by chemical stimuli-for example, pros- Nociceptive \ Mast cell taglandins, might have caused an increased nerve fibre ...... release of SP from sensory nerves, thereby (A.delta or C fibre) lowering the nerve content to below the threshold for detection by immunostaining...... In 1990 Mapp et al published a study on the innervation of human Spinal cord .. .-.X...neuropeptide synovium, Orthodromic signal using antiserum to protein gene product 9-5 to in response to noxious stimulus map the overall innervation of synovium.27 The eg. pin prick on skin advantages of using protein gene product 9-5 antisera are that it is present throughout the Axon reflex _...... neurone and has a cytoplasmic location.28 It also - nociceptive orthodromic signal allows the localisation of small diameter fibres causes antidromic signal in Target organ eg. skin (0 5-1.5 [Lm), such as unmyelinated C fibres. other branches of the same et were to nerve fibre Using this technique, Mapp al27 able show a greater number of nerve fibres than had Neurogenic inflammation been shown in previous studies in which neuro- http://ard.bmj.com/ - antidromic signals cause filament antisera were used. In normal synovium release leading to inflammation nerve fibres were seen terminating near to the intimal layer of the synovium with some extend- Figre 2 Participation of substance P in inflammation ing to the boundary of the joint space and the intimal layer. The distribution of SP immuno- reactive fibres was similar (but less numerous) adjuvant arthritis in the rat delayed the onset than protein gene product 9-5 immunoreactive

and reduced the severity of disease in the fibres. In rheumatoid synovia protein gene on October 2, 2021 by guest. Protected copyright. operated limb.2' Rees et al foumd the opposite, product 9 5 immunoreactive fibres were not however-that is, denervated limbs developed found in the synovial intimal layer but had a the disease earlier and more severely.22 perivascular innervation ofdeeper vessels similar Another indication of nenvous system in- to that found in normal tissue. The number of volvement is the exact symmetry of a number free fibres in deeper tissues was greatly reduced of arthritic conditions.23 Inflatmmation in one compared with normal synovia, however. In joint is often followed by ain inflammatory normal tissue SP immunoreactive fibres were response in the contralateral jjoint. A nervous most abundant at the synovial membrane with pathway with its precise pro)jection seems a only sparse innervation ofsynovial blood vessels, more likely explanation foir the observed and blood vessels and free fibres in deeper symmetry than the presence of some blood tissues. In rheumatoid samples SP immunoreac- borne factor. Levine et a124 suggested that tive fibres were not present in the synovial neural mechanisms may expl.ain three of the blood vessels or the synovial membrane and, clinical features of RA-namiely, (a) specific although present in deeper tissues (either peri- joints at high risk are more 1ikely to develop vascularly or as free fibres), immunoreactivity arthritis; (b) specific joints at high risk have was much weaker than in normal synovia. As more severe arthritis; and, as:stated above, (c) suggested by Gronblad and colleagues, this may RA is bilaterally symmetricial. It was also be due to increased release of SP from sensory suggested in this hypothesis that SP released nerves but may be a result of synovial prolifera- from peripheral terminals of noxciceptive afferent tion in joint inflammation outflanking growth of fibres may mediate the neurog3enic component the nerve. Pereira da Silva and Carmo-Fonseca29 of the joint inflammation. L.evine et al also confirm the observations of Mapp et al27 with found that in the adjuvant arthrritis model in rats the suggestion that rheumatoid synovial tissue the ankle joints (which are moire densely inner- has a disturbed neuronal control. The lack of vated by SP containing neurones than the knee innervation in rheumatoid synovia may be due 1016 Garrett, Mapp, Cruuys, Kidd, Blake

to the inactivation or insufficient release of the their findings by an increase in the metabolism

protease inhibitor, protease nexin I. It of tachykinins in arthritis. This explanation has Ann Rheum Dis: first published as 10.1136/ard.51.8.1014 on 1 August 1992. Downloaded from has been shown that human stromelysin, a con- been re-enforced by evidence from recent work nective tissue metalloproteinase released from in our laboratory.4' It has previously been human synovial fibroblasts in response to shown that the enzyme neutral endopeptidase inflammatory mediators such as interleukin 1,3` (EC 3.4.24.11) can hydrolyse neuropeptides can inactivate the serine protease inhibitor, such as Sp.42 Mapp et al,41 using both a human a, antitrypsin, which shares common polyclonal and monoclonal primary antibody to structural features with protease nexin 1.31 neutral endopeptidase and the avidin biotin Zhang et al showed that there was increased complex method of detection, did not detect proteolytic cleavage of a, antitrypsin in knee anyimmunostaininginnormalhumansynovium. joint synovial fluid from patients with RA.32 It In synovial samples from patients with RA, may be, therefore, that endogenous metallopro- however, there was intense staining of cells teinases inactivate protease nexin I in RA and surrounding blood vessels and weaker staining thus prevent neurite outgrowth, resulting in the in the supporting stroma cells. The increased absence of nerves to the superficial synovium. intensity of staining in cells surrounding blood An interesting finding in the study by Mapp27 vessels indicates a vascularorigin fortheinducing was the lack of innervation to lymphoid areas in stimulus. It is thought that synovial fibroblasts the rheumatoid synovium. Fink and Weihe are the source ofneutral endopeptidase. Another found neuropeptides, including SP, in nerves study of SP in synovial fluid was made by supplying the human lymph node.33 It is known Marshall et al.43 Four clinical groups of patients that SP increases proliferation and immuno- were studied: those with RA, osteoarthritis, globulin synthesis in lymphocytes from murine Reiter's syndrome, and arthritis following Peyer's patches, mesenteric lymph nodes, and trauma. It was found that apart from the spleen.34 Substance P reverses the reduced patients with Reiter's syndrome each group had regional lymph node antibody response caused significantly higher concentrations of SP in by , and it may be that there is a neuro- synovial fluid than were found in paired plasma genic control, particularly through nerves samples. Concentrations of SP in the group with containing SP, of normal lymph node function. RA were 946-6 (SEM 82 8) pg/ml in synovial Thus the lack of innervation and therefore fluid compared with 676-4 (58-1) pg/ml in control seen in RA may be responsible for the plasma. This suggested a localised intra-articular excessive immunoglobulin production seen in source of SP-namely, release from the peri- some arthritic patients. pheral C fibres. Another possible role for SP and the nervous It is apparent from the above mentioned system in arthritis was suggested by Lotz et al.35 studies that detection of SP in human synovial They showed that SP was capable of stimulating fluid is extremely difficult. Concentrations synoviocytes in RA. It was found that SP reported range from 0 to 0 1 [ig/ml (table). The

increased the release of prostaglandin E2 from diversity in results may be due to the inaccuracy http://ard.bmj.com/ synoviocytes in a manner dependent on dose. It of the method used (Chapman and Tsao using was also shown that SP increased total protein high performance liquid chromatography found synthesis, including an increase in the synthesis SP concentrations about 100 times greater than ofcollagenase. Synoviocyteproliferationinduced those of Marshall) or a lack of sensitivity of the by SP enhances pannus tissue formation and assay (Devillier et al used an enzyme immuno- increases connective tissue damage in the dis- assay for tachykinin-like immunoreactivity, not a in specific for SP immunoreactivity. There is the eased state, thereby identifying role for SP on October 2, 2021 by guest. Protected copyright. the pathology ofarthritis. possibility of cross reactivity with neurokinins A and B and SP fragments). Interpretation of the meaning of the concentrations of any sub- Concentrations of substance P in joints stance in synovial fluid poses a problem Is it possible to detect SP in inflammatory because, as yet, the production, degradation, fluids? Tissot et al used the carrageenan pleurisy and clearance rates of the system have not been animal model of inflammation and found raised elucidated." levels of SP-like immunoreactivity in the in- flammatory exudate,36 indicating the possible participation of SP in the inflammatory reaction. Substance P receptors The first report of SP in human arthritic If SP has a pathophysiological function in synovial fluid came from Chapman and Tsao,37 arthritis then it is reasonable to expect altered who used high performance liquid chromato- levels of SP receptors. There are currently three graphy techniques and found SP at a concentra- pharmacologically distinct tachykinin receptors tion of 01 ,ug/ml in knee joint synovial fluid -namely, neurokinin 1, 2, and 3. They are from an arthritic patient. In 1986 Devillier et al defined usually by the rank order of potency of reported increased levels of tachykinin-like immunoreactivity in the synovial fluid ofpatients Published results ofsubstance P (SP) concentrations detected with rheumatic inflammatory diseases (3-94 in rheumatoid synovial fluid. Results are given as mean (SEM 1-21) ng/ml compared with 191 (0-56) (SEM) ng/ml in non-inflammatory fluid).38 Reference Substance P (pglml) Larsson et al using competitive radio- Chapman and Tsao37 lO, immunoassay found no SP-like immunoreacti- Devillier el al" 3940 (121) vity in patients with arthritis or in their control Larsson et al39 40 Nil patients.39 40 Larsson and colleagues explained Marshall et al43 946 6 (82 8) Role ofsubstance P in inflammatory arthritis 1017

tachykinins in binding studies. Substance P has of specific SP receptor antagonists could have the greatest affinity for the neurokinin 1 receptor therapeutic value for the treatment of inflamma- but may also interact with neurokinin 2 and 3 tory joint disease in humans. Ann Rheum Dis: first published as 10.1136/ard.51.8.1014 on 1 August 1992. Downloaded from receptors, but only at a higher concentration. It has been shown that the number of receptors for SP/unit area of tissue is increased Conclusions in inflammatory diseases such as ulcerative This review has concentrated solely on the colitis.45 This is probably owing to an increase involvement of substance P, currently the most in blood vessel density in the tissue rather than intensively studied neuropeptide, in arthritic an increase in receptor density. Walsh et al inflammatory disease. It should be noted that have recently shown the presence of specific, SP is one of many neuropeptides (others include high affinity, low capacity binding sites for gene related peptide) which is colo- Bolton Hunter labelled '251I-SP on endothelial calised with SP in sensory neurones and is also a cells of human synovial blood vessels.46 Binding potent vasodilator.56 On the other hand, neuro- characteristics indicated that the receptors were peptide Y, which is associated with sympathetic of the neurokinin 1 class. It is known that the neurones and colocalises with the catecholamine binding of SP to its receptor is dramatically synthesising enzymes tyrosine hydroxylase and increased in the presence of divalent cations, dopamine-4-hydroxylase,57 is a vasoconstrictor. and it may be that changes in physiological The balance of release of these regulatory concentrations of ions such as Mg2e or Mn2+ peptides is undoubtedly critical. alter SP binding in vivo (Walsh et al, paper It is now widely accepted that there is a presented at European Neuroscience Associa- neurogenic mechanism in the pathophysiology tion, Stockholm, 1990). It has also been shown of arthritis and SP is a leading candidate for the that there is a large increase in SP receptors in role of mediator-it has been found in the rheumatoid synovium compared with normal required locations, it is present in increased synovium.47 The receptors were present in concentration in synovial fluid from rheumatoid greatest concentrations in blood vessels, on patients, and there are altered numbers of SP endothelial lining cells, and the surrounding receptors in rheumatoid tissue. There is also the smooth muscle cells. Cruwys et al showed the possibility of a synergistic effect between SP presence of a specific vascular receptor for SP in and other neuropeptides, such as calcitonin rat synovium,48 and it may be that SP partici- gene related peptide, to produce the observed pates in the vasoregulation of the synovium, inflammatory changes. It is hoped that further particularly in inflammatory arthritic disease. research will elucidate the role for neuropeptides The involvement of SP and its receptors in in arthritis. inflammation and arthritis has been studied using a number of animal models. Levine et a125 We are grateful to the Arthritis and Rheumatism Council for found that infusion of SP into rat knees their generous financial support. increased the severity of adjuvant induced the I Nilsson J, Von Euler A M, Dalsgaard C-J. Stimulation http://ard.bmj.com/ arthritis,25 but administration of SP analogue of connective tissue cell growth by substance P and sub- (d-Pro2, d-Trp7'9)-SP, a putative SP receptor stance K. Natue 1985; 315: 61-3. antagonist,49 produced moderate soft tissue 2 Ruff M R, Whal S M, Pert C B. Substance P receptor- only mediated chemotaxis ofhuman monocytes. Peptides 1985; 6 swelling and osteoporosis. Lam and Ferrell (suppl 2): 107-11. model 3 Hartung H P, Toyka K V. Activation of macrophages by showed that the carrageenan of acute substance P: induction of oxidative burst and throm- joint inflammation could be virtually abolished boxane release. Eur J Pharmacol 1983; 89: 301-5. by prior administration of the SP antagonist 4 Hartung H P, Wolters K, Toyka K V. Substance P: binding properties and studies on celiular responses in guinea pig on October 2, 2021 by guest. Protected copyright. d-Pro4, d-Trp79"10-SP (4-11).50 It has also macrophages. J Immunol 1986; 136: 3856-63. been shown that of into 5 Foreman J C, Jordan C C, Oehme P, Renner H. Structure- injection capsaicin the activity relationships for some substance P-related peptides synovial cavity of the rat knee inhibited the that cause wheal and flare reactions in human skin. inflammation seen SP J Physiol 1983; 335: 449-65. following injection. 6 Foreman J C, Jordan C C. Histamine release and vascular Capsaicin causes the release of SP from afferent changes induced by neuropeptides. AgentsActions 1983; 13: 105-16. C fibres and the degranuilation of mast cells, but 7 Von Euler U S, Gaddum J H. An unidentified depressor in this case might have caused a depletion of SP substance in certain tissue extracts. J Physiol 1931; 72: 577-83. receptors in the target tissue.5' Until recently, 8 Helke C J, Krause J E, Mantyh P W, Couture R, Bannon antagonists at SP receptors have been analogues M J. Diversity in mammali tachykinin peptidergic neurons: multiple peptides, receptors and regulatory of SP itself. Snider et al, however, reported that mechanismis. FASEB J 1990; 4: 1606-15. the non-peptide compound CP-96,345 was a 9 Krause J E, Chirgwin J M, Carter M S, Xu Z S, Hershey A D. Three rat preprotachykinin mRNAs encode the potent, reversible, and competitive antagonist neuropeptides substance P and neurokinin A. Proc Nad! of the neurokinin 1 receptor. Their in vivo rat Acad Sci USA 1987; 84: 881-5. 10 Cuello A C, Kanazawa I. The distribution of substance P studies showed that the compound had a lack of immnunoreactive fibres in the rat central nervous system. agonist activity and was selective in its antago- Comtp Neuol 1978; 178: 129-56. 11 Hokfelt T, Kellerth J 0, Nilsson G, Pernow B. Substance P: nism. It has also been found that the affinity localization in the central nervous system and in some of CP-96,345 for the neurokinin 1 receptor prinary sensory neurons. Science 1975, 190: 889-90. 12 Harmar A, Schofield J G, Keen P. Substance P biosynthesis is dependent on species." S4 The possibility in dorsl root ganglia; an immtunohistochemical study of therefore exists ofspecific neurokinin 1 receptor (C5S) and (3H) proline incorporation in vitro. Neuroscince 1981; 6: 1917-22. subtypes for different species. 13 Pernow B. Substance P. Phamacol Rev 1983; 35: 85-141. Andrews et al showed that neurogenic plasma 14 Keen P, Harmar A J, Spears F, Winter E. Biosynthesis, axonal transport and turnover of neuronal substance P. In: extravasation in the rat skin was mediated by Porter R, O'Conner M, eds. Substance P in the nevous the neurokinin 1 receptor.55 All this evidence .ystem. London: Pitman, 1982: 145-64. (Ciba Foundation Symposium, No 91.) suggests a role for SP receptors in joint in- 15 Hokfeft T, Kellerth J O, Nilsson G, Pernow B. Experimental flammation in the rat, and it may be that the use immunohistochemical studies on the localization and distri- 1018 Garrett, Mapp, Criuys, Kidd, Blake

bution of substance P in cat primary sensory neurons. Brain 37 Chapman L F, Tsao M U. Possible presence of substance P in Res 1975; 100: 235-52. the synovial fluid of the knee joint of an arthritic patient. 16 Lembeck F, Donnerer J, Colpaert F C. Increase in substance Fed 'roc 1980; 39: 603. P in primary afferent nerve fibres during chronic pain. 38 Devillier P, Weill B, Renoux M, Menkes C, Pradelles P. Ann Rheum Dis: first published as 10.1136/ard.51.8.1014 on 1 August 1992. Downloaded from Neuropeptides 1981; 1: 175-80. Elevated levels of tachykinin-like immunoreactivity in joint 17 White D M, Helm R D. Release of substance P from peri- fluids from patients with rheumatic inflammatory diseases. pheral nerve terminals following electrical stimulation of NEnglJMed 1986; 314: 1323. the sciatic nerve. Brain Res 1985; 336: 27-31. 39 Larsson J, Ekblom A, Henriksson K, Lundeberg T, 18 Jansco N, Jansco-Gabor A, Szolcsanyi J. Direct evidence for Theodorsson E. Immunoreactive tachykinins, calcitonin neurogenic inflammation and its prevention by denervation gene-related peptide and in human synovial and by pretreatment with capsaicin. BrJ Pharmacol 1%7; fluid from inflamed knee joints. Neurosci Lett 1989; 100: 31: 138-51. 326-30. 19 Thompson M, Bywaters E G L. Unilateral rheumatoid 40 Larsson J, Ekblom A, Henriksson K, Lundeber T arthritis following hemiplegia. Ann Rhewn Dis 1%2; 21: Theodorsson E. Concentration of substance P, neurokinin 370-7. A, calcitonin gene-related peptide, neuropeptide Y and 20 Glick E N. Asymmetrical rheumatoid arthritis after polio- vasoactive intestinal polypeptide in synovial fluid from myelitis. BMJ 1%7; iii: 26-9. knee joints in patients suffering from rheumatoid arthritis. 21 Courtright L J, Kuzell W C. Sparing effect of neurological ScandjRheumatol 1991; 20:326-35. deficit and trauma on the course of adjuvant arthritis in the 41 Mapp P I, Walsh D A, Cruwys S C, Kidd B L, Blake D R. rat. Ann Rheum Dis 1%5; 24: 360-7. Localisation of neutral endopeptidase to the human 22 Rees R G, Dixey J J, Lightman S L, Brewerton D A. The synovium. BrJRhewnatol 1991; 30 (suppl 2): 121. influence of unilateral nerve section on the course of 42 Katayama M, Nadel J A, Bunnett N W, Di Maria G U, adjuvant arthritis in rats. Arthritis Rhewn 1989; 32 (suppl): Haxhiu M, Borson D B. Catabolism of calcitonin gene- S54. related peptide and substance P by neutral endopeptidase. 23 Kidd B L, Mapp P I, Gibson S J, et al. A neurogenic Peptides 1991; 12:563-7. mechanism for symmetrical arthritis. Lancet 1989; ii: 43 Marshall K W, Chiu B, Inman R D. Substance P and 1128-30. arthritis: analysis of plasma and synovial fluid levels. 24 Levine J D, Collier D H, Basbaum A I, Moskowitz M A, Arthritis Rhewn 1990; 33:87-90. Helms C A. The nervous system may contribute to the 44 Mapp P I. Substance P and arthritis. Current Medical pathophysiology of rheumatoid arthritis. J Rheumatol 1985; Literature-Rheumatology 1990; 9:127-31. 12: 406-11. 45 Mantyh C R, Gates T S, Zimmerman R P, et al. Receptor 25 Levine J D, Clark R, Devor M, Helms C, Moskowitz M A, binding sites for substance P, but not substance K or Basbaum A I. Intraneuronal substance P contributes to the neuromedin K, are expressed in high concentrations by severity of experimental arthritis. Science 1984; 226: 547-9. arterioles, venules and lymph nodules in surgical specimens 26 Gronblad M, Kontinnen Y T, Korkala 0, Liesi P, Hukkanen obtained from patients with ulcerative colitis and Crohn's M, Polak J M. Neuropeptides in synovium of patients with disease. Proc NadAcad Sci USA 1988; 85:3235-9. rheumatoid arthritis and osteoarthritis. J Rheumatol 1988; 46 Walsh D A, Mapp P I, Wharton J, et al. Localisation and 15: 1807-10. characterisation of substance P to human synovium L7 MappPI, Kidd B L, Gibson S J,etal. Substance P-, calcitonin in rheumatoid arthritis. Ann RheumDisbindin# 1992;51:313-7. gene-related peptide-, and c-flanking peptide of neuro- 47 Mapp P I, Walsh D A, Kidd B L, et al. Receptors to the peptide Y- immunoreactive fibres are present in normal neuropeptide substance P in human synovium. Br J synovium but depleted in patients with rheumatoid arthritis. Rheumatot 1990; 29 (suppl 1): 44. Neuroscience 1990; 37: 143-53. 48 Cruwys S C, Kidd B L, Mapp P I, Walsh D A, Blake D R. 28 Wilkinson K D, Lee K M, Deshpande S, Duerksen-Hughes Substance P-a specific vascular receptor in rat synovium. P, Boss J M, Pohl J. The neuron-specific protein PGP 9-5 is Br Rhewnatol 1991; 30 (suppl 2): 44. a ubiquitin carboxyl-terminal hydrolase. Science 1989; 246: 49 Engberg G, Svensson T H, Rosel S, Folkers K. A synthetic 670-3. peptide as an antagonist of substance P. Nature 1981; 293: 29 Pereira da Silva J A, Carmo-Fonseca M. Peptide containing 222-3. nerves in human synovium: immnunohistochemical evidence 50 Lam F Y, Ferrell W R. Inhibition of carrageenan induced for decreased innervation in rheumatoid arthritis. inflammation in the rat knee joint by substance P J Rheumatol 1990; 17: 1592-9. antagonist. Am RheumDis 1989; 48:928-32. 30 Docherty A J P, Murphy G. The tissue metalloproteinase 51 Lam F Y, FerrelI W R. Capsaicin suppresses substance family and the inhibitor TIMP: a study using cDNAs and P-induced joint inflammation in the rat. Neurosci Lett recombinant proteins. Ann Rheun Dis 1990; 49:469-79. 1989; 105: 155-8. 31 Winyard P G, Zhang Z, Chidwick K, Blake D R, Carrell 52 Snider R M, Constantine J W, Lowe IIIJ A, et al. A potent R W, Murphy G. Proteolytic inactivation of human a, non-peptide antagonist of the substance P (NK1) receptor. antitrypsin by human stromelysin. FEBS Letu 1991; 279: Science 1991; 251:435-7. 91-4. 53 Gitter B D,Waters D C,Bruns RF,MasonNR,NixonJA, http://ard.bmj.com/ 32 Zhang Z, Winyard P G, Chidwick K, et al. Increased HowbertJ J. Species differences in affinities of non-peptide protelytic cleavage of(a antitrypsin (ax proteinase inhibitor) antagonists for substance P receptors. Eur J Pharmacol in knee-joint synovial fluid from patients with rheumatoid 1991; 197:237-8. arthritis. Biochem Soc Trans 1990; 18: 898-9. 54 Beresford I J M, Birch P J, Hagan R M, Ireland S J. 33 Fink T, Weihe E, Multiple neuropeptides in nerves supply- Investigations into species variants in tachykinin NKI ing mammalian lymph nodes: candidates for sensory and receptors by use of the peptide antagonist, CP-96, 345. BrJ autonomic neuroimmunomodulation? Neurosci Lett 1988; 1991; 90: 39-44. Pharmacol 104:292-3. 34 Stanisz A M, Befus D, Bienenstock J. Differential effects of 55 Andrews P V, Helme R D, Thomas K L. NK1 receptor vasoactive intestinal peptide, substance P and mediation of neurogenic plasma extravasation in rat skin. on immunoglobulin synthesis and proliferations by lym- BrJPharmacol 1989; 97:1232-8. on October 2, 2021 by guest. Protected copyright. phocytes from Peyer's patches, mesenteric lymph nodes 56 Brain S D, Williams T J, Tippins J R, Morris H R, and spleen. J Immunol 1986; 136: 152-6. Maclntyre I. Calcitonin gene-related peptide is a potent 35 Lot;. M, Carson D A, Vaughan J H. Substance P activation of vasodilator. Nature 1985; 313: 54-6. rheumatoid synoviocytes: neural pathway in pathogenesis 57 Lundberg J M, Terenius L, Hokfelt T, etal. Neuropeptide Y ofarthritis. Science 1987; 235: 893-5. (NPY)-like activity in peripheral noradrenergic neurons 36 Tissot M, Pradelles P, Giroud J P. Substance P-like levels in and effects of NPY on sympathetic function. Acta Physiol inflammatory exudates. Inflammation 1988; 12: 25-35. Scand 1982; 116:477-80.