Evidence for Both Histamine H1 and H2 Receptors on Human Articular Chondrocytes

Home , H2 antagonist, Histamine H2 receptor

Ann Rheum Dis: first published as 10.1136/ard.46.6.431 on 1 June 1987. Downloaded from

Annals of the Rheumatic Diseases, 1987; 46, 431-435

Evidence for both histamine H1 and H2 receptors on human articular chondrocytes

D J TAYLOR AND D E WOOLLEY From the University Department of Medicine, University Hospital of South Manchester, West Didsbury, Manchester, UK

SUMMARY Using specific histamine HI and H2 receptor antagonists, evidence is presented for the existence of both H1 and H2 receptors on human articular chondrocytes in vitro. Stimulation of the H1 receptor by histamine (range 0-18 to 17-8 imol/l) significantly increased prostaglandin E (PGE) production, while activation of the histamine H2 receptor increased intracellular cyclic adenosine-5'-monophosphate (AMP). The histamine H1 antagonists mepyramine and tripelennamine blocked the histamine induced increase in PGE production, and the H2 antagonists cimetidine and ranitidine prevented the increase in intracellular cyclic AMP. These observations suggest that mast cell-chondrocyte interactions mediated via histamine may contribute to some of the pathophysiological changes observed in joint disease.

Key words: adenosine cyclic monophosphate, prostaglandins E. copyright.

After our observation of mast cells at sites of was shown to be mediated not by a histamine H2 cartilage erosion in rheumatoid knee joints' we have receptor but by a type 1 receptor. examined the potential of mast cell components to affect chondrocyte metabolism. The addition of Materials and methods http://ard.bmj.com/ whole mast cell products, prepared from purified rat or dog mast cells, caused a marked increase in the Materials were obtained from the sources previously intracellular cyclic AMP levels of cultured chondro- given,2 with the addition of the following [3H]PGE2 cytes, an observation subsequently explained by the was obtained from Amersham International, Amer- demonstration of histamine H2 receptors on chon- sham, Bucks, UK; anti-PGE serum was obtained drocytes derived from human, canine, and fetal from Miles Laboratories Limited, Slough, UK; bovine articular cartilage.2 As with histamine H2 recombinant murine interleukin 1 (ILl) was a gift receptors found in gastric mucosa, rat uterus, and from Roche Products Ltd, Welwyn Garden City, on September 27, 2021 by guest. Protected guinea pig heart,3 stimulation by histamine activated Herts and was prepared as previously described.6 adenylate cyclase and produced an increase in (One unit of ILl activity induces 50% maximal intracellular cyclic AMP. Alterations in intracellular proliferative response in the thymocyte costimulation cyclic AMP levels have been shown to affect a assay.) number of cellular processes, including prostaglandin (PG) production by human adipose tissue.5 CELL CULTURE Since increased PG production is a common feature Human articular chondrocytes (HAC) were obtained of inflammatory joint disease we have investigated by proteolytic digestion of macroscopically normal the effect of histamine on PGE production by articular cartilage from femoral heads and condyles chondrocytes. We report here that histamine was obtained from remedial surgery as previously de- found to increase PGE production by human scribed.7 The cells were grown in Dulbecco's modi- articular chondrocytes, but surprisingly this process fled Eagle's medium (DMEM) with 10% (v/v) fetal calf serum supplemented with penicillin, strepto- Accepted for publication 8 January 1987. and Correspondence to Dr D J Taylor, University Department of mycin fungizone. Cultures were incubated at Medicine, University Hospital of South Manchester, West Didsbury, 37°C in CO2/air 1:19 in a water saturated atmosphere Manchester M20 8LR, UK. and when confluent subcultured using conventional 431 Ann Rheum Dis: first published as 10.1136/ard.46.6.431 on 1 June 1987. Downloaded from

432 Taylor, Woolley trypsin treatment and replating at reduced density. further 10 min at 4°C. After centrifugation at 1500 g The chondrocyte cultures used in these experiments for 10 min at 4°C the supernatant was decanted and were from primary and passaged cells, both of which counted in a liquid scintillation spectrometer con- were qualitatively similar in their response to nected to a Beckman DP5500 curve-fit processor. histamine. The sensitivity of the assay was 5 pg when deter- mined by two standard deviations at zero dose. CYCL IC AMP STUDIES For measurement of intracellular cyclic AMP the [-H]MEPYRAMINE BINDING STUDIES chondrocytes were incubated in DMEM containing Radioligand binding studies were performed on 800 itmol/l isobutyl-L-methylxanthine (IBMX) with confluent monolayers of chondrocytes using the and without histamine and antagonists for five method recently described for rheumatoid synovial minutes at 37°C. After removal of the incubation cells.14 After a 90 min incubation at 0-40C with medium the cells were precipitated with 6% (w/v) 10 nM [3H]mepyramine the medium was removed HCl04 and processed for cyclic AMP determination and the cells were rapidly rinsed with buffer before as described previously.8 The cyclic AMP was detaching by incubation at 37°C with 0-25% (w/v) measured by a competitive protein binding assay trypsin. The cells were transferred quantitatively to using the binder isolated from bovine adrenal vials for liquid scintillation counting. All treatments glands.9 were performed in duplicate or triplicate.

PGE STUDIE,S Results To examine the effect of histamine on PGE produc- HISTAMINE EFFECTS ON PGE PRODUCTION tion chondrocytes were preincubated for between 16 Histamine stimulated PGE production by HAC in a and 24 hours with 10 or 20% (v/v) synovial factor (SF), which has previously been to stimu- reported PGE late chondrocytic PG production.' This overcame Table I Histaminie stimullation of chondrocYte copyright. the problem of very low basal PGE production, production which was normally at the limit of assay sensitivity. Treaotneot P(E (ng/well1/h) SF was the medium removed after three days from a primary culture of adherent rheumatoid synovial DMEM 2^20 (0(08);), cells, known to be a good source of interleukin 1,11 Histamine (()- 18 tmou/l) 2 79 (0109)* hiistaminc (1 8 fimol/l) 464 (() 11)* and after centrifugation to remove cellular debris Histaminc (17-8 ltmol/1) 5 24 (0(09)*

was used without further treatment. Cells were also +Mcpsraminc (143 timol/1) 2 17 (0.05) http://ard.bmj.com/ preincubated for 20 hours with 16 units/ml murine +Cimctidinc (4(0 umol/) 4-30 (0-20)* ILI, which like SF markedly stimulated chondrocyte The chondrocvtes were prcincubatcd for 24 h with 20 (XIs) SF. prostaglandin E production. After removing the Thc activating mcdium was rcmovcd and the cclls wcrc washed activating medium the cells were washed three times with HiBSS bcforc adding histaminc and its antagonists for onc with Hanks's balanced salt solution (HBSS) and hour. DMEM was added with or without histamine and its :p<()-()|I. antagonists. This second incubation was for either ,Valucs (irc the mcan (SEM). n=4. on September 27, 2021 by guest. Protected one or two hours, after which the medium was Table 2 Histanminle stimtulatiotn of PGE produictiotn front removed for PGE measurement. The inclusion of chonidrocvtes preiticlbhated with SF or mnurinie ILI indomethacin (14 [tmol/l) reduced PGE production to non-detectable levels, which confirmed that PGE Freattient1t 'GE (Oigl'we/Ilh obtained during the histamine incubation represented new synthesis and not leakage of preformed material. SF activated I lI (activiated PGE was measured in culture media by 6-13 (0.) ; (1185) 12 DMEM 12(08 radioimmunoassay using an antiserum with similar Histaminc (4.5 umol/1) 15X81 ((097)'' 23.57 ( 2'09):' specificity towards the prostaglandins El and E2 and +Mcpyramine (0-5 pmol/1) 5-69 (-164) 14 76 (1)46) charcoal to bound +Tripeclennamine (0-5 ItmolI/) 4-75 ((1-2t9) 8 21 (05-7) using dextran coated separate +Cimctidinc(79 rmolIl) 1174 (1)-51) 188-1 (()89)* from free fractions.13 The antiserum was diluted in +Ranitidinc (6.3 umonlI) 14,-21 ((138)* 20-83 (2 64)' assay buffer (50 mM phosphate, 0 15 M NaCl, 0 1% (w/v) bovine serum albumin, and 0-05% (w/v) The chondroctes were prcincubated for 20) h with cithci 1)0 (x/v) at and 100 was added to the SF or 16 units/ml murine ILI. The activating mcdium wias rcmoxcd sodium azide pH 7.4) [tl and the cclls werc washcd with HBSS before adding histatminc and same volume of sample and [3HIPGE2. After 90 min its antaigonists for onc hour. at 4°C 1 mg of dextran coated charcoal was added to "Valucs significantlv incrcased abovc control at p<()-(l2. each tube, which after vortexing was incubated for a Valucs ,re thenmcan (SEM). n=3. Ann Rheum Dis: first published as 10.1136/ard.46.6.431 on 1 June 1987. Downloaded from

Histamine receptors on human chondrocytes 433 concentration related manner, with concentrations 6-o as low as 0-18 .mollI causing a significant increase above control (Table 1). In four preparations of C HAC, histamine (17.8 ,umol/l) increased PGE pro- .iI.. 5- duction by an average of 5.4 (range 2*3-11-3) times 0. on 4- that of the control. The stimulation was prevented E \0 by the receptor antagonists mepyramine and CV) HI o 3- tripelennamine, was not significantly reduced by a x 10-fold higher concentration of the H2 antagonist 1- ranitidine, and only slightly reduced by cimetidine c 2- (Tables 1 and 2). l- 0 The H1 antagonists chlorpheniramine, mepy- a-I ramine, and tripelennamine each caused a dose C. A~\u related inhibition of the histamine stimulated PGE production at concentrations below 10-6 mol/l (Fig. 2 1). When added without histamine the highest -log I [CNorpheniramine] (mol/l) concentration of histamine antagonists used in these experiments did not significantly change the control Fig. 2 Displacement by chlorpheniramine of values of PGE production (result not shown). The [3H]mepyramine bound to HA C. The curve shows the amount of[3Hjmepyramine bound to HA C as a function of the amount ofchlorpheniramine added. The [?HJmepyramine binding was performed as described in 25- 'Materials and methods'.

PGE response of histamine stimulated HAC was copyright. approximately doubled after preincubation with 20- either murine recombinant ILl or synovial factor (SF), an increase prevented by HI but not H2 receptor antagonists (Table 2).

[3H]MEPYRAMINE BINDING TO HAC

Five preparations of subcultured HAC bound sub- http://ard.bmj.com/ stantial amounts of [3H]mepyramine, which was displaced by unlabelled mepyramine, tripelennamine, and chlorpheniramine at concentrations above 1 imol/l (e.g., Fig. 2), but not by 2-5 mmol/I ranitidine or cimetidine. Preincubation of the HAC for 17 h with murine recombinant ILl (12 units/ml) 5- had no effect on the amount of [3H]mepyramine subsequently bound by the cells. on September 27, 2021 by guest. Protected

0 EFFECT OF HISTAMINE ON HAC INTRACELLULAR CYCLIC AMP oJ 6-5 6 Histamine (17.8 Rmol/l) in the presence of a -log )I oist (MOVIl)I phosphodiesterase inhibitor produced an approximately fivefold increase in HAC intracellular cyclic Fig. 1 Efject ofH, antagonists on the histamine stimulated AMP, a stimulation effectively inhibited by the two PGEproduction by HA C. The chondrocytes were H2 but not H1 receptor antagonists (Table 3). The preincubatedfor 24 h with 20% (vlv) SF. The activating histamine induced rise in HAC cyclic AMP was not medium was removed and the cells were washed with HBSS a consequence of increased PGE production as the before adding DMEMfor one hour. O= DMEM; A =DMEM+histamine (1788vmolll); inclusion of indomethacin (14 ,umolI1) failed to *= DMEM+histamine (17-8 tunolll) +mepyramine; prevent the increase. Furthermore, the chondro- *=DMEM+histamine (1788smolll)+chlorpheniramine, cytes used in the cyclic AMP studies had not been O=DMEM+histamine (178 pnolll)+tripelennamine. preincubated with SF and would have had negligible Results are means (SEM) (bars) oftriplicate values for PGE biosynthesis, especially over such determinations. short (five minute) experimental incubations. Ann Rheum Dis: first published as 10.1136/ard.46.6.431 on 1 June 1987. Downloaded from

434 Taylor, Woolley Table 3 Effect of H, and H, antagonists on the histamine intracellular cyclic AMP recently reported.2 The induced increase of intracellular cyclic AMP in human association of the H2 receptor with adenylate cyclase articular chondrocytes and the production of cyclic AMP has been demon- Treatment Cyclic AMP Inhibition of strated in homogenates from many tissues, including (pmollwell) ryclic AMP cardiac muscle, gastric mucosa, and brain, and also increase (%) in isolated cells such as adipocytes and certain white blood cells.3 Therefore the prevention of the hista- DMEM+IBMX* (0-8 mmolI1) 4-23 (0-38)t. mine induced rise in Histamine (17-8 Rmol/1) 22-54 (0-28) HAC cyclic AMP by H2 but not +Mepyramine (12.3 jAmol/l) 23-61 (1-27) 0 H1 receptor antagonists (Table 3), together with the +Tripelennamine (17.1 jimoll1) 23-03 (0-40) 0 parallel shift of the histamine dose-response curve +Ranitidine (15 7 ,umoUl) 5-71 (0-17) 92 by increasing concentrations of cimetidine,2 pro- +Cimetidine (19.8 imol/1) 6-71 (0-26) 87 vides conclusive evidence that chondrocytes express Histamine and its antagonists were added to the cells for five H2 receptors. minutes, after which the medium was removed and the cells This study presents evidence that HAC possess assayed for cyclic AMP. both H, and H2 histamine receptors and that the *IBMX=isobutyl-L-methylxanthine. tValues are the mean (SEM), n=3. histamine induced stimulation of cyclic AMP and prostaglandin production is mediated via two separate processes (H2 and H1 respectively). The pos- session of both receptor types on chondrocytes is not Discussion unique since rat glomerular mesangial cells in culture have recently been shown to respond to We have previously shown that histamine stimulates histamine by events mediated by both receptors.16 human articular chondrocytes to produce increased Further studies are required to assess whether these amounts of prostaglandin E, an event probably receptor induced processes are interactive. The

mediated via histamine HI receptors.'5 This study histamine induced rises in PGE and cyclic AMP maycopyright. has demonstrated the ability of three H, receptor influence many cellular processes,4 17 including the antagonists to cause a dose related inhibition of production of degradative enzymes such as collage- histamine stimulated PGE production at concen- nase and plasminogen activator described for human trations below 10-6 mol/l, an effect not shown by synovial cells.'8 19 In addition, recent studies have histamine H2 receptor antagonists. Although much shown that the local production of PGE2 transforms of the data have been obtained from 'activated' cultured fibroblasts to cells with dendritic morpho- this from a 20 logy and increased proteinase chondrocyte cultures, resulting hour expression.20 http://ard.bmj.com/ preincubation with SF or ILl, a similar histamine The finding that both human articular chondro- stimulation of PGE production was obtained with cytes and rheumatoid synovial cells'4 have surface control, 'non-activated' chondrocytes.'5 The posses- histamine receptors is an important observation in sion of H, receptors therefore appears not to be view of the recent reports of raised histamine levels dependent upon preincubation with SF or ILl, but is found in rheumatoid but not osteoarthritic synovial a feature of both control and 'activated' chondro- fluids.2' Cells obtained from trabecular bone also cytes. responded to high concentrations of histamine with The existence of H, receptors on HAC was increased cyclic AMP,22 and 10-2-10-3 M hista- on September 27, 2021 by guest. Protected confirmed by the 3H labelled H, antagonist mepy- mine was reported to reduce bone resorption in ramine, which bound readily and with specificity to cultured mouse calvaria.23 The extent to which all the HAC preparations tested. There was no histamine may influence the metabolism and de- evidence that ILl pretreatment increased the ex- gradative mechanisms in joint disease is uncertain, pression of histamine H, receptors by the chondro- but the recent reports of mast cells in synovial fluids cytes since no change in [3Hlmepyramine binding from patients with diverse arthritides,24 the increased was observed for cells incubated with or without numbers of mast cells in rheumatoid synovial ILl. Stimulation of the H, receptor in some tissues tissues,25 26Athobevand the observation of mast cells at sites has previously been associated with an increase in of cartilage erosion in rheumatoid joint specimens' intracellular cyclic guanosine-5'-monophosphate suggest that the mast cell may be implicated in some formation,3 but attempts to detect an increase in this of the pathophysiological changes observed in joint nucleotide from histamine stimulated HAC were disease. Indeed, as chondrocytes are stimulated to not satisfactory, and the PGE response proved produce collagenase by mast cell products27 and to be a much better index for H, receptor activation. possess both histamine HI and H2 receptors it is Evidence for the existence of H2 receptors on probable that histamine 'activation' could have HAC was provided by the histamine induced rise in profound effects on various aspects of chondrocyte Ann Rheum Dis: first published as 10.1136/ard.46.6.431 on 1 June 1987. Downloaded from

Histamine receptors on human chondrocytes 435 metabolism, an interaction that may well be relevant 13 Kahn D, Andrieu J M, Dray F. Evaluation of some binding to the aberrant chondrocytic behaviour observed in parameters of hapten-antibody complexes using dextran-coated charcoal to separate the bound and free fractions. Immuno- degenerative osteoarthritis and some examples of chemistry 1974; 11: 327-32. rheumatoid joint disease. 14 Taylor D J, Woolley D E. Histamine HI receptors on adherent We thank Mr H Bertfield (Wythenshawe Hospital, Manchester) rheumatoid synovial cells in culture: demonstration by radio- and Mr M A Morris (Devonshire Royal Hospital, Buxton) for the ligand binding and inhibition of histamine stimulated prosta- provision of surgical specimens, and Mrs M Williamson for glandin E production by histamine HI antagonists. Ann Rheum preparation of the manuscript. This work was supported by grants Dis 1987; 46: 425-30. from the Arthritis and Rheumatism Council and the North West 15 Taylor D J, Yoffe J R, Brown D M, Woolley D E. Histamine stimulates prostaglandin E production by rheumatoid synovial Regional Health Authority. cells and human articular chondrocytes in culture. Arthritis References Rheum 1986; 29: 160-5. 16 Sedor J R, Abboud H E. Histamine modulates contraction and 1 Bromley M, Fisher W D, Woolley D E. Mast cells at sites of cyclic nucleotides in cultured rat mesangial cells. J Clin Invest cartilage erosion in the rheumatoid joint. Ann Rheum Dis 1984; 1985; 75: 1679-89. 43: 76-9. 17 Malemud C J, Moskowitz R W, Papay R S. Correlation of the 2 Taylor D J, Yoffe J R, Brown D M, Woolley D E. Histamine biosynthesis of prostaglandins and cyclic AMP in monolayer H. receptors on chondrocytes derived from human, canine and cultures of rabbit articular chondrocytes. Biochim Biophys Acta bovine articular cartilage. Biochem J 1985; 225: 315-9. 1982; 715: 70-9. 3 Johnson C L. Histamine receptors and cyclic nucleotides. In: 18 Woolley D E, Taylor D J, Yoffe J R. Mammalian collagenases: Ganellin C R, Parsons M E, eds. Pharmacology of histamine aspects of regulation. Life Chem Reports 1984; 2: 181-235. receptors. Bristol: Wright, 1982: 147-216. 19 Hamilton J A, Leizer T, Lingelbach R. The stimulation of 4 Robeson G A, Butcher R W, Sutherland E W. Cyclic AMP. human synovial fibroblast plasminogen activator activity. In- New York: Academic Press, 1971. volvement of cyclic AMP and cyclooxygenase products. 5 Mitchell M D, Cleland W H, Smith M E, Simpson E R, Biochim Biophys Acta 1986; 886: 195-202. Mendelson C R. Inhibition of prostaglandin biosynthesis in 20 Gadher S J, Woolley D E. Comparative studies of adherent human adipose tissue by glucocorticoids. J Clin Endocrinol rheumatoid synovial cells in primary culture: characterisation of Metab 1983; 57: 771-6. the dendritic (stellate) cell. Rheumatol Int (in press). 6 Lomedico P T, Gubler U, Hellmann C P, et al. Cloning and 21 Frewin D B, Cleland L G, Jonsson J R, Robertson P W.

expression of murine interleukin-1 cDNA in Escherichia coli. Histamine levels in human synovial fluid. J Rheumatol 1986; 13: copyright. Nature 1984; 312: 458-62. 13-14. 7 Meats J E, McGuire M K, Russell R G G. Human synovium 22 Crisp A J. Studies of histamine in the rheumatoid joint. releases a factor which stimulates chondrocyte production of Rheumatol Int 1984; 4: 125-8. PGE and plasminogen activator. Nature 1980; 286: 891-2. 23 Crisp A J, Wright J K, Hazleman B L. Effects of heparin, 8 Taylor D J, Yoffe f R, Woolley D E. Histamine Hz receptors histamine, and salmon calcitonin on mouse calvarial bone on foetal-bovine articular chondrocytes. Biochem J 1983: 212: resorption. Ann Rheum Dis 1986; 45: 422-7. 517-20. 24 Malone D G, Irani A-M, Schwartz L B, Barrett K E, Metcalfe 9 Tsang C P W, Lehotay D C, Murphy B E P. Competitive D D. Mast cell numbers and histamine levels in synovial fluids

binding assay for adenosine 3',5'-monophosphate employing a from patients with diverse arthritides. Arthritis Rheum 1986; 29: http://ard.bmj.com/ bovine adrenal protein. J Clin Endocrinol Metab 1972; 35: 956-63. 809-17. 25 Crisp A J, Chapman C M, Kirkham S E, Schiller A L, Krane 10 Russell R G G, McGuire M K B, Meats J E. Ebsworth N M, S M. Articular mastocytosis in rheumatoid arthritis. Arthritis Beresford J. Intercellular messengers in joint tissues in rheuma- Rheum 1984; 27: 845-51. toid arthritis. Scand J Rheumatol (SupplI 1981; 40: 75-87. 26 Godfrey H P, Ilardi C, Engber W, Graziano F M. Quantitation 11 Wood D D, Ihrie E J, Hamerman D. Release of interleukin-1 of human synovial mast cells in rheumatoid arthritis and other from human synovial tissue in vitro. Arthritis Rheum 1985; 28: rheumatic diseases. Arthritis Rheum 1984; 27: 852-6. 853-62. 27 Woolley D E, Yoffe J R, Evanson J M. Mast cells and matrix 12 Levine L, Gutierrez-Cernosek R M, Van Vunakis H. Specif- degradation in the rheumatoid joint. In: Tschesche H, ed. icities of prostaglandins B1, F, and F2 antigen-antibody reactions. Proteinases in inflammation and tumour invasion. Berlin: on September 27, 2021 by guest. Protected J Biol Chem 1971; 246: 6782-5. Walter de Gruyter, 1986: 61-76.