-?-K

oxY6Eil RßD|CÊLS RllD T|SSUE DfiÌlÊfiE:

T]IE EFFECT OF OXYGEil RRDICRLS O]I T]IE STPUCTURE

ffr{D sYilTHEStS 0F }tYRLUR0lllC RCID

by

JULIÊN DAVID llcNElL H.8.8.5., F,tl.Íl.C.P.

Ê Thesls Submitted for the Degree of Doctor of Philorophy

in The Unirersity of Êdelaida.

Department of PathologyThe llnlrersity of Adelalde

December, 1986.

:.), \ \..1? 1i\'*i.:... ,! t,d\,.\u^o

TâBLE OF COTTE;Í5

DECtâRTTIOT

ACßTOTIEDGIIEXTS

5U1,IHâNT

GrosfìâRr

it

iv

vil

I I flTRoDllcTlox AtD lErl Elf¡:-- GaËtal lntroductlon 2'É.

oxuqtl ntdtctlt trú Jol nt Blolog y 4

l. lntroductlon?. Daflnltlonr tú clPmlttrg

t. 9uProxldc tnlonb. HlJdrryn Pcrod&c. Hgdroxgl rtdlæld. Slftgltt oxlsôn

3. Posslblciouræïú oøoryn radlcalt in thc lnlltrrd Jolntt. Tlp r¿lææ sl wW¡ radle¡lr from ætivtttd ptnqocgtts

b. pæsiblc ¡¡¡pn iÁttæl productlon dw to lln grærtnsl rúwl¿¡rts tld trtnsltlonmattl¡ rrithln t]æ lnfltm.d Joint

c. Could cgnovlal iæhæmit lcáO to o(Wn rdictl Prsd'ßllon1¡rtJ3.?4. Protactlvt mæhnitm¡

t. SuProxlü dltmuttætb, Gluttthtonc/gluttthiona Wrwl&rcc. Cattlæed. tllçr dB.lc¡wtc. fl¡<lûnt &lerrrs vlthf n thc Jolnt

5. Ttrgatt for oxg rdiæl tttækô. Protcintb. ilrælcic æld cnd DllA

c. l"tcmbrtna llÍrldtd. Extræcllultr mæromolæulcs otlçr thtn hpluronic æidÊ. Ccllultr targett

| ) Rd ælltÐ 0tlpr ælls

C. Hgtluronic Acidl. lntroductlon2. Chtmlctl rtrrcturc5, Hytluronlc æld ln qæow tolutlon--tho formatlon ú t ¡¡rtrß4. lñpllcttlontlor tymvltl lluld furction

ü. Lubrlætlonb. Yltcælcrtlc aroPrllstc. A codult for nutrlantr

5. Eiotqnthcalr6. Turmvcr7 . Hgtl uronic æid-æll I ntcrætiont8. Sgnovltl fluld hPluronlc acld

2?

D. St¡llæ on (hrgcn Rtdlæl lnducod Hpluronlc Acld hpolUrprlzrtlon 42l. lnltltl tttllæ2. Hlttlon lnduccd hpluronlc æld dcgrtûtlonÍ. fhldatlon- rcductloñ dcpol Umarlætlon (0RD ræctlon)4. Emgmatiællg producad oxpn radlælr tld lnhlbltorr5. lmprtaru of mrltl clrlatorr tnd cnttlnlltmmtorg drqc6. Effæt of suproxl&dhmutæc?. DcArdttlon of hrpluronlc æid bg ncutrophlb8. l'lolæular vcight ch¡rp¿s9. Conclusion

il. TltE DEpotTHEnratTt0x p[oDucTs 0F HTAtuRoilc Æ¡D âFTER EXP0!¡UnE T0 0DrR lf,TITNOFlntrøuctlon 52B. I'lrtarials tld t4cthoú 55

l. Hgaluronlc æld Purification2. 0xggrn radlctl çncrdlonä. Farrour lon tutocldttionb. xÛl¡þt,c. Stlmuldad trlUrprphonrrlær lcucocqtor

!, Vlrcomctrg4. Êal chromatogrtphu5. Uronlc ætd tnal¡pit6. Amlgtictl ultræcntrlfugttfon

C. Ræultc 57l. Visco¡ltu2, Ì4olccultr cxclurion çl chromctogrtphg

t. Scphtdcx f¡- 100b. $ptnræc 48c. *glnrw 2B

3. Amlgtictl ullrlrn¡ûriitrytionD. Discusrion 67

?2E. Summarg

ilt. GHAIGES tt TltE ]to[EcutâP tfEtGHT SPECTRâ 0F STf,fHlAt rtulD Hrâtumtlc tclDSEET IT ITTLAHHâTOPT.NITT D|5EâI¡Eå. lntroductlon 75B. llttaritls and mctfre& 77

l. SgnMtl fluidZ. Hgaluronic æid reptrttion5. Eclluloæ tcctata trl clcctrophoræit4. Gcl chromatogrcphg5. Uronic æid tmlUsis6. 0xgçn radictl çncrationRæultsDiæuæionSummrg

IY. I,IEASUREI'IE]IT OF REDIICIlG E]ID GROUPÍ¡ GEIEPATED 8I EXPfT!¡UTE OF HYAI.URO]IIC

ACID TO OXTGE]I RâDICâU¡A. lntroduction 93B. Hdcrltls and mattpdr 94

l. 0xggcn rdlctl çmrttionË. Fcrrous lon ¡utoxlútlonb. xDlJ#

c.D,

E.

82889l

2. Erugmtlc dcgrútlon of hçluronlc ældÊ. Tætlcultrhpluronlfutb. Strcptomgæs hgtluronldæt Ûlçttlon

5. Gcl chromttognPhg4. Colorlmatrlc $ttlp

Ë. Uronic ældb. Ptrk-Johmn c$tgc. l'lorpn-flæn ruætlon

5. C$nlü lcbtlllm6. Atccnding ptpr chrorutagruPhU

C. RtsultsD. DlæussionE. Sumrurg

T. DOES ÍITTGET NTDICâI EXPÍËURE âITER THE âBILITT OF HTâLUROTIC âCID TO

l'l0DULtTE I{åER0PHåGE Fc REGEPT0R BltDltG?A. lntroduction I 08B. lttcritls and methods 109

l. Adhcrcnt monæUtæ?. Fc rcccptor æsag

t. Praptrttion of antiMt¡ cottcd red blæd ællsb. Fc rcccPtor asstgc. Sttining tnd æuntiq

3. Hgtluronic æid prcparttiont. 0xr¡gcn radictl cxposed htpluronic æidb. Enzgmc cxPoscd hgaluronic æid

C. Rrsultr lllD. Discussion ll4E. Summtrg ll?

TI. ETPOSURE OF CUTTURED STMTIåI GETLS TO ONTGET RTDIGâI FTIHE!ì. ETMTüßIGITTâTD EFTEGT OlI IITAIUMXIG TCID PRODIETIO]IA. lntrodrætlon I 19

B. Matarlals and l'tthdr t l9l. S¡novltl æll culturt?. Chromium rularstæu¡3. fhgçn rtülctl çncntion4. Hgtluronic æid Prodtrtion5. Ctllulæa tætttu çl tlætrophor¿sis

C. Ræults I e5l. N0/HX llductd cgtotoxlcitg2. Effcct of oxgçn raüictl cxpoturc on lH production

D. Discussion I 5eE. Summaru I 36

il t.GEIERAI- DISCUSSIoX

APPEXDICESI Dctail of thc Colorimetric æsôUlll Titsue Culture l.1eúia

lll Publicctionr

9?t04t06

137

t49152r53

t?0BIBTIOGRAPHT

I

DECLRRRTIOII

Thls thesls contalnr no materlal uhlch hac bcÉî accepted or

submltted for the auard of any other degnee or dlploma ln any

lJnlrerslty. Furthumore, to the best of my knouledge and bellef, thls

thesls contalns no matenlal prerlously publlshed on r¡rltten by another

pÊnson, except uhere due referÊnce ls made ln the text.

The r*¡ork descrlbed hereln has been the subJect oî the follorrrlng

publlcatlons:-

HcÌ{ElL, J.D., tdlEEKltl, 0.1d., EETT9, ld.l'|. and CLELÊÌ{D, L.E. (1985)

Depoþerlzatlon products of hyaluronlc acid after Êxpoiureto oxygen derlved free radlcals.Ênnals of the Rheumatlc Dlseases 11, ?80-?Eg.

McllElL, J.D,, lrJlEEKll{, 0.hJ., CLELÊIiD, L.6. and 7ERN0i{-R0BERT3, E.

( 1986). The generation of reduclng ends by exposure ofhyaluronlc acid to oxygen derired free radicals.Êgents and Êctlons Supplements 18, 95- lg l.

and abstract¡:-tlcNElL, J.0., lrjlEEKll{, 0,lrJ., CLELÊHD, L.E. and EETTS' hj.ll. ( lgBl)

Ênalysls of hyaluronlc acld degradatlon products afterexposurË to an oxygen derlved free radical flux.Connective Tissue Research I l, 25A.

Hct{ElL, J.D., kJlEEKllt, 0.kJ., ELELAiID, L.E. and EETT3, [d,H. (1983)Hechanlsms of hyaluronlc acid degradatløt by oxygen derlvedfree radlcals. Aust. N. Z. J. Hed. 13, 2l l.

tlcHElL, J.D., CLELÊND, L.E. and EETTS, lrJ.ll. (1983)The cytotoxlclty of oxygen derlrred frce radlcal¡.Êust. ll, Z, J. Hed. l\ 225.

HcllElL, J.0., [ÂllEEKllJ, O.bJ., and ELELÊIJD, L.E. (1981)Ênalyrls of the molecular uelghts of hyaluronlc acld fromlnflamed and non-lnflamed synovlal flulds.Êust, ì1. Z. J. Hed. l{ (suppl, l'), 313.

llcNElL, J.0., lrJlEEKltl, 0.UJ., and CLELÊHD, L.6, (tgS5)The cytotoxic effect oî oxygen dertved lree radlcals on borinesynovial cells in culture. Aust. 11. Z. J. 357.

Slgned:J. D. HcHeil

December, l9EE

1i.

ßCKll0l¡lLEDGllEllTs

The urork degcrlbed ln thlr thesls uras :upported by grantt from

the Natlonal llealth and lledlcal Research Councll of Rustralla and

from an E. R. Dau¡es fellor¡shlp from the Royal Adelalde llospltal.

I hereby ulsh to thank my supÊrylsors Dr. Ole lrliebktn, Dr. Leslie

Cleland and Professor Barrle Pernon-Roberts. Dr. Ole lrjlebkln has

helped me r¡lth day to day teachhngr guldance, enthuslasm and

frlendshlp. Dr. Leslie Eleland has lnfluenced me touards a càreer

lnrolrlng laboratory resÊàrch and has provlded support, raluable

coun¡el and frlendshlp. Professor Earrle /ernon-Roberts has prorided

orzerall superrlslon and advlce and has provlded all the facilltles of hls

department for thls uork. I also r¡lsh to thank Proîessor J. L.

Skosey rr.rho provlded valuabla supervision and edltorlal as¡lstance

durlng the r¡rltlng up phare.

0ther people uho hare helped mË ärÊ Dr. ld. llenry Eettc urho

prorlded teachlng help and as¡lstance r¡lth the chemlstry of oxygen

radicals, Dr. Stephen Mllazzo r¡ho provlded rraluable interect, support

and cltnlcal teaching, llrr. S. C. lrJlabkln rr.rho provlded agglstance in

tcchnical matters, l1?s. Eenevieve LaPinska r¡ho provided uilling help

ln typlng thls manuscript, I1n. Dayid llaynes for help r¡ith the

macrophage Fc receptor binding àssåpr and Hr. Dale Carille r¡ho

contributed of his photographic talents. I haye also had raluable

discussions r¡ith Dr. lllchael lrJhltehouse for r¡hlch I am grateful.

I am also indebted to the South Êustrallan lleat Corporation

for thelr glft of bortne hocks and Pharmacla, South Seas, for thelr

glft of llealon.

111.

I r¡ould llke to make a speclal trlbute to my late father and my

mother, both of uhonr hare provlded me' orer many y¿a?\ r¡rlth an

lnexhaustlble supply of encouragËmÊ-nt and support. Ftnalþ' I t¡ould

llke to thank my ullfe, Jane, r.rho has patlently cndured so much so

that I could undertake thls proJect.

iv.

SU}IIIRRY

ln order to examlnc the role of oxygÉTr radlcalc ln lnflarnmatory

Jotnt damage, a study of oxygen radlcal lnduced hyaluronlc acld

depolymerlzatlon uas undertaken. Hyaluronlc acld u,às uscd because

It ls a maJor extracellular compon"nt of s¡moylal fluld and the

s¡morlal membrane and a prlme secretory product of s¡rnorlal cells.

Êlso lt ls kroum to be sensltlve to oxygen radlcal inducad

depol¡merlzatlon and ls altered ln lnflammatory Jolnt dlsease.

The methods of approach used uære:

a. Ênalysls of the products of oxygen radlcal lnduced hyaluronlc

acld depolymerlztlon h rltro to determlne changÊ ln molecular

r.r.lelght, generatlon of reduclng c,nds and, to a tlmlted extent, change

ln blologlcal propertle¡.

b. Synorlal fluld hyaluronlc acld from lnflamed synovlal flulds rr.ras

isolated and the spectrum of molecular rrrrlghtg idËïtfled in order to

make comparisons urlth those obtalned E yJl¡4.

c. The effect of an oxygen radlcal flux on s¡norlal cells ln

culture uas determlncd and the effect on hyaluronlc acid produced

by thesc cells r¡as examlned.

Ênaþsls by gel chrornatography of oxygen radical induced

hyaluronlc acld depolymerlzation products revealed polydlsperslty tn

sizÊ. The smallest molecutes detected had a molecular rlelght of lla,¡¡hlch ulas not reduced further by exposure to a second oxygen

radical flux. Conslstentþ a relatlyely rapld transition uJàs seÊn from

large to small materlal suggestlng àn ondered element to the

breakdoum process.

Ênaþrls by analytlcal ultrccentrlfugatlon conflrmed the

decrease ln slze of hyaluronlc acld due to Ëxpoiure to oxygen

radlcals; houerer, thls procedure reyealed that those ramples of

v

louest vlscoslty dld not exhlblt the louest sedlmentatlon values, ¡

posslbþ reflectlng oxygsn radlcal lnduced repoþerlzatlon. '^v ") "'"r; ")

Oxygen radlcal lnduced hyaluronlc acld depoþcrlzatlon lg

Ëccompânlcd by the generatlon of rcduclng cndr ât mcåtured by

both colorlmctrlc assayg and cyanlde labellng, There ls, houerÊr' no

detectable lncreage ln llorgan-Elson reactlvlty thur lmplylng that

the reduclng ends released ärË prÊlent on the D-glucurøtlc acld

corriFonÊnt of the hyaluronlc acld product. Thls ls therefors evldence

that lt ls the Þ | -3 glycosldlc llnkage that ls surceptlble to clearage

by oxygur radlcal¡.

Fractlonatlon oî s¡novlal fluld hyaluronlc acid from patlents

r¡.rlth both o¡teoarthrltls and lnflanmatory arthropàthlcs revealed

uronlc acld contalnlng materlal that u.rag retarded on Sepharose '18-

EL chromatography rn contrast to non-arthrltlc cadaver subJects in

r¡hlch cäsÊ all s¡novlal fluld hyaluronlc acld uas excluded on

Sepharose 'iE-CL. ln the patlent group no materlal u,lar present

belou a molecular r.uelght, of 2 x l0a and no correlatlong could be

made betuccn the amount of uronate retarded and cllnlcal lndlces of

inflarnmatory actlvlty. Extracted cadaver synovial fluld hyaluronlc

acld ¡¡as susceptlble to o,(ygcfi radlcal lnduced depoþarlzation þrlLÊg.

0xygen radlcal exposed hyaluronlc acld had the rêmc effect on

Fc receptor blndtng a5 non-oxygen radlcal exposed hyaluronlc acid of

the same molecular r.uelght. Thus, at least ln thls context, no

alteratlon tn the propertlec of hyaluronlc acld due to oxygcn radlcal

ÊxposurÊ uas dc'rnonstrable beyond that due to decreaslng molecular

u.relght.

Êppllcatlon of än oxygÊn radlcal flux, produced by the actlon of

xanthlne oxldasc on hypoxanthlne, ls cytotoxlc to cultured borine

vf

t!,norlàl cells. Êdmlnlstratlon of catalase, but not supenoxlda Lqa(L-

dlrmutase, ls protectlye thus lndlcatlng HzO¿ ls the agent nÊc"ssäry i.;';'" ,

for toxlclty. Sublethal oxygen radlcal fluxes r¡¡lll result ln tha

cessatlon of hyaluronlc acld s¡rnthesls before the cessatlon of

s¡.rnthesls of other glucosamlne-contalnlng products. No lor¡

molecular r.lelght hyaluronlc acld pnoducts could be detected ln the

supÊrnatants of oxygen radlcal exposed cells. These flndlngs mply

that the lor¡ered molecular r.lelght hyaluronlc acld seen ln

lnflammatory Jolnt flulds ls dut to extracellular depoþerlzation.

These results prorlde furthen erldence for the role of oxygËn

radlcals as agents of hyaluronic acld depol!¡rnerlzation þ ¿Ug.

vi1.GLOSSARY

âDP - adenoslne dlphosphateÊïP - adenosine trlphosphateEPS - bathophenanthrollne sulphonateConA- concanavallnÊCTÊE - cetyl trimethyl ammonlum bromideCSb - the f¡fth component of complement (actlrated)DETÊPÊC - diethylenetrlamine pentaacetic acidEDTÊ - ethylamlnedlamlne tetnaacetic acidFe (ll) - ferrous lonFe (lll) - ferric lonFllLP - formyl methlonyl leucyl phrnylalanine6ÊG - glycosaminoglycanGSI'I - reduced glutathioneIìSSE - oxidized glutathionellÊ - hyaluronlc acidl'lBZ - peroxyl radicalKay - partltlon coefflcient descrlbed by the fdrmula:-

Kay= Pe-Po

Vt-VoHsD- molecular ureight mËäsurÊd by sadimeryrtation-dlffu:ion

t{ÊD - nicotine adenine dinucleotldeNÊDPll - nicotlne adenine dinucleotide phosphateNr - relative vlscosltyNrp - specific vlscoslty0RD - oxygeî radical depolymerlzationO2- - superoxide anion

0ll. - hydroxyl radicalOCL-- hypochlorous lonùBZ- delta singlet oxygen

ZEZ - sigma singlet oxygËn

Pl'lÊ - phytohaemagglutininPllÊ - phorbol myristate acetatePPD - purified protein derirzatirzeRPlll - tissue culture medium [lhJÊKÊTÊ and 6RÊCE, 1961]500 - superoxide dismutaseUDP-tìlcÊ - uridinedephospho-D-glucuronic acidUDP-GlcNÊc-uridinedephorpho-N-acatyglucosamineUV - ultrarioletVe - elutlon volumeVo- vold volumeVt- total rolumeXOlllK - xanthine oxldase/hypoxanthineXO/X - xanthlne oxldaselxanthine

1

C]IfiPTER I

ßTRoDUCTloll RllD RElrlEtd

¡¡

R. 6E]IERRL I]ITRODUCTIO]I

Dlarthrodal Jolnts àre speclallzed connectlve tlssue structuras

that haye dereloped to allou moblllty for an organlsm r¡lth a rlgld

skeleton. The surfaces of the artlculatfng bones àrê corered r¡lth

hyallne cartllage and the bones are unlted both by a fìbrous capsule

and by llgaments. ln addltlon, ovÊrlyins muscles and tendons con-

trlbute to the stablllty of the Jotnt. The lnner surface of the Jolnt

capsule ls llned by thc rascular s¡rnorlal membrane, thus formlng the

Jolnt spac? r¡hlch ls totalþ enclosed by synorlal membrane and

artlcular cartllage. The JoÌnt spàcË contalns a small amount of ¡otnt

fluld r¡hlch ls broardly slmllar ln composltlon to blood plasma to r¡hich

has been added hyaluronlc acld IHcEÊRTY, l9E5]. llyaluronlc acld ls a

secretory product of the synorial linlng cell. lt imparts to s¡rnorial

fluld a dlstlnctlrre high rlscosity and contrlbutes to tha biologlcal

functions of s¡rnorlal fluid. '

Joint tnflammation results in characterlstlc changes in all of the

above-mentloned anatomlcal structures, leading to damage to the

rarlous components of the Joint and ultlmateþ to Jotnt deformity

and loss of Jotnt functlon. Earþ tnflammatory changes are seen ln

the s¡rnoylal mambrana and are rÊry soon reflected ln the s¡rnorial

fluld, uhich becomes both lncreased ln volume and ls often much less

r¡lscous than normal fluld.

A r¡lde varlety of effector agents havs beer examlned to

determlne the blochemlcal mechanlgm¡ lnvolv¿d ln the loss of

structural lntegrlty of Joint tlssues. Those examined most closely

havc been ËnzymÊ systemsr prlnclpally acld proteares and neutral

proteases IEÊRRETT and 9AKLÊTVÊLÊ, l9E5l. lt ha¡ become clear from

these studles that ulthln the tnflanrmatory proccli there arÊ a

multltude of effector systems (including effectors, their precursors,

3

actlråtorr ând äntagonlsts) that can contrlbute to lnflanrmatory

Jolnt destructlon.

One such system that has come to llght ln the last l3 years

lnvolves the productlon and release of orqgen radlcals by actlrated

inflarnmatory cells. Êlthough the exlstÊnce of oxygen radlcals in

chemical systems has been appreciated for ove? 100 years, until

recently they urÊrË thought to be too actlve to play a rolc in

biologlcal processes. lt is nou clear that stlmulated phagocytes, tn

partlcular poþmorphonuclear leukocytes, undergo a "respiratory

burst" r¡hich ls characterized by lncreased oxygen uptake,

chemiluminescence, lncreased hexosË monophosphate shunt actlvlty

and production of the superoxldÊ anlmr. Superoxide anions dlsmutate

either spontaneousþ or catalysed by th¿ enzpe superoxide

dismutase to form hydrogen peroxlde. Superoxide anlons may react

uith hydrogen peroxlde to produce slnglet oxygen and the hydroxyl

rad¡cal. These radlcals ârÊ iËcreted lnto the phagosonre and are

utlllzed tn mlcroblcldal kllllng. llydrogen peroxlde is a substrate for

phagosomal myeloperoxidase r¡hich produces hypochlorlte ions, r¡hich

also hare potent oxldant actirlty. 0xygen nadicals also escape into

tha extracellular space t.lhere, in the absrnce of complete sca-

renglngr adjacent host components may be exposed to their effects.

The intra-articular space of inflamed joints prorides an important

site to examlne the effects of oxygen radicals fon reasons outlined

orlginally by tlcE0RD I l9?{]. Flrstþ, lerels of endogenous scapengers

of oxygen radlcals ln s¡,movial fluid are lou. Superoxide dlsmutase,

catalase and the glutathione/glutathione peroxidase system are

predomlnantly intracellular and thelr lerels in extracellular fluids in

general, and synorlal fluld ln pärtlcular, a?e ;,cveral orders of mag-

nltude less than lntracellular levels [l'fcCORD, 1971i ELÊKE Êt Al.,

4

lgE lJ. Secondly, hyaluronic acld hag beÊt shoum to be surceptlble to

the effects of oxygen radlcals ln ways that parallel the changer reÊ'Tì

tn lnflammatory Joint dlsea:e, that ls, lou.rered vlscoslty and a de-

crÊâse ln the areragÊ molecular u.relght.

ln order to elucldate the role of oxygÊn radlcals ln lnflam-

matory Jolnt dlsease, I harre choscn to examtne in detail the effect

of oxygen radicals on the structure, propertle¡ and cellular pro-

ductlon of hyaluronlc acld þ tzltro, and to determlne r¡hether any of

these changes can be correlated ulth the changer found ln fnflamcd

s¡,morlal fluld hyaluronlc acld. The results of thege studleg form the

basl¡ of thls theslg,

E. 0xYf¡Elt RADTCßLs ßltD J0tltT 8t0L06Y

l. lntroductlon.

0xygen has been present ln slgnlficant quantltles in the earth's

atmosphere for an estimated 2.5 billion yearl, its appearâncË cor-

responds to the ËmËrgÉnce of the flrst photosynthetlc orgánlsms

IBERKNER and HÊRSI.|ALL, l9E'f]. Since that time the atmospheric

oxygËn concentration has gradually increased and lifa forms haye

dereloped that are dependent on aeroblc metabolism. Concurnantþ

there has bcen an erolutionary pressurÊ for these organlsms toderelop effectlve defenses agalnst the toxlc lntermedlates that are

formed during the reductlon of oxygen to uatar TFRANK and Ì1ÊSSÊR0,

l980l. Host of the oxygen consumÊd by resplnlng cells ls accounted

for by the mltochondrlal cytochrome C oxldase system [ÊNTON|N| Êt

d., | 9?0]. Thls system effects the tetrayalent reductlon of oxygen

to r¡ater r¡rlthout the nrleasc of eithar superoxlde or peroxlde

IFRlD0PlCll, lg?BJ. llor¡eyen thæe are blological processÊs thatproduca superoxlde anlon, and effectlve cellulan antloxldant defense

mechanlsms hare ¿volved to protect the cell from these reduced

Ð

forms of oxygen ISIES and CÊDE|'|AS, l985l. lronlcalþ, lt r.uas the

elucldatlon of the functlon of one of these antloxidant enz¡nner,

superoxlde dlsmutase [llcCORD and FRlD0PlCtl, l96El, that led to aulde appreclatlon of the role of partlally reduced forms of oxygen (as

oxygÊn radlcals) ln blology and thelr posslble pathoge¡etlc role ln a

number of dlsearË procËlrËt.

2.

Ê frer radlcal mày bc dcfined as any ipcclel capable of lnde-

pendcnt cxlstencc that has one or morÊ unpalred clectrong

II'IÊLLIIJELL and GUTTERIDGE, l9E5J. ln general, the preic'rìcË of an

unpalred electron accountg for thc high reactlvity of radicals. As

stated, thls deftnitlon lncludes the hydrogen atom and most

transltlon metals. Radlcals may be formed f a covalent bond,

conslstlng of tuo shared electrons, ls spllt such that one electron

stays r¡ith each atom (homolytlc fission). This may be considered an

"lnltlatlon" reactlon. lndirldual radlcals mày react u.lith non-radlcals

r¡lth the formatlon of another radlcal specles (propagatlon reactlons)

or tr¡o radlcals mäy react together r¡ith the fonmation of a non-

radlcal product (tarmlnatlon reaction).

The ground state oxygèn molecule ltself mày be consldered a

dl-radlcal slnce lt has turo unpalred electrong each located ln a

different r antibondlng orbltal. Eoth of thcse electrons hare the

râmê spln quantum number. Thus, rl oxygen atte'rnpts to oxidize

another atom or molecule by acceptlng a palr of electrons from it,both neu electrons must be of parallel spin so as to enter the

racancies tn the unpaired ø orbitals [HÊLLltdELL and EUTTERIDGE,

198{a]. Thls restrictlon on oxldatlons oî oxygur dictates thato,rygen ulll tend to accept electroni onÊ at a tlme, 1.e., uÍtlvaleryrt

6

reductlon, räthËr than to be dlrectly reduced to tlater by

tetravalent reductlon.

e- e- Ê- e-02>

+211+

The unlralent reductlon of oxygen leads to the productlon of three

maJor oxygÊn-centred radlcal lntermedlates r¡hlch are dlscussed

belor*l:

a. fuFtr¡¡ldlÊulctrhJhen oxygÊn accepts a slngle electron lt ls converted to the

superoxlde anion (Br-) or its protonated form, the hydr operoxyl

radical (ll0z.). ln aqueous solution at ä pll ol 1.Ê the turo forms are

present in a ratlo of l:1 and at physiological pll oriy l% is present in

the pronated form IFEE and I/ÊLEHTII{E, 197?]. Superoxide aníon

hor¡erer disappears from aqueous solution via a dismutation

reactlon, uhlch proceeds.

H0r' * Oz + ll+ -----) llr!, + E"

This reaction may occur :pontaneousllt in r¡hich ca:,e it uill occur

most rapidly at an acidic pll (due to the higher concentratlon of

l'102.). lt may also be catalyzed by the enz¡tme superoxide dismutase

[llcC0RD and FRlDlVlEll, lgEE], u¡hich is activ¿ over a r¡ride pll range.

b. Hyircgen Peroxide

Êdditlon of tuo electrons to molecular oxygen r.rllll lead tothe formatlon of the paroxide ion (0.2-). Tha pK¡ of the reactlon

describlng the equlllbrlum betr¡een 0.2; and llr0, is so high that atphysiological pH rzlrtually all ls present as llr0r. Sereral ?nzymesr

e.9., glucose oxldase, ârË capable of the direct dir¿alent reduction of

0, independent of the dismutation reaction [FRlD0PlCll, l9?E].

?'t

7

G.@Ê mcchanlsm r.,.lhereby a three electron reductlon product of

oxygÊn could be formed from H"E" uas flrst propored by IIABER and

hJElSS [193{]. Houerer the rate conctant for thls reactlon ln

âquÊoui solutlon has been ¡horrrn to be vlrtually ze?o [llÊLLIUJELL,

tgB l]. Thls ls not the caiÊ hor¡erer lf traces of a transltlon metal

âre prÊsent, in r¡hich case a Fenton reaction ls possible, i.e.,

I Fe(lll) + Az- Fe(ll) + gz

2 Fe(ll) , HzOz >

Nett ReactionFe++ soln

0i * ltrg z -----> 0z + 0l-1. + 0l'l -catalyst

5 lron catalyzedHaber hJei¡c reaction

Sincc the sum of reactions 1 & 2 is the ¡ame ai the l'laber lrjeisE

reaction this reactlon is al¡o knor*m as the lron-catalyzed l'laber

l¡Jeiss reactlon. Thls reaction r¡lll be promoted if the iron ls äppro-

priately chelated, Ë.9., by EDTÊ []1cCORD and DÊY, l9?E; EETT!ì and

ELELÊND, l382l. 0n the other hand chelators uith à very strong

affinity for Fe(lll) do not allou Fe to shuttla betr¡een Fe(lll) and

Fa(ll) and therefore inh¡bit the reaction [6UTTER|D6E, RlCllH0llD and

I'|ÊLL||¡JELL, l9?91. Physiological chelators such as ÊTP or ÊDP can act

as iron chelators and support the superoxide-driven hydroxyl radical

production IFLOYD, I gE3] at physiological iron concentrations Ín the

absence of EDTÊ IFLITTER, R0hJLEY and l-lÊLLIIJELL, l9E3]. The

hydroxyl radical is extremely reactive and u¡ill react r.rith any

molecule uithin the immediate vlclnity of lts productlon in contrast

to 0r- and llr0, r.,.rhich are relatlvely stable and can dtÍîuse away

from thelr site of productlon IÊNEAR and NETÊ, 196?J.

I

d. Slnglat 0xFgen

Slnglet o,rygcn ls a hlghþ reactlve form of oxygen that ts

formed uhcn molecular oxygÉn absorbg sufflclÊît cne?gy to cause a

shlft of one of lts tr.rlo unpalrcd electrons of slmllar spln to an orbltal

porltlon of hlgher lin¿?gy, Thlg ls agsoclated r¡lth the lnrzerglon of spln

of one of the electrong IKASHA and ERABHÊH, 19?91. lf the excltcd

electrons form a palr then lt ls deslgnated lAgO, (therefore, by

deflnltlon not a radlcal) and lf the electront are not ln the Jäme

orbltal 3Ig0r. Slnglet oxygÊTr ha¡ been reported to be produced är âresult of the non-catalyzed dlsmutatlon reactlør and durlng the

Haber lrJelss reactlon [KELL066 and FRlDOVlEll, l9?5J. ln addltlon, lthas been reported to be produccd by the myeloperoxlde-Hr0r-hallde

system of neutrophlls [R0SEtl and KLEBAT{0FF, l9??1, uhere tt ls

bellered to be produced by the lnteractlon of the hypochlorlte lon and

H.o.'

3. Posslble Source¡ of 0xpgËn Radlcals urlthln an lnflamed Jolnt

ln addltlon to the general exposure to oÅygen radlcalc that any

cell exlgt¡ng ln Ên oxygÉTr contalnlng enrlronment must endure, (for

rerleu¡s see; llÊLLlldELL and IìUTTERIDGE, l9E5; FRlDOTlCll, l9?E), cells

and extra-cellular componentg of thç lnflamed Jolnt ärË Êxposed to

oxygÊn radlcals at å result of lnflanrmatlon. Êg a baslc for dlcusslon of

these erents, lt is approprlate to conslder brlefly lromË lnflanmatory

mechanlsms that ulll lead to .Jolnt lnflanrmatlon.

ln some sltuatlons the lncltlng agent ls obvlous, îor example, a

bacterlum ln the case of scptlc arthrltls or ân lrnmune complex ln

assoclatlon r¡ith ä strum sickrress reactlon. llorË conrmonþ, horr.rerer,

the nature of the tncltlng agent ls unknourn. Nerertheless lt ls likely

that the tratn of events slmllar to that outllned by ZVAIFLER I l9?5]

for rhcumatold arthrltls ls set ln motlon. lle proposed that the

9

lncltlng antlgen becomes locallzed ln the artlcular carlty and

stlmulates the productlon of antlbody. These combtnt tn the s¡rnorlal

fluld or the synorlal membrane and actlvate the complement

pathuay thus generatlng chemotactlc agents, partlcularþ CSb.

Pol¡nnorphonuclear neutrophlls r¡lll then accumulate, partlcularly

r¡lthln the s¡morlal fluld, and along r¡lth type Ê s¡rnoriocytes, r.uill

ingcst the fmmunË complexes. ât the time that thls theory uJàs

flrst proposed lt r¡as bellered that lt r¡as the hydrolytlc Ênzym?s

that ur?rÊ released from þsosomes durlng the act of phagocytosls

that accounted for the pnollferatlve and destructlve changes

characterlstlc of rheumatold arthrltls [ldElSSllANNr ZURIER and

H0FFSTEIN, t9?2], lt has nou becomê âpparent that oxygÊn radicals

måy also be rehasrd at thls stage [l¡JElSSllÊNN el il., 19?91 and may

contrlbute slgnlflcantly tn the genesls of these changes,

a. @en radicals from actiyated phagggl¿l!å.

Phagocytes become actlyated on contact r¡ith particulate

stlmull, such as fmmune complexÊs or bacterla that have been

opsonlzed (1.e. coated r¡lth host-derlred proteins) [R00T and

HETCÊLF, l9??]. Several soluble factors may also lead to phagocyte

actlyatlon. Êctlration tnvolves the "sr¡itchlng on" of a number of

metabollc processes, the end result belng that the partlcle ls

engulfed lnto an organelle called a phagosome. Eranules present in

the cytoplasm of the phagocyte fuse r¡lth the phagosome and

dlscharge thelr contrnts lnto lt (degranulatlon) to effect bacterial

kllllng and/or degradatlon of the lngested materlal. BÊB|0R, KIPNES

and CÊRIIUTTE I l9?5] urÊrË the first to shor¡ that lncluded ln the

procËss of actlyatlon ls the nelease of superoxlde anion. This ¡s

formed às ä componËnt of the "resplratory burst" of thr neutrophll

r¡hlch lnvolres :

r0

l) increased O" uptake l¡hlch ls lndependent of mltochondrtal

resplratlon ISEÊRRA and KÊRi|OPSKY, 19591.

2) lncreased glucose oxldatlsn vla the hexorÊ monopho:phate

shunt r¡lth the resultlng productlon of i{ÊDPH IIYER, |SLA]I and

ouÊsTEL, t9EIl.

3) Êctlratlon of an llADPll oxldasc IEÊEl0R, l9?EJ, uhlch csnverts

molecular oxygÊn to superoxldc anlon according to the reacflon ¡

2llADPtl +?Oz---) ZHÊDP+ *2E;+ jtzl

Thls tnz¡mc lg sltuated on the lnner rurface of the plasma

membrane and accepts t{ÊDPll from the cell cytoplasm; lt also accepts

oxygËn and releätÊs ruperoxlde anion from the outer ¡urface of the

plasma membranr [BÊE|0R Êt Ê!., l9E I ]. This mechanicm Érrsures thatsuperoxlde anion r¡ill be generated and deposited on the in¡ide of the

phagosome as the plasma membrane inrzaglnates durlng the cellular

ingestion of a partlcle. Houerer it is clear that not all the

superoxlde produced ln this uay ls contalned by the phagosomÊ

[lrlEl35t1ÊllN Êt å1., l9?gl, posslbþ because the enzyme ls acUvated

before the phagosorrc has been sealed, but alco becauge ënz1yt"r,C úr

contlguous årcås of mcrynbränê ls also actlrated. Also, r¡lthln the

Jolnt' lt lr clear that lrnmuna complexes mäy adhere to gtructural

componÊnts, Ç.9., collagen, leadlng to a sltuatlon of 'Trustrated

phagocytosls" r,.rhereby phagosomal cørtentr, lncludlng superoxlde

anlon, cän Êrcäpe freely lnto the extracellular mllleu ÍZVAIîLER,

lg?51. l¡Je hare recently obtained somÊ evidence that this may

occur ulth particulate stimull in the prÊsencÊ of hlghly polymerized

hyaluronlc acid [t1cHElL, Cl-l0lrJ and SKOSEY, l9E6J.

The ability of an organic system to be able to generate as

hlghþ reactlve an agent as the hydroxyl radical r¡as flrst lnferred by

EEÊUCHAHP and FRIDO]/lCll IlS?0], uho exposed the thtoether

11

methlonal to the xanthlne oxldase/xanthlne oxygÈn radlcal

generatlng rystem and detected ethylene. Slnce thls could not be

accompllshed by the actlon of elther superoxlde anlon or hydrogen

peroxlde alone, they postulated that hydroxyl nadlcals urÉrË

produced. tlydroxyl radlcals hare been detected ln suspenslons of

actlvated phagocytes, also by the pnoductlon of ethylene from

mcthlonal ITÊUBER and EÊB|0R, l9??]. Tha most lilreþ sequence of

events ls that superoxlda anions produced by the phagocytes

undergo dlsmutatlon to form Hr0, and, ln the presÊnc? of lron or

another transltion metal às â cataþst, hydroxyl radicals are formed

by the iron-catalyzed Haber l,rJelss reactlon.

Eridence for the extracellular generatlon of thls agent comes

from the experlments of SÊL|N and HcCORD [13?5] uho shor'.red that

the auto-inactlvatlon of stlmulated phagocytes can be averted by

addltion of superoxide dlsmutace, catalase or Ol1. scavångers to the

suspenslon medlum. Thls suggests that the componcnts necellaîy

for hydroxyl radlcal productlon are released lflto the extracellular

mllleu by actlrzated phagocytes r¡here they can medlate cytotoxlc

and other effectc.

Ehronlc granulomatous dl¡ease uhich lg due to a non-functlonal

NÊDPll oxldase, provldes lnslghts lnto the lmportance of orlygen

radlcals ln mlcrobial kllling. At least tuo lnherlted forms of chronic

granulomatous disease exist. Eiochemlcalþ, the tuo forms of the

disease demonstrate an inability of neutrophils to effect a

resplratory burst. Thls manifests clinlcally as persistent infections

rr¡ith sereral non-peroxlde producing bacteria, typlcally staphlococcus

aurÊus. lrJlthout the production of superoxlde anions, hydrogen

peroxlde ls not arallable to the phagocyte and both hydroxyl radlcal

productlon and hypochlorous lon productlon rla the myeloperoxldase

LZ

Ên abnormallty ln lron utlllzatlon has long been appreclated

ln actlye rheumatold arthrltlg. Thls manlfests cllnlcally ar anaemla,

r¡hlch ls assoclated r¡lth a lor¡ sËrum lron and à concomltant raised

sËrum ferrltln INASC|OL| and ELÊCKEURN, l9B5]. The mechanlsm

äppËårr to be related to an lnadequate release of lron from

retlculoendothellal cells IBENNETT, l9??].

NUIRDEH and SENÊT0R [1966] have demonstrated that iron

deposlts rrËrË a constant feature of the histopatholog¡l of the

rheumatold synovlum. Electron mlcroscopy of the s¡morial

membranes of a number of thelr patients shor¡ed that the iron r¡as

present both as ferrltin and haemosiderin IHUIRDEN, 1966].

Speculatlon exlsts as to urhether the accumulation of thls lron is due

to local factors, l.e. repeated eplsodes of mlcrobleeding consequent

upon an abnormal microvasculature, or u.lhether there is a

generallzed dlsturbance of lron storage due to chronic systmúc

dlsease [0GlLPlE-llÊRRlS and F0RI{ÊS|ER, 1 g80l.

Êftcr abrorptlon from the gut, lron, ac Fc(lll), ls transported

in the plasma bound avldly to the camler protein transferrin.

Lactoferrln, prerent ln sereral body fluids, ls a protein similar to

tranferrln ln that tr¡o moles of Fe(lll) are bound per mole of protein.

Transferrin blnds to receptors on the cell surface and can br

internallzed. lrJhen the cytoplasmic yacuole so formed becomes

acidified, the lron ls released and may become bound to yanlous

cellular constltuents such as cltrate, ÊTP and 6TP forming a small

pool of non protein-bound iron. lrJ¡thin the cell, iron is stored in the

protein ferritin, uhlch comprlses a high molecular ueight protein shell

surrounding an hydrated Fe(lll) oxlde-phosphate complex. lron

enters femltln as Fe(ll) becomlng oxldlzed by the proteln to Fc(lll)

and deposlted tn the lnterlor; conversÊly, ln order to be removed,

t3

Fc(lll) naeds to be reduced to Fe(ll), r¡hlch ls believed to be

accompllshed by blologlcal reducfng agents such .ts àscorbate,

cystetne or reduced flavlns. Lysosomal degradation of ferritin leaves

the tnsoluble product, hacmosld?rtn. FerrltÌn has a hlgh storage

capaclty for lron and ls normally only I /5 saturated.

The nett effect of thls system ls to keep the intracellular pool

of non proteln-bound lron lou. IIALLIUJELL and 6UTTERID6E Il9S{]

hare suggested that thls may be requlred in order to restrlct the

ablllty of non proteln-bound lron, (chelated ulith, say, ÃfP, 6TP or

citrate) from reacting r¡ith HrO" tn a Fenton reactlon thus leading to

the productlon of hydroxyl radicals. EUTTERID0E, RO|.dLEY and

llALLllrJELL [l9BlJ hare developed an ä5iåy for unbound iron salts,

r¡hich utlllzes the requirement of Eleomycln for unbound iron as a

catalyst in Bleomycln-induced DNÊ degradation. Their results indicate

no such "catalytic" iron in serum (consistent r¡ith the iron being

tightly bound by transferrin); houerer, a lerel of 2.8 t 1.2 ¡tmol/L

non-protein bound iron ¡alts in rheumatoid synovial fluid. These

data shou that fî llzÛz becomes arailable t¡ithin the rheumatold

Joint, then a Fenton reactlon ls possible IIìUTTERIDBE, R0ITJLEY and

HÊLLIhJELL, t 9ËZl.

Other erldence for the role of lron in joint inflanrmation is less

dlrect. The adminlstratlon of iron to a rheumatoid patient, partic-

ularly by the parenteral route ls often associated r¡ith àn exacer-

batlon of artlcular Ínflanrmation IREDDY and LElrIlS, 1969J.

Conrersely, ELÊKE gt ê1. t l983(b)l hav¿ clalmed that administration of

a chelatlng agent to gutnea plgs uith antigen-induced chronic

arthritis is associated uith a transient exacerbation of their disease

but ls follor¡ed by . long term lmprorement. The samË group IELÊKE

L4

!.1 d., l9E{l claim å corrÇlatlon betucÊn ltlrìovlal lrsn deposlts and

outcome of dlsaase ln a group of patlents urlth rheumatold arthrltls.

Studlas of the pathologl of lnon orerload s¡rndromes hare

lndlcated radlcal-medlated mechanlsms IY0UNG and âISEN, lgBZ ¡ ]IEYS

and D0RÌ1RIiDY, l9E l]. ln addltlon, ln the treatment of these

condltlons, ascorbate admlnistration r¡lll enhance the effect of the

chelating agent desfenrloxamine (consistent r¡lth its ability to

retrlere Fe(lll) from lts stonage sltes). lf glren alone, hourerer,

ascorbate ulll hare a deleterlous, sometfmes fatal, effect [NlENllUlS,

l98l J. This has been linked to its abllity to interact r¡ith iron to

produce hydroxyl radicals, leading to increased lipid peroxidation.

Finallyr þ rltro eridenct shor¡s that unsaturated lactoferrln r¡ill

lnhlblt hydroxyl radlcal formatlon IEUTTERIDGE É-êl.r l9Ell. llowerer,

lf fully saturated, lactoferrin r¡ill enhancc hydroxyl radlcal production

and lipid peroxidation IÊHERUSO and J0HN50i{, l9E l]. ln surnmary,

tnflamed synorlum ls lron laden and thls ls assoclated ulth an

lncrease ln "blologlcalþ arallable" ¡ron that could lnteract ulth HrÛ,

lÐ vlvo to produce hydroxyl radicals vla a Fenton reactlon.

c. Could stmorial ischaÉmia lead to oxrrggn radical production in

vivo?

Thls thlrd mechanlsm of oxygen radlcal productlon ls much more

speculatire and is reviewed as an arÊa u¡arranting further inrzesti-

gatlon.

Xanthine oxldase has been widely used g vltro to produce a

flux of oxygen radicals because it r¡lll react r¡ith molecular oxygen

and one of a rarlety of substrates (e.g. hypoxanthfne) to produce 0..and llr0r. Houeyer lt äppËärs that þ yiyo "xanthine oxidase" is à

xanthtnc dehydrogÊnàsË tuhlch transfers electnons to NAD+ rathen

than o¡rygËn. Attack by proteoþtlc enz4¿i or oxldation of romÊ

I5

thlol groups durlng purlflcatlon ls bellered to account for ltsconverslon to an oxldase IHALLIh,ELL and GUTTERIDGE, lgESl. Durlng

lschaemla xanthlne dehydrogÊnàsË ls also converted to an oxldase,

prÇsumably also by proteolytlc attack. llypoxla algo results ln the

accumulatlon of one of lts gubstrater, hypoxanthlne, fronr the

degradatlon of ÊTP ln hypoxlc cells. PÊRKS gt gl. [l9Bl] and others

hare suggcsted that thls may account for tlssue damage occurring

durlng reperfuslon of lschaemlc tlsgueg. As olgen suppþ ls

restorcd to hypoxlc tlssue ruperoxldc anlon may be produced by the

damaged Ënzt¡mÊ.

GRÊI|BER, RUTILI and llcC0RD Il9El] shor¡cd that reperfuslon

tnJury of cat small lntestlna, a tlssue rlch fn thls enzyme, ls

dacreased lf superoxlde dlsmutase ls added to the reperfuslon fluld.

Thls mechanlsm has dlnect releyance ln sltuations of coronary

lschaemla IBÊRDNER t$-91., l9B5l and stroke [FLÊlll1 g!j|., l9?E].

Desplte lncreased blood flou to the s¡moylum äs à result of

lnflammatlon, therc exlsts a state of relatlre hypoxla r¡lthln the

s¡rnorrlal fluld and presumabþ also r¡lthln the s¡norium ILUND-OLESEN,

l9?0¡ TREUIIAFT and l1cCÊRW, l9? I ]. ln patients ulth seyÊrr actiye

rheumatold arthrltls a synorlal fluld p0. as lou as I mm llg has been

rcported IFÊLCI'IUK, 60ETZL and KULKÊ, l9?0]. The mlcroyascular

leslons of rheumatold arthrltls assoclated t¡lth lncreased demand of

lnflamed s¡ttrovlal tls¡ue for oxygc.n and glucose may lead to areas of

unstable oxygÊn supply r¡hlch mlght ln turn lead to areas of

reperfurlon lnJury. lndced focal lnfarctlon and necroslg harc becn

ured to explaln the appearance of s¡novlal fluld detrltus knor*m as

"rlcÊ bodles" and also the cllnlcal flndlng of a lou tcmperaturc effu-

slon fluld found ln sevare rheumatold arthrltls hac becn attrlbuted

to an lschaemlc synovlum [tlßRRlg, l9B5l.

t6

lnltlal studles on the dlstrlbutlon of xanthlne oxldase hare

lndlcated that lt ls extremely wldely dlstrlbuted betueen specler

lRlCHEgf and ITJESTERFIELD, 1951, ÊL-KllÊLlDl and C||AGLRSSlAtl, 19651.

It ls not, hotlerer found in the parenchyma of many human tiscuec

(apart from llrer and small lntestlne) but ls prese'nt ln human

endothellal cells. Llttle ls knoun of the xanthlne oxldage content of

s¡rnorlal tlssues.

The potential relerance of thls mechanlsm uarrants further

study, since the concept of enztmatlc denaturatlon of an enzqe

due to hypoxlc or proteolyllc damage may apply to other

flarroproteln dehydrogenases, e.g. aldehyde dehydrogenaie, and could

lead to oxidase actlvity uith superoxlde production [LÊH05, personal

communicatlon].

1.

Êppreclation that biologlcal systems cån generate highly

reactlve partlal reductlon products of oxygên has largeþ come

about by the demonstratlon of the cellular defense mechanlsms thatprotect agalnst them. The prlmary defcnse ts provlded by " systenr

of enztrmes. Thr superoxlde dlcmutases ulll ellminate superoxlde

anlon by catalyzfîg lts csnverslon to hydrogrï pËroxlde, I'lydrogen

peroxlde ls remored by catalase r¡hlch lt r¡lll csnvert to uater plus

oxygÊn and by peroxldaces h,hlch also reduce hydrogen peroxide to

uater using a rarlety ol reductants arailable to the cell. Efflcient

remoral of one of the first tr¡o fntermediates of oxygen reduction

prerents the formation of the thlrd, the hydnoxyl radlcal, rrlhich is so

reactire that enzymatlc scayenging uould be tmpossible.

a. fupæ,ËidsiiÐs!.êEe¡.The exlstencÊ of haemocupreln, ä coppËr containlng protetn

componËnt of borlne blood r¡lth a molecular r,uelght of 5,1,000 uac

l?

dlscorered ln l93E by tlÊNN and KElLlN. tloueyer, lt urar not unfll 30

yÊärt later that l'lcC0RD and FRIDOVIEII Il98El ldentlfled lts functton.

They demonstrated that haemocupreln, nou Jrnourn ås superoxrde

dismutase, rrräs able to cataþtlcalþ remove rruperoxlde anlon. i{oother substrate has been ldentlfled for thls enzqe uhtch ls ublq-

ultous ämong aeroblc organlsms. Three dlstlnct classes of superoxide

dismutase, each r¡lth a dlfferent metal at thelr acüye site, haye

no¡¡ been rÊcognlzed [FRlD0PlEll, l9?5]. Enz]¡mes r¡tth iron or

mangänËse at their actlre sltes äpË gËnÊrally chanactenistlc for

prokaryocytes and mltochondrla. They share amino acld sequ?nce

homologlÊs suggestlng a common erolutlonany orlgin. The coppen

contalnlng Ënzymr studlad by llcCord and Frldorlch has been shor¡n

to contaln zlnc as tr.rell, although thls ls belleyed to fulfìll a stnuctural

role and ls not essentlal fon enz¡rmatlc actlrlty [EEEÌ1, RIEHÊRDS0N

and RÊJÊGOPÊLÊN, l9??1. lts amino acid sequËncË is qurte unnelated

to that of the othen tr¡o anzymes and lt is found in the cytosol of

eukaryotlc cells IKEELE, Hcc0RD and FRlD0Plcll, l9?0]. The copper

contalnlng enzyme is lnhibited by cyanide but ls stable to ethanol

and chloroform uhereas ¡¡lth the mangånese and iron containing

enzr¡mes the rererse is the case IFRlDOPlCtl, 1982].

The mechanlsm of enzqe cataþsed dismutation of superoxide

anion aPPeårs to lnvolve alternate reduction and oxldatlon of the

transltlon metal lon at the actlve Elte [RgTlLl0, BRAY and FIELDEN,

l9?21. The three dlmenslonal structure of the proteln subunits of

bovine coPPËr containlng supËroxlde dlsmutase suggests a positirely

charged track that leads to the actlye site and a slmllar mechanlsm

ls llkely for the mängänËse and lron contalnlng enzqÊs [llÊLLll¡JELL

and GUTTERIDEE, 1985J.

I8

The superoxlde dlsmutasÊs arÊ essentialþ lntracellular Ënzqes,

Lerels found ln extracellular flulds haye been ve?y lor¡ [l{cC0RD, l9?{].

llore recentþ a hlgh molecular r.r.relght form of copper contatntng

superoxlde dlsmutase has been neported II1ÊRKLUND, l gEZ]. Thls form

ls present ln the extracellular tpacÊr predomlnantþ of the lung. lt'sexact role ls yet to be elucldated.

hb. BlutaÏone f[ìlutathlone oeroxldase).

Glutathlone peroxldase detoxlfles hydrogen peroxlde through

thc oxldatlon of reduced glutathlonc accordlng to the reactlon¡

ZGSII * HzEz >

glutathloneperoxidase

The oxidlzed glutathione is then reduced by ^ second cnzqeglutathione reductase uslng l{ÊDPll (generated by the hexose mono-

phosphate shunt) as the reduclng agent. 0ther peroxides can act

as substrates for glutathlone peroxldase e.!.r lipid hydroperoxides

r¡hich r¡lll be metabollzed to hydroxy fatty aclds.

ElutathlonÊ pËroxldase (l1lÁl 85,000) lt a selÉrrlum dependcryrt

enz¡,rmÉ found ln the cytosol. lt ls effectlve at lou concentratlons of

hydrogen peroxlde and thereforu probably constltutes the cell'c flrstllne of defense agalnst thls agent. Furthcr crldence of thlc comËt

from the fact that glutathione peroxidase deficient indiriduals sulfer

a dlsease slmllar to chronlc granulomatous disease [llOLt1ES Êt Al.,

ts?01.

c. Eatalase.

Catalase catalyze¡ the dlyalent reductlon of hydrogen

peroxide to uater accordlng to the reaction:

211282 >

t9

It hac à, very hlgh P¡¡¡¡ for hydrogen peroxlde l.e. reactlon

rates approach the dfffusion controlled limit and thu¡ lt is very

difficult to saturäte. tln the other hand, houever, it is not efficic'nt

at remorzlng lorrr concentratlons of hydrogÊn peroxHe [SCl'lONBAUll and

CllÊNCE, l9?61. llost purlfied catalacec consist of four protein

subunits, each of r¡hich has a molecular u.leight of 60,000 and

contains a haem group bound to its active slte [PÊlNSllTElN BÈ 3!.,

l98 ll. Catalases are intracellular enz],l¡"n.as found in most aerobic

cells and ln animals, particularly in liver and red blood cells.

Êcatalasemic indiriduals are relatively symptom free INÊTSUDÊ et ê!.,

l9?61, thus implying that catalase is not the cell's primrary defence

against oxldant stress. Nerertheless it rr.rill complement the

glutathione/glutathione peroxidase system particularly at high levels

of hydrogen peroxide.

d.@.Ê number of other mechanisms exist uhich appear to

functlon, at least in part, as antioxidants. Eaeruloplasmin, r¡hich is

present in serum, has the ability to act as aî oxygen radical

scâvenger. 60LD5TE!N et ¡!. [19?9] have claimed that in combination

uith copper ions it has the ability to act as a superoxide dismutase,

houeyer this has been disputed [EÊtlHlSTER gt ê!., lg80l. Êscorbic

acld may act as a reductant, in r¡hlch case lt forms the semi-

dehydroascorbate radical uhich is relatiyely unreactiv¿ and usually

undengoes disproportionation to ascorbate and dehydroascorbate.

ln this manner ascorbate may react urlth superoxide anlon actíng to

detoxlfy it and its subsequent more reactire radical products.

Êscorbate is found in high concentratlons in the lens of the eye,

r¡hich has lou superoxide di¡mutase levels. lts function there is

belierzed to be an , antioxldant. Êlso uric acid, at lerels found in

20

the blood (0.12-0.'15 nrmol/l), can act âs ä porr.rerful scarenger of

stnglet oxygËn and of hydroxyl radlcals [Êl'1ES Êt å1., lgBl].

e..llcCORD t l9?{l flrst reported that boylne s¡rnorial fluld con-

tains a barely detectable amount of endogÊnous catalasr (1.e., 0.05

¡rgm/ml or less) and rery lor¡ concentratlons of endogenous

superoxlde dlsmutase (l ¡.rgm/ml). These leyels are at least Z-5

orders of magnltude less than the lntracellular concentrations of

these enzymÊi. He also shouJed that the activity of the superoxlde

dlsmutase present could be inhibited by cyanide, thus suggesting

that it t¡.ras the copper containing enzqe. The endogenous

superoxide dismutase presert ln fresh borine synovial fluid in Hc0ord's

experiments uas enough to effect only partlal protection of synovial

fluld hyaluronlc acid from än exogenously applied srrygen radical flux.

The magnitude of the flux that he applled uas equizalent to thatproduced by 5xl0ó actirated neutrophlls/ml, E yJgg. Sqne

inflammatory synovial flulds contaln a number äs much as tuentyfold

greater than thls. Extracellular flulds other than synovial fluid,

includlng human iÊrum, human cerebro-spinal fluid and borine

äquËous humour, have simllar lor¡ larels of both ënzqes. ELAKE Etå!., Il9El] examined the sgmovial ffuid from eight human patients ¡¡ith

rheumatoid arthritis and found no detectable superoxide dismutase

actituity ås åssessed by abillty to inhibit reduction of nitroblue

tetrazolium. They also found lor¡ catalase actirity (0-55 units

catalase/ml). l6ÊRl et al. [1982], using inhibition of reduction of

ferrlcytochrome C, reported that lerels of superoxide dlsmutase in

the s¡rnorial fluid of patients uith rheumatoid arthniüs r¡as

lncreased compared to that found ln ostaoarthritls. Î{erertheless

the absolute lerels that they report Ðte r,¿yeral orders of

zt

mâgnltude less than lntracellular lerels [FRlD0flCH, l9?5J. ï¡novlal

fluld ceruloplasmln lerels urere found to be lncreased ln rheumatold

arthrltlc ISCUDDER Êt Al., l9?El. llouerer, the results reported by

ELÊKE Êt gl. I l9E I I ¡¡ould suggest thls ls not associated r¡lth slgnlf-

lcant (i.e., detectable) superoxlde dlsmutase actirlty. Total

ascorbate lerels ln rheumatold s¡morlal fluld ur?re found by BLAKE ç¡!

¡!. I l9E l] to be at the lor¡ end of the normal ràng". Reduced

ascorbate hJâs not measured.

5. Targets for Oxy Radical Êttack

A large numbcr of biologlcal moleculer have been reported to be

denatured ln somÊ way by rcactlon¡ r¡lth oxygen radlcals. lt ls my

lntentlon to drau attentlon to tho¡e morÊ relevant, to inflanrmatory

Jolnt dlgeare,

a. Ec,glElE¡,

Holecules that ärË uniaturated or contain a sulphur molety

ârÊ susceptlble to free radlcal attack [PRY0R, l9?6]. Thrse include

the unsaturated amino acids, tryptophan, tyroslne, phenylalanine,

and histidine, and sulphur containing amlno acids, methionine and

cysteine. Thus, enzryes r¡hich depend on these amino acids for their

reactlvlty arÊ llkely to bc lnactirated by oxygÉTr radicals [Lltt and

ÊRI15TR0N6, l9?E; EUCIIÊNÊI{ and ARHSTRONIì, l9?El. Tu¡o examples of

this are, flrstlyr the fnactivation of serum cr I -pnoteinase inhibitor

upon ?xposurË to an oxygen radlcal flux generated by stlmulated

neutrophlls [EARP and JÊNOFF, l9?9]. lnactlyation is due to oxidation

to a sulphoxlde of the methlonlne resldue at the actlre sitc of thls

ËnzlrmÊ tnhlbltor U0l'll-130H and TRÊ/15, lg?g]. This has been proposed

äs å mechanl¡m of lung damage ln cigarette smoke-lnduced

emphysema, slnce clgarette smoke ls a potent source oî oxygen

radlcals ICÊRP and JÊtl0FF, l9?EJ. Secondly, JÊ51Ì{ [l9EZJ has

z2

rcportcd the generatlon of lgG aggrËgäter, u.rlth the propertles of

irnmune complexes, upon expoture of lgG to the myeloperoxlde-ll"O"-

hallde system. Further uork on thls phenonrenffi by tlULLlNÊX and

NULLIi{ÊX tl9E5] ha¡ shoum the formatlon of dltyrosine crosslinks (1.e.,

lnterproteln dlsulphlde bond formatlon). Thls could represent an

antlgen-lndependent means of formatlon of "lmmune complexes", r¡hich

could lead to the production of rheumatold factor.

b'@'Ê considerable literature exlsts r¡hich deccrlbes cell mutatlon

and death from ionizing radiation [I1YERS, l9E0J. These effects are

primrarily due to free radical reactlonc r'rlth DHÊ. A number of

enzlrmatlc radical generating systems hare beer tested, including

the X0/llX system [ERÊlrJll and FRlDOVlEll, l9Ell, that hare shou¡n

DNÊ strand scission. llydroxyl radical scärÊngers have been

protectlre, thus lmpllcating this agent.

E. Membrane lipids.

The unsaturated bonds of membrane cholesterol and îattyacids can readily react r¡lth free radicals leading to lipid penoxidation

IFRÊNKEL, l9Ë0]. Êfter lnltlatlon r¡lth an âppropriateþ raactlre

agent, e.g, the hydroxyl radlcal, thls process ulll become

autocatalytlc, yielding a variety of products including lipld peroxides,

llpid alcohols, aldehydes. Ên increase in the senum and synorrial fluid

content of llpid peroxidation products ln actire rheumatoid arthritis

has been reported ILUNEC É å]., l9Ell. Llpid peroxidation has been

proposed as ä cytotoxic erent Jn a number of conditions. llou.lerer,

as HÊLLlhlELL and EUTTERIDGE t1981(b)l have pointed out, lipid

peroxldatlon is a common pathr.uay in the dissolution of a cell and the

presence of llpld peroxldatlon products per se, does not Ìmpllcate llpld

peroxldation as the primany cytotoxic eyent.

23

d. Extracallular mac olecules other than hr.raluronlc acld.

The depol¡¿merlzatlon of several mäcromolecules of blologlcal

lmportance by rxporunË to a rarlety of oxygen radlcal generaÜng

systems has been demonstnated. Perlodate oxldation, for example,

ulll lnduca a loss of vlscoslty ln solutlons of proteoglycans and

glycosaminoglycans ISCOTT, TIEhJELL and SÊJDERA, l9?u], alginate

IPÊINTER and LARSEN, l9?01 and methylcellulose [5C0TT and TlGtrlELL,

l9?51. 5c0TT and PÊ6E-T||0HÊS [19?6] have presented evidencr for

the productlon of hydroxyl radlcals tn àquÊous periodate solutlons

and ln general the depolymerlzatlons descrlbed could be inhibited by

the hydroxyl radlcal scayenger, propan-l-ol.

Ê decrease in riscoslty in solutions of lsolated proteoglycan

subunit from borine nasal cartllage occurs upon exposure to the

xO/l'lx onygen radlcal generatlng system itr ritro IBREENUJÊLD, l'1BY

and LÊZÊRUS, l9?61, EÊRTOLD, hJIEEKIN and TH0NÊRD I l9E{(a)] haye

demonstrated a decneasa ln hydrodynamlc size of a glngiral

proteoglycân upon exposur? to oxygen radlcals. These authors

suggesttd that glycosaminoglycan side chains urËre cleayed from the

corÊ protein.

GREENITJÊLD and l10Y Ilg?gl have ¡hor¡n that soluble collagen

after Êxposure to an oxygen radical flux (generated by the X0/l'lx)

does not gel normally l¡hen heated. These authors and

PENKÊTÊSUERAIIAI|lÊtl and J0SEPll [13777 have postulated that this-

effect ls due to cleavage of the telopeptides of the non-helical

region. ll0NEOlSSE É e!. [19e5; l9B{], hor¡eyer, hare reported thatexPosure of soluble collagen to the X0/llX oxygen radical generating

system leads to multlple clearages of the collagen mrcrofibrrls

lncludlng the trlple hellcal reglon. lt ls not äpparent from these

24

studles, houerer, hour susceptlble fully aggregôted collagen flbres

are to oxygÊn radlcals.

e.@' l) Red cells.

Both stlmulated phagocytes [5lNCl'l0ldlTZ and gPILEERE,

l9?9; trJElSS, l9E0l and non-cellular onygên radtcal generailng

systcms IKELL0GG and FRlDOylCH l9??] ârË cäpable of medtatÍng red

blood cell lysls. The blochemlcal mechanlsms leadlng to lysls appear tobe complex and not nccessarlly appllcable to other cell types, The

fir¡t conslderatlon ls the presencÊ of oxyhaemoglobin r¡lthln the red

crll. Thls Is bellerzed to be prcsent ä! iupÊroxy-fercihaeme compound

IC0LLHAN É å1., l9?6] uhlch may autoxldtz e provtding a conflnuous

sourcÊ of superoxlde anlon r¡lthin the cell, uhlch ls ellmhated by

endogenous supÊroxlde dlsmutase unless this autoxidaflon rs

accelerated, e.9., by haemoþtlc drugs such as phenylhydrazfne or

menadlone. Reaction of superoxide anlon l¡lth haemoglobin leads totha formatlon of haemoglobin breakdor*¡n products (including

methaemoglobh) rr.rhich bind to the red cell membrane and cause

lncreased osmotic fragillty of the cell IIìOLDEERE and STERi1, 197?7.

LYNCH and FRlD0rlcll [19?s] hare demonstrated the exrstence of

anlonlc membrane channels ulthln the erythrocyte membrane by

l¡hlch superoxlde anlon can galn free acces¡ to the cell interlor and

thus react ulth oxyhaemoglobin. ln addiflon, [rlEl53 [lgE0] has

postulated that phagocytË genËrated hydrogËn peroxlde, uhlch can

dlffuse freely acro55 the cell membrane, cän react ulth methacryno-

globin to fonm a cytotoxlc complex capable of haemolysrng the

erythrocyte. Thls has stemmed from experiments that hare shou¡n

accelerated phagocyte-lnduced red cell lysls ulth exogerìous

superoxide dlsmutase, or inhlbltion of endogenous cataläse or

z5

glutathlonc peroxldase and protectlon r¡lth exogÉ.nous catalase

[lrJElSS, l9E0l. 0ther uorkers have demonstrated damage to

membrane llplds and protelns u.¡lth the use of red cell ghosts, r.uhere

the effects of oxyhaemoglobin have been elimlnated. Thus, prlmary

llpld peroxldatlon ls an alternatlre mechanlsm fon red cell þsls

[KELL066 and FRlD0PlCll, l9??; FLYNN Êtå!., l9E3; DRS and NÊlR, l9E0;

R05EÌi, EÊREER and RÊUCKI1ÊN, l9E5l.

2) 0ther cells

6lren the nonspeclflc reactirzity orlygÊn radlcals, lt ls not

surprlsing that enz',matic and chemical superoxide generating

systems have been shoun to be cytotoxic to cells in culture. Eells

lines studled hare lncluded human gllal cells IDEL HÊESTR0 .Êt ê1.,

19801, human lyrnphocytes [HcNElL, CLELÊND and EETÏS, l9B5l, lung

cells [F0X and ÊUTOR, l9?8] and human f¡broblasts [5lt1ON,5E0EGlH and

PÊTTERS0N, l9E I l. Êlso, sublethal exposure to oxygeî radicals causes

a r¡ide rängÊ of alteratlons to cell function. For example, KRÊUT and

SÊ60NE Il9El] hare demonstrated that exposurË of þphocytecultures to the X0/X system reduces the ability of these cells to

form rosettes ulth sheep red blood cells, thus indicatlng cell

membrane receptor damage. EÊTES, LOUITHER and I.|ÊNDLEY I l9E{]

hare demonstrated that cultured borine artlcular chondrocytes

respond to exposure to the X0/llX oxygen radical generating system

r¿rlth reduced proteogþcan and proteln synthesls that pensists fon

at least 5 days follouÍng exposurÊ. ln general these effects haye

been completely prerentable by catalase IEÊTES, L0lrJTtlER and

|.|ÊNDLEY, l9E{; KRÊLT and SÊBONE, l9E2¡ Sll10N, SCOGG|N and

PÊTTERS0N, l9E I ¡ DEL Ì1ÊESTRO E! E!., l9E0l, thus implicatlng hydnogen

peroxlde äs å key lntermedlate. lt ls appropnlate to note that fn

these cases the oxygen radical generating system is applied

z6

externally to the cell and the non-charged nature of hydrogen

penoxlde may allor¡ lt to penetrate hydrophoblc reglons of the cell

membrane. lndeed the cellular alteratlons and cytotoxlc ereîts may

depend upon the slte of secondany oxyg?n radlcal generatlon, lnduced

by hydrogen pËroxlde. The factors llkely to be inyolyed here are

complex, Ê.9.r ät/äilabllity of transltllon metals and/or chelator,

cellular antioxldant defence rtatus and target susceptlbllity.

Systems utilizing neutrophlls as cellulan sourcÊs of oxygen

radlcals hare also been described. Êutocytotoxicity of stimulated

neutrophils can be inhibited by superoxide dismutase [5ÊLlN and

P1cCORD, I g?sJ. This effect and the presencÊ of the myeloperoxide-

Hr0.-hallde system create difficulties tn interpreting studies of the

effects of oxygen radicals generated by stimulated neutrophils.

hJElSS É ê1. Il9ElJ demonstrated that granulocyte-induced injury

to cultured endothelial cells, as measured by ='Cr release, brås

lnhlbited by catalase but not by superoxlde dlsmutàse. Êdditional

myetoperoxidase dld not enhance slCr release and myeloperoxidase

deficlent neutrophils shor¡ed slmilar effects to normal neutrophils.

Thls suggests a myeloperoxldase tndependent, hydnogen peroxide

medlated cytotoxlclty.

There are, hourey¿rt tlssue systems uhere superoxide

dismutase äppcars protecllve. Superoxide dismutase r.uill protect

against the increased microrasculature permeability observed in the

hamster cheek pouch u¡hen lt is perfused r¡ith XE/X IDEL NÊESTR0 Êtê1., l980l. lt uill also ameliorate reperfusion injury Ín the cat small

intestlne IPÊRI(S Êt ê1., 19Ê21. Clearly, as the systems studied

become more complex, it becomes more dlfficult to fnterpret the

results of scarenger studles ln a r.lay that leads to a clear

understanding of the important inltial cellular events leading to

z7

toxlclty and to detenmtne the radlcals lnrolrad.

It also appcêrt that oxygÉTt radlcal dependent cytotoxlclty

mäy medlate such lrnmunologlcal procËJr as neutrophll-rnedlated

antlbody-dependent cytotoxlclty IE0RREGAÊRD and KRAGBALLE, l9E0J

and posslbly natural klllcr cell cytotoxlclty [SUltlÊtfTtllRÊÌ1 gt tl.,198{] although the latter contentlon ls far frorn establlshed.

c. HYRLUR0ÎilC RC|D

l. ln$sd$s:llsn

The term "synovla" uJas colned by Paracelcu¡ to dcscrlbe Jolntfluld because he llkened lts vlscous nature to that of an egg

IR0DNÊN, EENEDEK and PÊÌIETTA, l96El. The component nesponsible

for the hlgh rlscoslty of s¡.rnorlal fluld ls hyaluronlc acld. Thls

u.rldespread polysaccharlde ls classlfled as a glycosamtnogþcan

although màny of lts characterlstics mank lt as belng unlque. ln

the follotrlng pàràgràphs I shall neylerr¡ the physlology of thls

màcromolecula, ln partlcular, lts structure, llkeþ functions,

blos¡,tnthasls, turnorer and lts lntaractlons r¡¡lth cells partlcularþ ln

relatlon to lts ppÊsËncê ln the synorlal fluid of dlathrodal jotnts.

Z. Chemlcal Structure

flyaluronlc acld uac dlscovered by lleyer and Palmer ln 193{ ln

borlne vltreous humour. lt hag since becn lsolated lrs¡¡t numeroug

tlssue¡ lncludlng human rltreous, skln, umblllcal cord, blood vessel

u.ralls, and ls also prescnt ln lot¡ concentratlon ln the serum. ln-

deed lt ls a ublqultout component of extracellular matrlces. lnltlal

che.rnlcal studles by IIEYER and PÊL!'1ER [ | 938] ¡houed that lt uac a

large molecular urelght, non-rulphated, acldlc polysaccharlde. ltconglstg of equlmolar amounts of D-glucuronlc acld and l'{-acctyl-

glucosamhe (2-acetamldo-2-deoxy-D-glucose) both ln the

pyrannose form [|'1EYER, 19,1?J.

z8

Studles by JEANLOZ and FORCIIELLI Il95ll, more reccntþ cørflrmed

by LOllBÊS and HEYER Il98ll, demonstrated that the polymer l¡

unbranched and that the tuo constltuent sugari occur alternately

all the r.uay along the molecule (Ftgure l-l). The nature of the

glycosldlc llnkages urepa shoun, r¡lth the use of speclflc

hyaluronldases, to be alternatmg $ | -5 and D | -{ glycosldlc llnkages.

Eoth testlcular and bacterlal hyaluronldases cleave at the D | -{llnkages and leech hyaluronldase cleaves at the Fl-5 ltnkage

ILÊURENI, lg?0]. llyaluronic acld pol¡mers shor¡ conslderable

yarlatlon ln length (1.e., polydispersltg) and therefore total molecular

rrelght, but apart from thls no chemlcal dlfferences have been

demonstrated ämong hyaluronlc aclds from varlous blologlcal sources

including bacteria IBR|iIÊCOHEE and IdEBEER' lgE'îJ. The lsolation

mcthods that have bec'n used to separate hyaluronic acld harze

produced a product that contalns rarlable amounts of protefn.

Using mlld lsolatlon technlqucs 0fì5T0H and STÊt{lER [1950] separated

a hyaluronate-proteln conrplex from ox Jqtrvlal fluld that contalned

?5-3O% protein. Because thls sedlmcnted at a sfngle peak on

ultracentrlfugatlør they suggested that comblnatlon ulth protein

mäy ba an tntegral part of the structure oî hyaluronlc acld and

nËcetiary for somÊ of lts functlons. The flndlng of a rarlatlon ln the

amount of proteln bound r¡lth change ln pll has led CUgfÊlH [l955l

and othcrs to suggest that the bulk of thlg lnteractlon ls lonlc.

Êlternatlvely, houerer, proteln may bc adsorbed (trapped) lnto

mlcelles of the randomþ colled hyaluronlc acld molecules [0GST0Ì{ and

STÊN|ER, l95ll. Using morc rlgorous technlques hyaluronlc acld can

be lsolated r¡lth as llttle as 8.35% proteln [SlrJÊNl{, l9EEJ. lrJhether

thls proteln ls coralentþ bound l¡ unclear [LÊUREfff, l9?0; lllKUlll-

TÊKAEÊK| and P00LE, lgEl; TSIBÊÌ103, yYNl0S and KALPÊX|S, l9E6l.

Flgure l-l STRUCTURE 0F HYALUR0NIC ACID

The structure of hgoìuronic ocid showing the two

component sugars D-glucuronic ocid ond N-ocetglglucosomine

joined ot the centre bg o ßl-3 glgcosldic link0ge. At the

opposite end of each sugar Ëre the components of 0 ßl-4glgcosidic linkage. These two sugars alternote along the

whole length of this unbranched molecule,

cH2oHcooHo

HHH

\

:

.3¡

NH H3

o

HOHH

80

lndeed rcmoral of proteltt by trypsln dlgestlon does not affect the

tntrtnslc rlscoslty of hyaluronlc acld solutlons IEÊLÊZS and SUIiDELÊD'

lg5gl,

3. Ëyaluronlc Êcld fn Êou

Hyaluronlc acld has a hlgh afflnlty for uater. hlhen hydrated lt

sr¡ells to occupy a volume onÊ thousand tlme¡ greater than lts

anhydrous yolume [065T0H and STÊlllER, l95lJ. Thus a tingle molecule

of sod¡um hyaluronate (t1lrJ = 5 x l0É) r¡lll olcupy a spheroidal domaln

r¡lth a diameter of 0.5 !r or onË gram dissolrred Tn physlologlcal ¡aline

r.lill completely occupy ã liters of solrent. ThereforÊ, irr

concentratlons abore 0.35 mg/ml there r'rlll be overlapping of the

molecular domains of macromolecular hyaluronate and the formatlon

of a polymer matrlx [EÊLÊ25, l9?4J. Ê coarse neturork of branching

ftbrtllar strands has been vlsuallzed by llÊDLER Êt å!. lJgAZl using

electron mlcroscopy of freeze etch replicas of hyaluronate solutions.

lnitlal studles based on vlscotltyr sedtmentatlon and llght

scattering data tuggerted that the molecule adopts a random coll

conformation ln solution [OttsT0]l and 5TÊl{lER, l95l; LÊUREHT' 1955;

ELUNEERIì and 0tì5T0lt' 195?J. Hore recent studies uslng xray

crystallography have demonstrated that tndlvidual chains of

hyaluronlc acld adopt a stable left handed hellx - most conrmonþ

r¡ith 5 disaccharide units per complete helical turn IhJIHTER' Sl'1lT]l

and ÊRNOTT, l9?5J. The chains tend to be fully extended and are

htghly stable regandless of degree of hydration or lonic strength ln

the preparatlons used for crystallography and therefore likeþ to be

found as such in free solution. ln addltion, structural considerations

uslng the fact that the pyrannose rlngs of hyaluronic acid are both

ln the chalr form, lndlcate that lntramolecular hydrogen bondlng

äcross each glycosldlc ltnkage cân occur [ÊTK|NS, HEADER and SCOTT'

8r

l9E0¡ UJINTER, SHlTTl and ARll0TT, lg?51. Erldence for a secondary

structure stablllzed fn this manner has been found by SCOTT Êt À1.

t t Sg ll ustng ll'l n.m.r., although that study uas done using dimethyl

sulphoxide rather than r¡¡ater as the solvent. ln additlon,

lntermolecular lnteractlon betr¡een hyaluronlc acld molecules

medlated by r.rater molecules, formfng bridges betueen hyaluronate

catlon chalns, has also been postulated [SHEEllÊtl, ÊTKlllS and

NIEDUSZYNSKI, l9?51. These tnterchaÍn lnteractions, although

translent, r.,.rould tend to lead to a degree of order through chaln

packlng ulthtn the matrix, albelt temporarlly. Ê full dlscussion of

these polnts ls prorlded by l-lÊDLER and NÊPIER [19??].

{' lua. L$.ÞrLç^illlsn

Ên appreciatlon of all the functlons of hyaluronlc acld in

synorrial fluld is far from complete. The rlscous nature that hy^-

luronate lmparts to synovlal fluld has led to the suggestion that itacts äs a cartllage-cartilage lubricant, llouerer, RÊDll{, St¡JÊl{tJ and

IJEISSER [19?0] clearly demonstrated that the cartilage lubricating

actlrzity of r¡hole synovial fluid ls present in the protein fraction and

is undimlnlshed by hyalurmridase treatment. SldÊlttt and couorkers

hare isolated tr.rro high molecular r,reight glycoproteins (lubricin I and

2) r¡hlch account for thls property l5tdAl{l-l and RÊDl11, 1912i StÂlÊtll-l gt

ê!., l9E l; ShJÊNN É ê1., lg8{]. l{evertheless, a role for hyaluronic

acld in prorldlng lubrlcation betr¡een the soft tissues of the joint

has also been postulated. þ vltro eridence for this has been

prorzlded by RÊDlltl 9È å!., [19?l] r.,.rho shor¡ed that, coîìpared urith

buffer, the coefflclent of frlction betueen a flat surface and an

approprlately loaded sectlon of synovlal tlssue uas decreased

equiralently by synovlal fluld and a solutlon of hyaluronic acid alone.

82

b.@FrrlErl'lyaluronlc acld lmparts to s¡novlal fluld lts characterlstlc

riscoelastlc propertles. The rheologlcal beharrior of s¡novlal fluld and

of proteln-free sodlum hyaluronate solutions are rirtually identical

lElEBS, IIERR|LL and SlllTll,l96El and thlc beharlor may be considered

as haylng a rlscous component (d¡mamic loss module) and an elastic

component (dynamic storage module). These parameters can be

meåsured orer a nànge of strain frequencies corresponding to

rr.ralklng and runnlng. Predominantþ viscous beharior ulll result in

energy being dlssipated to surrounding tissues as heat and elastic

beharior results in energy belng stored in the deformatlon of the

hyaluronic acld molecules lfor full discussion, see EALÊZS, l9?{]. hlhen

these modull are measured n synovlal fluids from normal (young) knee

joints an elastoriscous transformation ls observed, 1.e., a

tranEformation from predominantly viscous behavior (at lorr.t straln

frequencies) to predominantly elastic beharior (at strain frequencies

corrËspondlng to a îast rr.ralk). This conrersion to elactic beharior is

less marked ln synovlal fluids from older subjects and uas not

obserred at all n synovlal fluids from patients uith osteoarthritis

and inflammatory joint disease evën at strain frequencies

corrÊspondlng to running [EÊLÊ23, l9?{]. Elastic beharior mày serve

the important function of protecting synovial membrane and

articular cartilage from mechanlcal shock and deformation (i.e.,

acting as ä shock absorber).

c. Ê Conduit for l{utrients

The second main function of synovial fluid ls that of nutri-tion. lndeed the nutrients required by the surface layers of carti-

lage are derlred from the synovlal vasculature and therefore must

pass through the synorlal fluld hyaluronate matrix. The posslble

88

influence of hyaluronate on the transfer of nutrientc has been the

subJect of con¡lderable lnrzestlgatlon. For example, llÊDLER I l9E I ]

claimed that transport of gluco¡e through an hyaluronate matrlx is

factlltated (1.e. lt occurs fa¡ter than through uater alone). This

flndlng has, hot.rÊyer, been dlsputed [NoRToN' UREÊN and Ì1ÊRoUDÊS'

lgBzl. The synoylal membrane has a rlch blood supply IR0PES and

BÊUER, lg55l and the caplltarics hare a fenestrated structure ISUTER

and HÊJN0, l9E4l. Thus, although s¡novlal linlng cells have lntercel-

lular Junctlons [tìR0Tll, t9?5], they do not form a discrete layer

IEHÊDIÊLLY, l9?EJ. The result ls that, ln certaln areas, the "ground

substancË" of the synoyial membrane has free àccess to the joint

carlty. These morphological features suggest that the extracellulan

matrlx of the synorial membrane plays an lmportant nole in

regulating the outflor¡ of fluid and soluble blood comPonents fnom the

capillaries of the synorium. lndirect eridence r¡hich supports this

concept u¡as obtained by NETTELELÊDT, SUNDELÊD and J0N550N [1965]'

uho shoued that sedimentation of plasma through a solution of

hyaluronlc acld produced an ultrafiltrate u.¡ith composition similar to

that found in normal s¡novial fluid.

5. Eios,unthesis

In recent years our understanding of the biosynthesis of hyal-

uronic acid has increased conslderably. PREIIH [1983 and l9E'l] and

PlllL|PS0N and SCHIdÊgtZ Llgïl) have shor*m that hyaluronate

synthetase is situated on the interior surface of the plasma

membrane and that as the molecule ls synthesized it extends out

lnto the extracellular späce. lt ls syntheslzed at the reducing end

by alternate transfer of the substrates UDP-GlcNÊc and UDP-G|cA to

the hyaluronate chain. The substrates can initiate chain formation

and no protein primers are rËquired. This is in contrast to the

84

synthËsis of other màmmällan glycosamlnoglycäns r¡hlch ane

syntheslzed at the 6olgl apparatus and undergo post-s¡rnthetlc

modlflcatlon. The mechanlsm of chatn tenmlnatlon for hyaluronic acld

synthesls remalns unclear.

6. Turnorer

Lrlhllst lt has been establlshed for many yeart that hyaluronic

acld ls syntheslzed ln the tissues r¡hene lt ls found I6R0SSFIELD eÈ

ê!., l95EI, knouledgÊ concÊrnlng lts turnorer has developed only

recently r¡lth the arallablllty of radiolabelled high molecular ueight

hyaluronlc acid produced in cell culture IEÊXTER, FRÊSER and CLÊRR15,

19?31. The lnltlal study in this ärea uas from ÊiTT0HÊS, FRÊSER and

I'IUIRDEN [19?3], r¡ho follor¡ed the fate of radiolabelled high molecular

u.relght hyaluronic acid after tnjectlon into rabbit knees. They

detected radloactirzity fn the synorlum, cartllage, negional lymph

nodes and blood plasma. âppearance r¡rithin regional lymph nodes r.,.ras

obserred wlthln l5 minutes of injection. Hyaluronic acid entens the

genenal clrculatlon most probably fnom the l¡mph ILAURENT and

LÊURENT, l9E I l. The normal serum concentration is of the order of

200 ng/ml ILÊURENT and TEN6ELÊ0, l9B0] and labelled hyaluronic acid

is cleared very rapidly from the blood by the liyen r¡here it is taken

up by the endothelial cells IERIKSSON Ê!å!., l9E5] by. receptor

mediated mechanism [SHEDSR0D et g]., l9E{]. There it is degraded

rapidly and completely IFRÊSER G! è!., l9E I ] in lysosomes [SHEDSROD

g! al., l984l. The major products of hyalunonic acid catabolism by

endothelial cell in culture àre acetate and lactate [S|'1EDSR0D, l9E{],

although ln the presence of hepatocytes and in ¡¡ ¡liyo expeniments,

these are further metabolised to C0, and u¡aten.

85

?. Hpaluronlc acld-Cell lnteractlons

lnltlal demonstratlon of an hyaluronlc acld-cell lnteractlon

related to the demonstratlon that a rarlety of cell types hare a

perlcellular coat of hyaluronate [60LDEERG and T00LE, l9E{] r¡hich

has the ablllty to exclude partlcles. lt has bee,n suggested that this

cell surface hyaluronate may be important in such diverse cellular

propertles as äggregätion, resistance to viral infection and

tumourlgeneslg. lJlEEKll''{ and llUlR [19?5] demonstrated thathyaluronlc acld uas able to lnhlblt the synthesis of proteoglycans by

chondnocytes ln culture and therefore may have a regulatory role in

this chondrocyte functlon. The nature of the lnteraction is very

speciflc slnce no inhibition u.raE seerì r¡ith the other

glycosaminoglycans or índeed r+rlth chondroltln, the desulphated form

of chondroitln sulphate, ulhlch dflfers chemically from hyaluronic acid

only in the orlentation of the hydroxyl group at E4 on the N-

acetylglucoEamine.

Further eridence suggesting that hyaluronic acid influerrces

cellular actirzity was advanced by T00LE ll98?l, r¡ho demonstrated

that during chick embryo derrelopment, cell morement takes place

ulthln extnacellular matrlces rich in hyaluronate. For example,

inraslon of the chick embryo cornea by mesenchymal cells colncldes

r¡lth lncreased hydration and suellÍng of the acellular stroma atr¡hich stage a maJor component of the extracellular matrix is

hyaluronate. Subsequenlly, however, migration ceases, the stroma

loses r;later, hyaluronldase actlvlty increases and hyaluronate

content of the extracellular stroma falls. lt ls at thls stage thatcellular differentlation starts. Toole suggested that hyaluronate

probably lnhlblts cell-cell lnteractlons, thereby permlttlng migration

and prollferation but delaylng the onset of diffenentiation until a

86

conrêct sequencÊ of tlssue organlzatlon could ensue. UNDERIIILL and

T00LE I l9?91 hare demonstrated a cell receptor for hyaluronate, tn

slmlan rlrus transformed 5T5 cells. Thls neceptor has a relailreþhlgh afflnlty. They càlculated that approxtmateþ 5 x l0! molecules

could blnd to each cell and that hyaluronlc acld r.uas pnobabþ bound

at sereral sltes.

The functlons that these receptorr ärÊ llkely to subserve àre)

flrstlyr the attachment of the penlcellular hyalunonate layen.

Secondly, thay are bellered to medlate the endocytosls of

hyaluronate before degradation by lysosomal hyaluronldases. Th.y

are also llkely to modulate metabollc functlons r¡ithin the cell, for

exampfe, the aforementioned lnhlbltlon of chondrocyte proteoglycan

synthesis.

llyaluronate has also been shourn to affect a number of inflam-

matory cell functions. Studies in this area have generally been con-

ducted at hyaluronate concentratlons similar to those found in the

extracellular spacË, i.Ë., l-1 mglml. Houerer more recently

lmportant obserratlons hare beer made using concentratlons llkely

to be found ln plasma 1 l,é., 200 ng/ml.

ERÊNDT I lg?1] reported that in solutlons of isola ted s¡novial

fluld hyaluronate, the lngestlon of monosodlum urate crystals by

perlpheral blood polymorphonuclear leukocytes uas inhlbited propor-

tlonal to hyaluronate concentratlon ove? the range 8,2 - l,O mglml

This effect uas diminishe d by decreasing the molecular uleight of the

hyaluronic acid to belou 106 uith mild hyaluronidase treatment.Houerer at a molecular uleight ó¡

= x 105, he noted an enhanced

uptake of crystals. F0RRESTER and EÊLÊZS [1980] have extended

these studles. Uslng ellclted mouse perltoneal macrophage

phagocytosis of latex spheres, they demonstrated that vrscous

Ðz

solutlons of hlgh molecular r.lelght hyalurontc actd (.1.6 x lgi -2.8 x

106) caused dose-dcpendent lnhlbltlon of phagocytosls orcr the

concentratlons 0.1 - ?.5 mg1ñ but, agaln, at the molecular rleight

of 9 x l0a, a paradoxlcal stlmulatlon of phago cytotls uJät teËn.

The lnhlbltlon of phagocytosls uas not present ff the cells r¡ere

pre-lncubated uith hyaluronic acid r¡hlch r¡as then removed by

r.'.lashing. Nor ulas it setn lf other large polyanions such as heparin

or chondroitln sulphate uJere substituted for the hyaluronic acid,

thus lndlcatlng that lt r¡as not a charge effect, Thry postulated

that the lnhibitory affect r¡as due to molecular lnteraction betueen

large coils of poþmeric polysaccharide prerenting the attachment

phase of phagocytosls.

Neutrophlls hare also been studled ln this regard by FORREIIER

and LdILKINSON IlgSl] r*rho reported that hyaluronic acid, ln the jame

concentratlon ränge, lnhlblted both dlnected and random locomotion

of neutrophils tn a dose-dependent and molecular-uelght dependent

manner. lnhlbltlon uJås more pronounced u.rlth higher molecular

r.uelght chemoattractant¡ such as casein, than lor small

chemoattractant peptides such as FHLP. ln addlticlrt, hyaluronic acid

inhibited the binding of the chemotactlc factor, denatured human

serum albumln, to the neutrophll surface. Thls effect úJas îeversible.

It appears, therefore, that at these concentrations the physical

effects such as molecular exclusion provide the best explanation for

the results obtained.

EÊLÊZS and DÊRZYIiKlElÅJICZ Í,19731 have examined the effects of

hyaluronlc acld on rarlous lymphocyte properties. They shor.,.red, using

a slmple migratlon inhibition assay, that concentrations of hyaluronic

acld slmllar to those found tn s¡novtal fluld tnhrbrted lymphocyte

moblllty. Ês rrlell, a number of other lymphocyte functions are

38

depressed, lncludlng the rÊsponsÊ to mltogens such as PllÊ'

l¡mphocyte dependent cytotoxlclty and experimental graft vÊrsus

host dlsease. Êgaln, each of these actlyltles ls inhlblted most

effectlrely by hlgh molecular t lelght hyaluronlc acld and uras

proportlonal to the concentratlon of hyaluronic acid.' They found

that the effect of hyaluronic acid uas cimilar to that observed by

decreaslng the cell denslty. For example, hyaluronlc acid utas more

effectlrze ln suppresslng ¡tlmulatlon of l¡mphocytes using pokerr.reed

mitogen and PPD, r¡hlch don't cautÊ signlflcant leukoagglutination'

than lt r¡as In suppresslng PllÊ lnduced stlmulatíon, urhere

leukoagglutination is ¡een. 0ther mänoÊurrcs that increase cell

dcnsity also tended to decrea¡e the effecllveness of hyaluronic acid

in lnhibltlng the stimulatlon of these cells. Other parallels betr¡een

the effect of hyaluronic acld and decreasing cell-cell distance urerË

seen such as delayed onset of DNÊ synthesis, delayed appearance of

blasts and the effects of both urere additire. They felt that in this

sltuation, hyaluronic acid may act to decrease cell-cell interaction,

although the possibility of interference r¡ith the dÌffusabillty of an

intercellular messenger such as intenleukin-Z could be an alternate

explanation.

I'IÊKÊNSSOH, llÊLLEREl,l and 1/Etl6E t 19801 have reported that lflou concentnations of hyaluronic acld (betr¡een 5-508 ¡.rgm/llten)

uJere lncubated r¡lth r¡hole blood, a stïmulation of vanious neutrophil

functlons uJås sËËn lncludlng phagocytosls, adherence, random

mlgratlon and chemllumlnescence. Ê simllar effect ls seen after a

subcutaneous inJectlon of hyaluronlc acid. They hav¿ subseque'ntly

determined that the serum cofactor necessa?y to see this

stlmulatlon ls flbronecttn [|iÊKÊNSS0N and PENGE, 1985]. Thls effect is

apparently specific for neutrophils since ÄnnSfnSSlÊDES and

89

ROBERTSON I l9B{] hare shou¡n that at slmllar lou concentratlons of

hyaluronlc acid, mltogen lnduced l¡trnphocyte stimulatlon ls lnhlblted,

although these authors separated thelr mononuclear cell fractlon

before lncubatlon r¡lth hyaluronlc acld thus not allol¡lng the

hyaluronic acld to interact ulth ierum flbronectin.

ln summaryr therefore, there arË ä number of lmport ant ways

in r¡hlch hyaluronate ls llkely to lnteract r¡lth cells. Êt concentra-

tions llkely to be found in extracellular matrixes (up to 2 mg/ml)

undegraded hyaluronate causes inhibitlon of a number of inflam-

matory cell functions including random and directed locomotion,

phagocytosls and lymphocyte stimulatlon. The llkeþ cause of theEe

effects is the entanglement of hyaluronlc acid molecules forming a

matrix. This ulll act to increase cell-cell distance and uill alter the

abllity of ranious factors to interact uith cells. Êt much lower

concentratlons, analogous to those found in plasma, however,

hyafuronlc acid, in concert ulth flbronectln, appears to cause

neutrophil stlmulatlon or priming.

E. Su¡s riêlljUslhyslur oniejgid.-Hyaluronic acid lsolated from any glven iource is polydisperse

and substantlal difference¡ ln the arerage molecular rrleight of

hyaluronic acld occur in fluids and tissues of the same species

ICLELÊND,lg?0]. Studies of the molecular ueight of nonmal synorrial

fluid hyaluronic acid are r¡ell represented by EÊLÊZS g!, al. [ 196?] r¡ho

studied pooled fluids from 268 normal adult male yolunteers. Using

analytical ultracentrifugation, lntrlnsic rziscoslty meàsurements and

light scatteringr they estimated an arerage molecular ueight of 6.5-

10.9 x 106 r¡hich is essentially fn agreement r¡ith others u.rho haye

looked at synorlal fluld obtalned from fresh cadayers UO|iNSTON, 1955;

DÊl-lL €t ê1., l985l.

40

RÊ6AN änd |-1EYER [ 1919] urere the flrst to report a decreace tn

the arerage molecular size of synorlal fluld hyaluronlc acid from

patlents ulth rheumatoid arthrltls. lt has been long appneciated

cllnlcally that r¡hole s¡rnorrial fluld from patients r¡tith lnflammatory

arthropathles has a decreased rlscosity. llouerer the hyaluronate

concentratlon is also decreased (see Table l) thus no inference

about the slze of the hyaluronic acld present can be made from this

obserratlon alone. RÊGÊN and IIEYER determlned a quoflent (log

tziscoslty/hyaluronlc acid concentnation) r¡hich uas linear oyer the

concentrations measured and uas dependent upon the degree of

polymerization. Thus rhey urere able to demonstrate that the

degree of polymerization uras decreased in a series of patients urith

rheumatoid arthritis. 0ther workers have used other techniques

including intrinsic rriscosity, anaþtical ultracentnifugation and light

scatterlng to confirm this obserration. There has, horr.reyerr been

some raniation in the estimates ol average molecular r.ueight forlnflamed synorzial fluid hyaluronic acid. These studies åre summarized

in Table l. lnitial reports suggested that the decreåse in arerage

molecular ueight of hyaluronic acid cornelated uith the degree of

lnflammatlon present [RÊ6ÊN and HEYER, l g.lg]. Thls, houerer, has

not been confirmed by other authors ISUNDELAD, lgSS¡ DÊl'lL É al.,

1ggsJ.

Êlso apparent from these studies uJas an increase in the Foly-disperslty of the hyalurontc acid present. lt is only tn relatively

recent years that reports have appeared that shor¡

chromatographic profiles of sSnovial fluid hyaluronic acid. BALAZS,

ERILLER and DENLINGER I l98l] fracflonated synovial fluid from

patlents I¡lth arthrltls on glycerallzed controlled pone glass columns.

They shoued the presence of a considerable amount of material in

Table l-l ÊPERÊEE H0LECULÊR hJElGl-lT ESTIÌ1ÊT|ONS 0F NORHÊL ÊND PÊT|-|OL0G¡CÊL SYllOt/lÊL FLUID

HORHÊLSYTIOYIÊL FLUID

HÊ Conc.

PÊTHOLOGICÊLSY]IOYIÊL FLUID

HÊ Conc Hâ l',ll¡J

0.08-0.850 decr.eased**

l.?3x106

l.?x106

1.2-1.5x106

dccr.ea¡ed**

1.8-2.15x106

1.2-2.2x1O6

5.2-6.8x106(avc=4.8)

Reference

I RÊGÊII & T4EYER [I94S]

2 SUHDBLÊD Irss3]

3 FESSLER, OGSTON & 5TÊTIIER II95{I

+ JoHHSToH 1955

5 BÊRKER et al. [1985]

6 CAST0R [rs66l

? BÊLÊZS et al. [196?]

I DÊHL et El. [19S5]

l'lethod'*mg/ml

HÊ HUJ

mg/ml

v

lv.

ll,trÊU

ÊU

TC

tv

ly, Êu, Ls

GC

r.36

?.9?

r .+- I 0.0

2.gr -3.5t

t.Ê1-?.25

r.45-5.t 2

2.85xt 06

I -,tx I 0õ

8.+-21xl0Ë

2.?5-{.0x10Ê

6.5-l tt.9xl06

6.5-?.6x106(are=?.0)

0.5-5

t.g?-8.0

1.05-1.,+5

<l.g

0.t ?-1.52

*V = viscosity; l!¡ = ¡ntrlnsic riscosity; TC = turb¡d¡ty cu-lve; ÊU = analytical ultracentrifugation; LS = light scattering;GC = gel chromatography;

'r* These methods shour decreased parameters, correlating to molecular r,reight, in the diseased fluids.rÈ)J

42

the molecular ueight fractlons 0.5 x l0l - 1.5 x 106. ln addlilon they

shoued at least trr.ro characterlstlc proflles of uronate-containlng

materlal, one unlmodal ulth sker,,.rlng into the louer molecular u.reight

fractlons and the other blmodal ulth the presÊnc; of a lot¡er molec-

ular r¡elght peak. BJELLE, ÊNDERSS0N and 6RÊNÊTH I l9E2] also

demonstrated conslderable heterogenelty ln molecular uelght profiler

of l? patlents r¡ith rarlou¡ form¡ of lnflanrmatory arthropathy.

Hore recently DÊllL Êt Ê1. I l9E5] utlllzlng a speclflc radloass ay lorhyaluronlc acld and therefore dolng away r¡lth the necesslty îor

potentlally degra datlve lsolation procedures obtained similar results.

These latter studies also failed to find a corelation betueen degree

of lnflammatlon and the amount of small moleculan r.reight hyaluronic

acid present.

D. STUDIES OII OXYGEI{ RRDICÊL I]IDUCED ]IYffLUROI{IC RCIDDEPOLY]îER I ZRTI OI{

l. lnltial Studies

Shortly after the enzqatlc depolqerization of hyaluronlc acid

r¡as desmlbed [CHÊl¡l and DUTI'||E, 1939; tlEYER, DUE0S and SllYTll,

193?l a number of reports appeared descrlblng the non-ÊnzWatic

degradatlon of hyaluronlc acld containfng substaÍrces. R0EEtrfS0H,

ROPES and BÊUER I lg39] described a spontaneour loss of viscosity n

fresh cattle ritreous upon standing. Because this tis¡ue has a high

ascorblc acid content they reasoned that this may be the "mucoþtic

agent" and subsequently shoued that both s¡novlal fluid hyaluronic

acid and extracted hyaluronic acid lost viscosity upon addition of

ascorbic acid in the presence of phosphate buffer [R0EERI50]1, ROPES

and EÊUER, 1939; l91l l. These uorkers urere quick to appreciate

that the process uJas irrerersible. lt r¡as soon shoum that a

43

numbËr of other polysaccharldes includlng pectln and starch could be

depol¡,rmerized in this fashlon [R0BERIS0N, R0PES and EÊUER, I g4l J and

that other organic reductants such ös pyrogallol and hydrogen

sulphide could replace ascorblc acid. HcCLEAIJ and IIALE I l91l]reported that ascorblc acid-induced hyaluronic acid depolymerization

uas inhibited by catala¡e, r.,.lhlch r.las important evidence thathydrogen peroxide played an intermedlate role.

The next maJor step in the elucidation of the mechanism of this

mÊans of depolymerizatlon Lras taken by SKÊllSE and SU|.iDELÊD [ 1913]

urho shor^led that ascorbic acid-lnduced depolymerization of

hyaluronlc acid dld not occur tn an anaeroblc enrlronment, but

proceeded as soon as ot(,ygen L'.,a5 added. They also reported thatthe addltlon of coppËr greatly enhancrd the effect. ln 1919 JEi|SEH

reported that ferrous ions alone or femic ions in the presence of a

reducing agent (hydrazine sulphate) could degrade hyaluronic acid.

There r+,as considerable debate in these early papers as to the role

of hydrogen penoxide and ascorblc acld. Indeed part of this confusion

has subsequently been resolred because ule nouJ realize that the

phosphate buffers used in these experiments contain iron in sufficient

quantity to cataþze a Fenton reactlon [h,0ll6 É ê!., lgBl].

Ê major concern at this tlme r¡as the unexplained degradation

of hyaluronic acid associated l¡ith various extraction procedures and

"presenratires". Êysteine, f or example, [¡Jas used to actirate papain

r¡hlch r¡as utilized in the extraction of hyaluronlc acid from various

organic sources and merthiolate LJas used as a preservatlve. Eoth

could act as reductants IPIEHÊN and RlZVl, 1959; 065T0H and

Sl-lERt1AN, 19591. Perhaps of more interest uas the depolymertzatisn

seen upon lyophlllzatlon. This uas noted by P|6HÊH and RIZVI [1S59]

to be accelerated by the presence of phosphate ioni. Subsequent

44

urork by LTJEDLOCK Êt El. I l9E5] has shor,m that the þophlllzatlon

procedure is assoclated r¡ith the appeärancÊ of radicals as detected

by electron spln reronänce spectroscopy.

Z. Radiatlon lnduced llyaluronlc Êcid Degradatlon

RÊ6ÊN gt è1. I l9{?] urere the first to report that the vlscosity

of both human synorzial fluids and solutions of hyaluronic acid urerÊ

dlmlnlshed after exposure to x-lrradlatlon. EÊLÊZS and LÊURENT

I l95l] demonstrated a slmllar effect sËcn upon Êxposure to IJV llght.

Ntlther UP nor x-lmadlatlon r¡ere assoclatcd r¡lth any change ln the

polyanlonlc charactrr of hyaluronic acid, thus indicatlng that the

carboxyllc acld groups uJËrË not affected.

EÊLÊZS Êt d. [ 1959] reported that the molecular ueight of a

hyaluronlc acid preparation, calculated from sedimentation and

diffusion constants, decreased from 80,000 to l9r7Ûg after

inradiation r¡ith U/ light, l-le has al:o reported the appearance of a

dialyzable fraction after x-imadiation suggesting a soîìeurhat smaller

end product. SUNDELÊO and EÊLÊ25 Ll96El have suggested however

that the rlscosity lor.lering effect of lor¡ doses of x-irradiation may

not be solely due to depolymerizatlon of the macrofiiolecule but also

mäy be due to changes ln shape, deformablllty or lnternal structure.

EÊLÊZS Ê! ê!. [1959] hare reported a decreas¿ n totalhexosamine and ln total glucuronic acid of hyaluronic acid exposed to

0.5-4 x l0ó rads, thus suggesting a degree of destruction of these

sugârs due to lnradiation uith electrons. Reducing substances,

meäsured by the femicyanide method, lncreased after UV ircadiation.

Degradation of hyaluronic acid by x-irradiation could be

prerented by the addition of sodlum thiosulphate, an effect belteved

to be medlated by the scavenging of hydroxyl radicals,[ERIHKHÊH glê!., I 96l l. BÊLÊZS et al. I l9E?] reported the presence of transient

45

radlcal intermedlate¡ after the pulse radlolysls of âqueous solutlon¡

of hyaluronlc acld as detected by electron pärämagnetlc rÊsonancÉ.

They postulated that the radlcal lntermedlates obs¿rv¿d could be

best explalned by cleavage of the gþcosldic linkage and alro hydrogen

abstractlon from the E, carbon of the glucuronic acld moiety r¡ith

subsequent resonänce stablllzatlon rllth C..

5. pelt¿Esrizê3lgL(ORD Reaction).

This term t¡as formulated by PlGtlÊN's group uho publtshed a

number of papers on the non-enzt¡matic depoþerizatlsn of

hyaluronic acid ln the 1960's. These uorkers uJere among the flrst topostulate that ascorbate-induced hyaluronic acid depolymerizatton

r¡as due to an oxygen radical reaction, Seyerd lines of evidence led

to thls concluslon. Flrstly, they clted the fact that x-irradiation of

aqueous solutions of hyaluronic acid leads to depolymerizallon due to

the generation of radicals (the hydroxyl radical is generated along

r¡lth hydrated electrons). The requirement for oxygeryr and the

presencÊ of a reducing agent suggested that oxygeî radlcals may be

found r¡ith the ORD reaction. Secondly a number of hydroxyl radical

scarengers including ethanol, toluene and D-glucose kJere inhibitors of

the ORD reaction. Finalþ electron spin resonance studies

demonstrated the presÉncÊ of free radicals upon the oxidation of

ascorblc acid in the presÊnce of molecular oxygen ILAEERÊRÊNTZ,

1gE.f l.

llÊRRl5, HERP and PlGllÊH [19?1] demonstrated that Fe(lI)-EDTÊ

complex could be regenerated from Fe(llI)-EDTÊ at a platinum

electrode (cathode) and thls uould depolymerize hyaluronic acid and

alginic acid. This prorzides further support for cycllc reduction r¡ith

subsequent autoxldation of Fe(ll) hyaluronic acld depolymerization

induced by ascorbate and other reductänts.

46

A. EnZy¡gtlgÀlly Produced 0xp$en Radicals and lnhtbltors

HcCORD I l9?{] uas the flrst to repont the effect of an

enzymatlc oxygen radlcal generatlng system (X0/llx) on slrrìortal flutd

and hyaluronlc acld solutlons r¡hlch demonstrated loss of riscoslty.

Uslng thls system he shor¡ed that both supcroxlde dlsmutasa and

catalast prerentad tha loss of rlscoslty at catalytlc concentratlons

(superoxlde dlsmutase I ¡.rg/ml; catalase 0.05 ¡rg/ml). lle reasoned

that slnce both supenoxlde dlsmutase and catalase alone inhiblt thls

reactlon thcn nelther the superoxlde anlon nor hydrogen peroxldà

could be the depoþmerlzlng agent but must be present

slmultaneously for the process to occur. The impllcation being thatin this sytem the actual depolymerizing species is generated

secondarily by a reactlon betureen superoxide anion and hydrogen

peroxide. The most likely candidate being the hydroxyl radical

generated by a l-laber-l,rJeiss reaction (see page 7), llannitol, a

scärengen of the hydroxyl radical, u.ras also found to prevent loss of

rlscosity in hls system. llÊLLIITJELL [19?E] conflrmed these findings

t¡ith X0/llx uslng the hyaluronlc acid solutlons and de'rnonstrated the

dependence of the observed reductlon ln vlscoslty upon the prerence

of lron salts, albelt tn trace quantlties.

5. l¡nportance of Hetal C fllnflanrmatorr.r Drugs

l'lÊLLlt¡JELL [19?E] and subsequently EETT9 and CLELÊND I lgBZ]

hare shoun that the r.lay the metal lons are chelated is of vitalimportance in both the X0IHX induced depolymerization and other

systems dependent on the oxldaflon of fercous ions (e.g., ascorbic

acid induced hyaluronic acid depolryerizatlon in phosphate buffer). lfthe chelator binds the Fe(lll) tightly (as uith the chelators Eps,

DETÊPÊC or desfemloxamfne) then thls milltates âgainst the reduction

of Fe(lll) and reductant-driren iron-dependent hydroxyl radical

4r

production i¡ inhibited. 0n the other hand iron chelators r,.rhich allor¡

iron to shuttle betr¡een Fe(lll) and Fe(ll) by transfer of electrons

r¡ill facllltate the neaction. The concentration of metal and chelator

are of importance in detenmining this effect fon example in the XO/l-lX

system. EETTS and ELELÊND I l9B2] hare shor¡n that the

antirheumatic agent D-penicillamine can act as a chelator and u.¡ill

inhlbit hyaluronic depolymerization in concentrations greater than

5m11. CIn the other hand in the ferrous ion autoxidation syslem atconcentrations of less than 5ml1 D-penicillamine is stimulatory but athlgher concentrations is inhibitory. Other antirheumatic drugs

lncludlng salicylate, chloroquine, gold sodlum aunothiomalate and

indomethacin r¡ere in inhibltory ln both systems.

6.EflÈçl-sf¡s@l'lydroxyl radlcal productlon appeärs to proceed by dlfferent

pathurays in the ,19ll1>1 system and ln systems dependent on the

autoxidatlon of femou¡ ions. The clearect evldence for this ís thatX0/l'lX induced hyaluronic acid depolymerization is inhibitable by

superoxlde dismutase but ferrous ion autoxidation and ascorbic acid

induced hyaluronic acid depolymerizatisrt are not [HOFHAltlN and

SC|-|NUT, 19Ê01. This is also the case uith ascorbate índuced Dl{Ê

depolymerization [|.40R6ÊN, C0NE and ELIìERT, l9?E]. IJIFfIERE0URI'|E

[19?9] has suggested that iron-EDTÊ catalyzed hydroxyl radical

production from hydrogen peroxide and ascorbate could proceed by ^

mechanism largely independent of superoxide, uhereas the reduction

of Fe(lll) to Fe(ll) m the XU/l'lX system is effectedby superoxide

anion [8Efi5 and CLELÊI|D lïEZl. 0ther dlfferences exist betr¡een

these tr.rro oxygen radlcal generatfng systems for example X0 can

effect the dlrzalent reductlon of oxygen as r¡ell as the unirzalent

reductlon and the relatlrc amount of superoxide anion and hydrogeïr

48

peroxide producÊd directly are dependent on pll, pO" and substrate

concentration IFRlDOl/lCl-|, I 9?CI].

?. Deqnadation of Hualuronic Êcid bu Neutroohils

----GREENhJÊLD and t10Y I l9E0] reported that phorbol myristate

acetate (Pl'1Ê) stimulated neutrophils induced a progressive decrease

in riscosity that r¡as abolished by addition of superoxide dismutase

on mannltol. These cells do not contain hyaluronidase ISODER, l9?0].

Funther uork from this group IBREENIIÊLD and H0ÊK, lg84l has

demonstrated that Fe(ll) or Fe(lll) are required, and thatstimulants other than PIlÊ (¡ncluding zymosän, Eon Ê or Fl"lLP) nequine

added hydrogen peroxide. They also demonstrated that neutnophil

lysates do not lnduce loss of rriscosity such as r¡ould be expected if

thls r¡ere due to a simple enzyme effect.

E. llolecular lrjeight Changg

PlGtlÊN and RIZPI [1959] did not detect dialysable material

aften treatment of hyaluronic acid r¡ith ranious oxidation-reduction

systems yet they did, horr.lÊve?, detect a broadening of the

hyaluronic acid peak on ultracentrifugal and electrophoretic analysis,

similar to mild treatment r¡ith hyaluronidase. These results indicate

that a true depolymerization occurs. This conclusion is alEo in

keeping uith the obserration that the changes in intrin¡ic viscosity

urere irrerersible [Pl6PlÊN, RIZll and l-l0LLEY, 1961]. EREEI{IIJÊLD and

l10Y I l9E0] uJere the first to publlsh gel chromatograms of hyalurontc

acid expoEed to any of these axyg¿n radical generating systems.

They shoued that their lnitial ppËparätion of (human umbilical cord)

hyaluronic acid rr.ras large enough to be excluded from Sepharos e 2B-

cL. Êfter elrposure to X0/llX or to Pl'lÊ stimulated neutrophils,

materlal had a reduced hydrodynamlc slze being retarded

throughout the column. Ho such information uas available until this

49

current study [tlcNElL Ê! ê!., 19ts5] for the ferrous ion autoxidation

system. EREEN|dÊLD and t10Y I I gEO] shou¡ed the presÊnce of uronic

acid containing material rzirtually to the /1 on Sepharose ZB-EL lKav

. = 0.8) corresponding to a molecular r.leight of approximatdy ?-3 x

l0s. This uas achiered using a level of super oxide anion generation

that they had calculated could be attalned r¡ithin an inflamed joínt.

They reported that part of their reactlon product became dtalyzable

but the llmiting pore size of their dlalysls tubing u¡as not given.

Studles of the generation of reducing endE from hyaluronic acSd

after exPosurÊ to oxygen radicals have provided conflicting results.

Early studies r¡ith ascorbate induced hyaluronic acid

depolymerization suggested that reducing ends kJÊre not generated

lPlcLEÊN and llÊLE, 13411. Houerer, ELELÊI|D Ë! e!. [ 1969]

subsequently detected reducing ends using sodium borohydride

labelling ln hyaluronic acid depolymerlzed by ascorbic acid. GREEHITJÊLD

and l"l0Y [1980] dld not detect reducing ends an X1/llX exposed

hyaluronic acid.

9. Eonclusion

llost inrzestlgatlons into the oïygen radical induced hyaluronic

acld depolymerizatlon systems have used change in viscosity as an

index of depolymerization -and have centered upon mechanisms of

oxygen radical production. Relatirely little information exist s as tothe molecular r.rlelght changes and conflicting information exists as tothe generation of reduclng ends ln the reaction product.

The molecular r.leight of hyaluronic acid is however altered ín

inflamed synoruial fluid hyaluronic acid. ln this investigation I have

therefore posed the follouing questions:

l. hJhat are the moleculan u.reight changes induced ln

solutions of hyaluronic acid by jn ritro oxygen nadical generating

50

systËms that resemble those in the inflamed joint? (i.e., femous ion

autoxidation, X0/HX, activated neutrophils)

2. l,rJhat are the molecular rrleights of hyaluronic acid

extracted fnom lnflamed Joints? Could any neduction in size be as a

result of oxygen radical generating systems?

5. Êre reducing ends generated by exposure of hyaluronic

acld to oxygen radicals?

1. ls the biologlcal activlty of hyaluronic acid altered if it is

depolymerized by hyaluronidase or an oxygen radical system?

5. hlhat happens to synovial cells r¡hen they are exposed tooxygen radicals and in particular to the hyaluronic acid that they

produce?

5t

C]IRPTER I I

T}IE DEPOLYIIERIZfiTIOÌI PRODUCTS OF }IYÊLURO]IIC RCID RFTER

EXP05URE T0 oDFR !Ë ytTRo

t2

R. II{TRODUETIO]I

lluch information about the molecular r.leight changes in

hyaluronic acld exposed to oxygen radical generating systems har¡e

been inferred from changes in intrinsic riscosity. Ên exception to this

is represented by the tr,.ro Sephanose ZE-CL chromatograms

published by IìREENIJÊLD and l10Y I l9B0]. Thelr data cleanly

demonstrate an incnease in polydispersity but they do not shor¡

sequential changes, nor do they demonstrate the effect of a

maximal oxygen radical flux. Sereral studies [Pl6l1ÊN and RlZPl, lg53;

EREENhJÊLD and 110Y, lgÊ01 suggest that products as ¡mall as ffono-

and disaccharides are not obtained follor.rring ü1ygen radical induced

depolymerization. Ho¡¡eyer the size of the smallest products

obtainable has not been established.

This inrestigation uas therefore undertaken to define by gel

chromatography:-

l. the sequence of molecular r¡eight changes seen after exposure

of hyaluronic acid to increasing levels of oxygen radical flux.

?,. to determine the molecular r.leight of the smallest degradation

products thus obtained.

3. to determine the effect of re-exposure to a second oxygen

radlcal flux.

1. to determine changes in an umbilical cord hyaluronate

preparation uith a characterized starting molecular r.rreight.

ln onder to obtain a r¡ide range of relative hydrodynamrc size

resolutions three gels rr.rere used:- Sepharose ZE-EL (included rolume

for polysaccharides 105 - 28x10õ), sepharose flB-EL (3x1oa - Í¡x1o6)

and Sephadex Gt00 (103 - 105) [pttRRtlÊctÊ - PR0DUET lF{F0Rt1ÊTt0]-tl.

53

Three oxygËn radical generating systems, each of possible

biological relerance uJere studied i.e. the ferrous ion autoxidation

system, the X0/HÅ system and PF1Ê stimulated neutrophils. The

generation of 0ll' is thought to proceed by different mechanisms in

each of these systems, as discuEsed in Chapter l.

llolecular rleight r¡as al¡o determined by analytical ultra-centrifugation as än independent confirmation of the results of gel

chromatography in the ferrous ion autoxidation system.

ln addition, a purified sample of hyaluronate was prepared from

the commercially arzailable human umbilical hyaluronic acid in order tobetter reflect the molecular r.leight spectrum of normal synovial flu¡d

hyaluronic acid.

E. HÊTERtÊLS Êt{D HETH0DS

Hyaluronic acid (grade lll from human umbilical cordJ and xanthine

oxidase (grade lll fnom buttermilk) r,.lere obtained from the Sigma

Ehemical Eo., 5t. Louis, l1o. Sepharose EL-28 and EL-18, Sephadex E-

100 and Sephacryl 5-100 brÊre obtained from Pharmacia, Upp:ala,

Sr¡eden. Êll other chemicals brere of the highest purity available

from Sigma or ÊJax Chemicals, Sydney, Êustralia.

I . H,ygI¿ronls--ËEld_ps¡flsÊSjsÐ

l-lyaluronic acid r¡as further purified by pasøage through a

Sephacryl 5-100 preparatire column (rrolume ZEE mls) using 0.5 l-1

sodium acetate as the buffer. Fractions r¡ith a Kar betueen 0.1

and [1.25 ulere pooled, This represented a hydrodynamic size of

betueen ?x1O6 and 5x l0s. Protein contamination uaE reduced toless than l% wlv by predigestion uith pronase, a procedure shoun

by EÊRT0LD, l,',llEEKlN and Tl-l0NÊRD [1981(b)] to have no detectable

effect on the hydrodynamic slze of hyaluronic acid.

54

2. !¡¡¿gen radical Benenation

a. Ferrous lon autoxidatlon

The reaction mixture comprised:

(l) hyaluronic acid, I mg/ml

(Z) potassium phosphate buffer, 50 mH, pll ?.4 .v

(3) ferrous sulphate at concentrations rrarying betr¡een 5

¡.r11 and 1000 ¡lt1 .-.,,.. ,o û_,,.

(1) EDTÊ in a molar ratio to ferrous ions of l:1. ./r,ì,:r,ir.:i.i .

Stock solutlons of hyaluronic acid r¡ere prepared at an initial

concentration of 4 mg/ml ln sterile distilled r¡råtep. This slightly

cloudy solution r¡as cleared by passäge thnough a l.? l.r Hillipure

filter and r¡as then stored either fnozen or at {oC until nequired.

Prlor to use, a solution of Z mg/ml hyaluronic acid in 50 ml1

phosphate buffeF urås prepaned by the addttion of an equal rolume

of 100 mll potassium phosphate buffen to the inttial 4 mg/ml

hyaluronlc acld stock solution. Stock solutions of l0 mÌl EDTÊ urËre

prepared in phosphate buffer and stock solutions sf l0 mt1 and 1 mt1

fenrous sulphate uJerÊ prepared in u.rater, The reaction rolume in

each case uJas l.? ml. Under these conditions and in the presence of

atmospheric oxygen, autoxidation of ferrous ions pnoceeds r¡ith the

production of the hydroxyl radlcal (OH.J, the agent neponted to be

directly effecting hyaluronic acid depolymerization [llÊLLlldELL, lgTE; 1,..i.r,Ì'

EETTS and ELELÊND, l9Ë81. -The neaction proceeds to completion in

less than 5 minutes IEETTS and ÊLELÊND, lgB2].

b' X.g/l-lx

For the enzymatic system the reaction mixture comp risedz

(l) hyalurontc acid, I mg/ml

(2) potasElum phosphate buffer, 50 mH, pl1 7.i

55

(3) l'lx, E mll

(1) X0 at concentrations varylng betu.reen 5xl0-3 and I lJlml

The action of X0 on HX under aerobic conditions j¡ vitro leads

to the production of superoxide anion and hydrtrgtrt peroxide in

addition to xanthine and subsequently uric acid, Trace quantities of

lron, present ln the phosphate buffer used, allor¡ further reactions

leading to the production of hydroxyl radlcals IBETTS and CLELÊi{D,

l9EZl. The reaction uaE allor¡ed to proceed to completion (greater

than tr.renty minutes) [gETfS and CLELÊND, l9E2].

c. Stimulated poll,mor@For the Pl1Ê Etimulated polymorphonuclear leucocyte system

the reaction mixture comprised:

(l) hyaluronic acid, I mg/ml

(Z) Dulbecco's phosphate buffered saline, pl'l ?.{

(3) EDTÊ, 60 ¡M('î) ferric chlonide, t 0 ¡.rH

(5) phorbol myristate acetate (Pt1Ê), 200 ng/ml

Elood. r¡as obtained by rrenipuncture from healthy labunatory

personnel and r¡as anticoagulated r¡ith EDTÊ. Polymorphonuclean

leucocytËs uJere separated by lor.'.r speed centrifugation oyer Ficoll-

llypaque ln a single step procedure as outltned by FERRÊNTE and

Tl'ltlNG I lg?E]. The cells trene then r¡ashed three times in Dulbecco's

phosphate buffered saline. lmmediately prior to the experiment,

riability uräs ässessed by tnypan blue exclusion. 0nly cell

preparations exhlbiting greater than 9596 exclusion uJere used. Pl'1Ê,

200 ng/ml, r.,las used as the stimulant and superoxide production r¡as

confirmed by obserring the neduction of nitroblue tetnazolium [SEGÊL,

l9?'11. Graded fluxes Lrere obtained by varytng the number of cells

ã6

added to the reaction mixture betr¡een 0.Exl06/ml and l2xl06/ml.

The cells bJere incubated at 37oE for 3 hrs before the cells ¡¡ere

sedimented by centrifugation for l0 minutes at 1000 g. The

vlscosity of the supernatant r¡as then measured.

5. l/iscometrl¡-

The riscosities of solutions of undegraded and degraded hyalur-

onate urere meäsured simpþ and nepnoducibly by using a I ml

tuberculin syringe and a ?0 gauge needle at ZoC according to the

method of EEfiS and ELELÊND I l gBZ]. Relatir¡e riscosities rrene

calculated by dirziding the time taken for 0.Ë ml of test soution todrain from the syrlnge by the time taken for the same volume of

buffer to drain. Specific viscosity (llsp) uas then calculated by

subtracting I from all the relatire rlscosities (1.e., subtracting the

contribution of the buffer). Changes in Nsp were expressed a¡ the

percentage ratio of degraded material to undegraded material. The

t¡iscosities of solutions of oxygen radical exposed material r¡.rere

determined at 2 hrs for solution: exposed to the ferrous ion

autoxidation or XO/l'lX systems, and at 3 hrs for solutions exposed

to the cellular system.

4. 6el chromatograp_hy-

Êliquots of O.25 ml of the reaction products urere fractionated

on Sepharose 2Ê or 4E and Sephadex6-100 (30 cm x 0.? cm). Frac-

tions of 0.? ml I¡ere eluted at ? ml/hr r¡ith 0.5 l"l sodium acetate atpll 5.8. Characterizatlon of the columns uas determined r¡ith

Dextran Blue, S3s-sulphate and glucuronic acid. ln addition, ín order

to rerify the resolution of these small columns a series of standards

uJas run through both large (i.e., 1.5 x 90 cm) and ¡mall columns and

the K¿y's of the profile peaks compared.

67

5. Uronic acid anal,usis

Uronic acid uas estimated by the meta-hydroxydiphenyl method

of ELUIIENKRÊ]-|TZ and ÊSB0E-|-|ÊNSEH I tg?3] (see Êpp endix I ) or amodlfication of the automated method of R0SENTI-IÊL, BENTLEY and

ÊLBIN I Ig?EJ.

E. Ênalytical ultracentrifugation

The sedimentation coefficients of intact and depolymerized

hyaluronlc acld samples rJepe determined on a Beckman llodel E

analytical ultracentrifuge. Samples (0.5 ml in 58 mt1 phosphate

buffer) urÊre centrifuged at 59?Ê0 r.p.m. (Zïgrg1} g åt the cell

centre) in a l2 mm single ¡ector ccll r¡ith plane quartz uindor¡s in an

ÊnD rotor at lÊoÊ for up to ? hrs, Photographr of Schlieren

patterns uJere taken at E minute interrals and images measured on

a Nikon profile analyser r¡ith a magnification cmEtant of 2.5, The

arerage molecular r.r.reights of the peak material kJere calculated by

neference to standard preparations of hyaluronic acid kindly provided

by 0r. 11. llather¡s of Ehicago and to the ralu es derived from the

data of ELELÊND and kJÊNÊ [ 19?0] as r¡ell as a standard preparation

of l-lealon analysed by BÊRT0LD, lrlElEKlN and TI'|ONÊRD Ilg8l(b)].C. RESULTS

l. Piscositu

Ê substantial decrease in llsp r¡as obs erved rJpon exposure of

hyaluronic acid to oxyge'n radicals uhether generated by the

autoxidation of femous ions, enzymailcally r¡ith >18 /t1x or by

incubation r¡ith Pl1Ê activated polymorphonuclear leucocytes (Table

?-1). The decrËases in viscosity seen uJere proportional to the

intensities of the fluxes generated r¡ithin each system. Follorr;ing

exPosure to fluxes of maxlmal lntenslty the vlscosity of hyaluronlc

58

Table 2-l 1055 0F lzlSCOSlTY INDUCED EY EÊCl'l 0XYRÊD|CÊL EENERÊTINE

5Y5TET1

Autoxidrtion of ¡(0/Hlt fictirrtrd nrutrophibfrrnou¡ lon¡

Fr lonconc. (pl"lì

H¡P(r)

t{0(U/ml)

Crll conc.(cdl,¡lml)

Ìlrp$t

lkp(ß,

r00

65

60

56

6?

65

98

0

25

50

100

e00

500

r00

{?3lt9lt1

0

0.05

0.t

t.0

t00

2Í22

t2

0

0.6 x 106

6 x l0ól? x 106

te x toG (+soD)

12 x l0ô (no PMfi)

tu x t0õ ({oc)

ln thc fmnou¡ ion ¡utoxidrtion hyeluronic acld, I rng/ml, r¡Jrs rxposrd to fennous ionsin thc pFÇsenct of 50 mH potassium phosphetc buffer, pH ?.{. EDTA was pnesent in arrtio to fcrrsuc ions of l:1. In the X0/lll{ systcm, hyaluronic rcid, I mg/ml, r,.ras

exposod to ll0 in thr prtsêncc of potessium buffcr, pH ?.{, and llX, 6 ml'l. ln thecollulan syrtcm hy*luronic ecid, I mg/rnl bns rxpo¡rd to cclls ln thc presence ofDulbçcco'¡ phosphetr buffercd srlinc to uhich hrd brrn rddrd FrClg l0 ¡rM, EDTÊ 60

¡M end PHÊ 200 ng/ml. SOD r¡rs u¡rd rt e concrntnrtion of 60 units/ml.

59

acid approached that of the buffer alone. The decrease in Nsp using

the cellular system hras partly inhibited by superoxide dismuta¡e

thus implying that superoxide ions and/or derived oxygen radicals

uJeFe, at least in part, responsible.

2. llolecular exclusion gel chromatograp[y-

a. Ss,Fh.e.Cs¡-È.!00

Samples of hyaluronic acid r¡hich had been exposed to

different oxygen radical genenating systeTns urÊre applied to

Sephadex 6- 100 columns. fln elution r¡ith 0.5 N sodium acetate,

hyaluronlc acid from preFarations r¡ith Hsp's of lO% of control, or

greater, urere largely excluded from the gel, t¡hile a major proportion

of the hyaluronic acid from lor¡er viscosity samples had a hydro-

dynamic size sufficient to be retarded by this gel. The smallest

products uJere seen at a Kay of 0.66 uhich corresponded to a

molecular r.leight of 104 (Fag. Z-ll.Ê further experlment uas performed to determine r¡hether the

Êxposure of hyaluronic acld to tr.ro sequential otlygln radical fluxes

r¡ould reduce the hydrodynamic size yet further. llyaluranic acid

r¡as first exposed to the XE/11>1 system and an aliquot r¡as then

chromatographed (Fig. Z-?). The remainder r¡as then exposed to the

ferrous ion autoxldation system using I ml1 ferrous sulphate, After

this double exposure to an ffiygsîr radical flux no material r¡as

obserred past a Rav of 0.68 on Sephadex E- 100 although the overall

change in shape of the proflle seeî on Sepharose 4E r¡ith acharacteristic shift tor¡ards the total column volu'me (V¡) tndicated

that some further depolymerization of larger molecules had

occurred.

FIEURE Z- I SEPHÊDEX Iì- I OO CHROIIÊTOERÊPHY OF HYHLURONIE ACID

EXPOSED TO TTIE FERROUS ION ÊUTOXIDÊTION SYSTEÌ1

l'lyaluronic acid, I mg/ml uràs exposed to fernous ions ata concÉrìtratlon of 50 ¡rt1 (Nsp = ?B%) and I mH (Nsp = 6g¡

in the prËsÊnce of 50 ml"l potassium phosphate buffer' pH

?.4. EDTÊ uJäs ppÊsant in a ratio to ferrous lons of l: l.

Nsp IOO5

1

3

2

I

o

5

4

3

2

t

o

5

I

3

2

I

20 25 30

20 25 30

ro

ro

t5

r5

5

ôtIox

-E

oE

zIl-gÊ,]-zll¡(Jzouô-U

IzoÈÊ,3

Ntp 78

5

Ntp I

or0

Vo5 3025

Vr20t5

FIEURE Z-2 |'|YÊLUR0NIE ÊClD EXP0SED T0 Tl¡J0 SEOUENTIÊL 0XYBEN RÊD|CÊL

FLUXES.

llyaluronic acld urat "xposed

to 0.5 unitc/ml X0 and 6 mH

l'lx in 50 mH phosphate buffer, pl'l ?.4. Ên aliquot r¡ar then

fractlonated on SapharotÊ 4E and anothcr uas frac-

tionated on Sephadex 6- 100. These profllas are rhorgn ln

the middle panel. I'lyaluronic acld contalning fractlonl rJËrÊ

then poolcd, dlalyzed agalnst urater, thctt dried in Z.e¡¡¡g

ov?? phosphorus perrtoxide beforÊ ÊxPoiure to the ferrouc

ion autoxidation using a concÇntration of ferroug ions of I

mll. Profllec obtained after fractionation of this material

on the ramÊ gels is shoun in the bottom panel.

SEPHAROSE 1B

UNE)OOSED HA

ro 15 20

HA EXPOSED TO xO/H

r0 l5 20 25 3025 30 5

x

25

THEN

30

SEPHADEX GIOO

ôaI

9x

çË

5

1

3

2

I

o

5

1

3

2

I

o5

E

zIÉ,

zt¡¡UzoUô6

IzoolD

5

I

3

2

'|

0

5

1

3

2

5

5

r0 15

HA EXPOSED TO

l0

vor5

3020 510 152025

TO FERROUS ION AUTOXIDATION

l0 t5 20

vo

30

30

5

1

3

2

I

0

5

I

3

2

xo/nx

V¡

25

V1

52520

FRACTION

6Z

b. EPierus-1EFractionation of hyaluronic acid (from human umbilical cord,

Sigma type lll) on Sephacryl s-400 provided a nominally high

moleculan rrreight starting materiall however there r¡as still same

retarded material present on Sepharose 4E chrom atography, Using

this material, a series of chromatographic profile5 LJas obtained for

graded fluxes generated by the ferrous ion autoxidation, the HOlt¡lÅ

and the stimulated polymorphonuclear leucocyte systems. They

appear as Fig. ?-3.

The effect of an increasing anygen radical flux on these high

molecular uleight hyaluronlc acld preparations in each of the axygen

radical generating systemi, rrras to sh¡ft the amount of uronic acid

containing material from the rzoid rolume (V6) to the íncluded rolume,

i.e., t0 the night. Êny shift of the chromatographic profile to the

right uJas accompanied by a decrease in Nsp in each of the

generating systems. The greatest decrease in viscosity was ¡een in

the ferrous ion autoxidation system. l-lere an Nsp of 5% of control

r¡as obtained and at this r¡isc osity no uronic acid containing material

uJas monitored at Vs. Ê significant amount of material uJas' however

retarded by the gel (i.e., Kav = 8.17 corresponding to a molecular

rrrelght of ?xl0s). Êlso, r¡ith an Nsp as lou as 3?% of control, uronic

acid containing material remained present at Vs ndicating the

presence of material r¡ith a hydrodynamic size of .l06 still preserrt in

this sample.

llyaluronic acid exposed to the stimulat ed polymorphonuclear

leucocytÊs remained predominantly excluded by the gel orer the

rånge of viscosities achiered, however there wal an increa¡e in the

amount of material seen in the included rolume and in the samples

r¡ith Nsps of 639' and 56?ã of contnol, a small amount of material r¡as

FIEURE Z-5 SEPHÊROSE '{B CIIROIIÊTOBRÊPTIY OF I'IYÊLURONIE REID EXPOSED

TO EÊEII OF TTIE OXYEEN RÊDIEÊL BENERÊTIIIE SYSTEI'IS

ln the ferrous ion autoxidation system hyaluronic acid, I

mg/ml, bJas Êxposed to ferrous ions at concentrations of

100 ¡.rH (Nrp = 5l%l and I mH (l{sp = 5%) 'n the prese'nce of

50 ml-l potassium phorphata buffer, pH ?.4. EDTÊ r¡a¡

prescnt in a ratio to ferroug ions of l:1.

ln the X0/llX system, hyaluronic acid, t mg/ml, r.tas

expored to X0 at cørcertrations of 0.1 unit/ml (Nsp = 32%)

and I unit/ml (Ì{ry = ?E%J in the prÊiÊncÊ of 6 mH llX and

50 mll potassium phosphate buffer, pll ?,1.

ln the cellular system hyaluronic acid, I mg/ml, rr.ras

exposed to concentrations of polymorphonuclear leukocytes

of 0.6 x l0ô cells/ml (Hsp = El%J and l? x IEG cellslml (Nsp =

íEZ.J in the prÊrfllce of Dulbecco's phosphate buffered saline

to r¡hich had been added FeClS l0 ¡rÌ1, EDTÊ 60 ¡.rl't and PHÊ

200 ng/ml.

IERROUS ION AUTOXIDATION

Nsp IOO

ro r5 20 25 30 5

Nsp 5l

r5 20 25 30

Nsp 5

r0vo

t5 20

HYPOXANTHINE STIMUIATED POTYMORPHONUCTEARTEUCOCYTES

Nsp 100XANTHINE OXIDASE

Nsp lO0

Nsp 32

lo l5

Nsp 20

ôt'gx

E

oE

zIÉz¡¡¡uzoue|.)

9zoÉ.l

5

1

3

2

I

0

5

a

3

2

I

0

5

a

3

2

I

o

5

I

3

2

I

o

5

1

3

2

t

o

15 20 25 30r0 t5 20 2s 30 5 to

Nsp ó3

Nsp 5ó

5

5

1

3

2

o 20 25 30 lo5 t5

ì5 20

20 25 305lo5

5

1

3

2

I

0

5

I

3

2

I

0

5

4

3

2

I

0 25Vt

ì0Vo

530

53025vr

5t0\b

15 20 25Vt

30

FRACTION NUMBER

64

present at a Kar = 0.6Ê r¡hich r¡as not present in the starting

material.

c, &PhaCeåe 2E

Ê large series of hyaluronic acid ¡amples rrler e exPc.sed to

the ferrous ion autoxidation system in order to resolve changes in

the molecular ueight spectra induced by graded oxygen radical

fluxes. Small increments in fernous ion concentnation (and thenefore

oxygen radical flux) uJere used and chromatographic profiles r¡ere

obtained on Sepharose ?Ê. Ê representatire sample of these is

shor¡n in Fig. 2-4. Examination of these profiles indicated that thene

r¡Jere tr.lo maJor peaks, onÊ tor¡ard V s and the other aPProaching

V¡. Ês degnadation Progressed there rJas a rapid transition of

material from the ]/s to the 1/1 rather than a ProgrÊs¡ion acroíí

the profile. For example in the fractionation of the sample r¡ith an

Nsp of Ê7% of control the maj ority of uronic acld containing material

r¡as excluded r¡hilst a profile obtained for a sample r¡ith an l'lsp of 8%

of control shor¡s a peak of uronic acid containing material

approachlng P1. ln fact, betr¡een the relatively intact hyaluronic

acid and the lor¡ motecular r.r.leight degradation product, intermediate

sized material does not appear as a discrete peak. lrJhen the area

under the curre in the first, middle and final thirds of chromato-

graphic profiles from Sepharo se 2E separation are plotted against %

Nsp (Fig. Z-5), the area in the middle third does not increase above

33Z. of the total area.

Using a high molecutar r.leight fraction of hyaluronic acid,

obtained from a prepar ative Sephacryl 5-400 column and exposing it

to the autoxidation system a similar computation r¡as applied to the

first, middle and final thirds of a series of Sepharose 1Ê

FIGURE 2-{ Ê SELECTION OF SEPHÊROSE 2B EHROIIÊTOERÊPIIY PROFILES

ÊFTER EXPOSURE OF I.IYÊLURONIC ÊEID TO THE FERROUS ION

ÊUTOXIDÊTIBN SYSTEH U5IN6 5}1ÊLL INERET,IE].ÍTS IN FERROUS ION

EONCENTRÊTION

The experimerrtal condltiont hrÊrÊ the same at outlined in

Flg. l. The concentrations of ferrous ions used ¡n the

profiler shoun urÊrÊ Z0 ¡.rH (Nsp = A7%1r 50 pH (t{sp = 73%'1,,

100 ¡rl1 (Nsp = 3g%), 200 ¡rM (Nsp = 237<-) and 1000 ¡-rH (Nrp =

Bß).

TtxExE

a

3

2

1

3

2

5

a

2

2

I

o

5

1

3

2

I

o

o

6

3

NrP

N ¡p87

Nrp 73

N rp39

Np23

Nrp 0

-oÉ,Þ-¡¡¡2oIJ

-ê|J{I2oc,I

5

1

to

Y6

5

2

020 a,

FRACI|Ol{v1

FrfìuRE 2-5 ÊREÊ UI-IDER THE EURI/E ÊS Ê FUNCTI0N 0F 9ã Nsp FOR

IIYÊLURONIE AEID EXPOSED TB TtIE FERROUS ION AITTOXIDÊTION

SYSTEH ÊND THEN FRÊCTIBNÊTED ON SEPI.IÊROSE 2E.CL

A large saries of hyaluronic acid samples uert ÊxPosÊd to

thc ferrous lon autoxldatlon system and then fractlonated

on Sepharose 28. ln each profila obtalncd, thc årËå undar

tha curve ln each thlnd uJât mËåsurtd. This flgure sho¡¡s

thosa mÊäsurÊments plottad ås à functlon of ß Nsp.

The area for the flrst thlrd of each pnofile (r""r)therefonÊ rËprÊsents hlgh molecular uelght material. The

arÊa under the curve for the mlddle third of each

profile (r......r) represents middle molecular u.reight material

and the last third of each proflle (o"""'o ) rÊPrÊsents small

molecular r.r.laight material.

roo

Agoo-

.C(Joooooooo

oñv

t¡¡

É,fIllrIt-et¡¡ôzÐ

¡¡¡É

q

80

70

óo

50

&

30

20

t0

t

...

aj

E

.o*, *o.acr,.^fWEIGHT MATERIAI

iAIDDTE MOTECUTAR

WEIGHT MATERIAT

¡o..

¡. 4

tr

I 1+:.j

A

I{IGH MOI.ECUTAR

aMAlERtAl

1.......0 30 20 t0 o100 90 80 70 óo 50 10

SPECIFIC vlScOSlTY (Ntp Z )

67

fractionationg. Êgain no more than 13% of the uronic acid appeared

in the transition portion of the profiles (Fig. Z-Ê).

5. Ênalptical ultracentrifugation

Sedimentation of hyaluronic acid and its oxygen radical induced

degradation products uas achieved on the analytical

ultracentrifuge. The reciprocals of the sedimentation coefficients (5o-¡) and the extrapolations to derived M5g rralues [ÊLELÊl'lD and l,rJÊl-lG,

l9?0; EÊRTOLD, I,rJIEEKIN and TH0NÊRD, 1981(b)l are shor¡n in Tabte Z-2,

Thnee samples of hyaluronic acid hJere used, each of a diÍferent

starting molecular r.leight. They were all exposed to the ferous ion

autoxidation system. Êgain a louer molecular r.leight limit of

approximately lBa uas obtalned. The calculated moiecular uleightralues (t1sD) folloued a predictable overall drop r¡ith decreased

rziscosity. Hor¡erer in Experiment I the sample exposed to ?E yl1

ferrous ions (Nsp = E7% of control) had an I'l5g of 60,000, u.rhereas ín

the sample exposed to 100 ¡.rl'1 ferrous ions (Nsp = 3g% of control)the l15g ralue rr.ras Ê8,000. Similarly in ExperimEnt 5, u.rhere more

samples uJere analyzed, the lor¡est H5g rralue calculated r¡as 9,000 in

the sample that had been exposed to 125 ¡tP1 ferrous ions, r¡hereas

samples exposed to 150 and 2OO ¡tYl ferrous ions shor¡ed t15g yalues

to 39'000 and 't2'000 respectively prlor to a further drop occurring.

D. Dt5CU55tOil

The commercial hyaluronic acid used in this study uJas supplied

as a lyophilysate and Sepharose 4E chromatography shous this tobe polydlsperse as regards hydrodynamic size. Attempts to obtain ahigh molecular ueight hyaluronic acid preparation by fractionaflon on

Sephacryl 5-400 chromatography r¡ith subsequent dialysis and

lyophlllzatlon of the fraction¡ reduced the amount of material

FIGURE 2-6 AREÊ UNDER TtlE CURPE ÊS R FUNCTION 0F ß Nsp F0R PURIFIED

IIYÊLUROT{IC ÊCID EXPOSED TO TTIE FERROUS ION ÊLTTOXIDÊTION

5Y5TET1 TIIEN FRHCTIONÊTED ON SEPHÊROSE {B-EL

PERCENTAGE

OF

PROFILE

70

30

ro

óoLOW MOTECUTAR

WEIGHÍ MATÊRIAL

MIDDTE MOTECUTARWEIGHT MAIERIATA

.^

HIGH MOTECUIAR

WEIGHÍ MAÍERIAT

0 roo 90 80 70 óo 50 10 30 20 loNtP

0

69

Teblc 2-2 MOLECULÊR IJEIGHT ESTIMÊT|0]{ 0F HYÊLUP0HIC fiClD E1{P05ED T0 THE

FERROUS IOÌI ÊUTOIIIDÊTION SYSÍEM USITIG THE AIIÊLYTICAL ULTRÊCEI.ÍTRIFUÊE

E*P

?r cFFotJt ton

concrntretionmicnomoler

Sprcificvi=cosil,yHrp (f,)

Rrciprocel of thrccdirnçfitrtion

corff¡c¡rnt tt z¿noconcrntration (l lïo't)

C¡lculrtrdmoleculerrlcight(F1sD)

0t02g

100

020

t00I 000

t00918?ãg

100E9+tt0

r00E6?2565216352+t5

0.0600.0150.1858.118

0.29?0.50e0.+ee0.585

0.{280.{l ?0.1950.5180.89?t.t060.Ët00.5890.90Ê

l,+50,808g+,00060,u00EE,gOE

?

3

e5e,0ooe15.00095,000,15,0Ð0

02550?5

r00125t50200100

9e,00098,00065,00058.00015,0009,000

59,000,+2,000

15,000

For rach cxprnimcnt, thr rcection concçntmtion of hyelunonic rcid r¡¡s I mg/rnl,,EDTH wrs prusent in r netio to fcrnou¡ ions of l:l and the buffcn used r¡as 50 mMphosphate buffcn. Thr srdlmentation coefficicnts breFe then determined on concen-tnetions of hyeluronic ecid of I gm/ml, 0.5 mg/ml and 0.25 mg/ml, although et theletter concentnation in somG of the moFc dcpolymenl=ed samples no pcak r,.ras

ldentlfled rnd thrnrforc no rrlur fon thr srdlmrntrtion constent could be uscd.Sincr thr hyrlunonic rcld concrntretlon¡ u¡rd r,¡rnc nrletirrly lou.r, e linr¡nnrgnrssion of thr reciprocel of thr ¡rdimrntetion coefficient could br used todr[rrminr So'l [BÊnruUD, UJIEBKIII rnd Ttl0NÊRD, lg8{].

?0

retarded on Sepharose 4E chromatography but did not eliminate itentirely. This preparation uräs performed before the publication of

the r¡ork of hIEDL0CK et al. I I gE3]. They shor¡ed that lyophyltzation

' ls accompanled by radical generation and depolymerization.

Nerzertheless a large amount of material in the commencial pnoduut

that uas rrtarded by Sephanose {E-CL r¡as able to be eliminated.

Êll subsequent prËpåråtlons uJere made r¡lthout lyophylization.

The controllcd exposune of hyaluronlc acld to oxygen nadicals ls

åccompanlad by a predicted decreå,se in riscoslty and the

appËärance of a polydlspense population of breakdor¡n pnoducts. This

phenomenon is noted r¡hether the nadicals ane genenated by the

autoxldatlon of fernous ions, the action of xanthine oxidase on

hypoxanthine or by stimulated peripheral blood polymorphonuclear

leucocytes. EREENhJÊLD and HOY I l9E0] hare reported tr.ro

Sepharose ZB-EL chromatognaphic profiles r¡hich adequately

demonstrate ä genenalized reduction of hydrodynamic size of

hyaluronic acid exposed to the XO/l'lX system and to PHÊ stimulated

neutrophils. lrJe hare endearouned to define the pnogpess of oxygen

radlcal lnduced degradation by lncreasing the numben of profiles

analyzed. 0f the oxygen radical produclng systems studied, the

ferrous ion autoxidation system could ba most neadily manipulated

to proride a comprehensire serias of dagraded hyaluronic acid

samples. Less extensive studies using the X0/l-lX system produced

congruent findlngs compatible r¡ith a similar fonm of hydroxyl nadlcal

attack on hyalunonic acld in both systems. The nelatirely modest

degradatlon achiered by Pl'1Ê stimulated polymorphonculear leuco-

cytes could be partly explained by: (a) lessen fluxes of superoxide

7L

(not quantitated) and/or (b) less uniform production of oxygen

nadicals by the cellular system.

The effects of oxygen radicals on hyaluronic acid observed Ey

, concurrently monitoring several parameters (i.e., viscostty, change in

hydrodynamic profile sn several gels and l15g) indicated that change

in riscosity is a sensltive indlcator of early damage to hyaluronic

acid. For example, a sample r¡ith an Hsp = 3E% of control (Fig. 2-3)'still contains a signfficant amount of material that is excluded on

Sepharose 4E chromatography. Ên explanation of this may be

erident from the concept expounded by BfrLAZï Ug?4l and others

that hyaluronic acid in solution forms an interlocking molecular

matrix. lf sereral large chains of hyaluronic acid in this molecular

matrix are depolymerized, the matrix r¡ould become less ordered,

despite the persistence of some intact molecules. lndeed

fractionation of hyaluronic acid on Sepharose 1B after expasure to a

mild oxygen radical flux shor¡ed some excluded material, rep?e:,enting

the intact chains together r¡ith sorre material in the included volume

rÊpresÊnting the depolymerized chains. Yet as a r¡hole the ¡olution

r¡as dramatically altered in vtscosity, ln turn theEe changes are

likely to resr-tlt in alterations in physiological properties,

ln the pnesent study the smallest degradation products obtained

had a hydrodynamic size of the order of l0a. This lor¡er limit r¡as

not shifted by subsequent exposure to a second orlygln radical flux.

Ênalysis of the large Eeries of chnomatographic profiles

obtained r¡ith the ferrous ion autoxidatlon system suggested thatthe depolymerization of hyaluronic acid h radicals may be

at least partly ondered uith rapid progression from lange to small

72

material, such ås måy occun uith a chain nraction of clearrage

erents.

The obserred increase in sedimentation coefficient of louer vis-

cosity hyaluronic acid samples on analytical ultracentrifugation may

reflect some form of molecular interaction (aggregation) or

conformatlonal change due to crossllnklng. BALÊZS g! al. [ 196?] hare

postulated that, during pulse nadiolysis of hyalunonic acid, in addition

to direct oxidatire clearrage of the glycosidic linkage betr¡een

glucunonlc acid and N-acetylglucosamlne, uronate fnee radlcals may

be formrd by ellmlnatlon of the hydnogen at C5. This nadlcal r¡ould

then be nesonancÊ stabilized by lnteraction r¡ith the canboxyl group.

Thesr radicals could interact to form cross-llnked hyalunonate

degradation products. lndeed tha identiflcation of these branched

products in synorial fluid could be used to more directly implicate

radical-induced damage to macromolecules in inflamed joints,

Eertainly such structural alterations could have important conse-

quencÊs in terms of the altered physiological properties of the

hyaluronate matrix.

E. SUllllÊRY

Preparatlre chromatographic fractions of human umbilical cord

hyaluronic acid of a molecular r.r.reight of l0ô uJerÊ subjected to

graded oxygÊn radical fluxes produced by (a) the autoxidation of

ferrous ions, (b) the action of xanthine oxldase on hypoxanthÈe and

(c) by perlpheral blood polymorphonuclear leucocytes r¡hich had been

stimulated by PllÊ. Ênalysls by gel chromatography of the products

obtained r¡ith each of the oxygÊn nadlcal generating systems

rerealed polydispersity in size. The smallest molecules detected had

a molecular r.,.relght of 104. This limiting slze r¡as not reduced further

?8

by exposure to a second oxygen radical flux. The relatir¡e

proportions of large, medium and small degradation products urere

establiEhed for rarious lerels of oxygen radical flux. Consistently a

relatlr¿ely rapid transition from large to small material uJas seËìì on

Sepharose ZB chromatognaphy suggesting ån ordered element to the

breakdoun process, Êlthough the decrease ln molecular r.leight,

after oxygen nadical exposunÊ, uJas conflrmed by anaþtical

ultracentrifugation, thls pnocedune nerealed that samples of lor¡er

rziscosity did not necÊssanily exhibit lor¡er sedlmentation ralues,

posslbly reflecting oxygen radical-induced repolymerization. Ey

altering slze and possibly the conformational characteristics of

hyaluronic acid, oxygen radlcal exposure may be expected to alter

its biological propertles.

74

cllÊPTER lllCllRllGES lH II0LECULÊR hlElEllT SPECTRß OF SYll0V,ÃL FLUID

TIYÊLURO]IIC RCID SEEII I]I II{FLÊH}IßTORY JOI]IT DISERSE

76

R. lltrRoDucil0l{.

The changes in several parameters of synovial fluid hyaluronic

acid in resPonse to inflammation have been r¡ell char acterized.

These include changes in concentration, average molecular ueight

and ¡ntrinsic t¡iscosity and are summarized in Table 1-l' page 11'

llouerer, sparse and conflicting lnformation exists concerning the

effect of inflammation upon the polydispersity of molecular r.leights

of hyaluronic acld fnom inflamed synovlal fluids. One means of

lmptlcaflng or discounting the possibillty of oxygen radical induced

hyaluronic acid depolymerization ln inflamed joints is compari=on of

the moleculan r.rleight profiler of hyaluronic acid isolated from

lnflammatory synovlal effuslons to profilei leeî after exposure of

hyaluronic acid to radlcal fluxeE 'E yJ3,. The study descrihed in

Chapter ll prorzided a large range of profiles de+nonstrating the

changes noted in the molecular r.leight of hyaluronic acid as the

size of the oxygen radical flux uas increased. llowever no Euitable

data obtained from inflamed and non-inflamed synovial fluid

hyaluronic acid rJas arailable in the literature to enable a ralid

comparison.

CREENITJÊLD [ 19Ê0] has reported in abstract that in 2l patients

r¡ith rheumatotd arthrltls none of the synovial fluid hyalunonic acid

r¡as of sufflciently small hydrodynamic size to be retarded on

Sepharose ZB-EL chromatography. ln anothen study by tsÊLÊZS'

ERILLER and DENLINEER ItSel] r¡ho analysed molecular r.r.reight

profiles of synorial fluid hyaluronic acid fnom l5 horses r¡ith

traumatic arthritls and Z0 humans r¡.¡ith osteoanthritis' a lange

amount of hyalunonic acid r¡as identified r¡ithin the included

rolume of glycer allzed controlled pore glass columns calibrated

betrr.reen Ëx l0E and 5x 104. Their re¡ults therefore indicate the

presence of material r¡hich t¡ould be included upon Sepharose 2Ê

and Sepharose {E-ÊL chromatography. llore recently EJELLE,

z6

ÊNDERSSON and GRÊNÊTl'l t l3S2l have reported gel profiles of l?

"arthritic" patients, predominantly rheumatoid arthnitis, using a

composite gel column packed r¿¡ith consecutirre portions of

Sephanose ZE, Sepharose EE and agarose 0.58, cross-linked r*¡ith

dirzinyl-sulphone. This composite gel required the tneatment of

their samples r¡ith 2.5 11 sodium hydroxid. IPH < 12) and thein

column r¡as eluted using a buffer r¡ith a pl-l I l.E. This strong

alkaline treatment leayes this technique open to cniticism because

of the possibility of alkaline hydrolysis of the hyalunonic acid.

Houerer the authors claim this r¡ould har¡e been minimal due to

the napidity of analysis (< l2 hrs)' yet no internal standard r¡as

shoun to demonstrate this. Nerertheless they shor¡ed a r¡ide

rzariety of molecular ueight distributions uith material still

present at a Rav corresponding to a molecular r.leigh t of ? x l}aon their column. ¡

None of the above studies have included a control group of

"normal" fluids. The results of these latter tuo studies

suggested that a portion of hyaluronic acid from inflamed synorrial

flulds r¡oulå elute ¡¡ithin the included r¡olume aften fractionation

on Sepharose 4E-CL. The follotling study r¡as therefore

undertaken r¡ith the follor.ling aims:-

l. To determine the molecular r.leight profiles of synorial fluid

hyaluronic acid from a series of patients r¡ith inflammatory

arthropathies.

Z. To determine similar molecular ueight profiles for a grouP

of non-inflamed synorial fluids.

3. To determine r¡hether any correlation exiEts betueen the

amount of material retarded by these gels and clinical

parametens of inflammation (erythrocyte sedimentation rate and

synoruial fluid cell count).

77

4. To determine r¡hether extracted synovial fluid hyaluronic

acid r¡as susceptible to än i¡1 ritro oxygen radical flux in the same

uräy äs commencially arrailable human umbilical hyaluronic acid used

in the prerious study.

Non-lnflamed synovial fluid samples urere taken from fresh

cadarers. ln order to be included in the studlt there r¡as to be

no history of sepsis, malignancy or anthnopathy. ln addition thene

uas to be no mäcroscopic eridÊnce of osteoarthritis or othen

arthropathy at autopsy.

E. ]IETIIfIDS

Sepharose 4E-CL, and Sephadex Ê 100 LrÊre obtaÍned from

Pharmacia, Uppsala, Su.reden. Eellogel cellulo¡e acetate membraneE

rrlere obtained from Chemetron, Hllan, ltaly. Êlcian blue r¡as

obtained from Difco, Sumey, UK. Glycosaminoglycan standards

uJere a gift from Dr. H.E. llatther¡s, Ehicago. Êll other chemicals

LrerË of the highest punlty ar¡ailable from Sigma Chemical Eompany,

5t. Louis, l"lissouri or Êjax Ehemicals, Sydneyr Êustnalia.

l. Sunorial Fluid

Details of patients are given in Table 5- 1. The erythrocyte

sedlmentation rate r¡as determlned by the lrJestergren method

and synorlal fluld uhlte cells r¡ere counted using a

haemocytometen. Êfter aseptic asplration of the knee joint in

each case the synorial fluid uas placed into a sterile tube

(r¡ithout anti-coagulant) and diluted r¡ith an equål rrolume of

normal sallne. Details of the cadaver subJects are giren in Table

5-2. Synorrial fluid from these cadarers uas obtained r¡ithin E

hours of death by closed irrigation of the joint r¡ith l0 ml of

normal saline, after r¡hich the joint rra5 opened to establish the

lack of macroscopic evidence of joint pathology. The cellular

content of the fluids thus obtained uJas removed by

7E

Table 5-l CLll''llEÊL DETÊ|LS FR0ll PÊTlEffTS ldlTll ÊRTtlROPÊTllY

Petient Sex Ag.

(yru)

Duration ofdl¡tr,tt(yrù

Synovial lluidwhltr crll count

(ctll¡tr¡ni)

% oî totaluronic ¡cidrtt*drd by9rpherorr

{B-CL

E5R

(mm/hr)

Srnopositivr nhruTnetoid enthnitisPrF65leP2F?15P5MÊ615P{ M 45 llP5 F +'l e5PÊF?g16P?M60lePEF5218Pg 1.1 6? 21Pto F {E 19PilF56l+

Srnonrgetivc nhrumatoid crthrltisPre F 68 16PIs F Ê? IgPt{ F ?{ llPlS F eS Zmo.PI6MBtl¿

P¡orietic rrthnitisPI? H 81

Pr8 ll +5

Rcltrr'¡ syndromePlg 1''1 lg Zmo.

0stroarthritisP21 F

F22 F

Ênkylosing spondylitis uith pæiphcral arthropathyPzg F ?5 25 15 2,809

t09r08t05685039322120l9t2

8t?05tt5E

NÊ*Nâ

900Nâ

+,50031,900

t00NÊ

NR

{,500Nñ

20e9l6t0ete?t2t{tst{t9

I{II

+00500+00{00600

100

e?2ersI

r5

ll

35

t7

3l

62

I+

8,290I,7EÍ)

11Ê

ilÊ5g?9

2525

23

t3l+

tJlâ . not erailablo

6

z9

Table 5-Z DETÊ|LS FR0l'1 Tl'lE EÊDÊlzER SUEJECTS

Subject Sex Ag" Êutopsy diagnosis ß Haterial retardedby Srpharose 4B-CL

?

??

M

lt"l

F

M

cl

ce

c5

c{

55

5?

58

Myocardirl inferction

Myocerdial infanction

Rupturrd aonta

Sub¡rechnoid h¡cmomhage

t0

l0

l1

EO

centrifugation at lUB0 x g îor 15 min. The supernatant uas

collected and frozen until assayed.

2. Hualuronic acid seoaration

Synoruial fluid hyaluronic acid r¡as isolated by precipitation

r¡ith cetyl trimethyl ammonium brom¡de (ETÊE) according to the

method of SEOfi [ 1968]. Ê l% solution of CTÊE in 0.3 l'1 sodium

chloride r¡as added to synovid fluid in 0.3 l'1 sodium chloride in a

ratio of 5: l. This uas then centrifuged for 30 min at ll,OOO x g

to remove àny precipitate. lrlater uas then added dropuise to

the synor¡ial fluid: CTÊE mixture until the sodium chloride

concentration approached 0.0?5 H at uhich stage a heavy

precipitate formed. This contained the hyaluronic acid-ÊTÊB

complexes. The sodlum chloride concentration r¡a¡ then decreased

to 0.04? 11 to ensure that all the hyaluronic acid present r¡a¡ in

the pnecipitate. This uas then centrifuged (5,000 x g f or l0 min)

before remoral of the supernatant (3?). The precipitate r.las

then redissolrzed in 1 11 sodium chloride at r¡hich stage the

hyaluronic acid ETÊE complexes dissociatÊ. This solution r¡as then

neprecipitated r¡.rith 4 r¡olumes of ethanol at SoC and

recentrifuged. This procedure kras repeated twice, after r¡¡hich

the final pnecipitate uas redissolved in r¡ater (there being enough

salt present in the pellet to allou this to occur). lt r¡as then

ethanol precipitated, cettrifuged and dried Jn vacuo aveî

phosphorus pentoxide. The removal of the protein from the

synorial fluids rJas monitored by opttcal scanning of the

supernatants betr¡een 190-5?0 nm on a Perkin-Elmer

spectrophotometer. lt r¡as determined that material thatabsorbed at 280 nm hJäs removed u.rith supernatant 52 and the

final pnoduct had negligible absorption at this r¡ar¡elength.

8r

5. Eellulose acetate gel electrophgEEig

ln order to demontrate that the glycosaminoglycan in the

final product rr.ras hyaluronic acid, cellulose acetate gel

electnophoresis rr.ras penformed acconding to the method of

EÊRT0LD, hJIEEKIN and Tl-l0NÊRD [1382]. Ê characteristic "fingened"

Êlcian blue stalning band on electrophoresis connesponding to the

migration of a standard sample of hyaluronic acld r.rlas nequired

before the sample r'.las further analyzed.

4. Eel chromatogxgphy-

Êliquots oî 8,2 ml of a ¡olution of purified synovial fluid

hyaluronic acid (l mg/ml in buffer) rr.rere eluted on Sepharose 1Ê-

EL columns (30 x 0.? cm) at 7 ml/hr uslng 0.5 11 ¡odium acetate,

pll 5.8, as the buffer. Fractions of [J.? ml uere collected. The

columns rr.Jere characte rized using Dextran blue, 3J5-sulphate and

glucuronic acid. Retarded rolume material uas defined as uronic

acid containing material beyond fraction l2 (voad rolume atfractlon t 0). Êreas under the curre for each profile LJere

measured on a J'ler¡lett-Packard 9E?Ê digitizer.

5. Uronic acid anal,usis

Uronic acid r¡as determined by the meta-hydroxy diphenyl

method of ELUI1ENKRÊNTZ and ÊSBOE-HÊNSEN I l9?5] oF a

modification of the automated method of ROSEMI'IÊL, EENTLEY and

ÊLEtN I le?E].

E. Qðyl|en nadical qeneration-¡-E¡The ferrous ion autoxida tion systËm kJas used as.outlín ed in

Ehapter ll. Ê stock solutlon of 2 mg/ml r¡as made from the

extracted desiccated non-arthritlc cadaren synorial fluid

hyaluronlc acld ln potasslum phosphate buffer, 50 mll, pll ?.4. The

reaction mlxture consisted of hyaluronic acid, I mg/ml, buffen,

fernous sulphate at concentrations of 0, 50 and 1000 ¡.rÌ1 and EDTÊ

82

bJås prËsent in ä molar rätio to ferrous ionE of l: l. Viscasity was

measured after 2 hr as described in Ëhapter l.

E. RESULTS

Fractionation r¡ith subsequent analysts of rau ;,ynovial fluid

resulted in the prËsencÊ of another material or materials thateluted at least partly r¡lthin the lncluded volume and r¡hich

lnterfered r¡ith the uronlc acid assay used. Using the CTÊE

precipitation method, as outlined abore, no other

glycosaminoglycans urere detected by Êlcian blue staining of the

electrophoretically separated samples (Fig. 3- l) and no

interferencË occurred r¡ith the uronic acld a:.:,ay. lt r¡as alsg

established that degradation of contnol samples of human umbilical

cord hyaluronic acid did not occup during the separation

pnocedure.

Fractionatlon on Sepharose 4E-CL of synovial fluid hyaluronic

acld from control subjects demonstrated that the uronlc actd

containing material eluted at the V6 of the colurrn. Peaks ul,lre

either symmetrical (Flg. 3-?, Ã) or Lrere dlghtly skewed towardr

the included rolume. The mean % of total uronic acid in the

retarded material r¡as 10.3 (50 = 2.374, 5EP1 = 1.1%) [Fig. 3-3].

For the patients r¡ith inflammatory arthropathies the mean 96

of uronate in the included r¿olume after fractionation on

Sepharose {E-EL uas lE.0 (50 = ?.0, 5El'1 = 1.6) [Fig. 5-5]. This

r.ras significantly different. ln most casËs this r¡as pnesent ås a

sker.l.ring of the excluded rolume material peak (Fig. 5-?, E),

although in some cases there rJas a separate peak of netanded

unonic acld-containing matenial (F¡g. 5-Zr E). The Kar¿'s of these

peaks r¡arled betr¡een 0.2? and 0.?5, cornesponding to molecular

rrreights betr¡een 9x l0s and Zx l0{. No material r¡as seen r¡ith amoleculan r+reight lor¡er than ZxlE+. This finding rJas confirme d by

\

FIGURE 5- I ELECTROPI.IORESIS OF STÊNDÊRD 6LYEB5ÊIIINO6LYEÊN5 ÊND

EXTRÊETED PÊTIENT ÊND CTNTROL SÊHPLES.

5 microliters of each sample (l mg/ml) r,.ras applied to a

t.5 cm line on cellulose acetate strips and electnophoresed

at 5B r¡olts /cm in 0.? M calcium acetate. The strips urene

stained r¡ith 0.29 Êlcian blue.

Top panel: Standards

!r. l-lyaluronic acid

b. Ehondroitin 4 sulphate

c. Chondroitin E sulphate

d. Hixture of each of the aborre

Bottom panel: Sample patients and controls

Ê. l-lyalunonic acid standard

f. c2

9. Pt+

h. Ptg

i. P5

*ve

I trËI

dcbo

-ve

+llft {-+l e) a

hIre

FIÍìURE 5-Z CI.IROHÊTTERÊPI-IIC PROFILES OF EXTRÊETED SYNOI/IÊL FLUID

I.IYÊLURONIC ÊCID ÊFTER FRÊCTIONÊTION ON SEPI.IRROSE 4B-EL.

(lÊ) Control subject,

( I B) seronegatlrre rheumatoid arthritis,( lC) seroposltlre rheumatoid arthritis,( lD) psorlatic anthritis,

( lE) Re¡ter's syndrome,

( I F) patiant rejected as a contnol r¡ho died aften a

prolonged illness that included a lymphoma and a

mediastinal abscÊss complicated by an episode of

septicaemia.

voI

100

80

60

40

20

100

BO

60

40

20

100

80

60

40

20

Fig. 1Ac2

Fig. 18

Vt

I

?eoü,obo5z9l-ÊÞzl¡¡ozooeo9zoÊÐ

0

Fis. 1C

Fig. 1EPt9

Fig. 1F

P14

0

0

P3 c5

5 10 15 20 250 5 10 15 20 25 30

FRACTION NUMBER

Fis. 1DPr8

V

ItVe

I

FIBURE 5-5 PEREENTRGES OF IlÊTERIÊL RilÊRDED EY SEPIiÊROSE 4B-CL

IN EÊDÊPER SUBJECTS ÊND INFLÊHTIHTÍIRY ÊRTTIROPÊÏTIY

PÊTIENTS.

50

PERCENTMATERIALRETARDED

BY

40

SEPHAROSE 48

30

20

ro

oo

a

oa

oaooOo

.3.ooooo

0

ooaa

CADAVERS INFLAMMATORYARTHROPATHY

PATIENTS

86

fractionation on Sephadex 6100 as all material analysed from the

inflammatory arthropathy and control groupi uas excluded from

this gel. Fractionation of several samples an Sepharose ?ts-EL

(data not shor¡n) resulted in similar profiles and these r¡ere not

altered by fractionation under di=sociative conditions using 111

guanidine HEL as the elution buffer, Smaller molecular r.leight

material r¡a¡ found in all vaüeties of inflammatory arthropathy

¡tudied r¡hich included sero-positir¡e and sero-negatirre nheumatoid

arthritis (Fig. 5-2, E & E), psoriatic arthritis (Fig. 5-2, D), Reiten's

syndrome (Fig. 5-2, E) and ankylosing spondylitis r¡ith periphenal

arthropathy, Êlso in tr*ro patients r¡.rith osteoanthritis retanded

material ¡Jas present, although these patients uere not

othenr¡ise included in the analysis.

There uJäs no significant cornelation betuJeen the percentage of

uronate retarded on Sepharose {ts-CL from the synor¡ial fluids of

patients r¡ith inflammatory anthropathies and (a) the

erythrocyte sedimentation rate, (b) the synorrial fluid r¡hite cell

count, (c) age, or (d) duration of disease.

Extracted synovial fluid hyaluronic acid from cadaver subject

C I rJas exposed to the ferrous ion autoxidation system using

ferrous ion concentrations of 50 and 1000 ul"l. The results are

ilfustrated as Fig. 3-1. There is migration of the uronic acid

containing peak across the partially included volume in a manner

identical to that seen uith human umbilical cord hyaluronic acid.

No uronic acid-containing matenial from either the cadayer

subjects or the inflammatory arthropathy patients r¡as retarded

by Sephadex G100.

FIEURE 5-{ SEPHÊROSE .îB.EL ÊND SEPI.IÊDEX Iì TOO CI-IRBNÊTTIGRÊPI.IY OF

SYNBHIÊL FLUID I.IYÊLURIINIE ÊEID EXPOSED TO T}IE FERROUS

ION ÊUTOXIDATIBN SYSTEI.1.

Extracted synorrial fluid hyaluronic acid fnom contnol

subject ES uas exposed to the femous ion autoxidation

system at fernous ion concentrations of 50 Ft1 (E Nsp =

?6) and 1000 FH (96 Nsp = 9) under conditions outlined in

the text.

SEPHAROSE 4B-CL

l0

SEPHADEX G1OO

Nsp 100

=b-¡-

zIl-É,t-zt¡¡ozooeo

IzoÉ,f

Nsp 7ó

Nsp 9

5lzvt15 20 ,ì 2s 30 s

vr FRAcnoN

0

vot5 tol

vol5 20 30

E8

D. Dl5CU55l0l{

The initial studie¡ in this investigation using untreated

synorzial fluid highlight potential pitfalls in the use of colorimetric

assays for uronic acid upon untreated samples from human tiEsue

and body fluids. The detection of a substance that is retarded

on Sepharose 4E-CL and produces an interfering colour reaction in

the uronic acid assay used makes necessary a purification

procedure to isolate the hyaluronic acid before chrom atography

and uronic acid analysis. DÊllL et ê!. I I gS5] urere able to

circumrent this problem by using a specific radioa=say (,using

radiolabeled link protein). Ênother factor dictating the nece ssity

for a separation procedure is the fact that chondroitín sulphate

has been shor¡n to be present in normal synovial fluid IS|LPÊNÊTNÊ,

DUI{STONE and 065T0I{, lgE?l and to be present in elevated

concentrations in pathologlcal fluids ISEPPÊLÊ *. d., l97?l, This

mäy be cartilage derived or a secretory product of synovial cells

lt1ÊR5l'l e.[ ê1., l3?9; SÊÊRN| É ê!., 1gÊ0l. The fact that after

extraction this glycosaminoglycan could not be demonstrated by

Êlcian blue staining after electrophoretic separation indicates

that chondroitin sulphate is unlikely to account for any of the

uronic acid containing material retarded by Sepharose íÊ-EL.

The selection criteria for donor patients (i.e. lack of infectisn or

cancer in the prÊ-morbid illness) r.r.rere imposed due to the

possibility oî colonization of the synorzial fluid and/or synovial

membrane uith bacteria or metaEtatic carcinom a cells, both of

r¡hich may be sources of exogenous hyaluronic acid [5T00LNlLLEE

and DORFI1ÊN, l9?0; KltlÊTÊ É al., lg8'fl. These criteria excluded

one autopsy subject as a non-arthrltic control. This subject (Fig.

3-?., F) demonstrated a large amount of .hyaluronic acid (a3%J in

the included rolume on Sepharose 4E-CL chromat ography. ln this

89

case no orert sync//ial inflammation uas present at autop=y and

her synorial fluid uras analyzed before the full clinical details rrere

arzailable. Those clinical details revealed lymphoma, mediastinal

abscess and an episode of septicaemia. She r¡as not used as a

control in this study due to the possibility of bactenially-deriyed

hyaluronic acid. l'lot.lerer, another explanation for the lange

amount of lor¡ molecular r,r.reight hyaluronic acid in hen cäse is the

effect of a prolonged and serere illness r¡ith its associated

metabolic derangements upon synovtal cell hyaluronate production

(see Ehapter /l). This is an area that uarrants furtherinrestigation.

Êadarers cannot be considered an ideal so,Jrce of control

synorzial flu¡d due to the possibility of post-mortem changes. ltfJas, hor¡erer, considered unethical to expose rolunteers (if they

could be found!) to the risks of arthrocentesis, hourerer small.

Ner¡ertheless, our results shor¡.1 that the molecular rrleight of

synorzial fluid hyaluronate fnom non-arthritic subjects, rr.rho died

suddenly, uras uniformly high, r¡ith mainly symmetnical peaks at thePs on Sepharose 4E-EL. No mope than l{16 of the uronic acid

containing material r¡as netarded by Sepharose +E-CL. Thus

post-mortem depolymerization does not appear to hare been a

factor. DÊl'lL et ê!. t lSË5] hare independently reached the same

conclusion.

llolecular rr.reight profiles on Sepharose íE-CL demonstrate

considenable rariation in the shape. Ê striking feature, however,

uas that the predominant amount of uronic acid-containing

material uas excluded by Sepharose 4E-CL i.e, it represented a

hydnodynamic size of I x 106 or greater. Samples run on

Sepharose ZE-CL, ruhich can char acterize molecule s rp to a higher

limiting molecular r,leight, índicated that much of the material has

90

a hydrodynamic size as high as l0?. Therefore, in inflamed synovial

fluid a significant proportion of the hyaluronic acid present is still

of a high molecular u.reight and r¡ell able to interact to form a

polymer netr¡ork at concentrations normally found r¡ithin inflamed

joints, i.e., I mg/ml [EÊLÊ25, 1971].

The amount of material included on Sepha?ose íB-ÊL nanged

bet¡¡een E9ã and 35?6 of the total and the smallest material

present had a K¿y of 0.?5 comesponding to a molecular ueight of

? x l0'+, Thls ls r¡lthln the neglon r¡here hyalunonic acld is found

to be stimulatony of macrophage phagocytosis IF0RRESTER and

EÊLÊ25, lgË01.

There rJas no correlatlon betr¡een the relatíve amount of

smaller moleculan rr.leight hyaluronic acid and clinical indices of

serzerity of joint inflammation including erythrocyte ¡edimentatisn

nate and synorial fluid cell count

Extracted cadarer synovial fluid hyaluronic acid uJas susceptible

to oxygen radical induced depolymerization to the same degree a=

human umbilical cord hyaluronic acid. Oxygen radical exposure

results in depolymerization products that fall r¡ithin the saTfie

molecular ueight range as sÊen in extracted inflamed =ynovial fluid

hyaluronic acld. Ê mild atrygen radical flux ([Fe ++l = 1L]fi applied

to extracted cadarer synovial fluid hyaluronic acid produced a

Sepharose 4E-CL profile similar ín appearance to many extracted

lnfl¿med synorial fluld hyaluronic acid's. ThiE finding supports the

hypothesls that otlygsît radlcal depolymerization can cäu5e a

decrease in arerage molecular ueight (and increased

polydispersity) of ínflamed synovial fluid hyaluronic acid E vivç).

9t

E. SU]1]1ÊRY

llyaluronic acid uas isolated from the synovial fluid sf ?E

patients r¡ith inflammatory arthropathies, Z osteoarthritíc

patlents and 4 cadarer subjects. Fractionation of hyaluronic acid

from inflammatory arthropathy and osteoarthritis patients on

Sepharose 'fE-EL revealed that uronic acid containing materiat

r¡as retarded (ave. 17.3%). Synovial fluid hyaluronic acid from

cadarer subjects r¡ho died suddenly r¡as excluded from Sepharose

{E-CL (i.e., molecular ueight >5x 106). ln most patients retarded

unonate uJas present as a sker.ling of the excluded volu'ne peak

torr.rards the included rzolume, hou.lever in some patients Emall

discrete peaks uJÊre present. No material r¡as present belou a

molecular r.leight of Z x l0a and most material present had a

molecular r+leight greater than l0õ, 1.e., r¡lthin the region thathyaluronic acid r¡ill form a polymer netuork at concentrations

found in the inflamed joint. ln patlents r¡ith inflammatony

arthropathies, there uJas no significant correlation betr¡-leen the

percent of uronate retarded on Sepharose 4E-CL and (a) the

patient's erythnocyte sedimentation rate, (b) the synorrial fluid

r¡hite cell count, (c) age, on (d) disease duration. Extracted

synorzial fluid hyaluronic acid hJas susceptible to oxygen radical

induced depolymerization'n vttro to the säme degree as human

umbilical cord hyaluronic acid.

92

C]IRPTER I1'

IIERSURE}IEIIT OF REDUCI]IG EIID SROUPS GEIIERRTED EY

EXPOSURE OF ]IYRLUPOIIIC RCID TO OXYGEII RRD¡CRLS

98

Ê. iltTRoDUCTl0ll

Enzymatlc depol¡,rmerization of hyaluronic acid ls accompanied by

the release of reducing ends irrespective of the linkage cleaved.

Testiculan hyaluronidase (EC 5.2.1.35) and hyaluronate lyase from

Streptomyces hyaluronolyticus (EE 1.Z.Z.l) are both endoenztmes

that can clearze hyaluronlc acld at the þ1-1 glycosldlc llnkage

lll0UEK and PEÊRCE, 195?; LIHKER, NEYER and t'l0FFl1ÊH' lgsÊl. ln

the lattËr cårc ân unläturated uronide lr formcd at thc non-

reducing end. Thi¡ form of clcaragc r.rill"learc l1-acetylglucosamine

at the reducing end and can thus be meaEurcd by an increase in

Horgan-Elson reactivity, 0n the other hand leech hyaluronidase

(hyaluronic acid-endo-p-glucuronidase), also an endoenzyme,

clearrs at the F | -3 glycosidic linkage, thus leavng D-glucuronic

acid at the reduclng end IYUKI and FlSl-lMÊN, l9E3J. No increase in

llorgan-Elson reactirzity can be detected follor.uhg this form of

depolymerization, houerer an increase in total reducing ends r¡ill

be detected by less reEtricted a#ays of reducing ends' such as

that of PÊRK and J0lltl50N I tg49].

It iE not clear from the llterature if reduclng ends are

detected after depol¡rmerlzation of hyaluronic acid induced by

oxygËn radlcals. tlcELEÊN and llÊLE I l9+0] dld not dctect reducing

ends after ascorbata lnduced depolymenization of hyaluronic acid

and morr racently EREENhJÊLD and l10Y I l9B0] dld not detect

reducing ends after Ë,(poture of hyaluronic acld to the XCI/HX

oxygËn radical generatlng system. llor¡erer CLELÊiID qË åL t lgEgl

urerË able to dcmonstrate that the depolymerization of

hyaluronlc acld exposed to ascorbic acid at pll 5.5 is accompanied

by the release of reduclng ends as demonstrable by reaction r¡ith

sodium borohydride. The resolution of this discrepancy is

94

of importäncË in the further clucidation of thc mechanirm of

oxygËn radical induced dcpolymerization.

The aim of this inrzestigation r¡as to determine r¡¡hether

reducing ends brerÊ detectable in samples of hyaluronic acid thathad been exposed to oxygen radicals generated either by the

autoxldatlon of femous ions or by the actlon of XE upon l'lX. The

assays for neducing sugars uJere the ferricyanide assay of Park

and Johnson and ta?-cyanide labelling of reducing ends, both of

r¡hich r¡ill detect N-acetylglucosamine and glucuroflic acid at the

reducing end and the Horgan-Elson ä,ssäpr l¡hich is specific for N-

acetylglucosamine at the reducing end in this system.

E, }IET]IODS

l. gry

a'

l'lyaluronlc acid uJäs e¡(posed to ferroug ionE in the

prËsence of 50 ml1 phosphate buffer pll ?,1 under conditions

identical to thoEe outlined ln Ehapter ll. Fcrrous ion

concentrations betr.¡een 50- 1000 ul1 uJere used.

b, x0/Ëf-lyaluronic acld uras exposed to the XEll'lK system under

conditions similar to those outlined in Chapter ll, except that in

this series of experiments a fixed concentnation of X0 of 5 mU/ml

r¡as usÊd. The magnitude of the flux ¡¡as varied by altering the

inltial concentratlon of l'lX betueen l0- 1000 ull and the reaction

¡¡as allor¿.led to go to completion (i.e. I hr).

Z. En¿y:nggg-dËgr a dation of h,r.r alur onic acid

a.

Eorlne testlcular hyaluronldase (Slgma Typ, lV-1, EC,

3.?.1.35) r¡as added to hyaluronic acid, I tg/ml, in 50 mll

phosphate buffer, pll 7.1, and incubated at 3?oE at e'nzyme

e5

concÊntratlons betl¡een l0 and 500 ug/ml for 2 lts. The reactlon

r¡as stopped by changing the pl'l to I r¡ith the addition of Na0l'l

and placing the dlgestion tube on lce.

b. Stfs,plegylefiyalurÉnldase d¡ gestion.

Hyalunonate lyase from Streptomyc,æ hyalurolytlcus (EE

1.?.?.1) r¡as obtalned from Selkagaku Kogo Co., Ltd., Tokyo, Japan.

Solutlons of I mg/ml hyaluronlc acld ln 50 ml{ phosphate buffer' PH

8.0, u.rare lncubatcd uith a concËntratlon of hyaluroflate lyaæ of

l0 TRU/ml at E0oE for tlmes betueen 5 and 120 mlnutes. The

reactlon rräi ågäln stopped by raising the pll of the solution to B

r¡ith NaOH and placing the sample on icÊ.

5. Esl-hæ¡nÊlsgcaphy-

The arerage molecular r.leights of the hyaluronic acid

samples urere determined by gel permeation chromatography,

Êliquots of 0.2 ml of a I mg/ml solution of hyalursnic acid t¡¡ere

applied to Sepharose 'IE-CL or Sephadex 6100 (Pharmacia, Uppsala'

Sr¡eden) using perspelr columns (50 x 0.? cm). The columns rrÊre

eluted at ? ml/hr uslng 0.5 11 sodlum acetate, pll 5.8, as the

elutlon buffer and fractlons of 0.? ml r¡ere collected. The colurnns

rrêFÊ callbratcd ulth Dextran bluc, [353]-gulphate and D-

glucuronic acld. Ka,r's uJÊrË mÊàsurÊd and conrented to molecular

r.letghts by refarÊncË to data glrran by Flshar I l9B0] for dextPàns.

't'@Prlor to analy¡il thc samples urÊPÉ cxhaustlvely dialyzed

agalnst uJatËr.

a. !.@ig-ê.g!.d

Uronic acids urere determined by the metahydroxydiphenyl

method of ELUIIENKRÊNTZ and ASE0E-|'|ÊNSEN I l9?5].

96

b. Park-Johnson assal¿

A modlflcatlør of the ferlcyanlde method of Park and

Jolrnson as outlined by ÊSlllrJELL IlSS?l uag used. This assay is a

ve?y sensltlre mËêiurË of rcduclng rubstances. The modlfication

applied urâs èr outlined by llOUEK and PEÊRCE [lg5?] and entails

ensuring that thc carbonata-cyanlde reagent ha¡ a pll no higher

than 9. Êt a morË alkallne ptl, hydrolysls of the hyaluronic acld

occurs laadlng to falseþ eleyated ralues of thc reducing ends

obtained [R0ElN30N and tl0PlrJO0D, lg?31.

G. l'lorgan-Elson reactlon

The Rrissig modlflcation IREl33ltì, STR0NIIIGER and LEL0lR,

19551 of the llorgan Elson reactlon uag used. Thls assay ls

spcclflc because ln order to be porltlrc, ln addltlon to a frea

reduclng end, lt requlres the presÊncÊ of an acylamldo group and

a frce hydroxyl at the 'lth carbon of the rugar molecule.

ThereforË, ln thc case of hyalursnlc acld, lt r¡ill only be posltirze

r¡hen l{-acetylglucoramtne ig at the reduclng end.

5. Ê'¿¡niCcJsÞellhg

The method usad b,ä5 crsÉrrtlalþ that outllned by SlrJÊtlH Et

e!. If 9821. Degraded samples oî hyaluronlc acld r¡ere dlalyzed

agalnst 1000 volumes of E.? l'1 l{arE0r/ÌlallC0, buffer, pll 9.1. The

uronlc acld concertratlon of the solutlon uas then determlned

and an equlmolar conccntratlsn ol KtlCn (gO y}l.ml) was

incubated uith the mlxture for 3E hrs al ?2o8. Then unlabelled

KCÌ1 r¡as added to achleve a concentratlon four tlmes that of the

uronlc acld concetrtratlon. Êfter a furthrr l2 hour¡ the reaction

uras terminated by the addltlon of acetlc acid to glve a pll of 4.5.

The resultant sample uas thçrr fractlonated ør a Sapharose 48-

CL column ( I x E0 cnr) at l0 ml/hr urlng 0,5 H sodium acetate, pH

5.6, as the elution buffen. I ml fractlons uJËre collectad and

l

1

I

The reason that there is a discrepency betr¡een thedegr-e of Par*-Jofrnson reactlvlty for a gtvenmolecular r,reight rrcm¡in¡ unclear, although it maybe related to the inherent inaccuracy in using thebro¡d pcak, obtsined aftcr gcl chromatographl, lom"¡3un molecul¡n r,rcight,

97

ästayËd for ursnlc acid crtttrrt and radloactlvlty kJaJ counted

uslng a Beckman llquid sclntlllation counter.

E. Êscendlng Paper Êhromatograohp.

tlyaluronlc acld sampler tabelted uith '4C-CH urÉrÊ hydrolysed by

incubation at l00oÊ r¡lth Ën Êqual rolume of 4Ì{ llCl for .lE hrs

under nltrogen. MonoiaccharldÊr uJerÊ lsolatad å! a slngle peak

after fractlonatlon on Elogel P2. ÊscËîdlng chronratography oî

matcrlal from the monosaccharide peak rlas pcrformed on

hJhatman lJo I chronratography päpêr. Th¿ solve¡t uged bra¡

ethyl acetate I pyr6ne / r¡ater in a ratlo ol 12z3z1, Samples of

D-glucuronlc acld, l{-acetylglucoramlne and glucuronlc acid lactone

urÊr? labelled r¡ith I{C-CN as deccribed prevlously and uerÊ run

concurrcntly. Êfter lE hrs thc paper uJat dried and cut lnto lcm

strips and the radloactlvlly counted.

C. RESULTS

The rcsults of the colorhnetric assays are ¡houm in Flgure l- l,llyaluronlc acld degraded by either tasflcular hyaluronidase or

hyaluronate lyase from Streptomyces hyaluronoþticus shor¡s an

increase ln Parl<-Jolrnswr raactirlty. Thls ls closeþ reflectrd by a

slmilar lncrease ln l'lorgan-Elson rcactirlty. Thls ls in keeping ulith

the kno¡rrn actlon of thesa Ênzlrmer r¡hlch are both endocnz!¡m-es

that claaye haxosaminldlc llnkages thus learfng N-

acetylglucosamlnr at the reduclng cnd ILAUREM, l9?0]. ln both

oxygrtr radlcal generatlng systcms, hourerer, there is a marked

dlrergancc betr¡ecn the reduclng end to uronic acld raüo using _

the tr¡o assåys. Thene is a rise ln neducing ends uith decreasing

molacular r.lelght as mêåsurcd by the Park-Johnron assayr but

thls ls not mlrrored by a rlsa ln reducfng ends as mÊåsured by the

Relssig modlfication of the llorgan-Elson âssày.

FIGURE I- I ËENERÊTION OF REDUEINË ENDS HFTER DEPOLYI'IERIZRTION OF

HYÊLURONIC AEID EY EXPOSURE TO ENZYIIÊTIE RND OXYGEN

RHDIEÊL GENERÊTINE SYSTEIIS.

Reducing ends urere detected by the Park-Johnson assay

as modified by l-lOUCK and PEÊREE I lg5?], t ) and the

modification of the llorgan-Elson assay acconding to RElSSlE,

STRoIItNEER and LEL0IR ItgS?], (r-).Each point represents the mean of triplicate

determinations of the reducing end assay used, expressed

as å ratio of the mean of tniplicate determinations of the

uronic acid ralue of the sample.

The rar¡ data are shor¡n in Table 4- l.

STREPTOMYCES HYALURONIDASE TESTTCULAR HYALURONI0.15

o.Fl!(Ú 0. l0ú

.Jo 0.05

0.15

10 10 6

FERROUS ION

€

o.r{Éotr

!É

tr¡

è0c

.r.loЀc,É

ro5 to4

AUTOXIDATION

ro7 106 105 ro4

XANTHINE OXIDASE/HYPOXANTHINE

0. 10

0.05

10 1067 ro5 ro4 to7Molecular Weight

106 ro5 ro4

99

TRELE {- I COTIPÊRISON OF REDUÊINE EI*ID TO UROTIIS ÊgID RÊTIOS Ê5

DETERTIINED BY TTIE PÊRK-JOIINSIIN ÊSSÊY ÊND HOREÊ¡I-EL5OI'.I Ê55ÊY

Perk-Johncon UÊ natioHIJ x 5D

Morgan-Elson UA ratio1'4lJ x 5D

Fenrous ion autoxidationFc(ll) ¡M

I25I5

E

I23

1

5

6

0t0

100200500I 000

5xl.5x

g6góg6g5glga,lx

0.00r 60.00r 80.00830.01510.05560.015 r

0.0036g.alzE0.01Ê60.03820.055,1E.OS12

0.00060.0r 250.0190O.E?ZEo.l2180.t ?90

0.00810.00940.0{050.05130.t 1680.1 8t 5

0.0002û.00010.001 r0.0018E,ÊE+Z0.0080

E.EAZ10.008?0.00310.00220.00?g0.0040

0.00050.00100.00+00,00550.00t 10.0106

0.00110.00r 40.000Ê0.00180.00500.0060-

5xl.5x

lx(r(

1x1x

3xl 0ól.5xt 06

1x106lxlOj1x103SxlO+

2x13x[

00

0.sat20.00280,00130.01 t 5

0.00160.00150.008s0.0t 060.0 r 190.0159

t.001?0.026a.g?g+0.r 0200. r 5290.2325

0.00?50.029?0.12100.19630.3030.356

0,00020.00050.00100.001 I

0.0003

0.00030,0005Ð.00020.0010

0.00t 89.00?20.01820.0501E.O?92E.EBl?

0.000r8.00s20,0072a.a2350.005?0.0325

g6gÉ

6693gJga

lx(xlx

ilypoxanthinr/llenthine oxidrsrHx ¡rMl0

z2E3 t001 2905 5006 t000

3xl 0ó5xl 0ölxl06'lxl05lxl0s5xl 0a

2x1063x1053xl0alxl0a9x1038x103

2x1063x1059xl 0+5xl0aflxl0+Zxl0a

Testlcular HyalunonidaseConc(u9/ml)

102 t0550+ 1005 2006 500

g6g5gaga

93g3

gÉgsgag+g+ga

5x.5x9x8x

2x3x9x5x1x¿x

Strcptococcal Hyeluronid¡¡cTime(mins)

05

t530EO

tzE

r00

The rise in Park-Jol'rnson acllvlty ln hyaluronlc acld exposcd tothe ferrous ion autoxldatlon system uJèr further tested by

measurlng the Park-Johnron to uronic acld ratlo beforc and afterextenslra dlaþsls agalnst uJätÊr. Slnce alt the componcnt¡ of thl¡autoxldation systÊm årr of lor¡ molecular ueight they uould not

ba ratained by the dlaþsls bag (molecular r.relght cut off =

10,000). l{o changc in the Park-Johnson to uronlc acld ratlo uas

rcÉrr after this manoÉurrÊ (data not shoum).

Since lt has beer suggested that oxygcn radlcal cxposurÊ

mlght caulc port-clearägÊ modlflcatlon of the reductng c'nd

[tìREE]llÂlÊLD and ilOY, l980l, thls posslbllity uas tested. Ê rample

of hyaluronlc acld r¡a¡ exposed to a llrnited dlgestion uithhyaluronate lyase and uas then retrleyed by ethanol

preclpltatlon X3 and uag thcn drled þ yacqo oyer phosphorus

pentoxlda. Thls r¡as thcn expoted to the famous autoxldation

system (FIgure 1-21. Thc llorgan-Elson to uronlc acid raflo of

enz¡rmatlcally degraded hyaluronlc acld krêt unchanged by

ÊxposurÊ to the oxygÉrt radlcal gcneratlng rystcm, Thls

damonstrateg that post-cleavag¿ modflcatlon cannot account forthe dlrergs'ncÊ betr¡cerr Parle-Johngon and llorgan-Elssn

rcactlritlcs noted fn Flgure {-1.Flgura 'l-3 shor¡s the proflle, after fractlonatlør ør SepharosË

{B-CL of hyaluronic acld samples that hare bec,n reductirrely

labelled ¡¡ith l4c-cN. Labelllng utth l'c-cN occurr urth hyaluronic

acld that has been expored to both hyaluronldase digestion as

r¡ell as both the fcmous ion autoitdaflon and the X0ltlX oxygcn

radlcal ganaratlng systems. The proflle for laC mirrors the uronic

acid profih until P¡ rr.las reached r¡here fnee |{C-CN elutes. The

rate of lncrea¡e ln tha ratio of radloactlrity to uronlc acid

beturen fractlons Z0 - ZB r¡as drtcrmlned for each proflle by

FIEURE t-Z EFFECT OF ÊN IIXYBEN RÊDICRL FLUX ON TIIE I'IORIìÊN.ELSON

REÊCTIYIW OF ÊN ENZYHÊTIEÊLLY DEERÊDED SÊ]-IPLE OF

HYÊLURONIC RCID.

Ê seria¡ of samples of hyaluronic acid urere digested

r¡ith hyaluronate lyase from Streptomycei

hyaluronolyticus for t0 min under the conditlons outlined

in the text. The hyaluronic acid u.¡as then retrieved by

precipitation x3 r¡ith ethanol, drled in ZSC¡¡C ove?

phosphorus pentoxide and made up rrlith the appropriate

components of the ferrous ion autoxidation systc'rn at

ferrous ion concentrations of 100' 200 and 1000 ul1

respectively,

E-0-

E

I

o.r{

\¡JÉ(tloútn

F{€

f4 "-{o

ci <(ü@o¡{ 'r-loÉ=ot{

0. 15

0. 10

0.05

10 106Molecular

hyal uroni daseoxy radical

104105

I'lei ght

FtGuRE {-s sEpHÊRosE +E-cL cHRoHÊToGRÊpHv 0F I{c-cvR¡uDE LÊEELLED

IIYÊLURONIC ACID DEBRÊDÊTION PRBDUCTS.

The uronic acid value is closely reflected by the lerel of

I {E-cyanide until Pt is neached r¡hen unbound llC-C¡t

elutes.

lF{É

o0

50€o

o.r{É

I 10

00

10

50

10

50

0

STREPTOMYCES HYALURONIDASE

FERROUS ION AUTOXIDATION

XANTHINE OXI DAS N /HYPOX ANTHI NE

vo vt

5000

2 500

5000

2500

0

5000

ã0)ÈH.o0)ocr

(-f

tFÚ

3uroEH

I2 500

0

10 20

Fraction Number

30 35

FIGURE I-{ ÊSCENDINIì PÊPER EIIRBIIÊTOERÊPHY OF I'1ONI]SÊCEI'IÊRIDES

ISoLÊTED ÊFTER ÊEtD HYDRoLYSTS 0F l{C-CVR¡IIDE LÊBELLED

I.IYÊLURBNIC ÊCID DEGRÊDÊTION PRODUCTS.

The top panel shot¡s the results obtained after initial

depolymerization r¡ith the X0/HX system. The bottom

panel shot¡s the results obtained after initial digestion

r¡ith hyaluronidase from streptomyces hyaluronidasa. A

l{C-.yanide labelled N-acetylglucosamine standard

mlgrated to l? cm, l4C-cy"ttide labelled D-glucuronic acid

and D-glucuronolactone standards migrated to bet¡¡een

5 and ? cms.

120

100

80

60

40

20

ASCENDING PAPER CHROMATOGRAPHY

XANTH I NE OX I DASE/HYPOXANTH I NE

STREPTOCOCCAL HYALURONI DASEDPM/CM

5000

4000

3000

2000

1 000

5 10

DI STANCE ( C}4 )

20

t04

llnear rcgrcssion anaþsls. The slopËs årË all comparable

(hyaluronldase = lg.E, Fe2+ autoxldatlon = l{.0 and X0/llX = 19.6).

Slnce the only aldehyde that thc enzr¡rne uill rclease is at the

reduclng end, thls flndlng suggerts that aldehyde or ketone

formatlon ln thc onygÉn radlcal systems does not occur

slgnlflcantþ other than from the reduclng end crcated at the

clearage slte (1.c. cach clcaragc rlte = I aldehydc Aroup created).

Furthcr efforts uJGrË undertaken to ldcntlfy the rugar tor¡htch |{C-CN tr btndtng by hydrolystng the rtC-CN labellcd

degradatlon products. llonosaccharldes uJÇrË isolated by

fractlonatlon on Elogel PZ and urÊrê separated by ascandtng päpêr

chromatography. Flgure {-.t shot¡s that for hyaluronldase

dcgraded hyaluronlc acld the label mlgratas r¡lth a similarly

labelled l{-acetylglucosamlne standard. Thls ls ln keeping r.ulth the

knoum actlon of thls Ênz!¡mÊ. l¡Jlth hyaluronic acld exposed to the

X0/tlX system r{E-CÌ{ does not mlgrate ulth the N-

acatylglucosamine standard but ls spread oy?p the areas of

mlgratlon of the lrE-CN labellad D-glucurontc actd and a stmllarly

labelled glucuronlc acld lactone standard.

D" Dt3CU35l0il

The "reduclng Érrd" of a rugar molecule ls due to the præencÊ

of a frae or potcntlally free aldehyde group [LEHNIilGER, l9B3l,

The structurc of glx carbon tugarr ls usualþ deplcted as a ring

structura, hourercr r¡lth reductng rugær a small portiør ls

present in rolutlon al än open rlng structurç uith an aldehyde

group centered on the I tt carbsr. The tuo formg are fn an

equilibrlum that rtrongþ îavours thc closed ring form.

The colorlmetrlc assay of Park and Johnson rellas on the

ablllty of the aldehyde group to act ôs â reduclng agcrrt ¡¡ith ametal ¡on (Fe3*). Cyanldc r¡ill atso rcact u¡ith thc free aldehyde

r0õ

group of a sugar in a reaction first described by FISCIIER [1890]

and knoun ås cyanohydrin synthesis. lnltially utilized as a mêanj

of adding an extra carbon to a sugar molecule, this reaction r¡as

subsequently dereloped by H0YER and ISEELL Ilg5E] as means of

reductlr¡e labelllng of sugars. lt is clear from our rÊsults r¡ith

both of these assay;, that free aldehyde groups (i.e. reducing

sugars) arË prÊsent in oxygÉrr radical dcpolymcrtzed hyaluronic

acld"

The flndlng that one docs not sec Horgan-Elson reactlvity

after dcpolymcrizatlon of hyaluronic acid by the tr¡o olrygm

radlcal grnenating systems tested indlcateg that uhrn

depolymerization occurr, thç rcducing rugar gÉ'nÊrated ig not

centered on the l"l-acetylglucosamlne. Thls i¡ contrary to the

findlngs of ELELÊiiD Èt gl. I lgEg], r,.rho reported that for hyaluronic

acld depolymerlzed by an arcorbate drlren systen, thc $l-4hexosaminidlc llnkage rrras prefererttially cleaved (thus learzing l{-

acetylglucosamlne at the reducing eîd). ln Chapter l, I hare

draum attentlon to the similaritics betuJeen the ascorbate

syste'rn and the femous ion autoxidatlon system (ascorbate

reduces Fe3* to Fez+J. l'louerer, it may bc the dffferencer in

these systems that axplaln the dlvergent flndtngs c.f. the

posslbillty of a mopÊ direct reaction bet¡¡een ascorbate or its

oxidlzed products and hyaluronic acld.

0ur flndings apÊ consistent r¡lth the hypothecis that olrygerr

radicals clearc at thc glucoraminidic linkage thus learing D-

glucuronic acid at the rcducing etd. lndccd EALAZS gt tl. [ 198?l

hare postulatcd that it ir the this glycosidlc linkage that ig

cleared during pulsc radioylsis of aquÊous solutlon¡ of hyaluronic

acld. Eoth pulse radiolysis and the tuo o,(ygen radical generating

t06

tystËmr used in tha prescnt study are bellered to càusÊ

hyaluronlc acld dcpolymrrizatlon by OHt productlon.

It is al¡o posslbla to interpret from the colorimetnlc data thatclearrage mäy occup by lysls of the carbon rlng of elther sugar

uith the creatlon of a terminal aldehydc or ketone gnoupr both of

r¡hlch mäy also react r¡ith cyanida. Hor¡eyer this ls less llkely

than F | -3 glucosaminidlc llnkage ruptura because although the

laballed monosaccharides do not clearly separate on ascanding

paper chromatography solely as labelled D-glucuronic acid,

probably due to lactone formation duning hydroþsis, it is clean

that a considerable portion ls labelled D-glucuronic acid. Should

clearage occur by lysls of a sugar ringr than no such labelled

sugår should be present"

The ldentlflcatlon of reductng sugars by radloactire labelling and

subsrquent separatlon måy p?ov¿ to be a sultably úscfmlnatlng

method to determlne r¡hether hyaluronlc acid has been exposed to

oxyg"n radlcals !n EiEg. ,, "

,.-ï-rr"l,"';, ''E. SU]1HfiRY

Êfter depol¡rmerlzatlon of hyaluronlc acid Ey expoture to the

ferrous ion autoxldation syrtem and to X0/l'lX therc ic an

increase in reducfng cnd actlrlty ar mÊåiured by the Park-

Johnson assåy and by reductirue labelling ulth radiolabclled

cyanide. Therc is houJerer no increa¡e in llorgan-El¡on reactivityt

nor lg there any change ln l"lorgan-Elson reactirzity n hyaluronic

acid pre-exposed to hyaluronidase and subsequently exposed an

oxygËn radical generating system. These flndings uJÊre consistent

uith our hypothesis that oxygen radlcal lnduced depolymerization

of hyaluronic acid occurs by preferentlal clearage of the Þ 1-5

glucuronidlc glycosidlc llnkage thus learzlng D-glucuronic acid at thc

reducing end.

r0?

C]IRPTEN Y

DOES OXYEE]I RRDICRL EXPOSURE RLTER T]IE REILITY OF

llYRLUR0lllC RCID T0 ]{0DULÊTE }î0ll0cYTE Fc RECEPT0R

EHDHE?

t08

R. I]ITRODUCTIOil

The indicatlon that ÊxporurÊ to oxygcn derlved free radlcalg

mlght lead to clcaragc. of speciflc Aþcostdtc llnkager and posstbly

other altaratlont along the lcrrgth of the hyaluronic acid molccule

leads to the questlon:- Do the chemical requclae frsrn exposurt tooxygrrt frec radlcals lead to an alteratlon ln biologlcal or chemical

propertles apart from those rerultlng frorn decreased molecular

trlalght?

The obserratlon of F0RREÍTER and EÊLÊZ,ï l.lg80l thathyaluronlc acld r¡ould lnhlb¡t the phagocytosis of latex beads

demonstrated a speclflc cellular effect that ls of conslderablc

potcntlal blologlcal releyance, partlcularþ ln relaflon to

phagocytosls occurrlng ulthln an lnflamcd Jolnt. Thcy obserred

that thc phagocytorlr of latex bcads by eltcttcd rat perrtoneal

macrophågËs r¡as lnhlbltcd by hlgh molccular r.lclght (1.c. llhl = 1,8

x l0r - ?.f, x l0É), hyaluronlc acld at concrntraflons betueen 0, 1-

0.2 mg/ml. Paradoxtcalþ, hor¡erer hyaluronlc acid of molecular

r.rreight 9.0 x 101 had a gtimutatory effect at concerrtration¡

betr¡een 0.2 -0.5 mg/ml" Hyaluronic acld of thig ¡lze can be

expectcd folloulng depolymerlzatton r¡tth oxygrrt radtcals (see

Chapter ll) and ls found ln thc synovtal fluld of a considerablc

proportlon of patlerrts r¡¡lth infla¡matory arthropathies (see

Chaptcr ttl),

lrJe thercfore sought to verlîy the flndtngs of Fomcstar and

Ealazs by utlllzlng the Fc receptor blndlng assay. There urerÊ

several rÊasons r.rlhy thls assày uas used. Ftrst;ly the Fc receptor

medlates both blndlng and lnterrrallzatlon of l96 coated partlcles

and lgfì aggregates. lt ls a r¡cll characterlzcd membrane

rcceptor [ERlFFlll, ElAl{c0 and 9ILI/ERSTEII{, l9?5J. Fhalþ, the

paramatêr mÊåsurcd, l.e. Fc receptor binding is more dlrectþ

r0g

linked to the phagocyto¡ls of immune conrplexeg and ls thuE ls

morË clearþ llnked to the pathogcneslg of lnflanrmatory Jointdlseasc than the phagocytosls of latex baads.

E. NET}IODs

l" Adherent monocptas

l0 ml samplcs of yenous blood, antlcoagulated ulth llthlum

heparhate, urcre obtalned from healthy Red Cross blood donorg.

These bJGpÉ thcryr dlluted ¡¡lth Ën Êqual rolumc of l'lank's balanced

salt solutlon (833), placed ln a 50 ml slllcontzcd glass centrtfuge

tube and underlayed r¡lth ?.5 ml Flcoll Paque (Pharmacta, South

Seas, Sydncy). The tubes ur?rê thcn centrffuged at 1009 for 35

min at poom tanperaturc. The mononuclcar cell fraction uas

then asplratad fronr the Flcoll Paqua-Hank's 833 tnterfacc. Th.y

brÊrÊ then rrrashed three tlmes ln llank's 853. The flnal call ptlleturär r?suspendad ¡n RPHI-|E{0 medlum supplementad r¡tth l09E

fetal calf serum (FCS) and penlclllln/streptomyctn. They kr?rÊ thcn

transfemed to Z{-r¡ell tlssue culture plates ( l E mm dlam) Costar

(Cambrldge, llass, U.S.Ê). Êt the bottom of each r¡ell had baen

placed a small clrcular glars coversllp that had prertously beeryr

r¡ashcd ln l0% rulphurlc acld folloucd by 7O% ethanol.

âpproxlmataþ { xlOË cells ucrc added to each uell. The celb uJËrÊ

then lncubatad for I hour at S?oC tn a 5% C0, incubator, l{on-

adherent cells uJÊre then r¡ashed off r¡ith l{ank's BSS.

2. Fc receptor assap-

a. Preparatlon of anttbod blood cells

Ê mlxturÊ r¡ra! made conslgtfng of 50 ¡rl packed r¡a¡hed

red blood cells, 200 pl antl-D tËnum (Commonrr¡ealth Serum

Laboratories, Sydnay, Australia) and 2 ml Hank'¡ 835. This mixture

r.r.ras placed in a S?oC tncubator for I hour. Durlng the incubation,

the tube r¡as invertad ¿vety 20 minutes. The red cells r¡ere then

rt0brËthÊd three tlmeg in gterlle normal sallnc and flnalþ resurpended

ln 25 ml RPÌ1|.

b. Fc recaFtor_ansa}¿

500 ul of thc anttbody coated erythrocytes Lrere added

to aach r¡ashed mwtocyte monolayer. Thls uas then incubatc d forI hour after uhlch each uell r¡as r¡ashed three ttme¡ r¡ith llank's

ESS to rÊmova non-bound erythrocytes. Thc cells LJæÊ flxed by

tha addltion of 39ã gluteraldchyde in llank'¡ 833 for 30 min atroom temparature (or 12 hrs at 4oE).

c.@The corcrsllps r¡JcFË stainçd ulth Elemsa staln. Êfter Etl

seconds the corerrllps uJÊrÊ uaghed ln uatcr and allor¡ed tostand ln dlstlllcd llz0 for 50 mln, the uatcr being changed êvery

It mlnutes. Th.y LrÊrÇ drlcd and mountcd. The mlcrorcoplc

appeåpånce of these prËpËrètlons shouJcd rsadlly ldcnflflable

adherent monocytrs to r¡hlch urcrÊ attached rad blood celb,

eithar touchfng or partlalþ or completaþ lngested. Sinca the erd

point in thls inrzestigation r¡as rcceptor attaclrmrnt, no attemptuas made to dlstlngulsh attaclrmrnt fronr conrplete phagocytosls.

For each corersllp flra hlgh pouJÇr flelds brcre counted or cnough

to guarantce at least 100 mørocyles and thc comespondlng

number of erythrocytes. Thlr uas thrn ÊxprÊæed as a bindlng

ratlo. The results for cach Àstay uJÊrê exprÊJsed as the Blndlng

lndax.

EFtdlng lndex = ratlo REC per mmoc¡,¡te uith ïAratlo REC per monocyte r¡ithout I'lA

3,

a.

Purlflad hlgh molecular ludght hyalurontc acid (Healon: J"lhJ =

>2x l0t) r¡as tha gift of Pharmacia, south Seas, sydney, Êustralia,

rtlThis r¡as exposed to the fcrrous ion autoxrdation orlygen radlcal

gancratfng system under condltlon¡ ldentical to thosc outllned fn

Chapten ll. Êrerage molecular rletghts of the depol¡rmerized

samples urËrÊ determlned by fractionaflng allquots upon

Sepharose {E-CL columns ( l0 x 0,5 cm) and the K6y of the peak

uronlc acld raluÊ rrtâl mcasured. The correrponding molecular

uleight r¡as determlned by referÊncÊ to data for poþsaccharldrs

glrcrr by FISHER [19?0]" The ferrous lon concentraüon and the

corrËsPonding åvÇrâgc molecular weight of the product are shourn

ln Table 5- l.

b. Enzl¡me exoosed hpaluronlc acld

Ê preparation of llcalon uJa! ênposed to testicular

hyaluronidase (Sqma Typ, lV-ï, EE 3.2.1.35), 0.1 mg/ml for S mins

at EO.C and the reactlen u¡as stopped by decreastng the pll to8.0 and dropplng the temperature to .loC. Thls r¡as then

fractlonated on ü Sephacryl 51000 column. Poolad fractions uJêre

dlaþzrd extenslveþ agalnst uater, ethanol prectpttated and then

drlad h g¡Sl¡C. The average molecular rrrclghts of pooled fraction

uscd uas detcrmlned by fractlonatlon on à Sepharosa {E-EL

column as dascrlbed preriousþ. Charactarlsflcs of these

prËparètlsns ara glran ln Table 5- l. ln thls âssày onþ ont

hyaluronlc acld concentratlon of 0.5 mg/ml uas used.

C, RESULTS

Using undegraded high molecular rrralght hyaluronic acid there

l¡rËs å dose dependent lnhib¡tlon of Fc receptor blndlng orer the

concerrtration rângÊ 0.2 - I mg/ml (Fig. 5-l).Êt a concentratlon of 0.3 mg/ml there urar a biphasic effect

upon Fc neceptor blndfng dependent upon molccular rrleight.

lnhlbltlsn of Fc receptor blndlng h,ar iËÉrt r¡ith prcparatlons r¡lth

an äreràge molecular ¡¡elght of I x l0ó and abo v¿ and stimulaflon

rl2

Table 5- | P1BLECULÊR hJEIGHTS 0F THE I'|YÊLURON|E ÊClD PREPÊRÊT|0l-15 USED

Itl TllE Fc RECEPTOR EIND|NB Ê55ÊYs

0xygen nedical depolymerized JIA

Fcnr^ou¡ ion Êrcragc l"lLJ

concentration(micromol¡r) (xl O6)

Enzymc degradcd HÊ

Poohd Êrcragc MLJ

frectiong(xt O'6)

050

r00500

r 000

505

2

t.t.0.

?1.51.00.50.2

HEÊLOf'|

11- 2E5r- 80101-il0

Frf¡uRE 5- | EFEEff OF TTIE CONCENTRÊTION OF H FIIÍìII ¡TOLEEULÊR

l¡JEIEFfT |'|YÊLUR0NIC ÊElD PREPÊRÊT|0N (HEÊLON -PIIRRÌ1ÊC|Ê) UP0N Fc RECEPTOR BINDINE.

Each data polnt = mÊân (t 3.D.) of trlplicate

datcrmlnatlons.

BINDING

INDEX

r.5

r.0

0.5

0 0.5

HYALURONIC ACID CONC ns/nl

I

rl4of Fc recaptor blndlng uJär tÊÉrì ln preparaflons ulth Ën ärÊragÊ

molecular uletght of 5x 105 and belor¡ (Flg. S-z). Thls effect ts rËen

t¡lth both o¡rygÉTt radlcal exposed and enzyme dtgestcd hyaluronic

acld"

D" DtSCUSSt0il

The Fc receptor blndlng àssày rystcm used ln this studydfrTered fnom the latex phagocytosts âsray studted by F0RREIIER

and EÊLÊZS [1980] becausa a specrftc receptor for the uptake olpoþstyrene latex rphercs has not bcen demonstrated. Latex ls

consldered to be crrgulfad ln a non-spËclflc manner porstbþ rclatedto lts hydrophoblclty [FR0ilnfìEfIT, I g?11, The rcsults of thts rtudylndlcate that recrptor medrated brndrng ts tnhtbtted by htgh

molecular rrlelght hyaluronrc acrd (>lxl0ê), t{ouercr at a lou

concertratlon (1.e., 0.3nrg/ml) of lor¡¡ molacular ualght hyaluronlc

acld, a paradoxlcal silmulaflon of Fc rcceptor bindlng lr erident.Thls l¡ ln accord ulth the flndlngs of Fomcster and Ealaz¡ forlatex phagocytosls, houevcr the molecular r.leight at r¡hich thisstimulatlon flrst becomes errdent rs sqnreu.rhat higher in our

system (i.e. 5x105 companed ulth gxl0{).

Depol¡rnrarlzatlon of hyaluronlc acld by both enz4aflc dtgerflon

and oxygan radlcal exposupÊ resulted in hyalurørlc acld

dagradatlon products that urcrÊ lnhlbltory ãt htgh hyaluronic acld

molecular rrrelght yet caused paradoxlcal gtimulafiør at louered

molecular rlelghts. Oxygen radlcal depol¡,rmerlzed hyaluronlc acld

caused a greatar dagree of sttmulaflon of blndlng lndcx but thlg

dld not raach statlsflcal slgnlflcancÊ.

The hyaluronlc acld concerrtraflon of 0.r mglml uas choser-r

after consldaraflon of Fomester and Earaz's data to glre maxlmal

¡tlmulatlon at the decraased molccular tr,leight and it ls partinent

Ftt¡uRE 5-Z EFFEET 0F ENZYHRïC (ao....t) ÊND 0XYEEN RRDIEIIL

DEP0LYI.'IERIZED (- ) |'|YÊLUR0NIE ÊClD UPON Fc RECEPT0R

BINDINE ÊT H CONCENTRRTION OF 0.5 IIE/T11.

Each data point = mean t S.D. of triplicate detenminations.

BINDING

INDEX

r.5

r.o

t.0

MOLECULAR WEIGHT

r.5 2-Oo.2 0.5

x loó

rr6to nota that at thig concentratlon a matrlx of hyaluronlc acid is

unllkaþ to form aspeclalþ at louertd molacular r¡eights.

Clcarly mor? extenslyc studles äpê requlred to probe the

mechanlsms lnvolved ln both the lnhlbltory and the stimulatory

lffccts dcmonstratad. llouevar lt is pantlnent to conslder sareral

rcctnt studles, ln conslderlng the lnhlbltory effect, Forrcster and

Ealazs concluded that alectrostaflc repulslon (latex, lgG, and

hyaluronlc acld are electronegaüre) ls unllkeþ to haye an effect

since sulphated gþcosamlnogþcans r¡hlch arÊ more electronegatire

are t¡lthout effect. Stmilarþ an hyaluronic acid tnduced membrane

pertubation ls unllkaþ since other membrane modtfying agents

r.,crÇ r¡lthout cffect. Th.y favorrd a starlc excluslon of latex

from the area of the cell surface thus decreaslng the proxinnity of

call and latcx. Êlternataly they speculated that latax beads and

lndard lgtì and complement coatad parUdes are hydrophoblc thus

maklng lt easlcr for the cell to effect phagocytosts. tlyaluronic

acld, betng hydrophillc r¡¡ould dacreasa this phagocytic adrantage.

Eoth of thcse explanatlons could appþ to our àssày. Stmilarþ,

starlc excluslon effects have been lmpltcatad ln the lnhlbitory

affect of hyaluronic acld on mttogen induced lyrnphocyte

stimulation.

rr?

E. SUtltlÊRY'

l'ligh molecular r.reight hyaluronic acid (l'lealon-Pharmacia) caused

a dose dependent inhibitlon of macrophage Fc receptor binding

betr¡een the concentrations of E,Z - I mg/ml. Êt a concentration

of 0.5 mg/ml both oìlyg?n radical depolymerized and enzymatically

degraded hyaluronic acid cäule an inhibition of Fc receptor binding

at molecular r¡eights of lxl06 and 1.5 x 106. Jlowever r¡hen the

molecular rrleight r¡a¡ reduced to 2 x 10s and 5 x I g= by axygen

radical degradation, ¡timulation of Fc receptor binding occurred.

Stimulation uas al¡o ¡ean r¡ith enzymatically degraded hyaluronic

acid r¡ith a molccular r.rleight of 2.5 x l[Js.

rt8

C}IRPTER YI

EXPOSURE OF CULTURED SYÎIOYIRL CELIS TO OXTGE]I RßDICRL

FLUXES. CYTOTOXICITY .R]ID EFFECT OII IIYRLURO]IIC RCID

PR0DUST|oil

119

R. iltrRoDUCTt0H

Tha studtes descrlbcd so far hare explored tha posslblllty that

oxygÊn radlcals, generatcd ln the lnflamed Jolnt, can act to

depolyrnerlze fully s¡.mtheslzed hyaluronlc acld. Houerer' the

flndlng of hyaluronlc acld of decreased arerage molecular u.leight ln

inflamed s¡morlal flulds may also be cxplalned by the synthasls of a

lcss pol¡merlzed product. The rynorlal cell rÊprÇlctlts a potcntlal

target for damagc by oxidants releascd by leucocytss as part of

the lnflarnrnratory procÊts. lt is porsible that a synovlocyte

damaged by oxldants may contlnue to producË a smaller molacular

rlelght hyaluronlc acld product.

The aim of thls expertmcnt uas to dctermlne the extent of

thc susceptlbltlty of synovlal cells to an cnzymatlcalþ gcnerated

oxygÉrt radlcal flux þ rltro. and to detarmlne the molecular rleight

charactælstlcs of the hyaluronlc acld produccd by synovlocytes

glrcrr a sub-lcthal expoturË to an oxygërl radical flux.

E" HETHoDS

l.@Eoylne s¡rnorzlal cellg rrJcre obtained by trypsln larage of borlne

metacarpophalangaal Jolnts accordlng to a modlflcatlon of the

method of FRASER and tlcCALL [1965l, Fre¡h bavtng^ hocks uJÊrÊ Ë

gift fronr the South Êustrallan lleat Corporation. trllthln { hours

of slaughter they urËra u¡ashed and the skln uas dlssectcd frea

from the mctacarpophalangeal joint. Ên area of ¡ubcutaneou¡

tissue brts rurabbed ulth 7E% alcohol and an I E gaugc needle

lnserted. ThË metcarpophalangeal Jolnt uas laraged u.lith tuo 50

ml rolumes of normal sallne to remoyÊ l',norlal fluld and slnorlal

fluld cells. {0 ml of 0.25 trypsln (tyPe lll from boytne Pancrtâs'

Slgma Chemlcal to., St. Louls, Hlssourl) in Êa++ and llg++ free

phosphate buffcred saline (Dulbecco's formula) uas then inJected.

r?'0

Êpproxtmateþ 25 ml uas therr uithdraunr lnto the syrtnge and the

remaindar uJas left h sltu for 15 mlnutes. The Jolnt rr.ras gcrrtly

malragÊd initially to ensure that the trypsln solution ent+ed

rêcassËs in the synorial lpacÊ. The jolnt r¡as then rer¡ashed r.¡¡ith

the conrplete rolume of the trypsln solutlon u¡hlch r¡as r¡lthdrarr¡n

and cantrlfugad at 1000 x g.

The cell pellet obtalned r¡as then rËruspênded in lo tlssue

culture medlum (see Êppendlx ll) supplcmented r¡ith 20fi foetal

calf serum and seedad lnto a plastlc 25 cm2 tlssue culture flask

(tìrand lsland Elologlcal Ëompanlr 6rand lsland, ller¡ York). The cells

urerÊ thcn grourn at 5?oC ¡n 5% carbon dloxlde. Êftar Z days

scatterçd adherent calls could be seÊn u.¡lth charactarlstic sharp

cytoplasmlc processes" The cells brËre then r¡ashed tr¡ice ulth

Dulbecco's phosphata buffered sallne (PEg) and the I o tlssue

culture madlum replaced. Thc medlum uag changed ln thls mannËr

tr¡lce ucekþ. Tha prhnary culture urär maintained until the cells

had rcached confluÉncÊ (usually about l0- l{ days). Th.y urerÊ

then subcultured by uashlng tr¡lce in Dulbecco's PES and then

incubatlng ät 5?oC for 5 mln in 2-5 ml of O.25% trypsfn ¡n PES.

lrJhen lt u¡¡as obsenved that the cells had separatcd, l0 ml oî 2o

tlgsua culture medlum (see Appendlx ll), supplementad r¡¡th l0%

fetal calf sarum uJâr added. Thls ¡¡as than dlvided into tr¡o on

thrce allquots r¡hlch urêrÊ then added to fresh tlssue culture

flasks. Êll subsequcnt cultures urerÊ grourn ln Zo tlssue culture

medlum unlesg otherr¡lre rpecifled. Cells LJerË harrestad for the

chromium release arrËy betu¡cerr the Zttd and ?th subcultures. ln

thls r.r.lay a relatlveþ hornogÉ'nout populatlon of synoviacytss r¡¡ith

a flbroblast-llke åppËaräncË urÇre obtalncd IFRÊSER Êt å1., l9?31,

t2l

Z.

S¡morlal cells r¡crr haryestad for tht chromlum relcase

assåy by subculturlng ln the mânnÊr dascrlbed abovc. l0l calls (ln

1.5 ml tlssue culture medlum) þrËrÊ then added to aach urell of a

2{ r¡cll tlsrue culture plate (Eostar, lE nrm dlametcr u,relb,

Cambrldge, Hass.). Thegc urêrÊ thcn lncubated at 3?oC rî 5% Egz

for tuo days r¡hsn a confluerrt layer of cells could be seÉî at the

bottom of each u¡ell. Each ¡¡cll uas then u,¡ashed tu¡lce r¡lth

Dulbecco's PBS. Carc uas takcn durlng ulashlng to pofnt the tip of

the pasteur plpette at a rlngle polnt on the circumferencÊ of the

uall ln order to dlsturb the cell laycr as little as posrlble. 0ne-

tenth mllllllter of trBr (speclflc actltzlty = 50 ¡.rCllml, Amersham,

Êustralla) r¡as added to each r¡cll and lncubated for I hour at5?oC. Thc cell laycrs bJËrË thcî uashrd three tlmes ln Dulbecco'¡

PBS and the reagcnts for cach cxpcrlmcnt uerc addcd ln a total

rolume of I ml. Expcrlrn+nts uJerc performcd ln trlpllcata u¡lth

+ve controls (20?ã acctlc acld) and -rc controls (buffer) ln each

platc.

Êftar lncubatlon for the approprlatc tlme, the cell layerr

urÊrÊ lnspected to datcrmlne that they uJcrr stlll adherÉnt. Then

0.5 ml of thc supÊrîËtant (3) uas arplrated and placed in a poþ-

carbonate tube for countfng. The remalnder (R) uag then

asplrated and the cell laysrs uJcrÊ remored by frypslnlzatlon x2

r¡ith 0.5% trypsln ln PEg and a flnal r¡ash ulth uater tohypotonically lysc any rasldual cells. These urashtngs rdcrÊ then

pooled, the volumes equallzcd and the tuber uJerÊ counted on a

Roche Eammamatlc gamma countar.

Perce¡tagÇ llCr rclearÊ urâr obtalned by the formula

ß 3t Cr releasc = ?3 X 100

R+3

LZ2

Baclrgroun d %3lÊr releate k as determlned fronr the -ve control

and r¡as subtracted from the gros s % 3lCr release. The mean

and standard derzlatlon of trlpllcate determlnatlsnr uJÊre then

calculated.

3. gry

The oxygcn radlcal geteratlng systsm used t¡¡as the XOlllX

system as prerlously outllned (sae Ehapter ll). Protcoþtlc

contamlnatlon of the xanthlne oxldase prÊprratlons uJËrÊ

decrcased accordlng to the method of IìREENhJALD and NOY I 19?91

by Scphadcx [ì100 chrornatography, Bccauge all thc proteare

actlrlty uas not conrpletely rsmovcd by thls method and lndced

afforts to complrtely rernorÇ all protea¡e actlvlty wlll rcsult ln

loss of xanthlne oxldare actlrlty IBHTES, L0trfT?lER and llÊtJDLEY,

l9B'tI, the axpcrlmcrrts urrrÊ camled out ln the presÉncê of l%foetal calf sarum ln order to ¡uamp rerldual protc ase actlvlt,y

(ree rasults sactlon).

{"Ë

llyaluronic acld productlon uräs assesged urlng synovial cells

grouJn to confluÊïrcË ln 25cmz tlssue culture ftagks. After uashing

r¡lth Dulbecco's P85, 5 ml¡ llam's F l0 tlssue culture medlum

contalnlng 50 ¡rfi 3H-gluco¡amlne hydrochlorlde wac added. Thls

r¡as then lncubatad for 'lE hours before the supernatant uas

collccted aftcr gentla urashlng of thr cell laycr t¡lth th;rupernËtant.

Tha collacted rupËrîatant uas therr cmrcerrtrated ln a

Mllllpore 25 mm stlrred cell orer a Pclllcon F[-EC flltcr.

Hpproxlmateþ 2.5 ml of concentratad superîatant ua¡ thcn

appllad to a Sapharose'iE-CL colunrn (100 x I cm) and eluted r¡lth

'l 11 guanldlnlum chlorlde. Slnce the peak of radloactlvity identifled

as hyaluronic acld dld not changc slgnlflcantþ under assoclative or

rz3

dissoclatirc condltlons rubsequent clutlons u¡ere performed urlng

0.5 H NaÊc pll 5.8. Fractlons of I ml uere collected. 5 ,¡l of cach

fraction urÊpÊ thc.n added to 2 ml scintillation fluid and counted in

a Eeckrnann llquld scfntlllatlon countËr.

5. Cellulosc acctate gel electrophoreír

The hyalurwrlc acld containing pcaks after fractløration

brÊrÊ ldentlflcd by applylng 5 ¡rl of the peak fractlon to cellulose

acetate strips. Elactrophoresls uJät conducted a¡ descrlbed tn

Chapter lll. Êfter electrophorerls, standard hyalursnlc acid

samples bJere read by Êlclan blue staining t¡hereas radlolabelled

samples uJerÊ rcad by cuttlng the cellulosc acctate strlp nto ?

mm sectlons fronr the potnt of orlgln. Each strlp r¡as addcd to acountlng ylal ulth 2 ml of llquld sclntlllatlon fluld and counted.

C."RESULTs

l. f,I/HX induccd c].rtotoxlclt¡¿

The lnltlal part of thls experlment ua¡ to detcrmlne the

leyels of oxygcn radlcal flux that produced a cytotoxlc effect

uslng s¡morlal calls as targets. Flgure 6- I shor¡s thr resultg of a

tlma coupre experlmcnt ln ¡¡hlch I cmz uells of confluent

monolayar¡ of syrnavlocylss uJÊrÊ expored to l0 mU/ml xanthine

oxldase ln the prËsrncÇ of 2 mll hypoxanthlne. Êfter E hrs a

cytotoxic efract bJät mênfestcd by the releagc of stCr into the

medlum. Subsequmtly all measurements in the stf;r release assay

uJerÊ made at 12 hrs.

Eecause xanthlne oxldase prËparatlørs contaln tomÊ

proteoþtic actirltyr ln splte of chromatographlc puriflcation,

foetal calf serum uJas added to the mlxture to suamp this

actlrlty. Flgure E-Z shor¡s the results of an expertment ustng

progrËrrlvely lncreaslng ümounts of foetal calf serum up to 50ß.

Êbore l0% foetal calf sÊrum, the F s'Erreleasa is brought back

FIÍ¡URE 6-I KINETICS OF SICr RELEÊSE FROI1 OXYBEN RÊDIEÊL EXPOSED ÊND

UNEXPOSED SYNOI/ I OCYTES.

1

- ): X0 = l0 mU/ml, llx = Z ml.l

( .."".... )¡ llx (2 mll) alone"

(Data points rÊprÊsÊnt mean I 5.0.)

FIÍiURE 6-2 EFFEET OF FOETÊL CÊLF SERUI'I ETNEENTRÊTION ON XO/IIX

INDUEED SICT RELEHSE.

Hatchad bars: X0 = 5 mU/ml, HX = Z mH

Elank bars: flX onþ

r00

PERCENT

80

CHROMIUM

ó0

RELEASE40

TIME COURSE FOR CHROMIUM RETEASE AFTER EXPOSURE

TO AN OXYRADICAL FIUX

20

.¡.

o

r00

PERCENT

80

CHROMIUM

óo

RELEASE40

20

2 4 8

T¡ME

l0

HOURS

E FFECT OF FOETAT CAIF SERUM CONCENTRATION

oN xo/nx INDUcED cHRoMluM RELEASE

hotched bor: xo hx

blonk: control

6

Nofcs I%fcs 2%1cs 5%Ícs lO%fcs 50%fcs

rz6

to control lerels, using an xanthine oxldarË concËntration of

SrnU/ml. Subsequently l% foetd calf sÊrum r¡as added to the

reactlon mlxture r¡hen the cells urcrc exposed to the X0/HX

systenr.

Figure E-3 shor¡s thc affact of superoxlde dlsmutasË on XO/HX

lnduccd slCr releasc. Êt a superoxlde dlsmutasa concentratlon of

?5 U/ml, no protcctlrc effcct uJar rËGn ove? a xanthlne oxldage

concentratlon rangË oJ 5 to 100 mU/ml. Thcse l¿v¿ls of ruperoxtda

dlsmutasc hare becn rhoum to be protectlv¿ ln the XO/tlX induced

hyaluronlc acld degradatlon rystrm and X0/acetaldchydc lnduced

erythrocyte lysls (KELL0CB and FRlD0ylCll, lg11'), Catalasc, on the

other hand uras protcctlra at a concartratlon of 50 U/ml (Figure

B-'l)" Ilannltol had a mlnor protectlve eflect at lor¡ xanthlne

oxldase concrntratlons (Flgure 6-5).

Eased on the lndlcation from the aboye expertments thathydrogen paroxide r¡¡as the agent necËrsapy for demonstration of

a cytotoxlc effcct, the effect of glucose/glucose oxidasÊ r¡Jas

examlned. This É'nzlrmt producas the dlralent reduction of oxygen

to I'1r0, ulthout productlon of the suparoxlde anlon [N|L350N, PICK

and ERÊY, l3E9l. Ê cytotoxlc affact r¡as raadlþ damonstrated ata glucosa oxldasË concentratlons of 50 - 500 mU/ml. Thls r¡as

conrpletely lnhlblted by the addltlon of catalase at aconcentratlon of 50 U/ml (Ftgura E-6).

2. Effact of oxpgen radlc

Predr¡rllcD

Tha next phase ln thls serlcs of exper¡nients uag todetrrmlne the effect of an Ëxog€'nousþ applled oxygen radlcal flux

upon hyaluronlc acld productlon by sytavlal celb ln culture. Ì1on-

dlseased boyine synovlal cells, ag used in the cytotoxlclty studies,

uJerÊ incubated rrlith 3H-glucosamine hydrochloride. lnltlal

FIEURE E-3 EFFEET OF SUPEROXIDE DISHUTÊSE ON XO/I.IX INDUEED 5IËT

RELEÊSE.*

500 concentration = ?5 units/ml

FIfìURE E-{ EFFECT OF EÊTÊLÊSE OH XO/IIX INDUCED 5IC' RELEASE.,I

Eatalase concentration = 50 units/ml

FIfìURE 6.5 EFFEET OF NÊNNITOL ON XO/HX INDUEED 5IE' RELEÊsE.*

* The error bars rÉprÊtÊnt the mÊan t 5.0. of triplicate

determination¡. Th¿ % 5l Cr relaase in the +re controls in

these experiments = 91.2 ! 1.1.

100

PERCENT80

CHROMIUM

ó0

REIEASE

10

EFFECT OF SOD ON XØX INDUCED CHROMIUM RELEASE

soDNO SOD

05 20 50

20

100

PERCENT

80

CHROMIUM

óo

RELEASE10

20

XANTHINE OXIDASE CONC

,( lO-3 units/ml

EFFECT OF CATALASE ON XqhX TNDUCED CHROMIUM RELEASE

ro0

r0005 20

CATATASE

50

XANTHINE OXIDASE CONC

, lO-3 units/ml

CH RETEASE

NO MANNIÍOI

2OmM MANNIIOL

IOOmM MANNIIOL

EFFECT OF MANNTTOT ON XqfiX lNpUCEpr00

PERCENT

80

CHROMIUM

ó0

RETEASE10

5

XANTHINE OXIDASE

. lO-3 units/ml

20

coNc.

20

1

I

FtfìuRE 6-6 EFFEET 0F IìLUEosE/BLUcoSE oxtDÊsE 0N Slcr RELEÊSE.

[ìlucose oxidase = 50-500 mU/ml

[ìlucose = 8"5 mH

Catalase = 50 units/ml

100

PERCE NT

80

CHROMIUM

óo

RELE AS E

40

EFFEC T OF

ON

GLU COSE/GLUCOSE OXIDASE

CHROMIUM RELEASE

Il

20CATALAS E

GLUCOSE OXIDASE

unity'ml

50510

Peak Ê is also completeþ susceptible to digestionr,tlth hyaluronate -lyase from streptomyceshyelur.onolyticus (d¡t¡ not ¡hor^m).

rz8

ÊxpËrlmÉrlts, shorrm ln Flgurc E-? and E-8, rr.lere performed toldentlfy hyaluronlc acld tn thc call culture mpdtum. Thc

suPËrnätants fronr tr¡ò 25 cmz tl¡ruc culture flasks contalnlng

confluent synorlocytes brËrÊ pooled after a .18 hr laballing period.

Êftar concentratlon, these uJÊrÊ fracflonated on ä Sepharose 4E-

ÊL column under dlssoclaflre condlflons (i.e" { Ì1 guanidtnium

chloride)' The chromatggraphtc proftle demonstrated 2 peaks oflH-glucosamlne containlng matarial. Tha higher molecular peak

(peal A) elutad niar Po and a louer molecular rreight peak (peak

E) cluted in the lncluded rolume. Electrophoreüc separaüon of

material from each peak uJàs pËrformad on cellulosa acetate

strlps, (Flgure E-B), and hyalurontc actd r¡as ldenttfled tn paak Ê

but not peak E"

Eorlne s¡rnorlal cells thtrefore produced rslalclvely small

quantltlas of a hlgh molccular rlelght hyaluronlc acld that elutcr

at the Po of a Sepharo!Ç {E-CL colunn undcr dlsrociaflre

condltlons.

l¡Je next exposed confluÉît monolaye?s of sgtnovlocytes h Zs

cmz tlssuc culture flasks to the xEltlx axygeî radical generaüng

rystem. The xanthlne oxldasÊ concentratlon used uas 5 mU/ml ln

the preseïcÊ of 2 ml-l hypoxanthlne. Thts resutcá rn conrplete

abolitlon of both peaks and all the radtoactivity eluted near V¡

(Flgure 6-9).

Subsequently confluent monolayerr uJerÊ exposed to a xanthine

oxldase concerrtratlon of I mU/ml. Flgure 6-10 ghours the rcsults

one such expcrlment. Exposure to xanthine oxldagç alsne shok s

the prescncÊ of tr¡o peaks, a¡ ls rcÉrì ulth unËxpoicd cclls,

houerer ln the prËtcncË of 2 ml1 hypoxanthlne the htgher

molecular r.rlelght peak dlrappearr. The proflles rhoun ln Flgure E-

l0 dúrcr from those shoum in Ftgure 6-? in that the cell products

FIT¡URE E-? SEPI.IÊRI]SE {B-CL EI.IRONÊTOERRPI.IY IlF T}IE SUPERNÊTANTs OF

Stt-ctucûsÊÌ1rNE LÊEELLED EoprNE syNoptÊL EELLS.

Borine synorrial cells, grouJn to confluence in tu.ro ?5 cmZ

. tissue culture flasks. Th.y uJene then cultured in routine

tissue culture medium (see Êppendix lll) containing 50 ¡.rCi

SH-glu.osamine for 2{ hrs. The supernatants uJere then

pooled and concantratad, and the concentnate applied to aSepharose 4E-CL column r¡hich r¡as eluted r¡ith { H guanidine

chloride. Tr.,-ro peaks of glucosamine containing material are

apparent, a high molecular rrreight peak (Ê) and a lou.ren

molecular r.rraight paak (B).

FIEURE 6-fl CELLULOSE ÊEETÊTE GEL ELEETROPIIORESIS OF T1ÊTERIÊL FROTI

PEÊKS Ê ÊND E"

5 ¡.rl of fraction 2E 19- <) and 5 pl from fraction 5E (o"<1

from the material demsnstrated in Figure E-7 were

electrophoresed under conditions outlined in Ëhapter lll. Ê

standard of hyaluronic acid r¡as run concurnently and stained

r¡ith Êlcian Elue. The area of Êlcian Blue staining is indicated

at the bottom of the figure. Radioactirre hyaluronic acid r¡as

identified in peak A only.

3x- GTUCOSAMINE INCORPORATION BY UNEXPOSED SYNOVIAL CELLS

DPM 15,000

50¡l

10,000

5,000

SEPHAROSE 48 - CL CHROMATOGRAPHY

Vg

tPeak A

Peak B

a

0 10 20 30 40 50 60 70

CELLULOSE ACETATE CEL ELECTROPHORESIS OF HATERIALFROT PEAKS A&B FROT FIGURE I

DPil 3002mm

r-< Fraction 20 (F¡gue 1)

O€ Fract¡on 56 (FigurelJ

200

100

12 14 16 18 20

vt

t

B0 90 100

FRACTION NUiIBER

oI

I

I

I

I

I

I

I

I

II

I

o24TILLIMETERS FROT T}IE ORIGIN +VE

lAREA OF ALCIAN ELUE SIA¡IIIIIGFOR AI{ HYALUßOIIIC ACID STAÍ'IDARD

0

-VE

FI6URE 6-9 SEPIIÊROsE .18-CL EIiROHÊTOIìRÊPI'IY OF SYNOPIÊL CELL

SUPERNÊTÊNT:' LÊBELLED ÊFTER EXPOSURE TT] TTIE XÊNT}IINE

OXIDÊSE/HYPOXÊNfiINE OXYBEN RÊDICÊL IìENERÊTING SYSTEM.

Eovine s¡morial cells, gpouJn to confluÊncÊ in tr¡o ?5 cm2

tlssuc culture flasks uJÊFÊ axposrd to xanthlna oxidase 5

mU/ml X0 and hypoxanthinc 2 mll in Dulbecco's PES for 2 hours.

Thry uJÊrÊ then l¡aghed and cultured in routlne tissue culture

medium containing 50 ¡.rEi 5n-glucogamlne for 24 hrs. The

supÊrîätantg uJÊpÊ thcn pooled and concmtrated, and the

concentrate appliad to a Sepha?ol,e'{E-EL colurnn uhich r¡as

eluted r¡ith 4 l'1 guanidine chloride.

3H - GLUCOSAUINE INCORFORATION BY SYNOVIAL CELLS EXPOSED TO5 mU/ml XANTHINE OXIDASE

DPM 3OOO SEPI-IAROSE 48 - CL CHROMATOGRAPHY50¡¡l

Vg

IYt

I

2000

1000

0 10 20 30 40 50 60 70 80 90 100 110

FRACTION NUMBER

FIGURE E-IO SEPIIÊROSE +E-EL EI.IROI'IRTOIìRÊPI'IY FROH SINÍìLE E5 CHZ

TISSUE CULTURE FLÊSKS CONTAININE CONFLUENT SYNOTIÊL

CELLS ÊFTER EXPOSURE TO XÊNT}IINE OXIDÊSE RLIINE ÊND

XÊNTTI I NE OX I DÊSE /IIYPOXHNTTII NE.

The panel shor¡s results from a tissue cultura flask thatrdas Êxposed to I mU/ml X0 in Dulbecco's PE5 before a 21 hour

labelllng period uith 3tl-glu.osamhe as outlined in the legend

to Fig. 6-8. The bottom panel shous the ruper'natant fronr a

flask exposed to I mU/ml X0 in the prÊrencÊ of 2 ml"î tlX. The

columns urÊrË eluted r¡ith 0.5 11 NaÊc' pll 5.6. The l/s r¡as at

fraction Z I and /1 at fraction 93.

Hyaluronic acid uas not identifled by cellulose acetate gel

electrophoresis in peak E ln either profllc.

3H - GLucosAMtNE rNcoRpoRATroN By syNovtAl cELLsEXPOSED TO 1 mU /mI XANTHINE OXIDASE

-

DPM1o ¡l

300

250'

200

150

100

50

0

400

350

300

2æ

200

150

100

50

0

XO ALONE

XO/HX

10 20 30 40 50 60 70 80 90 100

FRACTION NUMBER

r32

fractionated arc the contents of one 25 ct¡tz tiscue culture flask

that has been labelled for 21 hrs. ln addition, fractlonation uras

performed under associatlre conditions (1.e. 0.51'{ l'la Ac, pll 5.8).

Thls has resulted ln moremcnt of peak E touards Vo suggetting

that the componmts of peak E hare the ablllty to âggrÊgate and

that thls ls not af?cctad by exposurc of the cell to the full X0/HX

system ln thc mËînËr dcgcrlbed. Efforts to ldcntíy lche Prclc-ncË

of hyaluronlc acld, by gd elactrophorcslr, ln thc rlnglc peak of thc

proflla sÊÉ'n tn the X0/tlX cxposed cell rupernatantl u,crÊ

unsuccÊrsful (data not shor.m).

The flnal serles of experlnrtnts, shorrnr ln Flgure E- I I ' LJÊre

deslgned to examlne thc effect of superoxlde dlgmuta¡e and

catalase in thls systrm" The crrdltlons urÊrÇ identical to those

descrlbed above. Suproxlde dlsmutase added at a concsntration

of 100 U/ml had no ef?ect, houerer ulth the addltion of 100 U/ml

catalase, tulo peaks uJGrÊ agaln erzldcnt after fractlonation.

D. DlSCUSSloll

The cytotoxlclty studles descrlbcd abore demonstrate that

catalase protects adhercnl s¡novlocytes from the cfïects of an

çrtraccllular oxygcn radlcal generatlng rysttm. Superoxlde

dlsmutasc r¡lll renrorc 02-r to a dcgrec that ulll lnhlblt a Fenton

rcactlon and thus hydroxyl radlcal productlon. Thc flnd¡ng that

superoxlde dlsmutasc uag not effcctlve ln thls tystsm uould

suggest that hydroxyl radlcal, generated extracellularly, uas not

nÊcÊrtäFy îor the lethal ererrt to occur. The fact that mannltol,

å icèrÊrrger of the hydroxyl radlcal, u.ras minimalþ effectlre r¡ould

support this ffndlng. The toxicity oî the glucoselglucose oxidase

system and lts ablatlon by catalase, conflrms the ablllty of

hydrogen peroxlder !ËhËpâted extracellularly, to act as a

cytotoxlc agent.

FIEURE E-I I EFFEET OF SUPEROXIDE DIST1IJTÊSE ÊND EÊTÊLÊSE ON XÊNTHINE

OXIDÊSE/I.IYPIIXÊNTTIINE INDUCED SUPPRESSION OF HYÊLURONIC

ÊEID PRODUETION BY CULTURED EOPINE SYNOPIAL CELLS.

This figure shous a series of profiles, each rcprÇsenting the

radloactlvll.y aften fractlonation of the concentrated

rupÊfiìatants of ringle 25 cnß tlg¡uc culture flaskr. The

elutlon buffer r¡ag 0.5 11 NaHc, pl'l 5.8.

The top panel rÇprÊrentg the cmrtrol l.e. unÊxposed cells.

Thc gecond shor¡r expolure to tha X0/l'lX system (X0, I mU/ml;

I'lX, 2 mH).

The third panel ¡hous the effect of the addltiør of 100 U/ml

500.

The fourth panel shouc the effect of the additim of 100 U/ml

catalase"

The tr¡o primary peaks as illustrated in Fig. 6-? are

apparerrt. Exposure to the X0/l'lX system result¡ in the

dlsappearance of the highar molecular r.ueight llÊ containing

peak uhlch is partialþ rertored by catala¡e but not by S0D.

500

DPM 050 rul

CONTROL

VO400t

300

200

100

500 XO/HX

400

300

200

100

0

400

200

500 SOD100 /ml

300

100

500 CATALASE

400 100 /ml

300

200

100

30 40 50

FRACTION NUMBER

0

0 20

r34

llydrogen peroxide ls a small and unchargcd molçculc and

rdatlrely stabla r¡hen cornparËd to thc hydroxyl radical

I'lydrogen peroxide may therefore be exptcted to be abla todiffuse auJày from lts extracallular slte of producflør and crors

the cell membrane. Thi¡ r¡lll raise intra-cellular hydrogcrr peroxide

lerzels and lf enough hydrogen peroxlde ls generatad, the lntra-cellular antloxldant defensÊs urould ba oyeruhelmed" I'lydrogen

peroxide uould then be frea, not only to madlate intnacellular

damage pËr lrÊr but also to fnteract l¡lth other radlcal specles

that arÊ gËnÊrated as by-products of ccll metabolism.

Suparoxidc anlsn for example, is genaratad by a number of

lntraccllular procersÊs. lf endogcnously produced superoxide anion

ls not adequateþ scaranged ln a sltuatlon of oxldatlre strass, ltcould lnteract '¡lth hydrogen peroxlde to lcad to hydroxyl radlcal

productlsn rla a llabtr-hJelss reactlsn. The prurc'nce of

tntracellular lrwr rr¡ould cåtalyse thls rcactlon (1.e. al a Feryttsn

reactlon). Erldence that supports thls mechanlsm has recerrtþ

been produced by SCI'IRÊUFSTHTTER gÈ el. [ | 985] uho hare recently

documented the depletlon of lntracellular l1ÊD and glutathione in a

cell linc exposed to hydrogÉn pcroxide. Thls lndicateg a depletion

of the capaclty of the glutathlone/glutathlone peroxidase system

to act as an oxldant defence mechanlsm.

The addition of ruperoxlde dismutase and catalase to tha

extracellular medlum ls unllkely to alter lntracellular superoxlde

dlsmutasc or catalage lcrels due to the size of thesc molecules.

llannltol, hoularer ls small and unchargcd and therefore likeþ tocrors the cell menrbrane. lt¡ mlnor protectlve ¿ff¿ct mËy be due

to hydroxyl radlcal rcar+rrglng at lntracallular sltcg. Also houeve?

its mlnor protactlve ;ffect-, tËÉrt at high conccntratlont, may be

slo

135

due to non-tpËciflc scayenging since so-called oxygÊn radlcal

tcêrffigers trÊ notorlousþ non-speciflc.

Chrsrnlum release ls a manlfastatlon of cell þsls slnce

lntracellular chromlum ls bound to protetns that ulll onþ be

released r¡lth loss of ccll mambrana lntegrlty [R0tlAl, l9E9]. Êell

menrbranÊr år? partlcularþ susceptlbla to the deleterlous efrects

of oxldants through thc procËss of llpld peroxidatlon. Nelthcr of

thesc obseryatlons ln themsalyes indlcate ¡¡hather mambrane

damage ls a I o or 2o erent in oxldant inducad cell damage.

llouever tha sansltlrlty of hyaluronic acld production to lor¡ lerels

of X0/HX r¡ould suggest call membrana damage uràs ån earþ

eve,nt ln oxldant damage" PREHI1 tlgB{l and PtllLlPSON and

SCtlURRfZ I l98ll hare demonstratad that hyaluronlc acld

synthatase is situatad on the lntarnal surface of the call

membrane and nerr.lly syntheslzed hyaluronlc acld ls secrated

dlrectly Ínto the extracellular spacÊ uhereas other

gþcosamlnoglycans arÊ s¡mtheslzcd ln the cytoplasm. Thercfore

the obsarratlon that ËxpoturÊ of s¡novlocytes to I mU xanthine

oxldase ln thc prËsÉ.ncÊ of hypoxanthlne rcsult¡ ln the lo¡s of

hyaluronata productlon, but not the productlon of other Il-

acetylglucosaminc contalnlng products, ruggcstg earþ membrane

damage at a stage r¡hctr cytoplasmlc conrpmeîtg are still

functlonal.

PUflRl0 Êt il. t l gEZl hare reported that synoviocytes groun

from patiants ulth rhcumatoid arthritis ulll produce lou molecular

uelght hyalunonlc acid uhen establlshed in culture. Synthesis of

such a product fi vlvo lg an alternatlre explanatlon for the flndfng

of decreased arerägÊ molecular rrlelght hyaluronlc acid found tn

lnflammatory arthropathles. 0ur flndlng that expoture tosublathal lerals of X0/HX rçsults in cessation of hyaluronic acid

r36

productlon rËthÊr than s¡mtheslg of a smalltr product uould

suggest that oxldant Ê,(posurÊ can be dlscounted a¡ a cause for

such an alteratlon ln synthesir. lt ls of intercst to note that

EATES, L0lrJTtlER and l'lÊllDLEY [19E5] hare also reported thatchondrocytes cxposad to sublethal lerels of oxygan radlcals uill

decrease the amount of protcogþcan produced, but r¡lll not

cffcct the glze of the proteogþcan produced by these calls.

E" SU]I}IRRY

Eorlnc s¡novlocyta¡ ln culture êrÊ tusceptlble to oxygen

radlcalg pnoduccd by X0/llX at mcäturcd by'lCr releage.

Êdnrlnlstratlon of catalasa uJat protactlre u¡herËêl luPËroxide

dlsmutasê uJât tneffectlvc and mannltol had a minlmal efl'ect. This

lndlcatr¡ that hydrogcn perorlde ls thc ag+nt nccattËPy for

toxlclty. Sublethal oxygcn radlcal fluxes ulll result ln ccssatlon of

hyaluronlc acld synthesls evsn I¡hct the r¡mthsglr of othcr lll-

glucosamlne contalnlng products contlnucg. l{o lor¡ molecular

r.rleight hyaluronic acld products could bc detectad in the

superîatants of oxygen radlcal cxposcd cells.

r3?

C}IRPTER YI I

Dtscu35toil

t3E

fi. lü vlTRo oxYtEil RÊDtcRL HDUCED HTRLUR0ilTC RC|D

DEPOLYIIER I ZÊTI O]I.

l. llolecular r¡clght studlas.

Tha shnple chemlcal structure of hyaluronlc acld belles lts

complax phyrlco-chemlcal propartles and blologlcal roles. Thasa

propartlÊ! ËrÊ duc to lts macromolecular structure ¡¡hlch

depends upon both lnter- and lntra- chaln interactlon¡ and r¡hich

lcad to thc formatlon of a matrlx ln âqucoul solutltrtg of

sufflclant contentratlon. lt ls also âppârËnt from thc data

presented in this thesis, that a consideration of macromolecular

structure is necersåry bafore an hypothasls of hyaluronic acid

degradation by oxygÉï radlcals can be formulated.

The findlngs that Émarge from the studles dme ør the

molacular r.rreight changes after oxygÉyr radical lnduced

depolymerization and r¡hich must be reconcllcd lnto ruch an

hypothesls are:

l. The smallast dapol¡merlzatlon product lr 104 molccular

r.lclght.

Z. Thera ls a rapld progragslon from large molccular ulclght

hyaluronlc acld to the gmallegt rlze dcgradatlon product.

3. Ê dagrce of rcpol¡merlzatlon of degradatlon products ls

probable.

The flrst flndfng ls the most clear cut and may be accounted

for by a number of depol,,¡merlzation mechanlsms, Ìlost obriou¡

are those mechanlsms that require that there be a site along

the hyaluronic acid molecule that ir particularþ susceptiblc toclearage by oxygen radicab.

Perhaps the most llkely propcrty of hyaluronic acid that¡¡¡ould lead to a serles of susceptible sites along its length is its

t89

tcrtlary structurc ln aquËoul solutlon. Ês outlfncd in Chapter I'

hyaluronate, in aqueou¡ solutlon, ls bellerued to adopt random coll

best characterlzed alt a left-handed hellx. The random nature of

the colling of the hyaluronate molecule ls one feature that may

lead to GxporurË (or increased accc¡gabttity) of a glycosidlc lÍnkage

ts the surroundlng cnrzironm?nt. This in turn could make such a

linkage susceptlble to oxldatirze damagc" Êlternately structural

cEnstralnts may be imposed upon the tertlary structura of a

hyaluronic acld molccule in solutlon by lts interactlon ulth other

hyaluronate molcculcr ln the rämË rolutlon. Structural studics

IhJ|NTER, Sl'llTTl and ÊRi{0TT, l9?5J hdlcate a degrec of stffneæ of

the hyaluronata molacule ln aqueous solution, clearþ then the

formatlon of a matrlx ln s.olutlon r¡ith so mäny "ttiff' moleculas

r¡ould cåurË lrregularitles,fcrtiary structura to occuP. l,rlhat is

claar fnom our studles ls that these susceptible sites must be a

conslderable distäncÊ apart to account for the smallest moleculan

r.lelght degradàtion products harlng.a molecular rr¡cight of 101

(i.e. approximateþ 50 saccharldes)" Thls hypothesis predlcts that

an hyaluronate chain of lesser length is not long enough to adopt

the tertlary structure that leads to the formatløt of a

susceptlble slte.

The chclatlon of lron ls clearly a critlcal factor in all the

oxygÉî radlcal reactlon¡ utlllzed in this studyt and manl E vlvo

reactlons lnrolrlng radlcals (gee Chapter l). lt may be that by

rlrtue of lts tartlary gtructurc that hyaluronlc acid can chelate

lror or another metal lør ln such a r¡Jäy that the gcneratlon of

tha hydroxyl radlcal occurs ln close proxlmlty to the hyaluronate

chaln. lt ls claar nouJ from a numbar of studles that hydroxyl

radlcal productlon ls nÊcÊssäry for hyaluronic acid

r40

depolymerlzatlon [EffT:t and CLELHND' I gEZ; HBFl"lÊNl{ ÊND SCHHUT'

l9E0l. The extremely high rcactirzlty of the hydroxyl radical

dlctates that lt must bc generated close to its target ¡ite since

It so lndlsrfminately neactire [FEE and PÊLEt{Tlt'lE' 197?1" This

sltuatlon r¡ould thereforË rÊPrcleltt a form of "site-specific

Fenton reaction" i.Ê. the site of chelation of a metal ion capable

of catalysing a Fenton neaction determines r¡hene hydroxyl radical

damage occurs. ln tha case of hyaluronic acld this r¡ould be close

to tha Fl-S glycosidic linkage as d¡scussed ln Ehaptar lP. This

hypothesls t¡ould thus neconcile the high and non-speciflc

reacür¡ity of the hydroxyl radlcal u¡lth a specific alycosidlc bond

clearage"

Hrn lmpllcation of theEe "susceptlble site" hypothesËs ls that

they may prorzide an explanatlon for the second statcd flnding of

the molecular r.lelght studles. Depolymerization proceedlng from

"susceptlble sltc" to "lulceptlble sitÊ", ln an Íneruitable chaln

reaction r¡ould regult in the obscrration of a rapid progresssion

of largc molecular r.leight hyalunonic acid to tha smallest size

degradation product r¡ith a ralatire paucity of intermediate sized

materlal.

0ther hypothetÊl arË porslbla. The gencration of lor¡er

molecular r.r,relght hyaluronic acld may interferc uith further

orlygÉTt rad¡cal production. This may occur lf the small molecular

rrletght products act as hlghly efflclent oxygÉTt radlcal lcarengers.

Alternately hyaluronic acid depolymerlzation products rnay

interfere uith chelator-rnetal bindingr thus prerenting further

hydroxyl radical generatlon. The chelator may be EDTÊ' or' äs

suggested prerlously, hyaluronlc acld ltself.

r4l

Finally the ruggestion that axygcn radical damaged

hyaluronlc acld may repolymerlze suggests a furthcr mcchanism

that r¡ould llmlt the slze of the degradatlon product to 104 . Ên

equlllbruim may be reached r¡ith depolymerizatlen predominating

at large molecular rrreights and repolymerization predominating at

small molecular rlalghts.

Ê number of there hypothese¡ are testable and lead to tha

folloulng speciflc questions being posed.

l. To r¡hat extent and hor¡ tlghtly does hyaluronic acid bind

(labelled) lron on other metal ions?

2. Can hyalunonic acld degnadatlon products interfcra r¡ith

chalator-metal bindlng?

3. Êre hyaluronic acid degradation productg morê efflcient

o¡(ygÉrr radlcal scavÉngÊrt on a mole for mole basls that

undepolymerlzed hyaluronate?

4. Can crolt llnked or branchcd hyaluronlc acid be found in

the depoll,merlzation products of oxygen nadlcal exposed

hyaluronic acid?

SAT0 and llll.dÊ (l9ES) hare addressed the question of the

nadical scarenglng capaclty of hyalunonlc acid (yas) and its

constlturnt sugars, D-glucuronic acld (No) and N-

acetylglucosamlne (yss¡. llyaluronlc acld dcAradatlør products (i.e.

of about 104 mohcular r.rlelght or gmaller ollgosaccharldcr)

houevçr harç not barn axamlned.

lrJa hara looked for crosslinked or "brânchad" hyalunonic acid

in oxygen radlcal exposed hyaluronlc acld. Thls r¡as dont by gel

chromatography of oxygan radlcal exposed hyaluronic acid that

uas axhaustlrely digested r¡ith streptomycÊs hyalunonidase.

llo¡¡ever our abillty to nesolve the oligosaccharides uas not

r42

sufficlent to allor¡ a concluslon. Ner¡ar methods using

polyacrylamide gel electrophonesis to separate hyaluronate

oligosaccharides may be helpful in this negand [l-lHllPS0N and

tìÊLLÊ61'lER, l9E5l. The findlng of a product, chanactenistically

altered by oxygen radical exposure r¡ould prorride a means to

establlsh unequirocally r¡hether lnflamed synorrial fluid hyalunonic

acid is depol¡,merized by oxygen nadical exposune in rivo.

Z. Reducino Ends.

-

The findlng in our study that reducing ends are released

during the oxygen radical induced depolymerization of hyaluronic

acid is clear cut since it has been obserred by both colorimetric

and radiolabeling techniques. Thls finding is in contrast tu some

othen early reponts IHcELEÊN and llÊLE, lg4tl; EREENITIÊLD and

H0Y, tg80l. Houerer ELELÊN0 et ê1, I lgEgl, using the aEcorbic acid

induced drpolymenization of hyaluronic acid r¡ene able to detect

the release of reducing ends uslng sodium bonohydnide labelling.

l.rJe dlffer fnom these authons hor¡erer in finding that it is the F l-5 glucuronidic linkage that is preferentially cleaved nather than

the F l-{ hexosaminidic linkage. l,rJe har¡e been careful not to let

clearage by the alkaline conditions required during both the

colorimetnlc and radiolabelling assays lead to misleading nesults.

ShJÊNN É êt t I 96Êl hare determined that hyaluronic acid is not

depolymerized unden the conditions used in oun study for cyanide

labelling. Ê similar cniterion r¡as also established for the

colorimetric assay of Park and Johnson (1.e. pl'l not greater than

9.0). This is of importance since the F l-3 glycostdic linkage of

hyaluronic acid is relatlvely labile in alkaline csndltions [R0EIN50H

and l'l0Pl,rJ00D, l9?31.

Tr¡o other llner of evidence r¡ould also support clcarage atthe F l-5 glycosidlc linkage. The first comÊs fnom the uork of

r43

BÊLÊZS et êt I lgE?] urho reported the electron spin resonance

spectroscopy findings of hyaluronic acid in aqueous solution

exposed to pulse radiolysis. Pulse radiolysis in aqueous solutions"

leads to the generation of the hydroxyl radical as r¡ell as the

aquÊous electron and superoxide anion. Their findings uJere best

explalned by the formatlon of tuo translent lntermedlates thatcould be produced by eithen the hydroxyl radlcal (relevant to our

oxygÉTr radical ge.nerating systems) or the aqueous electron (not

relerant). The first transient intermediate is produced by

remoral of an electron fnom the oxygen of the F l-5 glycosidic

llnkage. This r¡ould react furthen girring rise to clearage at this

linkage. The second intermediate, seen at high doses of nadiation,

inrolves labilization of the hydrogen on E5 of D-glucuronic acid.

This could be stabillzed by a Fesonance lnteraction uith the

caboxyllc acld group of this sugar. This latten intermediate måy

hare significance in any rÊpolymerization neaction seen as à result

of interaction of hyaluronic acid ¡¡ith the hydroxyl radical.

Secondly our results allor¡,an explanation of the findings of

EREENUJÊLD and t10Y I l9?9], r,.rho reported that hyaluronic acid

exposed to oxygen radicals is mone susceptible to dlgestion by the

neutrophil lysosomal enzyme, N-acetylglucosaminidase. This

enzyme uill remore N-acetylglucosamine fnom the non-neducing

end of hyaluronic acid ILINKER, |'1EYER and l.¡JElSSt1Ê¡lì 1955].

Therefore if oxygen radical depolymerization of hyaluronic acid

representE a situation r¡here only the F 1-ã glycosldic línkage ís

cleaved, then all cleavages r¡ill leare H-acetylglucosamine at the

non-reducing end. This uill allol¡ maximal subEtrate for this

enzymÊ and thus a maximal amount of N-acetylglucosamine r¡íll be

released upon digestion r¡ith this enzyme.

L44

E. STUDIES OF sY]IOV'ÃL FLUID }IYÊLUROIIIC RCID ISOLßTED

FRO}I IIIFLR]IED JOI]ÍTS. RRE OXYf¡E]I RRDICßLs RESPO]ISIELE?

Ê large number of studles hare reported the changes in

synovlal fluid that occur as a result of tnflanrmation. Êlthough a

numbcr have presentad dlrzergtnt results for the molecular

r.r.relght changcs lcËn ln synorrlal fluld hyaluronlc acld (see BÊLÊZS

ç3 åL I l9E?], DÊl'lL Êt åL I I 985]), a number of characteristlc

changes arË agrÊed upon. These arc:

" incretsed synovlal fluid rolume

. decreased synorlal fluid hyaluronic acld concentration (and

hence the clinical finding of a decreased viscosity of rau

synorzlal fluid).

. lncneased "total amount" of hyaluronlc acld presetrt ln the

synorzlal fluld as ê result of the lncreased rolume.

" decrças¿d avsragc molecular r.lelght of the hyaluronic acid

presant.

Thc molecular rlelght spectra of lnflamm atory synovlal fluid

hyaluronlc acld prescnted in thls thesis årË conilstent ulth other

studles demon¡tratlng that the decreasÇ rn average molccular

rrrelght, sÊÉrr during fnflarnmation ls smalf but dcfinlte. ln inflamed

synovial flulds, mort synovlal fluid hyaluronic acld rerflains excluded

on Sepharose 'tE suggesting a ve?y hlgh molecular ureight"

l{erertheless, ulth inflarnmatlon a greater proportlon elutes in the

fncluded rolume and material can be indentifled doum to amolecular uatght of 2x l0{. ttf addit¡onal intarest are the shapes

of the proflles. Host lncluded rolume materlal ls seen ås a sker.rllng

of the excluded voluma peak into the lncluded volume houever a

number of our lnflamed flulds hare ä sÉparäte pcak in the includad

rolume on Saphaross 18. Thls l¡ould be pradlcted îrsr¡ our flndlngs

r4õ

in the þ ritro studies that degradation procecds rapidly to the

smaflest degradation product tÊËn (¡.e. molecular r.leight of lElJ.

This second peak may rÊprËrênt small molecular ulelght breakdol^m

productsr prÊseît in sufflcient quantity before thay can be

cleared from the Jolnt spacË.

l¡Je, llke others ISUI{DELÊD, 1953; DÊl'lL Et åLlgA5] could not

flnd any rlgnlflcant corrclatlon bctLlËêt1 thc decr?atÊ in areragc

molccular uclght and othcr lndlces of lnflanrmation (crythrocyte

sedlmantation rate, haemoglobln, synorrial fluld r¡hlte cell counts)"

Eoth ¡idcs of this attempted correlation hou¡ever, represent

impreclse mËärurÊr" Systemlc mÇasurÊs of inflarnmation

(erythrocyte sedimentation rate and fall in haemoglobh) may be

at rariance r¡ith the dagree of synovltls 'nvolvlng the joint

sampled. Local meaiurcs (synovial fluid cell count) are abo

lmpreclse e.g. some ve?y lnflamed Joints hare no effuslon at all"

0ther factors, such as the productlon and clearance of

rarious slzcs of hyaluronic acid from tha synovial fluld are goÍng

to effect the molecular r.r;eight proflle of s¡novial fluid hyaluronic

acid sampled.at any particular time. l'lyaluronic acld is being

samplcd at one point in a dynamlc procËst depcndtnt upon the

ra-tes of both s¡,rnthesis of hyaluronlc acld and its clearancË from

the joint rpäcË. The peturbatlons of thcse latter Processes by

inflanrmatory change are not knoum, indecd llttle ig knot fi of

hyaluronlc acid turnorÉr. Studies by ÊNIOI.|AS' FRÊSER and

HUIRDEN [19?2] hare shor,rn the very rapid äPPearance ((15

mlnutes) of labelled hyaluronic acid in the draining lWph node of

the rabblt knee after intra-articular lnjection.' Thls suggests a

rapld rate of turnorer. Êny molecular r.r.relght change produced by

a depolymerizatlon procÊrs !tr vivo is depcndent in part upon the

r46

"length of time hyaluronic acid is exposed to the depolyrnerzlng

agent, Ê teleological explanation for a high rate of hyaluronic acid

turnorer is that this brill limit the extent of deterioration of the

synovial fluid hyaluronate matrix due to a depolymerization process

and thus r¡ill preserre its biological functions,

C" EFFECT OF OXYÍ¡EH RRDICRLS OIt SYIIOYIOCYTES ffI{D T}IEIR

PRODUCTIOII OF TIYffLUROIIIC RCID.

Finally there remaing the proposition that hyaluronic acid is

not depolymerized n vivo - that there are enough oxygen radical

scarenging mechanisms arzailable in the synovial fluid to prevent

post- synt h et ic de polymerization. This hypo th esi¡ s u g ge s ts th at

inflammatory damage to type 1l sgnovtocytes causes them to

synthesize hyaluronic acid r¡ith a decreased average molecular

r,.reight. ln support of this hypothesis PUORIO É 4 tl gÊ21 have

published the profiles of hyaluronic acid produced in culture by

synoviocytes obtained by explant culture of rheumatoid and normal

synovtum. They reported that synaviocytes obtained from inflamed

rheumatoid and joints inflamed for other neasons produced

hyaluronic acid of a lor¡er molecular r¡eight than synovtocytes

obtained from normal or non-inflamed synovium. The Sepharo:e 2B

profiles for the products of inflamed synoviocytes shobJ minimal or

no material in the excluded volume and the peak of the included

rolume materiEl has a Kar of approximately B.?5 rorresponding to a

moleculan r.reight of 105. Simply stated this material is too small to

compapÊ r¡ith hyaluronic acid extracted fnom inflamed synorial

fluids. Ê number of factors r¡ill affect N-acetylglucosamine

inconponation into hyaluronic acid under cultune conditions. These

include the grou.rth phase of the cells and the glucose

concentnation. Êlso it may be preferable in this circumstance to

considen orgån culture as suggested by HYERS I l9E5]. The findings

L 47

of PUORl0 et al. [1981] therefore are more likely to indlcate thatthe synorriocytes fnom inflamed joints äne moFe susceptible to in

ritro lnfluences that cÊusÊ an altenation in hyalunonic acid

synthesis nathen than pnorring that a smallen size hyaluronic acid

molecule is produced by inflamed synoriocytes h 4!4,.

Ue appnoached thls question by asking uhether cultuned

synorrial cells could be altered by exposure to oxygen radicals in

such a rJay as to produce a louer molecular rrreight hyalunonic acid"

ln short u¡e found that cells exposed to sublethal doses of oxygen

radicals shut off thein productlon of hyaluronic acid nather than

produced a smaller product. l'lyaluronic acid production bJas more

sensitlre to oxygen radlcal damage than other cellular synthetic

functions. This is in keeping u.rith the r¡ork of BÊTES gE gt t tg8ll

r¡ho found that upon elrposupe of chondnocytes to a sublethal

oxygen radlcal flux the nates of synthesls decreased but the si¡e of

the product r¡as unaltened. Clearly r¡e hare only tested one of a

possibly very large number of factors acting on the synoriocyte at

an inflammatory site and r¡hich includes other neutnophil secnetony

products, T cell and macrophage products, cytokines and so on. ltis possible that one or more of these could lead to an alteration in

the size of the hyaluronate produced. Êlso more understanding of

the mechanisms of lnitiation and cesEation of hyaluronate chain

spnthesis are nequined before this question can be completely

resolred. llyaluronate synthetase, by rirtue of its close

association r¡ith the cell membrane is likely to be susceptible to

membrane disruption. [Jne of these is the production of oxygen

radicals due to the propensity of membranes to undergo lipid

peroxldation. Ê logical extenslon to oun studies is therefore toexamine other membrane damaging agents for their effect on

hyaluronate synthetase function and also to examine the cell

14E

supernatants for the aPPearance of lipid peroxidation products

particularly ln relation to the cessation of hyaluronic synthesis.

D. COHCLUDIHf¡ REHRRKS.

The effect of oxidants generated r¡ithin a biologic al system

hare been seen as an interplay betr¡een the reactivity of the

oxldant generated and the defence or scarenger systerns arzailable

to protect against that oxidant. lt has emerged frsn the r¡ork in

thts thesis that a third important factor in this lnterplay is target

susceptibility, The evidence in our gtudies on isolated hyaluronic

acid has pointed tor¡ards a site specific susceptibtlity to

depolymerization. Specifically þl-3 glycosidic linkages spaced at

sites along the molecule are susceptible. Equally r¡ith synoviocytes

in monolayer culture the loss of cell rziability and cessation of

hyaluronate synthetase function are mediated by hydrogen

peroxide, underlining the impontance of access to a target site

r¡ithin the cell.

Ê study of the effects of oxidants r¡ithin the context of joint

inflammatlon has been undertaken by examining in detail an

important component of joint physiology, namely hyaluronic acid.

Such a reductionlst tactlc has been used since the number of

effector agents, modffying factors and structures at risk r¡ithin

the inflamed join t arelÐnumerous to study effectively. Yet thiE

approach has p?oveî useful in exposing target sltes susceptibile to

oxidant damage

Ên ultimate aim of research in the field of rheumatology is

prerrention, by therapeutic interrention, of joint deformity u.rought

by the lnflammatory Pnocess. Ê clear appreciation of the specific

tyFe and extent of this damage that can be attr'ibuted to

oxidants is required so that appropriate therapeutic steps can be

taken to prerzent this form of joint damage from occunning.

r49

ÊPPE]IDIX I

uRoÌilc fictD Ê55RY

The technique desribed by HLUiIEÌ{KRÊNTZ and Ê5BUE-|'|ÊNSEI{

[19?51 uag used.

Ê" REÊGENTS

l. Heta-hl¡droxydiph

Ên 0" 15% solution of mcta-hydroxydlphenyl (Eastman Kodak

Eo., Rocherter, New York) ¡n O.5% t{a0ll urar prÊpared and ¡tored

in the refrigerator r¡¡ith an aluminum foil corering. This solution

uJas relnade erery three monthE.

Z. il;FL+/tetraborate solution

Ên 0.0125 n golution of sodium tetraborate in concentrated

sulphuric acid.

Ë. METHOD

0.2 ml of unknoun and standards containing from ?.5 - 100

ug/ml D-glucuronic are added to 15 xl00 mm test tubes. 1.2 ml

of the sulphuric acid/tetraborate solution is then added and the

tubes are refrigerated on crushed ice. The tubes are then

¡haken in a Portex mixer and heated in a boiling r.r.later bath for 3

minutes. Êfter coollng on crushed ica for l0- 15 mlnutes, 20 ¡.rl of

tha m-hydroxydiphenyl reagent is added. The tubes are shaken

and the color alloued to derelop fon 5 minutes. Êbsorbance

mÊåturãïents are tht¡ read at 520 nll on a Perkin-Elmer (LS

1500) spectrophotometer. Ileasurements on sämples fnom

fractions after gel chnomatography are done in duplicate, all

other determinations are in triplicate.

t50

FERRICYßI{IDE IIETTIOD OF PRRK RIID JO}I]IsO]I

Thic technlqut l¡ a modlflcation of the àsray orignally described

by PÊRK and J0l'll'150H [ 19111 and sub¡equently modified by H0UEK

and PEÊRCE [ 195?l îor use r¡ith hyaluronic acid.

Ê. E!.Agg¡lå Three reagents arÊ prepared I

I ) Femic,uanide solution. 0.5 gt of potassium ferricyanide per

liter; stored in a bror*m bottle"

Z) Earbonate-cpanide ¡olution; 5.5 gm of sodium carbonate +

0"ES gm of potassium per liter of solution. ln our experÍments the

pll of this reagent r.rras adjusted to a pl'l of 9.0 [ll0UCK and

PEÊRCE, tg5?1"

5) Femic ion solution; 1.5 gm of ferric antmøtium sulphata + I

gm of Duponol (sodium monolauryl rulphate, Du Pont lnc.,

l,rlilmington, Delar¡are) in t liter of 0.05N ¡ulfuric acid.

E. Hathod

0.5 ml of the unknoun¡ and standards contaÍning I -ZE ug/ml

glucuronic acid are added to 15 x 100 mm PyrËx test tubes. 0.5 ml

potassium fcrricyanida and 0.5 ml carbonate-cyanide solutions are

then added. The mixtune is then shaken on a Pontex mixen and

heated in a boiling r.r.rater bath for 15 minutes. The tubes ane

then cooled by immersion in a r¡ater bath at room temperature

for a further 15 minutes. 2.5 mls of the ferric ion solution are

then added and the mixture again shaken on a Pontex mixÊr. The

colour reaction is then read at 690 nPl on a Perkin Etmer (LS 1500)

spectrophotometer. Êll measurements are done in triplicate.

r5t

Ìl0Rf¡Rl{ - Elsoll REHCTI0I{

Thls technique ig thc modificatlon of tha Horgan-Elson reaction

. according to RElSSltì, STR0|'4|NGER and LEL0IR [1955],

Ê. BEÊEESIE

l " ; Hn [J.E l'l potatsium

tetraborate solution ig madc and the pll adjustÊd to 9.1 ¡¡ith

potassium hydroxide.

Z. P - dimeth,ulam¡nob Dl'1ÊE)-el¡¡lieg; l0 9m Dt1ÊE

(Eastman Kodak, Rochester, Fleu York.) is dissolved in 100 ml

reagent grade glacial acetic acid l¡ith lZ.5% (rrolume per rrolume)

l0 N hydrochloric acid. This reagemt can be stored at ?oE for up

to a month in a dark bottle. Shortly bcfore uge the golution ig

diluted ¡¡ith I rolumes of neagant grade acetic acid.

E. TIËIEED,

0"ZS ml roluma¡ of unknorrms and standards contalníng 2.5 -50

ug/ml N-acatylglucosamine are added to 15 x 100 mm PyrÊx test

tubes" 50 ul of potassium tetraborate solution is then added.

The tube! ârÊ heated in a boiling ulater bath îar 3 minutes. Than

they are cooled in tap urater and 1.5 Dl'1ÊE ¡olution i¡ added and

the tubes are shakerr on a Vortex mixer. The tubes are then

heated in a r¡ater bath at 3E - 38oE for 2O minutes. They are

then read at SES nH on a Perkin Elmer (LS 1500)

spectrophotomcter. Êll measurements are done im triplicate.

r52

RPPEI{DIX II

T|55UE CULTURE 50LUTl0l{5

Ê. PRI]1RÊRY CELL CULTURE IIEDIUI'I"

- llam's F l0 medium [HÊH' l9E3]

- Z0t6 foetal caff serum.

(heat inactivated by incubating ät 56oE for 50 mins)

- gentamycin 0.2 mg/ml.

- L-glutamine 5 mH added fresh.

E. ROUTINE SYNOI/IÊL EELL CULTURE IIEDIUII.

- llam's F l0 medium.

l0% foetal calf serum. (heat inactirated)

- gentamycin 0.2 mg/ml.

- L-glutamine 5 mll added fresh.

E. DULEEEEO'S PI'IOSPI.IRTE BUFFERED 5ÊLINE.

NaEl 8.0 g/l KEI 0.2 g/l

CaCl2 - 0.1 g/l HgCl2.6ll2B - 0.1 g/l

Na2HP0'1.2H20 l.l5 g/l KH2P04 0.? g/l

McNeil, J. D., Wiebkin, O. W., Betts, W. H., & Cleland, L. G. (1985).

Depolymerisation products of hyaluronic acid after exposure to oxygen-derived free

radicals. Annals of the Rheumatic Diseases, 44(11), 780-789.

NOTE:

This publication is included in the print copy

of the thesis held in the University of Adelaide Library.

It is also available online to authorised users at:

http://dx.doi.org/10.1136/ard.44.11.780

McNeil, J. D., Wiebkin, O. W., Cleland, L. G., & Vernon-Roberts, B. (1986). The

generation of reducing ends by exposure of hyaluronic acid to oxygen derived free

radicals. Agents and actions. Supplements, 18, 95-101.

NOTE:

This publication is included in the print copy

of the thesis held in the University of Adelaide Library.

r?0

EIELIOÍ¡RRP}IY

Êtenrcr, J.11., PlcERctxy, C.Ê. and llunst, J.K. (l9E l)" Eiological

reactir¡lty of hypochlorous acid: impllcations for microbicidal

mechanisms of leucocyte myeloperoxidase" Proc. Natl. Êcad.

Sci.USÊ ?8, 210-e 14.

Êl-Kr¡uor, U"Ê.3. and CrnalßssroN, T.ll. ( 1965). The specles dlstribution

of xanthine oxidase. Eiochem. J. 9?, 3lE-521.

Êmanuso, D.R. and Jottttsolt, R.E. ( l9E l). Lactoferrin enhances hydroxyl

radlcal productlon by human neutnophils, neutrophil

particulate fractlons and an enzymatic generatlng system.

J. Ëlln. lnvest. 6?, 352-380.

Êmes, 8.N., EßTHcñRT, R.r Scrutens, E. and llocnsrutn, P. ( | 9E I ). Uric acid

prorldes an antioxidant defense in humans against oxidant

and nadical-caused aging and cancer: ä hypothesis. Proc.

Natl. Êcad. Sci. USH ?8, 6ESE-6862.

ÊnnsrRsstRoEs, T. and RoeeRTsoN, lrJ. ( l9E{). l.lodulation of mitogen-

dependent lymphocyte stimulatlon by hyaluronic acid. J.

Rheumatol. I l, ?29-?54.

ÊneRn, 11. and NFrñ, P" ( | 96?). Ê compilation of specific bimolecular

rate constants for the reactions of hydrated electrons,

hydrogen atoms and hydroxyl radicals r¡lth inorganic and

organlc compounds in aqueous solution. lnt. J. Êppl. Rad.

lsotopes lE,495-5U5.

ÊtnonRs, K.l'|., Fnf,sEp, J.R.E. and Hu¡noEN, K.D. ( l9?3). Distribution of

biologlcally labelled radioactlve hyaluronic acid injected into

joints. Ênn. Rheum. Dis. 32, 103- I I l.

Êtnonrnr, E., Bnunopr, l{., Gnegnuooo, C. and J'lnlmsrnom, E.E. ( 1970).

Catalytic mechanism of cytochrome oxidas¿. Nature (London)

228, 93ã-937.

r71

Êsnulrtt, 6. ( lg5?)" Colorimatric analysis of sugårs. llethods

Enzymol. 5, ?3- l29.

Êtxrtts, E"D.T., He Roen, D" and Scon, J.E. ( lgËB). llodel for hyaluronic

acid tncorporating four intramolecular hydrogen bonds. lnt.

J" Elol. Hacromol. 2, 5l B-519.

Eaeton, 8.L., RosEN, R.E., Hcllunnrcr, EJ., Penns, N.Ê. and BRBIoR, 8.11.

( l98 l). Êrrangement of respiratory burst oxidase in the

plasma membrane of the neutrophll. J. Cl¡n. lnrest. 6?,

r ?2{- I ?28.

EReton, E.N, (19?E). 0xygen dependent microbial klllingby

phagocytes" l{err.r England J. Hed. 298, 859-6ã81721-'?23.

BReton, 8.H., KtRttEs, R"5. and ÊRnnurE, J,T. ( l9?5), Biological defense

mechanisms. The productlon by leucocytes of superoxide, a

potent bactericldal agent. J. Êlln" lnrest. 52, ?{ l-?4'1.

ERtnzs, E.Ê. ( l9?'l). The physlcal propertles of s¡rnorrial fluid and the

speclal role of hyaluronlc acld. ln: Dlsorders of the Knee.

Helfer Ê. (Edltor) pp Ê3-?5. J.E. Lipplncott, Philadelphia.

ERrnzs, E.Ê. and DRnzvrr¡¡blcz, Z, (.19?3). The effect oÍ hyaluronic

acld on flbroblasts, mononuclear phagocytes and

l¡mphocytes. ln: Blology of the Fibroblast. Kulonen, E. and

PlkkaralnËn, J. (Ed¡tons) pp 25?-25?. Êcademlc Pness, London.

Entnzs, E.Ê. and Lnunttn, T.E. ( lg5 l )" Tha viscosity function of

hyaluronlc acid ås a polyelectrolyte. J" Polymer Sci. 6' 665-

EEE.

ERtRzs, E.Ê. and Suttoltno, L. ( 1959). Vlscoslty of hyaluronic acld

solutions contafning proteins. Êcta Soc. HÞd. Upsalien. 54,

r 5?- r 46.

L7 2

ERtRzs, E.Ê., BntLLEn, S.l. and DgtttH6Ep, J.L. (1381). Na-hyaluronatc

molacular size rariations in equine and human arthritlc

synorrlal flulds and the effcct on phagocytlc cells. Semin.

Êrthritis Rheum. I l(suppl. l), l4l- 145.

ERtRzs, E.Ê., LRuRENT, T"C., HorrlE, Ê.F. and Hnnen, L" (1959). lnnadiation

of muco-polysaccharides ulth ultraviolet llght and electpons.

Radlat. Res. I I, l {9- 16'1.

ERtnzs, E.Ê., DfiutEs, J.P", Ptltttes, 6.0. and Youne, 11.0. ( 196?).

Translent fntermediates in the radiolysls of hyaluronic acid"

Radlat. Res. 5l, 245-245.

Eatnzs, E.Ê., hIRTsoN, D, Durr, l.F. and RosemRn, 5. (l9E?). l'lyalunonlc

acld ln synorrlal fluld. l. Holecular paramcters of hyaluronic

acld ln normal and arthrltls fluids" Êrthritls Rhaum. l0' 35?-

5?6.

Ennnlsmn, J.V,, BRrrttrten, lJ.l1., H|LL, ll.Ê.0", llÊttooo, J.F., hllLtsottr R.L. and

hlolrgnoer, 8.5. ( l9E0). Does ceruloplasmin dismutase

superoxlde? FEES Lett. I lE, 12?-129.

ERnrEn, 5.Ê., EfryyuK, J.5., Enlmnco¡,leE, J.5., HAt.lrltts, C.F. and STRCEY, l'1.

( 1965). Fhgerprinting the hyaluronlc acid component of

normal and pathological synovlal flulds. Clin. Êhim. Êcta E,

902-989.

ERntRno, P., l'loutroFF, Ê.8. and l'lnr¡gnmf,H, D. ( lgEZl. Electronmtcroscopy

of the human synorrlal membranc. J. Eell Elol. I'1, 20?-220"

ERnnm, Ê.J. and SnrlmvRlR, J. ( l9Ë5). Protelnases in Jolnt dlseage.

ln: Textbook of Rheumatology, Znd Ed. Kelley' Ld.N., llanris,

E.0., Ruddy, S. and Sledgê, 8.8., (Edltors) pp lEZ- 192. hJ.E.

Saundrns Eompanyr Phlladalphla.

BRntoto, P.H., lrltEextn, 0.1¡J. and TxonRno, J.E. ( lgEZ). Holaculan r.lelght

estlmatlon of suJfated glycosamfnogþcans in human gingirae.

Eonn. Tlss. Res. 9, 165- l?2.

r7 3

Enntoto, P.l*|., lrJteertr, O.[rJ. and Tror'rnno, J.Ê. ( l9f.1a). The effect of

o,(ygËn derired frce radlcals on glngirzal proteoglycans and

hyaluronlc acld. J. Periodont. Res. 19, 390-400,

ERmoto, P.l*|., l,rJttertn, 0.trJ. and TttonRno, J.E. ( lgB4b). Eeharlor of

hyaluronlc acld from glnglral epithelium and conncctlre tissue

on thç analytlcal ultracentrlfuge. CoÍln. ïr!, Res. 12r 25?-

2Ê4.

ERtEs, EJ., LounxEn, D.Ê. and llnnoLEy, C,J. ( l9B4). Exygen fnee-

radlcals medlatc an lnhibitlon of pnoteoglycan synthesis in

culturcd artlcular cartilage" Élnn. Rheum. Dis. 45, {62-469.

BRxnn, E., FRESER, J.R.E. and l'lnnusr 6.5. ( l9?5). Fractionation and

recovÇry of secretions of synorzial cells synthesized in culture

¡¡lth radioactire precursors. Ênn" Rheum. Dis. 52, 35-{0.

BgRucttRnp, E. and Fntoourctt, l. (19?0). Ê mechanlsm fon the

production of ethylene from methional: the generation of

hydroxyl radlcal by xanthine oxidasa. J. Elol. Ehem. 245,

.lEl l -+E{8.

Eeem, K.11., R¡crnnosoÌr, D.C. and Ra.rn60pf,Lnn, R.V, (13771. lletal Eltes of

coppËp-zlnc superoxide dlsrflutast. Eiochc'rnistry 18, 1950-

I gEE.

Etrnm, R.l'î. ( 197?). llaematologlcal changes ln rheumatoid diseasc.

Elln. Rheumàt. Dls. 5, {55-{65.

BEnrnen, L.E. and Hnnsrntt, L.E. ( l9E+). The hlstory of oxygenic

concentratlon in the earth's atmosphare. Faraday Discuss.

Ehem. Soc. 5?, I 12.

Ems, lrJ.lJ. and EtetRro, L.6. ( lgÊZ). Effect of metal chelators and

antiinflammatory drugs on the degradation of hyaluronic

acld. Êrthritis Rheum. 25, l'lE9- 1'l?8.

L74

Elette, Ê., ÊNoEnsson, T. and GnnnfrTr, K. ( lg82). I'lolecular r.leight

distribution of hyaluronic acld of human synorial fluid in

rheumatic diseases. Scand. J. Rheumatol. l?, ISS- lSE.

EtRre, 0.R., llRLL, N.D., Tneey, D.Ê., llRLLil¡JELL, B. and Gunen¡oee, J.l1.E.

( l9Ë l). Protection against superoxide and hydnogen peroxide

in s¡rnorlal fluid fnom rheumatoid patients" Clfn" scl" 61, 485-

{Ë6.

EtRre, 0.R., ELßÌ{il, É1., ERcott, P.Ê., Ffinn, ll., 6ulrERtD6E, J"H.E. and

HRU¡r¡eu, E" ( 1985a). Femoxidase and ascorbate oxidaEe

actitzitles of ceruloplasmin ln synovial fluid frsm rheumatoid

patients. Clln. Scl. 84, 551-553.

BtRt<t, 0.R., llfrLL, ll.D., ERcott, P.4., DtEppE, P.Ê., l-lnll¡l¡et¡., B. and

[ìunrn¡oee, J.Fl.C. ( lg8rb). Effect of a specific tron chelating

agent on animal models of inflammation. Ênn, Rheum. Dis. 12,

Eg-95.

EtRre, D.R., BßLLfiGHEn, PJ., P0TÍER, 4.R., Egtt, l1J. and EficoN, p.Ê.

( l9E4). The effect of synovtal inon on the progresEion of

rheumatoid disease. Êrthritis Rheum. 27, lgí-581.Etumeenc, 8.5, and 0esror, Ê.8. ( 195?). The effect of proteolyflc

Ënzymes on the hyaluronic acld complex of ox synorrial fluid.

Biochem. J. 66, 5'12-5+8.

EtumsnxnnnTz, N. and Êsaoe-llRnsEn, G. ( l9?5). Ner¡ method for

quantitatire detenmlnatlon of uronic acids. Ênalyt. Eiochem.

54' 'l8'î-'lBg'Eonngc¡Rno, N. and KnResarre, K. ( l9E0). Role of oxygert in anfibody

depandent cytotoxicity medlated by monocytes and

nautrophils. J. Cl¡n. lnrcst. 66, E?6-685.

Ennnot, K.D. ( l9?.1). The effect of synorzlal hyaluronate on the

ingestion of monosodium urate crystals by leucocytes. Clin.

Ehim. Êcta 55, 30?-315.

rz5

Ënnun, K. and Fntooutcx, l. ( l9B l). DNÊ strand scisslon by enz¡,rmlcalty

generated oxygen radicals. Êrch. Eiochem. Elophys. 206,

{ 1.1-419.

En¡mRcot"tgE, J.5" and [rJeeeEn, J.11. ( l96rî). I'lyaluronic acid. ln:

l"lucopolysaccharldeg: chemical structure, dlstribuflon and

lsolatlonn Ehap. 3, pp ,{3-83. Elserler, Êmsterdam.

BntnrmRn, R., Lfim8Enrg, H"8., lrJRoetn J. fitt0 Zuloeueto, J. ( | gE l).

Ëontributions to the study of immediate and early x-îay

reactlons l¡lth regard to chemoprotectlon. l. Further

analysis of the immediate drop in inJection pressure on

moderate irradiation" lnt. J. Radiat. Elol. 5, ZE5-210.

Eucrnnnn, J.D. and Ênmyrnone, D.Í1. (19?E)" The radtolysis of

glyceraldehyde-5-phosphate dehydrogÊnä,se. lnt. J. Radiat.

Blol. 53, 409 labstl.

CRnl, ll. and JRnorr, Ê. ( l9?E). Posstble machanismt of emphysema in

smokers. þ ritno suppression of serum elastase-inhibitory

capaclty by fresh cigarette smoke and lts pnerention by

antloxidants" tlnnu. Rer. Resp. Dis. I lE, 6l?-621.

ERRo, H. and JRnorr, H. ( l9?9). þ r¡ltno supprÈssion of serum

elastase inhlbitory capacity by neactlre oxygen species

gencrated by phagocytosing poþmorphonuclean leukocytes.

J. Clln. lnyest. 65, ?95-?9?.

Efiston, E.hJ., Pntncr, R.K. and HnzEtrott, HJ. ( l96E). l'lyalunonic acid in

human synorial effusions: a sensitire indlcaton of altered

connectlve tlsstte cell functlør durlng lnflammation. Êrthrltis

Rheum. g, ?Ë5-?9+.

EttRltt, E. and Dutnle, E.5. ( 1959). Ê mucolytic enzyme in testio

extracts. Natuna 144, 9??-9?8.

r?6

Etetnno, R.L. ( lg?0)" Holecular r.leight dlstribution in hyaluronic

acid. ln: Chemistry and Holecular Biology of the lntercellular

Ilatrix. Ealazs, E.Ê., (Editon) pp. ?55-?{2. Êcademic" Pness,

London and Ner¡ York.

EtetRno, R.L. and lrJnne, J.L. (19?B). lonic polysaccharides. lll. Dilute

solution pnoperties of hyaluronic acid fractlons. Eiopolymers

g, ?99-ElB.

EtrtRro, R.L., STo0LMILLER, Ê.C., R00EH, L, and LnunEHT, T.C. ( 1963).

Partial characterlzatlon of rcaction products formed by the

degradatlon of hyaluronic acid r¡ith ascorbic acid. Elochim.

Blophys. Êcta 192, 585-59{.

Cottt"tRtt, J.P., EEfiuHfrH, J"1., llRteenn, T.R. and Susucx, K.S. (19?Ê).

Hature of E" and C0 bfnding to metalloporphyrins and hc'rne

proteins" Proc. l{atl. Êcad. Scl. U5Ê 13r,3333-3337.

CumRtr, C.C. ( 1955). The nature of the protein in the hyaluronic

complex of borine synovial fluid. Eiochem. J. 61, EÊE-E9E.

DRttt, L., DfiHL, 1.11.5., Engsrno¡r-Launert, Ê. and 6nRrRTr, K. ( lgBS).

Concentration and molecular r.leight of sodium hyaluronate ín

synorial fluid from patients r¡ith rheumatoid arthritis and

other arthropathies. Ênn. Rheum. Dls. 11, B|7-BZ?,

DRs, 3.K. and NRtn, R.t. ( l9A0). Superoxide dismutase, glutathione

peroxidase, catalase and lipid peroxidation of normal and

slckled erythrocytes. Erit" J. Haematöt 11, A?-92.

DglssenoTx, Ê. and Dounce, Ê.L. ('19?0). Eatalasez physlcal and

chemical properties, m;chanisms of catalysls and physiological

role. Physiol. Rer. 50, 319-5?5.

DeL llnesrn0, R.F., T}lRtt, H.H., EJoRK, J,, PtRnren, 11. and ÊnFons, K.E.

( l9E0). Free radicals as medlators of tlssue inJury. Êcta

Physiol. Scand. suppl. 492, {3-5?.

L77

Enrrssott, S., FRf,sEß, J.tl.E., LRunent, T.8., PERToFT, 11. and 3meo8R00, E.

( l9E5). Endothellal cells ârË â slte of uptake and

degradation of hyaluronic acld in the llren. Exp. Eell Res.

111, 223-224.

FRtcttut<, K.11., EoETaL, EJ, and Kulrf,, J.P. (19?0). Rerpiratary gaset

of synorrlal flulds. Êmep. J. Hed. '13, 225-U 51.

FEE, J.Ê. and PnleNTHE, J"3. ( lg??). Êhemlcal and physlcal properties

of superoxlde. ln: Superoxide and Superoxide Dismutases.

lllchelson, É1.H., llcEord, J. and Fridorich, l. (Editors) pp l9-80.

Êcademlc Press, London.

Fennntm, Ê. and Tnone, Y.H. ( I g?B). Ê napld one-step procedure for

purlficatlon of mononuclear and polymonphonuclear

leucocytes from human blood using a modlflcation of the

Hypaque-Flcoll technlque. J" lmmunol. llethods 21, 589.

Ftssten, J.ll., 0csroÌ{f Ê.8. and SrnntEnr J.E. ( 195'{). Some propenties of

human and other s¡rnorlal fluids. Elochtm. J. 58, 658-680.

Flscuen, E. ( lE90). Synthesan in der zuckcrgruppÊ. Eerichte der

dautcshen chemischen geselschaft. 25' 2l l{-21+ l.

Ftscmen, L. ( l9E0). Eel flltratlon chromatography. ln: Laboratory

Techniquas in Eiochemistry and Holecular Eiolog¡1. lrlork, T.S.

and Eurdon, R.ll. (Eeneral Edltors). Elserrien/North llolland

Elomedlcal Press, Êmstendam, Nerl Yonk, and 0xfond.

Ftnmm, E.5., DEM0pouL0st ll.E., SEL|EMRNr fí.L., Posen, R.6. and RRNsotoFr, J.

( I g?E). Fnee radicals ln cerebral ischaemla. Stroke 9, '1'{5-

44?.

Ftrrun, hl., RouLEy, D.Ê. and Hnl¡.tt'rELL, E. ( l9Ë5). Superoxide-

dependent formation of hydnoxyl radlcals in the presence of

lron salts. hJhat is the physiological lron chelator? FEBS

Latt. l5A, 510-31?.

t78

FtoYo, R.Ê. ( lgES). Direct damonstratlon that fernous ion

complexes of di- and triphosphate nucleotides catalyze

hydroxyl fnee nadical formation fnom hydrogen peroxide.

Êrch, Eiochem. Blophys. 225, 263-2?0.

FLynn, J.P., ÊLLEN, D.hl.r Jottttsott, 6J. and hlxre, J.6. ( lgES). 0xidant

damage of the llpids and protelns of the erythrocyte

membranes ln unstable hemoglobin dlsease. Eridence for the

role of llpld penoxidatlon. J. Elln. lnrest. ? l, lZ 15- l?25.

Fomr, C.3. (lg?9). Detectlon of slnglet olrygÉ'n in complex systems:

a critique. lnz Exygen: Eiochemical and Clinical Êspects.

Caughey, hJ.S" (Ed¡tor) p 805" Êcademic Press, l{eu York.

Fonnesren, J.7. and BntRzs, E"Ê. (1980)" lnhlbitlon of phagocytosis by

high molecular rrelght hyaluronatÊ. lmmunologg {0' 435-4{6"

Fonnesnn, J.P. and lrl¡t-xtxsolt, P.C. ( lgB I ). lnhibitlon of lucocyte

locomotion by hyaluronlc acld. J. Eell Sci. 48, 515-551.

FoN, Ê.1d. and Êuron, Ê.P. ( l9?E). guperoxide anlon mediated toxicity

in isolated lung cell preparatlons. flnn. Rer. Resp. -Dls. I 11,

535? [abst].

FnRnr,"L. and J{RssRgo, D. ( lg8tl). Exygen toxlcity. Élmer. J. lled. 89,

I r?-r26.

FnRrrtt, E.N. (19Ë0). Llpld oxidatlon. Progr, Lipld Rcs. 19, l-Z?.

Fnnsen, J.R.E., Etannts, BJ. and ERrtel, E. ( l9?g). Pattarns of induced

rarlatlon in the morpholo1!, hyaluronlc acld secretion and

lysosomal erzyme actlvlty of cultured human synovial cells.

Ênn. Rheum. Dii. 38, ?A7-291.

Fnnsen, J.R.E. and llcERtt , J.F, ( l9E5). Culture of synovlal cells iÐ

Eil¡e. Ênn. Rheum. Dis. 2'1, 551-559.

1?9

FnRsen, J.R.E., LRunem, T.8., PEmoFT, 11. and Enxrep, E. ( l9E l), Plagma

clcarance, tlrrua dlstributlon and metabollgm of hyaluronic

acid lnjected intrarenously in the nabbit. Eiochem. J. 200,

.115-{2+.

Fntooutcr, l" (19?0). 0uantltatlrç aspects of thc productlon of

superoxlde anlon radlcal by mllk xanthine oxidase. J. Elol.

Chem. ?15, 1853-'105?.

Fuooulctt, l. ( l9?5), Superoxlde dlgmutasçs. Annu, Rer. Elochem. 44,

l.î?- 159.

Fntooulctt, l. (19?E). 0xygen radicals, hydrogen peroxlde and oxygen

toxiclty. ln: Free radicals in biology, Vol. l. Pryor, Lil.H.

(Ed¡tor) pp 239-217. Êcadcmlc Press, lJer¡ York.

Fntooulctt, l. (lg?E). ThÇ biology of oxygm radlcals. Sclc,nce 201,

8?5-Ë80.

Fntooutctt, l. (19ÊZ). Supcroxlde dlgmutage ln blology and mcdlc¡ne.

ln; Pathology of 0xygen. Êutor, Ê, (Edltor) pp l - 19.

Êcademlc Prcsr, Hcr¡ York.

Fnor¡Rr¡or, 11.P. ( l9??). Êdsorptlon of human erythrocyte surf ace

glycoprotein on to polystyrene latex rpheres. J. Lab. Elin.

lled. 98, 321.

ERnonsn, T.J., SrEurßFT, J.R., ERsRte, Ê.5., DobIHEy, J.11. and CtlnmBEns, D.E.

( l9A5). Reduction of myocardial ischaemic injury r¡ith

oxygen-derived free radlcal scarengers. Surgery 31, 123-

1?7.

EnnotRtty, F.N. ( l9?E). Fine structure of joints. ln: The Joints and

Synorzlal Fluld, Pol. l. (sokoloff, L., Editon) pp 105- l?6.

Êcademlc Pness, Ner¡ Yonk.

6roas, D.Ê., HERRILL, E.hl. and Smm¡ K.Ê. ( l gEE). Rheology of

hyaluronic aÊld. Blopolymers Ë, ???-?9 t.

t80

Botoeene, E. and Srunn, A. ( l9??). Thc role of thc ruperoxide anlon

ås å toxic species ln the enythrocyte. Êrch. Biochem"

Elophys. l?8, 2l E-225.

Gotoeenc, R.L. and Toote, E.P. ( l9E4). Hyaluronate coat formation

and cell spreading ln rat flbrosarcoma cells. E*p. Ëell Res.

l5l, z5B-zâ3.

Gotomern, 1.H., KfipLfir, H.8., Eoeusolt, H.S" and ldg¡ssrtRrtt, E. (19?9).

Ceruloplasmin: a scarÊnger of superoxide anion radicalE. J.

Elol. Ehem. 231, 1E1O-1O13"

6nRncrn, 0.N., RuïLt, 6. and HcCono, J. ( I gB I ). Superoxide radicals in

feline intestlnal ischaemia. Gastroenterology E I, 22-29.

Gnee¡rt¡Rro, R.Ê" ( l9Ê I ). The role of oxygen-derived fnee radicals

(0DFR) in connectlve tissut degradation" lll. Studles on

hyaluronic acld depolymerlzatlon in lnflamed synorrial fluids.

Semin. Êrthnitis Rheum. I l(suppl. l), 9?.

Eng¡nl¡Rt o, R.Ê. and Honx, S.Ê. ( l9E4). Degradatlon of hyaluronlc acid

by neutrophils. lnflammation Research Êssoclation.

Therapeutic control of lnflanmatory di¡eages. ll. Êbst.

papers, p.21.

EneentJnlo, R.É1. and HoY, h,,hJ. (l9?9). lnhibition of collagen gelation

by the actlon of the superoxide radlcal. Arthnitls Rheum. 22,

25 I -259.

GnrenuRlo, R.Ê. and l-10Y, hJ.kI. ( 1980). Effect of orygen-derived free

radlcals on hyalunonic acid. Êrthritls Rheum. U5, '155-465.

Gn¡gnurRr-o, R.Fl., l1oy, lrl.lrl. and LRzRnus, D. ( l9?6). Degradation of

cartilage proteoglycans and collagen by supenoxide radlcal

[Êbstract]. Êrthritls Rheum. 19, ?99.

t8l

6ntrnn, F.11., BtRnco, E" and Srt pgnsrulr, S.E. ( l9?5). Characterization of

the macrophage receptor for complemert and demonEtration

of its functlonal independence from the receptor for the Fc

portlon of immunoglobulln B. J. Exp. Med. 111, 1269-12?7.

EnossrsLo, ll" (1958). Studies on production of hyaluronic acid in

tissue culture: the presence of hyaluronidase in embryo

extract. Exp. Cell Res. 14, 213-218.

6non, l'f.P. ( I g?5). Cellular contacts in tha synovlal membrane of

the cat and rabbit: an ultnastructural study. Eell Tlssue

RËs. 16'1, 525-5{ l.

Burrgnoc¡, J.fí.E", Rrcrmoro, R. and llRl¡.n¡ell, E. ( l9?9). lnhlbltion of

iron-catalyzed formatlon of hydroxyl radicals from

superoxide and of lipid peroxidatiøt by desfercio,ramine.

Eiochem. J. lE4, 469-1?2.

[ìurrenroee, J.11.8., Routey, D.Ê. and I'lRlt ¡l¡ell, E. ( l9E I ). Superoxide-

dependent formation of hydroxyl-radicals in the presence of

lron salts. Blochem. J. 199, 283-265.

6un¡n¡oee, J.l'î.C., Routey, D.Ê. and HRt-r"n¡¡r-t, B. ( l gEZ). Superoxide-

dependent fonmatlon of hydroxyl radlcals and lipid

peroxdation in the presence of iron salts. Detection of

"catalytic" lron and antloxidant actlvity in extracellular

flulds. Eiochem. J. ?OÊ, 805-609.

[ìungn¡oa¡, J.Ì1.C., Pmuneon, 5.K., SEEf,L, Ê.h1. and HnlltLrELL, E. ( l98l ).

lnhibition of lipid peroxidatlon by thc lron bindlng protein

lactoferrin. Elochem. J. l99, Z5g-ZÊ1.

llReen, F. and l,rlutss, J. ( | 934). The catalytlc decompoEition of

hydrogen peroxide by iron salts. Proc. Royal Soc. London,

Ser. Ê l+7,332-351.

llRoten, N.l'!. (19Êl). Enhanced dlffusionity of glucose in a matrix of

hyaluronic acid. J. Eiol. Chem. 255, 5552-5555.

182

llRoten, N.l{. and NRlttn, H.Ê. ( l9??). Structure of hyaluronlc acid in

synorzlal flu¡d and lts lnfluence on the mopÊmÊnt of solutes.

Semin. Êrth. Rheum. ?, l4l- 152.

HRotrn, N.J1., DounmRsrruH, R.R., Nenmur, H.P. and lrJttttRms, L.D" ( ßAZ),

Ultrastructure of a hyaluronic acid matrix" Proc. Ì{atl. Ãcad.

Sci. U5Ê ?9, 50?-509.

llRrnnssott, L. and Punce, P. ( l9Ê5). The combined action of

hyaluronic acid and fibronectfn stimulates neutrophil

migration. J. lmmunol. 135, 2?55-U?59.

HnrRnssotr, L., llÊL[6pEN, R. and Psner, P. ( | 980). Regulation of

granulocyte function by hyaluronic acid. þ r¡itro and þ riroeffacts on phagocytosls, locomotion and metabolism. J. Cl¡n.

lnrçst. 86, 298-505.

HRH, R. ( tgES). Ên lmprored nutnlent solution fon dlploid chlnese

hamsten and human cell llnts. Exptl. Cell Res. 2S, 515-526.

HRt"usoN, I and 6Rt tRengn, J. ( lgES). Separatlon of nadiolabelled

glycosamtnoglycan ollgosaccharldes by polyacnylamide gel

electrophoresis. Biochem. J. 2?l, E9?-?05.

f'lRul¡¡lt , E. ( l9?E). Superoxide-dapendent fomnation of hydroxyl

radicals in the prÊience of iron salts. FEES Lett. gE, Z3B-

21?,.

f'lnLL¡ueLL, B. (l98l). Free radicals, oìrygen toxiclty and aging. ln:

Êge Pigments. Sobel, R.5. (Editor) pp l-EZ, ElsevlerlNorth

Holland Elomedical Press, Êmsterdam.

l'lRttn¡Etr., E. and EunenloeE, J.J"l.C. ( I gË4a). Oxygen toxicityr oxygen

radlcals, transitlon metals and disease. Eiochem. J. 213' I -14.

llntttuletr, E. and Gurren¡oc¡, J.l'l.C. ( l9Ê4b). Llpid peroxidatlon, ot4ygen

nadlcals, call damage and antloxldant tharapy. Lancet

i:1596- 159?.

183

Hnlu¡¡gtt, E. and Eunen¡gee, J.ll.E. ( l9E5)" Free Radicals in Ëlology

and lledlcine. Elanendon Pness, Oxfond.

l'lnnnts, E.D, ( l9Ë5). Pathogenesis of rheumatold arthnitis. ln:

Textbook of Rheumatologyr ãnd Ed. Kelley, hJ.¡|., llarrls, E.0.,

Ruddy, 5. and Sledge, C.E. (Editors) pp EË6-9 t5. ld.E.

Saunders Eompany' Phllädelphla.

llRnnts, 11.J., HEnP, Ê. and PlemRn, hJ. ( 19? I ). Depolymerlzation of

polysaccharldas thnough the gcneratlon of free nadlcals at aplatlnum surface: a norel procedure for the controlled

productlon of fnee-radlcEl oxidatlons. Êrch. E¡ochem. Biophys.

l{2,615-62U.

llRnntsott, J.E. and Scruttz, J. ( | 9?Ë). Studlcs on the chlorinating

actlrlty of myeloperoxldasa. J. Elol. Cham. 251' l3? l - 15?{.

lleys, Ê.0. and DonmRnoY, T.L. ( l9E I ). Llpld peroxidatlon ln lron-

orerloaded spleens. Clln. Sci. 60' 295-301.

Hormnnn, ll. and Scrmut, 0. ( 1980). The lnabillty of superoxide

dismutase to inhiblt the depolymerizatlon of hyaluronic acid

by ferrous ions and ascorbate. [ìnaefes Êrch. Kl¡n.

0phthalmol. 211, I I I - I Ê5.

l{otmes, 8., Pf,nK, 8.H., l'lfiLlsrfr, S.E.r 0ul, P.G., Netsott, D.L" and 6000, R.Ê.

( I g?0). Chronlc granulomatous dlscage ln fc'males: a

deflclcncy af laucocytc alutathione peroxldase. l{crrl Engl. J.

Hed. 283, Zl?-Z?1,

lloucr, J.E. and PeRncl, R.ll. ( 195?). The mcchanlsm of hyaluronida¡e

actlon. Biochlm. Elophys. Êcta 25' 555-58?.

l6Rnr, T., Kf,HEDR, H., l'lontucttt,5. and 0no, S. (lgBU). Ê rtmarkable in-

crËase of superoxlde dlsmutase actlvlty in synovial fluid of

patlents uith rheumatoid arthritls. Clln. 0rthop. Rel. Res.

1ã?, ZB?-2Ë?.

IE4

IutrRtn, 5. and EnRce, J.T. (1981) Eultiratlon l¡ vitro of myelobla¡tE

from human leukaemia. N. Y" J, Jlad. 64' 22?9-2282.

lyER, 6.Y.N., lsLRH, H.F. and 0unsrct, J.ll. (lgEl). Elochemical aspects

of phagocytosis. Nature (London) 192' 535-541.

JRsttt, t1.E. (l9EZ). Eeneratlon of lgG aggregates by the

myeloperoxldase-hydrogen pcroxldc systcrn. Êrthritls Rheum.

Z5(suppl), 53 [abst].

JeRrtOZ, R.bJ" and Foncrettt, E" ( l95l). Studles on hyaluronic acld and

related substances. lP. Periodate oxidation. J. Eiol. Ehem.

190, 55?-5'+6.

Jrnsen, C.E" ( 1949). The influence of metalllc lons on the vlscosity of

hyaluronic acid solutions. Êcta Chem. Scand" 5' 5E{-588.

Jonnson, D. and TnRuts, J. ( l9?9)" The oxldatire tnactlratlon of

human a- | -protetnase lnhlbltor. J. Elol. Chtm. Z3+, 1DZ2-

{0?Ë,

Jornsror, J.P. (1955). Ê comparlson of the propcrtl¿s of hyaluronic

acid from normal and pathologlcal human synovial fluids.

Elocham. J. 59, EZE:635.

Jonnsnn, J.P. ( 1955). Tha vlscoslty of normal and pathologlcal

human synorial fluids. Eiochem. J. 59, 635-85?.

Ktete, 8.8., lfcEoRD, J.H. and FnlOolrlcn, l. (t9?0). Superoxide-dlsmutase

from Escherlchia Eoll B. Ê ner¡ manganese-containing

enzyme. J. Elol. Chem. 2{5' 6l ?6-E I E l.

K¡tloee, E.lJ. and Fuooutcr, l. (19?5). Superoxlde, hydrogen peroxide

and singlet oxygen in lipid peroxidation by a xanthine oxidase

system. J. Biol. Chem. 250' 8A lz-Bïn.Kettoee, E.hJ. and Fntooulcr, l. ( 19??). Llposome oxidation and

erythrocyte lysls by enzymlcally generated superoxlde and

hydrogen peroxlde. J. Elol. Ehem. 232, E72l-Ê728.

185

KtttRtR, K., lloNMfi, Y., 0KfiyßMñ, H., 0cunt, K., llozumr, Ì1. and Suzurt, S.

( l9Ë'{). lncreased synthesis of hyaluronic acld by mouse

mammåry carcinoma cell rarlants uith high metastatic

potentlal. Eancer Res. 45, 1311-1354.

KLssRHorr, S.J. (l9E?). lodinatlon of bacterla: a bactericidal

mechanlsm. J. Exp. Hed. l?6, t0E3- l0?8.

Kt¡sRnorr, 5J. ( l9A0). 0xygcn metabollsm and the toxlc properties

of phagotytes. Ann" lntern" HÊd. 93, {Ë0-4Ê9.

Klesnnorr, 5J. and Rosel{, H. ( l9?E). Ethylene formatlon by

pol¡rmorphonuclear elucocytes" Role of myeloperoxidase. J.

Exp" I1ed. l1A, 498-506.

KnRtn, E.ll. and SReore, Ê.L. ( lg8 I ). The effect of oxldant injury on

the lymphocyte membrane and functions. J. Lab" Elin. Hed.

98, 69?-?05.

LReencnRnfz, E. ( l9E't). Free radlcals tn the auto-oxldatlon of

ascorbic acid. Êcta Chem. Scand. lïr SEZ.

Lnunent, T.E. ( 1955)¡ Studles on hyaluronic acld tn the ritreous

body. J. Biol Ehem. 216, 265-2? l.

LRunert, T.C. ( l9?0). Structure of hyaluronic acld. ln: Chemtstry

and llolecular Elology of the lntercellular l'latrlx. Vol. ll,Glycosamlnoglycans and Proteoglycans. Ealazs, E. (Edltor)

pp ?03-?4?. Êcademic Presl, London and l{eu York"

Leunem, U.E.E. and LRUnEHT, T.C. ( lg8 l). 0n the orlgin of hyaluronate

in the blood. Elochemistry Internatlonal 2, 195- 199.

LRunem, U.8.6. and TeneBLRo, Ê. ( lgË0). Determinatlon of hyaluronate

ln biological samples by a speclflc radloassay technique.

Ênalyt. Eiochem. 109, 586-394.

Lenntrern, Ê.L. (19?5). Blochemistry'. The llolecular Easis of Ëell

Structure and Function, Znd Ed. lrJorth Publishers, lnc.,

Ner¡ York.

r86

LtN, lrl.5. and ÊnusrRoNc, D.Ê. ( lg?E). Effects of supenoxide

dlsmutasa, dithiothreltol and formate lon on the lnactiyatlon

of papaln by the hydroxyl and superoxide radicals in aerated

solutions. lnt. J. Radiat. Eiol. 55, 251-23{.

LrnxEn, Ê., HEyEn, K. and Horrunn, D. ( l95E). The production of

unsaturated uronldes by bacterlal hyaluronldases. J. Elol.

Ehem. Zl9, 15-25.

Lrnren, Ê., l.lEyEn, K. and l,rJerssmRn E. ( 1955). Enzryatlc formation of

monosaccharides from hyaluronate. J. Biol. Chem. Zl3,237-

Z,IE.

Lonc, C.Ê. and Etetsrt, E.l'lJ. ( l9Ë0). Rate of reaction of superoxide

radlcal '"ulth

chlorlde-contalnlng specles. J. Phys. Ehem. E{,

555"

LottcRs, H.0. and Heyen, K. (l3A 1)" Sequential hydrolysls oî

hyalunonata by p-glucuronldase and p-N-acetyl

hexosamlnldase. Blochem. J. l9?, 2?5-242.

Luno-0lesEN, K. ( l9?0). 0xygen tenslon in synorial fluids. Êrthritis

Rheum. 15, ?69-??6.

Lunec, J., HfittoRRNr S.P.r lrJrre, Ê.8. and Donmnnoy, T.L. ( l9E I ). Fnee

radlcal oxidation (peroxidation) products Ìn serum and

synorzial fluld in rheumatoid arthrltis" J. Rheumatol. E, 253-

215.

Lyttctt, R.E. and Fntooutcr, l. ( t 9?E). Effects of superoxide on the

erythrocyta membrane. J. Eiol. Ehem. 255, I ESE- I845.

l'lRnn, T. and Kelutt, D, ( l33E). llaemocuprefn and hepatocuprlen,

coppÊr-proteln compounds oî blood and llrer ln mammals.

Proc. Roy. 3oc., Ser.E:Eiol. Scl. l2E, 305-315.

HRnrtuno, 3.L. ( lgEZ). lluman coppËr-contalnlng superoxlde

dismutase of hlgh molecular rrlelght, Proc' llatl. Êcad. Sci.

u3Ê ?9, ?811-?83Ê.

187

J'1Rnsn, J,M., hltEsxnt, O.hJ., 6RtE, 5., l"lutR, I',l. and Hntnt, R.N. ( l9?g).

Synthesis of sulphated proteoglycans by nheumatoid and

normal synorrial tissue in cultune. Ênn. Rheum. Dis. 58, 166-

r ?0,

llnscrorr, E.Ê. and Elncreunn, E.L. ( l gES). l'{utrltlon and rheumatic

dlseases. ln: Textbook of Rheumätology, ?nd ed. Kelley, lÁ|.H.'

llarrls, E.D., Ruddy, S" and SledgË, 8.8., (Edltors) pp 552-560.

lrl.E" Saundars, Phlladelphla.

HRtsuoR, 1., 0Kn, T., TfiNl6ucHl, N., FunuYRmn, H., KRoRt'tñ, 5., ÊnRsllme, S. and

Hmsuynno, T. ( l9?E). Leucocyte glutathione penoxidase

deflciency in a male patient r¡ith chronic gnanulomatous

dlsease. J. Pedlat. 88, 5A I -583.

llcCRmv, D.J. ( ISES)" S¡morrlal fluld. ln: Êrthrltls and Êllled

Eondltlons, lOth Ed. l'tcEarty, DJ. (Edltor) pp 5a-75. Lea &

Febiger, Philadelphia.

llcCleRn, D. and llRte, C.hJ. ( l9'î0). lluclnase and tlssue Permeability.

Natune l't5, 86?-Ë88.

PlcClenn, D. and llRte, C.ld. ( l9{ I ). Studles on diffusing factons: the

hyaluronldase actirlty of testlcular extracts, bactrial

culture flltrates and other agents that increasa tissue

permaablllty. Blochem. J. 55, l59- I85.

HcEono, J.H. ( l9?4). Free radlcals and inflammatlon: protection of

synorzlal fluld by suparoxide dlsmutase. Science lE5' 529-55t.

ilcEono, J.J"l. and DfiY, E.D. ( l9?E). Superoxlde dependent production

of hydroxyl radlcal catalyzed by iron-EDTÊ complex. FEBS

Lett.86, 133-11?.

f'lcEono, J.11. and Fuooutcr, l.: The reduction of cytochrome C by milk

xanthlne oxidase. J. Blol. Ehem. l9EB; Z'13: 5?55-5?68.

I8E

l'lcNelu, J.D., EHotrI, D.Ë. and SrosEy, J.L. (lgEE). Effect of a high

molacular rrreight hyaluronic acid prËparation on the

stimulation of polymorphonuclear leucocytes. Fed. Proc. {5,

658 [abst].

llcNen, J.0., CLELfiHo, L.G. and E¡ns, hJ.l{. (lg8g). The cytotoxicity of

oxygËn derired fnee radlcals. Êust. N.Z. J" l'led. 15, ZZs

Iabst].

HcNen, J.D., hlrEaKrt, 0.hJ., Ems, hJ.l1. and EuetRno, L.g. ( 19ES).

Depolymerization products of hyaluronic acid after eilposure

to oxygen-derlred free radicals. Ênn. Rheum. Dis. 1+, ?tE-

?89.

lfeao, J.F" ( lg?E). Free radlcal mechanlsms of llpid damage and

consÊquÊrrceg for cellular membraneg. ln: Free Radlcab in

Elology, Vol, l. Pryor, hJ.Ê. (Edltor) pp 5l-EE. Êcademic

Press, Ner¡ York.

lfeven, K. ( | 94?). The biological significance of hyaluronic acid and

hyaluronldase. Physiol. Rev" 2?, 335-359.

Heven, K. and PRtmen, J,hJ. (1934). The polysaccharide of ritreous

humor. J. Biol. Chem. I B?, 829-85'1.

Hrven, K. and Pntmtn, J.lrJ. (1958). tln glycoproteins. ll. The polysac-

charldes of ritneous humon and of umbilical cord. J" Biol.

Chem. I l{, EÊ9-?05.

llEyEn, K., DuBos, R. and St,tyttt, E.N. ( 193?). The hydrolysis of the

polysaccharide acids of vltreous humon of umbillcal cord and

of streptococcus by the autolytic enzyme of pneumococcus.

J. Elol. Chem. I tE, ? l-?8.

Hrxult¡-TnKfrGßKr, Y. and Tools, E.P. ( l9Ë I ). llyaluronate-proteln

complex of rous rarcoma rlrus-transformed chlck embryo

flbroblasts. J. Eiol. Chem. 258, 8'{85-E'tEg.

t8g

Honeotsse, J.8., BRfiouET, P. and BoneL, J.P. (lgBl). 0xygen-free

radlcals as mediators of collage breakage. Êgents Êctions

15, 49-50.

llotmorsse, J.Ê., ERf,ouET, P., Rnnooux, Ê. and Eonet, J.P. ( l9E5). Non-

enzymatlc degradation of acld-soluble calf skln collagen by

superoxide lon: protectlre effect by flavnoids. Elochem,

Pharmac ol. 32, 53-58.

lloneRn, Ê.R., EoNE, R.L. and Eleenr, T.J'|. ( l9?E). The mechanism of DHÊ

strand breakage by ritamln C and superoxide and the

protective roles of catalase and superoxide dismutase.

Nucleic Êcid Res. 3, 1 159- I I {9.

lloven, J"D. and lseett, 11.3. ( I gSE). Determination of reducing sugars

and reduclng end gpoups in polysaccharides by neaction r¡ith

carbon- l {-labelled cyanlde. Ênalyt. Ehem. 50, l9?5- l9?8.

l"lutnotn, K.D. ( 1966). Ferritin in synorlal calls in patients r¡ith

rheumatoid arthrltls. Ênn. Rheum. Dls. 25, 3E?-401.

Hutnoen, K.D. and Senatonr 6.E. (1968). lron ln the s¡,rnovlal membnane

ln rheumatold arthrltls and othan Joint diseases. Ênn.

Rheum. Dis. 2?, 3E-{8.

Hutttnnx, F. and HutttttRx, 6.L. ( l9Ë5). Tyroslne-tyrosine Eross links in

lgG. Generation by peroxidation and occurnÊnce in RÊ synorrial

fluids" Êrthritis Rheum. ZE(suppl), S lE [abst.].

llyens, L.S. (1980). Free nadical damage of nucleic acids-and their

components: the dlrect absorption of enËrglr. ln: Free

Radicals in Eiology. Pryor, hJ.Ê. (Ed¡tor) pp 95- l l'1. Êcademic

Press, Ner¡ Yonk.

Nm¡le¡-Ror, E., Sunoetno, L. and Jonsson, E. ( 1963). Permeability of the

synorzial membrane to proteins. Êcta Rheum. Scand. 9, 28-

52.

r90

Nrentuls, Ê.hJ. ( l9E I ). l/ltamin E and iron. Nerrr England J. Hed. 504,

t?0-t?r.

I'lltssott, R., PtcK, F.J1. and Bnny, R.C. ( l969). EPR studies on reduction

of oxygen to superoxide by some biochemical system.

Elochim. Eiophys. Êcta 132, l45- l'16.

Nonton, J., UnBfiH, J. and HnnouoRs, Ê. ( | 9BZ). Ê failure to obserre

enhanced diffuslrzlty of glucose in a matrix of hyaluronic acid.

J. B¡ol. Ehem. 25?, 14151- l'{ 135.

0ol¡.yle-Hnnnts, D.J" and FonnfistEn, V.L. (1980). ï¡novial iron depoEition

in osteoarthritls and rheumatoid arthnitis. J. Rheumatol. ?,

30-56"

Ocsron, 3. and SrenmRr, T.F, ( 1959). Degradation of the hyaluronic

acid complax of s¡morzial fluid by protealytic Ënzqet and by

EDTÊ. Eiochem. J. 72' 301-505.

0esron, Ê.8. and Stnnttn, J.E. ( 1950). 0n the state of hyaluronic acid

in s¡rnorrial fluid. Biochem. J. {8, 56'f-5?6.

0cstott, Ê.6. and StRuen, J.E. ( l95lJ. The dimensions of the particle

of hyaluronic acid complex in synorial fluid. Eiochem. J. 49,

5Ê5-590.

Ptnr, J.T. and Jottttsott, ll.J. ( l9'19). Ê submlcrodeterminatlon of

glucose. J. E¡ol, Ehem. I E l, 149- l5l.

PRnrs, D.Ê., BULKLEY, 6.8., Gnnncen, D.ll., HñMILT0N' 5.R. and llcC0nD, J.11.

(l9EZ). lschemlc injury in the cat small lntestine. Role of

superoxide radlcal¡. Ba¡tnoenterology 82' 9- 15.

PrRn¡lnClR-PnOOUCT lnfonmmOn. Eel Filtration - Theory and Practice.

Pharmacia: Laboratory Separation Division, Plscatar.layr Netl

Jersey, U.S.Ê.

Pxtueson, L.ll. and Scru.lRmz, H.E. ( l9B't). Subcellular locallzation of

hyaluronate synthetase in ollgodendroglioma cellc. J. Eiol.

Ehem. ZSg, 5011-5023.

191

Pr6mRn, lrJ. and Rtzut, 5. ( 1959). llyaluronlc acld and the ORD

neactlon. Elochem. Eiophys. Res. Eommun. l, 59-{2.

PleNnn, l,rJ., Rtzut, 5. and Hortty, ll.L. ( l3E l). Depolymenization of

hyaluronic acid by the ORD reaction, Êrthritis Rheum. 4,

u +0-252"

Pnexm, P. ( l3A3). Synthesls of hyaluronate In differerntlated

tcratocarcinoma cells: mechanism of chain gror.r.lth. Eiochem.

J. ?l l, l9l-198.

Pnenm, P. ( | gB{). llyaluronate is synthesized at plasma

membranes. Eiochem. J. 220, 59?-800.

Pnvon, lrJ.Ê. ( l9?E). The role of free radical reactions in biological

systems. ln: Free Radlcals in Eiology, Pol. l. Pnyor, l,rJ.Ê.

(Edltor) pp l-49. Êcademlc Press, Ner¡ Yorì<.

Raottt, E.L., StIRNH, D.Ê. and lrJ¡¡ssEn, P.Ê. ( l9?0). Separation of a

hyaluronate-free lubnicatlng fraction from synorial fluid.

Natune 228, 5??-5?8.

RRottt, E.L., PÊuL, 1.L., SURtttt, D.Ê. and ScxorrsrngDnr E.S. (19?l).

Lubrication of synorial membrane. Ênn. Rheum. Dis. 50, 5e2-

. 3U5"

RReRn, C. and Heyen, K. (1949)" The hyaluronic acid of synovial fluid

in rheumatoid arthrltls. J. Elhl. lnregt. 28, 5Ë-59.

RRcRtt, C., DoHtRu,8.P., Coss, J.Ê. and 6nuetr, Ê.F. (131?r. Effects of x-

ray irradiation on viscosity of synovial fluid. Proc. Soc. Exp.

Eiol. Hed. 68, l?E-l??,

Rtooy, P.S. and Leulls, H.. ( l gEg). The adrerse effect of intr aveno.Js

irõn dextran in rheumatoid arthritls. Êrthritis Rheum. lZ,

454-45?.

R¡HNcnoNR, 5., SrEsJo, E.K. and S¡,lnr, 0.5. ( l9Ë0). Rererslble lschaerynia

of the braln: blochemical factors influencfng restitution.

Êcta Physiol. Scand. {9Z(suppl), 155- l'10.

r92

Retssre, J.L., STR0M¡N6ER' J.L. and Lttotn, L.P. ( 1955)" Ê modlfled

colorlmatric mathod for the estimatlon of N-acetylamino

sugars. J. Eiol. Ehem. Z l?' 959-986.

Rtcnent, D.Ê. and lrJesr¡nrtelû, l,lJ.U. ( tgS l). Xanthine oxldases in

different species. Proc. Soc. Exp. E¡ol. Hed. ?Ê,252-251.

RoseffisoN, hJ.E., Rotes, H.hJ. and Eauen, hl. ( 1953). Enzymatlc

hydrolysls of mucln by a bacterlal enzyme and by t¡itamin E.

Êmer, J. Physlol" l2E, 609 labst.l

RoggntTsoÌ{, hJ.E., Role s, l1.lJ. and ERuen, Ld. ( | 9'10). Huclnage: a

bactçrial enzyms ¡¡hlch hydrolysês synÊvial fluid mucln and

other muclns. J. Blol. Chem. l3,3' ?E|-Z?E.

Rogeprson, hJ.B., Roles, l1.hl, and"Enuen, hJ. (19'îl)' Thc degradation of

mucins and polysaccharides by ascorbic acld and hydrogen

paroxida. Elochem. J. 55' 9CI5-908.

Roelnson, H.E. and Hottloou, J.J. (l9?5). The alkallne clearage and

borohydnlde reductlon of cantllaga proteogþcen. Biochem. J.

155, ,{5?-+?0.

RoonRn, 6.P., EENEDEK, T.6. and Pnnetrf,, hl.E. (1386). The early history

of synorzial fuoint) fluid. Ênn. Intern. lled. 65' E? I -E{2.

RonRt, P.Ì1. ( lgEg). The elutlon of t'Cr from labellcd leucocytes - a

nËuJ ässây. Blood 53' 408-{ 15.

Root, R.K. and lletcRtF, J.Ê. ( l9??). H"O, release from human

. granulocytrr durfng phagocytoslg: rclatlonghlp to rup+oxldc

anlon formatlon and cellular metabollsm of llr9rt studles r¡lth

normal and cytochalasln E trrattd crlls. J. Clln. lnrest. 80,

r 286- I 2?9.

Rores, lf.hl. and Enuen, U. ( 1953). Synovial Fluid Changes in Joint

Dlsease. llarrard Unlrerslty Press, Eambrldge, HÊ,

r9g

Roser, E.H. and KLegnnorr, 5.J. ( lg??). Formation of singlet otrygen

by thc myeloperoxldase-medlated antlmlcroblal system. J.

Blol. Chem. 252, {805-'1810.

Rosttt, 6.J",l., BRneen, 11.J. and RRucxmnn, E.J. ( l9E5). Disnuptlon of

crythrocyte membranal onganizatlon by superoxide. J. Eiol.

Ehem. ?58, ZZ?5-2ZZË.

Rosenrrat, H.8., EEnnev, J.P. and Êtetn, E.E. ( 19'lE). Ê sensitlre auto-

mated procedure for the detenmination of glucuronic acid.

Conn. Tiss. Res. 4, 155- lE t"

Roltto, 8., Ënfiy, R.E. and Flrtoen, E.H. ( l9?2). Ê pulse radiolysis

study of superoxide dismutase. Eiochim. Eiophys. Êcta 268'

EB5.EBg.

SRRnnt, ll., TrMMr, H., [/u0nt0, E. and PerrnHEH, R" ( l9E0). Distribution of

glycosamtnoglycans in nheumatold cultures and effects of

cortlsol on lt. Scand. J. Rheumatol. 9, I l- 16.

Sncrs, T., l{0L00tJ ) C.F., CnRooocr, P.R., EobJERsr T.L. and JRcoB, l'1.S.

( t 9?E). 0xygelr radlcals mediate endothellal cell damage by

complement-Etimulated granulocytesz an in vltro model of

immune vascular damage. J. Elin. lnrest. E l, 1 lE I - I lE?.

Snun, J1.L. and llcEono, J.H. ( l9?5). Protection of phagocytosing

leukocytes by suparoxide dismutase. J. Clin. lnrest" 56,

t5l9-1525.

SRto, l. and NlttR, Y. The oxygen radical scarenging actlrzlty of

hyaluronic acid and its comPonÊnt sugars, in press.

SeRnnn, Ê.J. and KRnnousrv, 11.L. ( 1959). The biochemical basis of

phagocy-tosls. l. Hetaabolic changes during the lngestion of

particles by polymorphonuclear leucocytes. J. Biol. Ehem'

231, I 355- l1EZ.

194

Scronenum, G.R. and ExnncE, E. ( l9?E). Êatalase. ln: The Enzymei,

5rd Ed. Eoyer, P.D. (Edltor) Pol. 15, pp 365-459. Êcademic

Press, Ner¡ York.

ScrnRursrRTTER, 1.U., lilNsHfirr¡, 0.8., l'lystoP, P.Ê., SPnßGE, R.6. and CocrRRNE,

Ë.6. ( l9E5). Elutathlone cycle actlrrlty and pyrldine nucleotide

lerels ln oxldant-induced inJury of cells. J. Elln. lnvest, ?Ê,

I l5l-l t39.

Scor, J.E. ( 1980). Allphatlc ammonlum salts ln the a;.say of acidlc

polysaccharldes from tlssucg. Methods Elochern. Ãnalysis Ê,

I 45- r 9?.

Scorr, J.E. and Pnag-Tto¡1fis, D.P.P. ( l9?E). Spectrofluorimetric

detectlon and measurement of hydroxyl radicalc in periodate

solutlon. Earbohydrata Ras, 52, Z l4-? I E.

Scotr, J.E. and Ttcuett, HJ. ( lg?3). Periodate-induced riscosity

decreases in aquÊous solutions of acetal and ether linked

polymers. Êarbohydrate Res. 28, 53-59.

Scorf, J.E., TTcLIELL, ll.J. and Sn.lo¡nfi, UJ. (1372). Loss of basophilic

(sulphated) matenial from sections of cartilage treated r¡ith

periodate solutlon. l'listochemlcal J. 4, t55- 16?.

Scon, J.E., HEÊTLEv, F., Hoogcnorr, D, and 0tRuesen, ll. ( | 9E l).

Secondary structures of hyaluronate and chondroitin

sulphate. a lH n.m.r. study of NH slgnals in dimÇthylsulphoside

solutlons. Elochem. J. 199' 829-432.

Scuooen, P.R., ÊL-Ttmtmt, D., HcHunnfiY, lrJ., hlnm, A,8.,2008, E.E. and

Donmanov, T.L; ( l9?A). Serum coppÊr and rçlated rarlables in

rheumatoid arthrltls. Ênn. Rheum. Dis. 31, E?-79.

5e ent, Ê.lrJ. ( l9?'{). Nltroblue-tetrazolium tests. Lancet il, IUA-

1252.

r95

SerRRtR, P.0., KfinKKRTNENT J., LEHToNEN, Ê. and llRxtsñRfi, P. ( lg?Z).

Ehondroitln sulphate in normal and rheumatoid synorzial fluid.

Elln. Ehim. Êcta 58, 549-555.

SreexRn, J.K., ÊTKrNs, E.D.T. and Nl¡ouszYNsKl' T.A. ( l9?5). X-ray

diffraction of polysaccharldes. Tr¡o-dimensional packing

schemes for threefold hyalunonic chalns. J. Hol. Elol. 3l ' 155-

I 83.

SrEs, ll. and Eaoenfis, E. ( t 983). Biological basis of detoxlcation of

oxygen free radlcals. ln: Biological basis of detoxication.

Caldrr.rell, J. and Jakoby, hJ.E. (Edltors) pp lBl-Zl l. Êcademic

Press, San Dlego.

SnpnnRrnR, P", DuNsToHE, J.R. and 0egfor, Ê.8. ( l9E?). Fractionation of

a hyaluronic acld prÉparätion in a dansity gradlent: the

lsolation and ldentlflcatlon of a chondroitln sulphate.

Blochem. J. l0+, {0{-409.

SrmcrotJrrz, L. and Selteeno, l. (19?9). Evidence for the role of

superoxide radicals in neutrophil-medlated cytotoxicity.

lmmunologg 5?, 301-509.

Srt-lon, R.ll., ScoGGtH, E.l.l. and Pnrrglsot{, B. ( l9E l). llydrogn peroxlde

cäuses fatal inJury to human fibroblasts exposed to oxygen

radlcals. J. Biol. Chem. 256' ? lE l-? lEË.

SrRnse, E. and SunoetRo, L. ( l9'{5). Oxidatlre bneakdor¡n of

hyaluronlc and chondroitln sulphurlc acld. Êcta Physiol. Scand.

E,5?-51.

Smeosnoo, 8., PEnroFT, l''l., Enlrssott, 5., FRfrsEn, J.R.E. and LnunENTr T.C.

( l9E'1). gtudles þ ritro on the uptake and degradatlon of

sodlum hyaluronate in rat lirer endothellal cells. Elochem. J.

2?1, El?-Ê?E.

196

Sooen, P.0., NoRD, C.E., LuttoBuRo, 6. and K,leutmRn, 0. ( l9?0J.

lnrzestigations on lysozgffie, protease and hyalunonidase

actirzity in extracts from human leucocytes. Êcta Ehem.

Scand.2{, l29-158.

Sroou¡,tu.gn, Ê.C. and DonrMfiN, Ê. ( l9?0). The bios¡mthesis of

hyaluronlc acld ln group Ê streptococcl. ln: Chemlstny and

Holecular Elology of the lntercellular Hatrlx, Pol. 2. Ealazs'

E.Ê. (Edltor) pp 7A3-791. Êcademlc Press, Het¡ York.

SunoatRo, L. ( lg53). Studles on hyaluronic acid in synovial fluids.

Êcta Soc. Hed. Upsalla 58, I 15-256.

SunoeuRo, L. and ERtRzs, E.Ê. ( lgËE). Ehemical and physical changes of

glycosamlnoglycans and glycoprotelns caused by oxidation-

rcducction systems and radlatlon. ln: The Amino Sugars, Pol.

28. l'letabolism and lntenactions. Balazs, E.fl. and Jeanloz,

R.hl. (Edltors) pp 229-250. Êcademic Press, Neu York and

London.

Swen, E.R. and lln.lno, 6. ( l9E'l). Ultrastructure of the joint capsule

in the rat: presence of t¡¡o kinds of capillaries. Nature 202,

920-92 l.

SunRnrHnßN, 1"1., Sotot'tolt, 5.0., hJtLltfiMs, P.5., RuElN, 4.L., N0u06n00sKYr Ê.

and SrenzEL, K.11. (198.1). liydroxyl radlcal s:avËngers inhibit

human natural killer cell actirzity. lïature 381r 21ã-??A.

Sl¡RNn¡ D,Ê. (1968). Studles on hyaluronic acid. ll. The protein

component(s) of rooster comb hyaluronic acid. Eiochim.

Biophys. Êcta 160, 96- 105.

SrrtRnx, D.Ê. and Rfioltt, E.L. ( 197?). The molecular basis of articular

lubrication. l. Purification and properties of a lubricating

fractlon from borine synovial fluid. J. Eiol. Ëhem. Z1?, AEE9-

B0?3.

I9?

SuRnn, D.Ê., I'|EN0nEH, R.8,, Rnotn, E.L., SoTMfiN, 3.L. and Duon, E.A. ( l9E l).

The lubrlcating actirlty of synovial fluid glycoproteins.

Êrthrltis Rhaum. 21, 22-38.

SttRnn, D.Ê., SrLuER, F.ll., Stott,tRlt, S.L. and Henmnnn, l'1. (lgE?).

Heasurements of reduclng end groups on borzine ritreous-

humour hyaluronic acid by reaction r¡ith l"E7-cyanide.

Eiochem. J. ?87, '{09-{ l'1.

St*lRtttt, D.Ê., ELocH, KJ., Sutnoett, D. and Srone, E. ( l9E{). The

lubricatfng actlvity of human synorial flulds. Êrthritls

Rheum. 2?, 552-558.

TRuaen, Ê.1. and ERelon, B.l'1. (19??). Eridence lor hydroxyl radlcal

productlon by human neutrophlls. J. Elln. lnvest. 60' 5?{-

5?9.

Toote, E.P. ( I gEZ)" Derzalopmental nolc of hyaluronätê. Conn. Tlss.

Res. 10,95-100.

TnuexRrt, P.5. and HcCRnrv, D. ( 19? I ). Synorial fluid pll, lactatc,

oxygen and carbon dioxlde: partlal pressure n various joint

dlseases. Êrthritls Rheum. 14, 4?5-443.

TsrcRltos, C.P., l/yntos, D.l'|. and KRt-pf,Nts, D.L. ( l9EE). Rooster comb

hyaluronate-protein a non-coralently llnked complex.

B¡ochem. J. 235, l17-123.

Unoennttt, E.E. and Toote, E.P. ( l9?9). Einding of hyaluronate to the

surface of cultured cells. J. Eell BIol. 82, {?5-48'1.

PR¡nsnnrn, 8.K., lteur-ÊonHyfrN, hJ.R., EnnYnln, P., PnGtN, Ê.Ê. and Gntetnro,

Ê"1. (19Ël). Three dlmanslonal structure of the enzyme

catalase. Natune 295, 'l I l-'t 12.

PgnrntnsuonftMfrNrñN, K. and JosERtt, K.T. ( I g??). Êctlon of slnglet

oxygÊn on collagen. lnd. J. Elochem. Elophys. l{' 2l?-2?0.

198

Puonro, E., EINoLR, s., HfrKKfiRfitNEn, s. and pe nntnen, R. ( lg8z). synthesis

of underpolymerized hyaluronlc acid by fibnoblasts culturedfrom rheumatoid and non-rheumatoid synorritis. Rheumatol.

lnt. 2, g?- l0Z.

l,rJeotocr, D.J,, PHtLLtpg, 8.0., Dautes, Ê., EoRHf,LLy, J. and l¡Jytt-JoHEs, E,

( l9E3). Depolyrmertzaflon of godtum hyaluronatc durrng

freeze drying" lnt. J. Blol. Hacromol. E, I EE- lËÊ.

lrlelss, S.J. ( lg80)" The role of superoxidc ln the de¡truction ofeythrocyte targets by human neutrophils. J. Eiol. Ehem.

255, 99 t?-ggt?.

hlulss, 5.J., RusrfiGt, P.K. and Logu6Lt0, Ê.F. ( l g?E). Human granulocyte

generation of hydroxyl radical. J. Exp. Hed. r4?, gr6-sag.

l,rJetss, sJ., YouN6, J., Loeuetto, 4.F., sLruKfr, Ê. and N¡HSH, N.F. ( lgE I ).

Role of hydrogen peroxide in neutrophil-mediated destructionof cultured endothelial cells. J. clln. Inrest. 68, ? l4-?z l.

hJerssmRrn, 8., ZuRrER, R.E. and llorrsTurN, s. ( lg?z). Leucocyte

pnoteases and the lmmunological release of þsosomalenzymes. Élmep. J. Pathol. 6Ë, 539-560.

l,rJelss¡lnnn, G., KoRcHfrK, l1.l't., Punez, ri.D., 3M0LEN, J,E., fìotosrurn, 1.11. and

llorrsntn, s.T. ( l9?9). LeukocytÊr ä5 secreiory organs of

inflammatlon. Êdr. lnflanrmation Res. l, g1-l lZ,

|rJrerrtn, 0.1¡,l. and Hurp, ll. ( lg?s). lnfluence of the cells on thepericellular ertvironment. The effect of hyaluronic acid on

proteoglylan synthesis and secreilon by chondrocytes ofadult cartilage. Phil. Trans. Royal soc. Lond. E zllr zgS-zgl,

hJrmen, hl.l., SMFH, PJ, and Ên¡rolT, s. (lg?s). llyaluronic acid:

structure of a fulþ extended 5-fold helical sodium salt and

comparison uith the less extended l-fold helical forms. J.

llol. Elol. 99, 219-255.

r9g

lrJrnrensounnE, E.E. ( lg?9). Comparison of suparoxlde and other

reduclng agents in the biological productlon of hydnoxyl

radicals. Elochem. J. I E2' EZ5-628.

hlOfe, S.F., HfiLLtbJELL, 8., Rlcttt.,tOno, R. and SfOlrlRoNECKr lrJ.R. (l3E l). The

role of superoxide and hydroxyl radicals in the degnadation

of hyalunonlc acld tnduced by metal Ions and by asconblc acid.

J. lnorganic Blochem. I {' lZ?- 154.

YoUne, 5.P. and Êtsen, P. ( | $EZ). The llrzer and lron. ln: The Lirzer:

Elology and Pathobiology. Êriäs, 1., Popper, ll.r Schacter, D.

and Shafrltz, D.Ê. (Edltors) pp 393-40'1. Raren Press, f'ler.,.l

York.

Yurcr, l'1. and Frsnman, hJ.l1. ( lSEg). Purlflcatlon and charactcrlzatlon

of lcach hyaluronlc acld-endo-p-glucuronldagc. J. Elol. Chem.

238, lA?1-18?9.

ZuRlrten, N. ( l9?5). The lntmunopathology of joint lnflammatlon in

rheumatoid arthrltiE. Êdr. lmmunol. l6' 285-338.

P. ?, r"cact¡on (3)

P. 10, reectlon llne I

P. 18, line ?

P. I g, linc 28

P. 139, linc I {

P. ll0, Table 5-l

P. l{1, line 2

P. I {8, line 22

P. 90, lnc 2l ([Fee+] = 50 ¡rL) should read ([Fe?+l = 50 ¡rl'l)

P. g?, |:me 29, Insert aftcr'llorgan-Elson assay'The reason that ther.e is a discnepency betr¡een thedegree of Park-Johnson reactlvlty for a glvenmolecular r.reight r.emains unclear, although it meybe related to the inherent inaccuracy in using thebroad pcak, obtained after gcl chromatography, tom?¡rur1r molcculan urcight.

P. 128, linc 15, Inscrt after 'but not Peak B'Peek Ê is also completeþ susceptible to digestionr.rlth hyaluronate lyase fnom streptomyceshyeluronolyticus (d¡ta not shor,rn).

ERRâTA SHEEÏ

0a- + Ha0a 0a + 0H.+ 0H-

zltÊDPH + 20¿ zltÊDP+ + 202 + HaO

'Glutatione' should r-ed'Glutathione'

'to bc ån an antioxidant' ¡hould re¡d 'to be ¡nantioxidant'

inscrt 'in' after 'irægularities'

headtng factor ls 'x l0-6'

'r',epolymenized' shou|d read'nepolymerize'

'so' should rced 'too'