distribution and characterisation rat choroidal mast cells · british_journalofophthalmology...

British_Journal ofOphthalmology 1994; 78: 211-218

ORIGINAL ARTICLES - Laboratory science

Distribution and characterisation of rat choroidalmast cells

R J Steptoe, P G McMenamin, C McMenamin

AbstractDespite the implication that choroidal mastcells are involved in the onset of experimentalautoimmune uveoretinitis (EAU), a widelyused animal model of uveoretinitis, little isknown of these cells. In the present study thedistribution, total number, regional density,and phenotype of choroidal mast cells wereexamined in Lewis, Wistar Furth, PVG/c, andbrown Norway rats. Choroidal mast cells werepredominantly associated with arteries andarterioles of more than 30 pm diameter whichlie in the outer (sclerad) choroid. The densityof mast cells was greatest in the posteriorchoroid with density diminishing anteriorly.The choroid of male Lewis rats containedsignificantly greater number of mast cells thanthat of females (p<001). Histochemical(Alcian blue/safranin) and immunohistochemi-cal (anti-rat mast cell protease I and II mono-clonal antibodies) studies revealed choroidalmast cells were of the connective tissue type.However, granule proteinase content appearedless than that of well characterised connectivetissue mast cell populations such as those inmesentery and skin. Lewis rats exhibited thehighest density of choroidal mast cells (23.6(SD 1.2)/mm2), Wistar Furth approximatelyhalfthat ofLewis (13-5 (0.7)/mm2) while PVG/cand brown Norway rats had very low densities(3.06 (0.3); 1-95 (0.2)/mm2 respectively). Thesestudies provide valuable choroidal mast celldata for rats which may have implications forour understanding of experimental models ofintraocular inflammation and clinical uveitis.(BrJr Ophthalmol 1994; 78: 211-218)

Department of Anatomyand Human Biology,University of WesternAustralia, Nedlands, WA6009, AustraliaR J SteptoeP G McMenamin

Division of Cell Biology,Western AustralianResearch Institute forChild Health, PrincessMargaret Hospital forChildren, Subiaco, WA6008, AustraliaC McMenaminCorrespondence to:Dr P G McMenamin,Department of Anatomy andHuman Biology, University ofWestern Australia, Nedlands,WA 6009, Australia.Accepted for publication23 September 1993

Mast cells are central to IgE mediated immediatehypersensitivity reactions and have been increas-ingly implicated in a wide range of inflammatoryresponses. 2 Mast cells may also participate in a

variety of normal physiological processes.3The presence of mast cells in the lid, con-

junctiva, and uveal tract of the mammalian eyehas been reported for a number of speciesincluding rats,' guinea pigs, rabbits, monkeys,4and dogs.7 Mast cells within human ocular tissuehave not been widely studied; however, they are

present in variable numbers throughout theuveal tract - namely, the iris,8 ciliary body,9 andchoroid."

Allansmith and associates6 estimated the num-ber of mast cells within various tissue compart-ments of the rat eye and adnexa by analysis of

tissue sections. These authors reported the vastmajority ofmast cells were located in the externalstructures - the eyelids, conjunctiva, and orbit.Mast cells observed within the uveal tract werelargely restricted to the choroid.6 While earlyinvestigations primarily studied the presence andvariation in the number of choroidal mast cellsbetween species, some information pertaining tothe distribution of mast cells within the choroidhas been reported. Smelser and Silver,5 using anovel flat mount technique, reported an antero-posterior gradient of choroidal mast cell densityin a number of species. Godfrey9 described ratchoroidal mast cells in a periarterial location.While the distribution of mast cells within extra-ocular tissues is consistent with the usualstrategic location below surfaces exposed to theexternal environment" and their role in IgEmediated responses, the function of mast cellswithin the uveal tract is still unclear.

Distinct mast cell subpopulations are distin-guishable by functional and morphological char-acteristics.'213 In the rat, two phenotypes arewidely recognised: the connective tissue mast cell(CTMC), found predominantly in the skin,peritoneal cavity, and other areas of connectivetissue'4 and the mucosal mast cell (MMC) whichoccurs in the mucosa of the gastrointestinal'5 16and respiratory tracts.'7 18 These phenotypes maybe partially distinguished by the histochemicalstaining characteristics of their proteoglycancontent,'6 but more definitively by analysis ofserine proteinases within the intracellular storagegranules.The phenotype of choroidal mast cells has not

been established to date, although it may beimportant to the role of these cells in intraocularimmunological processes and inflammation dueto, for example, the differing mediator produc-tion20 and T cell dependence'4 of the two sub-populations.

Variations in choroidal mast cell numbersduring the course of experimental autoimmuneuveoretinitis (EAU) in Lewis rats led de Kozakand associates2' to propose that choroidal mastcells have a role in the onset of this inflammatorydisease. Supportive evidence of such a role wasprovided by Mochizuki and colleagues22 whodemonstrated a positive correlation betweenchoroidal mast cell numbers and EAU sus-ceptibility in a number of rat strains. Highlysusceptible strains, such as Lewis, had up to afivefold greater density of choroidal mast cellsthan strains of low susceptibility.As a prerequisite to investigating the function

211

on June 29, 2020 by guest. Protected by copyright.

http://bjo.bmj.com

/B

r J Ophthalm

ol: first published as 10.1136/bjo.78.3.211 on 1 March 1994. D

ownloaded from

http://bjo.bmj.com/

Steptoe, McMenamin, McMenamin

of choroidal mast cells in intraocular inflamma-tory disease it was considered appropriate toobtain more detailed information on the normaldistribution, location, and phenotype of thesecells in a number of rat strains.

Materials and methodsMale and female Lewis and female PVG/c,brown Norway, and Wistar Furth rat strains(specific pathogen free), aged 10-12 weeks, wereobtained from the Animal Resources Centre,Murdoch, Western Australia. Six animals ofeachstrain were used to obtain choroidal mast celldensity data from flat mounts. For all strains,other than Wistar Furth, both eyes wereexamined and an animal mean determined. ForWistar Furth, one eye from each of six animalswas used to generate choroidal mast cell densitydata. Three animals each of Lewis, WistarFurth, PVG/c, and brown Norway strains wereused to determine the histochemical andimmunohistochemical (albino strains only) stain-ing characteristics of choroidal mast cells inparaffin sections. A further group of Lewis ratswas used for immunohistochemical staining oftissue whole mounts (choroid and mesentery).

Following sodium pentobarbitone inducedanaesthesia, animals were perfused with coldheparinised phosphate buffered saline, followedby cold 4% paraformaldehyde in sodium caco-dylate buffer. Eyes were rapidly enucleated andfurther fixed for 6-8 hours, then transferred to70% ethanol for 12-18 hours. Eyes were pre-pared either as choroidal flat mounts or embed-ded in paraffin wax. Mesentery spread mountswere prepared from portions of ileum and pro-cessed in a similar manner.

HISTOCHEMICAL STAINING OF CHOROIDAL FLATMOUNTSRat mast cell phenotypes can be partially distin-guished by differential staining of granule pro-teoglycans following staining with a sequence ofAlcian blue followed by safranin. The granules ofmature CTMC have a high heparin content andstain red due to safranin displacing Alcian bluefrom heparin and binding in its place. The morehighly sulphated proteoglycans of MMC andimmature CTMC stain blue as Alcian blue is notdisplaced from these proteoglycans by safranin.

Flat mount preparations, consisting of thechoroid and attached sclera, were rehydrated indistilled water for 5 minutes and then processedas a single free floating unit. Alcian blue/safraninstaining was performed as previously described.23Briefly, free-floating preparations were acidifiedin 0 7M HCI (5 minutes) followed by immersionin 01% Alcian blue (Gurr, England) in 0-7 MHCI (60 minutes). Preparations were then brieflyrinsed in distilled water and immersed in 0-5%safranin (Hopkins and Williams, England) in0-125 M HC1 (5 minutes). Finally, preparationswere dehydrated through graded alcohols,cleared in toluene, and mounted. Mesenteryspread mounts were also stained using the proto-col applied to choroidal flat mounts.Owing to pigmentation in PVG/c and brown

Norway strains, bleaching of choroidal prepara-

tions was necessary to enable visualisation ofmast cells. Before Alcian blue/safranin stainingchoroid preparations were bleached with 5%chromic acid (3 x 25 minutes), rinsed in distilledwater (5 minutes) and further bleached in 5%oxalic acid (10 minutes). The preparations werethen rinsed in distilled water (5 minutes). Mastcell numbers in Alcian blue/safranin stainedpreparations were compared in bleached andunbleached choroidal flat mounts from six albinoanimals (Wistar Furth) to determine the effectsof the bleaching protocol. No significant differ-ence was found (p>0 53).

Toluidine blue staining was performed onchoroids from a further two Lewis rats. Prepara-tions were placed in 0.5 M HCI (5 minutes)before staining in 0-5% toluidine blue (Hopkinsand Williams, England) in 0 5 M HC1 (10minutes). Preparations were then rinsed (3 x dis-tilled water), dehydrated, and mounted.

HISTOCHEMICAL STAINING OF PARAFFIN SECTIONSA number ofeyes were embedded in paraffin waxusing routine methods. Sections (5 iim) were cutand mounted on slides. Following dewaxing andrehydration sections were stained with Alcianblue/safranin and toluidine blue following theprotocols used for choroidal flat mounts.

IMMUNOHISTOCHEMICAL CHARACTERISATION OFCHOROIDAL MAST CELLSAlthough histochemical staining differentiatesmature CTMC from MMC and immatureCTMC a more definitive method to distinguishmast cell phenotype involves analysis of thedistinct serine proteinases within cytoplasmicgranules. Rat mast cell protease I (RMCP I)which occurs inCTMC and rat mast cell proteaseII (RMCP II) present inMMC can be detected byimmunohistochemical staining.

Paraffin sections were stained with anti-RMCPI and anti-RMCP II according to a method

Superior

Temporal X Nasal

Inferior

Figure I Distribution ofmast cells in the Lewis rat choroidwith major arterial vessels shown. This figure demonstratesthe distribution ofmast cells within a single choroid, one dotrepresenting a single mast cell. This distribution isrepresentative ofall choroids examined. Note the superiorand inferior long posterior ciliary arteries and large branchsupplying a major portion ofthe nasal regions. Note also, theincreased density ofmast cells within the nasal regions.

212


http://bjo.bmj.com

/B

r J Ophthalm


ownloaded from

http://bjo.bmj.com/

Distribution and characterisation ofrat choroidal mast cells

^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. .:........

.....:.-

>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. .....,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.......

Figure 2 Photomicrograph ofaflat mount offemale Lewis choroid (toluidine blue). Notethe abundance ofmast cells associated with arterial vessels (A) and the absence ofmast cellsassociated with venous vessels (V); x45.

previously described.2' The extraocular musclesand surrounding connective tissue of the ratcontain large numbers of mast cells, whichserved as an internal positive control for RMCP Istaining. Sections of small intestine preparedfrom rats infected with Nippostrongylusbrasiliensis, which display a characteristic MMChyperplasia,"4 provided positive control sectionsfor RMCP II staining. One section of thismaterial was included on each slide prepared forRMCP II immunostaining. Negative controlswere prepared by substituting phosphate buf-fered saline in place of primary monoclonalantibodies. A number of sections were examinedand then counterstained with Alcian blue accord-ing to the first stage of the Alcian blue/safraninsequence. In addition, to determine the propor-tion of choroidal mast cells expressing RMCP I,choroidal flat mounts were prepared, immuno-histochemically stained for RMCP I and counter-stained with Alcian blue.

MAST CELL DENSITY AND DATA ANALYSISMast cell numbers were obtained by countingstained cells in entire choroidal flat mounts.Areas of the choroidal flat mounts were deter-mined using a digitiser pad in conjunction withan area calculating software package (Sigma-

60 rE 50 -

ETL~40a'

3030

Uo204-110

Scan; Jandel Scientific, USA). Statistical signifi-cance was tested using Student's t test.

Results

DISTRIBUTION, LOCATION, AND DENSITY OFCHOROIDAL MAST CELLSThe choroidal vascular supply was consistentwithin and between strains of rat. A single largebranch of the ophthalmic artery pierced thesclera nasal to the optic nerve head and bifur-cated, forming superior and inferior long pos-terior ciliary arteries (LPCA) (Fig 1). The LPCArun anteriorly to join the major arterial circle ofthe iris. Posteriorly, the LPCA were locatedwithin a depression on the internal aspect of thesclera, while anteriorly the LPCA were situatedwithin the dense connective tissue of the sclera.Normally one, and occasionally a second, majorbranch arose from either LPCA very close to theoptic nerve head and supplied a large portion ofboth the superior and inferior nasal quadrants ofthe choroid (Fig 1). Small arterial branches oftheLPCA supplied the remainder of the choroid.

In Lewis rats mast cells were irregularly dis-tributed throughout the choroid (Fig 1). Largenumbers ofmast cells were present in the connec-tive tissue surrounding the branch of the oph-thalmic artery, at the point where it penetratesthe sclera, and the two LPCA for a short distanceafter bifurcation. Consistently larger numbers ofmast cells were observed along the inferiorLPCA. In addition, the inferior LPCA had aregion of increased mast cell density near itsiridial termination (approximately 50 mast cellsalong the final 1 mm). Mast cell density wasconsistently greater in the nasal portion asso-ciated with the major arterial branch(s) supply-ing this region. In addition, whole mounts ofLewis anterior uveal tract were examined; theserevealed low numbers ofmast cells present in theciliary body (approximately 30 per eye) and iris(0-1 per eye).On high power examination of stained choroi-

dal flat mounts, mast cells were observed pre-dominantly in a perimural location associatedwith arteries and arterioles greater than 30 ,um indiameter (Fig 2). In particular they werearranged along the lateral and not the sclerad orvitread margins of vessels. This location wasconfirmed in histological sections.The density ofmast cells within the choroid of

Lewis rats varied in an anteroposterior manner(Fig 3), density being greatest (42/mm2) in theequatorial, or middle, regions of the choroid.From this region, density diminished rapidly to aminimum (12/mm') at the anterior margin of thechoroid. A distinctive focal region of high mastcell density (40/mm2) occurred around the opticnerve head.

In paraffin sections, mast cells were observedin association with the large and small arteries ofthe choroid and were largely absent from thechoriocapillaris (Fig 4). This contrasts with thelocation in mesentery spread mounts where mastcells were distributed evenly throughout thetissue with small numbers associated with veinsand venules larger than 25 ,um (Fig Sa, b).

Sexual dimorphism was evident in the Lewis

u 12 24 36 48Distance from optic nerve head (mm)

Figure 3 Posteroanterior gradient ofmast cell density in thechoroid of the Lewis rat. Columns represent mean withSEM.

213


http://bjo.bmj.com

/B

r J Ophthalm


ownloaded from

http://bjo.bmj.com/

Steptoe, McMenamin, McMenaminX~~~~~~~~~~~~~~~~. .;.;..;.

iN:f:0:;:;=fR5a .¾

S

p...''...;

*rS X,4t

*s

.Y .. .~~

7U.

a. 4t.-

-5b *& 41'' "I'

I. ...... V*

*

..9. . .:: .: ' .:-:.. :_:': K .:;..'

ti : .$;: ...... .: 's *i i:

t:

W.:.................. ....... _ P .... . . .. _ . ... .. .. aa.mo9..

.a

:: :.:.: .' . .. . ...a . . .. .: . . ..... . .... . ; ... . ....... ..... .. .. .... ................. . .. ;.... ; ; ;i .. .. .... .......* :: ; : .: .. ::.. ....... ... . ... .. i, :;: . :..... ...,-.; ........ .. . ................ : ...??,'sAm ? ...... ta A .... . . ; . .S > 4 t >f ; w:-ir ;;o. ;+-g5Z

4*.

*6 vo,i

a .K,, I ~~~~~~~~~~~~~~4.. ,.

10i ..

4Wc-

el

*.S

'p

214

A.'r,.gt,

__om-

.'0


http://bjo.bmj.com

/B

r J Ophthalm


ownloaded from

http://bjo.bmj.com/


Figure 4 Safranin positive mast cell adjacent to a choroidalarteriole in the Lewis rat; R=retina, C=choroid (paraffinembedded, Alcian bluelsafranin); xI 100.Figure Sa Safranin positive connective tissue mast cells in amesentery spread mount (Lewis rat). Note the pattern ofdistribution ofmast cells both close to and distantfrom a largevenule (Alcian bluelsafranin); x 100.Figure 5b RMCP I positive mast cells in mesentery spreadmount; x250.Figure 6 Choroidal flat mount (Lewis rat) demonstratingperiarteriolar mast cells containing a mixture ofblue and redstaining granules (Alcian bluelsafranin); x350.Figure 7 Mast cells in aflat mount preparation (Lewis rat).Note the predominantly periarteriolar location (toluidineblue); x250.Figure 8a Blue staining mucosal mast cells in the smallintestine ofNippostrongylus brasiliensis infected ratfollowingAlcian bluelsafranin sequence; x 100.Figure 8b Mucosal mast cells in the small intestine ofNippostrongylus brasiliensis infected rat demonstratingpositive RMCP II immunostaining; x 100.Figure 9a RMCP I positive mast cells adjacent toarterioles in the choroid (Lewis rat, paraffin embedded);x 1100.Figure 9b Positive control section (paraffin embeddedmesentery) immunostainedfor RMCP I. Note the intensityand uniformity ofimmunopositive granules throughout thecytoplasm (Lewis rat); x400.Figure 10 Choroidflat mount (Lewis rat) sequentiallystained with anti-RMCP I (brown) and Alcian blue (bluelgreen). Note the variable quantity ofRMCP I positivegranules throughout the cytoplasm; x250. Inset; highermagnification; x 600.Figure 11 Choroidflat mount (brown Norway rat)demonstrating blue staining mast cells (arrowheads); x200.

rats where both total choroidal mast cell numberand density (Table 1) were both significantlygreater in males (p<001 in both cases). Noobservable difference existed in the distributionand microanatomical arrangement of mast cellsthroughout the choroid between sexes.

MAST CELL MORPHOLOGY, HISTOCHEMISTRY, ANDPHENOTYPEChoroidal mast cell profiles were clearly evidentin flat mount preparations and were most com-monly ovoid (7x20 pm), although circular (12-15 ptm diameter) and elongated (5 x 30 p,m)profiles were occasionally noted (Figs 6 and 7). Ingeneral, the shape of choroidal mast cellsappeared related to two factors, the Alcian blue/safranin staining characteristics and the prox-imity to arterial vessels. Periarterially locatedmast cells containing predominantly blue stain-ing granules were most frequently thin andelongated. However, mast cell profiles becameincreasingly rounded as the distance from vesselsbecame greater or the degree of safranin stainingincreased (Figs 6 and 7). A minor subpopulationof small, round (7-10 pm diameter), blue stain-ing mast cells was observed distant to arterialvessels, most commonly in the anterior portionsof the choroid.The majority of mast cells (89%) contained a

variable mixture of both red and blue stainedgranules ('mixed staining') (Fig 6), blue stainingcells constituted approximately 6% of choroidalmast cells, while red staining cells made upapproximately 5% of the total. Choroidal mastcell granules in paraffin sections were pre-dominantly safranin positive (Fig 4), only rarely

were blue staining mast cells observed, asexpected from their rarity in flat mounts.Mast cells in toluidine blue stained material

exhibited intense staining and distinctive purplemetachromasia (Fig 7). Details of intracellulargranules and nuclear profiles were oftenobscured due to the intensity of staining (Fig 7),although occasionally, a large pale centralnucleus was observed. The margins of mast cellswithin toluidine blue stained material weresmooth and clearly delineated. Evidence of mastcell degranulation was minimal, only 3-6% (SEM0 25%) of mast cells displayed a degranulatedappearance in toluidine blue stained paraffinsections.During the course of this study the question

arose whether both staining protocols werevisualising all mast cells present in rat choroid.Therefore, a comparison of the ability of theAlcian blue/safranin and toluidine blue stainingprotocols to determine choroidal mast cellnumber was performed. No significant differ-ence was observed between the total number ofchoroidal mast cells visualised by toluidine blueand Alcian blue/safranin in consecutive serialsections of Lewis choroid.

Stining with anti-RMCP II antibodies failedto reveal any immunopositive cells within thechoroid or extraocular tissues with the exceptionof some limbal conjunctival mast cells. Controltissue (N brasiliensis infected gut) displayedpositive staining ofMMC (Fig 8a, b).Immunohistochemical studies of histological

sections and flat mounts both revealed choroidalmast cells to be RMCP I positive and generallyassociated with large and small arterial vessels(Fig 9a). In addition, RMCP I positive mast cellswere observed surrounding limbal vessels and inextraocular tissue. The immunostaining inten-sity ofchoroidal mast cells was variable and oftenless intense than that displayed by mast cells inpositive control tissue such as lymph node ormesentery (Fig 9b). Choroidal flat mountsimmunostained for RMCP I and counterstainedwith Alcian blue provided a novel method ofintegrating histochemical and immunohisto-chemical procedures. In these preparations allmast cells contained at least a small number ofRMCP I positive granules; however, a widerange in the content ofRMCP I positive granules(brown reaction product) was present (Fig 10).In addition, peripherally located RMCP I nega-tive granules which were Alcian blue positive(Fig 10) corresponded to the peripheral bluegranules seen following Alcian blue/safraninstaining (Fig 6). Mast cells within the choroidcontrasted with those in mesentery spreadmounts which had a uniformly high content ofRMCP I positive granules (Fig Sb) which asindicated above contained almost exclusivelysafranin positive granules (Fig Sa).

STRAIN VARIATIONS IN CHOROIDAL MAST CELLNUMBERPrevious studies have suggested strain specificvariations in choroidal mast cell number may beimplicated in the susceptibility to EAU, there-fore we chose to re-examine this issue with regardto mast cell heterogeneity in low, medium, and

215


http://bjo.bmj.com

/B

r J Ophthalm


ownloaded from

http://bjo.bmj.com/


Table I Mean mast cell number and densities in the choroidand mesentery ofthe rat strains examined

Mean mast cell density,Mean choroidal ceUs/mn2 (SEM)mast cell nwnber

Strain (SEM) Choroidt Mesenteryt

Lewis(male)* 1813(116) 37 0 (3-3) NDLewis(female)* 1180 (51) 23-6(1-2) 102 (7)Wistar Furth 610 (52) 13-5 (0-7) NDBrown Norway 106 (9) 3-06 (0-3) 110 (7)PVG 69 (8) 1-95 (0 2) 97 (8)

*Significant difference between male and female (p<0c01).tSignificant differences exist between all strains (p<0001,except PVG/c and BN, p<0 01).:No significant difference between any strains; ND=notdetermined.

high IgE responsive rat stains: Lewis (low),PVG/c and Wistar Furth (moderate), and brownNorway (high). The choroid of Lewis rats con-tained the greatest number of mast cells; how-ever, in comparison the choroid of both PVG/cand brown Norway rats contained low numbersofmast cells (Table 1). The number of choroidalmast cells in Wistar Furth rats was approx-imately half that of Lewis rats. These strainvariations in mast cell density are apparentlytissue specific as no significant difference wasobserved between the density ofmast cells withinthe mesentery spread mount preparations fromLewis, PVG/c, or brown Norway rats (Table 1).Choroidal mast cells in all strains demon-

strated the same periarterial location describedpreviously for Lewis rats. Mast cells within thechoroid of Wistar Furth rats displayed the samegeneral distribution seen in Lewis rats; however,in those strains of rat with a very low number ofchoroidal mast cells (PVG/c, brown Norway) thedistribution was most commonly restricted to thelarge arterial vessels in the region of the opticnerve head. In brown Norway rats a smallnumber of mast cells'20 were observed in themid-choroidal region of the nasal quadrants,associated with the large branch of the longposterior ciliary ateries supplying this area. InPVG/c rats choroidal mast cells were restrictedalmost exclusively to the proximal portions ofthelarge arterial vessels and the long posterior ciliaryarteries.Mast cells in bleached choroidal flat mounts

failed to display safranin positive granules (Fig11); however, in paraffin sections of PVG/c andbrown Norway rat eyes choroidal mast cellsdisplayed characteristic safranin positive asdescribed in sections of Lewis choroid (notshown).

DiscussionThe present study was prompted by the increas-ing awareness of the diverse role of mast cellsbeyond classic IgE mediated hypersensitivity,together with the need for a more precise know-ledge of their distribution, location, and pheno-type in the choroid. The present study expandsprevious reports by providing, not only a moredetailed account of the strain specific variationsin choroidal mast cell density, distribution, andmicroanatomical location in rats but, moreover,presents novel data with regard to the histo-chemical and immunohistochemical phenotypeof these cells.

The physiological significance of the associa-tion of mast cells with arterial vessels and theregional variation in distribution is unclear at thisstage. It is possible, however, that local regula-tion ofblood flow within the choroid, in responseto the requirements of the overlying retina, maybe among the normal physiological roles of thesecells. Interestingly, CTMC have been identifiedalong the margins ofthe middle meningeal arteryin the rat dura mater,25 a tissue analogous to thesclera. It has been suggested that mast cells inthis region may have a role in the regulation ofblood flow and the pathogenesis of headache,26 27although this is a matter of controversy. Localcontrol of blood flow and secretory activity bymast cells has been demonstrated in the ratadrenal gland28 and both rat29 and hamsterovary.'

Degranulation of periarteriolar mast cells inthe hamster cheek pouch results in diapedesis ofleucocytes exclusively from 'downstream' post-capillary venules. Leucocytes then migratetoward adjacent arterioles along chemotacticgradients generated by the mast cell degranula-tion.3' The results of the present study areconsistent with speculation that the activation ofchoroidal mast cells reported at the onset ofEAU2" may result in mediators being carrieddownstream from the long posterior ciliaryarteries to the extensive capillary beds of theciliary body,32 where extravasation of inflamma-tory cells may be effected. It is possible amechanism such as this may mediate the largescale inflammatory cell infiltration of the iris,ciliary body, and anterior chamber observed inEAU.33 As both the iris and ciliary body of Lewisrats contain few mast cells, local release of mastcell mediators seems an unlikely explanation ofsuch a response. It is interesting to note thatrabbits, which have little connection between thechoroidal circulation and that of the anteriorsegment, develop only minimal inflammatoryinfiltration of the anterior chamber duringEAU.3' Activation of mast cells has been shownto result in the production of a range of cyto-kines,35 '3 which if released from choroidal mastcells, may have a 'downstream' effect on theextensive network of macrophages and majorhistocompatibility class II antigen bearing den-dritic cells in the anterior uveal tract.3738

It is interesting to note the presence of mastcells in the vascular and connective tissue sur-rounding both the eye and brain. Mast cells havealso been identified within selected regions ofbrain parenchyma,39 whereas mast cells areabsent from the neural retina. Furthermore, theretina is isolated from the mast cell rich choroidby the retinal pigment epithelium. These factorsmay indicate the need to protect the retinalvasculature and delicate photoreceptors from theaction of mast cell mediators.

Sexual dimorphism in choroidal mast cellnumber has not previously been described. How-ever, sexual dimorphism ofmast cell content hasbeen reported for a number of tissues, includingrat peritoneal cavity,4' hamster and mouseharderian gland,4243 and rat brain.39 Thesetissues, however, differ from the choroid as mastcell content is invariably higher in females than inmales. The reduced number ofmast cells in these

216


http://bjo.bmj.com

/B

r J Ophthalm


ownloaded from

http://bjo.bmj.com/


tissues has been attributed to an androgen depen-dent mechanism.42"' The functional significanceof higher choroidal mast cell numbers in males isuncertain but may affect the susceptibility to, orthe severity of, intraocular inflammation.The presence of safranin positive granules

within choroidal mast cells is strongly suggestiveof CTMC phenotype. However, the presence ofRMCP I positive cytoplasmic granules andabsence of RMCP II immunostaining is moredefinitive evidence of the CTMC phenotype ofchoroidal mast cells. The coincidence of intensestaining for RMCP I and the exclusively safraninpositive nature of mesenteric mast cells suggestsRMCP I may be restricted to safranin positivecytoplasmic granules. It is possible the Alcianblue positive/safranin negative and Alcian bluepositive/RMCP I negative granules representimmature granules which have not yet attainedheparin and RMCP I contents.45" The variableand thus, perhaps, reduced content of RMCP Iwithin choroidal mast cells could serve to limitbystander damage, resulting from the proteolyticactivity of this mediator,4748 to the delicate cells inthe immediate microenvironment such as theretinal pigment epithelium, an important com-ponent of the blood-ocular barrier. The presenceof heparin and the susceptibility to compound48/80 reported for rat choroidal mast cells byGodfrey9 are both features suggestive of CTMCphenotype and concur with the findings of thepresent study.

In Lewis rats a minor subpopulation of un-usually small mast cells which contain exclu-sively Alcian blue positive granules was locateddistant to arterial vessels. The features of thesemast cells suggest they were in an early stage ofmaturation from mast cell precursors.45"Although the lifespan of rat CTMC is greaterthan 6 months,49 the presence of these small bluestaining cells may indicate that mast cell precur-sors are being continually recruited, possiblyfrom venous vessels,50"5 in the anterior choroid,from where they may subsequently migrate totheir periarteriolar location and assume a morecharacteristic mature form.The rat strains chosen in the present study

demonstrate a range of IgE responsiveness andsusceptibility to EAU. For example, Lewis ratsare highly susceptible, PVG and Wistar Furthmoderately susceptible, and brown Norwayexhibit low susceptibility to EAU,2"53 while IgEproduction in response to antigenic stimulationshows an inverse trend.""57 In the present studythese strains were examined to determine thepossibility of an association between IgE respon-siveness, choroidal mast cell number, and EAUsusceptibility. The mean total number of mastcells within the choroid of Lewis rats determinedin the present study (1180 (SD 51)) is similar tothat reported by Godfrey.9 A discrepancy exists,however, between the choroidal mast cell densitydetermined in the present study (23-6 (1 -2)/mm2)for Lewis rats and the value reported (50-2 (5-4)/mm2) by Mochizuki and associates,22 a valuewhich is close to the mast cell density in theposterior choroid, possibly indicating incom-plete sampling of the choroid. A possibleexplanation for the previously reported densityin brown Norway (0-2 (0 1)/mm2)22 being lower

than our value (3-06 (0 3)/mm2) may be thebleaching protocol employed by Mochizuki andco-workers22 which, in our hands, caused exten-sive disruption of choroidal mast cell morphol-ogy. Moreover, the fixation protocol employed(10% formalin) is known to reduce the number ofstained mast cells.58 Choroidal mast cell numbersor density in PVG/c and Wistar Furth have notbeen reported before the present study.The results of the present study are in agree-

ment with the results of Mochizuki and col-leagues22 which. suggest that strain specificvariation in the choroidal mast cell density istissue specific. However, the results of this studyare not consistent with the suggestion that sus-ceptibility to EAU is directly correlated withchoroidal mast cell number.22 The moderateresponsiveness of PVG/c rats does not reflect thechoroidal mast cell number of this strain whichwas the lowest of the strains examined in thepresent study. Indeed, the susceptibility to EAUshows an inverse relation to IgE responsiveness.Studies performed in this laboratory to investi-gate the role of choroidal mast cell dynamicsduring EAU in rat strains of differing suscep-tibility indicate their importance in the patho-genesis of EAU may be stain specific.59The authors wish to thank Professor J V Forrester for hisinvaluable input during the initiation and formulation of thisstudy.

1 Stevens RL, Austen KF. Recent advances in the cellular andmolecular biology of mast cells. Immunol Today 1989; 10:381-6.

2 Marshall JS, Bienenstock J. Mast cells. Springer SeminImmunopathol 1990; 12: 191-202.

3 Galli SJ. New approaches for the analysis of mast cellmaturation, heterogeneity, and function. Fed Proc 1987; 46:1906-14.

4 Levene RZ. Mast cells and amines in normal ocular tissues.Invest Ophthalmol 1962; 1: 531-43.

5 Smelser GK, Silver S. The distribution of mast cells in thenormaleye: amethodofstudy. ExpEyeRes 1963; 2: 134-40.

6 AllansmithMR, Baird RS, Kashima KA, Bloch KJ. Mast cellsin ocular tissues of normal rats and rats infected withNippostrongylus brasiliensis. Invest Ophthalmol Vis Sci 1979;18: 863-7.

7 Louden C, Render JA, Carlton WW. Mast cell number innormal and glaucomatous canine eyes. AmJ Vet Res 1990;51: 818-9.

8 Rodrigues MM, Hackett J, Donohoo P. Iris. In: Jakobiec FA,ed. Ocular anatomy, embryology, and teratology. Philadelphia:Harper and Row, 1982: 285-302.

9 Godfrey WA. Characterisation of the choroidal mast cell.TransAm Ophthalmol Soc 1987; 85: 557-99.

10 Feeney L, Hogan MJ. Electron microscopy of the humanchoroid I. Cells and supporting structures. AmJ Ophthalmol1961; 51: 1057-72.

11 Metcalfe DD, Kaliner M, Donlon MA. The mast cell.CRC CritRev Immunol 1981; 2:23-74.

12 Ennerback L, Miller HRP, Mayrhofer G. Methods for identi-fication and characterisation of mast cells by light micro-scopy. In: Befus AD, Bienenstock J, Denburg JA. eds.Mast cell differentiation and heterogeneity. New York: RavenPress, 1986: 405-18.

13 Stevens RL, Katz HR, Seldin DC, Austen KF. Biochemicalcharacteristics distinguish subclasses of mammalian mastcells. In: Befus AD, Bienenstock J, Denburg JA, eds. Mastcell differentiation and heterogeneity. New York: Raven Press,1986: 183-203.

14 Galli SJ. Biology of disease: new insights into the "riddle of themast cells": microenvironmental regulation of mast celldevelopment and phenotypic heterogeneity. Lab Invest1990; 62: 5-33.

15 Ennerback L. Mast cells in rat gastrointestinal mucosa: effectsof fixation. Acta Path Microbiol Scand 1966; 66: 289-302.

16 Ennerback L. Mast cells in rat gastrointestinal mucosa: dyebinding and metachromatic properties. Acta Path Microbiol

_ _ Scand 1966; 66: 303-12.17 Barrett KE, Metcalfe DD. Heterogeneity of mast cells in the

tissues of the respiratory tract and other organ systems.Am Rev Respir Dis 1987; 135: 1190-5.

18 Wilkes LK, McMenamin C, Holt PG. Postnatal maturation ofmast cell subpopulations in the rat respiratory tract.Immunology 1992; 75: 535-41.

19 Gibson S, Miller HRP. Mast cell subsets in the rat distin-guished immunohistochemically by their content of serineproteases. Immunology 1986; 58: 101-4.

20 Irani AA, Schwartz LB. Mast cell heterogeneity. Clin ExpAllergy 1989; 19: 143-55.

217


http://bjo.bmj.com

/B

r J Ophthalm


ownloaded from

http://bjo.bmj.com/


21 de Kozak Y, Saint-Laudy J, Benveniste J, Faure J-P. Evidencefor immediate hypersensitivity phenomena in experimentalautoimmune uveoretinitis. EurJ Imnmol 1981; 11: 612-7.

22 Mochizuki M, Kuwabara T, Chan C-C, Nussenblatt RB,Metcalfe DM, Gery IG. An association between sus-ceptibility to experimental autoimmune uveitis and choroi-dal mast cell numbers.JImmunol 1984; 133: 1699-701.

23 Strobel S, Miller HRP, Ferguson A. Human intestinal mastcells: evaluation of fixation and staining.J Clin Pathol 1981;34:851-8.

24 Miller HRP, King SJ, Gibson S, Huntley JF, Newlands GFJ,Woodbury RG. Intestinal mucosal mast cells in normal andparasitised rats. In: Befus AD, Bienenstock J, Denburg JA,eds. Mast cell differentiaion and heterogeneity. New York:

2 Raven Press, 1986: 239-56.25 Dimlich RVW, Keller JT, Strauss TA, Fritts MJ. Linear

arrays ofhomogenous mast cells in the dura mater of the rat.J Neurocytol 1991; 20: 485-503.

26 Theoharides TC. Mast cells and migraines. Perspectives inBiology and Medicine 1983; 26: 672-5.

27 Dimitradou V, Lambracht-Hall M, Theoharides TC. Histo-chemical and ultrastructural characteristics of rat brainperivascular mast cells stimulated with compound 48/80 andcarbachol. Neuroscience 1990; 39: 209-24.

28 Hinson JP, Vinson GP, Pudney J, Whitehouse BJ. Adrenalmast cells modulate vascular and secretory responses in theintact adrenal gland of the rat. J Endocrinol 1989; 121:253-60.

29 Jones RT, Duvall D, Guillette LJ. Rat ovarian mast cells:distribution and cyclical changes. Anat Rec 1980; 197:489-93.

30 Krishna A, Terranova PF. Compartmentalised mast celldegranulations in the ovarian hilum, fat pad, bursa andblood vessel regions of the cyclic hamster: relationships toovarian blood flow. ActaAnat 1991; 141: 18-25.

31 Raud J, Lindbom L, Dahlen S-E, Hedqvist P. Periarteriolarlocalisation of mast cells promotes oriented interstitialmigration of leukocytes in the hamster cheek pouch.AmJPathol 1989; 134: 161-9.

32 Morrison JC, deFrank MP, Van Buskirk EM. Comparativemicrovascular anatomy of mammalian ciliary processes.Invest Ophthalmol Vis Sci 1987; 28: 1325-40.

33 Gery I, Mochizuki M, Nussenblatt RB. Retinal specificantigens and the immunopathogenic processes they pro-voke. ProgRetRes 1986; 5: 75-109.

34 Wacker WB, Rao NA, Marak GE. Immunopathologic res-ponses of rabbits to retinal 's' antigen. I. Phlogistic charac-teristics associated with antigen source and sensitizing dose.Ophthalmic Res 1981; 13: 302-11.

35 Burd PR, Rogers HW, Gordon JR, Martin CA, Jayaraman S,Wilson SD, etal. Interleukin 3-dependent and -independentmast cells stimulated with IgE and antigen express multiplecytokines. JExpMed 1989; 170: 245-57.

36 Wodnar-Filipowicz A, Heusser CH, Moroni C. Production ofthe haemopoetic growth factors GM-CSF and interleukin-3by mast cells in response to IgE receptor activation. Nature1989; 339: 150-2.

37 McMenamin PG, Holthouse I. Immunohistochemical charac-terisation of dendritic cells and macrophages in the aqueousoutflow pathways of the rat eye. Exp Eye Res 1992; 55:315-24.

38 McMenamin PG, Holthouse I, Holt PG. Class-II majorhistocompatibility complex (ia) antigen-bearing dendriticcells within the iris and ciliary body of the rat eye -distribution, phenotype and relation to retinal microglia.Immunology 1992; 77: 385-93.

39 Goldschmidt RC, Hough LB, Glick SD, Padawer J. Mast cellsin rat thalamus: nuclear localisation, sex difference and left-right asymmetry. Brain Res 1984; 323: 209-17.

40 Davis RH, McGowan L. Comparative peritoneal content asrelated to species and sex. Anat Rec 1968; 162: 357-62.

41 Jaques R, Ruegg M. Age and sex difference in mast cell countand histamine content. Agents Actions 1970; 1: 144-7.

42 Payne AP, McGadey J, Johnston HS, Moore MR, ThompsonGG. Mast cells in the hamster harderian gland: sex differ-ences, hormonal control and relationship to porphyrin.JAnat 1982; 135: 451-61.

43 Shirama K, Kohda M, Hokano M. Effects of endocrine glandsand hormone replacement on the mast cell count of theharderian gland of mice. Acta Anat 1988; 131: 327-31.

44 Gaytan F, Carrera G, Pinilla L, Aguilar R, Bellido C. Mast cellsin the testis, epididymis and accessory glands of the rat.Effects of neonatal steroid treatment. J Androl 1989; 10:351-8.

45 Combs JW, Lagunoff D, Benditt FP. Differentiation andproliferation of embryonic mast cells of the rat. J Cell Biol1%5; 25: 577-92.

46 Katz R, Austen KF, Caterson B, Stevens RL. Secretorygranules of heparin-containing rat serosal mast cells alsopossess highly sulfated chondroitin sulfate proteoglycans.BiolChem 1986; 261: 13393-6.

47 Befus D, Fujimaki H, Lee TDG, Swieter M. Mast cellpolymorphisms. Present concepts, future directions. Dig DisSci 1988; 33: (Suppl) 16-24.

48 Dietsch GN, Hinrichs DJ. Mast cell proteases liberate stableencephalitogenic fragments from intact myelin. CellImmunol 1991; 135: 541-8.

49 Schwartz LB, Austen KF. Structure and function of thechemical mediators of mast cells. Prog Allergy 1984; 34:271-321.

50 Marchesi VT, Gowans JL. The migration of lymphocytesthrough the endothelium of venules in lymph nodes: anelectron microscopic study. Proc Roy Soc Lond 1964; 159:283-90.

51 Cho Y, De Bruyn PH. The endothelial structure of thepostcapillary venules of the lymph node and the passage oflymphocytes across the venule wall. J Ultrastruct Res 1979;69: 13-21.

52 de Kozak Y, Thillaye B, Renard G, Faure JP. Hyperacuteform of experimental autoimmune uveoretinitis in Lewisrats; electron nicroscopic study. Graefes Arch Klin ExpOphthalmol 1978; 208: 135-42.

53 de Kozak Y, Sakai J, Thillaye B, Faure JP. S antigen-inducedexperimental autoimmune uveo-retinitis in rats. Curr EyeRes 1981; 1: 327-37.

54 Karlsson T, Ellerson JR, Dahlbom I, Bennich H. Analysis ofthe serum IgE levels in nonimmunised rats of various strainsby a radioimmunoassay. ScandJ Immunol 1979; 9: 217-28.

55 Karlsson T, Ellerson JR, Haig DM, Jarrett EEE, Bennich H.A radioimmunoassay for evaluation of the IgE and IgGantibody responses in the rat. Scand J Immunol 1979; 9:229-38.

56 Abadie A, Provost-Danon A. Specific and total IgE responsesto antigenic stimuli in Brown-Norway, Lewis and Sprague-Dawley rats. Immunology J980; 39: 561-9.

57 Bergstrand H, Bjornsson A,Trick I-M, Lundquist B, PauwelsR, Bazin H. Anti-IgE and Con A-induced histamine releasefrom mast cells of four rat strains: Correlation with totalserum IgE. AgentsActions 1982; 12: 612-8.

58 Gleich GJ, Ackerman SJ, Hisamatsu K, Leiferman KM. Mastcell heterogeneity: biological and clinical significance. In:Befus AD, Bienenstock J, Denburg JA, eds. Mast celldifferentiation and heterogeneity. New York: Raven Press,1986: 347-56.

59 Steptoe RJ, McMenamin C, McMenamin PG. Choroidal mastcell dynamics during experimental autoimmune uveo-retinitis in rat strains of differing susceptibility. OcularImmunol Inflamm 1994 (in press).

218


http://bjo.bmj.com

/B

r J Ophthalm


ownloaded from

http://bjo.bmj.com/

distribution and characterisation rat choroidal mast cells · british_journalofophthalmology...

Documents