J Clin Pathol 1991;44:194-199

Papers

Histopathology and immunohistochemistry of thecaecum in children with the Trichuris dysenterysyndrome

T T MacDonald, M-Y Choy, J Spencer, P I Richman, T Diss, B Hanchard,S Venugopal, D A P Bundy, E S Cooper

Department ofPaediatricGastroenterology,St Bartholomew'sHospital, LondonECIA 7BET T MacDonaldM-Y ChoyDepartment ofHistopathology,St Bartholomew'sHospital, LondonP I RichmanDepartment ofHistopathology,University CollegeLondon and MiddlesexSchool of Medicine,LondonJ SpencerT DissDepartment ofPathology, Universityof the West Indies,Kingston, JamaicaB HanchardDepartment ofSurgery, University ofthe West Indies,Kingston, JamaicaS VenugopalDepartment ofPureand Applied Biology,Imperial College ofScience, Technologyand Medicine, LondonD A P BundyDepartment ofPureand Applied Biology,Imperial College ofScience, Technologyand Medicine, Londonand TropicalMetabolism ResearchUnit, University of theWest Indies, Kingston,JamaicaE S CooperCorrespondence to:Dr T T MacDonaldAccepted for publication4 October 1990

AbstractCaecal biopsy specimens from Jamaicanchildren with the Trichuris dysenterysyndrome (TDS) and age matchedJamaican controls were investigated byimmunohistochemistry and by lightmicroscopy. Biopsy specimens from allchildren (with TDS and controls) showeda mild to moderate increase in inflam-matory cells. Except in the vicinityof the worm, where the epithelium wasflattened, there was no other epithelialabnormality. Compared with controls,children with TDS had increased IgMlamina propria plasma cells anddecreased intraepithelial T cells. Therewas also an increase in crypt epithelialcell proliferation. Lamina propria Tcells (both activated and non-activated)were no more common in children withthe Trichuris syndrome than controls.Epithelial cell HLA-DR and VLA-1expression (which are increased in othercolitides) were the same in both groups.Despite the presence of large wormburdens and chronic dysentery, there-fore, only minor changes were seen inthe caecal mucosa of children with TDS.

Trichuris trichiura is a nematode parasite ofthe human colon. It occurs throughout theworld but is more common in the Tropics. Inlight infections worms are found only in thecolon, but in heavier infections all of the colonand rectum can be colonised. Intense infec-tion (greater than 200 worms) is morecommon in children between the ages of 2 and10 years and is associated with a colitis termedthe "Trichuris dysentery syndrome" (TDS).'-3Affected children pass frequent, bloody,heavily mucoid stools, may present with rectalprolapse, and are often severely growthretarded.4 Antihelminthic treatment elimi-nates the worms, restores normal colonicfunction, and leads to a rapid growth spurt.4

Chronic diarrhoea and growth suppressionare also associated with the idiopathic colitidesand enteropathies such as Crohn's disease andulcerative colitis in children in the developedworld.56 In these diseases the mucosalinflammatory responses are well characterised.To our knowledge, however, there is littleinformation on the histology of the colon in

chronic trichuriasis. This is of interestbecause, unlike the idiopathic colitides, theagent responsible for the Trichuris dysenterysyndrome is known. We describe here thehistopathology and immunohistology of thecaecum in the Trichuris dysentery syndrome.

MethodsColonic (caecal) biopsy specimens wereobtained at colonoscopy from children with aclinical history suggestive of TDS.2 The studywas approved by the ethical committee of theUniversity of the West Indies and the biopsyspecimens were taken at the TropicalMetabolism Research Unit, University of theWest Indies, Jamaica.The study was carried out in two parts.

Firstly, caecal plasma cell numbers weremeasured on a group offive children with TDS(cases 5-9) (table 1) and four age matchedcontrols (cases 1-4 (table 1)). This part of thestudy was carried out using formalin fixed,paraffin wax embedded caecal biopsyspecimens. Sections 4 gum thick were cut,dewaxed, and IgA, IgM, IgG plasma cells

Table I Patient details

Case No Age Sex

A Controls Diagnosis1 5 9 M Normal2 2-5 F Non-specific colitis3 4-2 F Mild proctitis4 17 M Normal.

Patients Worm burden5 4 5 M 9326 3-8 M 9887 2-6 F 3918 6-6 M 35439 5-7 F 2700B Controls Diagnosis

1 2-2 F Non-Trichuris dysentery2 4 0 F Proctitis3 4-9 M Non-Trichuris dysentery4 5 0 M Polyps5 2-7 F Rectal prolapse6 7-1 M Small bowel disease

Patients Worm burden7 6-3 F 8938 3-2 F 15359 2-5 M **10 50 M **11 2-9 F 127512 5 0 F 190413 3-0 F 71514 4-3 M 97015 3-2 F 720w twins16 3-2 F 942}17 4-6 F 1431

**Stools were not collected for these children afteranthelminthic treatment, but large numbers of worms werevisible in the caecum at colonoscopy.

194

on Decem

ber 6, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.44.3.194 on 1 M

arch 1991. Dow

nloaded from

The caecum in children with the Trichuris dysentery syndrome

Table 2 Details of monoclonal antibodies used

Antibody Specificity Source

UCHTI CD3 ICRF-HTIG,1B5 HLA-DR fUniversity College Hospital, LondonLeu 3a CD4 QBecton Dickinson Ltd,Leu 2a CD8 Mountain View, CaliforniaTCRC1 yb TcR T Cell Sciences,Anti-VLA-l VLA-1 os chain fCambridge, MAKi67 Dividing cells Dako Ltd,Anti-Tac CD25 fHigh Wycombe, Bucks

identified by immunoperoxidase immuno-histochemistry as described elsewhere.7 Theantisera used were isotype specific, affinitypurified, rabbit antisera purchased from DakoLtd, High Wycombe, Buckinghamshire. Aperoxidase conjugated goat anti-rabbit IgG(Dako Ltd) was used as a second layer.Frozen sections were used in the second part

of the study. This permitted the use of a greaternumber of phenotypic markers. Biopsyspecimens obtained at colonoscopy were snapfrozen in liquid nitrogen and stored at - 700C.Tissue was studied from 11 children withtrichuriasis (cases 7-17) (table 1) and six con-trols (cases 1-6) (table 1). Cryostat sections6 gm thick were cut and stained using thealkaline phosphatase anti-alkaline phosphatase(APAAP) method, as described previously.7The APAAP technique was used becauseendogenous peroxidase in inflammatory cells inthe mucosa of both groups of patients -madeimmunoperoxidase stained sections impossibleto interpret. The monoclonal antibodies usedin this study are shown in table 2. Secondaryantibody used was a rabbit anti-mouse IgG(Dako Ltd) and the monoclonal APAAPantibody used was also purchased from DakoLtd. Fast red was used to visualise antibodybinding.Immunostained plasma cells, CD3 +, or

CD25 + cells in the lamina propria werecounted as the percentage of stained cells of thetotal number of nucleated cells in the laminapropria. At least 500 cells were counted in eachbiopsy specimen. Differences between groupmeans were determined by analysis ofvariance.The relative numbers of cells expressing

CD3, CD4, CD8, or yb TcR in the caecalepithelium were quantified by counting thenumber of positively stained cells in a givenlength of epithelium (both surface and crypt).Length of epithelium was measured with acalibrated micrometer eyepiece. At least 200positively stained lymphocytes were counted ineach biopsy specimen. Differences betweengroup means were determined by analysis ofvariance.HLA-DR and VLA-1 staining of epithelial

cells was graded semi-quantitatively on a fourpoint scale (±, +, + +, or + + +). The rangeofstaining was from very weak cytoplasmic andbasolateral membrane staining (±) to intensecytoplasmic and membrane staining (+ + + ).For statistical analysis the grades were conver-ted to numbers (± = 0-5, + = 1, + + = 2,and + + + = 3), and the groups were com-pared using the Kolmogorov-Smirnov 2 grouptest.

Ki67 staining of dividing crypt epithelialcells was carried out as described previously.'Differences between group means weredetermined by analysis of variance. Whereapplicable results are shown as mean +1standard error.

ResultsHISTOPATHOLOGICAL FINDINGSHaematoxylin and eosin stained sections ofcaecal biopsy specimens from nine children(five with Trichuris and four controls (table 1))were coded and examined blind by one of us(PIR). All biopsy specimens showed a mild tomoderate increase in chronic inflammatory cellsin the lamina propria. In sections where wormswere visible epithelial cells were distortedadjacent to the worm, but otherwise there wasno evidence of architectural damage or activecryptitis. In biopsy specimens from bothnormal controls and childred infected withTrichuris there was occasional mild goblet celldepletion. The colon of a patient with tri-churiasis is shown in fig 1. Specimens from thetransverse colon, descending colon, and rectumof children with TDS were also examined andshowed the same histological features as thoseof caecal mucosa.

Plasma cell numbers in the lamina propriaThe percentage of plasma cells containingIgM, IgG, and IgA were determined in caecal

*:.:::.

A'.t v

Figure I Histological appearance of the caecal mucosain a child with trichuriasis. Although there is an increasein lamina propria inflammatory cells, the glands are longand straight and wellfilled with goblet cells. Noteepithelial synctium (arrow) forming around the head ofthe worm (haematoxylin and eosin).

195

on Decem

ber 6, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.44.3.194 on 1 M

arch 1991. Dow

nloaded from

MacDonald, Choy, Spencer, Richman, Diss, Hanchard, Venugopal, Bundy, Cooper

Table 3 Plasma cells in caecal lamina propria

% Lamina propria cells

Plasma cell isotype

IgA IgM IgG

Controls 26 3 (6 4) 7-2 (3 0) 6 2 (3 0)Patients 34-8 (5 3) 12-3 (3-9)* 3-7 (3-1)

*p < 0 05 compared with IgM in controls (pooled estimate ofvariance).

biopsy specimens from four controls and fivechildren with TDS (table 3). IgA positiveplasma cells predominated in both groups.IgM positive plasma cells were more abundantthan IgG plasma cells. There were significantlymore IgM plasma cells in the caecum of thechildren with TDS than controls.

Lamina propria CD3+ and CD25+ T cellsCD3 + cells were counted to establish if worminfection was associated with an increase in thedensity of lamina propria T cells. Those cellsexpressing CD25 (the # chain of the IL-2receptor) were also quantified to determine ifany lamina propria cells were activated. CD25is expressed on activated T cells and macro-phages; but these two cell types can be differen-tiated as the CD25 + macrophages are usuallylarge, foamy cells which predominate in thesubepithelial layer.9 There were no differencesin the prevalence of CD3 + and CD25 + cellsin the lamina propria of controls and childrenwith TDS (fig 2). In both groups CD3 + cellsmade up about 20-60% of the lamina propriacells (fig 3). The prevalence of CD25 + cells inthe lamina propria varied widely in bothgroups.

Epithelial T cellsDespite a wide variation in the numbers ofintraepithelial CD3+ T cells in the controls, inchildren with TDS there was a significantdecrease in their number (fig 4). Most

w*'+w... _~~~~~~~}}

70r

60 F

50

-!

0

M

..Q

co.E

308

20 -

10

u'

0

S0

000 00

00

00

0

*I. 0

0

0

:1 01

_I 0

00

0* 99

0Ilk

0~0?I

.

C' 0

U-

A

CD3 CD25

Figure 2 The percentage of lamina propria cellsexpressing CD3 or CD25 in controls or children withTDS. Although there is a wide range in both groups, nodifferences are apparent.

intraepithelial lamina cells are CD8+ and asexpected, there was also a decrease in epithelialCD8 + intraepithelial cells in TDS. CD4 +intraepithelial cells were rare in both groups. ybTcR + intraepithelial cells were also decreasedin patients with TDS. This result also showsthat most of the CD3 + cells in the colon usethe ac TcR.

HLA-DR and VLA-1 expression on epitheliumThe epithelial expression of HLA-DR andVLA-1 can be used as an index of colonic

Figure 3 (A) CD3+ T cells in the lamina propria and epithelium immediately below a worm (APAAP).(B) CD3+ T cells in the lamina propria and epithelium of a control child (APAAP).

196

I

.t

on Decem

ber 6, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.44.3.194 on 1 M

arch 1991. Dow

nloaded from

The caecum in children with the Trichuris dysentery syndrome

90

80

70

60E

X50

E

. 40a

w

30

197

*

p<0-0050

8

. 820 F10

0

888t

0 ~~~~~~~~~~00%CD3 IC,

CD3 CD4'

0

P<0008 e0 pc05

* o

8

CD82

CD8

"I

0b,

inflammation.'""3 As expected, HLA-DRexpression was absent or only weakly positivein controls (table 4). Most of the children withTDS also showed weak or absent epithelialstaining with HLA-DR, although one childshowed strong positive staining. VLA-1expression showed a similar pattern, with nodifference between the controls and patients(table 4).

Epithelial cell proliferationThe antibody Ki67 was used to assess epithelialcell proliferation.'4 In control children 8-2 (SD2 0)% of the crypt epithelial cells were Ki67positive. In contrast, in children with TDS,26 1 (SD 4-5)% of the cells were Ki67 positive,indicating a three-fold increase in cellproliferation.

DiscussionWe provide here a detailed histopathological

Table 4 HLA-DR and VLA-1 expression on caecal epithelium

HLA-DR expression VLA-1 expression

Location of epithelium

Surface Crypt Surface Crypt

Controls1 + ++ +2 + + + +3 - - + +4 + - + +5 - ++ +6 ++ - ++ +

Patients1 - - + +2 + + +++ +3 - - + +4 + + + +5 ++ + _6 + +++ +7 + + ++ +8 + - + +9 ++ + ++ +10 - - +11 + + +

-, no staining; +, very weak staining; +, clearly positive, but weak; + +, moderately positive;+ + +, strongly positive.

and immunohistological study of the caecalmucosa in children with trichuriasis. A prin-cipal observation is that despite the presence ofheavy worm burdens, and the passage offrequent bloody mucoid diarrhoeal stools,there is little evidence of an inflammatoryresponse in the colonic mucosa, apart from inthe immediate vicinity of the worm, whereepithelial damage does occur. These studiesconfirm an earlier necropsy report on twochildren with massive trichuriasis (as well asother disorders)."5 In a review of trichuriasis,Pawlowski mentioned that infiltration of lym-phocytes, plasma cells, and eosinophils is seenin heavy infections with Trichuris trichiura, butno illustrations or quantitative data were

presented.'6 Here we show that using standardhistopathological features it was impossible toattribute the cellular infiltrate seen in the colonto Trichuris as most of the control children hadthe same changes.The mild non-specific inflammatory changes

in the caecum of Jamaican control children, inthe absence of any kind of colonic disease, isperhaps surprising (such as cases 1 and 4 intable 1). At colonoscopy these children did nothave any worms in their caeca. It is unlikelythat they would have been missed for theworms are 4-5 cm long and are easily visiblethrough the endoscope. We, of course, cannotrule out the possibility that the controls were

recovering from mild trichuriasis and had justeliminated their worms. The prevalence of Ttrichiura in children in western Jamaica is55%,17 but figures for Kingston and thesurrounding rural areas are not available. Mostchildren also only harbour a few worms.23 Thebest explanation for the increased inflammatoryinfiltrate in the control children is that it reflectsthe poor standards of hygiene and the con-taminated environment in which they live,which in turn reflects the low socioeconomicstatus of most of the children.Bloody diarrhoea in the developed world is

often associated with the idiopathic colitides,ulcerative colitis or Crohn's colitis. None ofthefeatures characteristic of these disorders,'8 9

however, such as crypt abscesses in ulcerativecolitis, or a dense lymphocytic infiltrate andgranulomata of Crohn's colitis, was apparent inthe mucosa of the children with trichuriasis.The mucoid diarrhoea is undoubtedly due toincreased mucus secretion. We have shownthat increased crypt cell proliferation isindicative of increased cell production andturnover in the colon of these children. It isknown that immune complexes and IgEmediated anaphylaxis in the gut can causeincreased release ofmucus from goblet cells.202'Children with trichuriasis have high titres ofIgG, IgM, and IgE antibodies to Trichuris intheir serum,22 so it is possible that thesehumoral events are important in goblet cellsecretion. Increased epithelial crypt prolifera-

tion in the gut can be induced by activated Tcells in the lamina propria,8 but this is unlikelyto account for the increased proliferation seenin trichuriasis because activated T cells wereuncommon in the lamina propria. A more likelyexplanation is that worm damage to the surface

Figure 4 Numbers ofepithelial T cells arereduced in children withTDS.

a %J

on Decem

ber 6, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.44.3.194 on 1 M

arch 1991. Dow

nloaded from

MacDonald, Choy, Spencer, Richman, Diss, Hanchard, Venugopal, Bundy, Cooper

epithelial cells induces a compensatory increasein crypt cell proliferation.The blood in the stools of the children

probably leaked out from areas where the wormhead had burrowed into the mucosa, becauseno ulcers were seen in the mucosa. It has beenshown, however, that blood loss in trichuriasisis only 5 pl/worm/day,23 so that in a child with1000 worms only 5 ml would be lost a day. It isdifficult to explain how such a small volumecould cause overtly bloody stools. Electronmicroscopical examination shows thaterythrocytes can be seen to cross intact largebowel epithelium in mild ulcerative colitis.24Perhaps non-specific increased colonic bloodflow in trichuriasis could result in red cellleakage from the sub-epithelial vasculararcades and across the epithelium. Electronmicroscopical studies are currently under wayto address this problem.Animal experiments have shown that T cells

have an important role in defence againstparasitic nematodes.25 We had therefore expec-ted to identify activated T cells around theanterior end of the worm embedded below theepithelium. This, however, was not the case. Tcell numbers in the lamina propria were thesame in controls and childred infected withTrichuris, even in the areas around the wormhead, and activated T cells were rather uncom-mon in most of them. Because Trichuris tri-chiura lives with its anterior end buried withinthe colonic epithelium we considered that theparasite might induce an intraepithelial T cellinfiltrate, but this was not the case and thenumbers of these cells were actually reduced inthe infected children. This could be due toincreased loss of intraepithelial cells into thelumen or decreased migration into the epith-elium.There has been intense interest in cells

expressing the yb TcR + in the gut epitheliumin recent years. In man these cells only con-stitute about 10% of the CD3 + intraepithelialcells; but this proportion is increased inuntreated coeliac disease.2"2 There are alsoincreased numbers of yb TcR + T cells in thechronic lesions of leprosy and leishmaniasis.9Intraepithelial yb TcR+ T cells, however,were also decreased in children with TDS.Taking all the results together, there was

little immunohistological evidence for a localinflammatory response in the mucosa. IgA andIgG plasma cell numbers were the same as incontrols, although there was a slight increase inIgM plasma cells; T cell numbers were thesame; and the stigmata of local immune re-actions seen in the idiopathic colitides, such asincreased epithelial HLA-DR and VLA-1expression, were not apparent. This suggeststhat there is no cell mediated anti-parasitemechanism operating in the mucosa of thesechildren. There is in fact no evidence foracquired immunity to Trichuris. The childrenwith TDS may be a selected population: theirheavy worm burdens may result from failure tomount a protective response (as opposed to aserum antibody response which may be of noprotective value). On the other hand, thechildren with heavy infestations may merely be

the ones who are exposed to a lot of eggs.Studies are in progress to determine if childrenwith TDS have parasite specific T cells in theirblood and colonic lamina propria.

In conclusion, our findings concur withprevious descriptions of trichuriasis as anessentially non-pathological helminthic infes-tation.' The children examined in this studywere selected in the sense that they had excep-tionally high worm burdens which causedTDS and growth retardation. Even in theseselected children, however, the colonicimmunopathological changes were minimal.Why the antigenic load of hundreds andsometimes thousands of metabolically active,4-5 cm long, metazoan parasites living withtheir heads buried in the colonic epitheliumshould fail to elicit mucosal inflammation is ofextreme interest and is the subject of ongoingstudies.

This work was supported by the Wellcome Trust.

1 Ramsey FC. Trichuris dysentery syndrome. West IndMedJ1962;11:235-9.

2 Bundy DAP. Epidemiological aspects of Trichuris andtrichuriasis in Caribbean communities. Trans R Soc TropMed Hyg 1986;80:706-18.

3 Bundy DAP, Cooper ES. Trichuris and trichuriasis inhumans. Adv Parasitol 1989;28:107-73.

4 Cooper ES, Bundy DAP, MacDonald TT, Golden MHN.Growth suppression in the Trichuris dysentery syn-drome. Eur J Clin Nutr 1990;44:285-91.

5 Silverman FN. Regional enteritis in childhood. AustPaediatr J 1966;11:20.

6 O'Donogue DP, DawsonAM. Crohn's disease in childhood.Arch Dis Child 1977;52:627-32.

7 Isaacson PG, Wright DH. Immunocytochemistry of lym-phoreticular tumours. Immunocytochemistry. In: PolakJM, van Noorden S, eds. Practical applications in path-ology and biology. Bristol: John Wright & Sons, 1983:249.

8 MacDonald TT, Spencer J. Evidence that activated mucosalT cells play a role in the pathogenesis of enteropathy inhuman small intestine. J Exp Med 1988;167:1341-9.

9 Choy M-Y, Walker-Smith JA, Williams CB, MacDonaldTT. Differential expression of CD25 on mucosal T cellsand macrophages distinguishes the lesions in ulcerativecolitis and Crohn's disease. Gut 1990;31:1365-70.

10 Selby WS, Janossy G, Mason DY, Jewell DP. Expression ofHLA-DR antigens by colonic epithelium in inflammatorybowel disease. Clin Exp Immunol 1983;53:614-18.

11 Scott H, Sollid LM, Fausa 0, et al. Expression of majorhistocompatability Class 11 subregion products by jejunalepithelium in patients with coeliac disease. Scand JImmunol 1987;26:563-71.

12 MacDonald TT, Weinel A, Spencer J. HLA-DR expressionin fetal human small intestinal epithelium. Gut1988;29:1342-8.

13 MacDonald TT, Horton MA, Choy MY, Richman PI.Increased expression of the laminin/collagen receptor(VLA-1) on epithelium of inflamed human intestine.J Clin Pathol 1990;43:313-15.

14 Gerdes J, Lemke H, Baisch H, Wacker H-H, Schwab U,Stein H. Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the mono-clonal antibody Ki-67. J Immunol 1984;133:1710-15.

15 Hartz PH. Histopathology of the colon in massive trichoce-phaliasis in children. Doc MedGeorg Trop 1953;5:303-13.

16 Pawlowski ZS. Trichuriasis. In: Adel AF Mahmoud,Warren KS, Warren S, eds. Tropical and geographicalmedicine. New York: McGraw Hill, 1984:380-5.

17 Speed JC, Culpepper V, Thompson DE, Henson R, Wint B,Bundy DAP. A community based study ofgastrointestinalhelminth and protozoan infection in western Jamaica.West Ind Med J 1987;36:73-5.

18 Rappaport H, Burgoyne FH, Smetana HF. The pathologyof regional enteritis. The Military Surgeon 1951;109:463-502.

19 Lockhart-Mummery HE, Morson BC. Crohn's disease(regional enteritis) ofthe large intestine and its distinctionfrom ulcerative colitis. Gut 1960;1:87-105.

20 LakeAM, Bloch KJ, Sinclair KJ, WalkerWA. Anaphylacticrelease of intestinal goblet cell mucus. Immunology1980;39:657-65.

21 Walker WA. Role of the mucosal barrier in antigen handlingin the gut. In: Brostoff J, Challacombe SB, eds. Foodallergy and intolerance. London: Bailliere Tindall,1987:209-20.

22 Bundy DAP, Bianco AE, Lillywhite JE, Didier JM,Simmons I. Human antibody responses to Trichuris

198

on Decem

ber 6, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.44.3.194 on 1 M

arch 1991. Dow

nloaded from

The caecum in children with the Trichuris dysentery syndrome

trichiura: age-dependency of infection status and serumantibody level. Trans R Soc Trop Med Hyg (in press).

23 Layrisse M, Aparcedo L, Martinez-Torres C, Roche M.Blood loss due to infection with Trichuris trichiura. Am JTrop Med Hyg 1967;16:613-19.

24 Donnellan WL. Early histological changes in ulcerativecolitis: a light and electron microscopic study. Gastro-enterology 1966;50:519-40.

25 Ogilvie BM, Love RJ. Cooperation between antibodies andcells in immunity to a nematode parasite. Transplant Rev1974;17:147-68.

26 Spencer J, Isaacson PG, Diss TC, MacDonald TT. Expres-sion of disulphide linked and non-disulphide linked formsof the T cell receptor gamma/delta heterodimer in humanintestinal intraepithelial lymphocytes. Eur J Immunol1989;19:1335-9.

27 Halstensen TS, Scott H, Brandtzaeg P. Intraepithelial Tcells of the TcRy5 CD8- and V 1/Jb1 + phenotypes areincreased in coeliac disease. Scand J Gastroenterol1989;30:665-72.

28 Jarry A, Cerf-Bensussan N, Brousse N, Selz F, Guy-GrandD. Subsets ofCD3+ T cell receptor (af or yb) and CD3-lymphocytes isolated from normal human gut epitheliumdisplay phenotypical features different from their counter-parts in peripheral blood. Eur J Immunol 1990;20:1097-104.

29 Modlin RL, Pirmez C, Hofman FM, et al. Lymphocytesbearing antigen-specific yb T-cell receptors accumulate inhuman infectious disease lesions. Nature 1989;339:544-8.

30 Warren KS, Mahmoud AAF. Algorithms in the man-agement and diagnosis ofexotic diseases. IX. Trichuriasis.J Infect Dis 1976;133:240-3.

Eponym in pathology . . .

von LEYDIG, Franz (1821-1908) was aGerman comparative anatomist, born inRothenburg. He qualified from Wurzburg in1844. He was first a lecturer in Wurzburg, then,in 1855, was appointed professor of anatomy.In 1857 he moved to Tubingen and in the sameyear he described the interstitial cells of thetestis (Leydig cells). In 1875 he was appointedprofessor of comparative anatomy at Bonn, afield in which he was a pioneer.

von KUPFFER, Karl Wilhelm (1829-1902)was a German anatomist and embryologist,born in Munich. He was professor of anatomysuccessively at Kiel (1867), Koenigsburg(1875), and Munich (1880). It was while inKoenigsburg in 1876 that he first described thestellate cells lining the sinusoids of the liver(Kupffer cells). As well as this, he made impor-tant contributions to embryology, and theembryonic outgrowths from the Wolffian ductswhich form the ureters also bear his name.

199

on Decem

ber 6, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.44.3.194 on 1 M

arch 1991. Dow

nloaded from