196 J Clin Pathol 1990;43:196-200 Mast cells in skin lesions J Clin Pathol: first published as 10.1136/jcp.43.3.196 on 1 March 1990. Downloaded from

I A Cree, G Coghill, J Swanson Beck

Abstract prominent, there is an influx of , The variability ofmast cell density within proliferation is evident, and and between leprosy skin lesions was formation occurs.7 Patients of the same classi- examined as a basis for future studies, fication on the Ridley-Jopling scale vary in and whether the number of mast cells in their susceptibility to reversal reactions. the lesion was determined by local or It has long been known that mast cells are systemic factors was evaluated. The mast intimately concerned with the production of cell density in the , skin oedema, one ofthe features ofreversal reaction. appendages, and intervening dermis was Their possible role in the pathogenesis of such assessed by counting mast cells in glycol reactions was investigated by Chowdhury and methacylate sections stained with Ghosh,5 who found a lower density ofmast cells Giemsa stain and relating these counts to in tuberculoid lesions in reaction compared area measurements obtained by plani- with non-reactive tuberculoid lesions. There is metry. In biopsy specimens taken from also some evidence of degranulation the edge of established lesions the density in erythema nodosum leprosum (ENL, type II of mast cells within the granulomata was leprosy reactions).9 O The aim ofthis investiga- considerably higher than that in the tion was to examine the variability of mast cell intervening dermis and was comparable density within and between leprosy skin lesions with that found in the appendages. No as a basis for future studies, and to ascertain major differences in mast cell density whether the number of mast cells in the lesion were found between unaffected skin and is determined by local or systemic factors. the centre or edge of individual lesions. Three methods of staining mast cells are Mast cell densities in biopsy specimens currently in use: (a) metachromatic staining of from the edge of different lesions on the their granules"; (b) histochemical demonstra-

same patient were also similar, suggest- tion of enzymes12; and (c) immunoperoxidase http://jcp.bmj.com/ ing that the mast cell density within the staining using monoclonal antibodies.'"" As granulomata is independent of the site of methods (b) and (c) both require frozen sec- the lesion and is determined systemically. tions metachromatic staining was the only option available for this study. The method developed by Vogt et al" using glycol meth- The skin lesions of leprosy contain many if not acrylate sections stained with Giemsa stain was all of the cell types which occur in normal skin, chosen for this study because it combined the on September 28, 2021 by guest. Protected copyright. in addition to those which migrate into the excellent morphological detail of plastic lesions from peripheral as part of the embedding with a reliable method of showing granulomatous process. The contribution of the presence of metachromatic granules. cells normally present in the dermis to granu- loma formation and progression is uncertain. Mast cells have received little attention in leprosy recently, but evidence linking them Methods with the development of delayed hypersen- Two groups of 4 mm skin punch biopsy sitivity reactions' raises the possibility that they specimens were obtained under local anaes- might be of some importance in leprosy lesions. thesia with 10O lignocaine from patients with Mast cells are heterogeneous with respect to leprosy in India and Bangladesh. All of the their morphology, biochemistry, and func- patients were untreated or had received treat- tion.2" The differences between them appear to ment (WHO multiple drug therapy regimen) on the tissue and species from which for less than one month. Multiple biopsy Department of depend Pathology, Ninewells they are derived, so that is is not possible to specimens were taken from individual patients; Hospital and Medical apply the results of studies of mast cell dis- for the examination of intralesional variation in School, Dundee, tribution or function in one site or animal to mast cell numbers, specimens were taken from Scotland I A Cree another with any certainty. the centre of the lesion, the edge of the lesion, G Coghill Type 1 reversal reactions are thought to and from the apparently unaffected skin out- J Swanson Beck reflect increased delayed hypersensitivity to side the lesion at a point 2 cm from the nearest Correspondence to: Mycobacterium leprae and account for much of identifiable margin of the lesion in 22 patients Dr I A Cree, Department of of interlesional Pathology, Ninewells the nerve damage which occurs in borderline (biopsy series 1); for the study Hospital and Medical leprosy patients.5 Clinically there is inflamma- variation specimens were taken from the edge School, Dundee, DD1 9SY, same in Scotland. tion of both skin lesions and affected nerves, of two different lesions on the patient was made to Accepted for publication with pain, swelling, and tendemess on pal- 35 cases (biopsy series 2). An effort 31 August 1989 pation.6 Histologically, intercellular oedema is choose lesions of disparate size and appearance Mast cells in leprosy skin lesions 197

Table 1 Age and sex distribution ofpatients and Ridley-Jopling classification BIOPSY SERIES 1 The average mast cell density in biopsy Age Sex Ridley-Jopling classification specimens from unaffected skin was 17 6 per ratio Mean (SD) (M:F) Idt TT BT BB BL LL Total mm2. Specimens taken from the centre of the J Clin Pathol: first published as 10.1136/jcp.43.3.196 on 1 March 1990. Downloaded from lesion had an average of 18 7 mast cells per Series 1 36 (13 0) 16:6 1 0 14 0 3 4 22 mm2, while at the edge of the lesion the average Series 2 38 (15 1) 28:7 0 1 19 2 1 12 35 mast cell density was 19 5 per mm2. There was Idt = indeterminate; TT = tuberculoid; BT = borderline-tubeculoid; BB = borderline; no significant difference between the densities BL = borderline-lepromatous; LL = lepromatous. in either of these sites, suggesting that the total number of mast cells in the skin does not where possible to determine the maximum change greatly within the lesion and is similar variation which might occur. The number, age, to that in normal skin. When the component and sex distribution, and Ridley-Jopling clas- mast cell densities from different sites in the sification of the patients in each series is shown same lesion (granuloma, appendages, and in table 1. intervening dermis) were compared in each After being transported to Dundee in 4% biopsy specimen there was no discernible pat- buffered formaldehyde each biopsy specimen tern in either paucibacillary or multibacillary was bisected: one half was embedded in paraf- patients. fin wax, the other in glycol methacrylate." Paraffin wax sections (5 gim) were stained with Adikim haematoxylin and eosin and by the Wade-Fite 10oorIr ..de r Aa. W..Awam method for diagnostic purposes, while 2 gm 7vr A.. sections were cut from the glycol-methacrylate ... tow embedded half of each biopsy specimen and a stained for mast cell quantitation using the Giemsa method developed by Vogt et al." 4Jt ..S The dermal area and the areas of the section 5 occupied by granuloma and by skin append- ages (sebaceous glands, sweat glands, and hair follicles) were measured by planimetry.'6 Mast i cell counts were performed on non-overlap- I ping fields at x 400 magnification by scanning I the whole section using a 25 square eyepiece A._ grid. To avoid overestimation of mast cell density cells were counted only if part of the nucleus could be seen, and anucleate fragments were ignored. Mast cell counts were expressed for each Auk -Aftl http://jcp.bmj.com/ component (dermis, granuloma, and append- sV.w_ age) as cells per mm.2 The results for paucibacillary and multibacillary cases were Figure I Part of an epithelioid cellgranuloma in a compared using the Mann-Whitney U test for biopsy specimen takenfrom a patient with borderline- non-parametric data. As lesion selection may tuberculoid leprosy. Two mast cells (arrows) within the have occurred when taking biopsy specimens gramuloma show prominent nucleoli and many granules. (Giemsa.) on September 28, 2021 by guest. Protected copyright. from different lesions (biopsy series 2) the results in this group were randomised before the relation between mast cell densities in -e: -e individual patients was assessed by linear regression analysis.

Results The appearance of the mast cells in leprosy lesions is very variable. Several mor- phologically distinct variants were present: "active mast cells" with large ovoid nuclei, prominent nucleoli, and extensive cytoplasm with many granules; "fusiform mast cells" with elongated dark-staining fusiform nuclei and little cytoplasm; and "resting mast cells" with small dark nuclei, scanty cytoplasm, and few granules. These cell types were most com- monly found in inflammatory foci (fig 1), applied to sebaceous or sweat glands (fig 2), and in the unaffected dermis (fig 2), respectively. There were too many cells of intermediate Figure 2 A fieldfrom the unaffected dermis in a case of morphology, however, to allow these types of borderline-tuberculoid leprosy showing a dermal mast cell. to be differentiated and counted (arrow) with dense granules and an elongated mast cell mast cell applied to the side of a capillary adjacent to a sweat separately on morphological grounds alone. gland (arrowhead). (Giemsa.) 198 Cree, Coghill, Swanson Beck

Table 2 Average mast cell densities (per mm2) in skin 400- and component parts of specimens takenfrom edge of leprosy lesions

350- J Clin Pathol: first published as 10.1136/jcp.43.3.196 on 1 March 1990. Downloaded from Total Grade n = Granulonia Appendage Dermis skin PBL 35 1147 773 10-3 173 300- SD (79 5) (65 6) (4 3) (6 6) MBL 22 75.6 65 1 8 5 17 2 SD (73 7) (80 2) (6 9) (11.3) 250- 0 total 57 99.7 72-8 9 6 0 17 3 S SD (790) (70 8) (5 5) (8 6) e4 E *000 E 200- 0 0 S BIOPSY SERIES 2 U' 0@ The overall mast cell densities in biopsy ' 150- 0 000 S specimens from the edge of different lesiions on *000 0 the same patient showed some sinnilarity xoo (r = 0 559, slope = 0 534) (fig 3a), desk?ite the 000 0* 0 0 differences in mast cell density known t(D occur 50- 0000 0 between different anatomical locations iin nor- 0@@ 0.. mal human skin."7 When the componerit mast 0*: 0 were 0- 0 *00000 cell densities compared between diifferent PBL lesions, the strongest correlation was found MBL within the granulomata (r = 0-854, Figure 4 Comparison ofgranuloma mast cell densities for edge biopsy specimensfrom paucibacillary (PBL) and slope = 0 827) (fig 3b), and there waIs little multibacillary (MBL) patients, showing pronounced correlation in the mast cell densities fr4om the variation in granuloma density between patients. The appendages (r = 0 107, slope = 0 0()7) or granuloma mast cell is significantly lower in intervening dermis (r = 0A481, slope multibacillary patients (p < 0-04). = 0354). To examine possible relations betwee:n mast from the edge of the lesion in biopsy series 1 cell density and Ridley-Jopling classifiication, (table 2). In those from the edge of the lesion the results from edge biopsy specimenLs from the density of mast cells was greatest in the different lesions (biopsy series 2) were suimmed granulomata (average density = 99-7 per and considered with the results from those mm2), but large numbers were also found around skin appendages (average den- sity = 72 8 per mm2). In the intervening Figure 3 Correlations (with regression lines dermal connective tissue mast cells were shown) between mast cell uncommon (average density = 9 6 per mm2) 0 http://jcp.bmj.com/ densities in biopsy m and were usually of "resting" morphology. specimensfrom the edge of There was no significant difference between two separate lesions on the IX same patient. (biopsy 0. the overall mast cell densities in biopsy series 2). 'CO specimens from the edge of paucibacillary or 0 ~ (a) Total skin mast cell multibacillary patients (table 2). The gran- densities (r = 0 559, .0 E uloma mast cell density (fig 4) is reduced in p < 0001) E (b) Granuloma mast cell m patients with a multibacillary compared on September 28, 2021 by guest. Protected copyright. densities (r = 0 854, .1;E paucibacillary patients (Mann-Whitney U- p < 0 001). . test, p < 0-04), as is the mast cell density in the .0. intervening dermis compared with the gran- uloma mast cell density (Mann-Whitney U- test, p < 0-02). The mast cell densities within -_ the appendages showed no appreciable dif- 40 ference between paucibacillary and mul- Most (Biopsy A) tibacillary patients. To investigate the pos- sibility that mast cells might be moving into the granuloma from either the appendages or the intervening dermis, the relation between com- m * ponent mast cell densities was examined, but there was no inverse correlation between the cL 0a._ granuloma, appendage, and intervening der- 'CB . mal mast cell densities in individual patients.

E . E in Discussion The methacrylate sections stained with

0- Giemsall used in this study provided excellent IX histological detail. All cells containing meta- chromatic granules could be counted easily, even ifthe cells were partially degranulated and i only a few granules were present. Although 0 100 400 Vogt et al state that human can be detected using this method, their differentia- Mast cells in leprosy skin lesions 199

tion from mast cells in human tissues is odological differences. The average number of problematical and relies on their polymor- mast cells at the centre and the edge of the phonuclear morphology, which is not easily lesion is apparently slightly smaller than it is appreciated, even in thin sections." outside the lesion, although this may reflect the J Clin Pathol: first published as 10.1136/jcp.43.3.196 on 1 March 1990. Downloaded from Unequivocal basophils were not seen in the loss of skin appendages at the centre of tuber- sections examined during this study and all culoid and borderline lesions. The degree of cells containing metachromatic granules were migration of mast cells into leprosy gran- included in the counts. ulomata, either from the surrounding dermis or The role of mast cells in delayed hypersen- as precursor cells from the blood, may sitivity reactions has attracted considerable therefore be small in relation to the overall interest in recent years' from those studying numbers present in the skin. Comparison ofthe mast cell function in experimental animals. component densities at different sites within Interleukin 3 (IL-3) production by T lym- the lesions showed no rise or fall in mast cell phocyte is implicated in mast cell precursor density within the granuloma, appendages, or maturation and that T lymphocytes may intervening dermis. This finding contrasts promote mast cell division and that other with both granuloma fraction and apoptotic mediators cause mast cell degranulation which density, both ofwhich are highest at the edge of is not IgE dependent."' the lesion.2' In leprosy mast cells were first investigated Biopsy specimens from the edge of different by Sen Gupta and Ghosh, who published their lesions in the same patient showed considerable observation that mast cell granules could be variation in overall mast cell density, in keeping stained by acid fast methods such as the Fite- with previous studies of the distribution of Faraco method.'9 Using toluidine blue stained human mast cells in skin.22 The correlation in paraffin wax sections, Chowdhury and Ghosh mast cell densities within the granulomata from showed that the density of mast cells within different lesions, however, is unexpectedly tuberculoid lesions was lower if the lesions strong (r = 0 854), with a slope close to that were in reaction.8 They noted that these which would be produced by complete observations mirrored the content of agreement. This suggests that the number of non-reactional lesions and those in reaction. mast cells within leprosy granulomata is Mast cells which have partially degranulated, independent of the site of the lesion, and that however, possess few remaining granules420 the mast cell density within granulomata is and may be missed in paraffin wax sections mainly determined by the systemic response to stained with toluidine blue. Their results infection by M leprae and not by local factors. probably indicate widespread degranulation of The density of mast cells within the gran- mast cells in reactions. The presence ofoedema ulomata is considerably higher than that in the may also be important as this would increase intervening dermis and is comparable with that

dermal volume and result in an apparent found in the appendages. Several of the cases http://jcp.bmj.com/ decrease in mast cell density. A similar increase studied show much higher mast cell densities in dermal volume in multibacillary patients is within their granulomata than average (fig 4). likely to be the cause of the lower granuloma In the absence of sequential studies of biopsy mast cell density in these patients compared histology during treatment and reactions in with paucibacillary patients. leprosy the clinical importance of these dif- Mast cells are a heterogeneous population of ferences is not clear, but the fact that such

cells. In different tissues they exhibit very differences exist is of interest, and the results of on September 28, 2021 by guest. Protected copyright. different biochemistry and show functional the seminal study by Chowdhury and Ghosh8 differences.24 In the biopsy specimens suggest that differences in mast cell density examined here they exhibited a wide range of might be related to a tendency to develop morphological types which seemed to be com- reactions. The hypothesis that those patients mon in different sites. Although total mast cell with higher granuloma mast cell densities degranulation occurs in vitro, mast cell activa- might be susceptible to more damaging rever- tion in tissues is thought to provoke partial sal reactions due to the release of mediators degranulation and continuous production of from the mast cells seems attractive and is various inflammatory mediators, including worthy of further study. and histamine.34 Their func- tional activity in the lesion may well be more We are most grateful to LEPRA for funding the collection ofthe important than their overall number, and fur- biopsy specimens. We thank Dr J Mehta, C A Gardiner, C ther studies of mast cell morphology in leprosy Fisher, I Cochrane and R Hart for their assistance. We also thank Mr R Fawkes, Mrs S Gibbs, and Mr S MacPherson for lesions using both light and electron micros- preparation of the figures. copy would be worthwhile. The quantitative data show that in biopsy specimens from the edge and centre of leprosy 1 Galli SJ, Dvorak AM. What do mast cells have to do with delayed hypersensitivity? Lab Invest 1984;50:365-8. lesions and outside the lesion, the total number 2 Barrett KE, Metcalfe DD. Mast cell heterogeneity: evidence of mast cells present does not change sig- and implications. J Clin Immunol 1984;4:253-61. 3 Friedman MM, Kaliner M. In situ degranulation of human nificantly. Although it was not possible to nasal mucosal mast cells: ultrastructural features and cell- compare the results of this study with those cell associations. J Allergy Clin Immunol 1985;76:70-82. 4 Friedman MM, Metcalfe DD, Kaliner M. Electron micro- from normal subjects, the mast cell density in scopic comparison of human nasal and lung mast cell clinically unaffected skin was found to be 17-6 degranulation. In: Befus AD, Bienenstock J, Denburg JA, eds. Mast cell differentiation and heterogeneity. New York: per mm2, a figure comparable with the mast cell Raven Press, 1986:367-78. density found in normal skin by other work- 5 Ridlcy DS. The pathology of leprosy. In: Hastings RC, ed. Leprosy. Edinburgh: Churchill Livingstone, 1985:100-33. ers,7 22 if allowances were made for meth- 6 Pfaltzgraff REP, Bryceson A. Clinical leprosy. In: Hastings 200 Cree, Coghill, Swanson Beck

RC, ed. Leprosy. Edinburgh: Churchill Livingstone, technique using a murine against 1985:1 34-76. tryptase. Am J Pathol 1986;124:427-35. 7 Ridley DS, Radia KB. The histological course of reactions in 15 Harvima IT, Naukkarinen A, Harvima RJ, Fraki JE. borderline leprosy and their outcome. Int J Lepr Other Immunoperoxidase and enzyme-histochemical demon- Mvcobact Dis 1981;49:383-91. stration of human skin tryptase in cutaneous mast cells in J Clin Pathol: first published as 10.1136/jcp.43.3.196 on 1 March 1990. Downloaded from 8 Chowdhury SK, Ghosh S. Distribution of tissue mast cells normal and mastocytoma skin. Arch Dermatol Res in "reaction in tuberculoid leprosy". Bull Calcuitta School 1988;280:363-70. Trop Med 1968;16:13-14. 16 Cree IA, McDougall AC, Coghill G, Beck JS. Quantitation 9 Mabalav MC, Helwig EB, -Tolentino JG, Binford CH. The of the granuloma fraction in leprosy skin biopsies by histopathology and histochemistry ofErythema Nodosum planimetry. Int J Lepr Other Mycobact Dis 1985;53:582-6. Leprosum. Int . Lepr Other Mycobact Dis 1965;33:28-49. 17 Cowen T, Trigg P, Eady RAJ. Distribution of mast cells in 10 Ridley DS. Erythema nodosum leprosum. In: Ridley DS, human dermis: development of a mapping technique. Br J ed. Pathogenesis of leprosy and related diseases. London: Derniatol 1979;100:635-40. Wright, 1988:123-34. 18 Haig DM, McMenamin C, Redmond J, et al. Rat IL-3 11 Vogt RF Jnr, Hynes NA, Dannenberg AM Jnr, Castracane stimulates the growth of rat mucosal mast cells in culture. S, Weiss L. Ienproved technique using geimsa stained Imminiunology 1988,65:205-1 1. glycol methacrylate tissue sections to quantitate basophils 19 Sen Gupta PC, Ghosh S. Tissue mast cells. Nature and other leukocytes in inflammatory skin lesions. Stain 1963;197:506-7. Technol 1983;58:193-205. 20 Friedman MM, Kaliner M. Nasal mucosal mast cell 12 Sanderson IR, Leung KBP, Pearce FL, Walker-Smith JA. degranulation. NER Allergy Proc 1984;5:1 19-25. Lamina propria mast cells in biopsies from children with 21 Cree IA, Gardiner CA, Beck JS, Mehta J. Studies of cell Crohn's disease. J Clin Pathol 1986;39:279-83. death () and cell division in leprosy . 13 Rimmer EF, Turberville C, Horton MA. Human mast cells Itot JILepr Other Mx'cobact Dis 1986,54:607-13. detected by monoclonal antibodies. J Clin Pathol 22 Eadv RAJ, Cowen 1', Marshall TF, Plummer V, Greaves 1984;37: 1249-55. MW. Mast cell population density, blood vessel density 14 Craig SS, Dc Blois G, Schwartz LB. Mast cells in human and histamine content in normal human skin. Br J keloid, small intestine, and lung by an immunoperoxidase Dermatol 1979;100:623-33. http://jcp.bmj.com/ on September 28, 2021 by guest. Protected copyright.