INTERNATIONAL. JOIIRNAL or LEPROSY^ volume 51, Number 2Primed in the

The Degradation of Mycobacterium leprae by aComparison of Its Staining Properties'

Marian J. Ridley2

Acid-fast stains whereby the disintegra-tion of Mycobacterium !mac may be as-sessed have for many years provided a valu-able means of indexing chemotherapeuticactivity, and of estimating the cessation ofactive infection and risk of infectivity. Al-though the rate of degradation estimated byacid-fast stains formed the main basis ofclassification at the lepromatous end of thespectrum (") little is known of the processof degradation of the organism or of theelimination of its degraded products. Suchproducts are thought to play a role in theformation of extravascular immune com-plexes in the pathogenesis of erythema no-dosum leprosum (ENL) reactions ( 2). More-over, in vitro studies of lymphocytetransformation using whole or sonicated Al./eprae antigen have demonstrated that hy-persensitivity reactions may be dependenton bacterial disintegration. Reactions thatinvolve skin have high lymphocyte re-sponses to whole Al. Ieprae. Reactions thatinvolve mainly nerves respond to sonicatedAl. !mac, while mixed reactions involvingboth skin and nerves respond to whole andto sonicated M. leprae antigen (').

This study was undertaken to follow theprocess of degradation of Al. leprae by com-paring the results obtained with a) acid-faststains, b) methenamine silver to reveal cellwalls ( 5), and c) the immunoperoxidasetechnique using an antimycobacterial anti-body to identify non-particulate compo-nents. An immunoperoxidase technique wasrecently used to locate `Al. 'mac antigen'in tissues ( 10 ).

MATERIALS AND METHODSPatients. Sixty-nine patients comprising

different groups in the leprosy spectrum were

' Received for publication on 21 September 1982;accepted for publication on 14 December 1982.

M. J. Ridley, B.Sc., Ph.D., Research Assistant,Hospital for Tropical Diseases, St. Pancras Way, Lon-don NW1 OPE, England.

available for study. They included 16 lep-romatous (LL) patients, 8 active untreatedpatients (LLA) and 8 treated patients withadvanced quiescent disease (LLR); 20borderline lepromatous (BL) patients, 15untreated and 5 treated; 8 untreatedmid-borderline (BB); 8 untreated border-line-tuberculoid (BT), and 8 untreated tub-erculoid (TT) patients. Five `histoid . lepro-mas were also studied and four positiveMitsuda skin test biopsies, taken 30 daysafter intradermal injection of 0.1 ml of heat-killed Al. /eprae suspended in saline, wereobtained from four BT patients. Twenty pa-tients undergoing ENL reactions, 15 in theacute stage and five resolving, were alsoavailable for study.

The biopsies were received from the M RCUnits at Sungei Buloh, Malaysia, and AddisAbaba, Ethiopia; from the Hospital forTropical Diseases, London, and from PapuaNew Guinea.

Smears. Slit-skin smears from live pa-tients treated for lepromatous leprosy inwhom no acid-fast bacilli were detected werepost stained with anti-BCG antiserum andthen with methenamine silver.

Five smears of Al. leprcze from the ar-madillo were fixed, two in formol saline andtwo in formol-mercuric-chloride-acetic acid(FMA). These were stained with anti-BCGantiserum. The fifth smear (the control) washeat fixed and stained by Ziehl-Neelsen.

Tissues. Biopsies were fixed in FMA ( 12 )

for routine examination by hematoxylin-eosin and a modified Fite-Faraco acid-faststain ("). Gomori-Grocott ( 7) methenaminesilver impregnation was also used. The for-mol-mercuric-chloride-acetic acid mixtureis used routinely in leprosy because of itsexcellent preservation of cytoplasmic detail.Recently it has been shown to be the fixativeof choice for the immunoperoxidase tech-nique ( 3 ' 6• 9), preserving both intracyto-plasmic and some cell surface components.

Immunoperoxidase technique. Consider-able experience is necessary to carry out this

211

international Journal of Leprosy^ 1983

technique. The method described here hasbeen successfully used by several technicalassistants with precise reproducibility. Thefinal preparation should have no back-ground staining. Many heme-containingmolecules, including hemoglobulin and cat-alase, oxidize diaminobenzidinc-hydrogenperoxide (DAB) to form the insoluble poly-mer. This reactivity varies with fixation, in-cubation time, and pH. Therefbre appro-priate controls are essential for the specificity01 the method.

Perfectly flat serial sections cut at 4 pmand air dried were used in this study. Ex-cessive heat was avoided at all stages in thepreparation of the tissue. The peroxidase-antiperoxidase (PAP) procedure of Stern-berger. et al. ('') and Burns ( 4 ) was carriedout with minor modifications. Trvpsin didnot intensify either the strength or the spec-ificity of staining after FMA fixation and itwas omitted.

Antiserum. Anti-.1/. tem -a(' antiserumraised in the laboratory gave too weak areaction to be of use, and pooled serum fromfive active LL patients was used instead asa source of antibody to A/. //Taw. Thesesections were examined directly after in-cubation with the antibody using peroxi-dase labelled anti-human 1g.

Anti-BCG antiserum was used to detectAl. /eprue antigen since Al. leprae is knownto share antigenic determinants with thisllycobacterium(`). Anti-BCG was obtainedfrom DAKO (Mercia Brocades, Watford,England). it was raised without Freund'scomplete adjuvant and the specificity ofantisera obtained from this source has beendescribed by a number of workers [reviewedby Mason, et al. (u)].

Serial sections of two active LL granu-lomas were used to standardize the tech-nique. dilutions were obtained us-ing a checkerboard system of testing serialdilutions of antiserum against other re-agents. The other reagents were similarlytested for optimal efficiency. Anti-BCG wasused at a dilution of 1/200 in Tris buffer.

Staining method. Sections are adhered toacid-cleaned, dry slides.

1. Take sections to water removing alltrace of wax.

2. Remove mercury with iodine, 2 min.

3. Wash. Remove iodine with 5% sodiumthiosulphate.

4. Block endogenous peroxidase with 30volumes hydrogen peroxide in metha-nol, 1% H2O, in 99 ml methanol for 30min (room temperature).

5. Wash very well in running water fol-lowed by distilled water. 15-20 min.

6. Block excess reaction with normal swineserum diluted 1/5 in Tris, 45 min.

Tris60.57 g Tris/500 ml H 20376 ml N/HCIAdjust pH 7.6Make up to 1 liter with H 20Use: dilute 1/10 with normal saline

7. Tip off swine serum, without washing.8. Treat sections with optimally diluted

rabbit antiserum. 30 min.9. Wash several times in Tris butler.

10. Treat sections with swine anti-rabbit Ig1/20 in Tris, 30 min.

11. Wash very well at least 30 min, chang-ing buffer 3 times.

12. Treat sections with PAP 1/50 (opti-mum dilution), 30 min.

13. Wash in Tris.14. Make up DAB.

DAB (Sigma)Dissolve 0.006 g DAB/ 10 ml TrisDilute 30 volumes H 20 2 : 1 ml

H 20, + 10 ml TrisAdd 3 drops dilute H 20, to DAB

15. Treat sections with DAB-peroxide forup to 5 min. Wash in tap water as soonas they turn brown.

16. Wash in running water, 10 mill.17. Counterstain nuclei in Mayer's hem-

alum. 15 sec.18. Wash to blue.19. Dehydrate, clear and mount.

All sections stained with anti-BCG weretreated as described. However, Mitsuda skinbiopsies required overnight staining withanti-BCG: the sections being kept in a moistchamber at 4°C.

Controls. One section was stained for en-dogenous peroxidase by a solution of DAB-peroxide. A second control was normal rab-

51, 2^ Ridley: Degradation of M. leprae^ 213

bit serum used in place of the antiserum. Athird section was treated with serum afterabsorption of antibody by I3CG (Glaxo,vaccine); 5 pg/m1 13CG was incubated,mixed, centrifuged at 2500 X g, filtered, andused.

Finally, the method has been testedthrough the spectrum of leprosy and it wasMt that this provided a control within thesystem ( 15 ' 16 . 17 ).

Enumeration of bacilli. The Bacterial In-dex was determined as follows:

6+ 1000+ AFB in every field5+ 100-1000 in every field4+ 10-100 in every field3+ 1-10 in every field2+ 1 AFI3 in every 10 fields1+ 1 AFB in every 100 fields

The solid, fragmented, and granular (SFG)index ( 14 ) was used to assess the ratios ofsolid-staining bacilli, to fragmented rods, togranular debris.

RESULTSControl sections were all negative.Sections stained with pooled LL serum as

antibody were usable but gave weaker re-sults than those obtained with anti-BCG se-rum.

Intact bacilli stained equally well by allthree methods. BCG-positive bacilli had abeaded appearance. Bacillary products couldonly be demonstrated by special tech-niques—methenamine silver for cell walls,acid-fast stain for cytoplasm, and anti-I3CGfor degraded or soluble cytoplasmic com-ponents. The means of the Bacterial Indicesare recorded in Figure I.

Spectrum of leprosy. There was a gradualrise in numbers of bacilli from BT to BL,and in untreated patients equal numbers ofbacilli were seen as rod-shaped organismsby all three staining methods. BCG-positivebacilli consistently appeared granular. Somemacrophages contained diffuse BCG-posi-tive debris, which was especially notable inthe BL group. In this group after chemo-therapy there was a diminished number ofbacilli still recognizable by all these meth-ods, but fragmented and granular acid-fastforms were increased, and the abundant in-tracellular I3CG component was diflitserather than granular. Some cells containedonly this degenerative product.

O Silver6

4

if, 3

2

0

TT^BT^BB^BL thistoim LLA IENLI LLR (AM' RI

Spectrum of leprosy

FIG. I. The degradation of .1/. /eprae in untreatedpatients through the spectrum of leprosy, histoid lep-romas and in ENL.

LLA = active lepromatous leprosyLLR = lepromatous leprosy in long-treated pa-

tients in regressionAdv. R. = advanced resolution in lepromatous lep-

rosy

In all the active LL lesions there was anequally high density of bacilli which weresolid by acid-fast stains, intact by silver, andgranular by anti-BCG. In most cases no dif-fuse BCG material was evident, although ina few cases it was striking. Following pro-longed chemotherapy, regressing lepromashad negligible numbers of solid bacilli andfewer fragmented or granular acid-fast ba-cilli. The level of diffuse BCG material incells was much lower, although on occasionit was extracellular. There was no differencebetween active and regressing lesions in thelarge number of cell walls demonstrated bysilver impregnation (Fig. 2). Advanced res-olution was marked by clumps ofaggregatedand diffuse silver, cell-wall deposit.

Localization of antigen. The TT and BTlesions were marked by epithelioid cellgranulomas, while the lepromatous groupcomprised bacteria-laden macrophages. In-termediate in the 1313 group. the lesion wascomposed of activated macrophages.

TT-BT-BB. Solitary acid-fast organismswith intact cell walls were found in the sub-epidermal zone or in small nerve trunks. Ingeneral, macrophages contained intact acid-fast rods which stained with a beaded ap-pearance by anti-BCG. In some BT and BB

r^1,rit

atta, •

'

.°111; ,^

•e.^4111

IliOre"•

Mir^..... k

• r^ a'

FIG. 3. BL leprosy. Diffuse BCG-positive debris(It) appears in the phagocytic vacuole on disinte-gration of the bacillus (t). Immunoperoxidase anti-BCG ( X750).

• 4 • rilbst• 90^•^.ItoOf.

.1111111* 11-w

214^ International Journal of Leprosy^ 1983

FiG. 2. Active lepromatous leprosy. Large numbersof intact cell walls are seen among aggregates of cellwall material. Methenamine silver ( X750).

cases these macrophages also contained asmall amount of diffuse BCG-positive de-bris. Very noticeable was the appearance ofabundant diffuse BCG material in endo-neurial and perineurial macrophages ofnerves in the three groups, more than wasapparent elsewhere in the granuloma or inthe nerve itself.

BI, LL. Intact, acid-fast, BCG-positivebacilli were found in macrophages. DiffuseBCG material when present was seen inphagocytic vacuoles which sometimes ap-peared at the macrophage cell margin (Fig.3). Granular or diffuse BCG componentscorresponding to acid-fast bacillary prod-ucts were demonstrated in nerve trunks andin cells infiltrating the endoneurium or peri-neurium. Silver impregnation revealed in-tact and aggregated cell walls. Some en-dothelial cells of medium-to-large bloodvessels contained only BCG-positive de-bris, and some of this material was seen

close to the nucleus in muscle cells of thelarger blood vessels.

In the granuloma of the epidermal-der-mal junction, especially in LL lesions, solid-staining, acid-fast bacilli were located in thecenter of the lesion with more granular or-ganisms at the periphery. Peripheral mac-rophages contained more BCG-positivedebris and dense deposits of silver-impreg-nated, cell-wall material. This arrangementof cellular activity was reversed on passinglower into the dermal-subcutis region, wherethe most heavily infected macrophages werefound in the advancing edge of the lesion.Extracellular diffuse BCG debris was some-times present in active lesions but silver-impregnated, cell-wall debris was not seenoutside the macrophage. Isolated macro-phages in intercollagenous spaces containedfew granular acid-fast bacilli and moreabundant diffuse BCG-positive material(Fig. 4).

Ilistoid leproma. There was a uniformlyhigh density of intact and solid-staining,acid-fast bacilli which appeared granularwith anti-BCG. Extracellular diffuse BCGdebris was demonstrated especially in thesubepidermal border and in discrete foci

lir

Cvelitv '14/P• •

Aritke•_•

51, 2^ Ridley: Degradation of M. leprae^ 215

• '^+^qr

•

^

.^.

• ' 44AO,

. .

••1 .1

R4,9

1^j• *I I

.^.

1 4 111FIG. 4. Lepromatous leprosy. Isolated macro-

phages in intercollagenous spaces contain diffuse I3CG-positive debris (*). Immunoperoxidase anti-I3CG( X250).

within the granuloma. The material was seenin isolated neutrophils in these lesions.

Erythema nodosum leprosum (ENL). Thenon-reacting area of the lesion was markedby large numbers of intact cell walls, a lessernumber of which were granular with theacid-fast stain. The demonstrable BCGcomponent was variable, mostly intracel-lular, granular and diffuse. This contrastedwith the appearance of bacilli in the reactingareas where much diffuse BCG-positivc ma-terial could be detected extracellularly andin macrophages, especially around fat lob-ules (Fig. 5). Very little acid-fast granulardebris was seen in the degenerating mac-rophages, although there were some intactrods and large aggregates or a diffuse ma-terial by silver impregnation. Neutrophilsin the reacting area contained only BCG orsilver-stained diffuse components.

In resolving ENL lesions there were a fewresidual foam cells containing degraded ba-cillary products, either cell walls or acid-fastor BCG-positive debris. Isolated residualmacrophages in intercollagenous spacescontained I3CG-positive debris.

Nlitsuda skin tests. Small amounts of dif-fuse BCG-positive material were found ina few macrophages scattered among accu-

FIG. 5. ENL. Diffuse BCG-positive material is seenaround fat lobules in the subcutis (*). Immunoperox-idase ( X400).

mulations of epithelioid cells. No cell wallsor acid-fast organisms were detected.

Smears. The five smears negative by acid-fast stains did not stain with BCG. Howevervariable numbers of cell walls were seen af-ter treatment with silver.

Smears from the armadillo contained sol-id or fragmented acid-fast bacilli. FMA fix-ation gave an enhanced granular staining ofAI. leprae, compared to formalin, after anti-BCG.

DISCUSSIONThe immunoperoxidase technique has

been used previously to compare 13CG-pos-itive itt. leprac with those that stain by Fite-Faraco's acid-fast stain. No quantitation wasattempted but the acid-fast technique wasthought to be slightly superior ('"). The dif-ference in the results with anti-BCG in thatstudy compared to those in this study maybe due to technique.

In the present investigation it was difficult

216^ International Journal of Leprosy^ 1983

to stain H. leprae with frprae-specilicantibody. With I3CG antiserum granularstaining was obtained. This suggests thatthere is difficulty of access to the mycobac-terial determinant in the cytoplasm of anintact bacillus.

Degradation of M. leprae by the macro-phage is seen here in three stages: acid-fast(solid, fragmented, and granular); non-acid-fast cell walls revealed by silver im-pregnation, and non-acid-fast, BCG-posi-tive debris. "Solid" bacilli stain by all threemethods but their appearance with anti-BCG is always granular. When the acid-fastorganism becomes fragmented, diffuse I3CGdebris can be seen. In LL the cell wall is stillintact and remains so often until the ter-minal stage of disintegration, by which timeacid-fast and BCG staining is negative.

The rates of degradation vary at differentpoints on the leprosy spectrum, reflectingcell-mediated immunity. Thus they are morerapid in the borderline than in the lepro-matous group. In BT, 1313, and 13L the deg-radation of cell walls, loss of acid-fastness,and total elimination of bacilli are an almostsimultaneous process. In the borderline partof the spectrum the last component of thebacillus to disappear was the BCG-positivedebris; whereas in the lepromatous part ofthe spectrum. it was the cell wall. In re-gressing lepromatous patients, it is evidentthat the macrophages are effete and non-functional, as demonstrated by the lack ofimmunological factors shown by immuno-peroxidase ( 1 "). They would therefore lackthe enzymes required to digest the cell walls,and the leakage of BCG-positive materialfrom damaged cell walls would be a slowprocess. In BB and BT the viable and activegranuloma cells possess the enzymes re-quired to digest the bacterial cell walls. To-tal absorption follows rapidly.

In BB and BT disappearance of BCG-pos-itive material is, however, slower in nervesthan in the epithelioid cell granuloma of thedermis, no doubt due to the barrier to ab-sorption presented by the nerve and its peri-neurium. This would explain the observa-tion that in reversal reactions associatedmainly with nerves the response to soni-cated Al. leprae is greater than that to intactorganisms, and vice versa in reversal reac-tions associated mainly with skin ('). Son-icated and non-sonicated organisms would

both exhibit cell walls, but only in the for-mer would the 13CG-positive matter be ex-posed.

Although histologically a 30-day Mitsudareaction is similar to the granuloma of atuberculoid lesion, 13CG-positive materialwas still present. This indicates that thebreakdown of the bacilli had only recentlybeen accomplished. The Mitsuda reactionis presumably a response to the event.

In ENL there is increased degradation ofbacilli represented by a rapid fall in acid-fast organisms with a concurrent elevationof BCG-positive debris, relative to the sit-uation in non-reacting lesions at a corre-sponding stage of regression. The release ofBCG-positive debris into the extracellularspace by altering the antigen-antibody ratioprecipitates the ENL reaction (''). The rea-son why there should be a rapid build upof degradation products at this point is lessclear. It seems likely that the first stages ofdecay following death affect the morphologyof Al. leprae but not the ratio of acid-fastto BCG-positive products. At some point,often perhaps related to time after com-mencement of chemotherapy, the organ-isms lose their acid-fastness with a propor-tional increase of BCG-positive debris. Bythis time the macrophages are also degen-erate, and if they then release their degra-dation products the ENL cycle is likely tosupervene, subject to there being a suitableantigen-antibody ratio.

As a final technical point, the non-specificstaining of plasma cells (''') (and reportedby other workers) can be blocked by ade-quate exposure to normal swine serum. Thispoint needs critical examination when set-ting up the technique.

SUMMARYThe disintegration of Mycobacterium lep-

rae is revealed by a study of its acid-fastcomponent, its non-acid-fast cell walls us-ing methenamine silver, and its BCG-pos-itive cytoplasmic component. Solid bacillistain by the three stain techniques used toidentify these products, but the I3CG com-ponent is demonstrated only with difficultyand appears granular.

Degradation of Al. leprae is fairly rapidin BT, BB, and BL, and clearance of bacil-lary products occurs almost simultaneouslybecause of the destruction of the cell walls.

51, 2^Ridley: Degradation ()I'M. leprae^ 217

However, clearance is slower in nerves andBCG-positive material persists. The break-down of cell walls is slow in LL and theirclearance is delayed, but I3CG-positive ma-terial is cleared as fast as it leaks out. ENLappears to coincide with a more rapidbreakdown and release of disintegrationproducts from degenerate macrophages. TheMitsuda reaction appears as an epithelioidcell granuloma after complete degradationof IL hp/we with residual BCC; positivematerial at 30 days.

RESUMENSe visualizó la degradación del .11yeobacterium lep-

rae en base al estudio de su componente acido-resis-tente, de sus paredes celulares no-ãcido-resistentes

(usando metenamina argentica). y do su componentecitoplasmico I3CG-positivo. Los bacilos sólidos se ti-

nen las tres tecnicas usadas para identiticar estosproductos pero el componente I3CG solo se demuestra

con dilicultad y aparcce granular.

La degradación del M. leprue es bastante rapida en

los casos 13T, BB, y BL, y la eliminación de los pro-

ductos bacilares ocurre casi simultancamente debido

a la destrucciOn de las paredes cclulares. Sin embargo,la depuración es mas lenta en nervios, persistiendo elmaterial BCG positivo. El rompimiento de las paredes

celulares es lento en los casos LL y su eliminación esretardada, pero el material BCG positivo es eliminado

tan rapid() como se libera. Durante el estado reactionaltipo ENL el rompimiento y la liberación de productos

de degradaciOn a partir de macrOlagos en degeneración

parecen ocurrir mas rapidamente. La reacciOn Mitsuda

aparece comp un granuloma de celula epitelioide des-pués de la degradación total del .11. leprae con material

positivo residual BCG a los 30 dial.

RESUMEOn peut mettrc en evidence la désintegration de My-

cobacterium leprae en Ctudiant le constituant acido-resistant du bacille, les parois cellulaires non-acido-resistantes colorees par lc methenamine d'argent, et le

constituant cytoplasmique positif pour le BCG. Les

bacilles solides se colorent par les trois techniques decoloration utilisees pour identifier ces produits, mail

le constituant I3CG ne peut etre mis en evidence qu'avecdifficult& et revele un aspect granulaire.

La degradation de leprae est plutót rapide chez

les malades BT. BB, et BL. L'Climination des produits

bacillaires survient presque simultanement par suitede la destruction des parois cellulaires. Néanmoins,l'élimination est plus lente dans les nerfs, et du materiel

positif au BCG persiste. La degradation des parois cel-

lulaires est lente chez les malades LL, et !cur elimi-nation est retardee, tandis que le materiel positif auBCG est élimine au fur eta mesure. aussitôt qu'il est

rejete a l'exterieur des bacilles. II semble que l'Ery-theme Noueux LCpreux coincide as cc une accelaration

de In degradation et de In liberation des produits

desintegration provenant des macrophages dégenerés.La reaction Mitsuda apparait comme un granulome

de cellule epithelioide apres la degradation complete

du .11. leprae avec du materiel positif de résidu de BCG

apres 30 jours.

Acknowledgments. I am most grateful to Dr. D. S.

Ridley for reading the manuscript, for independent as-

sessment of the results, and for useful advice. I wouldlike to thank Miss Susan Langeveld and Mr. Adrian

Moore for their technical assistance, and UCH for fi-

nancial support.

REFERENCES

I. BARNETSON, R. S., MUNE, G., PEAnsos, J. M. and

KRosvAt.I., G. Antigenic heterogenecity in pa-

tients with reactions in borderline leprosy. Br. Med.J. 4 (1975) 435-437.

2. 133onvA -ryN, B., 13AnsLisos, R. S.. I:nosy/sit., G.,

Zi'mTn, R. H. and LANII1ERT, P. H. Immune com-

plexes and complement hypercatabolism in pa-tients with leprosy. Clin. Exp. Immunol. 26 (1976)388-396.

3. BossiAN, F. T., LINDEMAN, J., KUIPER, G., VAN

DER WALL, A. and KEENING, J. The influence of

fixative on immunoperoxidase staining of plasmacells in paraffin sections of intestinal biopsy spec-

imens. Histochemistry 53 (1977) 57-62.4. 13t Icxs. J. Background staining and sensitivity of

the unlabelled antibody-enzyme (PAP) method.Comparison with the peroxidase sandwich meth-

od using formalin fixed paraffin embedded ma-

terial. Histochemistry 43 (1975) 291-294.5. COLLINS, F. M., MORRISON, N. E., Diwit:, A. M.

and WA -rsoN, S. R. Microscopic counts carried

out on Mivabacterium leprae and .1/. tuberculosis

suspensions. A comparison of three staining pro-cedures. Int. J. Lcpr. 48 (1980) 402-413.

6. CtIRRAN, R. C. and GREGORY, J. Effects of fixation

and processing on immunohistochemical dem-onstration of immunoglobulin in paraffin sectionsof tonsil and bone marrow. J. Clin. Pathol. 33

(1980) 1047-1057.7. Gncarr, R. G. A stain for fungi in tissue sections

and smears. Am. J. Clin. Pathol. 25 (1975) 975-979.

8. HARDOE, M., MSIIANA, R. N., CLoss, 0.,

KnosvALI., G. and ALENSEN, N. H. Cross reac-

tions between mycobacteria. II. Crossed immu-noelectrophonetic analysis of soluble antigens ofBCG and comparison with other mycobacteria.

Scand. J. Immunol. 9 (1979) 115-124.

9. MASON, D. Y., BELL, J. I., CIIRISTENSSON, B. and

BlIWRHELD, P. An immunohistochemical studyof human lymphoma. Clin. Exp. Immunol. 40

(1980) 235-248.10. MsnANA, R. N., BELE1-11.1, A., STONER. G. L.,

HARitoE, NI. and HAREGLwois, A. Demonstra-

218^ International Journal of Leprosy^ 1983

Don of mycobacterial antigen in leprosy tissues.

Int. J. Lepr. 50 (1982) 1-10.

I I. RIDLEY, D. S. and JOMING, W. H. A classification

of leprosy for research purposes. Lcpr. Rev. 33

(1962) 119-128.

12. RIDLEY, D. S. and RIDLEY, Fixation of skin

biopsies. Lepr. Rev. 46 (1975) 309-310.

13. R1D1IN, I). S. Skin biopsy in leprosy. Histological

interpretation and clinical application. Documen-

ta Geigy, Basel, 1977.

14. RIDII:1", D. S. The SFG (solid, fragmented and

granular) index for bacterial morphology. Lepr.

Rev. 42 (1971) 96-97.

15. RIDLLy, M. J., RtIssELL, D. and RIDLLY, D. S. An

immunoperoxidase study of immunological fac-

tors in skin lesions across the spectrum aleprosy.

Int. J. Lcpr. 50 (1982) 11-17.16. RIDLEY, M. J. and Russul., D. An immunoper-

oxidase study in high immune and low resistancegranulomas in leprosy. J. I'athol. 137 (1982) 149-

157.17. RILEY, NI. J. and RIDLEY, D. S. Unique expres-

sion of I l LA-DR (la-like) antigen in the lesions of

polar tuberculoid leprosy. Lepr. Rev. 53 (1982)249-252.

18. RioLLY, NI. .1. and RIDI.11v, D. S. The immuno-

pathology of erythema nodosum leprosum (ENL).The role of extravascular complexes. Lepr. Rev.

54 (1983) 95-107.19. SrERNBERcil.R, L. A., HARDY, P. 11., .In., CuctJus,

J. J. and MEYER, H. G. The unlabelled antibody

method Of im munohistochem istry: preparation and

properties of soluble antigen-antibody complex

(horseradish peroxidase-antihorseradish peroxi-dase) and its use in identification of spirochetes.

J. Histochem. Cytochcm. 18 (1970) 315-333.