American Journal of Pathology, Vol. 127, No. 2, May 1987Copyright © American Association of Pathologists

Histopathologic Changes Induced by Vaccination inExperimental Cutaneous Leishmaniasis ofBALB/c Mice

MANOEL BARRAL-NETTO,LUIZ ANTONIO R. de FREITAS, and

ZILTON A. ANDRADE

Highly susceptible BALB/c mice became partially re-sistant to Leishmania mexicana amazonensis infectionafter intravenous immunization with solubilized ho-mologous promastigote antigen. Immunized BALB/cmice exhibited mixed mononuclear cell reactions,with granulomatous inflammation, collagen deposi-tion, and fibrinoid necrosis at the site of infection. Incontrast, naive animals displayed a monomorphic pic-ture composed of largely vacuolated and parasitized

SEVERAL ATTEMPTS have been made to correlatehistopathologic changes with prognosis in human cu-taneous and mucocutaneous leishmaniasis.'-3 Thefindings observed during the experimental infectionofsusceptible and resistant inbred strains ofmice canalso indicate which are the main histologic changesassociated with prognosis.

Despite the clearcut differences in the course oftheinfections with agents ofcutaneous leishmaniasis ob-served in these strains ofmice, it is difficult to discrim-inate factors really involved in resistance from thosedue to diverse genetic backgrounds, but not impli-cated in specific immunity.

Another way to address this question is to use ani-mals of the same inbred strain, with identical geneticbackgrounds, but differing in susceptibility to theparasite. Such is possible by immunizing susceptibleanimals. In the present study we took advantage ofanimmunization protocol which rendered BALB/cmice partially resistant to Leishmania mexicana am-azonensis5 and compared the evolution ofthe lesionsto their susceptible inbred counterparts.

Materials and MethodsAnimals

Inbred BALB/c mice were obtained from theCPqGM colony (Salvador-Bahia) and were used at

From Gon(alo Moniz Research Centre (FIOCRUZ) andFederal University ofBahia, Bahia, Brazil

macrophages with areas of coagulative necrosis. Elec-tron microscopy revealed an increased number of eo-sinophils, sometimes in close contact with parasitizedmacrophages, in immunized animals. These findingsillustrate that histologic changes reflect host immunestatus in cutaneous leishmaniasis, and that susceptibil-ity of BALB/c mice to L m amazonensis, although de-pendent on genetic background, can be artificiallymodified. (AmJ Pathol 1987, 127:271-278)

10-14 weeks of age. Animals were maintained onpellet mouse ration and water ad libitum.

Parasites

The "Josefa" strain (MHOM/Br/76/Josefa) ofLmamazonensis (as characterized by isoenzyme pat-terns, a panel of monoclonal antibodies, as well askDNA restriction analysis) was used for both infec-tion and antigen preparation.

Antigen Preparation

Stationary-phase promastigotes were obtained onliver infusion tryptose medium supplemented with5% fetal calf serum (LIT-FCS). After washing threetimes in phosphate-buffered saline with 2% glucose(PBS-G), parasites were resuspended in "lysingbuffer" (50 mM Tris, pH 7.7; 0.12 N NaCl; 0.5%NP-40; 0.25% sodium deoxyxholate; 5TIU/ml apro-tinin; 5 mM EDTA) at a concentration of 109 pro-

Supported by the Brazilian National Research Council(CNPq) Grant 40-0634/85.Accepted for publication December 16, 1986.Address reprint requests to Dr. Manoel Barral-Netto,

Hospital Professor Edgard Santos, Rua Joao das Botas s/nCanela, 40000 Salvador, Bahia, Brazil.

271

272 BARRAL-NETTO ET AL

mastigotes/ml. The material was incubated for 10minutes at room temperature (under occasional mix-ing) and then submitted to dialysis against 500 vol-umes of PBS at 4 C. Concentration of the material isexpressed as "parasite equivalent per milliliter" (pe/ml), as a function of initial number of promastigotesand final volume of the preparation.

Immunization and Infection

Mice were given intravenous injections in 3 weeklydoses of solubilized antigen at 5 X 107 pe/ml in vol-umes of0.2 ml (unimmunized mice received an equalvolume ofPBS at the same dates). One week after thelast immunizing dose, mice were challenged subcuta-neously in the left hind footpad with 5 X 106 viablepromastigotes in 0.025 ml. Footpad thickness was

measured periodically with a dial gauge micrometercaliper (C. Starret, Athol, Mass). Differences betweenthe infected footpad and the contralateral uninfectedone were expressed as "lesion size" in millimeters.

Assay for Anti-L m amazonensis Antibody Titers

An enzyme-linked immunosorbent assay (ELISA)was performed with homologous promastigote anti-gen. Microtiter plates were sensitized with 1 ,ug ofprotein per well in a carbonate-bicarbonate buffer,pH 9.6. After washing three times with PBS contain-ing 0.05% Tween 20 (PBS-Tween) serial dilution ofsera were incubated for 1 hour at 37 C, followed bythree washings with PBS-Tween. Peroxidase-conju-gated rabbit anti-mouse IgG (Sigma, St. Louis, Mo)diluted 1: 1000 in PBS + 10% newborn bovine serumwas incubated for 1 hour at 37 C. After further wash-ing the substrate solution (0.04% ortho-phenilenedia-mine, 0.012% hydrogen peroxide in pH 5.0 citratephosphate buffer) was added. Reaction was stoppedby adding 8N sulphuric acid (0.025 ml/well) after a30-minute incubation at room temperature in thedark. Plates were read in a Titertek Multiskan spec-trophotometer (Flow Laboratories, Ayreshire, Scot-land, UK) at 495 nm.

Assay for Delayed Type Hypersensitivity (DTH)

L m amazonensis promastigotes cultured on LIT-FCS were used as the antigen source, and the footpadswelling test performed as described elsewhere.6 Inbrief, mice received 50 ,ug protein os leishmanial anti-gen (0.025 ml into the ventral aspect ofthe uninfectedhind footpad). Measurements were performed with a

dial gauge caliper as indicated above. Results are pre-sented as the thickness (mm X 10-2) measured 24

AJP * May 1987

hours after antigen injection minus the thickness be-fore injection (A DTH).

Histopathology

At 2, 10, and 13 weeks after infection the lesionswere removed and placed in Bouin's fluid for about 1hour. After hardening they were recut and trimmedand left in the fixative for another 12 hours. Thebone-containing lesions were decalcified in EDTA.Tissues were embedded in paraffin, and the sectionswere stained with hematoxylin and eosin (H&E).

Electron Microscopy

Small fragments from the footpad lesions were im-mediately fixed in ice-cold 2% glutaraldehyde in0.1 M cacodylate buffer for 1 hour, postfixed in OSO4,washed in the buffer, dehydrated in graded ethanol,and embedded in Spurr's resin. Semithin sectionswere cut on a Reichert-Jung Ultracut E microtomewith glass knives. They were mounted on glass slidesand stained with methylene blue-azur II. Ultrathinsections were cut with a diamond knife, picked up oncopper grids, and contrasted with uranyl acetate andlead citrate. Observations were carried out on a ZeissEM- 109 electron microscope at 50 kv.

Statistical Analysis

Group means were compared by the Student t test.For comparison of antibody titers log transformeddata were used.

Results

Growth of Primary Lesion

Groups of 10 BALB/c mice were immunized intra-venously with three doses ofsolubilized promastigoteantigen, at weekly intervals, or maintained withoutimmunization. After challenge infection with 5 X 106viable promastigotes, immunized mice exhibitedmuch smaller lesions than unimmunized animals(Figure 1A). Unimmunized animals had progressivegrowth of lesions, reaching huge volumes, associatedwith distention of the covering skin and exhibitinglarge necrotic areas. On the other hand, lesions inimmunized mice were small and ulcerative and neverappeared as large masses.

DTH Responses

Unimmunized BALB/c mice developed a transientlow level DTH response at 7 weeks after infection,

VACCINATION IN CUTANEOUS LEISHMANIASIS 273

C

O 6 12

W E E K S P O S T - I N F E C T I 0 NFigure 1-Demonstrative of lesion size (a), DTH reaction (b), and antibody titers (c) in immunized and unimmunized BALB/c mice infected with L mamazonensis. , unimmunized; ----immunized.

followed by a decreased response up to the end oftheobservation period. In contrast, immunized mice ex-

hibited stronger DTH reactivity, which was persis-tently positive (Figure 1B). Differences were statisti-cally significant at 10 weeks (P < 0.05) and 13 weeks(P < 0.02) after infection.

Antibody Production

Immunization by itself induced high IgG anti-leishmanial antibody titers, explaining the big differ-ence between immunized and naive animals beforethe infection. After 6 weeks of infection, the differ-ence in antibody titer was no longer significant; andtiters were virtually identical 12 weeks after infection(Figure 1C).

Histopathology

The lesions observed in the footpad of immunizedand naive mice 2 weeks after inoculation looked alike.They were represented by a moderate and diffuse in-filtration of lymphocytes and a few plasmocytes andpolymorphonuclear eosinophils, which were locatedat the dermis, but extended for some distance into theskeletal muscles. Among these cells there were several

small collections of parasitized macrophages. Theselatter cells exhibited a large, clear, and vacuolatedcytoplasm containing numerous Leishmania orga-nisms. However, some differences could be noted be-tween the two groups of animals when details of themicroscopic picture were comparatively analysed. Inthe immunized mice the clusters of parasitized mac-rophages were less frequently seen and containedfewer parasites as compared with naive mice. Also, inthe immunized mice the diffuse mononuclear infil-tration was in some cases predominantly lympho-plasmocytic, rather than lymphocytic, and a mild de-gree of fibroblastic proliferation and collagendeposition could be observed in focal areas, as well asa few microscopic foci of fibrinoid necrosis. On theother hand, small patchy areas ofcoagulative necrosiswere detected in all nonimmunized mice and in onlyone immunized mouse.A clear-cut difference appeared in the lesions ob-

served 10 and 13 weeks after inoculation. At this timethe changes in the naive mice were represented by anextensive and monomorphic collection ofvacuolatedand parasitized macrophages that resembled fatty tis-sue when seen with low-power microscopy (Figure2A). This monotonous picture was only altered by thefrequent appearance of areas of coagulative necrosis.

A B4

10

6

E

---

bJN

z0

Lui-J

N0

..0

-

IX

n

w

00do

_ le.I-.z4c

0

H10

Vol. 127 * No. 2

274. BARRAL-NETTO ET AL AJP * May 1987

q. f 4I;I_{'m'9' <.t

4E~~~~~~~~~41 'H'Ss *e'v5 't21 -vS't,

41~~~~~~~~~*

S S ; ' ;4 ' v ''t44 i'' > t

IL ;*4 ' ,/ 2-'.

Figure 2a-Extensive collection of vacuolated and parasitized macrophages with focal areas of purulent necrosis, a typical picture of the established lesioncaused byLm amazonensis in the naive BALB/c mouse. (H&E, X 1 00) b-The most frequent histologic picture seen in BALB/c mice immunized against andchallenged with L m amazonensis, a mixed cell reaction with diffuse fibrosis, giant cell formation, and scattered parasitized macrophages. (H&E, X200)

VACCINATION IN CUTANEOUS LEISHMANIASIS 275

These focal areas, which appeared eosinophilic andpoorly stained, were frequently invaded by numerouspolymorphonuclear leukocytes presenting variabledegrees of disintegration ("purulent necrosis"). Clus-ters ofparasitized cells were seen dissociating the skel-etal muscle fibers and sometimes invading the bonemarrow more deeply situated over the footpad. At theperiphery of the lesion a cuff of small lymphocytescould be noted, but there was no definite fibrous cap-sule. Fibrinoid necrosis was never observed in theleishmanial lesions of these naive mice.

In the vaccinated animals the lesions were repre-sented by a mixed-cell inflammatory reaction, withonly a few scattered parasitized macrophages. Thesmall lymphocyte was the predominating cell, butmany nonparasitized macrophages, plasmocytes, andeosinophils were also present. The infiltration tendedto assume a focal pattern, the collection ofinflamma-tory cells sometimes appearing separated by densefibrous septa. Frequently a granulomatous reactionwas noted, with accumulation of undifferentiatedmacrophages, epitheliod cells, and multinucleatedgiant cells (Figure 2B). The number of eosinophilswas variable from case to case. Usually the eosino-phils appeared in moderate numbers and with a dif-fuse distribution. Around and within the areas of fi-brinoid necrosis they could be seen focallyaccumulated. Focal fibrinoid necrosis was an out-standing feature. Besides the eosinophilic fibrinoidfilaments, disintegrating parasites and a few cellscould also be noted within the necrotic areas. Aroundthese areas there appeared some fibroblasts, whichseemed to be involved in an organizing process inlater lesions. Fibroblast proliferation and collagenformation appeared as an active process throughoutthe lesion, giving rise to a more or less dense stromafor the inflammatory cells, as well as forming fibroussepta and an external capsule. In one instance thefootpad lesion was represented by a fibrous nodule, inthe interior ofwhich there were isolated small collec-tions ofmononuclear inflammatory cells, including afew parasitized macrophages. No involvement ofneighboring skeletal muscles or bone was seen in vac-cinated mice.

Electron Microscopy

The material from early lesions (2 weeks) showedmore parasitized macrophages and eosinophils, aswell as some necrotic macrophages, in the immunizedthan in unimmunized animals (Figure 3A).

Later lesions (13 weeks) disclosed large macro-phages containing several well-preserved Leishmaniaorganisms within parasitophorous vacuoles. Rarely,

Leishmania could be observed within the macro-phages in the absence of a defined parasitophorousvacuole. A few polymorphonuclear eosinophils ap-peared among the macrophages, sometimes withclose contact oftheir respective external membranes.

Lesions in the immunized animals showed manynonparasitized macrophages with many inderdigitat-ing cytoplasmic processes, a few primary lysosomes,residual bodies, and well-developed endoplasmic re-ticulum. Fibroblasts and collagen fibers were fre-quently observed. Some macrophages were necroticand in their vicinity Leishmania appeared degener-ated in the interstitial tissue or phagocytosed by poli-morphonuclear, especially eosinophils (Figure 3Band C).

Discussion

Intravenous immunization of BALB/c mice withsolubilized promastigotes altered immunoregulatorymechanisms in these animals, rendering them par-tially resistant to leishmanial infection.S7 Althoughthe precise mechanism of action is not known, thereare some likely explanations for the phenomenon.First, the effect does not seem to depend solely on areduction of the initial parasite burden. It has beenshown that BALB/c mice develop the characteristicuncontrolled course ofthe disease even when infectedwith only 50 promastigotes of Leishmania / major.7In the present case, both the gross and the histologicpictures were drastically altered. The monotonoushistopathologic aspects of the lesion, composed ofvacuolated macrophages, which characterized theleishmanial infection ofBALB/c mice,4 changed to apicture that is generally described for resistantstrains.4'8 It is interesting to note that some featurestaken as representative for a good prognosis inhuman3'8"' and experimental cutaneous leishman-iasis,4'9 such as fibrosis, fibrinoid necrosis, and granu-lomatous inflammation, were outstanding findings inthe immunized animals.The present findings stand as a good example of

how histopathologic features may reflect variations inthe host immune status. While the lesions in the naiveand susceptible BALB/c mice tended to present a ho-mogeneous and regular appearance, as described pre-viously both in human cutaneous diffuse leishman-iasis'0"'1 and in susceptible animals,4"12 in theimmunized animals the lesions were characterized bytheir variability. Such variability from one animal toanother probably represented different degrees ofim-munity brought about by vaccination. Confirmingprevious reports,24 6 the histologic features probablyrelated to resistance were the presence of a mixed

Vol. 127 * No. 2

276 BARRAL-NETrO ET AL

"2' S

&

.i.

Figure 3a-Electron micrograph showing a vaccuolated macrophage and eosinophils in its proximity, as frequently seen in immunized animals. (X3000) band c-Eosinophils phagocytosing Leishmania organisms which are exhibiting variable degrees of degenerative changes. (X7000)

AJP * May 1987

.4p. .":. p *

VIV P.-4

-W, A&

Vol. 127 * No. 2 VACCINATION IN CUTANEOUS LEISHMANIASIS 277

mononuclear cell reaction, fibrinoid necrosis, granu-lomatous inflammation, and collagen deposition.There also were signs that the majority ofLeishmaniaorganisms were killed outside the macrophages, ininterstitial tissue, in areas of necrosis, and phagocy-tosed by polymorphonuclear cells, eosinophils, andneutrophils. In fact, eosinophils were frequently ob-served at the ultrastructural level in the lesions ofvaccinated mice. They not only appeared phagocytos-ing degenerating Leishmania organisms, but were inintimate contact with the external membrane of theparasitized macrophages. Whether some importantcooperation between these cells develops after immu-nization is a question for future research.

Secondly, it is possible that complement activationin animals with high antibody titers is responsible forattracting larger numbers ofpolymorphonuclear leu-kocytes in the inflammatory response observed in im-munized mice. The lack of a protective effect of serafrom immunized animals injected into naive recipi-ents'3 militates against this hypothesis. If polymor-phonuclear cells are important in controlling parasitegrowth, 14 their recruitment is probably due to cellularmechanisms, through lymphokine production, forexample. The latter could also account, probably, fora more efficient Leishmania-clearing mechanism-tissue necrosis.2" This refers us to the third point, thatof the T-cell function. Concerning the role of the Tcells, two different explanations exist related to therather opposing views ofthe immunopathogenesis ofleishmaniasis in BALB/c mice. If one considers thedisease in such a strain of mice as a consequence ofgeneration of a potent suppressor T cell,'5"16 the im-munization may have circumvented its generation,or, alternatively, may have stimulated a contrasup-pressive mechanism. In keeping with the view ofexac-erbated T-helper function as a cause of the uncon-trolled disease in BALB/c mice, 7'19 the protectiveeffect would follow the generation ofa suppressor cell.Ifthis is the case, such a suppressor cell may be able todecrease, but not abrogate, the T-helper activity, be-cause immunized mice remain with positive DTHresponses and produce high levels ofanti-leishmanialantibodies. It is noteworthy that the above explana-tions for the disease in BALB/c mice were derivedfrom studies using animals infected with L major, anddifferent immunoregulatory mechanisms may be op-erative with Leishmania species from the NewWorld.20Our data show that immunization can partially

overcome the susceptibility of BALB/c mice to L mamazonensis, a putative model of human diffuse cu-taneous leishmaniasis. Although intravenous immu-nization is hardly to be recommended for use in

humans, it can open the possibility of immunopro-phylaxis in selected populations.

References

1. Bryceson ADM: Diffuse cutaneous leishmaniasis inEthiopia: I. The clinical and histological features ofthedisease. Trans R Soc Trop Med Hyg 1969,63:708-737

2. Ridley DS, Marsden PD, Cuba CC, Barreto A: A histo-logical classification of mucocutaneous leishmaniasisin Brazil and its clinical evaluation. Trans Soc TropMed Hyg 1980, 74:508-521

3. Destombes P:Application du concept de "systematiza-tion polaire" aux leishmanioses cutanees. Bull SocPathol Exot 1960, 53:299-301

4. Andrade ZA, Reed SG, Roters S, Sadigursky M: Im-munopathology of experimental cutaneous leishman-iasis. Am J Pathol 1984, 114:137-148

5. Barral-Netto M, Reed SG, Sadigursky M, SonnenfeldG: Specific immunization of mice against Leishmaniamexicana amazonensis using solubilized promasti-gotes. Clin Exp Immunol (In press)

6. Barral A, Petersen EA, Sacks DL, Neva FA: Late meta-static leishmaniasis in the mouse. A model for mucocu-taneous disease. Am J Trop Med Hyg 1983, 32:277-285

7. Howard JG, Nicklin S, Hale C, Liew FY: Prophylacticimmunization against experimental leishmaniasis: I.Protection induced in mice genetically vulnerable tofatal Leishmania tropica infection. J Immunol 1982,129:2206-2212

8. Ridley DS, Ridley MJ: The evolution of the lesion incutaneous leishmaniasis. J Pathol 1983, 141:83-96

9. Andrade ZA, Reed SG, Roters S, Sadigursky M: Pato-genia da leishmaniose cutanea experimental: A impor-tancia da necrose na eliminacao dos parasitos dasles6es. Rev Soc Bras Med Trop 1984, 17:187-197

10. Convit J, Pinardi E, Rondon AJ: Diffuse cutaneousleishmaniasis: A disease due to an immunological de-fect of the host. Trans R Soc Trop Med Hyg 1972,66:603-610

11. Bittencourt AL, Andrade ZA: Aspectos imunol6gicosna leishmaniose cutaneo-mucosa. Hospital (Rio de Ja-neiro) 1967, 71:89-98

12. Perez H, Labrador F, Torrealba JW: Variations in theresponse of five strains of mice to Leishmania mexi-cana. Int J Parasitol 1979, 9:27-32

13. Howard JG, Liew FY, Hale C, Nicklin S: Prophylacticimmunization against experimental leishmaniasis: II.Further characterization of the protective immunityagainst fatal Leishmania tropica infection induced byirradiated promastigotes. J Immunol 1984, 132:450-455

14. Grimaldi G Jr, Soares MJ, Moriearty PL: Tissue eosin-ophilia and Leishmania mexicana mexicana eosino-phil interactions in murine cutaneous leishmaniasis.Parasite Immunol 1984, 6:397-408

15. Howard JG, Hale C, Liew FY: Immunological regula-tion of experimental cutaneous leishmaniasis: III. Thenature and significance of specific suppression of cell-mediated immunity. J Exp Med 1980, 152:594-607

16. Howard JG, Hale C, Liew FY: Immunological regula-tion ofexperimental cutaneous leishmaniasis: IV. Pro-phylactic effect of sub-lethal irradiation as a result ofabrogation of suppressor T cell generation in mice ge-netically susceptible to Leishmania tropica. J Exp Med1981, 153:557-568

17. Liew FY: Specific suppression of responses to Leish-mania tropica by a cloned T-cell line. Nature 1983,305:630-632

278 BARRAL-NETTO ET AL AJP * May 1987

18. Titus RG, Lima GC, Engers HD, Louis JA: Exacerba-tion of murine cutaneous leishmaniasis by adoptivetransfer of parasite-specific helper T cell populationscapable of mediating Leishmania major-specificdelayed-type hypersensitivity. J Immunol 1984,133:1594-1600

19. Titus RG, Ceredig R, Cerottini JC, Louis JA: Thera-peutic effect of anti-L3T4 monoclonal antibody GK1.5 on cutaneous leishmaniasis in genetically-suscepti-ble BALB/c mice. J Immunol 1985, 135:2108-2114

20. Alexander J, Kaye PM: Immunoregulatory pathways

in murine leishmaniasis: Different regulatory controlduring Leishmania mexicana mexicana and Leish-mania major infection. Clin Exp Immunol 1985,61:674-682

Acknowledgments

We are indebted to Ms. Silva Cardoso for excellent tech-nical assistance and to Mrs. Rosimari Keysselt for typingthe manuscript.