- Volume 72 , Number 3
- Page: 327–30
False positive reaction of the immunohistochemistry technique using anti-BCG polyclonal antibodies to identify Mycobacterium leprae in wild nine-banded armadillos
ABSTRACT
The authors studied 66 wild nine-banded armadillos f rom Brazil. The ear samples were collected and Ziehl-Neelsen or Fite-Faraco stains were performed, as well as immunostaining using polyclonal BCG antibody, to avaluate the presence of the Mycobacterium leprae. The AFB were not detected by the Ziehl-Neelsen or Fite-Faraco staining, neither immunoexpression of the BCG marker. However, many normal structures f rom the ears of the nine-banded armadillos, such as condrocytes, condroblasts, fibroblasts and endothelial cells, and Gram positive bacteria cocci, showed false positive reaction by the BCG marker. The authors discuss the use of the immunohistochemical studies with the polyclonal BCG antibody to identify M. leprae antigens in wild armadillos.RÉSUMÉ
Les auteurs ont étudié 66 tatous à neuf bandes sauvages du Brésil. Les échantillons d'oreille furent prélevés et des colorations de Ziehl-Neelsen et de Fite-Faraco, ainsi qu'un immunomarquage utilisant un sérum polyclonal dirigé contre le bacille de Calmette et Guérin (BCG), afin d'évaluer la présence de Mycobacterium leprae. Des bacilles acido-résistants (AAR) ne furent pas détectés par le Ziehl-Neelsen ou le Fite-Faraco et une immuno-expression du marqueur BCG ne fut pas observée non plus. Cependant, de nombreuses structures normales des oreilles de tatous à neuf bandes, telles que des chondrocytes, des chondroblastes, des fibroblastes et des cellules endothéliales, ainsi que des coques bactériennes positives à la coloration de Gram, ont montré une réaction faussement positive au marqueur BCG. Les auteurs discutent de l'utilité des études immunohistochimiques avec un anticorps polyclonal dirigé contre le BCG pour identifier des antigènes de M. leprae chez les tatous sauvages.RESUMEN
Se estudiaron 66 armadillos silvestres de 9 bandas de Brasil. Se colectaron muestras de las orejas, se seccionaron y se tiñeron por Ziehl-Neelsen o Fite-Faraco, y se evaluó la presencia de Mycobacterium leprae utilizando un anticuerpo policlonal anti-BCG. No se detectaron bacilos ácido-resistentes por las tinciones de Ziehl-Nelseen/Fite-Faraco y tampoco se encontraron bacilos con el suero anti-BCG. Sin embargo, se observó que muchas estructuras nor-males de las orejas de los armadillos, tales como condrocitos, condroblastos, fibroblastos y células endoteliales, y bacterias Gram negativas, dieron reacciones falsas-negativas con el reactivo anti-BCG. Los autores discuten el uso de los estudios inmunohistoquímicos con el suero policlonal anti-BCG para identificar antígenos de M. leprae en los armadillos silvestres.To the Editor:
Current identification of Mycobacterium leprae is done by Fite-Faraco or Ziehl-Neelsen stains, but low numbers of bacillus in the tissue could make identification difficult. Some authors have reported that immunohistochemistry techniques demonstrate antigens of M. leprae instead of the whole bacillus, increasing the sensibility of the histopathology diagnose mainly in paucibacillary forms (1, 5, 7). Takahashi, et al. (7) identified more then 20% of the AFB from of the skin biopsy of the indeterminate leprosy (paucibacillary form) using immunostaining with anti-BCG (bacilli of Calmette-Guérin) antibodies. Using the same technique, Barbosa Jr., et al. (1) found acid-fast bacilli (AFB) in 50% of the skin biopsies where they did not find AFB before by Fite-Faraco stainning. The antigens were detected in several sites of the dermis, mainly in the perivascular region. Kutzner, et al. (4) described the anti-BCG immunostaining technique as a promising screening tool for the detection of common infectious microorganisms in humans. The anti-BCG antibodies do not react with the normal structures from human skin nor with inflammatory cells, and produce little background, although Schettini, et al. (6) observed that melanophages and mast cells could be stained. Some authors have demonstrated cross reaction between polyclonal antibodies and a large spectrum of fungi and bacteria (2, 3, 4, 9). Bonenberger, et al. (2) established that some microorganisms such as Mycobacteria and Nocardia were strongly stained by immunostaining with anti-BCG; Criptococcus, Pythium and Prototheca were partially stained; spirochetes and protozoa (Leishmania) were not stained.
We analyzed 66 wild nine-banded armadillos, Dasypus novemcinctus species, from the State of Espírito Santo, localized in Southern region of Brazil. The purpose of this research was to evaluate if immunostaining with a polyclonal anti-BCG antibody was an appropriate diagnostic tool for M. leprae detection in these animals. After the armadillos had been captured, the ears, margins were collected and fixed in 10% neutral buffered formalin.
The paraffin-embedded samples were sectioned and slides were stained with Ziehl-Neelsen or Fite-Faraco stains and immunohistochemistry. We used the immunostaining with the anti-BCG on human positive-control biopsies from multibacillary patients, to compare with the samples from the armadillos. After blocking of endogenous peroxidase, the slides were incubated with the poly-clonal rabbit antibody (anti-Mycobacterium bovis, DAKO A/S, Denmark), Dilution of 1:100,000 was done in PBS buffered pH 7.4 with bovine serum albumin (BSA). After this, the slides were incubated with the secondary antibody, biotinylated goat antimouse/rabbit Ig (StrepABComplex/HRP Duet mouse/rabbit, DAKO A/S, Denmark), diluted 1:200 in PBS and with the strepavidinBiotin-peroxidase complex 1:200 in PBS. The antigens were visualized after incubation with the diaminobenzidine tetrahydrochloride 60 mg% (SIGMA, EUA), DMSO 1 ml, 6% H2O2 20 volume and PBS 100 ml for five minutes at room temperature. Negative controls were prepared by omission of the primary antibody and substitution with BSA. The readings were carried out on coded, "blinded" slides.
The histopathological examination of the ear biopsies showed an inflammatory infiltrate of the lympho-histiocytes and neutrophils, but no AFB was found by Ziehl-Neelsen or Fite Faraco staining, or by immunohistochemical study. However, nuclei of many normal cells, such as condrocytes, condroblasts, fibroblasts, endothelial cells (Fig. 1), and Gram positive bacteria cocci produced immunostaining with the anti-BCG marker (Fig. 2). In all the material, the primary antibody was replaced by BSA to demonstrate that there are no reactions due to the secondary antibody or slide background. The human controls showed macrophages immunostained with the anti-BCG marker (Fig. 3).
Fig. 1. Immunohistochemistry findings in ear fragment. Nucleous of the condrocytes, condroblasts, fibroblasts, endothelial cells presenting immunoexpression to the anti-BCG marker. (Objective 40 x).
Fig. 2. Immunohistochemistry findings in ear fragment. Inflammatory infiltration of the lymphohistiocytes and neutrophils, and Gram positive bacteria as coccus presenting immunoexpression to the anti-BCG marker (Objective 20 x).
Fig. 3. Immunohistochemistry findings in human skin fragment from leprosy multibacillary patient, the human positive control (Objective 40 x).
The immunohistochemical technique using polyclonal anti-BCG antibody for detection of M. leprae antigens in ears from wild armadillos, or other armadillo tissues, has not been reported before. We found several structures presenting immunoexpression to the anti-BCG marker in the epidermis and the dermis of the ear biopsies, without the morphologic features presented by the M. leprae immunostained antigens previously described. Probably in wild nine-banded armadillo, microorganisms other than Mycobacteria could be responsible for false positive results in these tissues. These findings agree with previous reports (2, 4). Another interesting feature observed is that the polyclonal anti-BCG antibody stains Gram positive bacteria. Some animals showed very similar structures to the morphology of coccus. This similarity can create doubt about the use of the immunohistochemical methods using the polyclonal anti-BCG alone to detect the M. leprae infection in wild armadillos. We do not know if this technique would be useful to detect mycobacterial antigens in tissue other than ears, such as skin, lymphnodes, or liver. In conclusion, the immunohistochemical technique, using anti-BCG antibodies, is a very useful tool for the diagnosis of human paucibacillary or indeterminate forms of leprosy, but in wild nine-banded armadillos immunostaining of normal structures of the ear and cross reaction with different infectious agents could difficult its utilization for Mycobacterial antigen detection difficult.
Acknowledgment. The authors would like to thank Dr. Carlos Faria from Santa Casa de Misericordia Medical School, Vitória, Brazil; Dr. Fausto Edmundo Lima Pereira from Federal University of Espírito Santo, Vitória, Brazil for precious critical suggestions; Dr. Charles K. Job from the Department of Pathology, St. Thomas Hospital and Leprosy Center, Chettupattu, India for the histopathologic review, and Ms. Suely Nonogaki from Adolpho Lutz Institute, São Paulo, Brazil for technical assistance.
-Patrícia D. Deps, M.D., Ph.D.
Department of Social Medicine, Infectious Disease Unit,
Federal University of Espirito Santo, Vitória, Brazil.
-Nilceo S. Michalany, M.D.
Department of Pathology, Federal University of São Paulo,
São Paulo, Brazil.
-Jane Tomimori-Yamashita, M.D., Ph.D.
Department of Dermatology-Federal University of São Paulo,
São Paulo, Brazil.
REFERENCES
1. BARBOSA, JR. A. A., SILVA, T. C., PATEL, B. N., SANTOS, M. I. R., WAKAMATSU, A., and FERREIRA, V. A. Demonstration of mycobacterial antigens in skin biopsies from suspected leprosy cases in the absence of bacilli. Path. Res. Pract. 190 (1994) 782-785.
2. BONENBERGER, T. E., IHRKE, P. J., NAYDAN, D. K., and AFFOLTER, V. K. Rapid identification of tissue micro-organisms in skin biopsy specimens from domestic animals using polyclonal BCG antibody. Vet. Dermatol. 12 (2001) 41-47.
3. BYRD, J., MEHREGAN, D. R., and MEHREGAN, D. A. Utility of anti-bacillus Calmette-Guérin antibodies as a screen for organisms in sporotrichoid infections. J. Am. Acad. Dermatol. 44 (2001) 261-264.
4. KUTZNER, H., ARGENYI, Z. B., REQUERA, L., RUTTEN, A., and HUGEL, H. A new application of BCG antibody for rapid screening of various tissue microorganims. J. Am. Acad. Dermatol. 38 (1998) 56-60.
5. MSHANA, R. N., BEHELU, A., STONER, G. L., HARBOE, M., and HAREGEWOIN, A. Demonstration of mycobacterial antigens in leprosy tissues. Int. J. Lepr. 50 (1982) 1-10.
6. SCHETTINI, A. P. M., FERREIRA, L. C. L., MILAGROS, R., SCHETTINI, M. C. A., PENNINI, S. N., and REBELLO, P. B. Enhancement in the histological diagnosis of leprosy in patients with only sensory loss by demonstration of mycobacterial antigens using anti-BCG polyclonal antibodies. Int. J. Lepr. Other Mycobact. Dis. 69 (2001) 335-340.
7. TAKAHASHI D. I., ANDRADE, JR. H. F., WAKAMATSU, A., MANINI, M., and DE BRITO, T. Treated-cured indeterminate leprosy: a search for predictive histopathological and immunohistochemical parameters in skin biopsies taken from patients at admission and at clinical discharge. Acta Leprol. (Gen-eve) 8 (1992) 95-102.
8. TAKAHASHI, D. I., ANDRADE, JR. H. F., WAKAMATSU, A., SIQUEIRA, S., and DE BRITO, T. Indeterminate leprosy: histopathologic and immunohistochemical predictive parameters involved in its possible change to paucibacillary or multibacillary leprosy. Int. J. Lepr. Other Mycobact. Dis. 59 (1991) 12-19.
9. WILEY, E. L., BECK, B., and FREEMAN, R. G. Reactivity of fungal organims in tissue sections using anti-mycobacterial antibodies. J. Cutan. Pathol. 18 (1991) 204-209.
Reprint requests to: Address:
Patrícia D. Deps, M.D., Ph.D.
Núcleo de Doenças Infecciosas, Centro Biomédico
Av. Marechal Campos 1468
Vitória, ES, Brasil
Email: pdeps@ndi.ufes.br / pdeps@uol.com.br
Received for publication on 12 February 2004.
Accepted for publication on 26 April 2004.