Reversal reaction (RR, type-1 reaction) can occur in the borderline (BT-BL) area of the spectrum of leprosy to alter the chronic, uneventful course of the disease. It consists of acute episodes, which may involve skin and/or nerves and often amount to medical emergencies demanding immediate treatment to prevent progression of nerve damage to the irreversible stage. The underlying immunologic alteration is a localized Th1 cell-mediated immune reaction against mycobacterial antigens. In a previous work,1 we have shown that vascular endothelial growth factor (VEGF) is a major player throughout the natural history of RR. In this study we have extended these data on RR to study tissue expression of cyclooxygenase (COX)2, an inducible enzyme synthesizing eicosanoids in inflammation. In experimental models, VEGF readily stimulates expression of COX2 and prostaglandin (PG)E synthethase in response to cell activation.2 Our aim was to show the role of COX2 in RR in view of the possible adoption of selective COX2 inhibitors to reduce tissue edema and prevent nerve damage progression.
Skin biopsies from seven patients with RR (six BT and one BL) were retrieved from the files of the Italian Reference Center for Leprosy of Genoa. Additional leprosy biopsies without histologic signs of RR were retrieved from seven patients with BT, three with BL, and four with LL. The above 21 biopsies were subsequent entries in our Registry. The specimens had been fixed in 10% formaldehyde for 24 hours. Sections were stained with hematoxylin and eosin for general microscopic examination and a modified Fite-Faraco procedure for M. leprae. The Ridley-Jopling leprosy classification scheme3 was applied to define the leprosy diagnosis across the spectrum of the disease. In particular, RR was recognized histologically when conspicuous edema with dispersion of the granuloma and foreign body type giant cells were present.
After routine deparaffinization and rehydration, immunochemistry was performed on 5-μm sections of paraffin-embedded tissue mounted onto slides coated with polylysine and subjected to microwave antigen retrieval in sodium citrate 10 mmol/L pH 6/0.05% Tween 20. Endogenous peroxidase was blocked with 3% H2O2 in phophate-buffered saline (PBS) 10 mmol/L followed by immersion in 10% goat serum. The antibody was anti-COX2 monoclonal (Cayman Chemical, Ann Arbor, MI) at 1:80 dilution (starting at 0.5 mg/mL) in PBS 10 mmol/L/5% goat serum. The antibody was layered onto the sections at 4°C overnight. Detection was performed with polyclonal goat antimouse antibody biotinylated and diluted 1:300 in PBS 10 mmol/L using the StreptABC/HRP kit and diaminobenzidine (Dako Cytomation, Glostrup, Denmark) as substrate solution. Sections were counterstained with Mayer’s hematoxylin. Sections without primary antibody were prepared as negative controls. Sections of lichen planus were used as positive controls for COX2 expression.
COX2 was consistently expressed in cells of the mono-nuclear–macrophage lineage, especially macrophages and epithelioid cells. The granuloma cells were positive across the spectrum of leprosy, from the vacuolated macrophages of the lepromatous area to the predominantly epithelioid cell infiltrate of the tuberculoid area. RR specimens showed positivity of cells of mononuclear–macrophage lineage in a fashion similar to that of specimens pertaining to the rest of the leprosy spectrum. The accompanying lymphocytes were negative. COX2 positivity in the cells of mononuclear–monocytic lineage of the leprosy granuloma has already been reported by Kiszewski and others,4 who claim that it is more pronounced in lepromatous cases.
In our experience, the characteristic finding in cases of RR was COX2 expression in the vessels and nerves of the dermis and subcutis. Microvessels, especially the tall-endothelium ones implied in vascular dilation and tissue edema, were distinctly positive in RR, in contrast to the vessels encountered in nonreactionary leprosy (Figure 1). In addition, nerve bundles and isolated nerve fibers were distinctly positive for COX2 (Figure 2). In cases of RR, our group previously described positivity for VEGF in the vessels and, at a lower degree, in the nerves.1 Vascular changes, leading to tissue edema, characterize RR at both early and conclamate stages of development. Edema, which is minimal in nonreactionary leprosy, is a major diagnostic marker of RR. It starts in the early phase and is most conspicuous in the conclamate phase of RR.5 With progression of RR, edema occurring at nerve fibers and bundles may lead to permanent nerve damage, the most important, long-term sequela of RR.
Experimental models support the view that VEGF and COX2 expression are related events. The VEGF enhancement of PG production through COX2 stimulation and PG synthase expression is associated with a dose-dependent increase in the expression of COX2 and a parallel membrane-associated PG synthase mRNA increase in cell culture.2 High PG levels induce vasodilation, a main feature of RR, pain, and inflammation. COX2 is an immediate-early gene product, which is undetectable in normal tissues, but can be readily induced in response to cell activation by cytokines, growth factors, and tumor promoters.6 Furthermore, treatment with the COX2 inhibitor NS-398 abolishes the VEGF-enhanced prostaglandin synthase expression. Finally, PG2 stimulates VEGF expression.7 These considerations suggest that selective COX2 inhibitors, which are currently used in several inflammatory conditions, could be considered for RR treatment, particularly at its early stage, to reduce acute symptoms and possibly prevent long-term nerve damage. Furthermore, they could be useful to prevent RR recurrence in unstable forms of the disease.
Address correspondence to Carlo Pesce, DISTBIMO, University of Genoa Medical School, Largo Benzi 10, Pad. IV, 16132 Genova, Italy. E-mail:
Note: Supplemental figures of non-reactional BT leprosy can be found online at www.ajtmh.org.
Authors’ addresses: Carlo Pesce, Myriam Grattarola, and Stefano Menini, DISTBIMO, University of Genoa Medical School, Largo Benzi 10, Pad. IV, 16132 Genova, Italy, E-mails:
Fiallo P, Clapasson A, Favre A, Pesce C, 2002. Overexpression of vascular endothelial growth factor and its endothelial cell receptor KDR in type 1 leprosy reaction. Am J Trop Med Hyg 66 :180–185.
Sakurai T, Tamura K, Kogo H, 2004. Vascular endothelial growth factor increases messenger RNAs encoding cycloxygenase-II and membrane associated prostaglandin E synthase in rat luteal cells. J Endocrinol 183 :527–533.
Kiszewski AEC, Becerril E, Baquera J, Ruiz-Maldonado R, Hernandez Pando R, 2003. Expression of cycloxygenase type 2 in lepromatous and tuberculoid leprosy lesions. Br J Dermatol 148 :795–798.
Ridley DS, 1985. Reactions. Ridley DS, ed. Skin Biopsy in Leprosy. Basel, Switzerland: Ciba-Geigy, 53–58.
Sakurai T, Tamura K, Kogo H, 2005. Stimulatory effects of eisosanoids on ovarian angiogenesis in early luteal phase in cycloxygenase-2 inhibitor-treated rats. Eur J Pharmacol 516 :158–164.