361 Am. J. Trop. Med. Hyg., 61(3), 1999, pp. 361–366 Copyright  1999 by The American Society of Tropical Medicine and Hygiene A NOVEL Fab-BASED ANTIVENOM FOR THE TREATMENT OF MASS BEE ATTACKS R. G. A. JONES, R. L. CORTELING, H. P. TO, G. BHOGAL, AND J. LANDON Department of Chemical Pathology, St. Bartholomew’s Hospital, London, United Kingdom Abstract. The frequency of mass bee attacks has dramatically increased in the Americas following the introduction and spread of the aggressive Africanized ‘killer’ bee (Apis mellifera scutellata). As yet no specific therapy is available, which led us to develop an ovine Fab-based antivenom as a potential new treatment. Sera from sheep immunized against the venom contained high levels of specific antibodies, as demonstrated by ELISA and by small-scale affinity chromatography, against both whole (A. m. mellifera) venom and purified melittin. A nerve muscle preparation was used to show the myotoxic effects of the venom and neutralization by the antivenom. Antivenom neutralizing ability was also demonstrated using assays for venom phospholipase A 2 and in vivo activities. Venom from both European and Africanized bees appeared identical when analyzed by acid-urea gel electrophoresis. This antivenom may therefore provide the first specific therapy for the treatment of mass envenomation by either European or Africanized ‘killer’ bees. The accidental introduction of the native African bee Apis mellifera scutellata into Brazil in 1957 and its subsequent displacement and hybridization of the long-established Eu- ropean bees A. m. mellifera and A. m. ligustica has resulted in the highly aggressive Africanized ‘killer’ bee. 1,2 It is better adapted to warmer climates than its European counterparts and, as a result, has been highly successful and spread rap- idly throughout Latin America. In 1992, the Africanized bee crossed the border into the United States and is now found in Texas, Arizona, New Mexico, and southern California. 3 It is predicted to eventually be distributed as far east as North Carolina. 2 These bees attack much more readily and in greater num- bers than the European varieties, 1 and thousands of stings may be inflicted on a single individual. 4,5 About 50 simul- taneous stings can cause systemic envenoming and about 500 are probably necessary to cause death by direct toxici- ty, 4,6–10 although one death in Arizona was due to only 125 stings. 11 By 1985, it was estimated that these bees had caused between 700 and 1,000 deaths, 12 and in Mexico there have been more than 190 such deaths between 1988 and 1993, with future estimates of 60 deaths per year. 13 Clinical fea- tures can include rhabdomyolysis, intravascular hemolysis, respiratory distress, hepatic dysfunction, hypertension, myo- cardial damage, shock, and/or renal failure. 4,5,8,11 Each Africanized bee injects an average of 94 g of ven- om in less than a minute, compared with 147 g for the slightly larger European bee, and circulating venom levels in patients of up to 3.8 mg/L have been recorded. 4,14,15 The venom contains many biologically active components such as melittin, phospholipase A 2 , apamin, mast cell degranulat- ing peptide, hyaluronidase, histamine, and dopamine. 16 Of these, melittin is predominant (50% by dry weight) and can disrupt cell membranes and produce pores even at low concentrations. 16,17 Phospholipase A 2 enhances this mem- brane disruption by forming powerful membrane-active products, and these relatively nonspecific membrane effects are thought to be responsible for many of the clinical find- ings. 18–20 No specific therapy is available currently and a safe, ef- fective treatment is urgently required. 4,21 We describe here the preclinical development and assessment of a specific A. mellifera antivenom that will hopefully prove suitable for treatment of mass bee attacks. MATERIALS AND METHODS Bee venoms. Freeze-dried A. m. mellifera venom and its purified venom components were obtained from Latoxan (Rosans, France) and Africanized bee venom was obtained from Southwest Venoms (Tucson, AZ). Antivenom. Antibodies were raised by immunizing nine, half-bred, Welsh ewes in groups of three with up to a max- imum of either 0.25 mg, 1 mg, or 4 mg of A. m. mellifera venom every 28 days. Venom was dissolved in phosphate- buffered saline (1.3 ml/sheep) and mixed extensively with Brij 35 (0.2 ml/sheep) and Freund’s adjuvant (3.25 ml/sheep) to form a stable water-in-oil emulsion. Freund’s complete adjuvant was used for the primary immunization, and Freund’s incomplete adjuvant was used for re-immunization at monthly intervals. The sheep were immunized intramus- cularly in each forelimb and hindlimb and subcutaneously on either side of the neck. Ten-milliliter blood samples were collected from each sheep two weeks after each immuniza- tion, and once adequate antibody levels had been achieved, the sheep were bled (10 ml/kg of body weight). The serum was separated and passed through a 0.2-m filter and pooled prior to storage at -20°C. The IgG fraction was partially purified by precipitation with sodium sulfate (final concen- tration = 18%) at 25°C for 15 min with mixing. The precip- itate, after pelleting by centrifugation, was washed twice with 18% sodium sulfate, reconstituted in saline (0.9% NaCl), and the IgG was cleaved completely with papain to yield Fab fragments. 22 After dialysis against saline, Fab pro- tein concentrations were determined by optical density mea- surements at 280 nm using an extinction coefficient of 1.4 for ovine Fab. Final antivenom concentrations were typically between 20 and 30 g/L. Serum titer estimation. Serum antibody titers directed against purified melittin or whole venom were determined on the pooled monthly samples or bleeds by ELISA accord- ing to the method of Sjostrom and others. 23 Polystyrene mi- crotiter plates (Nunc, Roskilde, Denmark) were coated with antigen (2 g/ml) and washed three times with ELISA wash buffer (137 mM NaCl, 2.68 mM KCl, 8.1 mM Na 2 HPO 4 , 0.25 mM thimerosal, 0.1% Tween 20). Doubling dilutions of immune serum diluted with ELISA wash buffer were per- formed in duplicate. A reference antiserum (from a pool of sera with an established titer) and a normal sheep serum