Proteolytic activity of africanized honeybee (Apis mellifera: hymenoptera, apidae) venom

Original paper

Proteolytic activity of africanized honeybee (Apis mellifera: hymenoptera, apidae) venom

P. R. M. DE LIMA, M. R.¹, BROCHETTO-BRAGA^1,2,3,

J. CHAUD-NETTO^1,2,3

1 Department of Biology, Institute of Biosciences of Rio Claro, State of São Paulo, Brazil; 2 Center of Study of Social Insects - CEIS - UNESP, State of São Paulo, Brazil; 3 Center for the Study of Venoms and Venomous Animals - CEVAP-UNESP, State of São Paulo, Brazil.

ABSTRACT: Some properties of a Africanized honeybee venom proteases were determined by enzymatic assays in solution, electrophoresis in SDS-PAGE, and gel filtration. Bee venom extracts were obtained by reservoir disruption, selective dialysis (cut off 12 kDa) to eliminate small components, such as the protease inhibitor present in the venom, and then fractionation of the dialyzed extract by gel filtration on a Sephadex G-100 column. The optimal conditions for the caseinolytic assays were pH 9.5, 2-hour digestion at 37 °C, and 1% casein concentration. The proteolytic activity was also determined by electrophoresis in SDS-PAGE with co-polymerized gelatin with three major bands of 66.0, 41.6, and 25.1 kDa. A principal serine-protease-like mechanism was revealed in the enriched fraction of proteolytic activity.

KEY WORDS: venom, africanized bee, Apis mellifera, proteolytic activity, gel filtration chromatography.

INTRODUCTION

Protease activity has been commonly related to several functions, such as protein transport, cellular and tissue structuring, defense (blood clotting, immune system), and degradation/activation of proteins and enzymes (1).

Proteases from venoms, mainly from snakes, have been characterized by electrophoretical, chromatographic studies and biochemical assays, being associated with hemorrhagic, coagulating/anticoagulating effects, and necrosis caused by the venom (13,19). Insect venoms enzymes, including those of Hymenoptera, are poorly studied and characterized comparatively to other venom components, such as the allergenic enzymes. High levels of protease activity were detected in venoms of some wasps of the genus Polistes (16), but very little is known about the action of these enzymes on insect venoms. Hoffman and Jacobson (8) studied venom from the bumblebee Bombus pensylvanicus and detected and characterized a significant activity of tryptic amidase associated with a potent immunogenic action. On the other hand, nothing is known about those enzymes of Apis bee venom.

The proteolytic activity of the venom from africanized honeybees (Apis mellifera) has been detected by our laboratory. In this study, we analyzed some biochemical properties in order to begin an understanding of its function in that system.

MATERIALS AND METHODS

BEE COLLECTION AND VENOM EXTRACTS. All the experiments performed here were repeated at least twice. About two thousand 25 to 30-day old africanized honeybee workers descended from naturally mated queens were used each time. These bees derived from a pool of 10 colonies were collected from the apiary of the Institute of Biosciences of UNESP, Campus of Rio Claro, São Paulo State, Brazil. They were immobilized by quick freezing at -20°C. The venom reservoirs were extracted at 4°C by dissecting the stinging apparatus, lyophilized, and stored at -20°C until required. Venom sacs were re-suspended in distilled and deionized water and extracts of whole bee venom (WBV) were made by reservoir disruption under rapid defrosting and light pressure with a glass rod. These samples were centrifuged at 10,000g at 4°C for 5 minutes, and the supernatants were used as protein and enzyme sources.

PROTEIN DETERMINATION. The total protein was determined by the Coomassie Blue Binding method described by Sedmak and Grossberg (17).

PROTEOLYTIC ACTIVITY. The proteolytic activity was mainly assayed on casein in a range of pH 6.0 to 10.0, with two buffers 0.1 M sodium phosphate (pH 6.0 to 8.0) and 0.1 M glycine-NaOH (pH 8.0 to 10.0). Venom extracts (0 to 400 mg of protein) were incubated with 0.25 to 1.50 % of casein at 37°C from 0 to 200 minutes in a total volume of 1 ml. A typical reaction contained 150 mg of venom protein and was incubated at 37°C, pH 9.5, with 1% casein for two hours. The reactions were stopped by adding 110 ml of cold 50% trichloroacetic acid and the tubes were allowed to stand for 20 min at 4ºC to ensure complete precipitation of the remaining casein and its fragments. The precipitates were removed by centrifugation (4,500g, 10 minutes), and 900 ml of supernatant was analyzed for the free amino acid content and liberated peptides by proteinase action, according to McDonald and Chen (14). A standard tryptophane curve was used as reference to convert optical density to micromoles of tryptophane.

One unit of proteolytic activity was defined as the quantity of liberated amino acids in ng/mg protein per hour under standard assay conditions.

Other substrates, such as hemoglobin and azocasein were tested at pH 9.5 (0.1 M glycine-NaOH buffer), using 100 mg of WBV protein/assay and incubation at 37ºC from 0-10 hours (15). The final concentration of hemoglobin or azocasein per assay was 1.6% and 2.5%, respectively. One unit of activity was defined for both reactions as the necessary amount of protein to produce an increase of 1.0 in 280 nm of optical density under assay conditions. The three methods were compared based on increases in optical densities after discounting the blanks of reaction.

DETERMINATION OF THE PROTEOLYTIC ACTIVITY IN SDS-PAGE CO-POLYMERIZED GELATIN. The venom proteolytic activity was also assayed by electrophoresis in SDS-PAGE (11% acrylamide) with co-polymerized gelatin (6). The venom samples (100 mg of protein) diluted in 0.05 M TRIS-glycine buffer, pH 6.8, containing 10% glycerol, 0.025 M SDS, and 0.5 mg/ml phenol red were directly applied on the gel. Soon after the run at 150 V for about 3 hours at 4ºC, the gel was then (a) incubated at 37°C in 0.1 M glycine-NaOH, pH 9.5, for 5 hours, and (b) developed for proteolytic activity with 0.1% Amido Black in methanol:acetic acid:water (3:1:6; v:v:v). The gels were destained with methanol:acetic acid:water (3:1:6; v:v:v) until the clear zones of proteolysis could be seen.

REMOVAL OF SMALL COMPOUNDS FROM WBV BY DIALYSIS. About 30 mg of WBV protein were dialyzed in cellulose sacs (cut off 12.0 kDa) at 4ºC for 24 hours against 50 mM ammonium acetate buffer, pH 6.8, at proportion of 1:200 (v/v) with two buffer changes. This extract was called dialyzed bee venom (DBV).

The effects of dialysis on the specific proteolytic activity were analyzed by the two methodologies: casein digestion and SDS-PAGE co-polymerized gelatin.

GEL FILTRATION CHROMATOGRAPHY. Soon after dialysis, DBV was fractionated by gel filtration on a Sephadex G-100 (54 x 2.5 cm) column, which was equilibrated with 50 mM ammonium acetate, pH 6.8, and calibrated with the following standard molecular weights: Bovine Soroalbumin (66.0 kDa), Ovalbumin (450 kDa), Trypsin (24.0 kDa), and Ribonuclease (14.5 kDa). Fractions of 3 ml were collected at a flow of 15 ml/h. The separation was followed by absorbance at 280 nm and by monitoring protease activity as previously described, except by using of 200 ml of each fraction, and time reaction of 18 h instead of 2 h, at 37°C. Pooled fractions from G-100 column with higher levels of proteolytic activity were referred as PFV (Pooled Fractions of Venom).

COMPARISON OF THE LEVELS OF PROTEOLYTIC ACTIVITY IN WBV, DBV, AND PFV. PFV was assayed comparatively with WBV and DBV for caseinolytic activity and by elecrophoresis in SDS-PAGE with co-polymerized gelatin. Twenty-five, 25, and 10 mg of protein from WBV, DBV, and PFV, respectively, were used for the analysis of the caseinolytic activity. Thirty, 30, and 10 mg of protein from WBV, DBV, and PFV, respectively, were applied in each channel of the gel for electrophoresis in SDS-PAGE.

PROTEASE INHIBITOR TESTING. To characterize the africanized honeybee venom proteases in relation to their mechanism, the PFV was assayed with inhibitors added to the casein digestion reactions at the following final concentrations: 0.01 mg/ml Pepstatin A (aspartic-proteases), 2.5 mM EDTA (ethylenediaminetetraacetic acid;metalo-proteases), 0.01 mg/ml Aprotinin (serine-proteases), 0.01 mg/ml Soybean Trypsin inhibitor (serine-proteases), 1 mM PMSF. (phenylmethylsulphonyl fluoride; serine-proteases), and 1 mM PMSF plus 10 mM DTT (dithiothreitol; for reversal of inhibition caused by 1 mM PMSF to cysteine-proteases).

RESULTS AND DISCUSSION

CASEINOLYTIC ACTIVITY IN WBV. Caseinolytic activity assayed in extracts of WBV produced high values at pH 6.5, but its highest level was at pH 9.5 (glycine-NaOH 0.1 M buffer). Schmidt et al.(16) also detected high levels of proteolytic activity at pH 7.0 and 9.2 in some wasp and ant venoms, while in acid pH nothing was observed by these authors

The effect of the venom amount on caseinolytic activity was analyzed in terms of absorbance at 700 nm and also of specific activity (Figure 1). While the absorbance values were considered high using more than 300 mg of WBV/assay, the values of specific activity began to decrease with 50 mg of protein. Those results can indicate or (1) a limitant substrate concentration (0.5%, w/v) used in these assays for those protein levels, or (2) maybe a possible inhibitory action deriving from some component present in the WBV, such as a protease inhibitor previously described by Shkenderov (18).

FIGURE 1.
Effect of the amounts of africanized honeybee whole venom (WBV) on the caseinolytic activity. Assays were performed as described in Material and Methods to the typical reactions using casein as substrate. U=ng of free aminoacids/µg of venom.

Figure 2 shows the effect of casein concentration in the assays from 0 to 1.5% (w/v) on the rate of venom caseinolytic activity. It can be observed that the initial rate becomes essentially independent of substrate concentration above 0.5% casein, so that the working concentration chosen was 1% of casein. The Km value determined by these experiments, according to a double-reciprocal (Lineweaver-Burk) plot, was 0.055%.

FIGURE 2.
Effect of casein concentration 0% to 1.5% (w/v) in the assays of caseinolytic activity in WBV. Lineweaver-Burk curve and Michaelis-Menten (inserted curve). V=ng of free aminoacids/µg of venom protein per hour.

WBV was assayed at different times of reaction with the substrate casein from 0 to 200 minutes, with a linear correlation between both parameters, proteolytic activity and time reaction.

Table 1 shows the comparison of three methods for the estimation of WBV proteolytic activity in a 2-hour digestion of hemoglobin, casein, or azocasein at 37ºC, pH 9.5, with 100 µg of protein. It can be seen that the highest increase in optical density resulted from the casein digestion method adopted in this work.

TABLE 1.
Comparison of the three methods for determining increases in optical density resulting from WBV proteolytic activity in a 2-hour digestion of hemoglobin, casein, or azocasein at 37ºC and pH 9.5.

GEL FILTRATION CHROMATOGRAPHY DBV. Since a serine-protease inhibitor was demonstrated in honeybee venom with molecular weight of 9.0 kDa (18), we considered the possibility that this inhibitor could, in part, account for the low levels of proteolytic activity documented for honeybee venoms (7,9,16). Based on this, we submitted the crude venom (WBV) to a selective dialysis (cut off 12.0 kDa), attempting to eliminate or diminish any inhibition effects caused by that or other minor venom components on the proteolytic activity.

Figure 3 shows the chromatographic profile of DVB on Sephadex G-100 (2.5 x 54.0 cm) column for caseinolytic activity and total protein followed by enzyme activity and absorbance at 280 nm, respectively. The unique and wide peak of proteolytic exhibited a maximum about 40.0 U/h. Due to its width, this single peak was suspected of representing one overlapping of more than one proteolytic enzyme. A putative molecular weight found for this peak was of 40.0 kDa, based on the standard molecular weight curve used to calibrate this column. The total protein profile was almost the same as that found for the proteolytic activity, but a little bit displaced.

FIGURE 3.
Chromatographic profile of about 30mg DVB on Sephadex G-100 column (54 x 2.5 cm) for caseinolytic actitivity (•••) and total protein (¾). U=ng of free aminoacids/µg of venom protein.

COMPARISON OF THE LEVELS OF PROTEOLYTIC ACTIVITY IN WVB, DBV, AND PFV. To determine and quantify any possible enrichment in the proteolytic activity of the africanized honeybee venom extracts, using the sequential steps of (a) venom extract preparation, (b) dialysis, and (c) gel filtration chromatography, we analyzed comparatively these samples by enzymatic assays and SDS-PAGE with co-polymerized gelatin. Table 2 shows the results of the enzymatic assays with casein as substrate. A 3.7 fold increase in caseinolytic specific activity can be observed in PFV over WBV. This increase, though small, naturally means an enrichment in the proteolytic activity due to each treatment. However, the elimination of possible inhibitors could also be suggested, although it should be tested.

TABLE 2.
Comparative analysis of the WBV, DBV, and PFV extracts for caseinolytic activity.

Figure 4 shows the analysis of the proteolytic activity present in africanized honeybee extracts, as well as a comparative experiment in SDS-PAGE co-polymerized gelatin with the same samples from each step of treatment mentioned in Table 2. Three major clear zones resulting from proteolysis of gelatin, exhibiting apparent molecular weights of 66.0, 41.6, and 25.1 kDa can be observed in all the extracts. With regard to the intensity of clear zones on the gel, these results practically correspond to those previously obtained for caseinolytic activity (Table 2).

FIGURE 4.
Profile of africanized honeybee proteolytic activity on SDS-PAGE (11%) of acrylamide) containing co-polymerized gelatin as substrate. In each lane of the gel, from left to right, were, respectively, applied (1) 30 µg of protein of WBV, (2) 30 µg of DBV, and 10 µg of PFV, (3)

DETERMINATION OF A PRINCIPAL MECHANISM FOR THE ENRICHED PROTEOLYTIC ACTIVITY ON THE PFV. In spite of the fact that we did not make any attempt to purify or isolate specifically any of the proteases observed here for the africanized honeybee venom, we tried to determine the existence of a principal mechanism of action for these enzymes (observed in SDS-PAGE). Specific protease inhibitors were incubated with samples of PFV, as described in Material and Methods.

The results obtained in this experiment (Table 3) strongly indicate a general mechanism of serine proteases like to the bee venom proteolytic activity because the two specific serine-proteases inhibitors, aprotinin and soybean trypsin inhibitor, caused considerable inhibition levels of 59.3 and 52.1% , respectively to the proteolytic activity. No inhibition was caused by 1mM EDTA, suggesting the absence of metalo-proteases in this venom. Very low inhibition of only 2.6% was detected with pepstatin, a specific aspartic-protease inhibitor. PMSF is known as a less effective serine-protease inhibitor (15). We also observed very mild levels of inhibition of about 14.3% caused by this compound. Another possible explanation for this could be that this inhibitor is easily inactivated in buffer solutions (10). PMSF also inhibits cysteine proteases, which can be reverted by reduced thiols, such as DTT (4). So, the proteolytic activity over 100% observed in the presence of 10 mM DTT could be explained by the existence of some cysteine-protease activity in this venom extract.

TABLE 3.
Protease inhibitors tested in the enriched fraction of proteolytic activity (PFV) from africanized honeybee venom.

Although a high level of purification of proteases from africanized honeybee venom would be necessary for more detailed studies on their physiological role in the venom system, these results are important to demonstrate for the first time the proteolytic activity in honeybee venom, as well as to establish some of its properties. Recently, Hoffman and Jacobson (8) detected and characterized a significant activity of tryptic amidase associated with a potent immunogenic action in bumblebee venom of Bombus pensylvanicus. These authors also reported that this enzyme contains all the active site-associated residues of serine proteases.

Data from literature as well as some of our experimental results have suggested that this proteolytic activity found in honeybee venom could participate in the process of maturation or activation of venom compounds. These assumptions are based on the following:

1. In bee venom reservoirs or sacs, molecular precursors have been detected with high molecular weights (2,3,5,11,12), in opposition to the detection of smaller molecular forms obtained by electric stimulation (this one considered as the processed venom and almost equivalent to the venom injected in sting victims). Kreil and Bachmeyer (11) and Kreil et al. (12) demonstrated that in queen bee venom glands the peptide melittin, which corresponds to 40-50% of the dry weight of the venom, is synthesized as a large precursor and is possibly processed by a dipeptidylpeptidase, which was detected by these authors. However, this was not fully characterized.

2. On the other hand, we have observed that levels of proteolytic activity detected in the extracts of bee venom reservoir are always 1.5 to 3.0 fold higher than those from venoms obtained by electric stimulation. Naturally, the activity in the reservoirs can be compartmentalized or associated with some cellular organelle. If one of the activities observed here corresponds to the same dipeptidylpeptidase mentioned by Kreil et al. (12), this is under our investigation.

ACKNOWLEDGEMENTS

This study was supported by grants from FAPESP (Proc. 93/4620-8), CNPq (Proc. 302429-88.3 and 300703/81), and CAPES (Fellowships in Master degree: P. R. M. de Lima). The authors are grateful to Dr. E. C. Carmona and Dr. J. A. Jorge for fruitful discussion, and to A. Rodrigues for technical assistance.

Received 10 November 1998

Accepted 17 December 1998

• 01 AVILÉS FX., GUASH A., VENDRELL J. Activación de precursores de proteínas. Invest. Cienc., 1994, 210, 74-81.
• 02 BACHMEYER H., KREIL G., SUCHANEK G. Synthesis of promelittin and melittin in the venom gland of queen and worker bees: patterns observed during maturation. J. Insect Physiol., 1972, 18, 1515-21.
• 03 BANKS BEC., SHIPOLINI RA. Chemistry and pharmacology of honeybee venom. In: DIEK, T. Venoms of the hymenoptera: biochemical, pharmacological and behavioral aspects. London: Academy Press, 1986, 7: 329-416.
• 04 DUNN BM. Determination of protease mechanism. In: BEYNON RJ., BOND JS., Ed. Proteolytic enzymes:a practical approach Oxford: IRL, 1989: 57-81.
• 05 GMACHL M., KREIL G. The precursor of the bee venom constituents apamin and MCD peptide are encoded by two genes in a tandem which share the same 3’-exon. J. Biol. Chem., 1995, 270: 12704-8.
• 06 HEUSSEN C., DOWDLE EB. Electrophoretical analysis of plasminogen activators in polyacrylamide gels containing sodium dodecil sulfate and co-polymerized substrates. Anal. Biochemi., 1980, 102: 196-202.
• 07 HOFFMAN DR. Hymenoptera venom proteins. Nat. Toxins, 1996, 2: 169-86.
• 08 HOFFMAN DR., JACOBSON RS. Allergens in Hymenoptera venom XXVII: Bumblebee venom allergy and allergens. J. Allergy Clin. Immunol., 1996, 97: 812-21.
• 09 HOFFMAN DR., SHIPMAN WH. Allergens in bee venom I. Separation and identification of the major allergens. J. Allergy Clin. Immunol., 1976, 58: 551-62.
• ¹⁰
  10 JAMES G T. Inactivation of the protease inhibitor phenylmethylsulfonyl fluoride in buffers. Anal. Biochem., 1978, 86: 574-9.
• ¹¹
  11 KREIL G., BACHMEYER H. Biosynthesis of melittin, a toxic peptide from bee venom. Eur. J. Biochem., 1971, 20: 344-50.
• ¹²
  12 KREIL G., HAIML L., SUCHANEK G. Stepwise cleavage of the pro part of promelittin by dipeptidylpeptidase IV. Eur. J. Biochem., 1980, 111: 49-58.
• 13 LOMONTE B., GUTIÉRREZ J M. La actividad proteolítica de los venenos de serpientes de Costa Rica sobre la caseína. Rev. Biol. Trop., 1983, 31: 37-40.
• ¹⁴
  14 MCDONALD CE., CHEN LL. The Lowry modification of the folin reagent for determination of proteinase activity. Anal. Biochem.,1965, 10: 175-7.
• 15 SARATH G., LA MOTTE S., WAGNER FW. Protease assay methods. In: BEYNON RJ., BOND JS. Ed. Proteolytic enzymes: a practical approach. Oxford: IRL, 1989: 25-55.
• ¹⁶
  16 SCHMIDT JO., BLUM MS., OVERAL WL. Comparative enzymology of venoms from stinging Hymenoptera. Toxicon, 1986, 24: 907-21.
• ¹⁷
  17 SEDMAK JJ., GROSSBERG SE. A rapid, sensitive and versatile assay for protein using Coomassie Brilliant Blue G-250. Anal. Biochem., 1977, 79: 544-52.
• 18 SHKENDEROV S. A protease inhibitor in bee venom. FEBS Lett. 1973, 33: 343-7.
• ¹⁹
  19 TANIZAKI MM., KAWASAKI H., SUZUKI K., MANDELBAUM FR. Purification of a proteinase inhibitor from the plasma of Bothrops jararaca (jararaca). Toxicon, 1991, 29: 673-81.

CORRESPONDENCE TO:

M.R. BROCHETTO-BRAGA - Departamento de Biologia do Instituto de Biociências, Caixa Postal 199, CEP 13506-900, Rio Claro, São Paulo, Brazil.

Publication Dates

• Publication in this collection
  25 Feb 2000
• Date of issue
  2000

History

• Accepted
  17 Dec 1998
• Received
  10 Nov 1998