Aedes aegypti is the main vector for dengue and yellow fever in tropical and subtropical regions of the world (Honório et al., 2003; de Lima-Camara et al., 2006; Kaur et al., 2008; Maidana and Yang, 2008; Troyo et al., 2008; Vasilakis and Weaver, 2008). These viral human diseases have become major international public health concerns because of the geographical spread of mosquitoes that readily vector dengue and yellow fever viruses (Shope, 1991; Gubler, 2002; Lourenço-de-Oliveira et al., 2004; Maciel-de-Freitas et al., 2006; Wilder-Smith and Gubler, 2008). Dengue is the most common mosquito-borne viral disease of humans; globally 2.5 billion people live in areas where dengue viruses can be transmitted (WHO, 2008). In 2007, over 830,000 cases of dengue were reported in the American continent alone (PAHO, 2008). A. aegypti is widespread in Brazil transmitting three of the four dengue virus serotypes (Nogueira et al., 2000; Ríos-Velásquez et al., 2007).
During insect development, proteases play central roles in numerous processes, such as digestion, oogenesis, and metamorphosis (Terra and Ferreira, 1994; Borovsky and Mahmood, 1995; Nakajima et al., 1997). A number of proteases have been identified in both adults and larvae of the order Diptera. Serine proteases have been described in A. aegypti (Yang and Davies, 1971; Graf et al., 1986; Graf and Briegel, 1989; Ho et al., 1992; Borovsky and Meola, 2004), Anopheles (Han et al., 1997; Rosenfeld and Vanderberg, 1998; Abraham et al., 2005; Okuda et al., 2005; Rodrigues et al., 2007), Oestrus ovis (Tabouret et al., 2003), Culex pipiens pallens (Gong et al., 2005), Lutzomyia longipalpis (Fazito do Vale et al., 2007), and Dermatobia hominis (Pires et al., 2007). In addition, cysteine peptidases, leucine-aminoprotease, carboxypeptidase A, and carboxypeptidase B have also been described in A. aegypti (Cho et al., 1999; Noriega et al., 2002) and metalloproteases have been reported in Anopheles stephensi, Drosophila melanogaster, and D. hominis (Rosenfeld and Vanderberg, 1998; Vierstraete et al., 2003; Pires et al., 2007).
Despite the existence of valuable information about the presence of proteases in specific tissues of A. aegypti, as far as we know, no data have been reported on the proteolytic enzyme profile of whole extracts of pre-imaginal stages of this species. This study employed zymographic analyses (gelatin-SDS-PAGE) to characterize and compare the proteolytic expression patterns in the four larval stadia and pupae of A. aegypti.
MATERIALS AND METHODS
Stock solutions of 1,10-phenantroline (200 mM) and pepstatin A (1 mg/ml) were prepared in ethanol, whereas trans-epoxysuccinyl L-leucylamido-(4-guanidino) butane (E-64, 1 mM) was prepared in water. Phenyl-methyl sulfonyl-fluoride (PMSF, 250 mM) was diluted in isopropanol, and N-α-Tosyl-L-lysine chloromethyl ketone hydrochloride (TLCK, 100 mM) and N-p-Tosyl-L-phenylalanine chloromethyl ketone (TPCK, 100 mM) were prepared in methanol. All protease inhibitors were maintained at −20°C.
Insects and Collecting Methods
All experiments were conducted using larvae (L1, L2, L3, and L4) and pupae of A. aegypti obtained from a colony originating from insects captured in the Brazilian state of Rio de Janeiro and maintained in the Laboratório de Transmissores de Hematozoários, Instituto Oswaldo Cruz, Rio de Janeiro.
Larvae and pupae were washed twice with phosphate-buffered saline (PBS), pH 7.2, and homogenized in lysis buffer containing 10% glycerol, 0.6% Triton X-100, 100 mM Tris-HCl pH 6.8, and 150 mM NaCl. The homogenates were centrifuged at 14,000g for 40 min at 4°C to remove insoluble material, and proteins were resolved as previously described (Cuervo et al., 2008). Briefly, the supernatants were mixed with sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer (125 mM Tris, pH 6.8, 4% SDS, 20% glycerol, 0.002% bromophenol blue) and resolved in 12% SDS-PAGE co-polymerized with 0.1% porcine gelatin. The gels were loaded with 20 µg of protein per well. After electrophoresis at a constant voltage of 110 V and at 4°C, the gels were washed twice for 30 min at 4°C in either 100 mM sodium acetate buffer pH 3.5 or 5.5 containing 2.5% Triton X-100 or 100 mM Tris-HCl buffer pH 7.5 or 10.0 plus 2.5% Triton X-100. Protease activities were detected by incubating the gels in a reaction buffer containing 100 mM sodium acetate (pH 3.5 or 5.5) or 100 mM Tris-HCl buffer (pH 7.5 or 10.0) at 37°C for 30, 60, and 120 min for larvae, and 24 h for pupae homogenates. Bands of hydrolyzed gelatin were visualized by staining the gels with 0.2% Coomassie blue R-250, 40% methanol, 10% acetic acid, and destaining them with 10% acetic acid. The molecular masses of the proteases were estimated by comparison with the mobility of molecular mass standards (Fermentas Life Science, Hanover). The results are representative of five independent experiments carried out in triplicate.
After electrophoresis, gels of larvae extracts were incubated at 4, 10, 37, 50, 60, or 85°C for 60 min in preheated 100 mM Tris-HCl pH 7.5 reaction buffer.
Protease Inhibition Assays
Larvae and pupa homogenates were pre-incubated (before electrophoresis) for 30 min at 37°C with one of the following protease inhibitors: 10 µM E-64, 1 mM PMSF, 100 µM TLCK, 100 µM TPCK, 10 µM pepstatin-A, or 10 mM 1,10-phenanthroline. Inhibitor was also added to the reaction buffer at the same concentration. Samples were then resolved as described above.
Zymographic Profiles and Time-Courses of Proteolytic Activities From Larval Instars
The zymographic profile from L1, L2, L3, and L4 larval instars was first analyzed after a 1-h reaction at pH 7.5. The protease activities resulted in the appearance of six to eight bands ranging in molecular masses from 20 to 250 kDa (Fig. 1). Qualitative differences in the proteolytic profiles were observed between the larval instars. Time-course assays were performed to evaluate how the length of the incubation time impacted the enzymatic activity (Fig. 2). The intensity of proteolysis increased progressively during incubation from 30 to 120 min. Although enzymatic activities were detected in all experimental conditions, the protease composition of all larval instars was clearly pronounced in gels that were allowed to incubate for 60 min. Thus, a reaction time of 60 min was used for all subsequent larval enzymatic assays.
Influence of pH on the Proteolytic Activities of Larval Instars
To investigate the pH dependence of the protease activities, the gels were incubated for 1 h in buffers ranging from pH 3.5 to 10.0. Proteolytic activities were detected at all pHs in each of the four larval instars (Fig. 3). However, the intensities of the proteolytic profiles at pH 3.5 and 5.5 were drastically reduced when compared to those obtained at pH 7.5 and 10.0. At pH 10.0, some of the proteolytic zones overlapped in L3 and L4 instars, whereas some bands with molecular masses between 55 and 95 kDa were more pronounced in L1 and L2 instars.
Effect of Temperature on the Proteolytic Activity From Larval Instars
The effect of temperature on the proteolysis was followed at 4–85°C (Fig. 4). Low enzymatic activity was observed at 4 and 10°C when compared with the standard assay at 37°C. Proteolytic activities reached a maximum at 50 and 60°C, and decreased at 85°C to a lower level that observed at 37°C.
Enzymatic Inhibition Assays in Larval Instars
The effect of a number of protease inhibitors on larval proteolytic activities was determined (Fig. 5). Enzymatic activities in the larval homogenates were not affected by 10 µM E-64, 10 µM pepstatin A, 100 µM TPCK, or 10 mM 1,10-phenanthroline. On the contrary, 100 µM TLCK and 1 mM PMSF strongly inhibited the enzymatic activities in all larval instars.
Effect of pH and Inhibitors on the Proteolytic Activity Profile of the Pupa
The proteolytic profile during the pupa stage was evaluated after incubation of the gels in buffers at pH 3.5–10.0 for 24 h (Fig. 6A). Protease activities were detected at a pH range of 5.5–10.0, with no proteolytic bands detected at pH 3.5. Differences in the number and intensity of the activity bands were observed across pH 5.5, 7.5, and 10.0. The most complex proteolytic profile from the pupal stage was observed at pH 10.0, with at least 11 bands ranging from 25 to ∼200 kDa resolved. The activity of a proteolytic band appearing to have a molecular mass of 200 kDa was strongly increased at pH 7.5 and 10.0, while strong proteolytic activity exhibited by an enzyme of approximately 40 kDa was uniformly intense in a pH range of 5.5–10.0. The minimum incubation time to detect protease activities from the pupal stage was fixed at 24 h, as no enzymatic activity was detected when homogenates were incubated for 30, 60, or 120 min (data not shown), indicating very low activity. The enzymatic profile exhibited by pupa homogenates was strongly inhibited by 1 mM PMSF or 100 µM TLCK (Fig. 6B). Proteolytic activities were not affected by 10 µM E-64, 100 µM TPCK, 10 µM pepstatin A, or 10 mM 1,10-phenanthroline (data not shown).
This study reports the biochemical characterization and comparative analysis of diverse serine proteases in larvae and pupae of A. aegypti at different developmental stages using gelatin-SDS-PAGE. The detection of larval protease activities is time and pH dependent. Although a time-course experiment demonstrated that activities increased with longer incubation time, no additional bands were observed after 1 h of incubation. In addition, even though all activities were strongly detected at pH 10.0 and some proteolytic bands from L1 and L2 instars were better visualized at this pH (when compared to pH 7.5), the hyperactivity of the proteases at pH 10.0 produced overlapping zones and smears in several regions of the gel, precluding an accurate analysis of the proteolytic profile expressed by different larval instars. Therefore, the optimal experimental conditions to detect proteolytic activities in different larval instars were incubation for 1 h at pH 7.5. Protease inhibitors PMSF and TLCK strongly inhibited the proteolytic activities during different larval stadia. The inhibition of the proteolytic pattern by PMSF (an inhibitor of chymotrypsin and trypsin), and TLCK (a specific inhibitor of trypsin), indicates that the main proteolytic enzymes of larval of A. aegypti are trypsin-like serine proteases. These results agree with previous descriptions of the presence of trypsin in larval stages of A. aegypti, which have been detected using other biochemical methods (Yang and Davies, 1971; Ho et al., 1992; Borovsky and Meola, 2004). In fact, it has been reported that the gut of larval stages of A. aegypti synthesizes mostly trypsin-like enzymes (∼90%) and also other proteases in a lesser extent (Borovsky and Meola, 2004). The experimental conditions used here were intended for the detection of proteases from whole larvae and pupa extracts. Using dissected guts from pre-imaginal stages of A. aegypti and other Diptera species, other authors have also detected trypsin and chymotrypsin-like enzymes (Mahmood and Borovsky, 1992; Tabouret et al., 2003; Borovsky and Meola, 2004; Fazito do Vale et al., 2007; Pires et al., 2007).
Due to the fact that SDS-substrate-gel electrophoresis detects active proteases, proteins are not completely denatured and reduced. Therefore, the molecular weight calculated for proteolytic activities is only an estimated. Here we detect trypsin-like serine proteases ranging from 20 to 250 kDa whereas genome data revealed molecular masses between 24 and 35 kDa (Nene et al., 2007). Because the protein samples are not boiled in the presence of SDS and β-mercaptoethanol, the proteolytic activities of high molecular masses detected here could be due to protein aggregates or to proteases still bound to some membranes that precipitated at the top of the gels or migrated very slowly. In addition, it is possible that binding of the proteases to the substrate also impeded the migration on the gel. Such factors may slow the migration of the proteases, giving an illusion of a higher molecular mass (Nauen et al., 2001).
Differences in the observed intensities of the proteolytic banding patterns in the intra- and inter-stadium samples may indicate that some proteases are differentially expressed and/or exist in distinct isoforms with specific catalytic features. Supporting this hypothesis, quantitative differences in trypsin biosynthesis in the gut of A. aegypti larvae were previously reported (Borovsky and Meola, 2004). Such quantitative differences could correspond to our observation of differences in proteolytic band intensity in distinct larval instars. Additionally, the expression of functionally different proteases may be associated with distinct processes such as nutrition, development, and defense (Ho et al., 1992; Yano et al., 1995, Nakajima et al., 1997; Borovsky and Meola, 2004; Philip et al., 2007).
Comparison of protease activities from pupae and larvae show that both developmental stages express trypsin-like serine proteases that exhibit optimal activities at alkaline pH. However, the larval proteases exhibit a broader pH range of activity than that of the pupal enzymes. Enhanced resolution of proteolytic bands from pupa was obtained at pH 10.0; on the other hand, larval instars were resolved adequately at pH 7.5. Additionally, the composition of the proteolytic profile from pupae is qualitatively different from that exhibited by larval stadia. Although the pupa presents at least 11 main bands at pH 10.0, the larval stadia present only eight proteolytic bands. Also, the incubation time required for detecting proteolytic activities differs greatly between the stages as proteases of larval stadia can be detected after just 1 h of incubation, while detection of pupal protease activities requires 24 h of incubation. This observation could indicate the presence of highly active digestive proteases in larvae, the expression of which diminishes after larval–pupal ecdysis. It has been previously suggested (Yang and Davies, 1971) that highly active serine proteases at the larval stage may be necessary to accomplish the digestion of food particles at the low temperatures of the aquatic environment, and that the production of these enzymes would decrease upon the termination of active digestion, i.e., when larval–pupal ecdysis is complete. On the contrary, other authors have shown that the level of trypsin-like enzymes in the gut of pupae is similar to that found in the fourth instar larva (Borovsky and Meola, 2004). These differences between our work and these previous results could be due to: (i) the activities detected here are from whole larvae and pupae extracts, whereas the other authors used dissected guts; (ii) proteases detected in larval instars and pupae displayed different kinetics; (iii) the previous work was a quantitative comparison of total trypsin synthesis (active and inactive) in the gut of larvae and pupae, whereas in this study we detected only active proteases; and (iv) an endogenous serine protease inhibitor synthesized by pupae may be released when this stage is homogenized.
Despite notable differences in proteolytic profiles between pupae and larval stadia, common features, such as a band migrating at ∼40 kDa, were present in both the four larval stadia and in the pupa stage. These results suggest that both developmental stages constitutively express a trypsin-like serine protease that could perform similar functions throughout the life cycle of the insect. Thermal stability experiments illustrated that despite decreases in enzymatic activity at 4 and 10°C, the proteolytic profile of the four larval instars could still be observed. Enzymatic activity increased between 37 and 60°C, and decreased at 85°C, indicating that serine proteases of A. aegypti larvae are well adapted to extreme temperature conditions. Similar results were described for serine proteases of O. ovis (Angulo-Valadez et al., 2007), and Tenebrio molitor (Elpidina et al., 2005; Tsybina et al., 2005).
Trypsin and chymotrypsin serine proteases have been identified in specific organs from different stages of A. aegypti (Yang and Davies, 1971, 1972; Terra and Ferreira, 1994; Noriega et al., 1996; Noriega and Wells, 1999; Borovsky and Meola, 2004); however, to our knowledge, the application of zymography to the biochemical characterization and comparative study of the expression of proteases from the whole extracts of this species has not been reported. Our findings demonstrate that zymographic analysis is a practical and reliable methodology that could be applied to study changes in the expression of active proteases in species that progress through several developmental stages during their life cycle. In addition, this method also allows: (i) the study of the effect of pH, temperature, and reaction time on the protease profile, (ii) the detection of protease isoforms, and (iii) the isolation of specific active proteases (Zhao and Russell, 2003; Cuervo et al., 2008). Finally, this methodology together with other protein identification methods, such as mass spectrometry, could be a powerful approach for the analysis of active forms of proteases compared to traditional biochemical methods (Zhao et al., 2004; De Jesus et al., 2009).
We thank Bruno Esquenazi and Rodrigo Mexas (Laboratório de Produção e Tratamento de Imagem-FIOCRUZ) for their valuable contributions.