In vitro activation and enzyme kinetic analysis of recombinant midgut serine proteases from the Dengue vector mosquito Aedes aegypti

Abstract

Background: The major Dengue virus vector Aedes aegypti requires nutrients obtained from blood meal proteins to complete the gonotrophic cycle. Although bioinformatic analyses of Ae. aegypti midgut serine proteases have provided evolutionary insights, very little is known about the biochemical activity of these digestive enzymes.

Results: We used peptide specific antibodies to show that midgut serine proteases are expressed as zymogen precursors, which are cleaved to the mature form after blood feeding. Since midgut protein levels are insufficient to purify active proteases directly from blood fed mosquitoes, we engineered recombinant proteins encoding a heterologous enterokinase cleavage site to permit generation of the bona fide mature form of four midgut serine proteases (AaET, AaLT, AaSPVI, AaSPVII) for enzyme kinetic analysis. Cleavage of the chromogenic trypsin substrate BApNA showed that AaET has a catalytic efficiency (k(cat)/K(M)) that is ~30 times higher than bovine trypsin, and ~2-3 times higher than AaSPVI and AaSPVII, however, AaLT does not cleave BApNA. To measure the enzyme activities of the mosquito midgut proteases using natural substrates, we developed a quantitative cleavage assay based on cleavage of albumin and hemoglobin proteins. These studies revealed that the recombinant AaLT enzyme was indeed catalytically active, and cleaved albumin and hemoglobin with equivalent efficiency to that of AaET, AaSPVI, and AaSPVII. Structural modeling of the AaLT and AaSPVI mature forms indicated that AaLT is most similar to serine collagenases, whereas AaSPVI appears to be a classic trypsin.

Conclusions: These data show that in vitro activation of recombinant serine proteases containing a heterologous enterokinase cleavage site can be used to investigate enzyme kinetics and substrate cleavage properties of biologically important mosquito proteases.

Figure 1

Serine proteases in Ae. aegypti mosquitoes are processed to generate the mature active forms of the enzymes. A) Western blot of midgut epithelium (MgEp) and food bolus protein extracts obtained from blood fed mosquitoes at 24 hr PBM using antibodies specific for AaSPVI, AaSPVII, and AaLT as previously described [13]. Each lane contains protein from 0.2 midgut equivalents obtained from pooled blood fed mosquitoes. Note that it is not possible to use a protein loading control for these two extracts because the amount of protein differs by >5-fold (MgEp versus food bolus), and moreover, the MgEp proteins are from mosquito origin and the food bolus proteins are from bovine origin. B) Amino acid sequence of the nascent polypeptide (zymogen) for the four mosquito serine proteases (AaET, AaSPVI, AaSPVII, AaLT) and bovine trypsin (BvT). The protein leader sequence was predicted by SignalP 3.0, whereas the propeptide sequence was defined as the sequence between the leader sequence C-terminus and the N-terminus of the mature form of the proteases based on the known amino acid sequence of the BvT mature form. The predicted propeptide cleavage site is shown. C) Amino acid sequence of the Mek forms of the four recombinant mosquito proteases showing the engineered EK propeptide sequence.

Figure 2

Representative gel of purified active mature mosquito proteases and bovine trypsin used in the enzyme assays. Each lane contains 10 μg of protein. The asterisk denotes the protein band corresponding to the expected mature protease based on Western blots of Ae. aegypti midgut extracts from blood fed mosquitoes [13].

Figure 3

Partial proteolysis of BSA using BvT or the activated forms of each mosquito enzyme. A) Partial proteolysis of BSA in the absence (Cntrl) or presence of BvT for 15 or 60 min. The reactions were terminated with SDS-PAGE loading buffer and boiling for 3 min. The gel is stained with GelCode Blue reagent. B) Partial proteolysis products of BSA generated by in vitro cleavage reactions using each of the indicated enzymes for 15 or 60 minutes. The mass ratio of BSA:protease was 10:1 for all reactions except for the AaSPVII reaction, in which the mass ratio of BSA:protease was 298:1 owing to the high BSA cleavage activity of this enzyme relative to the others.

Figure 4

Representative gels of albumin (BSA) and hemoglobin (Hb) digestion by activated proteases for use in quantitating enzyme activities using natural blood protein substrates. The doublet in the Hb digestion reactions corresponds to the α and β subunits, which were both present in the commercial Hb preparation. For these experiments, all five enzymes were used at the same mass ratios, which were 922:1 for BSA:protease and 214:1 for Hb:protease. Rates of digestion from similar digestion reactions were determined using image analysis quantitation as described in Materials and Methods. Digestion data from multiple reactions run in triplicate are presented in Table 2.

Figure 5

Structural homology modeling of AaLT and AaSPVI using the Protein Model Portal (PMP) developed by the Swiss Institute of Bioinformatics. A) Structure of the heel fly collagenase (PDB 1hyl) and AaLT model, which was generated by PMP using the heel fly collagenase structure as a scaffold. The heel fly collagenase and AaLT proteins are 40% identical at the amino acid level. Amino acid residues in the catalytic triad in the heel fly collagenase (His57, D102, S195) and AaLT model (His69, D116, S209) are shown in red CPK, along with the discerning serine residue at the bottom of the specificity pocket, which is shown in magenta CPK. Three distinct amino acids located near the extended substrate-binding site in AaLT and AaSPVI are shown in stick style and labeled. B) Structural model of AaSPVI generated by PMP using the salmon cation trypsin structure (PDB 1mbq) as the scaffold. The AaSPVI and salmon proteins are 44% identical at the amino acid level. The catalytic triad residues are shown in red CPK (His81, D125, S221), and the discerning aspartate in the specificity pocket is colored salmon CPK. C) Structural alignment of the AaLT and AaSPVI models shown in the same ribbon colors as in A and B. The S203 (AaLT) and D215 (AaSPVI) residues in the specificity pockets are shown in CPK. The structural alignment was rendered using ICM BrowserPro (Molsoft).