Regulation of phenoloxidase activity by high- and low-molecular-weight inhibitors from the larval hemolymph of Manduca sexta

Abstract

Insect phenoloxidases (POs) generate quinones and other reactive intermediates to immobilize and kill invading pathogens and parasites. Due to the presumed cytotoxicity of these compounds, PO activity and its proteolytic activation have to be regulated as a local, transient reaction against nonself in order to minimize damage to the host tissues and cells. We identified a Manduca sexta cDNA encoding a polypeptide sequence with its carboxyl-terminal 33 residues similar to the housefly phenoloxidase inhibitor (POI). The recombinant POI, secreted into the Escherichia coli periplasmic space along with its fusion partner DsbC, was released by osmotic shock and isolated by nickel affinity chromatography. Following enterokinase digestion and protein separation, the POI was purified to near homogeneity in a soluble form which inhibited M. sexta PO at a high concentration. We then produced the inhibitor using a modified baculovirus-insect cell system and isolated the glycoprotein from the conditioned medium. Deglycosylation coupled with inhibition assay revealed that O-glycosylation only moderately increased its inhibitory activity. While this led us to speculate the role of Tyr(64) hydroxylation, we were unable to modify the recombinant protein with tyrosine hydroxylase or purify M. sexta POI (Tyr(64)dopa) from the larval plasma. Instead, we isolated a low-M(r), heat-stable compound which strongly inhibited PO. The wavelength of maximum absorbance is 257 nm for the inhibitor. These data suggest that the down-regulation of PO activity in M. sexta is achieved by two mechanisms at least.

Fig. 1

M. sexta POI and its comparison with homologous sequences from other insects. (A) cDNA and deduced amino acid sequence. The predicted signal peptide is underlined, and the O-linked glycosylation site is marked “◼”. Shaded Tyr⁶⁴ corresponds to dopa³² in the housefly POI, which is considered to be critical for its activity. (B) Alignment of the disulfide-knotted regions in homologous molecules. Ms, M. sexta (BE015616); Bm, Bombyx mori (BY939736); Md, M. domestica (AAB33998); Ag, Anopheles gambiae (CD747521); Aa1 and Aa2, Aedes aegypti (EB099073 and EB102275, respectively); He, Heliconius ethilla (DT668523.3); Dm1 and Dm2, Drosophila melanogaster (BK002735 and BK002734, respectively). Six Cys residues that form three disulfide bonds (Cys₁–Cys₄, Cys₂–Cys₅ and Cys₃–Cys₆) in M. domestica POI (Daquinag et al., 1999) are conserved in all of the sequences. The highly similar residues (Gly, His, Asp, Ser, Leu, Tyr, Lys, Val) are shaded and the conserved Tyr is boxed.

Fig. 2

Isolation of M. sexta POI from E. coli. (A) SDS-PAGE and immunoblot analysis of the DsbC-POI fusion protein in the periplasmic extract (lane 1), Ni-NTA elution fractions (lane 2) and enterokinase digest (lane 3). Left panel, silver staining; right panel, immunoblot analysis using S-protein antibodies. Positions and sizes of the M_r markers are indicated. The fusion protein and DsbC are marked by arrows. (B) Purification of recombinant POI by reverse phase HPLC on a Bio-Rad Hi-Pore RP-318 column. Flow rate: 1.0 ml/min; buffer A: 0.1% TFA in 5% acetonitrile; buffer B: 0.1% TFA in 95% acetonitrile; gradient: 0–75% B in 75 min; UV detection: 214 nm.

Fig. 3

Inhibition of M. sexta PO by recombinant POI from E. coli. (A) Concentration-dependent inhibition by the purified protein. (B) Effect of POI on proPO activation. Lane 1, proPO (with a small amount of PO cleaved after prolonged storage); lane 2, PO activated by PAP-3 and SPHs; lane 3, PO activated by PAP-3 and SPHs in the presence of recombinant POI (60 ng/µl).

Fig. 4

Purification of recombinant POI produced in baculovirus-infected Sf21 cells. The recombinant POI from insect cells were enriched and purified by affinity chromatography on an Ni-NTA column. As described in Fig. 1, eluted proteins were further separated by reverse phase HPLC (A). The gradient was 0–40% B in 40 minutes. (B) Inhibitory activity.

Fig. 5

Detection of O-linked glycosylation in M. sexta POI from Sf21 cells. (A) SDS-PAGE analysis of the recombinant POI before (lane 1) and after (lane 2) deglycosylation. (B) Effect of glycosylation on POI activity. 1, purified POI; 2, purified POI treated with O-glycosidase; 3 and 4,O-glycosidase and buffer controls.

Fig. 6

Isolation of a low-M_r PO inhibitor from the larval hemolymph. (A) Elution profile. column: Bio-Rad Hi-Pore reverse phase column RP-318; flow rate: 1.0 ml/min; mobile phase: 10% methanol; UV detection: 214 nm. The activity peak is indicated by underlied bar. (B) Inhibition of mushroom tyrosinase (left axis, ●—●) and M. sexta PO (right axis, □--□) by the low-M_r compound. (C) Effect of proteinase treatment on the recombinant POI and low-M_r inhibitor. (D) UV absorption spectrum. The low-M_r inhibitor was scanned between 210 and 300nm on a Beckman DU-520 UV/vis spectrophotometer using 20mM Tris-HCl, pH 7.5, as blank.

Fig. 7

Expression profiles of M. sexta POI in fat body at different immune states or development stages. (A) CF and CH, fat body and hemocytes from the naïve larvae; IF and IH, fat body and hemocytes from the larvae injected with bacteria 24 h before. (B) 4_e, fourth instar day 1; 4_m, fourth instar day 3; 4_l, fourth instar day 5; 5_e, fifth instar day 1, 5_m, fifth instar day 3; 5_l, fifth instar day 5, W_e, W_m and W_l, early, middle and late wandering stages; P_e, P_m and P_l, early, middle and late pupal stage.