Involvement of Manduca sexta peptidoglycan recognition protein-1 in the recognition of bacteria and activation of prophenoloxidase system

Abstract

Although the importance of peptidoglycan recognition proteins (PGRPs) in detecting bacteria and promoting immunity is well recognized in Drosophila melanogaster and other insect species, such a role has not yet been experimentally established for PGRPs in the tobacco hornworm, Manduca sexta. In this study, we purified M. sexta PGRP1 from the baculovirus-insect cell expression system, tested its association with peptidoglycans and intact bacteria, and explored its possible link with the prophenoloxidase activation system in larval hemolymph. Sequence comparison suggested that PGRP1 is not an amidase and lacks residues for interacting with the carboxyl group of meso-diaminopimelic acid-peptidoglycans (DAP-PGs). M. sexta PGRP1 gene was constitutively expressed at a low level in fat body, and the mRNA concentration became much higher after an injection of Escherichia coli. Consistently, the protein concentration in larval plasma increased in a time-dependent manner after the immune challenge. Purified recombinant PGRP1 specifically bound to soluble DAP-PG of E. coli but not to soluble Lys-type PG of Staphylococcus aureus. In addition, this recognition protein completely bound to insoluble PGs from Micrococcus luteus, Bacillus megaterium and Bacillus subtilis, whereas its association with the bacterial cells was low even though their peptidoglycans are exposed on the cell surface. After PGRP1 had been added to plasma of naïve larvae in the absence of microbial elicitor, there was a concentration-dependent increase in prophenoloxidase activation. Phenoloxidase activity, as usual, increased after the plasma was incubated with peptidoglyans or bacterial cells. These increases became more prominent when insoluble M. luteus or B. megaterium PG or soluble E. coli PG and PGRP1 were both present. Statistic analysis suggested a synergistic effect caused by interaction between PGRP1 and these PGs. Taken together, these results indicated that PGRP1 is a member of the M. sexta prophenoloxidase activation system, which recognizes peptidoglycans from certain bacteria and initiates the host defense response. The unexplained difference between the purified PGs and intact bacteria clearly reflects our general lack of understanding of PGRP1-mediated recognition and how it leads to proPO activation.

Introduction

Arthropods rely solely on their innate immunity to fight off invading pathogens (Gillespie et al., 1997, Lavine and Strand, 2002, Lemaitre and Hoffmann, 2007). This system is elicited by conserved molecular patterns associated with microbes, such as peptidoglycan (PG), lipopolysaccharide, and β-1,3-glucan. Peptidoglycan, a cell wall component of bacteria, is a high molecular weight polymer consisting of unbranched glycan strands cross-linked by short stem peptides (Vollmer et al., 2008). The glycan strands are composed of alternating β-1,4-linked N-acetylglucosamine and N-acetylmuramic acid. The stem peptides attached to N-acetylmuramic acid residues contain a lysine or meso-diaminopimelic acid (DAP) at the third position, which connects these peptides directly or indirectly through a peptide cross-bridge. Lys-type PGs are found in most Gram-positive bacteria, whereas DAP-PGs are present in Gram-negative bacteria and the Gram-positive genera Bacillus and Clostridium. Different bacteria vary in stem peptide, cross-bridge, and physical properties (e.g. thickness, elasticity, porosity) of their peptidoglycans.

Peptidoglycan recognition proteins (PGRPs) are immunity-related molecules conserved from insects to humans (Royet and Dziarski, 2007, Chaput and Boneca, 2007). The founding member of this protein family was isolated from the silkworm, Bombyx mori (Yoshida et al., 1996). The carboxyl-terminal PG-binding domain of B. mori PGRP is approximately 165 residues long and homologous to lysozyme of bacteriophage T7 (Ochiai and Ashida, 1999, Kang et al., 1998). Due to the lack of key catalytic residues in T7 lysozyme, the silkworm protein does not cleave an amide bond between N-acetylmuramic acid and l-alanine of the stem peptide. Its binding of Lys-type PG from Micrococcus luteus, however, triggers a serine proteinase cascade that leads to prophenoloxidase (proPO) activation and melanin formation. In Drosophila, PGRP-SA and -SD bind Lys-PG to induce Toll pathway; PGRP-SA is also involved in proPO activation; PGRP-LC and -LE bind DAP-PG fragments and induce Imd pathway; PGRP-LC stimulates phagocytosis (Charroux et al., 2009). On the other hand, Drosophila PGRP-SB, -SC, and -LB hydrolyze the amide bond between N-acetylmuramic acid and l-alanine in the stem peptide to reduce the amount of immunogenic muropeptide and down-regulate the immune responses (Royet and Dziarski, 2007). In Tenebrio molitor, clustered PGRP-SA detects Lys-type PG and activates the proPO cascade and Toll pathway (Park et al., 2007).

Three-dimensional structures of PG-binding domains in the Drosophila and human PGRPs reveal an α/β mixed fold similar to that of T7 lysozyme (Kim et al., 2003, Reiser et al., 2004, Guan et al., 2004, Guan et al., 2005, Lim et al., 2006, Cho et al., 2007, Leone et al., 2008). This general fold consists of a central β-sheet and three peripheral α-helices, parts of which form the walls and bottom of a PG-binding groove found in most PGRPs. The groove is composed of His, Tyr, His, Thr and Cys/Ser residues that correspond to His17, Tyr46, His122, Lys128 and Cys130 of T7 lysozyme (Mellroth et al., 2003). Some of these residues are responsible for Zn²⁺-binding and N-acetylmuramoyl-l-alanine amidase activity of T7 lysozyme and hydrolytic PGRPs. Structural analysis of free or muropeptide-bound PGRPs, along with site-directed mutagenesis and binding assays, suggests that certain residues in the binding groove are responsible for the differential recognition of Lys- and DAP-PGs (Guan et al., 2004, Lim et al., 2006, Cho et al., 2007).

In Manduca sexta, cDNA cloning disclosed a 19 kDa protein most similar in sequence to B. mori PGRP (identity: 54%) (Yu et al., 2002, Zhu et al., 2003). While its protein level in larval plasma increased after an injection of M. luteus, immunological function of M. sexta PGRP1 remains unclear. In fact, unpublished data indicated that addition of PGRP1 to plasma failed to enhance melanization triggered by M. luteus (Kanost et al., 2004). To explore its possible role in bacteria recognition and initiation of proPO activation, we expressed M. sexta PGRP1 in insect cells and tested its binding to soluble and insoluble PGs from various bacteria and to intact bacterial cells. By correlating the binding data with proPO activation upon exposure to these elicitors, we confirmed the unusual binding properties of M. sexta PGRP1 and established its role as a sensor of the proPO activation system. Implications of our findings regarding recognition of bacteria are discussed as well.