The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tydell, C. C.
Right arrow Articles by Selsted, M. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tydell, C. C.
Right arrow Articles by Selsted, M. E.
Right arrowPubmed/NCBI databases
*Substance via MeSH
The Journal of Immunology, 2006, 176: 1154-1162.
Copyright © 2006 by The American Association of Immunologists

Bovine Peptidoglycan Recognition Protein-S: Antimicrobial Activity, Localization, Secretion, and Binding Properties1

C. Chace Tydell2,*, Jun Yuan*, Patti Tran* and Michael E. Selsted*,{dagger},{ddagger}

* Department of Pathology and Laboratory Medicine, {dagger} Department of Microbiology and Molecular Genetics, and {ddagger} Center for Immunology, University of California, Irvine, CA 92697


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Peptidoglycan (PGN) recognition proteins (PGRPs) are pattern recognition molecules of innate immunity that are conserved from insects to humans. Various PGRPs are reported to have diverse functions: they bind bacterial molecules, digest PGN, and are essential to the Toll pathway in Drosophila. One family member, bovine PGN recognition protein-S (bPGRP-S), has been found to bind and kill microorganisms in a PGN-independent manner, raising questions about the identity of the bPGRP-S ligand. Addressing this, we have determined the binding and microbicidal properties of bPGRP-S in a range of solutions approximating physiologic conditions. In this study we show that bPGRP-S interacts with other bacterial components, including LPS and lipoteichoic acid, with higher affinities than for PCP, as determined by their abilities to inhibit bPGRP-S-mediated killing of bacteria. Where and how PGRPs act in vivo is not yet clear. Using Immunogold electron microscopy, PGRP-S was localized to the dense/large granules of naive neutrophils, which contain the oxygen-independent bactericidal proteins of these cells, and to the neutrophil phagolysosome. In addition, Immunogold staining and secretion studies demonstrate that neutrophils secrete PGRP-S when exposed to bacteria. Bovine PGRP-S can mediate direct lysis of heat-killed bacteria; however, PGRP-S-mediated killing of bacteria is independent of this activity. Evidence that bPGRP-S has multiple activities and affinity to several bacterial molecules challenges the assumption that the PGRP family of proteins recapitulates the evolution of TLRs. Mammalian PGRPs do not have a single antimicrobial activity against a narrow range of target organisms; rather, they are generalists in their affinity and activity.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Innate immunity provides the first line of antimicrobial defense in animals (1), and many components of this system are evolutionarily conserved from insects to humans (2). Pattern recognition molecules, encoded in the germ line, are essential components of innate immunity, because they distinguish invading pathogens based on the recognition of microbial surface molecules not present on host cells (3). The recently characterized peptidoglycan (PGN)3 recognition proteins (PGRPs), constitute an important family of pattern recognition molecules (4, 5, 6, 7, 8, 9, 10, 11), some members of which exhibit antimicrobial activity in vitro (6, 10) and in vivo (12). PGRPs have been classified into three categories based on transcript length: short (PGRP-S), short (PGRP-I), and long (PGRP-L). Formerly known as bovine oligosaccharide binding protein (bOBP), bovine PGRP-S (bPGRP-S) is an 18.7-kDa ortholog of human and murine PGRP-S.

PGRPs, first discovered in insects, were named for their ability to bind to PGN and Gram-positive bacteria (13, 14). Recent studies suggest that microbial recognition by PGRPs may not depend on PGN exclusively. Bovine PGRP-S has been shown to kill microorganisms in which PGN is either buried (Gram-negative bacteria) or absent (Cryptococcus neoformans) (10), and Holotrichia PGRP-S has been shown to trigger an insect immune response by specifically binding 1,3-{beta}-glucan (15). In addition, soluble murine PGRP-L can mediate macrophage responses to LPS (16). Moreover, PGRPs mediate Drosophila immune responses to both Gram-negative and Gram-positive bacteria (4, 5, 9, 11, 17), even though PGN is only surface exposed on certain Gram-positive organisms. Although the roles of PGRPs in insects are well documented, less is known about the functions of these proteins in mammalian systems.

To date, mammalian PGRPs have been identified in humans (6, 7, 13), rats (18), mice (13), cattle (10), camels (19), and pigs. PGRPs have structural homology to bacteriophage T7 lysozyme, and several PGRPs are reported to digest PGN (16, 20, 21, 22, 23, 24). Analysis of the human genome predicts the existence of four PGRPs of varying lengths (7, 17, 25). Although primary sequence data led to predictions that human PGRP-Ia and -b and PGRP-L were membrane bound, recent studies suggest that they are probably soluble, secreted proteins (16, 26). Crystallographic studies of PGRP-I revealed that the putative transmembrane domain may actually form a second PGN binding domain (26). In mammals, PGRP proteins have been found in peripheral white blood cells (10, 12) and Peyer’s patch tissue (27) and the mRNA is present in liver, esophagus (7), and oral epithelium (28).

A clear understanding of PGRP action in vivo requires knowledge of the locations where the proteins are stored and where they act. Our previous studies (10) demonstrated that PGRP-S is found in neutrophils and eosinophils and has the same immune staining pattern as {beta}-defensin (29), which is a secreted protein found in bovine dense/large granules. We therefore evaluated PGRP-S in secretion studies and also definitively localized PGRP-S in both naive and phagocytic neutrophils by Immunogold electron microscopy.

An antimicrobial role for mammalian PGRPs has been shown in studies of PGRP-S knockout mice, which exhibit both increased susceptibility to bacterial infection and deficiency in killing of phagocytosed bacteria by neutrophils (12). However, bPGRP-S is the first PGRP shown to be microbicidal in vitro (10), acting against a range of organisms. Purification of PGRP-S from bovine blood provides milligram quantities of natural protein for study, avoiding the use of recombinant material with multiple histidines at the terminus. To further elucidate the role of PGRP-S in innate immunity, we analyzed the binding and microbicidal activities of bovine PGRP-S under conditions approximating the extracellular milieu by modulating cations, ionic strength, osmolality, and serum concentration.

The complete set of molecular targets for PGRPs has yet to be elucidated. Our previous results provided evidence that bPGRP-S could act in a PGN-independent manner. In this study we report the results of competition assays that identify additional bacterial target molecules and show that they are recognized with higher affinity than PGN. Using a direct precipitation assay, we demonstrate that the interaction between bPGRP-S and LPS, one of these alternate molecules, is both dose dependent and saturable. Finally, bPGRP-S was shown to lyse heat-killed bacteria; however, bacterial killing is independent of this function.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Bovine granulocytes

Granulocytes were purified from fresh citrated bovine blood as described previously (10, 30). Preparations contained an average of 1 x 109 granulocyte-enriched leukocytes/l whole blood, of which 93 ± 3% were neutrophils and 4 ± 1% were eosinophils.

Purification of bovine PGRP-S

Bovine PGRP-S was purified from 10% acetic acid extracts of bovine granulocytes by a modification of our previous method (10). Acid extracts of 3 x 108 cell equivalents of bovine leukocytes were loaded onto a 10 x 25-cm Delta Pak reverse phase HPLC C18 cartridge (Waters) equilibrated in 0.1% trifluoroacetic acid (solvent A) at a flow rate of 15 ml/min. A linear gradient of acetonitrile containing 0.1% trifluoroacetic acid (solvent B) was applied at 2.33%/min from 0–35%, then held at 35% B for 10 min, followed by 0.2%/min from 35–42%. The PGRP-S-containing fraction was determined by Western blotting, and purity of >99% was confirmed by gel electrophoresis and reverse phase HPLC as previously described (10). Purified PGRP-S was lyophilized and resuspended in 0.01% acetic acid (HOAc) for storage at –70°C. The protein concentration was quantified spectrophotometrically (1 mg/ml = 1.28A280) (10).

PGRP-S secretion

Granulocyte-enriched populations of bovine peripheral leukocytes were purified as described above and suspended to 1.2 x 107 cells/ml in HBSS (137 mM NaCl, 5.6 mM glucose, 5 mM KCl, 4 mM NaHCO3 1 mM CaCl2, 0.5 mM MgCl2, 0.4 mM KH2PO4, 0.4 mM Na2HPO4, and 0.4 mM MgSO4 (pH 7.4)). Aliquots of the cell suspension were incubated for 60 min at 37°C in a final volume of 500 µl containing one of the following stimulants: 100 nM PMA, 20 µg/ml lipoteichoic acid (LTA) from Bacillus subtilis, 160 µg/ml muramyl dipeptide (MDP) (Sigma-Aldrich), 100 µg/ml LPS from Staphylococcus typhimurium (List Biological Laboratories), 110 µg/ml butyric acid, or 5.4 x 108 CFU/ml nonopsonized bacteria (Staphylococcus aureus 502a or Staphylococcus typhimurium 10428 PhoP–). The viability of the leukocytes was ≥94% for each condition, as determined by trypan blue exclusion at the beginning and the end of the incubation. Bacteria were separated from the cell pellet by centrifugation at 50 x g for 5 min at 4°C. The resulting bacterial suspensions, verified to be free of contaminating leukocytes by light microscopy, were then sedimented at 22,500 x g. Supernatants were acidified by addition of acetic acid to a final concentration of 10% (v/v). Both leukocyte and bacterial pellets were extracted with 10% acetic acid for 36 h with rotation at 4°C. Aliquots of each pellet extract or supernatant (4 x 104 cell equivalents) were lyophilized, boiled in SDS-tricine sample buffer for 10 min, separated by tricine-SDS PAGE and analyzed by Western blotting using anti-bPGRP-S IgG as previously described (10).

Immunogold labeling

Leukocytes exposed to serum-opsonized bacteria and naive leukocytes were evaluated by electron microscopy. S. aureus and S. typhimurium were opsonized with a 50/50 mixture of HBSS and autologous bovine serum for 30 min on ice before an equal volume of the bacterial suspension was added to the leukocytes in HBSS. Phagocytosis was allowed to proceed for 60 min at 37°C. The mixture was then centrifuged, washed, and pelleted. White blood cells were fixed with 2% glutaraldehyde in cacodylate buffer (0.1 M sodium cacodylate, 1 mM MgSO4, and 3 mM CaCl2 (pH 7.4)), washed, then dehydrated and embedded in LR White (London Resin) according to manufacturer’s instructions. Sections (750–900 Å) were prepared using an Ultracut E microtome (Reichert). Primary staining was performed using 1/32 anti-bPGRP-S IgG for 2 h, then washed for 10 min. Secondary staining was performed by an additional 1-h incubation with a 1/75 dilution of goat anti-rabbit IgG conjugated to 15-nm gold beads (BB International) and washed again. Sections were stained with 2% uranyl acetate and Reynold’s lead citrate before evaluation on a Philips EM 201 transmission electron microscope. A negative control staining was performed by substituting preimmune rabbit IgG in the primary incubation.

Microbicidal assays

Listeria monocytogenes 967 and S. typhimurium 10428 PhoP–, suspended in 10 mM Tris-HCl and 5 mM glucose (pH 7.4), were used as target organisms in microbicidal suspension assays as previously described (30, 31). The final concentration of bPGRP-S was 100 mg/ml for S. typhimurium and 50 mg/ml for L. monocytogenes. The effects of ionic strength, divalent cations, and osmolality on microbicidal activity were evaluated by the addition of increasing concentrations of one of the following: NaCl or KCl (0–150 mM), MgCl2 or CaCl2 (0–2 mM), or sucrose (0–250 mM).

The effects of bacterial cell envelope constituents on bactericidal activity were analyzed by preincubating bPGRP-S with one of the following before the protein was added to the suspension of target bacteria: 0–240 µg/ml smooth LPS from S. typhimurium (List Biological Laboratories) and LTA from Bacillus subtilis or MDP (Sigma-Aldrich). Alternatively, 0–4.5 mg/ml PGN from S. aureus (Fluka) was added.

Microbial binding by PGRP-S

One-hundred microliter aliquots of log-phase organisms, suspended in buffers that optimize their survival, were incubated with 200 µg/ml bPGRP-S for 60 min at 37°C with continuous agitation. Candida albicans 16820, Cryptococcus neoformans 271a, S. aureus 502a, and Escherichia coli ML35 were suspended in 10 mM PIPES and 5 mM glucose (pH 7.4). Similar incubations were conducted with S. typhimurium 10428 PhoP– and Listeria monocytogenes 967 in 10 mM Tris-HCl and 5 mM glucose (pH 7.4). The organisms, either untreated or opsonized, were suspended in buffer alone or in buffer supplemented with 2 mM MgCl2 or 150 mM NaCl. Opsonization was accomplished with 50 or 100% fresh, autologous, bovine serum on ice for 30 min before centrifugation and removal of the supernatant. Additional binding assays were performed in HBSS with or without 50% fresh bovine serum or in 100% serum.

For each incubation condition, bacteria were centrifuged after 60 min at 22,500 x g for 10 min at 4°C, washed once with buffer, pelleted again, resuspended in HBSS, and transferred to a new microcentrifuge tube. The pellet resulting from a third centrifugation was boiled for 5 min in SDS-tricine sample buffer, vortexed for 30 s, and boiled for another 5 min. Solubilized samples were resolved by tricine-SDS PAGE (15% acrylamide), and the amount of bPGRP-S present was estimated by determining the signal (MultiImage Light Cabinet; Alpha Innotech) obtained by Western blots with anti-bPGRP-S IgG (10), using purified bPGRP-S as standard.

PGRP-S interaction with LPS

Biosynthetically tritiated rough LPS (80 dpm/pmol recombinant bovine LPS (rbLPS)) from an rb strain of E. coli (List Biological Laboratories) was suspended according to the manufacturer’s directions in 10 mM Tris-HCl buffer (pH 7.4) at concentrations from 10 to 100 µM. To minimize nonspecific binding of tritiated LPS, microcentrifuge tubes were coated with a 200 µg/ml solution of unlabeled, smooth LPS (List Biological Laboratories) in 10 mM Tris-HCl for 1 h with rotation. The coating solution was removed immediately before the tubes were used. Thirty-milliliter aliquots of LPS were added to equal volumes of 0–6.7 µM bPGRP-S in 0.01% HOAc. Samples were prepared in triplicate. After the mixtures were incubated at room temperature for 2 h, each sample received 140 µl 10 mM Tris-HCl and was centrifuged at 22,500 x g for 15 min, and 190 µl of supernatant was removed. Tritiated LPS in each supernatant and residual fraction were quantified by scintillation counting (LS9000; Beckman Coulter) using Packard Ultima Gold scintillation mixture (Sigma-Aldrich). Precipitated LPS was quantified by measuring tritiated LPS counts in the residual 10-µl aliquot of bPGRP-S-containing samples and subtracting tritiated LPS counts in 10-µl control aliquots lacking PGRP-S. At the highest LPS concentration tested (50 µM), spontaneous sedimentation accounted for 6–16% of the input material.

Bacterial cell wall lysis

Dissolution of bacterial cell walls was measured by the lysoplate method of Osserman and Lawlor (32), except that 10 mM HEPES or Tris-HCl (pH 7.4.) with 0–154 mM NaCl was used in place of PBS for preparation of the buffered 1% agarose. Each sample well of the lysoplate received 15 ml of lysozyme standard or bPGRP-S. After incubation for 20 h at room temperature, the radius of each zone of lysis was measured to the nearest 0.5 mm. Lysozyme solutions were prepared in distilled water, and bPGRP-S solutions were prepared in 0.01% HOAc. HOAc (0.01%) was used as a negative control. To verify the enzymatic nature of the lysis, we tested protein solutions that had been boiled for 10 min and rapidly cooled on ice.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Secretion of PGRP-S by neutrophils

It has not been clear whether mammalian PGRPs are secreted to act in the extracellular milieu, or they function exclusively in an intracellular compartment. We previously demonstrated (10) that bovine PGRP-S has the same immunohistochemical staining pattern as {beta}-defensin, a secreted protein stored in the large/dense granules of bovine neutrophils (29). To address the possibility that PGRP-S may be mobilized for secretion by activated leukocytes, we analyzed bPGRP-S secretion by neutrophils stimulated with 100 nM PMA, 20 mg/ml LTA, 160 µg/ml MDP, 100 mg/ml LPS, 110 mg/ml butyric acid, or ~45 nonopsonized bacteria/leukocyte (S. typhimurium or S. aureus). Opsonized bacteria were not used in this assay because this would have contaminated the leukocyte fraction with phagocytosed bacteria. Leukocytes, bacteria, and the resulting supernatants were evaluated by Western blotting. PMA induced a nearly complete release of PGRP-S from leukocytes (Fig. 1). This could not be accounted for by cell lysis, because the treated cells were 98% viable as determined by trypan blue exclusion. Stimulation of granulocytes with nonopsonized S. aureus or S. typhimurium also induced measurable secretion (Fig. 1). In contrast, PGRP-S secretion was not detected in supernatants of granulocytes treated with purified butyric acid, LPS, MDP, or LTA (data not shown). Therefore, PGRP-S appears to be secreted in response to specific extracellular stimuli. Under these experimental conditions, bPGRP-S was not found to bind to bacterial pellets.



View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 1. Secretion of PGRP-S from neutrophils. Aliquots of 3 x 106 granulocyte-enriched bovine leukocytes were incubated for 1 h in buffer alone (sample marked 60 min) with 100 nM PMA or nonopsonized bacteria before separating the leukocytes from the supernatant and bacteria. Leukocyte and bacterial pellets were acid extracted, and the resulting supernatants were acidified before aliquots (6 x 104 CE) were analyzed by Western blotting. Sample in buffer alone was collected at time zero as a control. The samples are as follows: C, purified bPGRP-S standard; P, leukocyte pellet; S, reaction supernatant; B, bacterial pellet.

 
Cytoplasmic localization of PGRP-S in neutrophils

Evidence that PGRP-S is secreted by neutrophils does not, by itself, identify the intracellular location of the protein. Identification of the granule type in which bPGRP-S is stored would, however, offer insight into how the neutrophil uses the protein. We sought to determine the intracellular address of PGRP-S using Immunogold transmission electron microscopy with anti-bPGRP-S IgG. As shown in Fig. 2, A and B, gold particles were concentrated over the unique large granules (also known as dense granules or tertiary granules) observed in neutrophils of cattle (33). A small amount of gold present in the cytosol may be background or may indicate the presence of PGRP-S in small secretory vesicles. The immunostaining pattern is virtually identical with that observed in the Immunogold localization of bovine neutrophil {beta}-defensin-12 (29). Neutrophils that were incubated with S. aureus or S. typhimurium (opsonized and suspended in a 50/50 mixture of serum and HBSS) contained numerous ingested bacteria and showed marked degranulation of PGRP-S-containing dense granules (Fig. 2C). PGRP-S was associated with S. aureus bacteria in phagolysosomes (Fig. 2, C and D), but it is not clear whether PGRP-S was deposited into the phagolysosome directly or whether secreted PGRP-S bound to bacteria before engulfment. Conversely, although the neutrophils that phagocytosed S. typhimurium had similar depletion of large/dense granules and cellular PGRP-S (Fig. 2E), the protein was not associated with the bacteria, and PGRP-S was not detected inside these phagolysosomes (Fig. 2F). The few remaining large granules in these cells were positive for PGRP-S. In the control, preimmune IgG did not bind to the dense granules or to S. aureus inside the phagolysosome (Fig. 2, G and H). These experiments provide additional evidence that PGRP-S is secreted from neutrophils upon stimulation by either Gram-negative or -positive bacteria. The data also suggest that bovine PGRP may enter the phagolysosome after binding to bacteria in the extracellular milieu.



View larger version (106K):
[in this window]
[in a new window]
  		FIGURE 2. Immunogold localization of PGRP-S. A, Immunogold labeling of a bovine neutrophil incubated with anti-bPGRP-S IgG reveals specific labeling of the abundant dense granules. B, Enlargement of the boxed area of A. C, Immunogold labeling, with anti-bPGRP-S IgG, of a neutrophil that has phagocytosed S. aureus reveals depletion of dense granules and bPGRP-S from the cytoplasm, but label is found inside phagolysosome associated with bacterial cells. D, Enlargement of the boxed area of C, showing phagosome. E, Immunogold labeling, with anti-bPGRP-S IgG, of a neutrophil that has phagocytosed S. typhimurium, showing loss of dense granules and PGRP-S from the cell. The phagolysosome containing the bacterial cells shows no PGRP-S. F, Enlargement of the boxed area of E, showing phagosome. G, Immunogold labeling, with preimmune IgG, of a neutrophil that has phagocytosed S. aureus shows the absence of nonspecific Immunogold labeling. H, Enlargement of the boxed area of G, showing phagosome. Phagosomes are indicated by filled arrows. C and D, Electron-lucent areas in cells (open arrows; A and C) are LR White embedding artifacts.

 
Solute modulation of PGRP-S-mediated microbial killing

To date, bovine PGRP-S is the only PGRP reported to kill microorganisms in vitro (10). As a secreted protein, it may perform this function in the extracellular milieu. Previous microbicidal studies were performed in low ionic strength medium that may not reflect the in vivo conditions. We therefore analyzed PGRP-S-mediated killing of bacteria in a range of buffer conditions reflective of the extracellular milieu to determine the effects of increasing ionic strength and divalent cations on PGRP-S-mediated killing. As shown in Fig. 3, A and B, increasing concentrations of NaCl or KCl markedly reduced, but did not ablate, PGRP-S-mediated killing of bacteria. Killing of L. monocytogenes was reduced from 4 logs (99.99% killing) to slightly >1 log by 30 mM NaCl or 60 mM KCl, but bovine PGRP-S continued to kill >90% of the bacteria at salt concentrations up to 150 mM. Similarly, killing of S. typhimurium by PGRP-S was reduced from nearly 5 to 2 logs by 30 mM NaCl, but >1 log PGRP-S-mediated killing occurred even at physiologic concentrations of NaCl (150 mM). The effect of KCl on PGRP-S-mediated killing of S. typhimurium could not be determined because KCl concentrations >60 mM were toxic to the bacteria. Incubation buffers modified with sucrose had no impact on PGRP-S-mediated killing (Fig. 3, E and F), indicating that NaCl and KCl interfered with killing by changing the ionic environment, rather than by their effects on solvent osmolality.



View larger version (34K):
[in this window]
[in a new window]
  		FIGURE 3. Solute modulation of PGRP-S-mediated bactericidal activity. Log-phase bacteria (L. monocytogenes and S. typhimurium) were incubated with bPGRP-S in buffer at the indicated concentrations of NaCl or KCl (A and B), MgCl2 or CaCl2 (C and D), or sucrose (E and F) for 2 h at 37°C before plating. Microbicidal activity was determined by colony counting of plates after 16–18 h of incubation at 37°C. Hand-plating of undiluted aliquots of the suspension allowed quantitation of killing to 5 logs.

 
In contrast, the addition of MgCl2 or CaCl2, ablated PGRP-S-mediated killing of S. typhimurium (Fig. 3D) in a dose-dependent manner, while having no impact on bovine PGRP-S’s activity against L. monocytogenes (Fig. 3C). These data are consistent with the hypothesis that divalent cations interfere with PGRP-S-mediated killing through stabilization of the Gram-negative outer membrane (34), rather than by interacting directly with the protein.

Binding of PGRP-S to microorganisms

We previously reported evidence that bovine PGRP-S can kill bacteria even in the absence of exposed PGN. Therefore, we investigated the prerequisites for PGRP-S binding to target cells using a direct binding assay. PGRP-S was seen to bind not only to Gram-positive bacteria (S. aureus and L. monocytogenes), which express exposed PGN, but also to Gram-negative bacteria (E. coli and S. typhimurium), on which PGN is buried beneath an LPS bilayer, and to fungi (C. neoformans), which lack PGN altogether (Fig. 4). The specificity of this binding was indicated by the fact that C. albicans, a yeast that is neither inhibited nor killed by PGRP-S (10), did not bind PGRP-S even in salt-free buffer.



View larger version (10K):
[in this window]
[in a new window]
  		FIGURE 4. Binding of PGRP-S to microorganisms. Aliquots of bacteria or fungi were incubated with bPGRP-S in the presence or the absence of 150 mM NaCl or 2 mM MgCl2 in buffers optimal for the survival of the microorganism. Tris buffer is 10 mM Tris-HCl and 5 mM glucose (pH 7.4); PIPES denotes 10 mM PIPES and 5 mM glucose (pH 7.4). Bacterial pellets were analyzed by Western blotting. Microorganisms incubated without bPGRP-S did not show a positive signal on Western blots (not shown). Aliquots represent 1 x 106 CFU Salmonella or Listeria, 1 x 107 CFU E. coli, 2 x 106 CFU S. aureus, or 4 x 105 CFU Cryptococcus or Candida.

 
The addition of 2 mM MgCl2 to the incubation buffer ablated PGRP-S-mediated killing of S. typhimurium (see above), but did not eliminate the binding of PGRP-S to L. monocytogenes or S. typhimurium (Fig. 4). This suggests that binding of PGRP-S to target organisms is not sufficient for bacterial killing. In contrast, 150 mM NaCl, which permitted 90–95% killing of L. monocytogenes or S. typhimurium, reduced PGRP-S binding to levels undetectable by Western blotting (Fig. 4). This result may indicate that very low concentrations of PGRP-S are required to kill bacteria or that bacterial death may release PGRP-S under high salt, serum-free conditions. NaCl inhibited binding of PGRP-S to each organism tested. Consistent with these results, HBSS, a buffer that contains physiologic levels of mM NaCl (137 mM), Mg2+ (0.9 mM) and Ca2+ (0.9 mM), antagonized PGRP-S binding to both L. monocytogenes and S. typhimurium (Fig. 4).

To assess the binding of PGRP-S to bacteria under conditions most similar to the extracellular milieu, we analyzed the effects of opsonization and serum on PGRP-S binding to target organisms. Opsonized/washed and nonopsonized S. aureus or S. typhimurium were incubated with bPGRP-S under three conditions: in fresh bovine serum, in HBSS, or in 50/50 HBSS/serum. Opsonization of S. aureus markedly increased PGRP-S binding to the bacteria, and this effect was enhanced by the presence of serum in the incubation buffer (Fig. 5). Binding of PGRP-S to S. typhimurium also increased under these conditions (Fig. 5). These results suggest that binding of bPGRP-S to bacteria is likely to occur in the extracellular environment in vivo.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 5. Effect of serum on PGRP-S binding. Aliquots (4 x 106 CFU each) of nonopsonized or serum-opsonized S. aureus or S. typhimurium were incubated with bPGRP-S (1 µg) in HBSS, a 50/50 mixture of HBSS and serum, or 100% serum. Bacterial pellets were analyzed by Western blot with anti-bPGRP-S IgG. The purified protein standard is 100 ng of bPGRP-S.

 
Antagonism of microbicidal activities by bacterial molecules

The ability of bovine PGRP-S to bind to Gram-negative bacteria and fungi suggests that PGRP-S might recognize microbial molecules other than PGN. To identify targets of PGRP-S binding, we determined whether prebinding of PGRP-S with purified bacterial molecules could competitively inhibit PGRP-S-mediated killing of bacteria. To assess the specificity of any competition, killing was assayed on Gram-negative as well as Gram-positive bacteria. Before incubation with bacterial suspensions, PGRP-S was mixed with 0–25 µg/ml LPS, LTA, PGN, or MDP, then tested for its ability to kill bacteria (see Materials and Methods). As shown in Fig. 6A, LPS and LTA inhibited the killing of a Gram-positive organism (Listeria) by PGRP-S in a dose-dependent manner. Surprisingly, PGN and MDP had little effect; concentrations of PGN up to 448 µg/ml decreased killing of L. monocytogenes by only 2% (data not shown), whereas nearly complete inhibition of the listericidal activity occurred if PGRP-S was preincubated with 12 µg/ml LPS or 24 µg/ml LTA (Fig. 6A).



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 6. Inhibition of PGRP-S-mediated killing by bacterial molecules. bPGRP-S was preincubated with 0–25 mg/ml purified LPS, LTA, PGN, or MDP for 30 min. Preincubated bPGRP-S was added to log-phase organisms, Gram-positive L. monocytogenes in A and Gram-negative S. typhimurium in B, and incubated for 2 h at 37°C before plating. Microbicidal activity was determined by colony counting of plates after 16–18 h of incubation at 37°C. Inhibition of activity was defined as the percentage of CFU killed without preincubation minus the percentage CFU killed with preincubation divided by the percentage CFU killed without preincubation.

 
In contrast, killing of Gram-negative bacteria (S. typhimurium) was not inhibited by preincubation of PGRP-S with LTA (Fig. 6B). Preincubation with LPS showed less inhibition of PGRP-S-mediated killing in the Gram-negative assay than in the Gram-positive assay. As shown in Fig. 6B, 12 µg/ml LPS reduced the PGRP-S-mediated killing of S. typhimurium by ~50%, but at the highest LPS concentration tested, killing was inhibited by <60%. Under these same conditions, 448 µg/ml PGN inhibited killing of S. typhimurium by only 10% (data not shown).

PGRP-S interaction with LPS

The ability of bovine PGRP-S to bind to Gram-negative bacteria and the marked inhibition of PGRP-S-mediated killing by preincubation of the protein with LPS suggested that LPS and PGRP-S might have a specific binding interaction. To clarify this relationship, we determined the effect of PGRP-S on LPS solubility. In three separate assays, triplicate samples of biosynthetically radiolabeled rbLPS (0–50 µM) were incubated with purified bPGRP-S (750 nM or 3.35 µM.). PGRP-S efficiently precipitated up to 84% of the tritiated LPS in a sample (Fig. 7). Kd values for binding affinity could not be calculated by the Scatchard method, because there was no binding plateau. Suggesting specificity in the interaction between LPS and PGRP-S, precipitation failed when the LPS/PGRP-S molar ratio exceeded 3 at 3.35 µM PGRP-S or 7 at 750 nM PGRP-S (Fig. 7). The kinetics of PGRP-S-induced precipitation of LPS are similar to the hook effect or prozone phenomenon, wherein cross-linking of Ag by divalent Ab can be inhibited by excess Ag. A higher concentration of PGRP-S shifts the precipitation curve to the left (Fig. 7), reflecting more effective precipitation of LPS. Recent crystallization data suggest that PGRPs may function as multimers (21) or have tandem PGN-binding domains (26), which may explain PGRP-S-mediated precipitation and the greater efficiency of interaction at higher PGRP-S concentrations.



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 7. Precipitation of LPS by PGRP-S. Biosynthetically radiolabeled rbLPS from E. coli was suspended in buffer at concentrations from 10–100 µM. Aliquots (30 µl) of LPS were added to equal volumes of 0.01% HOAc (negative control) or bPGRP-S (1.5 or 6.7 µM) dissolved in 0.01% HOAc. Samples were performed in triplicate and had negligible SDs. Bovine PGRP-S (750 nM) was used in two separate assay dates ({circ} and •). Results are graphed as the percentage of the total LPS in the sample precipitated vs the LPS/bPGRP-S molar ratio.

 
Bacterial cell wall lysis

Among the known PGRPs, there are some that have been reported to digest PGN (20, 22, 23). Other PGRPs, such as bPGRP-S, lack a critical zinc-binding residue and were predicted to be incapable of enzymatic activity. The finding of carboxypeptidase activity of Drosophila PGRP-SA (24), which lacks the zinc-binding amino acid, led us to test bovine PGRP-S for lytic activity. Bovine PGRP-S has been shown to be sensitive to buffer conditions in microbial binding and killing assays, so we tested for bacterial cell wall lysis by the lysoplate method (32) using buffer lacking NaCl. Assays using lysoplates made with 10 mM HEPES (pH 7.4) demonstrated that lysozyme and bPGRP-S have opposite salt requirements for optimal activity (Fig. 8 and data not shown). The addition of 50 or 154 mM NaCl to HEPES buffer improved the lytic activity of lysozyme and interfered with bPGRP-S-mediated lysis of bacterial cell walls (data not shown). Although lysoplates made with PBS appear to be the optimal substrate for egg white lysozyme activity, bPGRP-S was not able to digest Micrococcus lysodeikticus cell walls under these conditions (data not shown).



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 8. Bacterial cell wall lysis. The lysoplate method was used to evaluate lysis of bacterial cell walls by bOBP. Lyophilized M lysodeikticus was suspended in molten 1% agarose made with 10 mM HEPES (for bOBP assays) or in 10 mM HEPES containing 154 mM NaCl (for lysozyme assays) and poured into petri dishes. Wells cored into the cooled agarose received 12 µl of either chicken egg white lysozyme or bOBP at the indicated concentrations. The radius of the lytic zone is plotted against the log of the protein concentration.

 
To substantiate that lysis of the cell wall preparation was the result of enzymatic activity, both the lysozyme and bPGRP-S solutions were boiled for 10 min and rapidly iced before they were used in the lysoplate assay. Boiling of the protein samples reduced lysozyme’s ability to generate a lytic zone by 50-fold. The lytic activity of bPGRP-S was reduced 20- to 30-fold (data not shown). Boiled bPGRP-S was tested in microbicidal assays to explore the relationship between bacterial cell wall lysis and bPGRP-S-mediated killing. Although boiled bPGRP-S lost bacterial cell wall lytic activity, it maintained 100% of its microbicidal activity against Listeria (data not shown), indicating that PGRP-S-mediated killing of bacteria is independent of this lytic activity.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Bovine PGRP-S, an ortholog of human and murine PGRP-S, is secreted from neutrophils, is stored in neutrophil large/dense granules, and is also found associated with bacteria in the phagolysosome. Purification of bPGRP-S (formerly termed bOBP) from bovine blood allows for the evaluation of natural PGRP-S in in vitro assays. This protein is verified to bind to and kill a range of microorganisms in a PGN-independent manner in solutions approximating physiologic conditions. Among the bacterial components tested, bPGRP-S demonstrated the highest affinity for LPS. This PGRP showed significant affinity for LTA and little for PGN in killing inhibition assays.

The first PGRP was named for its ability to bind to Gram-positive bacteria and PGN. As the family of known PGRPs has grown, investigations have shown that members of this protein family recognize a range of microorganisms and cell envelope constituents (4, 5, 9, 10, 17, 25), leading researchers to emphasize the discovery of specific molecular targets for each PGRP analogous to the TLR family (35). However, bPGRP-S recognizes multiple microbial components, as do Holotrichia PGRP-S (15) and Drosophila PGRP-LC (4), suggesting that PGRP recognition of pathogens is less specific than that reported for TLRs. It may be that PGRPs bind common moieties or secondary structures in their target molecules or, alternatively, that they have binding sites for more than one microbial component. The report that Holotrichia PGRP-S binds laminaripentaose, a component of the fungal cell wall (15), is consistent with the hypothesis that PGRPs recognize diverse microbial ligands by conserved or similar oligosaccharide moieties. Unlike the TLRs, mammalian PGRPs have not been shown to act in signaling, although Drosophila PGRPs are involved in both immune signaling cascades (4, 5, 9, 17, 36, 37).

A major question about PGRPs has been the anatomical site in which they are active or stored. Our staining localizes bPGRP-S to large granules, in agreement with differential centrifugation studies of the murine ortholog (12). Large granules of bovine neutrophils, equivalent to the tertiary/dense granules of humans and mice, contain the oxygen-independent bactericidal proteins of these cells (33, 38). Electron microscopy of Immunogold-labeled neutrophils that were exposed to opsonized bacteria in 50/50 HBSS/serum also confirms our data showing that PGRP-S is secreted. Specifically, neutrophils that have phagocytosed either Gram-negative or Gram-positive bacteria show marked loss of their large granules and nearly complete loss of PGRP-S, which is not primarily translocated into the phagolysosome. Neutrophils that phagocytosed bacteria had PGRP-S inside the phagolysosome associated with S. aureus, but not S. typhimurium. The absence of PGRP-S from phagolysosomes containing S. typhimurium is not consistent with our analysis of microbial binding by PGRP-S, suggesting that either PGRP-S has been degraded in the phagolysosome or that the Ab epitope has become unavailable, possibly through interaction with LPS.

Evidence that bPGRP-S is secreted and also present in the phagolysosome may indicate that the protein re-enters the cell bound to susceptible microorganisms. Adding weight to this hypothesis is the fact that the killing capacity of neutrophils from PGRP-S knockout mice was reconstituted to the level of wild-type mice by the addition of exogenous PGRP-S to the incubation medium (12). Our data do not, however, exclude the possibility that PGRP-S might be both secreted and translocated into the phagosome.

Although other PGRPs have shown bacteriostatic activity (6, 12, 39), bPGRP-S is the only PGRP shown to be microbicidal. Dziarski et al. (12) demonstrated that murine PGRP-S participates in the intracellular killing of bacteria in neutrophils. However, they found only inhibition of bacterial growth by protein with a multiple His tag using the salt-free assay described in this study. Recently, Cho et al. (39) have demonstrated that recombinant human PGRP-S cooperates with lysozyme to inhibit bacterial growth. This difference between the measured microbicidal activity of natural bPGRP-S protein and the recombinant human and murine PGRP-S may be due to the chemical modification of the cloned human and mouse PGRPs. A terminal His tag or the salt precipitation method used to purify the recombinant proteins may affect PGRP-S-mediated microbial killing. Microbicidal studies of natural, purified human and mouse PGRPs would address this issue, although it is possible that the primary sequence of the bovine protein (murine and bovine PGRP-S are 64% identical) makes it unique among PGRPs in this function.

Bovine PGRP-S has been demonstrated to kill Gram-positive bacteria, Gram-negative bacteria, and also fungi in vitro (10). Additional microbicidal assays using a range of buffer conditions reveal that although the microbicidal activity of bPGRP-S does not require a hypotonic milieu, killing is reduced by NaCl concentrations as low as 30 mM. Interestingly, even though 150 mM NaCl reduced the binding of bPGRP-S to target bacteria below the level of detection by Western blot analysis, bPGRP-S was still able to kill 90–99% of the bacteria at this salt concentration. This may indicate that very little protein is required for efficient killing, bacterial death causes the release of bound PGRP-S under high-salt, serum-free conditions, or this method of measuring binding is not correlative with killing. Sensitivity to ionic concentration is a quality that bPGRP-S shares with other microbicidal proteins, such as defensins (40, 41). NaCl may interfere with the interaction between bPGRP-S and cell envelope constituents through reducing the electrostatic interaction between the negatively charged microbial surface and the positively charged protein, pI 9.38 (10).

This reduction of bPGRP-S-mediated killing by NaCl could suggest that the protein has another role in vivo, the remaining bactericidal activity (90–99% killing) is adequate for host defense, or the presence of serum ameliorates the salt effect. To address this question we evaluated the binding activity of PGRP-S in physiologic salt buffer with and without serum using opsonized bacteria. We found that opsonization facilitated PGRP-S binding to S. aureus, but not S. typhimurium (Fig. 5). Additional studies of binding suggested that physiologic levels of salts in simple buffers (Tris-HCl or HBSS) interfered with PGRP-S binding to microorganisms, but when fresh serum was used, binding of PGRP-S to S. aureus was enhanced over opsonization alone (Fig. 5). This verifies that bPGRP-S is likely to bind to microorganisms in the extracellular milieu. The serum-rich extracellular milieu contains many components that may be responsible for enhancing target binding. Such factors might also enhance bPGRP-S-mediated killing. One relevant serum constituent may be PGRP-L, recently identified as a component of normal mammalian serum (16). Indeed, other PGRPs, such as PGRP-LE and PGRP-LC of Drosophila, have been found to act synergistically (37).

Physiological levels of divalent salts may play another role in PGRP-S interaction with target organisms. Concentrations of MgCl2 or CaCl2 as low as 0.8 mM ablate bPGRP-S-mediated killing of S. typhimurium (Fig. 3D), but have no effect on killing of Gram-positive L. monocytogenes at concentrations up to 2 mM (Fig. 3C). The moderate reduction of bPGRP-S binding caused by 2 mM MgCl2 (Fig. 4) does not explain this effect, because binding to L. monocytogenes and S. typhimurium is similarly affected. Protection of Gram-negative, but not Gram-positive, bacteria by MgCl2 and CaCl2 suggests that these divalent cations do not act on bPGRP-S directly, but, rather, exert an effect on the LPS outer membrane. This protective effect may be related to the decrease in LPS bilayer fluidity conferred by divalent cations (34). Complete protection of S. typhimurium from the bactericidal activity of bPGRP-S by MgCl2 concentrations that only moderately reduce protein binding indicates that binding alone is not sufficient for PGRP-S-mediated killing of bacteria.

The ability of bovine PGRP-S to bind to intact yeast and Gram-negative bacteria confirms that PGRP-S does not require PGN for recognizing and binding to a range of microorganisms. Although bPGRP-S binds to and kills C. neoformans (10), another fungus, C. albicans, resists antimicrobial activity and is the only organism tested that does not bind PGRP-S in salt-free buffer. This result suggests that microorganisms may escape PGRP-S-mediated killing by avoiding binding.

The range of microbial cell envelope constituents that may bind PGRP-S was explored by testing a variety of bacterial components for their ability to inhibit bPGRP-S-mediated killing of Gram-positive and -negative bacteria. Not only was inhibition of killing not selective for PGN or its primary constituent, MDP, but PGN was found to be a poor inhibitor of killing. In fact, LPS (S. typhimurium) and LTA (B. subtilis) were up to 100-fold more potent inhibitors of bacterial killing than PGN from S. aureus on a molar basis. These findings suggest that binding of PGRP-S to Gram-positive bacteria might be mediated as much by surface-exposed teichoic acids, which constitute 10–50% of the mass of the Gram-positive cell wall (42), as by PGN. The fact that LTA effectively inhibited killing of Gram-positive L. monocytogenes yet had no effect in assays with Gram-negative S. typhimurium suggests that bovine PGRP-S has greater affinity for LPS than for LTA. The incomplete inhibition of bPGRP-S-mediated killing of S. typhimurium by LPS, a molecule also present on the surface of these bacteria, is expected in a competition assay.

The interaction between the PGRP binding domain(s) and microbial constituents are beginning to be elucidated. The affinity of bPGRP-S for LPS, LTA, and an unidentified fungal constituent is not in conflict with any published data, although the mechanism of this interaction is not immediately apparent. The microbial molecules that interact with bPGRP-S share only limited oligosaccharide moieties, but may share significant structural similarities. Analysis of the PGRP-LB crystal structure indicates that poor conservation of amino acids predicted to line the binding cleft would generate widely varying specificities of individual PGRPs (21, 43). This variation may explain the affinities of various PGRPs to ligands other than PGN, but it does not explain the interaction with multiple ligands by bPGRP-S, Drosophila PGRP-LC (4), and Holotrichia PGRP-S (15). The prozone-like kinetics of bPGRP-S precipitation of LPS suggests that the protein has a specific affinity to LPS and adds weight to structural data suggesting that PGRPs function as multimers (21) or are comprised of tandem binding domains (26). The superior efficiency, judged by their molar ratio, of LPS precipitation at a higher PGRP-S concentration may be due to improved protein dimerization.

Some PGRPs have been shown to have enzymatic activity. Kim et al. (21) have predicted that a bound catalytic zinc ion was essential for activity. Although this appeared to explain zinc-dependent amidase activity in some PGRPs (16, 20, 21, 22, 23), it did not predict the zinc-independent L,D-carboxypeptidase activity of Drosophila PGRP-SA (24). Like Drosophila PGRP-SA, bPGRP-S has a serine substitution at a cysteine residue predicted to stabilize a zinc ion. The bacterial cell wall lytic activity of bPGRP-S demonstrates that the findings of Chang et al. (24) are not anomalous and that, contrary to the prevailing hypothesis, mammalian PGRPs do not require stabilization of a zinc ion for enzymatic activity. Verification that these two proteins lyse PGN in HEPES, a salt-free buffer, suggests that other short PGRPs might be analyzed under similar conditions to determine whether they exhibit lytic activity.

Bovine PGRP-S appears to have multiple functions and microbial affinities. A secreted protein, bPGRP-S, is stored in neutrophil large/dense granules and is also found associated with bacteria in the phagolysosome. Bovine PGRP-S binds to a range of microbial components and kills diverse microorganisms. This PGRP has greater affinity for LPS and LTA than for PGN or MDP. Although bPGRP-S digests bacterial cell walls, bPGRP-S-mediated killing of Listeria appears to be independent of the protein’s cell wall lytic activity. Our analysis of bPGRP-S adds to the building evidence that PGRPs, rather than being a collection of proteins with highly selective affinities (in the manner of TLRs), may be generalists in their antimicrobial binding and activity.


   Acknowledgments
 
We acknowledge the assistance of Patrick Koen and Jean Edens (Caltech Electron Microscope Facility) in performing the electron microscopy. We are most indebted to Ellen V. Rothenberg, Rochelle Diamond, Marianne Bronner-Fraser, Mary Yui, and Jon Moore for critical reading of the manuscript.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by National Institutes of Health Grants A122931 and AI54699-01 and the Hewitt Foundation. Back

2 Address correspondence and reprint requests to Dr. C. Chace Tydell, Division of Biology, California Institute of Technology, Pasadena, CA 91125. E-mail address: chacetyd{at}caltech.edu Back

3 Abbreviations used in this paper: PGN, peptidoglycan; PGRP, peptidoglycan recognition protein; bOBP, bovine oligosaccharide binding protein; bPGRP, bovine PGRP; LTA, lipoteichoic acid; MDP, muramyl dipeptide; PGRP-I, intermediate PGRP; PGRP-L, long PGRP; PGRP-S, short PGRP; rb, recombinant bovine; HOAc, acetic acid. Back

Received for publication September 8, 2005. Accepted for publication October 28, 2005.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Hoffmann, J. A., J. M. Reichhart. 2002. Drosophila innate immunity: an evolutionary perspective. Nat. Immunol. 3: 121-126. [Medline]
  2. Kimbrell, D. A., B. Beutler. 2001. The evolution and genetics of innate immunity. Nat. Rev. Genet. 2: 256-267. [Medline]
  3. Janeway, C. A., Jr. 1989. Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harbor Symp. Quant. Biol. 54: 1-13.
  4. Choe, K. M., T. Werner, S. Stoven, D. Hultmark, K. V. Anderson. 2002. Requirement for a peptidoglycan recognition protein (PGRP) in Relish activation and antibacterial immune responses in Drosophila. Science 296: 359-362. [Abstract/Free Full Text]
  5. Gottar, M., V. Gobert, T. Michel, M. Belvin, G. Duyk, J. A. Hoffmann, D. Ferrandon, J. Royet. 2002. The Drosophila immune response against Gram-negative bacteria is mediated by a peptidoglycan recognition protein. Nature 416: 640-644. [Medline]
  6. Liu, C., E. Gelius, G. Liu, H. Steiner, R. Dziarski. 2000. Mammalian peptidoglycan recognition protein binds peptidoglycan with high affinity, is expressed in neutrophils, and inhibits bacterial growth. J. Biol. Chem. 275: 24490-24499. [Abstract/Free Full Text]
  7. Liu, C., Z. Xu, D. Gupta, R. Dziarski. 2001. Peptidoglycan recognition proteins: a novel family of four human innate immunity pattern recognition molecules. J. Biol. Chem. 276: 34686-34694. [Abstract/Free Full Text]
  8. Ochiai, M., M. Ashida. 1999. A pattern recognition protein for peptidoglycan: cloning the cDNA and the gene of the silkworm, Bombyx mori. J. Biol. Chem. 274: 11854-11858. [Abstract/Free Full Text]
  9. Ramet, M., P. Manfruelli, A. Pearson, B. Mathey-Prevot, R. A. Ezekowitz. 2002. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416: 644-648. [Medline]
  10. Tydell, C. C., N. Yount, D. Tran, J. Yuan, M. E. Selsted. 2002. Isolation, characterization, and antimicrobial properties of bovine oligosaccharide-binding protein: a microbicidal granule protein of eosinophils and neutrophils. J. Biol. Chem. 277: 19658-19664. [Abstract/Free Full Text]
  11. Hoffmann, J. A.. 2003. The immune response of Drosophila. Nature 426: 33-38. [Medline]
  12. Dziarski, R., K. A. Platt, E. Gelius, H. Steiner, D. Gupta. 2003. Defect in neutrophil killing and increased susceptibility to infection with nonpathogenic Gram-positive bacteria in peptidoglycan recognition protein-S (PGRP-S)-deficient mice. Blood 102: 689-697. [Abstract/Free Full Text]
  13. Kang, D., G. Liu, A. Lundstrom, E. Gelius, H. Steiner. 1998. A peptidoglycan recognition protein in innate immunity conserved from insects to humans. Proc. Natl. Acad. Sci. USA 95: 10078-10082. [Abstract/Free Full Text]
  14. Yoshida, H., K. Kinoshita, M. Ashida. 1996. Purification of a peptidoglycan recognition protein from hemolymph of the silkworm, Bombyx mori. J. Biol. Chem. 271: 13854-13860. [Abstract/Free Full Text]
  15. Lee, M. H., T. Osaki, J. Y. Lee, M. J. Baek, R. Zhang, J. W. Park, S. Kawabata, K. Soderhall, B. L. Lee. 2004. Peptidoglycan recognition proteins involved in 1,3-{beta}-D-glucan-dependent prophenoloxidase activation system of insect. J. Biol. Chem. 279: 3218-3227. [Abstract/Free Full Text]
  16. Xu, M., Z. Wang, R. M. Locksley. 2004. Innate immune responses in peptidoglycan recognition protein L-deficient mice. Mol. Cell. Biol. 24: 7949-7957. [Abstract/Free Full Text]
  17. Michel, T., J. M. Reichhart, J. A. Hoffmann, J. Royet. 2001. Drosophila Toll is activated by Gram-positive bacteria through a circulating peptidoglycan recognition protein. Nature 414: 756-759. [Medline]
  18. Rehman, A., P. Taishi, J. Fang, J. A. Majde, J. M. Krueger. 2001. The cloning of a rat peptidoglycan recognition protein (PGRP) and its induction in brain by sleep deprivation. Cytokine 13: 8-17. [Medline]
  19. Kappeler, S. R., C. Heuberger, Z. Farah, Z. Puhan. 2004. Expression of the peptidoglycan recognition protein, PGRP, in the lactating mammary gland. J. Dairy Sci. 87: 2660-2668. [Abstract/Free Full Text]
  20. Mellroth, P., J. Karlsson, H. Steiner. 2003. A scavenger function for a Drosophila peptidoglycan recognition protein. J. Biol. Chem. 278: 7059-7064. [Abstract/Free Full Text]
  21. Kim, M. S., M. Byun, B. H. Oh. 2003. Crystal structure of peptidoglycan recognition protein LB from Drosophila melanogaster. Nat. Immunol. 4: 787-793. [Medline]
  22. Wang, Z. M., X. Li, R. R. Cocklin, M. Wang, K. Fukase, S. Inamura, S. Kusumoto, D. Gupta, R. Dziarski. 2003. Human peptidoglycan recognition protein-L is an N-acetylmuramoyl-L-alanine amidase. J. Biol. Chem. 278: 49044-49052. [Abstract/Free Full Text]
  23. Gelius, E., C. Persson, J. Karlsson, H. Steiner. 2003. A mammalian peptidoglycan recognition protein with N-acetylmuramoyl-L-alanine amidase activity. Biochem. Biophys. Res. Commun. 306: 988-994. [Medline]
  24. Chang, C. I., S. Pili-Floury, M. Herve, C. Parquet, Y. Chelliah, B. Lemaitre, D. Mengin-Lecreulx, J. Deisenhofer. 2004. A Drosophila pattern recognition receptor contains a peptidoglycan docking groove and unusual L,D-carboxypeptidase activity. PLoS Biol. 2: E277[Medline]
  25. Werner, T., G. Liu, D. Kang, S. Ekengren, H. Steiner, D. Hultmark. 2000. A family of peptidoglycan recognition proteins in the fruit fly Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 97: 13772-13777. [Abstract/Free Full Text]
  26. Guan, R., E. L. Malchiodi, Q. Wang, P. Schuck, R. A. Mariuzza. 2004. Crystal structure of the C-terminal peptidoglycan-binding domain of human peptidoglycan recognition protein I{alpha}. J. Biol. Chem. 279: 31873-31882. [Abstract/Free Full Text]
  27. Lo, D., W. Tynan, J. Dickerson, J. Mendy, H. W. Chang, M. Scharf, D. Byrne, D. Brayden, L. Higgins, C. Evans, et al 2003. Peptidoglycan recognition protein expression in mouse Peyer’s patch follicle associated epithelium suggests functional specialization. Cell. Immunol. 224: 8-16. [Medline]
  28. Uehara, A., Y. Sugawara, S. Kurata, Y. Fujimoto, K. Fukase, S. Kusumoto, Y. Satta, T. Sasano, S. Sugawara, H. Takada. 2005. Chemically synthesized pathogen-associated molecular patterns increase the expression of peptidoglycan recognition proteins via toll-like receptors, NOD1 and NOD2 in human oral epithelial cells. Cell. Microbiol. 7: 675-686. [Medline]
  29. Yount, N. Y., J. Yuan, A. Tarver, T. Castro, G. Diamond, P. A. Tran, J. N. Levy, C. McCullough, J. S. Cullor, C. L. Bevins, et al 1999. Cloning and expression of bovine neutrophil {beta}-defensins: biosynthetic profile during neutrophilic maturation and localization of mature peptide to novel cytoplasmic dense granules. J. Biol. Chem. 274: 26249-26258. [Abstract/Free Full Text]
  30. Selsted, M. E., Y. Q. Tang, W. L. Morris, P. A. McGuire, M. J. Novotny, W. Smith, A. H. Henschen, J. S. Cullor. 1993. Purification, primary structures, and antibacterial activities of {beta}-defensins, a new family of antimicrobial peptides from bovine neutrophils. J. Biol. Chem. 268: 6641-6648. [Abstract/Free Full Text]
  31. Selsted, M. E., D. Szklarek, R. I. Lehrer. 1984. Purification and antibacterial activity of antimicrobial peptides of rabbit granulocytes. Infect. Immun. 45: 150-154. [Abstract/Free Full Text]
  32. Osserman, E. F., D. P. Lawlor. 1966. Serum and urinary lysozyme (muramidase) in monocytic and monomyelocytic leukemia. J. Exp. Med. 124: 921-952. [Abstract]
  33. Gennaro, R., B. Dewald, U. Horisberger, H. U. Gubler, M. Baggiolini. 1983. A novel type of cytoplasmic granule in bovine neutrophils. J. Cell Biol. 96: 1651-1661. [Abstract/Free Full Text]
  34. Snyder, S., D. Kim, T. J. McIntosh. 1999. Lipopolysaccharide bilayer structure: effect of chemotype, core mutations, divalent cations, and temperature. Biochemistry 38: 10758-10767. [Medline]
  35. Beutler, B., K. Hoebe, X. Du, R. J. Ulevitch. 2003. How we detect microbes and respond to them: the Toll-like receptors and their transducers. J. Leukocyte Biol. 74: 479-485. [Abstract/Free Full Text]
  36. Bischoff, V., C. Vignal, I. G. Boneca, T. Michel, J. A. Hoffmann, J. Royet. 2004. Function of the drosophila pattern-recognition receptor PGRP-SD in the detection of Gram-positive bacteria. Nat Immunol. 5: 1175-1180. [Medline]
  37. Takehana, A., T. Yano, S. Mita, A. Kotani, Y. Oshima, S. Kurata. 2004. Peptidoglycan recognition protein (PGRP)-LE and PGRP-LC act synergistically in Drosophila immunity. EMBO J. 23: 4690-4700. [Medline]
  38. Gennaro, R., L. Dolzani, D. Romeo. 1983. Potency of bactericidal proteins purified from the large granules of bovine neutrophils. Infect. Immun. 40: 684-690. [Abstract/Free Full Text]
  39. Cho, J. H., I. P. Fraser, K. Fukase, S. Kusumoto, Y. Fujimoto, G. L. Stahl, R. A. Ezekowitz. 2005. Human peptidoglycan recognition protein-S is an effector of neutrophil-mediated innate immunity. Blood 106: 2551-2558. [Abstract/Free Full Text]
  40. Bals, R., M. J. Goldman, J. M. Wilson. 1998. Mouse {beta}-defensin 1 is a salt-sensitive antimicrobial peptide present in epithelia of the lung and urogenital tract. Infect. Immun. 66: 1225-1232. [Abstract/Free Full Text]
  41. Goldman, M. J., G. M. Anderson, E. D. Stolzenberg, U. P. Kari, M. Zasloff, J. M. Wilson. 1997. Human {beta}-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis. Cell 88: 553-560. [Medline]
  42. R. W. Wheat, Jr, ed. Composition, Structure, and Biosynthesis of Bacterial Cell Envelope and Energy Storage Polymers 1992 Appleton and Lange, East Norwalk.
  43. Guan, R., Q. Wang, E. J. Sundberg, R. A. Mariuzza. 2005. Crystal structure of human peptidoglycan recognition protein S (PGRP-S) at 1.70 A resolution. J. Mol. Biol. 347: 683-691. [Medline]



This article has been cited by other articles:


Home page
 BloodHome page
L. P. Sashchenko, E. A. Dukhanina, Y. V. Shatalov, D. V. Yashin, T. I. Lukyanova, O. D. Kabanova, E. A. Romanova, S. V. Khaidukov, A. V. Galkin, N. V. Gnuchev, et al.
Cytotoxic T lymphocytes carrying a pattern recognition protein Tag7 can detect evasive, HLA-negative but Hsp70-exposing tumor cells, thereby ensuring FasL/Fas-mediated contact killing
Blood, September 15, 2007; 110(6): 1997 - 2004.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
S. Cho, Q. Wang, C. P. Swaminathan, D. Hesek, M. Lee, G.-J. Boons, S. Mobashery, and R. A. Mariuzza
Structural insights into the bactericidal mechanism of human peptidoglycan recognition proteins
PNAS, May 22, 2007; 104(21): 8761 - 8766.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
M. Wang, L.-H. Liu, S. Wang, X. Li, X. Lu, D. Gupta, and R. Dziarski
Human Peptidoglycan Recognition Proteins Require Zinc to Kill Both Gram-Positive and Gram-Negative Bacteria and Are Synergistic with Antibacterial Peptides
J. Immunol., March 1, 2007; 178(5): 3116 - 3125.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
X. Li, S. Wang, H. Wang, and D. Gupta
Differential Expression of Peptidoglycan Recognition Protein 2 in the Skin and Liver Requires Different Transcription Factors
J. Biol. Chem., July 28, 2006; 281(30): 20738 - 20748.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if