Gene Silencing and Overexpression of Porcine Peptidoglycan Recognition Protein Long Isoforms: Involvement in -Defensin-1 Expression
http://www.100md.com
感染与免疫杂志 2005年第11期
Department of Anatomy and Physiology, Kansas State University, Manhattan, Kansas 66506
ABSTRACT
Peptidoglycan recognition proteins (PGRPs) are a group of newly identified proteins with emerging functions in mammalian innate immunity. Here we report the identification and characterization of two long isoforms of porcine PGRP. Their complete cDNA sequences encode predicted peptides of 252 and 598 residues and are named pPGRP-L1 and pPGRP-L2, respectively. These porcine isoforms share identical PGRP domains at their C terminus, which are highly conserved with human and mouse orthologs. pPGRP-L1 is expressed constitutively in several tissues, including bone marrow, intestine, liver, spleen, kidney, and skin. pPGRP-L2 is highly expressed in the duodenum and liver, and expression in intestinal tissues is increased by Salmonella infection. In intestinal cells, expression of both pPGRP-L1 and pPGRP-L2 is increased by bacterial infection. Recombinant pPGRP-L1 and pPGRP-L2 have N-acetylmuramoyl-L-alanine amidase activity. Loss-of-function and gain-of-function experiments indicate that these two pPGRPs are involved in expression of the antimicrobial peptide -defensin-1. Silencing of pPGRP-L2 in intestinal cells challenged with Listeria monocytogenes results in downregulation of -defensin-1. Conversely, overexpression of pPGRP-L1 or pPGRP-L2 dramatically upregulates expression of -defensin-1. Collectively, these findings suggest that porcine PGRPs are involved in antimicrobial peptide expression.
INTRODUCTION
Innate immunity provides the first line of defense against infectious disease (1, 3, 16). The cells and effector molecules responsible for innate immunity have the essential role of identifying and eliminating microbial pathogens and signaling the adaptive immune system of their presence. The first critical step in initiating these responses is recognition of the invading pathogen. This recognition occurs via protein receptors that recognize motifs or patterns on microorganisms that are not present on eukaryotic cells (27, 33). In addition to the well-known Toll receptors and Toll-like receptors (TLR), peptidoglycan recognition proteins (PGRPs) represent another group of pathogen-binding proteins, and some PGRP isoforms in Drosophila have been shown to mediate the production of antimicrobial peptides (30, 39).
The first PGRPs were identified in silkworms (Bombyx mori) and moths (Trichoplusia ni) through their high-affinity binding to peptidoglycan and were implicated in innate immunity because of their ability to activate the protease cascade that leads to phenoloxidase activation in response to peptidoglycan (17, 47). To date, data from the fruit fly (Drosophila melanogaster) genome project have revealed 13 PGRP homologs on three chromosomes (7, 18). Based on their predicted peptide sequences, they are classified into two groups: seven short PGRPs (PGRP-S), which are predicted to be small extracellular proteins that are exported after removal of an N-terminal signal peptide, and six long PGRPs (PGRP-L), which encode intracellular or membrane-spanning proteins (7, 18). It is now clear that some of these pattern recognition proteins respond to several microbial species via interactions with peptidoglycan and mediate the production of antimicrobial peptides through the Toll or immune deficiency (IMD) pathways. For example, PGRP-SA and -SD respond to different groups of gram-positive bacteria (4, 12, 29), and the PGRP-LC gene was discovered through its induction of antimicrobial genes after a gram-negative bacterial infection (6, 12, 13, 32). In addition, PGRP-LE stimulated the IMD pathway through interaction with diaminopimelic acid-type peptidoglycan, which is present in many gram-negative and some gram-positive bacteria (40). Loss-of-function studies, using RNA interference, showed that PGRP-LC is involved in phagocytosis of gram-negative Escherichia coli but not gram-positive Staphylococcus aureus (6, 12, 13, 32). Gain-of-function studies showed that overexpression of PGRP-LC or PGRP-LE upregulated the production of insect antimicrobial peptides (6, 40). In addition, some Drosophila PGRP isoforms, such as PGRP-SC1B, have N-acetylmuramoyl-L-alanine amidase activity (28), and PGRP-SA has L-carboxypeptidase activity (5).
Several PGRPs have also been identified in mammals. Four PGRPs have been identified in humans and mice: one long, one short, and two mammalian-specific intermediate isoforms (7, 23, 24, 26). In addition to binding peptidoglycan, murine PGRP-S, which is expressed in neutrophils, inhibits growth of gram-positive bacteria and therefore has been suggested to function as a neutrophil intracellular antibacterial protein (8). Studies using PGRP-S knockout mice indicate that murine PGRP-S is involved in clearance of nonpathogenic gram-positive bacteria but not pathogenic gram-positive or gram-negative bacteria (8). Similarly, the bovine PGRP ortholog has antimicrobial activity against gram-positive and -negative bacteria and yeasts (41). Like Drosophila PGRP-SC1B, human and mouse PGRP long isoforms have been shown to function as N-acetylmuramoyl-L-alanine amidases (11, 28, 42). In addition, peritoneal macrophages from PGRP-L-deficient mice produce less interleukin 6 and tumor necrosis factor alpha when stimulated with E. coli or lipopolysaccharide (46). Collectively, these findings indicate that PGRPs are involved in mammalian innate immune responses. However, unlike the case for Drosophila, the association between recognition of the bacterial threat by PGRPs and elicitation of an antimicrobial peptide response has not been established for mammals. Here we report the identification and characterization of porcine PGRP long isoforms and show, by silencing and overexpressing these proteins, that they are associated with the expression of the antimicrobial peptide -defensin-1.
MATERIALS AND METHODS
Cell culture and reagents. Porcine intestinal cells (IPEC-J2) were grown in Dulbecco-Ham F-12 medium (1:1, vol/vol) containing 5% fetal bovine serum and supplemented with insulin (10 μg/ml), transferrin (5.5 μg/ml), sodium selenite (0.67 μg/ml), and epidermal growth factor (5 ng/ml) (GIBCO Invitrogen Corp., Gaithersburg, MD) (34). Cells were maintained in a 95% air-5% CO2 humidified atmosphere at 37°C and subcultured at 4-day intervals.
Gene identification and analysis. To begin our characterization of the porcine PGRP family, we first screened the porcine expressed sequence tag (EST) database at GenBank, using human and bovine PGRP sequences (GenBank accession numbers AF384856, AY035376, AY035377, and AY083309) (24, 41). Two porcine EST sequences (accession numbers BE013528 and BF442868) that contained PGRP-homologous sequences in their translated regions were identified. Two pairs of primers were designed based on the EST sequences (outer sense, 5'-GTA GCC CAA AGC CAC TGA AG-3'; outer antisense, 5'-GAG TTG TGG TCA CGT GTG T-3'; inner sense, 5'-AAG CCA CTG AAG CTG CCA CT-3'; inner antisense, 5'-ATC TGA GCC CAC CAC GAA AC-3'). To obtain the full-length cDNA sequences, we used a modified rapid amplification of cDNA ends (RACE) (Ambion FirstChoice RLM-RACE kit) protocol, which selects for nontruncated 5'-capped mRNAs, and adopted a similar nested PCR strategy to obtain the 5'- and 3'-terminal cDNA sequences from total RNA purified from porcine liver. The nested PCR products were purified with a column-based PCR purification kit (QIAGEN, Inc., Valencia, CA) and cloned into plasmids with pGEM-T Easy Vector systems (Promega Co., Madison, WI), followed by transformation into E. coli (JM109; Stratagene Co., La Jolla, CA) and colony screening with PCR. Primers used for clone screening PCR were the sense and antisense gene-specific primers. Plasmids were extracted from bacterial cultures derived from the identified single colonies, and inserts were sequenced with SP6/T7 vector primers on an ABI 3700 DNA analyzer at the Kansas State University Sequencing and Genotyping Facility (Manhattan, KS). Each cDNA sequence was generated from the sequencing results for at least five identical plasmid extracts from individual colonies. The full-length cDNA sequence was then generated by ligation of 5' and 3' sequences and deletion of the adaptor and overlapped regions. Open reading frames (ORFs) were predicted with ORFfinder and Genescan (http://www.ncbi.nlm.nih.gov/gorf/orfig.cgi and http://genes.mit.edu/GENSCAN.html). Translated peptide sequences were analyzed using SignalP and PSORT (ExPASy; http://au.expasy.org/tools/#proteome) for signal peptide and cellular location predictions, respectively.
Expression in tissue. Tissues were obtained from healthy and Salmonella-challenged 5-week-old pigs as previously described (45, 48, 49), using procedures approved by the Kansas State University Institutional Animal Care and Use Committee. Tissue samples were collected, placed immediately in liquid nitrogen, and stored at –135°C until use. Total RNA was extracted with TRI reagent (Sigma-Aldrich, St. Louis, MO) after grinding the frozen tissues in liquid nitrogen. A one-step reverse transcription-PCR (RT-PCR) was used to detect expression of target transcripts. Briefly, total RNA was treated with RQ1 RNase-free DNase I (Promega), and RNA samples (250 ng) were then used in a 25-μl RT-PCR mixture with 0.1 μM of each sense and antisense primer designed to distinguish pPGRP-L1 (sense, 5'-CCC AGG TTC CTT CTT GGA TT-3'; antisense, 5'-GTG GAA GCG CTG CAT AGA G-3'; cDNA nucleotides 51 to 70 and 551 to 533, respectively) and pPGRP-L2 (sense, 5'-GCC CAG ACT GAA AGC AAT TC-3'; antisense, 5'-AGC CCA CCA CGA AAC TGT AG-3'; cDNA nucleotides 848 to 867 and 1502 to 1483, respectively). Conventional one-step RT-PCR was performed using an AccessQuick RT-PCR system kit (Promega). cDNA synthesis and predenaturation were performed for one cycle at 48°C for 45 min and 95°C for 3 min, respectively. Amplification was carried out for 40 or 25 (glyceraldehyde 3-phosphate dehydrogenase [GAPDH]) cycles at 95°C for 30 s, 55°C for 30 s, and 72°C for 30 s, and final extension was at 72°C for 10 min in a 25-μl reaction mixture. After amplification, 10 μl of each reaction mixture was analyzed by 2% agarose gel electrophoresis, and bands were visualized by ethidium bromide staining. For Southern analysis, PCR products were transferred onto Hybond N+ filters (Amersham Pharmacia Biotech), hybridized with PGRP cDNA probes labeled with [-32P]dCTP (3,000 Ci/mmol), and detected using autoradiography as previously described (48).
Bacterial infection and real-time RT-PCR. Bacteria used were Escherichia coli (ATCC 25922), Listeria monocytogenes (ATCC 19115), and Salmonella enterica serovar Typhimurium (ATCC BAA-185). Cells cultured in 24-well plates were infected by adding medium containing 104 CFU/ml at 0.25 ml per well. Plates were centrifuged at 800 x g for 10 min and incubated for 2 h at 37°C in a CO2 incubator. Cells then were washed with medium containing 100 IU penicillin and 10 μg/μl streptomycin solution to remove extracellular bacteria. The cells in replicate wells were cultured and sampled through 2 to 24 h, and total RNA was extracted using Trizol reagent (45).
Real-time quantitative RT-PCR was performed on a SmartCycler (Cepheid, Sunnyvale, CA) as previously described (36) with a real-time one-step RT-PCR kit (QIAGEN, Valencia, CA). Gene-specific primers for pPGRP-L1 were sense, 5'-CTG TGG TAG CAG CAG CTT CTC TTT-3', and antisense, 5'-GCT GCA GAA GCA ATC CAT ACA GGA-3' (cDNA nucleotides 12 to 35 and 256 to 233), and those for pPGRP-L2 were sense, 5'-AGC CCA GAG TTT CAA GGC CTG ATT-3', and antisense, 5'-TCA GGA ACT GTG GCT CCA ATG TCT-3' (cDNA nucleotides 368 to 391 and 602 to 579). Primers for -defensin-1 were as described previously (48). DNase-treated total RNA (500 ng) was used in each 25-μl RT-PCR mixture. RT-PCR cycling conditions were 30 min at 50°C and 15 min at 95°C, followed by 45 cycles of 15 s at 95°C, 30 s at 56°C, and 30 s at 72°C with the optic on for fluorescence data collection. Specific amplification was confirmed by sequencing the amplicons before performing real-time RT-PCR. Threshold cycle (CT) values were determined by exponential product amplification and subsequent increased fluorescence intensity above background. Relative gene expression data were normalized against CT values for the housekeeping (GAPDH) gene, and the relative index () was determined by comparison to the average expression levels for control samples with the index defined as 1.0 (25).
Silencing of PGRP expression. Small interfering RNAs (siRNAs) were generated (QIAGEN-Xeragon) against the isoform-specific and -identical regions. Preliminary experiments showed that only two siRNAs from isoform-identical regions exhibited significant suppression with isoform preference. The template sequences of these two siRNAs are 5'-AAT GTG AAA CCA AGA ACT GCC-3' (cDNA nucleotides 885 to 905 and 1787 to 1807 for pPGRP-L1 and -L2, respectively) and 5'-AAA CTG GAA GAC CAC AAT GGT-3' (cDNA nucleotides 169 to 189 and 91 to 111 for pPGRP-L1 and -L2, respectively). To optimize the transformation procedure and for a negative control, we used a rhodamine-labeled scrambled siRNA (5'-AAT TCT CCG AAC GTG TCA CGT-3'). Porcine intestinal cells (IPEC-J2 cells at 5 x 104 cells/well in a 24-well plate) at 60% confluence were transfected with siRNAs by using QIAGEN siRNA transfection reagent. After 24 h, cells (except control wells) were infected with Listeria monocytogenes (104 CFU/ml at 0.25 ml per well) for 2 h at 37°C and then washed with Hanks' balanced salt solution containing 1x antibiotic/antimycotic solution (GIBCO) to remove extracellular bacteria (21). siRNA transfection mixtures were added back to the appropriate wells, and cells were cultured for another 24 h. Cells were then collected, and RNA was isolated for one-step RT-PCR assay with gene-specific primers.
Overexpression of PGRP and cellular localization. Open reading frames of pPGRP-L1 and pPGRP-L2, were generated from cDNA clones with HindIII and BamHI cloning sites at their 5' and 3' termini, respectively, by PCR. Inserts were subcloned into HindIII- and BamHI-digested and phosphatase-treated pEGFP-C3 (Clontech) (2). E. coli (JM109) transformed with the constructs were selected with kanamycin (30 μg/ml; Sigma), and positive clones were identified by PCR. Purified plasmids were then used to transfect the porcine cell lines. In brief, control pEGFP-C3, PGRP-L1, or PGRP-L2 ORF plasmid constructs (2 μg) were introduced into cells at 60 to 80% confluence in six-well plates by using Fugene6 transfection reagent (Roche Diagnostics). All DNA-Fugene6 incubations were optimized at a ratio of 1 μg of DNA to 2 μl of Fugene6 according to the manufacturer's recommended protocol. Transformation efficiency was determined by enhanced green fluorescent protein (EGFP) expression using inverted fluorescence microscopy. After transformation (24 h), cells were collected and transformants were sorted by fluorescence-activated flow cytometry as previously described (31). Overexpression of control EGFP and EGFP-fused pPGRP-L1 or pPGRP-L2 was confirmed by both RT-PCR and immunoblotting. Primers for EGFP were derived from the pEGFP-C3 sequence (GenBank accession number U57607) and are 5'-TCT TCT TCA AGG ACG ACG GCA ACT-3' (sense), and 5'-TGT GGC GGA TCT TGA AGT TCA CCT-3' (antisense). For immunoblotting, proteins were extracted with a mammalian protein extraction reagent (Pierce). Briefly, cells were lysed (0.4 ml of extraction reagent/106 cells) and centrifuged at 12,000 x g for 10 min, and proteins in the supernatant, which include cytoplasmic, nuclear, and some membrane-associated proteins, from 105 EGFP-positive cells were separated with 4 to 20% precast sodium dodecyl sulfate (SDS)-polyacrylamide gels (Pierce) and transferred onto polyvinylidene difluoride blots. EGFP and EGFP-fused proteins were then detected with monoclonal anti-EGFP antibodies (1:8,000; Clontech) and visualized using a color development reaction catalyzed by alkaline phosphatase-conjugated secondary antibodies.
To determine the cellular localization of the EGFP-tagged pPGRPs, porcine intestinal cells were cultured and transfected as described above in a four-well culture slide (BD Falcon). Twenty-four hours after transfection, slides with >20% EGFP-positive cells were fixed and examined using confocal laser scanning fluorescence microscopy. In brief, media were removed, cells were washed once with Dulbecco's phosphate-buffered saline (DPBS) (pH 7.4), and 1 ml of 4% paraformaldehyde-DPBS was added to each well. Cells were fixed for 30 min at 22°C, nuclei were counterstained with Hoechst 33258 (0.12 μg/ml), and cells were rinsed twice with DPBS. Slides were mounted with a drop of ProLong antifade solution (Molecular Probes) and examined with a confocal laser scanning microscope (LSM 510; Zeiss).
Peptidoglycan binding assay. EGFP and EGFP-recombinant pPGRP-L1 and pPGRP-L2 were purified on anti-GFP-conjugated agarose bead affinity columns (Vector Lab Inc., Burlingame, CA) and evaluated on Western blots using monoclonal anti-EGFP antibodies as described previously (11, 17). Insoluble peptidoglycan (0.3 mg; Sigma) from S. aureus was incubated with 10 μg of EGFP or EGFP-tagged pPGRP-L1 and pPGRP-L2 in a 300-μl reaction mixture buffered by 20 mM Tris-HCl saline (pH 7.9). After rocking incubation at 4°C for 5 h, bound and unbound proteins were separated by centrifugation. Proteins were resolved from peptidoglycan by boiling in 2x SDS-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer and subjected to 4 to 20% SDS-PAGE. Proteins were detected with anti-EGFP antibodies on Western blots.
N-Acetylmuramoyl-L-alanine amidase activity. N-Acetylmuramoyl-L-alanine amidase activity was measured as described for murine PGRP-L and Drosophila PGRP-SC1B (11, 15). Briefly, peptidoglycan (50 mg, intact or digested with lysozyme) (Merck) (15) was incubated with 20 μg/ml recombinant pPGRPs in a reaction mixture containing 1 mM ZnSO4 20 mM MgCl2, and 50 mM Tris-HCl, pH 7.9, at 37°C. Samples (200 μl) were removed at each time point, brought to a 0.5-ml volume with 1.0 M NaOH, and incubated at 38°C for 30 min. Fifty microliters of 0.5 M H2SO4 and 5 ml of concentrated H2SO4 were added, and the samples were placed in a boiling water bath for 5 min. Samples were cooled, and 0.05 ml of 4% (wt/vol) CuSO4 · 5H2 O and 0.1 ml of 1.5% (wt/vol) p-hydroxydiphenyl (in 95% ethanol) were added. Samples were then incubated at 30°C for 30 min, and absorbance was determined at 560 nm. A standard curve consisting of 0 to 20 μg of muramic acid (Sigma) was used in the assay (22).
Nucleotide sequence accession numbers. The cDNA sequences reported in this paper have been registered in the GenBank/EBI data bank with accession numbers AF541955 (pPGRP-L1) and AF541956 (pPGRP-L2).
RESULTS
cDNA and amino acid sequences of pPGRP isoforms. Using 5'- and 3'-RACE in combination with primer pairs generated from a porcine EST clone, we identified two PGRP candidates from porcine fetal liver RNA. The cDNA of the first porcine PGRP is 1,008 bp (pPGRP-L1, GenBank accession number AF541955), and the longer isoform is 1,910 bp (pPGRP-L2, GenBank accession number AF541956). Comparison of these two sequences explained why two cDNA sequences were amplified with these primers; 722 bp of their 3'-terminal sequences are identical (Fig. 1A), and our gene-specific RACE primers are matched within this identical region. However, pPGRP-L1 has a unique 80-bp sequence at its 5' terminus, and pPGRP-L2 has an additional 982-bp insert incorporated into its coding region. These two cDNA sequences encode predicted peptide sequences of 252 and 598 amino acid residues, respectively (Fig. 1B). The C-terminal 242 amino acids of the two peptides, which contain the conserved PGRP domains, are identical. In addition, pPGRP-L1 has 10 and pPGRP-L2 has 356 unique amino acids at their N termini. Computational predications suggest that pPGRP-L1 is a cytoplasmic protein and pPGRP-L2 is a transmembrane protein. Porcine PGRP-L2 contains one obvious transmembrane helix at its N-terminal 55 to 76 amino acid residues. The predicted signal peptide region of pPGRP-L2 is located before the N-terminal 31 amino acid residues.
Species comparison. Porcine PGRP-L1 and pPGRP-L2 share identical PGRP domains, which include PGRP domains I, II, and III (Fig. 2). The PGRP domains of porcine long isoforms are 85% and 78% identical to human and murine orthologs, respectively. However, the porcine long isoforms are more similar to the human long isoform in that they have six conserved cysteines, in contrast to the murine long isoform, which has five conserved cysteines. Compared to Drosophila long isoforms, pPGRP domains are most similar to PGRP-LB and -LE, and PGRP-LE has been linked to antimicrobial peptide expression. Overall, the entire peptide sequence of pPGRP-L2 shows 70% and 64% identities to human and murine long isoforms, respectively. In addition, several residues (His42, His152, Cys160, and Thr158 [referring to the positions in the Drosophila PGRP-LB sequence]), which are important either for zinc-binding or peptidoglycan-binding activities (14, 19), are conserved in porcine PGRP sequences.
Bacterial infection and pPGRP expression. After obtaining the full sequences of porcine PGRP long isoforms, we examined their tissue expression profiles in healthy animals and in animals infected with Salmonella enterica serovar Typhimurium. Porcine PGRP-L1 is expressed in all tissues tested, although at a lower level in the livers of control pigs (Fig. 3). Conversely, in control pigs PGRP-L2 is expressed in the duodenum, liver, and kidney and is upregulated in these tissues and in the rest of the small intestine in Salmonella-infected pigs.
To extend our evaluation of the effect of bacterial infection on pPGRP and antimicrobial peptide expression, we infected porcine jejunal cells with gram-positive and -negative bacteria. Infection with E. coli and S. enterica serovar Typhimurium enhanced pPGRP-L1 expression at 2 and 6 h after infection (Fig. 4). In cells infected with L. monocytogenes, enhanced PGRP-L1 expression was sustained for at least 24 h. pPGRP-L2 was stimulated by E. coli at 24 h and by S. enterica serovar Typhimurium at 2 and 6 h after infection was and markedly increased by L. monocytogenes infection. (Fig. 4). The degree and kinetics of -defensin-1 expression were more closely associated with increased pPGRP-L1 expression than with increased pPGRP-L2 expression in E. coli-infected cells. Through 12 h after infection, -defensin-1 expression coincided with increased expression of both pPGRP long isoforms in L. monocytogenes-infected cells (Fig. 4).
Silencing of pPGRP-L2 decreases -defensin-1 expression. Because porcine intestinal cells infected with L. monocytogenes exhibited robust expression of pPGRP-L1 and pPGRP-L2, we evaluated the influence of decreasing pPGRP expression on -defensin-1 expression by using RNA interference in cells challenged with L. monocytogenes. Bacterium-induced increases in pPGRP expression were mitigated by transfecting jejunal cells with siRNAs targeted to both long isoforms of pPGRP (Fig. 5). Control siRNA (scrambled) did not influence pPGRP expression. Similarly, two other siRNAs that we tested (matching gene-specific regions of pPGRP-L2) did not suppress pPGRP-L2 (data not shown). Silencing of pPGRP-L2 expression reduced -defensin-1 expression significantly, almost to the level of expression observed in noninfected control cells. In contrast, pPGRP-L1 is not involved with -defensin-1 expression in L. monocytogenes-infected jejunal cells. These findings suggest that pPGRP-L2 is linked to -defensin-1 expression in jejunal cells.
Cellular location of pPGRPs. To further evaluate the involvement of pPGRP long isoforms in porcine innate immunity, we expressed pPGRP-L1 and pPGRP-L2 ORFs fused with the C terminus of EGFP in a pEGFP-C3 vector. Intestinal cells were transfected with these constructs, and the translated proteins were visualized by confocal laser scanning microscopy. As predicted, pPGRP-L1 was localized predominantly to cytoplasmic areas, especially peripheral regions of nuclei (Fig. 6B). In contrast, pPGRP-L2 was located abundantly in vesicular cytoplasmic structures of some cells (Fig. 6C) and was distributed on the cell surface in other positive cells (Fig. 6D). In comparison, control EGFP shows a preference for nuclear location (Fig. 6A).
pPGRPs bind peptidoglycan. Recombinant pPGRPs from transfected cells displayed the correct molecular weights for EGFP-tagged proteins (Fig. 7A). In addition to the fused protein bands, a band was detected at a size similar to that of the EGFP tags in the PGRP-L2 lane (Fig. 7A, arrow). This band likely represents processed signal peptides fused with EGFP tags and suggests that the overexpressed proteins are processed by the signal peptidase. As previously described for other mammalian PGRPs (11, 14, 17), pPGRP-L1 and pPGRP-L2 bind to the S. aureus-derived insoluble peptidoglycan at a similar capacity (Fig. 7B).
pPGRPs have N-acetylmuramoyl-L-alanine amidase activity. Although both pPGRP long isoforms bind peptidoglycan, they differ in their ability to degrade peptidoglycan. pPGRP-L1 released two- to threefold more muramic acid from either intact (Fig. 8A) or lysozyme-treated (Fig. 8B) S. aureus-derived peptidoglycan.
Overexpression of pPGRPs increases -defensin-1 expression. Overexpression of both porcine long isoforms of PGRP was coupled with a significant increase in -defensin-1 expression in IPEC-J2 intestinal cells (Fig. 9).
DISCUSSION
First identified and characterized from insects, PGRPs are a family of proteins that are important in innate immunity (7, 17, 18, 38). In addition to insects, several short and long PGRP isoforms have been described for mammals, including humans, rodents, and cattle (3, 7). Here we report the identification and characterization of two long isoforms of porcine PGRPs, and we provide evidence that these proteins bind and degrade peptidoglycan and that expression of these proteins is associated with porcine innate immunity.
Porcine PGRP long isoforms are 85% and 78% identical to human and mouse orthologs, respectively, in their PGRP domains. However, pPGRP long isoforms have distinctive features. For example, in contrast to the case for the porcine genome, only one long PGRP isoform has been identified in humans and mice (7, 24). Porcine PGRP long isoforms likely represent alternative splicing variants. Analysis of the full cDNA sequences suggests that the two identical regions of pPGRP-L1 and pPGRP-L2 are transcribed from the same exons; however, the initial 5'-terminal 79 bp of pPGRP-L1 and the middle 982 bp (positions 170 to 1151) of pPGRP-L2 are likely from isoform-specific exons. The two pPGRP long isoforms also show differential expression and regulation patterns in tissues from control and Salmonella-challenged animals. pPGRP-L1 is expressed constitutively in most tissues, and gene expression is not influenced by Salmonella infection. pPGRP-L2 has a more discrete tissue expression pattern and is markedly upregulated in intestinal tissues by Salmonella infection. In contrast, in an intestinal cell line both isoforms are influenced by in vitro challenge with E. coli, S. enterica, and, most dramatically, L. monocytogenes. The difference between the in vitro and in vivo findings related to Salmonella challenge is likely associated with the kinetics of the in vitro challenge (0 to 12 h) and the in vivo infection (48 h postchallenge).
Different activities have been described for PGRP isoforms. For example, Drosophila PGRP-SC1B is an N-acetylmuramoyl-L-alanine amidase (28), an activity that also is conserved in human and mouse PGRP-L isoforms (11, 42). Both porcine long PGRP isoforms are similar to human and mouse PGRP long isoforms in the domain regions that have been shown to function as N-acetylmuramoyl-L-alanine amidases (11, 28, 42). In particular, several amino acid residues (His411, Cys419, His436, Tyr447, and Cys530) which are required for amidase activity of the human long isoform of PGRP (11, 42) are also conserved in the porcine long PGRPs. To confirm this predication, we performed peptidoglycan binding and lytic assays with the expressed recombinant porcine PGRPs and showed that they have N-acetylmuramoyl-L-alanine amidase activity. Why pigs have two PGRP long isoforms with N-acetylmuramoyl-L-alanine amidase activity is not known. Our predictions suggest that pPGRP-L1 is a cytoplasmic protein and pPGRP-L2 is a transmembrane protein, likely associated with the Golgi. Thus, it is possible that the two porcine isoforms function in different cellular locations. Similarly, the cellular location of the porcine PGRP long isoforms likely influences their spectrum of activity against intracellular and extracellular bacteria.
As their name implies, recognition of microbial patterns is another potential function associated with PGRPs. Drosophila PGRP-LC, -LE, -SA, and -SD are pattern recognition receptors that in response to bacterial challenge signal through the Toll or IMD pathway, resulting in the expression of antimicrobial peptides (4, 6, 7, 12, 13, 29, 32, 40). Moreover, overexpression of Drosophila PGRP-LC or PGRP-LE, without pathogen stimulation, results in upregulation of antimicrobial peptides (12, 13, 32, 40). For Drosophila, the general thought is that a PGRP may have either peptidoglycan-sensing activity (such as with PGRP-SA and -LC) or amidase activity (such as with PGRP-SC1B). However, recent reports indicate a bifunctional aspect of PGRP-SA. In addition to sensing bacterial infection and activating the Toll signaling pathway, it also digests diaminopimelic acid-type peptidoglycan from gram-negative bacteria (5). Besides its potential amidase activity, the crystal structure of PGRP-LB indicates that a hydrophobic groove at the back face of its active cleft may function in peptidoglycan signaling (19, 28).
Recently, TLR2-deficient mice were found to display normal antibacterial capabilities in response to polymicrobial or L. monocytogenes peritonitis (9, 44). Toll-like receptors and receptors of the interleukin 1 family transduce signals through myeloid differentiation factor 88; however, myeloid differentiation factor 88-deficient mice still control S. aureus and L. monocytogenes infections (37, 43). In addition, peritoneal macrophages from PGRP-L-deficient mice produce less interleukin 6 and tumor necrosis factor alpha when stimulated with bacteria, which are considered TLR2-mediated events (20, 46). Taken together, these findings suggest that other pattern recognition receptors, perhaps PGRPs, are involved in these bacterial infections (35). Because PGRP amidase activity is not sufficient to lyse intact bacteria (19), our data further imply that overexpression of porcine long PGRP stimulates the expression of other innate immune effectors, such as antimicrobial peptides, which in turn may limit bacterial growth.
In addition to the two porcine PGRP long isoforms described in this report, the porcine genome also has two short PGRP isoforms, pPGRP-S1 and pPGRP-S2 (GenBank accession numbers AY598969 and AF541957). Porcine PGRP-S1, which was originally identified as porcine neutrophil cationic protein (10), is deposited as a partial sequence in GenBank (accession number AJ310355), and we subsequently obtained the full cDNA sequence (accession number AY598969). The regulation and function of these two short PGRP isoforms remain to be determined.
In conclusion, we have identified two long isoforms of porcine PGRPs, which are highly conserved with human and mouse orthologs. Our findings indicate that pPGRP-L1 is expressed constitutively in several tissues and that expression of pPGRP-L2 is increased by Salmonella infection. Both long isoforms of porcine PGRP bind peptidoglycan and have N-acetylmuramoyl-L-alanine amidase activity. Loss-of-function and gain-of-function experiments showed that these two pPGRPs are involved in -defensin-1 expression. Taken together, our findings indicate that PGRPs potentially mediate porcine innate immune responses through regulation of the expression of antimicrobial peptides.
ACKNOWLEDGMENTS
We thank Danielle Goodband and Ling Zheng for their excellent technical support and J. Ernest Minton for assistance with collection of tissues from Salmonella-infected animals.
This work was supported in part by National Research Initiative Competitive Grant 2001-35204-10818 (to F.B. and C.R.R.) from the U.S. Department of Agriculture.
REFERENCES
1. Aderem, A., and K. D. Smith. 2004. A systems approach to dissecting immunity and inflammation. Semin. Immunol. 16:55-67.
2. Arnold, A. S., M. Gueye, P. Ronde, J. M. Warter, P. Poindron, and J. P. Gies. 2002. Construction of a plasmid containing human SMN, the SMA determining gene, coupled to EGFP. Plasmid 47:79-87.
3. Beutler, B. 2004. Innate immunity: an overview. Mol. Immunol. 40:845-859.
4. Bischoff, V., C. Vignal, I. G. Boneca, T. Michel, J. A. Hoffmann, and J. Royet. 2004. Function of the Drosophila pattern-recognition receptor PGRP-SD in the detection of Gram-positive bacteria. Nat. Immunol. 5:1175-1180.
5. Chang, C. I., S. Pili-Floury, M. Herve, C. Parquet, Y. Chelliah, B. Lemaitre, D. Mengin-Lecreulx, and J. Deisenhofer. 2004. A Drosophila pattern recognition receptor contains a peptidoglycan docking groove and unusual L,D-carboxypeptidase activity. PLoS Biol 2:E277.
6. Choe, K-M., T. Werner, S. Stven, D. Hultmark, and K. V. Anderson. 2002. Requirement for a peptidoglycan recognition protein (PGRP) in Relish activation and antibacterial immune responses in Drosophila. Science 296:359-362.
7. Dziarski, R. 2004. Peptidoglycan recognition proteins. Mol. Immunol. 40:877-886.
8. Dziarski, R., K. A. Platt, E. Gelius, H. Steiner, and 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.
9. Edelson, B. T., and E. R. Unanue. 2002. MyD88-dependent but Toll-like receptor 2-independent innate immunity to Listeria: no role for either in macrophage listericidal activity. J. Immunol. 169:3869-3875.
10. Fornhem, C., C. G. Peterson, and K. Alving. 1996. Isolation and characterization of porcine cationic eosinophil granule proteins. Int. Arch. Allergy Immunol. 110:132-142.
11. Gelius, E., C. Persson, J. Karlsson, and H. Steiner. 2003. A mammalian peptidoglycan recognition protein with N-acetylmuramoyl-L-alanine amidase activity. Biochem. Biophys. Res. Commun. 306:988-994.
12. Gobert, V., M. Gottar, A. Matskevich, A., S. Rutschmann, J. Royet, M. Belvin, J. A. Hoffmann, and D. Ferrandon. 2003. Dual activation of the Drosophila toll pathway by two pattern recognition receptors. Science 302:2126-2130.
13. Gottar, M., V. Gobert, T. Michel, M. Belvin, G. Duyk, J. A. Hoffman, D. Ferrandon, and J. Royet. 2002. The Drosophila immune response against Gram-negative bacteria is mediated by a peptidoglycan recognition protein. Nature 416:640-644.
14. Guan, R., A. Roychowdhury, B. Ember, S. Kumar, G. J. Boons, and R. A. Mariuzza. 2004. Structural basis for peptidoglycan binding by peptidoglycan recognition proteins. Proc. Natl. Acad. Sci. USA 101:17168-17173.
15. Hazenberg, M. P., and H. de Visser. 1992. Assay for N-acetylmuramyl-L-alanine amidase in serum by determination of muramic acid released from the peptidoglycan of Brevibacterium divaricatum. Eur. J. Clin. Chem. Clin. Biochem. 30:141-144.
16. Janeway, C. A., Jr., and R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197-216.
17. Kang, D., G. Liu, A. Lundstrm, E. Gelius, and H. Steiner. 1998. A peptidoglycan recognition protein in innate immunity conserved from insects to humans. Proc. Natl. Acad. Sci. USA 95:10078-10082.
18. Khush, R. S., F. Leulier, and B. Lemaitre. 2002. Pathogen surveillance—the flies have it. Science 296:273-275.
19. Kim, M. S., M. Byun, and B. H. Oh. 2003. Crystal structure of peptidoglycan recognition protein LB from Drosophila melanogaster. Nat. Immunol. 4:787-793.
20. Knapp, S., C. W. Wieland, C. van 't Veer, O. Takeuchi, S. Akira, S. Florquin, and T. van der Poll. 2004. Toll-like receptor 2 plays a role in the early inflammatory response to murine pneumococcal pneumonia but does not contribute to antibacterial defense. J. Immunol. 172:3132-3138.
21. Krull, M., R. Nost, S. Hippenstiel, E. Domann, T. Chakraborty, and N. Suttorp. 1997. Listeria monocytogenes potently induces up-regulation of endothelial adhesion molecules and neutrophil adhesion to cultured human endothelial cells. J. Immunol. 159:1970-1976.
22. Lee, J. C., and G. C. Stewart. 2003. Essential nature of the mreC determinant of Bacillus subtilis. J. Bacteriol. 185:4490-4498.
23. Liu, C., E. Gelius, G. Liu, H. Steiner, and 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.
24. Liu, C., Z. Xu, D. Gupta, and R. Dziarski. 2001. Peptidoglycan recognition proteins: a novel family of four human innate immunity pattern recognition molecules. J. Biol. Chem. 276:34686-64694.
25. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(–C(T)) method. Methods 25:402-408.
26. Mathur, P., B. Murray, T. Crowell, H. Gardner, N. Allaire, Y. M. Hsu, G. Thill, and J. P. Carulli. 2004. Murine peptidoglycan recognition proteins PglyrpI-alpha and PglyrpI-beta are encoded in the epidermal differentiation complex and are expressed in epidermal and hematopoietic tissues. Genomics 83:1151-1163.
27. Medzhitov, R., and C. A. Janeway, Jr. 2002. Decoding the patterns of self and nonself by the innate immune system. Science 296:298-300.
28. Mellroth, P., J. Karlsson, and H. Steiner. 2003. A scavenger function for a Drosophila peptidoglycan recognition protein. J. Biol. Chem. 278:7059-7064.
29. Michel, T., J. M. Reichhart, J. A. Hoffman, and J. Royet. 2001. Drosophila Toll is activated by gram-positive bacteria through a circulating peptidoglycan recognition protein. Nature 414:756-759.
30. Pasare, C., and R. Medzhitov. 2004. Toll-like receptors and acquired immunity. Semin. Immunol. 16:23-26.
31. Ramanathan, B., H. Wu, C. R. Ross, and F. Blecha. 2004. PR-39, a porcine antimicrobial peptide, inhibits apoptosis: involvement of caspase-3. Dev. Comp. Immunol. 28:163-169.
32. Rmet, M., P. Manfruelli, A. Pearson, B. Mathey-Prevot, and R. A. Ezekowitz. 2002. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416:644-648.
33. Reis e Sousa, C. 2004. Activation of dendritic cells: translating innate into adaptive immunity. Curr. Opin. Immunol. 16:21-25.
34. Rhoads, J. M., W. Chen, P. Chu, H. M. Berschneider, R. A. Argenzio, and A. M. Paradiso. 1994. L-Glutamine and [scap]l-asparagine stimulate Na+-H+ exchange in porcine jejunal enterocytes. Am. J. Physiol. 266:828-838.
35. Sabroe, I., R. C. Read, M. K. Whyte, D. H. Dockrell, S. N. Vogel, and S. K. Dower. 2003. Toll-like receptors in health and disease: complex questions remain. J. Immunol. 171:1630-1635.
36. Sang, Y., M. T. Ortega, F. Blecha, O. Prakash, and T. Melgarejo. 2005. Molecular cloning and characterization of three beta-defensins from canine testes. Infect. Immun. 73:2611-2620.
37. Skerrett, S. J., H. D. Liggitt, A. M. Hajjar, and C. B. Wilson. 2004. Cutting edge: myeloid differentiation factor 88 is essential for pulmonary host defense against Pseudomonas aeruginosa but not Staphylococcus aureus. J. Immunol. 172:3377-3381.
38. Steiner, H. 2004. Peptidoglycan recognition proteins: on and off switches for innate immunity. Immunol. Rev. 198:83-96.
39. Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. Immunol. 21:335-376.
40. Takehana, A., T. Katsuyama, T. Yano, Y. Oshima, H. Takada, T. Aigaki, and S. Kurata. 2002. Overexpression of a pattern-recognition receptor, peptidoglycan-recognition protein-LE, activates imd/relish-mediated antibacterial defense and the prophenoloxidase cascade in Drosophila larvae. Proc. Natl. Acad. Sci. USA 99:13705-13710.
41. Tydell, C. C., N. Yount, D. Tran, J. Yuan, and 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.
42. Wang, Z. M., X. Li, R. R. Cocklin, M. Wang, M. Wang, K. Fukase, S. Inamura, S. Kusumoto, D. Gupta, and R. Dziarski. 2003. Human peptidoglycan recognition protein-L is an N-acetylmuramoyl-L-alanine amidase. J. Biol. Chem. 278:49044-49052.
43. Way, S. S., T. R. Kollmann, A. M. Hajjar, and C. B. Wilson. 2003. Cutting edge: protective cell-mediated immunity to Listeria monocytogenes in the absence of myeloid differentiation factor 88. J. Immunol. 171:533-537.
44. Weighardt, H., S. Kaiser-Moore, R. M. Vabulas, C. J. Kirschning, H. Wagner, and B. Holzmann. 2002. Cutting edge: myeloid differentiation factor 88 deficiency improves resistance against sepsis caused by polymicrobial infection. J. Immunol. 169:2823-2827.
45. Wu, H., G. Zhang, J. E. Minton, C. R. Ross, and F. Blecha. 2000. Regulation of cathelicidin gene expression: induction by lipopolysaccharide, interleukin-6, retinoic acid, and Salmonella enterica serovar Typhimurium infection. Infect. Immun. 68:5552-5558.
46. Xu, M., Z. Wang, R. M. Locksley. 2004. Innate immune responses in peptidoglycan recognition protein L-deficient mice. Mol. Cell. Biol. 24:7949-7957.
47. Yoshida, H., K. Kinoshita, and M. Ashida. 1996. Purification of a peptidoglycan recognition protein from hemolymph of the silkworm, Bombyx mori. J. Biol. Chem. 271:13854-13860.
48. Zhang, G., H. Hiraiwa, H. Yasue, H. Wu, C. R. Ross, D. Troyer, and F. Blecha. 1999. Cloning and characterization of the gene for a new epithelial -defensin. J. Biol. Chem. 274:24031-24037.
49. Zhang, G., C. R. Ross, S. S. Dritz, J. C. Nietfeld, and F. Blecha. 1997. Salmonella infection increases porcine antibacterial peptide concentration in serum. Clin. Diagn. Lab. Immunol. 4:774-777.(Yongming Sang, Balaji Ram)
ABSTRACT
Peptidoglycan recognition proteins (PGRPs) are a group of newly identified proteins with emerging functions in mammalian innate immunity. Here we report the identification and characterization of two long isoforms of porcine PGRP. Their complete cDNA sequences encode predicted peptides of 252 and 598 residues and are named pPGRP-L1 and pPGRP-L2, respectively. These porcine isoforms share identical PGRP domains at their C terminus, which are highly conserved with human and mouse orthologs. pPGRP-L1 is expressed constitutively in several tissues, including bone marrow, intestine, liver, spleen, kidney, and skin. pPGRP-L2 is highly expressed in the duodenum and liver, and expression in intestinal tissues is increased by Salmonella infection. In intestinal cells, expression of both pPGRP-L1 and pPGRP-L2 is increased by bacterial infection. Recombinant pPGRP-L1 and pPGRP-L2 have N-acetylmuramoyl-L-alanine amidase activity. Loss-of-function and gain-of-function experiments indicate that these two pPGRPs are involved in expression of the antimicrobial peptide -defensin-1. Silencing of pPGRP-L2 in intestinal cells challenged with Listeria monocytogenes results in downregulation of -defensin-1. Conversely, overexpression of pPGRP-L1 or pPGRP-L2 dramatically upregulates expression of -defensin-1. Collectively, these findings suggest that porcine PGRPs are involved in antimicrobial peptide expression.
INTRODUCTION
Innate immunity provides the first line of defense against infectious disease (1, 3, 16). The cells and effector molecules responsible for innate immunity have the essential role of identifying and eliminating microbial pathogens and signaling the adaptive immune system of their presence. The first critical step in initiating these responses is recognition of the invading pathogen. This recognition occurs via protein receptors that recognize motifs or patterns on microorganisms that are not present on eukaryotic cells (27, 33). In addition to the well-known Toll receptors and Toll-like receptors (TLR), peptidoglycan recognition proteins (PGRPs) represent another group of pathogen-binding proteins, and some PGRP isoforms in Drosophila have been shown to mediate the production of antimicrobial peptides (30, 39).
The first PGRPs were identified in silkworms (Bombyx mori) and moths (Trichoplusia ni) through their high-affinity binding to peptidoglycan and were implicated in innate immunity because of their ability to activate the protease cascade that leads to phenoloxidase activation in response to peptidoglycan (17, 47). To date, data from the fruit fly (Drosophila melanogaster) genome project have revealed 13 PGRP homologs on three chromosomes (7, 18). Based on their predicted peptide sequences, they are classified into two groups: seven short PGRPs (PGRP-S), which are predicted to be small extracellular proteins that are exported after removal of an N-terminal signal peptide, and six long PGRPs (PGRP-L), which encode intracellular or membrane-spanning proteins (7, 18). It is now clear that some of these pattern recognition proteins respond to several microbial species via interactions with peptidoglycan and mediate the production of antimicrobial peptides through the Toll or immune deficiency (IMD) pathways. For example, PGRP-SA and -SD respond to different groups of gram-positive bacteria (4, 12, 29), and the PGRP-LC gene was discovered through its induction of antimicrobial genes after a gram-negative bacterial infection (6, 12, 13, 32). In addition, PGRP-LE stimulated the IMD pathway through interaction with diaminopimelic acid-type peptidoglycan, which is present in many gram-negative and some gram-positive bacteria (40). Loss-of-function studies, using RNA interference, showed that PGRP-LC is involved in phagocytosis of gram-negative Escherichia coli but not gram-positive Staphylococcus aureus (6, 12, 13, 32). Gain-of-function studies showed that overexpression of PGRP-LC or PGRP-LE upregulated the production of insect antimicrobial peptides (6, 40). In addition, some Drosophila PGRP isoforms, such as PGRP-SC1B, have N-acetylmuramoyl-L-alanine amidase activity (28), and PGRP-SA has L-carboxypeptidase activity (5).
Several PGRPs have also been identified in mammals. Four PGRPs have been identified in humans and mice: one long, one short, and two mammalian-specific intermediate isoforms (7, 23, 24, 26). In addition to binding peptidoglycan, murine PGRP-S, which is expressed in neutrophils, inhibits growth of gram-positive bacteria and therefore has been suggested to function as a neutrophil intracellular antibacterial protein (8). Studies using PGRP-S knockout mice indicate that murine PGRP-S is involved in clearance of nonpathogenic gram-positive bacteria but not pathogenic gram-positive or gram-negative bacteria (8). Similarly, the bovine PGRP ortholog has antimicrobial activity against gram-positive and -negative bacteria and yeasts (41). Like Drosophila PGRP-SC1B, human and mouse PGRP long isoforms have been shown to function as N-acetylmuramoyl-L-alanine amidases (11, 28, 42). In addition, peritoneal macrophages from PGRP-L-deficient mice produce less interleukin 6 and tumor necrosis factor alpha when stimulated with E. coli or lipopolysaccharide (46). Collectively, these findings indicate that PGRPs are involved in mammalian innate immune responses. However, unlike the case for Drosophila, the association between recognition of the bacterial threat by PGRPs and elicitation of an antimicrobial peptide response has not been established for mammals. Here we report the identification and characterization of porcine PGRP long isoforms and show, by silencing and overexpressing these proteins, that they are associated with the expression of the antimicrobial peptide -defensin-1.
MATERIALS AND METHODS
Cell culture and reagents. Porcine intestinal cells (IPEC-J2) were grown in Dulbecco-Ham F-12 medium (1:1, vol/vol) containing 5% fetal bovine serum and supplemented with insulin (10 μg/ml), transferrin (5.5 μg/ml), sodium selenite (0.67 μg/ml), and epidermal growth factor (5 ng/ml) (GIBCO Invitrogen Corp., Gaithersburg, MD) (34). Cells were maintained in a 95% air-5% CO2 humidified atmosphere at 37°C and subcultured at 4-day intervals.
Gene identification and analysis. To begin our characterization of the porcine PGRP family, we first screened the porcine expressed sequence tag (EST) database at GenBank, using human and bovine PGRP sequences (GenBank accession numbers AF384856, AY035376, AY035377, and AY083309) (24, 41). Two porcine EST sequences (accession numbers BE013528 and BF442868) that contained PGRP-homologous sequences in their translated regions were identified. Two pairs of primers were designed based on the EST sequences (outer sense, 5'-GTA GCC CAA AGC CAC TGA AG-3'; outer antisense, 5'-GAG TTG TGG TCA CGT GTG T-3'; inner sense, 5'-AAG CCA CTG AAG CTG CCA CT-3'; inner antisense, 5'-ATC TGA GCC CAC CAC GAA AC-3'). To obtain the full-length cDNA sequences, we used a modified rapid amplification of cDNA ends (RACE) (Ambion FirstChoice RLM-RACE kit) protocol, which selects for nontruncated 5'-capped mRNAs, and adopted a similar nested PCR strategy to obtain the 5'- and 3'-terminal cDNA sequences from total RNA purified from porcine liver. The nested PCR products were purified with a column-based PCR purification kit (QIAGEN, Inc., Valencia, CA) and cloned into plasmids with pGEM-T Easy Vector systems (Promega Co., Madison, WI), followed by transformation into E. coli (JM109; Stratagene Co., La Jolla, CA) and colony screening with PCR. Primers used for clone screening PCR were the sense and antisense gene-specific primers. Plasmids were extracted from bacterial cultures derived from the identified single colonies, and inserts were sequenced with SP6/T7 vector primers on an ABI 3700 DNA analyzer at the Kansas State University Sequencing and Genotyping Facility (Manhattan, KS). Each cDNA sequence was generated from the sequencing results for at least five identical plasmid extracts from individual colonies. The full-length cDNA sequence was then generated by ligation of 5' and 3' sequences and deletion of the adaptor and overlapped regions. Open reading frames (ORFs) were predicted with ORFfinder and Genescan (http://www.ncbi.nlm.nih.gov/gorf/orfig.cgi and http://genes.mit.edu/GENSCAN.html). Translated peptide sequences were analyzed using SignalP and PSORT (ExPASy; http://au.expasy.org/tools/#proteome) for signal peptide and cellular location predictions, respectively.
Expression in tissue. Tissues were obtained from healthy and Salmonella-challenged 5-week-old pigs as previously described (45, 48, 49), using procedures approved by the Kansas State University Institutional Animal Care and Use Committee. Tissue samples were collected, placed immediately in liquid nitrogen, and stored at –135°C until use. Total RNA was extracted with TRI reagent (Sigma-Aldrich, St. Louis, MO) after grinding the frozen tissues in liquid nitrogen. A one-step reverse transcription-PCR (RT-PCR) was used to detect expression of target transcripts. Briefly, total RNA was treated with RQ1 RNase-free DNase I (Promega), and RNA samples (250 ng) were then used in a 25-μl RT-PCR mixture with 0.1 μM of each sense and antisense primer designed to distinguish pPGRP-L1 (sense, 5'-CCC AGG TTC CTT CTT GGA TT-3'; antisense, 5'-GTG GAA GCG CTG CAT AGA G-3'; cDNA nucleotides 51 to 70 and 551 to 533, respectively) and pPGRP-L2 (sense, 5'-GCC CAG ACT GAA AGC AAT TC-3'; antisense, 5'-AGC CCA CCA CGA AAC TGT AG-3'; cDNA nucleotides 848 to 867 and 1502 to 1483, respectively). Conventional one-step RT-PCR was performed using an AccessQuick RT-PCR system kit (Promega). cDNA synthesis and predenaturation were performed for one cycle at 48°C for 45 min and 95°C for 3 min, respectively. Amplification was carried out for 40 or 25 (glyceraldehyde 3-phosphate dehydrogenase [GAPDH]) cycles at 95°C for 30 s, 55°C for 30 s, and 72°C for 30 s, and final extension was at 72°C for 10 min in a 25-μl reaction mixture. After amplification, 10 μl of each reaction mixture was analyzed by 2% agarose gel electrophoresis, and bands were visualized by ethidium bromide staining. For Southern analysis, PCR products were transferred onto Hybond N+ filters (Amersham Pharmacia Biotech), hybridized with PGRP cDNA probes labeled with [-32P]dCTP (3,000 Ci/mmol), and detected using autoradiography as previously described (48).
Bacterial infection and real-time RT-PCR. Bacteria used were Escherichia coli (ATCC 25922), Listeria monocytogenes (ATCC 19115), and Salmonella enterica serovar Typhimurium (ATCC BAA-185). Cells cultured in 24-well plates were infected by adding medium containing 104 CFU/ml at 0.25 ml per well. Plates were centrifuged at 800 x g for 10 min and incubated for 2 h at 37°C in a CO2 incubator. Cells then were washed with medium containing 100 IU penicillin and 10 μg/μl streptomycin solution to remove extracellular bacteria. The cells in replicate wells were cultured and sampled through 2 to 24 h, and total RNA was extracted using Trizol reagent (45).
Real-time quantitative RT-PCR was performed on a SmartCycler (Cepheid, Sunnyvale, CA) as previously described (36) with a real-time one-step RT-PCR kit (QIAGEN, Valencia, CA). Gene-specific primers for pPGRP-L1 were sense, 5'-CTG TGG TAG CAG CAG CTT CTC TTT-3', and antisense, 5'-GCT GCA GAA GCA ATC CAT ACA GGA-3' (cDNA nucleotides 12 to 35 and 256 to 233), and those for pPGRP-L2 were sense, 5'-AGC CCA GAG TTT CAA GGC CTG ATT-3', and antisense, 5'-TCA GGA ACT GTG GCT CCA ATG TCT-3' (cDNA nucleotides 368 to 391 and 602 to 579). Primers for -defensin-1 were as described previously (48). DNase-treated total RNA (500 ng) was used in each 25-μl RT-PCR mixture. RT-PCR cycling conditions were 30 min at 50°C and 15 min at 95°C, followed by 45 cycles of 15 s at 95°C, 30 s at 56°C, and 30 s at 72°C with the optic on for fluorescence data collection. Specific amplification was confirmed by sequencing the amplicons before performing real-time RT-PCR. Threshold cycle (CT) values were determined by exponential product amplification and subsequent increased fluorescence intensity above background. Relative gene expression data were normalized against CT values for the housekeeping (GAPDH) gene, and the relative index () was determined by comparison to the average expression levels for control samples with the index defined as 1.0 (25).
Silencing of PGRP expression. Small interfering RNAs (siRNAs) were generated (QIAGEN-Xeragon) against the isoform-specific and -identical regions. Preliminary experiments showed that only two siRNAs from isoform-identical regions exhibited significant suppression with isoform preference. The template sequences of these two siRNAs are 5'-AAT GTG AAA CCA AGA ACT GCC-3' (cDNA nucleotides 885 to 905 and 1787 to 1807 for pPGRP-L1 and -L2, respectively) and 5'-AAA CTG GAA GAC CAC AAT GGT-3' (cDNA nucleotides 169 to 189 and 91 to 111 for pPGRP-L1 and -L2, respectively). To optimize the transformation procedure and for a negative control, we used a rhodamine-labeled scrambled siRNA (5'-AAT TCT CCG AAC GTG TCA CGT-3'). Porcine intestinal cells (IPEC-J2 cells at 5 x 104 cells/well in a 24-well plate) at 60% confluence were transfected with siRNAs by using QIAGEN siRNA transfection reagent. After 24 h, cells (except control wells) were infected with Listeria monocytogenes (104 CFU/ml at 0.25 ml per well) for 2 h at 37°C and then washed with Hanks' balanced salt solution containing 1x antibiotic/antimycotic solution (GIBCO) to remove extracellular bacteria (21). siRNA transfection mixtures were added back to the appropriate wells, and cells were cultured for another 24 h. Cells were then collected, and RNA was isolated for one-step RT-PCR assay with gene-specific primers.
Overexpression of PGRP and cellular localization. Open reading frames of pPGRP-L1 and pPGRP-L2, were generated from cDNA clones with HindIII and BamHI cloning sites at their 5' and 3' termini, respectively, by PCR. Inserts were subcloned into HindIII- and BamHI-digested and phosphatase-treated pEGFP-C3 (Clontech) (2). E. coli (JM109) transformed with the constructs were selected with kanamycin (30 μg/ml; Sigma), and positive clones were identified by PCR. Purified plasmids were then used to transfect the porcine cell lines. In brief, control pEGFP-C3, PGRP-L1, or PGRP-L2 ORF plasmid constructs (2 μg) were introduced into cells at 60 to 80% confluence in six-well plates by using Fugene6 transfection reagent (Roche Diagnostics). All DNA-Fugene6 incubations were optimized at a ratio of 1 μg of DNA to 2 μl of Fugene6 according to the manufacturer's recommended protocol. Transformation efficiency was determined by enhanced green fluorescent protein (EGFP) expression using inverted fluorescence microscopy. After transformation (24 h), cells were collected and transformants were sorted by fluorescence-activated flow cytometry as previously described (31). Overexpression of control EGFP and EGFP-fused pPGRP-L1 or pPGRP-L2 was confirmed by both RT-PCR and immunoblotting. Primers for EGFP were derived from the pEGFP-C3 sequence (GenBank accession number U57607) and are 5'-TCT TCT TCA AGG ACG ACG GCA ACT-3' (sense), and 5'-TGT GGC GGA TCT TGA AGT TCA CCT-3' (antisense). For immunoblotting, proteins were extracted with a mammalian protein extraction reagent (Pierce). Briefly, cells were lysed (0.4 ml of extraction reagent/106 cells) and centrifuged at 12,000 x g for 10 min, and proteins in the supernatant, which include cytoplasmic, nuclear, and some membrane-associated proteins, from 105 EGFP-positive cells were separated with 4 to 20% precast sodium dodecyl sulfate (SDS)-polyacrylamide gels (Pierce) and transferred onto polyvinylidene difluoride blots. EGFP and EGFP-fused proteins were then detected with monoclonal anti-EGFP antibodies (1:8,000; Clontech) and visualized using a color development reaction catalyzed by alkaline phosphatase-conjugated secondary antibodies.
To determine the cellular localization of the EGFP-tagged pPGRPs, porcine intestinal cells were cultured and transfected as described above in a four-well culture slide (BD Falcon). Twenty-four hours after transfection, slides with >20% EGFP-positive cells were fixed and examined using confocal laser scanning fluorescence microscopy. In brief, media were removed, cells were washed once with Dulbecco's phosphate-buffered saline (DPBS) (pH 7.4), and 1 ml of 4% paraformaldehyde-DPBS was added to each well. Cells were fixed for 30 min at 22°C, nuclei were counterstained with Hoechst 33258 (0.12 μg/ml), and cells were rinsed twice with DPBS. Slides were mounted with a drop of ProLong antifade solution (Molecular Probes) and examined with a confocal laser scanning microscope (LSM 510; Zeiss).
Peptidoglycan binding assay. EGFP and EGFP-recombinant pPGRP-L1 and pPGRP-L2 were purified on anti-GFP-conjugated agarose bead affinity columns (Vector Lab Inc., Burlingame, CA) and evaluated on Western blots using monoclonal anti-EGFP antibodies as described previously (11, 17). Insoluble peptidoglycan (0.3 mg; Sigma) from S. aureus was incubated with 10 μg of EGFP or EGFP-tagged pPGRP-L1 and pPGRP-L2 in a 300-μl reaction mixture buffered by 20 mM Tris-HCl saline (pH 7.9). After rocking incubation at 4°C for 5 h, bound and unbound proteins were separated by centrifugation. Proteins were resolved from peptidoglycan by boiling in 2x SDS-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer and subjected to 4 to 20% SDS-PAGE. Proteins were detected with anti-EGFP antibodies on Western blots.
N-Acetylmuramoyl-L-alanine amidase activity. N-Acetylmuramoyl-L-alanine amidase activity was measured as described for murine PGRP-L and Drosophila PGRP-SC1B (11, 15). Briefly, peptidoglycan (50 mg, intact or digested with lysozyme) (Merck) (15) was incubated with 20 μg/ml recombinant pPGRPs in a reaction mixture containing 1 mM ZnSO4 20 mM MgCl2, and 50 mM Tris-HCl, pH 7.9, at 37°C. Samples (200 μl) were removed at each time point, brought to a 0.5-ml volume with 1.0 M NaOH, and incubated at 38°C for 30 min. Fifty microliters of 0.5 M H2SO4 and 5 ml of concentrated H2SO4 were added, and the samples were placed in a boiling water bath for 5 min. Samples were cooled, and 0.05 ml of 4% (wt/vol) CuSO4 · 5H2 O and 0.1 ml of 1.5% (wt/vol) p-hydroxydiphenyl (in 95% ethanol) were added. Samples were then incubated at 30°C for 30 min, and absorbance was determined at 560 nm. A standard curve consisting of 0 to 20 μg of muramic acid (Sigma) was used in the assay (22).
Nucleotide sequence accession numbers. The cDNA sequences reported in this paper have been registered in the GenBank/EBI data bank with accession numbers AF541955 (pPGRP-L1) and AF541956 (pPGRP-L2).
RESULTS
cDNA and amino acid sequences of pPGRP isoforms. Using 5'- and 3'-RACE in combination with primer pairs generated from a porcine EST clone, we identified two PGRP candidates from porcine fetal liver RNA. The cDNA of the first porcine PGRP is 1,008 bp (pPGRP-L1, GenBank accession number AF541955), and the longer isoform is 1,910 bp (pPGRP-L2, GenBank accession number AF541956). Comparison of these two sequences explained why two cDNA sequences were amplified with these primers; 722 bp of their 3'-terminal sequences are identical (Fig. 1A), and our gene-specific RACE primers are matched within this identical region. However, pPGRP-L1 has a unique 80-bp sequence at its 5' terminus, and pPGRP-L2 has an additional 982-bp insert incorporated into its coding region. These two cDNA sequences encode predicted peptide sequences of 252 and 598 amino acid residues, respectively (Fig. 1B). The C-terminal 242 amino acids of the two peptides, which contain the conserved PGRP domains, are identical. In addition, pPGRP-L1 has 10 and pPGRP-L2 has 356 unique amino acids at their N termini. Computational predications suggest that pPGRP-L1 is a cytoplasmic protein and pPGRP-L2 is a transmembrane protein. Porcine PGRP-L2 contains one obvious transmembrane helix at its N-terminal 55 to 76 amino acid residues. The predicted signal peptide region of pPGRP-L2 is located before the N-terminal 31 amino acid residues.
Species comparison. Porcine PGRP-L1 and pPGRP-L2 share identical PGRP domains, which include PGRP domains I, II, and III (Fig. 2). The PGRP domains of porcine long isoforms are 85% and 78% identical to human and murine orthologs, respectively. However, the porcine long isoforms are more similar to the human long isoform in that they have six conserved cysteines, in contrast to the murine long isoform, which has five conserved cysteines. Compared to Drosophila long isoforms, pPGRP domains are most similar to PGRP-LB and -LE, and PGRP-LE has been linked to antimicrobial peptide expression. Overall, the entire peptide sequence of pPGRP-L2 shows 70% and 64% identities to human and murine long isoforms, respectively. In addition, several residues (His42, His152, Cys160, and Thr158 [referring to the positions in the Drosophila PGRP-LB sequence]), which are important either for zinc-binding or peptidoglycan-binding activities (14, 19), are conserved in porcine PGRP sequences.
Bacterial infection and pPGRP expression. After obtaining the full sequences of porcine PGRP long isoforms, we examined their tissue expression profiles in healthy animals and in animals infected with Salmonella enterica serovar Typhimurium. Porcine PGRP-L1 is expressed in all tissues tested, although at a lower level in the livers of control pigs (Fig. 3). Conversely, in control pigs PGRP-L2 is expressed in the duodenum, liver, and kidney and is upregulated in these tissues and in the rest of the small intestine in Salmonella-infected pigs.
To extend our evaluation of the effect of bacterial infection on pPGRP and antimicrobial peptide expression, we infected porcine jejunal cells with gram-positive and -negative bacteria. Infection with E. coli and S. enterica serovar Typhimurium enhanced pPGRP-L1 expression at 2 and 6 h after infection (Fig. 4). In cells infected with L. monocytogenes, enhanced PGRP-L1 expression was sustained for at least 24 h. pPGRP-L2 was stimulated by E. coli at 24 h and by S. enterica serovar Typhimurium at 2 and 6 h after infection was and markedly increased by L. monocytogenes infection. (Fig. 4). The degree and kinetics of -defensin-1 expression were more closely associated with increased pPGRP-L1 expression than with increased pPGRP-L2 expression in E. coli-infected cells. Through 12 h after infection, -defensin-1 expression coincided with increased expression of both pPGRP long isoforms in L. monocytogenes-infected cells (Fig. 4).
Silencing of pPGRP-L2 decreases -defensin-1 expression. Because porcine intestinal cells infected with L. monocytogenes exhibited robust expression of pPGRP-L1 and pPGRP-L2, we evaluated the influence of decreasing pPGRP expression on -defensin-1 expression by using RNA interference in cells challenged with L. monocytogenes. Bacterium-induced increases in pPGRP expression were mitigated by transfecting jejunal cells with siRNAs targeted to both long isoforms of pPGRP (Fig. 5). Control siRNA (scrambled) did not influence pPGRP expression. Similarly, two other siRNAs that we tested (matching gene-specific regions of pPGRP-L2) did not suppress pPGRP-L2 (data not shown). Silencing of pPGRP-L2 expression reduced -defensin-1 expression significantly, almost to the level of expression observed in noninfected control cells. In contrast, pPGRP-L1 is not involved with -defensin-1 expression in L. monocytogenes-infected jejunal cells. These findings suggest that pPGRP-L2 is linked to -defensin-1 expression in jejunal cells.
Cellular location of pPGRPs. To further evaluate the involvement of pPGRP long isoforms in porcine innate immunity, we expressed pPGRP-L1 and pPGRP-L2 ORFs fused with the C terminus of EGFP in a pEGFP-C3 vector. Intestinal cells were transfected with these constructs, and the translated proteins were visualized by confocal laser scanning microscopy. As predicted, pPGRP-L1 was localized predominantly to cytoplasmic areas, especially peripheral regions of nuclei (Fig. 6B). In contrast, pPGRP-L2 was located abundantly in vesicular cytoplasmic structures of some cells (Fig. 6C) and was distributed on the cell surface in other positive cells (Fig. 6D). In comparison, control EGFP shows a preference for nuclear location (Fig. 6A).
pPGRPs bind peptidoglycan. Recombinant pPGRPs from transfected cells displayed the correct molecular weights for EGFP-tagged proteins (Fig. 7A). In addition to the fused protein bands, a band was detected at a size similar to that of the EGFP tags in the PGRP-L2 lane (Fig. 7A, arrow). This band likely represents processed signal peptides fused with EGFP tags and suggests that the overexpressed proteins are processed by the signal peptidase. As previously described for other mammalian PGRPs (11, 14, 17), pPGRP-L1 and pPGRP-L2 bind to the S. aureus-derived insoluble peptidoglycan at a similar capacity (Fig. 7B).
pPGRPs have N-acetylmuramoyl-L-alanine amidase activity. Although both pPGRP long isoforms bind peptidoglycan, they differ in their ability to degrade peptidoglycan. pPGRP-L1 released two- to threefold more muramic acid from either intact (Fig. 8A) or lysozyme-treated (Fig. 8B) S. aureus-derived peptidoglycan.
Overexpression of pPGRPs increases -defensin-1 expression. Overexpression of both porcine long isoforms of PGRP was coupled with a significant increase in -defensin-1 expression in IPEC-J2 intestinal cells (Fig. 9).
DISCUSSION
First identified and characterized from insects, PGRPs are a family of proteins that are important in innate immunity (7, 17, 18, 38). In addition to insects, several short and long PGRP isoforms have been described for mammals, including humans, rodents, and cattle (3, 7). Here we report the identification and characterization of two long isoforms of porcine PGRPs, and we provide evidence that these proteins bind and degrade peptidoglycan and that expression of these proteins is associated with porcine innate immunity.
Porcine PGRP long isoforms are 85% and 78% identical to human and mouse orthologs, respectively, in their PGRP domains. However, pPGRP long isoforms have distinctive features. For example, in contrast to the case for the porcine genome, only one long PGRP isoform has been identified in humans and mice (7, 24). Porcine PGRP long isoforms likely represent alternative splicing variants. Analysis of the full cDNA sequences suggests that the two identical regions of pPGRP-L1 and pPGRP-L2 are transcribed from the same exons; however, the initial 5'-terminal 79 bp of pPGRP-L1 and the middle 982 bp (positions 170 to 1151) of pPGRP-L2 are likely from isoform-specific exons. The two pPGRP long isoforms also show differential expression and regulation patterns in tissues from control and Salmonella-challenged animals. pPGRP-L1 is expressed constitutively in most tissues, and gene expression is not influenced by Salmonella infection. pPGRP-L2 has a more discrete tissue expression pattern and is markedly upregulated in intestinal tissues by Salmonella infection. In contrast, in an intestinal cell line both isoforms are influenced by in vitro challenge with E. coli, S. enterica, and, most dramatically, L. monocytogenes. The difference between the in vitro and in vivo findings related to Salmonella challenge is likely associated with the kinetics of the in vitro challenge (0 to 12 h) and the in vivo infection (48 h postchallenge).
Different activities have been described for PGRP isoforms. For example, Drosophila PGRP-SC1B is an N-acetylmuramoyl-L-alanine amidase (28), an activity that also is conserved in human and mouse PGRP-L isoforms (11, 42). Both porcine long PGRP isoforms are similar to human and mouse PGRP long isoforms in the domain regions that have been shown to function as N-acetylmuramoyl-L-alanine amidases (11, 28, 42). In particular, several amino acid residues (His411, Cys419, His436, Tyr447, and Cys530) which are required for amidase activity of the human long isoform of PGRP (11, 42) are also conserved in the porcine long PGRPs. To confirm this predication, we performed peptidoglycan binding and lytic assays with the expressed recombinant porcine PGRPs and showed that they have N-acetylmuramoyl-L-alanine amidase activity. Why pigs have two PGRP long isoforms with N-acetylmuramoyl-L-alanine amidase activity is not known. Our predictions suggest that pPGRP-L1 is a cytoplasmic protein and pPGRP-L2 is a transmembrane protein, likely associated with the Golgi. Thus, it is possible that the two porcine isoforms function in different cellular locations. Similarly, the cellular location of the porcine PGRP long isoforms likely influences their spectrum of activity against intracellular and extracellular bacteria.
As their name implies, recognition of microbial patterns is another potential function associated with PGRPs. Drosophila PGRP-LC, -LE, -SA, and -SD are pattern recognition receptors that in response to bacterial challenge signal through the Toll or IMD pathway, resulting in the expression of antimicrobial peptides (4, 6, 7, 12, 13, 29, 32, 40). Moreover, overexpression of Drosophila PGRP-LC or PGRP-LE, without pathogen stimulation, results in upregulation of antimicrobial peptides (12, 13, 32, 40). For Drosophila, the general thought is that a PGRP may have either peptidoglycan-sensing activity (such as with PGRP-SA and -LC) or amidase activity (such as with PGRP-SC1B). However, recent reports indicate a bifunctional aspect of PGRP-SA. In addition to sensing bacterial infection and activating the Toll signaling pathway, it also digests diaminopimelic acid-type peptidoglycan from gram-negative bacteria (5). Besides its potential amidase activity, the crystal structure of PGRP-LB indicates that a hydrophobic groove at the back face of its active cleft may function in peptidoglycan signaling (19, 28).
Recently, TLR2-deficient mice were found to display normal antibacterial capabilities in response to polymicrobial or L. monocytogenes peritonitis (9, 44). Toll-like receptors and receptors of the interleukin 1 family transduce signals through myeloid differentiation factor 88; however, myeloid differentiation factor 88-deficient mice still control S. aureus and L. monocytogenes infections (37, 43). In addition, peritoneal macrophages from PGRP-L-deficient mice produce less interleukin 6 and tumor necrosis factor alpha when stimulated with bacteria, which are considered TLR2-mediated events (20, 46). Taken together, these findings suggest that other pattern recognition receptors, perhaps PGRPs, are involved in these bacterial infections (35). Because PGRP amidase activity is not sufficient to lyse intact bacteria (19), our data further imply that overexpression of porcine long PGRP stimulates the expression of other innate immune effectors, such as antimicrobial peptides, which in turn may limit bacterial growth.
In addition to the two porcine PGRP long isoforms described in this report, the porcine genome also has two short PGRP isoforms, pPGRP-S1 and pPGRP-S2 (GenBank accession numbers AY598969 and AF541957). Porcine PGRP-S1, which was originally identified as porcine neutrophil cationic protein (10), is deposited as a partial sequence in GenBank (accession number AJ310355), and we subsequently obtained the full cDNA sequence (accession number AY598969). The regulation and function of these two short PGRP isoforms remain to be determined.
In conclusion, we have identified two long isoforms of porcine PGRPs, which are highly conserved with human and mouse orthologs. Our findings indicate that pPGRP-L1 is expressed constitutively in several tissues and that expression of pPGRP-L2 is increased by Salmonella infection. Both long isoforms of porcine PGRP bind peptidoglycan and have N-acetylmuramoyl-L-alanine amidase activity. Loss-of-function and gain-of-function experiments showed that these two pPGRPs are involved in -defensin-1 expression. Taken together, our findings indicate that PGRPs potentially mediate porcine innate immune responses through regulation of the expression of antimicrobial peptides.
ACKNOWLEDGMENTS
We thank Danielle Goodband and Ling Zheng for their excellent technical support and J. Ernest Minton for assistance with collection of tissues from Salmonella-infected animals.
This work was supported in part by National Research Initiative Competitive Grant 2001-35204-10818 (to F.B. and C.R.R.) from the U.S. Department of Agriculture.
REFERENCES
1. Aderem, A., and K. D. Smith. 2004. A systems approach to dissecting immunity and inflammation. Semin. Immunol. 16:55-67.
2. Arnold, A. S., M. Gueye, P. Ronde, J. M. Warter, P. Poindron, and J. P. Gies. 2002. Construction of a plasmid containing human SMN, the SMA determining gene, coupled to EGFP. Plasmid 47:79-87.
3. Beutler, B. 2004. Innate immunity: an overview. Mol. Immunol. 40:845-859.
4. Bischoff, V., C. Vignal, I. G. Boneca, T. Michel, J. A. Hoffmann, and J. Royet. 2004. Function of the Drosophila pattern-recognition receptor PGRP-SD in the detection of Gram-positive bacteria. Nat. Immunol. 5:1175-1180.
5. Chang, C. I., S. Pili-Floury, M. Herve, C. Parquet, Y. Chelliah, B. Lemaitre, D. Mengin-Lecreulx, and J. Deisenhofer. 2004. A Drosophila pattern recognition receptor contains a peptidoglycan docking groove and unusual L,D-carboxypeptidase activity. PLoS Biol 2:E277.
6. Choe, K-M., T. Werner, S. Stven, D. Hultmark, and K. V. Anderson. 2002. Requirement for a peptidoglycan recognition protein (PGRP) in Relish activation and antibacterial immune responses in Drosophila. Science 296:359-362.
7. Dziarski, R. 2004. Peptidoglycan recognition proteins. Mol. Immunol. 40:877-886.
8. Dziarski, R., K. A. Platt, E. Gelius, H. Steiner, and 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.
9. Edelson, B. T., and E. R. Unanue. 2002. MyD88-dependent but Toll-like receptor 2-independent innate immunity to Listeria: no role for either in macrophage listericidal activity. J. Immunol. 169:3869-3875.
10. Fornhem, C., C. G. Peterson, and K. Alving. 1996. Isolation and characterization of porcine cationic eosinophil granule proteins. Int. Arch. Allergy Immunol. 110:132-142.
11. Gelius, E., C. Persson, J. Karlsson, and H. Steiner. 2003. A mammalian peptidoglycan recognition protein with N-acetylmuramoyl-L-alanine amidase activity. Biochem. Biophys. Res. Commun. 306:988-994.
12. Gobert, V., M. Gottar, A. Matskevich, A., S. Rutschmann, J. Royet, M. Belvin, J. A. Hoffmann, and D. Ferrandon. 2003. Dual activation of the Drosophila toll pathway by two pattern recognition receptors. Science 302:2126-2130.
13. Gottar, M., V. Gobert, T. Michel, M. Belvin, G. Duyk, J. A. Hoffman, D. Ferrandon, and J. Royet. 2002. The Drosophila immune response against Gram-negative bacteria is mediated by a peptidoglycan recognition protein. Nature 416:640-644.
14. Guan, R., A. Roychowdhury, B. Ember, S. Kumar, G. J. Boons, and R. A. Mariuzza. 2004. Structural basis for peptidoglycan binding by peptidoglycan recognition proteins. Proc. Natl. Acad. Sci. USA 101:17168-17173.
15. Hazenberg, M. P., and H. de Visser. 1992. Assay for N-acetylmuramyl-L-alanine amidase in serum by determination of muramic acid released from the peptidoglycan of Brevibacterium divaricatum. Eur. J. Clin. Chem. Clin. Biochem. 30:141-144.
16. Janeway, C. A., Jr., and R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197-216.
17. Kang, D., G. Liu, A. Lundstrm, E. Gelius, and H. Steiner. 1998. A peptidoglycan recognition protein in innate immunity conserved from insects to humans. Proc. Natl. Acad. Sci. USA 95:10078-10082.
18. Khush, R. S., F. Leulier, and B. Lemaitre. 2002. Pathogen surveillance—the flies have it. Science 296:273-275.
19. Kim, M. S., M. Byun, and B. H. Oh. 2003. Crystal structure of peptidoglycan recognition protein LB from Drosophila melanogaster. Nat. Immunol. 4:787-793.
20. Knapp, S., C. W. Wieland, C. van 't Veer, O. Takeuchi, S. Akira, S. Florquin, and T. van der Poll. 2004. Toll-like receptor 2 plays a role in the early inflammatory response to murine pneumococcal pneumonia but does not contribute to antibacterial defense. J. Immunol. 172:3132-3138.
21. Krull, M., R. Nost, S. Hippenstiel, E. Domann, T. Chakraborty, and N. Suttorp. 1997. Listeria monocytogenes potently induces up-regulation of endothelial adhesion molecules and neutrophil adhesion to cultured human endothelial cells. J. Immunol. 159:1970-1976.
22. Lee, J. C., and G. C. Stewart. 2003. Essential nature of the mreC determinant of Bacillus subtilis. J. Bacteriol. 185:4490-4498.
23. Liu, C., E. Gelius, G. Liu, H. Steiner, and 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.
24. Liu, C., Z. Xu, D. Gupta, and R. Dziarski. 2001. Peptidoglycan recognition proteins: a novel family of four human innate immunity pattern recognition molecules. J. Biol. Chem. 276:34686-64694.
25. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(–C(T)) method. Methods 25:402-408.
26. Mathur, P., B. Murray, T. Crowell, H. Gardner, N. Allaire, Y. M. Hsu, G. Thill, and J. P. Carulli. 2004. Murine peptidoglycan recognition proteins PglyrpI-alpha and PglyrpI-beta are encoded in the epidermal differentiation complex and are expressed in epidermal and hematopoietic tissues. Genomics 83:1151-1163.
27. Medzhitov, R., and C. A. Janeway, Jr. 2002. Decoding the patterns of self and nonself by the innate immune system. Science 296:298-300.
28. Mellroth, P., J. Karlsson, and H. Steiner. 2003. A scavenger function for a Drosophila peptidoglycan recognition protein. J. Biol. Chem. 278:7059-7064.
29. Michel, T., J. M. Reichhart, J. A. Hoffman, and J. Royet. 2001. Drosophila Toll is activated by gram-positive bacteria through a circulating peptidoglycan recognition protein. Nature 414:756-759.
30. Pasare, C., and R. Medzhitov. 2004. Toll-like receptors and acquired immunity. Semin. Immunol. 16:23-26.
31. Ramanathan, B., H. Wu, C. R. Ross, and F. Blecha. 2004. PR-39, a porcine antimicrobial peptide, inhibits apoptosis: involvement of caspase-3. Dev. Comp. Immunol. 28:163-169.
32. Rmet, M., P. Manfruelli, A. Pearson, B. Mathey-Prevot, and R. A. Ezekowitz. 2002. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416:644-648.
33. Reis e Sousa, C. 2004. Activation of dendritic cells: translating innate into adaptive immunity. Curr. Opin. Immunol. 16:21-25.
34. Rhoads, J. M., W. Chen, P. Chu, H. M. Berschneider, R. A. Argenzio, and A. M. Paradiso. 1994. L-Glutamine and [scap]l-asparagine stimulate Na+-H+ exchange in porcine jejunal enterocytes. Am. J. Physiol. 266:828-838.
35. Sabroe, I., R. C. Read, M. K. Whyte, D. H. Dockrell, S. N. Vogel, and S. K. Dower. 2003. Toll-like receptors in health and disease: complex questions remain. J. Immunol. 171:1630-1635.
36. Sang, Y., M. T. Ortega, F. Blecha, O. Prakash, and T. Melgarejo. 2005. Molecular cloning and characterization of three beta-defensins from canine testes. Infect. Immun. 73:2611-2620.
37. Skerrett, S. J., H. D. Liggitt, A. M. Hajjar, and C. B. Wilson. 2004. Cutting edge: myeloid differentiation factor 88 is essential for pulmonary host defense against Pseudomonas aeruginosa but not Staphylococcus aureus. J. Immunol. 172:3377-3381.
38. Steiner, H. 2004. Peptidoglycan recognition proteins: on and off switches for innate immunity. Immunol. Rev. 198:83-96.
39. Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. Immunol. 21:335-376.
40. Takehana, A., T. Katsuyama, T. Yano, Y. Oshima, H. Takada, T. Aigaki, and S. Kurata. 2002. Overexpression of a pattern-recognition receptor, peptidoglycan-recognition protein-LE, activates imd/relish-mediated antibacterial defense and the prophenoloxidase cascade in Drosophila larvae. Proc. Natl. Acad. Sci. USA 99:13705-13710.
41. Tydell, C. C., N. Yount, D. Tran, J. Yuan, and 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.
42. Wang, Z. M., X. Li, R. R. Cocklin, M. Wang, M. Wang, K. Fukase, S. Inamura, S. Kusumoto, D. Gupta, and R. Dziarski. 2003. Human peptidoglycan recognition protein-L is an N-acetylmuramoyl-L-alanine amidase. J. Biol. Chem. 278:49044-49052.
43. Way, S. S., T. R. Kollmann, A. M. Hajjar, and C. B. Wilson. 2003. Cutting edge: protective cell-mediated immunity to Listeria monocytogenes in the absence of myeloid differentiation factor 88. J. Immunol. 171:533-537.
44. Weighardt, H., S. Kaiser-Moore, R. M. Vabulas, C. J. Kirschning, H. Wagner, and B. Holzmann. 2002. Cutting edge: myeloid differentiation factor 88 deficiency improves resistance against sepsis caused by polymicrobial infection. J. Immunol. 169:2823-2827.
45. Wu, H., G. Zhang, J. E. Minton, C. R. Ross, and F. Blecha. 2000. Regulation of cathelicidin gene expression: induction by lipopolysaccharide, interleukin-6, retinoic acid, and Salmonella enterica serovar Typhimurium infection. Infect. Immun. 68:5552-5558.
46. Xu, M., Z. Wang, R. M. Locksley. 2004. Innate immune responses in peptidoglycan recognition protein L-deficient mice. Mol. Cell. Biol. 24:7949-7957.
47. Yoshida, H., K. Kinoshita, and M. Ashida. 1996. Purification of a peptidoglycan recognition protein from hemolymph of the silkworm, Bombyx mori. J. Biol. Chem. 271:13854-13860.
48. Zhang, G., H. Hiraiwa, H. Yasue, H. Wu, C. R. Ross, D. Troyer, and F. Blecha. 1999. Cloning and characterization of the gene for a new epithelial -defensin. J. Biol. Chem. 274:24031-24037.
49. Zhang, G., C. R. Ross, S. S. Dritz, J. C. Nietfeld, and F. Blecha. 1997. Salmonella infection increases porcine antibacterial peptide concentration in serum. Clin. Diagn. Lab. Immunol. 4:774-777.(Yongming Sang, Balaji Ram)