Journal of Leukocyte Biology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published online as doi:10.1189/jlb.1205702 on June 7, 2006

Published online before print June 7, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1205702v1
80/2/415    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sweet, L.
Right arrow Articles by Schorey, J. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sweet, L.
Right arrow Articles by Schorey, J. S.
(Journal of Leukocyte Biology. 2006;80:415-423.)
© 2006 by Society for Leukocyte Biology

Glycopeptidolipids from Mycobacterium avium promote macrophage activation in a TLR2- and MyD88-dependent manner

Lindsay Sweet and Jeffrey S. Schorey1

Department of Biological Sciences, Center for Tropical Disease Research and Training, University of Notre Dame, Indiana

1Correspondence: Department of Biology, University of Notre Dame, 130 Galvin Life Science Center, Notre Dame, IN 46556. E-mail: schorey.1{at}nd.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The Toll-like receptors (TLRs) are key components in the immune response against numerous pathogens. Previous studies have indicated that TLR2 plays an essential role in promoting immune responses against mycobacterial infections. Prior work has also shown that mice deficient in TLR2 are more susceptible to infection by Mycobacterium tuberculosis, Mycobacterium bovis bacillus Calmette-Guerin, and Mycobacterium avium. Therefore, it is important to define the molecules expressed by pathogenic mycobacteria, which bind the various TLRs. Although a number of TLR agonists have been characterized for M. tuberculosis, no specific TLR ligand has been identified in M. avium. We have found that glycopeptidolipids (GPLs), which are highly expressed surface molecules on M. avium, can stimulate the nuclear factor-{kappa}B pathway as well as mitogen-activated protein kinase p38 and Jun N-terminal kinase activation and production of proinflammatory cytokines when added to murine bone marrow-derived macrophages. This stimulation was dependent on TLR2 and myeloid differentiation primary-response protein 88 (MyD88) but not TLR4. M. avium express apolar and serovar-specific (ss)GPLs, and it is the expression of the latter that determines the serotype of a particular M. avium strain. It is interesting that the ssGPLs activated macrophages in a TLR2- and MyD88-dependent manner, and no macrophage activation was observed when using apolar GPLs. ssGPLs also differed in their ability to activate macrophages with Serovars 1 and 2 stimulating inhibitor of {kappa}B p38 and phosphorylation and tumor necrosis factor {alpha} (TNF-{alpha}) secretion, while Serovar 4 failed to stimulate p38 activation and TNF-{alpha} production. Our studies indicate that ssGPLs can function as TLR2 agonists and promote macrophage activation in a MyD88-dependent pathway.

Key Words: glycolipids • MAPK • cytokine • NF-{kappa}B


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Pattern recognition receptors (PRRs) expressed on macrophages and other leukocytes play a fundamental role in stimulating a host immune response against foreign invaders [1 ]. This group of receptors recognizes molecular patterns, which are commonly expressed by pathogens. Recognition of these pathogen-associated molecules by macrophage PRRs can lead to phagocytosis of the pathogen and host cell signaling processes [1 ]. Some of the more well-studied PRR members include the mannose receptor, complement receptors, and Toll-like receptors (TLRs).

Since the discovery in 1996 that the Toll receptor in Drosophila was important for antifungal activity [2 ] and subsequent studies in humans and other mammals that homologs of the Toll receptor exist, there has been much interest in what is now designated as the TLR family. Initial studies, which indicated that TLR4 was required for a macrophage response to lipopolysaccharide (LPS) and that mice deficient in TLR4 expression were hyporesponsive to LPS, have only intensified the interest in TLRs [3 ]. At present, there are 11 members of the TLR family, which recognize a diverse set of molecules from RNA and DNA to lipids, glycolipids, and lipoproteins [4 ].

The TLRs have been studied extensively in regard to Mycobacterium tuberculosis infections. Ligands have been identified for TLR2, TLR4, and TLR9 and include the 19-kDa lipoprotein, heat shock protein 65, and mycobacterial DNA, respectively [5 , 6 ]. The importance of TLR2 and TLR4 in controlling a M. tuberculosis infection using a mouse model has also been studied extensively [5 ]. However, the importance of the TLRs in controlling other mycobacterial infections and the TLR ligands present on these different mycobacteria is less well defined. Control of Mycobacterium bovis bacillus Calmette-Guerin in mice after an intraperitoneal infection was dependent on TLR2 with a minimal role for TLR4 [7 ]. Mycobacterium avium, which is a major opportunistic pathogen in AIDS patients and a significant cause of increased morbidity and mortality in human immunodeficiency virus-infected individuals, has also been evaluated in the context of TLRs. Studies by Feng et al. [8 ] determined that mice deficient in TLR2 had increased bacterial load and increased susceptibility to M. avium infection compared with wild-type (WT) mice. TLR4-deficient mice were similar to WT mice in susceptibility to M. avium infection [8 ]. This is in agreement with in vitro studies, which indicate that M. avium can stimulate a response through TLR2 but not TLR4 [9 ]. It is interesting that myeloid differentiation primary response protein 88 (MyD88) knockout (KO) mice failed to control M. avium growth and succumbed to infection by 9–14 weeks [8 ]. MyD88 is a common adaptor molecule coupled to many of the TLRs [10 ].

The in vitro and in vivo results clearly indicate that TLR2 signaling is important in controlling a M. avium infection. However, no identification of a TLR2 ligand on the M. avium surface has been identified. We hypothesized that glycopeptidolipids (GPLs), expressed in conspicuous amounts on the M. avium surface, might be a TLR2 ligand. GPLs are expressed in a number of mycobacterial species including M. avium, Mycobacterium intracellulare, Mycobacterium scrofulaceum, Mycobacterium smegmatis, Mycobacterium chelonae, and Mycobacterium fortuitum. The GPLs consist of a tripeptide amino alcohol core modified with an amide-linked fatty acid, a methylated rhamnose (Rhap), and a 6-deoxytalose (6-dTal). In M. avium and M. intracellulare, GPLs can be modified further in length and composition of sugars attached to the 6-dTal residue. This variation has been used to classify these two species into 31 serotypes or serovars using antibodies specific for the different GPLs (for review, see ref. [11 ]).

The hypothesis that GPLs may be a TLR2 ligand was based on their known exposure to the extracellular environment and that GPLs have a lipopeptide core, and lipoproteins are known ligands for TLR2 [12 ]. Our data, using purified GPLs, indicate that these glycolipids can indeed promote macrophage activation and that this activation is TLR2- and MyD88-dependent.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Growth of M. avium strains
M. avium strains, kindly provided by Delphi Chatterjee and Julia Inamine (Colorado State University, Fort Collins), were streaked onto Middlebrook 7H11 (Difco, Detroit, MI) agar supplemented with 0.2% glycerol (v/v) and 10% oleic acid-albumin-dextrose-catalase as described [13 ]. Petri dishes were sealed with parafilm and incubated at 37°C for 6–8 weeks.

Extraction of GPLs
Mycobacteria were scraped from 7H11 plates and transferred directly to sterile screw-capped glass culture tubes and suspended in chloroform:methanol (2:1) at a cell:volume ratio of ~10:1. Lipids were extracted at 37°C overnight, and the lipid-containing organic layer was removed by centrifugation at 3000 revolutions per minute for 10 min. This step was repeated once more, and the organic layers were combined and then subjected to two to three Folch washes to remove hydrophilic contaminants. The organic layer was then dried down and weighed and resuspended in chloroform:methanol (2:1) at a known concentration. To check total lipid profiles, lipids were applied to Silica gel 60 thin-layer chromatography (TLC) plates (EM Sciences, Canada) and developed in chloroform:methanol:H2O (30:8:1) to separate the apolar and polar GPLs, which were visualized with a solution of 1% {alpha}-naphthol and 5% H2SO4 in ethanol and heated to 100°C to detect the characteristic pink-purple color as a result of the inherent 6-deoxyhexoses. Lipids were detected by spraying with a solution of 50% H2SO4 in water and heating to 100°C. Lipid-containing spots appeared yellow or brown.

Purification of GPLs
GPLs were purified from total lipid extracts by a two-step procedure. First, total lipid extracts were applied to preparative 20 x 20 Silica gel 60 TLC plates and separated under the conditions described above; however, the {alpha}-naphthol/sulfuric acid spray was only used to visualize a cut portion of the plate for the 6-deoxyhexose-containing lipids. The corresponding bands were then scraped from the remaining portion of the plate, and the lipids were extracted from the silica gel with chloroform:methanol (2:1). The silica was removed by centrifugation, the organic layer dried down, and the purity of the bands checked by TLC. Preparative TLC-purified bands of interest were pooled, dried down, dissolved in 0.5 mL ethyl acetate, and subsequently applied to a SEP-PAK cartridge (Waters Corp., Milford, MA) containing silica equilibrated with ethyl acetate. The glycolipids were eluted with ~33% methanol in ethyl acetate, dried down, and resuspended in 2 mL ethyl acetate. The partially purified GPLs were subjected to further purification using high-performance liquid chromatography (HPLC; Waters Corp., 600E), which was carried out on a 250 x 4.6-mm column packed with 5 µm YMC PVA-Sil (Waters. Corp.) [14 ]. The column was equilibrated with ethyl acetate, and the bound glycolipids were eluted using a gradient of 0–20% methanol in ethyl acetate over 40 min followed by a 20–100% methanol elution over 10 min [15 ]. Fractions (0.5 mL) were collected, checked by TLC, and pooled accordingly. The purified GPLs were dried down, weighed, and resuspended in methanol at known concentrations. All solvents were HPLC grade (Sigma Chemical Co., St. Louis, MO).

Sugar analyses
Deacylation of the HPLC-purified GPLs was performed using 0.2 N NaOH in methanol at 37°C for 40 min as described previously [16 ]. We used previously published TLC profiles of different serovar GPLs to help us make an initial determination as to whether the purified material was the serovar-specific (ss)- or nonspecific (ns)GPL. However, further analyses were required to confirm this designation. Therefore, alditol acetates of deacylated bands were prepared by carbohydrate hydrolysis (2 M trifluoroacetic acid, 121°C for 2 h), reduction [10 mg/mL sodium borodeuteride in ethanol:1 M aqueous ammonium (1:1), 25°C overnight], and acetylation (acetic anhydride, 100°C for 2 h). The prepared alditol acetates were then dissolved in chloroform, Folch-washed to remove excess salt, dried down, and then redissolved in minimal amounts of chloroform prior to gas chromatography-mass spectrometry (GC-MS) analyses. Glycosyl composition analyses were carried out on a Hewlett Packard HP6890 series GC connected to a JOEL JMS-GCmale MS. Derivatized sugars were separated on a 30-m ID 0.32-µm HP-5 column with an initial temperature of 50°C raised to 250°C at a rate of 12°C/min and held at 250°C for 2 min.

Bone marrow-derived macrophage (BMM{Phi}) isolation and cultivation
BMM{Phi}, used in all experiments, were isolated from 6- to 8-week-old C57BL/6 and BALB/c mice as described [13 ]. In brief, BM was isolated, and fibroblasts and mature macrophages were removed by selective adhesion. Isolated monocytes were cultured in Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY), supplemented with 20 mM HEPES, 10% fetal bovine serum, 100 U/mL penicillin, 100 µg/mL streptomycin, 1x L-glutamine, and 20% L cell supernatant as a source of macrophage-colony stimulating factor. Mature macrophages were harvested and frozen at –140°C after 7 days in culture. Thawed macrophages were cultured on nontissue culture plates for 3–7 days and then replated at 3 x 105 cells/35 mm tissue-culture plate.

GPL exposure to BMM{Phi}
Purified GPLs suspended in HPLC-grade methanol were coated onto 35 mm TC plates at the amount of 10 µg lipid per plate. The methanol was allowed to evaporate completely prior to adding macrophages, which were plated onto the TC plates as indicated above and incubated at 37°C in 5% CO2 for times indicated.

Western blots
GPL- and non-GPL-exposed macrophages were harvested at defined time-points. For tumor necrosis factor {alpha} (TNF-{alpha}) determination, the culture mediumwas collected and saved for subsequent enzyme-linked immunosorbent assays (ELISAs). To obtain cell lysates, the macrophages were washed with ice-cold phosphate-buffered saline three times and then treated for 5–10 min with ice-cold lysis buffer [150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, 1 µg/mL aprotinin, 1 µg/mL leupeptin, 1 µg/mL pepstatin, 1 mM pervanadate, 1 mM EDTA, 1% Igepal, 0.25% deoxycholic acid, 1 mM NaF, and 50 mM Tris-HCl (pH 7.4)]. Cell lysates were removed and stored at –20°C. Equal amounts of protein, as defined using the MicroBCA protein assay (Pierce, Rockford, IL), were loaded onto 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels, electrophoresed, and transferred onto a polyvinylidene difluoride membrane (Milipore, Bedford, MA). The membranes were blocked in Tris-buffered saline with 0.05% Tween 20 (TBST), supplemented with 5% skim powdered milk, and then incubated with primary antibodies against p38, extracellular signal-regulated kinase (ERK) 1/2, nuclear factor (NF)-{kappa}B, as well as phosphorylated p38, ERK 1/2, stress-activated protein kinase (SAPK)/Jun N-terminal kinase (JNK), inhibitor of {kappa}B{alpha} (I{kappa}B{alpha}), and NF-{kappa}B (Cell Signaling, Beverly, MA) in 5% bovine serum albumin as described [17 ]. The blots were washed with TBST, incubated with a secondary antibody, horseradish peroxidase-conjugated anti-rabbit or anti-mouse immunoglobulin G (Pierce), in TBST plus 5% powdered skim milk. Bound antibodies were detected using supersignal West Fempto-enhanced chemiluminescence reagents (Pierce).

ELISAs
Levels of cytokines secreted into the culture medium by lipid-exposed macrophages were measured by ELISA using the BD PharMingen (San Diego, CA) for TNF-{alpha}, Biosource (Camarillo, CA) for regulated on activation, normal T expressed and secreted (RANTES), and eBioscience (San Diego, CA) for interleukin (IL)-6 and IL-10 mouse ELISA kits. Cytokine levels in the culture medium were analyzed according to the manufacturer’s protocol, and cytokine concentrations were determined against a standard curve.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The M. avium 104 ssGPL activates the mitogen-activated protein kinase (MAPK) and NF-{kappa}B pathways and stimulates TNF-{alpha} release in BMM{Phi}
To determine if M. avium GPLs can stimulate a macrophage response, total lipids were extracted from M. avium 104 (Serovar 1) with chloroform:methanol (2:1) and subjected to preparative TLC and HPLC to purify apolar/ns- and ssGPLs as described in Materials and Methods. The purity of ns- and ssGPLs was determined by TLC (Fig. 1A ) and GC-MS. The GC spectrum of M. avium 104 ssGPL confirmed the presence of 3,4-di-O-methyl-rhamnose (m/z 131, 190, 234), 6-dTal, and rhamnose (both m/z 129, 171, 231), which constitute the Serovar 1 GPL (Fig. 1B) . GC-MS of the deacylated nsGPL band indicated a mixture of apolar GPLs with different methylation patterns (data not shown).


Figure 1
View larger version (45K):
[in this window]
[in a new window]
  		Figure 1. Purification of Serovar 1 GPLs and their activation of BMM{phi}. (A) TLCs showing the M. avium 104 total lipid (TL) profile and HPLC-purified, mixed apolar/ns (NS)- and ss (SS; Serovar 1)GPLs. Each sample (10 µg) was spotted, and the TLC was run in a CHCl3:CH3OH:H2O (30:8:1) solvent. The first TLC was stained with {alpha}-naphthol-H2SO4 in ethanol to detect 6-deoxyhexoses, and the second TLC was treated with 50% aqueous H2SO4 and heated to 100°C to detect total lipids. (B) GC chromatogram of the M. avium 104 ssGPL, confirming the presence of 3,4-di-O-methyl rhamnose (m/z 131, 190, 234), 6-dTal, and rhamnose (both with m/z 129, 171, 231), which constitute the Serovar 1 GPL. (C) BMM{Phi} were exposed to 10 µg M. avium 104 total lipid extract, HPLC-purified ns- and ssGPLs, or controls, as described in Materials and Methods. Cells were lysed, and the cell lysates were analyzed for phosphorylated (p)p38, SAPK/JNK, ERK 1/2, I{kappa}B{alpha}, and NF-{kappa}B by Western blot. Total p38, ERK 1/2, and NF-{kappa}B were detected as loading controls. Me, Methanol-treated plates; NT, nontreated macrophages. (D) BMM{Phi} were exposed to M. avium 104 total lipids, GPLs, or controls for 24 h. The conditioned media were removed and analyzed for TNF-{alpha} by ELISA. Values are expressed as mean + SD. Data are representative of three separate experiments.

 
The unfractionated total lipids as well as purified ns- and ssGPLs were suspended in methanol and coated onto 35 mm tissue-culture plates at 10 µg per plate. Murine BMM{Phi} were transferred onto the lipid-coated plates or control methanol-treated or untreated plates, and 6 h later, cells were lysed and analyzed for MAPK activation by measuring for phosphorylated p38, SAPK/JNK, and ERK 1/2. Total levels of p38 and ERK 1/2 were also determined. Results demonstrate that the M. avium 104 total lipid extract and Serovar 1 GPL activate p38 and JNK significantly greater than observed for nsGPL or control-treated cells (Fig. 1C) . In contrast, phosphorylated ERK 1/2 was observed in BMM{phi} treated with total lipid, ssGPL, and nsGPL but not in solvent control-treated cells (Fig. 1C) . We also evaluated the lysates for phosphorylated p65 NF-{kappa}B and I{kappa}B{alpha} as well as total NF-{kappa}B. Prior studies have shown that activation of the NF-{kappa}B pathway requires I{kappa}B{alpha} phosphorylation prior to its release from NF-{kappa}B. This release exposes the nuclear translocation signal of NF-{kappa}B and allows the transcription factor to translocate into the nucleus for subsequent phosphorylation and activation [18 ]. We found that p65 and I{kappa}B{alpha} were phosphorylated in BMM{phi} treated with total lipid extract or Serovar 1 GPL to a significantly greater extent than observed in cells treated with nsGPL or solvent alone (Fig. 1C) .

In separate experiments, culture supernatants were collected 24 h after BMM{phi} were plated on total lipid, GPL, or control wells, and this conditioned media was assayed for TNF-{alpha}. We found that total lipid and Serovar 1 GPL stimulated TNF-{alpha} production, and BMM{phi}, plated on nsGPL as well as control methanol and untreated plates, gave little to no detectable TNF-{alpha} (Fig. 1D) .

Serovar 1 GPL activation of the p38 and NF-{kappa}B pathways as well as the release of TNF-{alpha} is dependent on TLR2 but not TLR4
To determine whether the ssGPL-stimulated macrophage activation was through a TLR-dependent manner, we exposed C57BL/6 WT, BALB/c WT, TLR2–/– (C57BL/6 background), and TLR4–/– (BALB/c background) BMM{phi} to M. avium 104 unfractionated total lipids and ns- and ssGPL as described above. BMM{Phi} were lysed after 6 h, and phosphorylated p38, I{kappa}B{alpha}, and NF-{kappa}B were examined by Western blot. Using WT C57BL/6 BMM{phi}, phosphorylation of p38, I{kappa}B{alpha}, and NF-{kappa}B in cells treated with total lipids and the Serovar 1 GPL was above what was observed for nsGPLs or on methanol-treated plates (Fig. 2A ). In contrast, no p38, I{kappa}B{alpha}, or NF-{kappa}B phosphorylation above control levels was observed for any of the lipid samples when using TLR2–/– BMM{phi} (Fig. 2A) . Nevertheless, not all the signaling responses induced by the GPLs were TLR2-dependent, as ERK 1/2 activation was still observed in TLR2–/– BMM{phi} treated with nsGPL or ssGPL (data not shown). As expected, TNF-{alpha} was present in culture media of C57BL/6 WT BMM{phi} plated on M. avium 104 total lipid and ssGPL, but no TNF-{alpha} was detected when TLR2–/– BMM{phi} were used (Fig. 2B) . In contrast, the total lipid extract and ssGPL, but not the nsGPLs, activated p38, I{kappa}B{alpha}, and NF-{kappa}B in TLR4–/– BMM{phi}, similar to the levels observed in the BALB/c WT cells (Fig. 2C) . Moreover, TNF-{alpha} production was also secreted by TLR4–/– and WT BMM{phi} after stimulation with total lipids and the ssGPL (Fig. 2D) . Knocking out TLR4 did not diminish a response to M. avium 104 lipids, contrary to what was observed in TLR2–/– BMM{phi}, indicating a specific role for the TLR2 receptor.


Figure 2
View larger version (41K):
[in this window]
[in a new window]
  		Figure 2. Stimulation of macrophage activation by Serovar 1 GPL requires TLR2. (A) BMM{phi} isolated from WT or TLR2–/– C57Bl/6 mice were plated on M. avium 104 total lipid, nsGPL, ssGPL, or solvent (i.e., methanol)-coated tissue-culture plates or on untreated plates, as described in Materials and Methods. Six hours after BMM{phi} were added, cells were lysed, and the cell lysates were analyzed for phosphorylated p38, I{kappa}B{alpha}, and NF-{kappa}B by Western blot. Total p38 and NF-{kappa}B were used as loading controls. (B) BMM{phi} isolated from WT or TLR2–/– C57Bl/6 mice (BL/6) were plated on total lipid, nsGPL, ssGPL, or controls as described above. Twenty-four hours post-treatment, conditioned media were removed and analyzed for TNF-{alpha} levels by ELISA. (C and D) Experiments were performed as described for A and B, except BALB/c WT and TLR4–/– BMM{phi} were used as the source of BMM{phi}. TNF-{alpha} values are expressed as mean + SD. Data are representative of three separate experiments. RC, non-treated, resting macrophages.

 
TLR2-mediated BMM{phi} activation by M. avium 104 ssGPL is dependent on MyD88
As TLR2 is important for p38, I{kappa}B{alpha}, and NF-{kappa}B phosphorylation as well as TNF-{alpha} release after exposure to M. avium 104 total lipids and Serovar 1 GPL, we also tested whether this response was MyD88-dependent. Previous studies with TLR2 ligands have indicated an essential role for the adaptor protein MyD88 in the signaling response by this TLR [10 ]. MyD88–/– and WT BMM{phi} were exposed to the total lipid extract and ns- and ssGPL. As observed in Figure 2A , p38, I{kappa}B{alpha}, and NF-{kappa}B are phosphorylated in WT BMM{phi} upon exposure to total lipids and ssGPL but not upon exposure to nsGPL (Fig. 3A ). However, no activation of these signaling molecules was observed when the MyD88–/– BMM{phi}s were used (Fig. 3A) . Moreover, TNF-{alpha} production was observed in WT but not MyD88–/– BMM{phi} in response to total lipid and ssGPL (Fig. 3B) .


Figure 3
View larger version (21K):
[in this window]
[in a new window]
  		Figure 3. The TLR2-mediated activation of BMM{phi} by Serovar 1 GPL is dependent on MyD88. (A) WT or MyD88–/– C56Bl/6 BMM{phi} were plated on M. avium 104 total lipid, nsGPL, ssGPL, or controls as described in Figure 2 . Western blot analyses were performed on macrophage cell lysates using antibodies specific to phosphorylated p38, I{kappa}B{alpha}, and NF-{kappa}B. Total p38 and NF-{kappa}B were used as loading controls. (B) TNF-{alpha} levels in culture media from WT, TLR2–/–, and MyD88–/– BMM{phi} plated for 24 h on the M. avium 104 total lipid, nsGPL, ssGPL, or controls. TNF-{alpha} values are expressed as mean + SD. Data are representative of three separate experiments.

 
Production of RANTES and IL-6 by BMM{phi} is stimulated by Serovar 1 GPL in a TLR2- and MyD88-dependent manner
To determine whether the Serovar 1 GPL stimulated macrophages to produce other inflammatory mediators, BMM{phi} were added to tissue-culture wells coated with total lipid extract, nsGPL, or ssGPL as described above. Twenty-four hours later, culture supernatants were removed and analyzed by ELISA for RANTES, IL-6, or IL-10. As shown in Figure 4 , RANTES and IL-6 production was induced in BMM{phi} treated with total extracted lipid and ssGPL but not by nsGPL. We did not observe any IL-10 production with any of the lipid treatments (data not shown). As observed for TNF-{alpha} production, RANTES and IL-6 production was TLR2- and MyD88-dependent, as treatment of TLR2 or MyD88 KO BMM{phi} with the total lipid or ssGPL did not lead to RANTES or IL-6 production (Fig. 4A and 4C) . However, similar levels of the chemokine and cytokine were induced between treated WT and TLR4 KO BMM{phi} (Fig. 4B and 4D) .


Figure 4
View larger version (20K):
[in this window]
[in a new window]
  		Figure 4. Serovar 1 GPL stimulates BMM{phi} to produce RANTES and IL-6 in a TLR2/MyD88-dependent manner. The level of RANTES in the culture media from C57BL/6, TLR2–/–, and MyD88–/– BMM{phi} (A) or BALB/c and TLR4–/– BMM{phi} (B) after 24 h on M. avium 104 total lipid, nsGPL, ssGPL, or controls as defined by ELISA. (C and D) The same culture media used above were analyzed for IL-6 levels by ELISA. Cytokine values are expressed as mean + SD. Data are representative of three separate experiments.

 
Different ssGPLs stimulate BMM{phi} via a TLR2- and MyD88-dependent signaling response
To determine if the macrophage response was specific to Serovar 1 GPL, we isolated GPLs from two other serovars; M. avium 2151 smooth opaque (SmO; Serovar 2) and M. avium A5 (Serovar 4). Total lipids and GPLs were extracted and purified as depicted above. ns- and ssGPLs were again purified by preparative TLCs and HPLC, and the purity and composition of the GPLs were confirmed by TLC (Figs. 5A and 6A ) and GC-MS (data not shown). Sugar analyses confirm that the nsGPL band is a mixture of apolar GPLs with different methylation patterns. Analyses of the Serovar 2 GPL confirmed the presence of 3,4-di-O-methyl-rhamnose, 2,3-di-O-methyl fucose (m/z 118, 143, 203), 6-dTal, and rhamnose in equal amounts. Moreover, GC-MS analyses demonstrated equal amounts of 3,4-di-O-methyl rhamnose, 2-O-methyl fucose (m/z 118, 275), 4-O-methyl rhamnose (m/z 131, 262), 6-dTal, and rhamnose for the Serovar 4 GPL, confirming the purity of the ssGPLs (data not shown). Total lipids and GPLs were dissolved in methanol and exposed to BMM{phi}, as described in Materials and Methods. BMM{phi} isolated from C57BL/6 WT, TLR2–/–, and MyD88–/– were exposed to total lipids and ns- or ssGPLs of the different serovars. Confirming our previous data, Western blot analyses indicate that the nsGPLs did not stimulate significant phosphorylation of p38, I{kappa}B{alpha}, or NF-{kappa}B in any of the macrophage populations (Figs. 5B and 6B and data notshown). However, the total lipid extracts from Serovars 2 and 4 stimulated p38, I{kappa}B{alpha}, and NF-{kappa}B phosphorylation as well as TNF-{alpha} production when WT BMM{phi} were used (Figs. 5 and 6 and data not shown). It is interesting that Serovar 2 GPL-treated WT BMM{phi} showed increased p38 and I{kappa}B{alpha} phosphorylation and TNF-{alpha} production above untreated cells (Fig. 5B and 5C) , and Serovar 4 GPL-treated WT BMM{phi} showed increased I{kappa}B{alpha} phosphorylation but no p38 activation or TNF-{alpha} production (Fig. 6B and 6C) . As was observed for M. avium 104, the total lipid extract and ssGPLs from M. avium strains 2151 and A5 did not activate a signaling response in TLR2–/– (not shown) and My88D–/– BMM{phi} (Figs. 5B and 6B) , indicating that these lipids are also signaling through a TLR2- and MyD88-dependent pathway. Moreover, TNF-{alpha} production was dependent on TLR2 and MyD88, as no TNF-{alpha} was detected in culture supernatants of KO BMM{phi} following treatment with total lipid extracts or purified Serovar 2 GPL (Fig. 5C) .


Figure 5
View larger version (14K):
[in this window]
[in a new window]
  		Figure 5. Serovar 2 GPL activates p38 and NF-{kappa}B and stimulates TNF-{alpha} release in a TLR2- and MyD88-dependent manner. (A) TLC showing M. avium 2151 total lipids and HPLC-purified ns- and ss-2 GPLs. (B) WT or myD88–/– C56Bl/6 BMM{phi} were plated on M. avium 104 total lipid, nsGPL, ssGPL, or controls, as described in Figure 2 . Western blot analyses were performed on macrophage cell lysates using antibodies specific to phosphorylated p38 and I{kappa}B{alpha}. Total p38 (Tp38) was used as a loading control. (C) Culture supernatants from WT, TLR2–/–, and MyD88–/– BMM{phi} plated for 24 hours on M. avium 2151 total lipid, and GPLs or controls were analyzed for TNF-{alpha} levels by ELISA. TNF-{alpha} values are expressed as mean + SD. Data are representative of three separate experiments.

 

Figure 6
View larger version (12K):
[in this window]
[in a new window]
  		Figure 6. Serovar 4 GPL stimulates I{kappa}B{alpha} phosphorylation in a MyD88-dependent manner but fails to activate BMM{phi} to secrete TNF-{alpha}. (A) TLC showing M. avium A5 total lipid as well as HPLC-purified ns- and ss-4GPLs. (B) BMM{phi} isolated from WT and MyD88–/– C57Bl/6 mice were plated on M. avium 104 total lipid, nsGPL, ssGPL, or controls as described in Figure 2 . Macrophage cell lysates were analyzed by Western blot for phosphorylated p38 and I{kappa}B{alpha}. Total p38 was used as a loading control. (C) TNF-{alpha} levels in the culture supernatants of WT, TLR2–/–, and MyD88–/– BMM{phi} exposed for 24 h to M. avium A5 total lipid, nsGPL, or ssGPL. TNF-{alpha} values are expressed as mean + SD. Data are representative of three separate experiments.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Mycobacteria have a long history as infectious organisms and are the etiologic agents of numerous human diseases. One such disease is caused by the M. avium complex (MAC), which consists of three mycobacterial species: M. avium, M. intracellulare, and M. scrofulaceum. MAC is one of the most common, opportunistic pathogens found in AIDS patients, particularly in individuals with low CD4+ cells (50–100 cells/mm3) and is associated with increased morbidity and mortality in these patients [19 , 20 ]. Although all three species of mycobacteria classified under MAC have been isolated from AIDS patients, M. avium appears to account for the majority of infections, up to 90% in one study [21 ]. This may reflect the ubiquitous nature of M. avium in the environment, where it can be found in the water, soil, and house dust [22 , 23 ].

The GPLs are major surface components of M. avium and other mycobacterial species but are absent from M. tuberculosis. For M. avium and M. intracellulare, GPLs are known to be antigenic and show variation in the length and composition of sugars attached to the D-allo-threonine residue of the lipopeptide core. This variation has been used to classify MAC isolates into 31 serovars using antibodies specific for the different GPLs [11 ]. Of these, no less than 13 distinctive, serovar-specific oligosaccharides of MAC GPLs have been characterized in detail [24 ]. Moreover, the genetic locus responsible for glycosylation and methylation of the GPL lipopeptide core has been mapped to the M. avium ser gene cluster [25 , 26 ].

What is less clear about GPLs is their importance in M. avium pathogenesis. These glycolipids are present in copious amounts on the cell surface, and it has been speculated that the abundance of these molecules make cryptic many of the other cell wall lipoglycans (lipoarabinomannan, lipomannan, and phosphatidylinositol mannosides). Studies have shown some serotypes of GPLs to be isolated more frequently from AIDS patients than others (i.e., Serotypes 1, 4, and 8), suggesting an increased exposure or increased virulence of these serotypes [11 ]. Furthermore, GPLs have been associated with drug resistance [27 ]. There is also evidence that purified GPLs can modulate macrophage signaling pathways. GPLs from Serotype 8 induce high levels of prostaglandin E2 (PGE2) in treated human peripheral blood mononuclear cells (PBMC), and GPLs from Serovars 4 and 20 failed to induce PGE2 production. Serovar 8 GPL also induced TNF-{alpha} production in PBMC [28 ]. However, in this study, the receptors engaged by the GPLs and the macrophage signaling pathways initiated were not elucidated.

Our studies confirm that at least some ssGPLs can induce a proinflammatory response, and based on our current studies and previously published work, GPLs can stimulate human PBMC and mouse BMM{Phi} [29 ]. Moreover, our early studies indicated that p38 and NF-{kappa}B were required for TNF-{alpha} production following a M. avium infection [30 ]. ssGPLs appear to work through a similar signaling pathway to promote cytokine production, as ssGPLs can stimulate the p38, JNK, and NF-{kappa}B pathways.

p65 NF-{kappa}B and MAPK activation has been clearly linked to engagement of TLRs, and M. avium is known to signal via TLR2. Furthermore, previous studies indicate a high level of GPL surface expression [31 ]. Together, the data suggest that GPLs could be one of the TLR2 ligands expressed by M. avium. The experiments described in Figure 2 support this possibility with a lack of cell signaling and cytokine production following ssGPL treatment when using TLR2–/– murine macrophages. However, this is likely not the only TLR2 ligand, as total extractable lipids isolated from GPL-deficient M. avium strains can also induce a TLR2-dependent activation of macrophages (data not shown). This could be a result of other known TLR2 ligands shared among the different mycobacteria species, including phosphatidylinositol mannoside, which isolated from M. tuberculosis, has been shown to signal through TLR2 [32 ].

The adaptor molecule MyD88 is critical for TLR2 signaling and for signaling by other TLRs, although TLR4 can signal through a MyD88-independent manner as well [10 , 33 ]. Our studies also indicate that the ssGPL signaling via TLR2 requires MyD88 and suggests that the signaling complex, consisting of IL-1 receptor-associated kinase and TNF receptor-associated factor 6, is likely functioning downstream of ssGPL/TLR2 engagement. TLR2 typically functions as hetrodimers in combination with TLR1 or TLR6, and there appears to be some ligand specificity with lipoproteins, usually binding via TLR1/TLR2 and peptidoglycan via TLR6/TLR2 [34 ]. Defining which combination of TLRs is functioning to engage ssGPLs awaits further study.

Our results with Serovar 4 GPL signify that there are differences between GPLs in promoting a macrophage response. This parallels studies, which indicated differences between GPL serovars in their induction of PGE2 in human PBMC [28 ]. What accounts for this difference in macrophage response to various ssGPLs? An answer to this question awaits a better understanding of the molecular interaction between the GPLs and TLR2. However, it suggests that the sugars attached to the 6-dTal can influence the TLR2-mediated response. Nevertheless, even Serovar 4 GPL can function through TLR2 and MyD88 to promote NF-{kappa}B activation. This is not sufficient, however, to promote TNF-{alpha} production. We have previously shown that MAPK activation is required for TNF-{alpha} production by BMM{phi} following a mycobacterial infection, indicating that NF-{kappa}B is necessary but not sufficient for cytokine production [17 ].

In summary, we have found that ssGPLs isolated from different M. avium strains can function as TLR2 ligands and can stimulate a macrophage signaling response, which includes NF-{kappa}B and in some cases, p38 and JNK activation and TNF-{alpha} secretion. This is the first TLR2 ligand isolated specifically from M. avium and suggests that the differences between ssGPLs and how they initiate a TLR2 signaling event may account for some of the variations we see in a macrophage response to different M. avium serovars.


   ACKNOWLEDGEMENTS
 
This work was supported through Grants AI056979 and AI052439 from the National Institute of Allergy and Infectious Diseases. We are very grateful to Dr. Bill Boggess for his help and expertise in our GC-MS analysis of isolated GPLs.

Received November 30, 2005; revised March 16, 2006; accepted April 17, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Gordon, S. (2002) Pattern recognition receptors: doubling up for the innate immune response Cell 111,927-930[CrossRef][Medline]
  2. Lemaitre, B., Nicolas, E., Michaut, L., Reichhart, J. M., Hoffmann, J. A. (1996) The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults Cell 86,973-983[CrossRef][Medline]
  3. Poltorak, A., He, X., Smirnova, I., Liu, M. Y., Van Huffel, C., Du, X., Birdwell, D., Alejos, E., Silva, M., Galanos, C., Freudenberg, M., Ricciardi-Castagnoli, P., Layton, B., Beutler, B. (1998) Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene Science 282,2085-2088[Abstract/Free Full Text]
  4. Underhill, D. M. (2003) Toll-like receptors: networking for success Eur. J. Immunol. 33,1767-1775[CrossRef][Medline]
  5. Quesniaux, V., Fremond, C., Jacobs, M., Parida, S., Nicolle, D., Yeremeev, V., Bihl, F., Erard, F., Botha, T., Drennan, M., Soler, M. N., Le Bert, M., Schnyder, B., Ryffel, B. (2004) Toll-like receptor pathways in the immune responses to mycobacteria Microbes Infect. 6,946-959[CrossRef][Medline]
  6. Bulut, Y., Michelsen, K. S., Hayrapetian, L., Naiki, Y., Spallek, R., Singh, M., Arditi, M. (2005) Mycobacterium tuberculosis heat shock proteins use diverse Toll-like receptor pathways to activate pro-inflammatory signals J. Biol. Chem. 280,20961-20967[Abstract/Free Full Text]
  7. Heldwein, K. A., Liang, M. D., Andresen, T. K., Thomas, K. E., Marty, A. M., Cuesta, N., Vogel, S. N., Fenton, M. J. (2003) TLR2 and TLR4 serve distinct roles in the host immune response against Mycobacterium bovis BCG J. Leukoc. Biol. 74,277-286[Abstract/Free Full Text]
  8. Feng, C. G., Scanga, C. A., Collazo-Custodio, C. M., Cheever, A. W., Hieny, S., Caspar, P., Sher, A. (2003) Mice lacking myeloid differentiation factor 88 display profound defects in host resistance and immune responses to Mycobacterium avium infection not exhibited by Toll-like receptor 2 (TLR2)- and TLR4-deficient animals J. Immunol. 171,4758-4764[Abstract/Free Full Text]
  9. Lien, E., Sellati, T. J., Yoshimura, A., Flo, T. H., Rawadi, G., Finberg, R. W., Carroll, J. D., Espevik, T., Ingalls, R. R., Radolf, J. D., Golenbock, D. T. (1999) Toll-like receptor 2 functions as a pattern recognition receptor for diverse bacterial products J. Biol. Chem. 274,33419-33425[Abstract/Free Full Text]
  10. Akira, S., Hoshino, K. (2003) Myeloid differentiation factor 88-dependent and -independent pathways in Toll-like receptor signaling J. Infect. Dis. 187(Suppl. 2),S356-S363
  11. Chatterjee, D., Khoo, K. H. (2001) The surface glycopeptidolipids of mycobacteria: structures and biological properties Cell. Mol. Life Sci. 58,2018-2042[CrossRef][Medline]
  12. Heldwein, K. A., Fenton, M. J. (2002) The role of Toll-like receptors in immunity against mycobacterial infection Microbes Infect. 4,937-944[CrossRef][Medline]
  13. Bohlson, S. S., Strasser, J. A., Bower, J. J., Schorey, J. S. (2001) Role of complement in Mycobacterium avium pathogenesis: in vivo and in vitro analyses of the host response to infection in the absence of complement component C3 Infect. Immun. 69,7729-7735[Abstract/Free Full Text]
  14. Christie, W. W., Urwin, R. A. (1995) Separation of lipid classes from plant tissues by high performance liquid chromatography on chemical bonded stationary phase J. High Resolut. Chromatogr. 18,97-100
  15. Nishiuchi, Y., Kitada, S., Maekura, R. (2004) Liquid chromatography/mass spectrometry analysis of small-scale glycopeptidolipid preparations to identify serovars of Mycobacterium avium-intracellulare complex J. Appl. Microbiol. 97,738-748[CrossRef][Medline]
  16. Barrow, W. W., Ullom, B. P., Brennan, P. J. (1980) Peptidoglycolipid nature of the superficial cell wall sheath of smooth-colony-forming mycobacteria J. Bacteriol. 144,814-822[Abstract/Free Full Text]
  17. Roach, S. K., Schorey, J. S. (2002) Differential regulation of the mitogen-activated protein kinases by pathogenic and nonpathogenic mycobacteria Infect. Immun. 70,3040-3052[Abstract/Free Full Text]
  18. Mercurio, F., Manning, A. M. (1999) Multiple signals converging on NF-{kappa}B Curr. Opin. Cell Biol. 11,226-232[CrossRef][Medline]
  19. Abrams, E. J. (2000) Opportunistic infections and other clinical manifestations of HIV disease in children Pediatr. Clin. North Am. 47,79-108[CrossRef][Medline]
  20. Horsburgh, C. R., Jr (1999) The pathophysiology of disseminated Mycobacterium avium complex disease in AIDS J. Infect. Dis. 179(Suppl. 3),S461-S465
  21. Yajko, D. M., Chin, D. P., Gonzalez, P. C., Nassos, P. S., Hopewell, P. C., Reingold, A. L., Horsburgh, C. R., Jr, Yakrus, M. A., Ostroff, S. M., Hadley, W. K. (1995) Mycobacterium avium complex in water, food, and soil samples collected from the environment of HIV-infected individuals J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 9,176-182[Medline]
  22. von Reyn, C. F., Maslow, J. N., Barber, T. W., Falkinham, J. O., III, Arbeit, R. D. (1994) Persistent colonization of potable water as a source of Mycobacterium avium infection in AIDS Lancet 343,1137-1141[CrossRef][Medline]
  23. Yakrus, M. A., Good, R. C. (1990) Geographic distribution, frequency, and specimen source of Mycobacterium avium complex serotypes isolated from patients with acquired immunodeficiency syndrome J. Clin. Microbiol. 28,926-929[Abstract/Free Full Text]
  24. Chatterjee, D., Hunter, S. W., McNeil, M., Jardine, I., Brennan, P. J. (1989) Structure and function of mycobacterial glycolipids and glycopeptidolipids Acta Leprol. 7,81-84
  25. Belisle, J. T., McNeil, M. R., Chatterjee, D., Inamine, J. M., Brennan, P. J. (1993) Expression of the core lipopeptide of the glycopeptidolipid surface antigens in rough mutants of Mycobacterium avium J. Biol. Chem. 268,10510-10516[Abstract/Free Full Text]
  26. Eckstein, T. M., Silbaq, F. S., Chatterjee, D., Kelly, N. J., Brennan, P. J., Belisle, J. T. (1998) Identification and recombinant expression of a Mycobacterium avium rhamnosyltransferase gene (rtfA) involved in glycopeptidolipid biosynthesis J. Bacteriol. 180,5567-5573[Abstract/Free Full Text]
  27. Khoo, K. H., Jarboe, E., Barker, A., Torrelles, J., Kuo, C. W., Chatterjee, D. (1999) Altered expression profile of the surface glycopeptidolipids in drug-resistant clinical isolates of Mycobacterium avium complex J. Biol. Chem. 274,9778-9785[Abstract/Free Full Text]
  28. Barrow, W. W., Davis, T. L., Wright, E. L., Labrousse, V., Bachelet, M., Rastogi, N. (1995) Immunomodulatory spectrum of lipids associated with Mycobacterium avium serovar 8 Infect. Immun. 63,126-133[Abstract]
  29. Horgen, L., Barrow, E. L., Barrow, W. W., Rastogi, N. (2000) Exposure of human peripheral blood mononuclear cells to total lipids and serovar-specific glycopeptidolipids from Mycobacterium avium Serovars 4 and 8 results in inhibition of TH1-type responses Microb. Pathog. 29,9-16[CrossRef][Medline]
  30. Yadav, M., Roach, S. K., Schorey, J. S. (2004) Increased mitogen-activated protein kinase activity and TNF-{alpha} production associated with Mycobacterium smegmatis-but not Mycobacterium avium-infected macrophages requires prolonged stimulation of the calmodulin/calmodulin kinase and cyclic AMP/protein kinase A pathways J. Immunol. 172,5588-5597[Abstract/Free Full Text]
  31. Rulong, S., Aguas, A. P., da Silva, P. P., Silva, M. T. (1991) Intramacrophagic Mycobacterium avium bacilli are coated by a multiple lamellar structure: freeze fracture analysis of infected mouse liver Infect. Immun. 59,3895-3902[Abstract/Free Full Text]
  32. Jones, B. W., Means, T. K., Heldwein, K. A., Keen, M. A., Hill, P. J., Belisle, J. T., Fenton, M. J. (2001) Different Toll-like receptor agonists induce distinct macrophage responses J. Leukoc. Biol. 69,1036-1044[Abstract/Free Full Text]
  33. Kawai, T., Adachi, O., Ogawa, T., Takeda, K., Akira, S. (1999) Unresponsiveness of MyD88-deficient mice to endotoxin Immunity 11,115-122[CrossRef][Medline]
  34. Akira, S., Hemmi, H. (2003) Recognition of pathogen-associated molecular patterns by TLR family Immunol. Lett. 85,85-95[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Leukoc. Biol.Home page
D. J. Weiss, C. D. Souza, O. A. Evanson, M. Sanders, and M. Rutherford
Bovine monocyte TLR2 receptors differentially regulate the intracellular fate of Mycobacterium avium subsp. paratuberculosis and Mycobacterium avium subsp. avium
J. Leukoc. Biol., January 1, 2008; 83(1): 48 - 55.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
S. Bhatnagar and J. S. Schorey
Exosomes Released from Infected Macrophages Contain Mycobacterium avium Glycopeptidolipids and Are Proinflammatory
J. Biol. Chem., August 31, 2007; 282(35): 25779 - 25789.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
Y. Miyamoto, T. Mukai, Y. Maeda, N. Nakata, M. Kai, T. Naka, I. Yano, and M. Makino
Characterization of the Fucosylation Pathway in the Biosynthesis of Glycopeptidolipids from Mycobacterium avium Complex
J. Bacteriol., August 1, 2007; 189(15): 5515 - 5522.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
S. L. Young, L. Slobbe, R. Wilson, B. M. Buddle, G. W. de Lisle, and G. S. Buchan
Environmental Strains of Mycobacterium avium Interfere with Immune Responses Associated with Mycobacterium bovis BCG Vaccination
Infect. Immun., June 1, 2007; 75(6): 2833 - 2840.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
M. Yadav and J. S. Schorey
The beta-glucan receptor dectin-1 functions together with TLR2 to mediate macrophage activation by mycobacteria
Blood, November 1, 2006; 108(9): 3168 - 3175.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1205702v1
80/2/415    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sweet, L.
Right arrow Articles by Schorey, J. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sweet, L.
Right arrow Articles by Schorey, J. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS