Differential Involvement of Th1 and Th17 in Pathogenic Autoimmune Processes Triggered by Different TLR Ligands

Guangpu Shi, Barbara P. Vistica, Lindsey F. Nugent, Cuiyan Tan, Eric F. Wawrousek, Dennis M. Klinman and Igal Gery
J Immunol July 1, 2013, 191 (1) 415-423; DOI: https://doi.org/10.4049/jimmunol.1201732

Abstract

The interaction between TLRs and their cognate ligands triggers both the innate and adaptive immune systems, and thus can play a pivotal role in the defense against pathogen invasion. This work investigates the differentiation of naive CD4 cells into Th1 or Th17 phenotypes in mice treated with different TLR ligands. We use a model system in which naive transgenic cells specific to hen egg lysozyme are adoptively transferred into recipients that express hen egg lysozyme in the lens of the eye. The transferred naive T cells induce ocular inflammation only in recipients treated with TLR ligands. Treatment with LPS preferentially stimulated IL-17 production, whereas CpG oligodeoxynucleotide and polyinosinic:polycytidylic acid primarily stimulated Th1 cells. Peptidoglycan stimulated the two Th subpopulations equally. The preferential induction of Th1 or Th17 by the four ligands was detected in the spleen (where a major portion of the adoptively transferred cells homed) and in the eyes, where activated Th cells initiate inflammation. Analysis of the cytokines present in recipient mice suggests that Th1 induction is elicited by IL-12 and/or IFN-α, whereas Th17 generation is preferentially mediated by IL-6. Importantly, we show in this article that treatment with LPS selectively promoted in the recipient mice the generation of IL-6–producing activated B cells. An inverse correlation was found between the level of regulatory T cells and severity of inflammation induced by the donor cells. Taken together, our data show that specific TLR ligands differentially activate the immune system as evidenced by the generation of distinct Th phenotypes from naive CD4 cells.

Introduction

The family of TLRs plays a central role in innate immune defense by recognizing pathogen-associated molecular patterns and initiating response against them (1–4). Thus far, 11 TLRs have been identified in humans and 13 in mice, with different or overlapping ligands (2, 5). In addition to their role in protection against pathogens, however, TLRs have been found to be involved in the pathogenesis of diseases with autoimmune causative agents. Specifically, relationships were noted between TLR2 and inflammatory bowel diseases (IBDs), rheumatoid arthritis (RA), and multiple sclerosis (6–9), between TLR3 and RA and type 1 diabetes (10, 11), between TLR4 and IBD and RA, and between TLR9 and IBD, type 1 diabetes, and systemic lupus erythematosus (6, 7, 12, 13). It is assumed that the pathogenic responses are triggered, at least in part, by the interaction of TLRs with their ligands, and studies with experimental animals provide evidence to show that microbial products function as the TLR ligands in these presumed processes (4, 14, 15).

At the cellular level, the development of autoimmune disease is a complex process that involves escape from tolerance in the thymus and activation followed by proliferation of the self-reactive effector T cells (16). During the activation process, the self-reactive CD4+ T cells differentiate into subsets of Th cells, mainly Th1 and Th17, and these cells initiate the pathogenic process by recruiting a variety of immune cells into the target organ. The exact role TLRs play in the above processes is not entirely known. It has been reported that TLRs are abundantly expressed on some innate immune cells, such as macrophages and dendritic cells (DCs), as well as on epithelial cells (17), differentiated T cells (18), and Tregs (19, 20). Addition of ligands for TLR2, TLR3, TLR4, and TLR9 with DCs into naive CD4 T cell cultures was shown to promote both Th1 and Th17 differentiation, with Th17 differentiation being more prominent (18, 21). Unlike TLR ligand–induced Th17 differentiation, in which IL-6/TGF-β–mediated pathway is the major known mechanism, TLR ligand–induced Th1 differentiation can be mediated by two distinct pathways, through IL-12 or IFN-α (22–25). Activation of TLR4 by LPS and TLR9 by CpG oligodeoxynucleotide (ODN) induce Th1 responses through IL-12p70 release (26), whereas stimulation of IFN-α release by TLR3, TLR4, TLR7, and TLR9 has been shown to be another important driving force of TLR-mediated Th1 responses (23, 24).

In a previous study (15), we provided evidence to show that treatment with TLR ligands converts naive CD4 cells into pathogenic effector cells, capable of initiating inflammatory processes. In the experimental system we used (15), naive TCR-transgenic (Tg) CD4 cells with specificity to hen egg lysozyme (HEL) are adoptively transferred into recipients that expressed HEL in their eyes. Treatment with TLR ligands activates the naive cells to initiate inflammation in the recipients’ eyes. This study was aimed at using this experimental system to investigate the patterns of differentiation of naive CD4 cells into either Th1 or Th17 phenotypes in recipient mice treated with different ligands. The tested ligands included LPS, which is specific toward TLR4, polyinosinic:polycytidylic acid [poly (I:C)], specific to TLR3, peptidoglycan (PGN), specific for TLR2 and the CpG-ODN, specific to TLR9. Analysis of the differentiation was performed on the donor cells collected from the recipients’ inflamed eyes, as well as their spleens. Remarkable differences were observed among the cell populations collected from recipients treated with the four tested TLR ligands.

Materials and Methods

Mice

Tg mice expressing membrane-associated HEL in the lens under control of the αA-crystallin promoter, on an FVB/N background, were generated as described previously (15, 27). HEL-specific TCR-Tg mice (designated “3A9”), on a B10.BR background, were a gift from Mark Davis (Stanford University). Tg mice from each line were mated to produce (FVB/N × B10.BR) F1 hybrids that express HEL in their lens, or HEL-specific TCR, or both, or neither transgene. Only F1 mice were used in this study and included those expressing the HEL gene (“HEL-Tg”), the 3A9 TCR gene (“3A9”), or neither gene (“WT”). All manipulations were performed in compliance with the National Institutes of Health Resolution on the Use of Animals in Research.

Adoptive transfer of CD4 cells and induction of disease

Naive CD4 cells were isolated from spleens and lymph nodes of 3A9 mice using T cell columns and MACS beads as described in detail elsewhere (15, 27). Isolated CD4 cells (5 × 106) were injected i.v. into naive HEL-Tg mice. On the following day, recipient mice were injected i.v. with HEL and TLR ligands. The doses we used were 100 μg HEL (Sigma), 300 μg poly(I:C) (Sigma), 50 μg LPS (Difco), 300 μg PGN (Sigma), 240 μg CpG ODNs, GCTAGACGTTAGCGT and TCAACGTTGA, and they were synthesized at the Center for Biologics, U.S. Food and Drug Administration, core facility. All ODNs were free of endotoxin and protein contamination, as determined by chromogenic Limulus amebocyte lysate and bicinchoninic acid protein assays (Pierce Biotechnology, Rockford, IL), and were administered together. Eyes were collected on day 7 after cell transfer, and development of inflammation was determined histologically (15).

Preparation of ocular single-cell suspensions

On day 7 after adoptive transfer, eyes of recipients were collected, dissected, and digested in RPMI 1640 medium containing 10% FBS, 1 mg/ml collagenase D (Roche), and DNase I (Sigma-Aldrich) for 2 h. After digestion, ocular cells were mechanically dissociated and filtered. The single ocular cell suspensions were used for flow cytometric analysis.

Intracellular cytokine flow cytometry

To analyze intracellular IL-17 and IFN-γ expression, we stimulated splenocytes, isolated as described elsewhere (15, 27, 28), and eye-infiltrating cells, isolated as described earlier, with 20 ng/ml PMA and 1 μM ionomycin (Sigma-Aldrich) in the presence of GolgiStop (BD Biosciences) for 4 h, before staining. Cells were surface stained using mAb against murine CD4 conjugated with PerCP-Cy5.5 (BD Pharmingen) and a clonotypic mAb specific for the Tg TCR of the 3A9 mice, designated “1G12,” a gift from Emil Unanue (Washington University), conjugated with FITC. Ag-specific donor CD4 cells in recipients were identified by staining with anti–CD4-PerCP-Cy5.5 and 1G12-FITC. Anti-CD16/CD32 (BD Pharmingen) was used to block FcRs in all staining. Intracellular cytokines were stained using mAbs against IL-17 conjugated with PE and against IFN-γ conjugated with allophycocyanin (BD Pharmingen).

To analyze intracellular IL-6 expression, we stimulated isolated splenocytes with 10 μg/ml LPS and GolgiStop for 4 h. FcRs on cells were blocked using anti-CD16/CD32, followed by staining with anti–CD19-PE (BD Pharmingen) and anti–CD-80-FITC (BD Pharmingen). Intracellular IL-6 was stained using anti–IL-6 conjugated with allophycocyanin (BD Pharmingen). Data were acquired using FACSCalibur and analyzed with FlowJo (Tree Star).

For Foxp3 staining, isolated splenocytes or ocular cells were fixed and permeabilized with the Fixation/Permeabilization buffer for 1 h at 4°C before intracellular staining with allophycocyanin-conjugated anti-Foxp3 Ab following the procedure recommended by the manufacturer (eBioscience). Data were acquired and analyzed as described earlier.

Surface Ag analysis in eye-infiltrating cells

Single-cell suspensions of tested eyes were obtained as described earlier and were suspended in FACS buffer at 2 × 106/200 μl in each tube. Cells were incubated with anti-CD16/CD32 Ab for 5 min at room temperature to block FcRs, followed by incubation for 20 min at room temperature with a set of conjugated Abs including 1G12-FITC, anti–F4/80-FITC (eBioscience), and the following Abs from BD Pharmingen: anti–CD8-FITC, anti–Gr-1–FITC, anti–CD4-PE, anti–CD3-PE, and anti–CD11b-PE. Stained cells were washed twice in 2 ml FACS buffer. Data were acquired and analyzed as described earlier.

Testing the effect of anti–IL-6R Ab

Anti–IL-6R Ab, a gift from Dr. Rachel Caspi, National Eye Institute/National Institutes of Health, and IgG control (Jackson Immunoresearch Laboratories) were administered i.p. on days 0 and 2 after cell transfer at 0.8 mg/mouse. On day 3 after cell transfer, splenocytes were isolated, and intracellular IL-17 and IFN-γ expression were determined as described earlier.

ELISA

Blood was withdrawn from recipient mice 3 or 24 h after ligand injection. Serum was used to measure the levels of IL-6, IL-12 (p70), and IFN-α, using the appropriate ELISA kits (BioLegend) and according to the manufacturer’s instructions.

Statistical analysis

Unpaired two-tail t test was performed for comparison of the proportions of eye-infiltrating CD4 T cells or Tregs from donor or host. A p value ≤0.05 was defined as statistically significant.

Results

Differences among TLR ligands in their capacity to trigger pathogenicity by naive CD4 cells

We have previously shown that TCR Tg naive CD4 cells specific against HEL adoptively transferred into recipients expressing HEL in their eyes acquire pathogenic capacity when triggered in vivo by TLR ligands, to respond against HEL and mediate inflammation in the recipient eyes (15) (also shown in Fig. 1A). To learn more about the pathogenic response triggered by the different TLR ligands, we collected the eye-infiltrating cells of the different recipient groups and counted their numbers after differentiating between the CD4 T cells of the donor or host origin by immunostaining with the clonotypic mAb 1G12, which is specific for TCR of the donor cells (Fig. 1B). The inflammation severity level, as well as the proportion of donor cells in the infiltrate, were found to be the highest in eyes of mice treated with LPS; lower levels were found in eyes of mice treated with CpG, poly(I:C) and PGN, respectively.

FIGURE 1.
FIGURE 1.

Adoptively transferred naive CD4 T cells specific against HEL induce various levels of ocular inflammation in recipient mice expressing HEL in their lens after treatment with the tested TLR ligands. (A) Histological sections of eyes from recipient mice, 7 d after transfer of 5 million naive CD4 cells. Shown are the retinas at the optic nerve head area (a major entry site for the activated lymphocytes). Inflammatory cells infiltrating the retina and vitreous of ligand-treated mice, but not of the controls treated with PBS. (B) Flow cytometric analysis of cells collected from the recipient eyes, determining the proportions of CD4 cells of the donor (1G12+) and host (1G12−) cells among the total retrieved ocular cells. (C) Mean numbers of donor and host CD4 cells collected from inflamed eyes, indicating the various levels of severity of the inflammatory response. L, Lens; O, optic nerve head; R, retina; V, vitreous. *p < 0.05.

Differential involvement of Th1 and Th17 in ocular inflammation triggered by CD4 cells activated by different TLR ligands

To investigate the involvement of Th1 and Th17 cells in the immunopathogenic processes triggered by TLR ligands, we analyzed the eye-infiltrating cells for expression of IFN-γ and IL-17, respectively, in eyes of recipients treated with the four TLR ligands. We carried out the analysis by intracellular immunofluorescence and identified the donor cells by using the clonotypic mAb, 1G12. Data of a representative experiment are demonstrated in Fig. 2; similar observations were made in three other independent experiments. Remarkable differences were seen among the T cell populations collected from inflamed eyes of the four mouse groups in the proportions of Th1 or Th17, expressing IFN-γ or IL-17, respectively. Analysis of donor cells revealed predominance of Th17 cells in eyes of mice treated with LPS, similar proportions of Th1 and Th17 in eyes of mice treated with PGN, but predominance of Th1 in eyes of mice treated with CpG or poly(I:C). Unlike with donor cells (1G12+), host cells (1G12−) infiltrating the recipient mouse eyes predominantly included Th1 cells in all four treated mouse groups (Fig. 2). It is also of interest that the proportions of double-positive cells, expressing both IFN-γ and IL-17, were considerably higher among donor cells than among recipient cells (Fig. 2A).

FIGURE 2.
FIGURE 2.

Identification of Th1 and Th17 among the CD4 cells infiltrating the recipient eyes. (A) Intracellular flow cytometric analysis of cells collected from eyes of TLR ligand–treated recipient mice, on day 7 after cell transfer. A representative experiment, showing the remarkable differences among the mice treated with the four ligands. (B) Calculated mean ratios between Th17 and Th1 among donor and host cells, in four repeated experiments. Remarkable differences between Th1 and Th17 levels among donor cells in mice treated with the TLR ligands, but minimal variety among the host cell populations.

The differences among the mouse groups treated with the four TLR ligands are also summarized in Fig. 2B, in which the calculated ratios of Th17/Th1 in eyes of the different groups in repeated experiments are shown.

Different proportions of donor Th1 and Th17 cells in spleens of recipients treated with the different TLR ligands

Before invading the eye, adoptively transferred cells home to the recipient spleen, where they proliferate, presumably after exposure to their target Ag, HEL (15, 29), which is also injected into the recipient animals. Assuming that portions of the donor cells are also polarized during the activation process, whereas being exposed to the cytokine milieu of the spleen, we determined the proportions of cells expressing IFN-γ or IL-17, on day 3 after cell transfer. The data of a representative experiment are shown in Fig. 3A.

FIGURE 3.
FIGURE 3.

Identification of Th1 and Th17 among the CD4 cells in the recipient spleen. (A) Intracellular flow cytometric analysis of cells collected from spleens of TLR ligand–treated recipient mice, on day 3 postcell transfer. A representative experiment, showing differences among the mice treated with the four ligands, despite the Th1 population being predominant in spleens of all tested treatments. (B) Calculated mean ratios between Th17 and Th1 among donor and host cells, in three repeated experiments. Remarkable differences between Th1 and Th17 levels among donor cells in mice treated with the TLR ligands, but essentially no variation among the host cells.

The proportions of donor cells expressing IFN-γ or IL-17 differed among the four groups of mice, but to lesser degrees than those seen among the eye-infiltrating CD4 cells (Fig. 2). Interestingly, the donor Th1 population was predominant in all four mouse groups, and differences were noted only in the Th1 dominance level, with mice treated with CpG and poly(I:C) exhibiting the highest ratio between Th1 and Th17. It is of note, however, that the portion of cells staining for IL-17 was the highest in spleens of recipients treated with LPS. Similar effects of LPS on donor cell populations were observed in recipient mice treated with lower doses of the endotoxin, that is, 10 and 2 μg/mouse (data not shown). These quantitative differences among the four treated groups, observed in repeated experiments, are summarized in Fig. 3B. Also shown in Fig. 3 are data concerning the host (1G12−) CD4 subpopulations. As expected, essentially no differences were noted among the four groups in the dominance of the host Th1 subpopulation, because the host T cells did not recognize HEL and could not participate therefore in the polarization process.

The four TLR ligands induce different patterns of cytokine production by the recipient mouse cells

TLR ligands affect the immune system by interacting with TLRs expressed on lymphoid cells and activating these cells to produce and release cytokines including those that activate lymphocytes. These cytokines include IL-6, the major polarizing cytokine for Th17 differentiation (30, 31), as well as IL-12 and IFN-α, cytokines that play major roles in generation of Th1 cells (23, 31). We tested for the production of these cytokines by measuring their serum levels in treated recipients, 3 and 24 h after treatment with the TLR ligands, and by quantitative PCR analysis of their transcripts in the recipient spleens on days 1 and 3.

Expression analysis for the three transcripts in recipient spleens revealed that treatment with the ligands had no measureable effect on the substantial “baseline” levels of these transcripts, expressed by the control mouse spleens (data not shown). The TLR ligands, however, initiated secretion of the polarizing cytokines into the recipient serum. As shown in Fig. 4, the four tested ligands differed remarkably in the pattern of their effect on the production of the polarizing cytokines, IL-6, IL-12, and IFN-α. Treatment with LPS was superior among the four ligands in its stimulating IL-6 release into the serum, reaching excessive levels at 3 h and dropping to much lower levels at 24 h after ligand injection. LPS also stimulated high levels of IL-12p70, but only in the serum sample tested at 3 h. Low levels of IL-12p70 were also found in the serum of mice treated with CpG, but unlike in LPS-treated mice, their serum levels of IL-12p70 at the 24-h time point were higher than those at the 3-h time point. IFN-α, the other cytokine involved in the polarization toward Th1 phenotype (23, 24), was detected only at the 3-h time point in mice treated with LPS, CpG, and, in particular, poly(I:C).

FIGURE 4.
FIGURE 4.

Production of potentially polarizing cytokines by mice treated with the TLR ligands. Groups of HEL-Tg mice were injected with PBS or the four tested ligands, and blood samples were collected at the 3- or 24-h time points after cell injection. Cytokine levels were determined in the serum samples by ELISA.

Treatment with LPS selectively increased the proportion of IL-6–producing B cells

To examine the possibility that the mode of action of the TLR ligands is related to their selective effect on certain populations of APCs in the recipient mice, we analyzed by flow cytometry the proportions of DCs, macrophages, and B cells in spleens of recipients treated with the four TLR ligands. Data of a representative experiment are summarized in Fig. 5A. Spleen cells from recipients treated with LPS were inferior to spleen cells of recipients treated with the other TLR ligands in their proportions of activated DCs, and their percentages of activated macrophages were lower than those of recipients of CpG or poly(I:C). LPS recipients, however, were superior to recipients of other ligands in the proportions of activated B cells (CD19+CD80+) in their spleen (Fig. 5A). To further analyze the effect of LPS treatment, we determined by flow cytometry the proportion of IL-6–producing cells among spleen cells of LPS-treated mice. As shown in Fig. 5B, a considerable proportion of activated B cells (CD19+CD80+) expressed intracellular IL-6, whereas nonactivated B cells (CD19+CD80−) did not.

FIGURE 5.
FIGURE 5.

LPS treatment selectively activates IL-6–producing B cells. (A) Flow cytometry analysis of spleen cells from recipient mice, 3 d after naive CD4 cell transfer (on day 0) and treatment with the ligands (day 1). CD11c+ cells were costained for CD40, CD80, and CD86; F4/80 were costained for CD40; and CD19+ cells were costained for CD80. Different patterns of staining by cells treated with the four TLR ligands. Importantly, splenocytes of LPS-treated mice exhibit a unique staining pattern, characterized by high expression of CD80 in B cells. (B) Spleen cells from LPS-treated mice were gated on CD19+ cells and analyzed for IL-6 production, showing selective production by CD80+ cells. (C) Spleen cells from recipients treated with anti-IL6R Ab or IgG controls were analyzed for intracellular expression of IFN-γ and IL-17. Considerable selective reduction in IL-17 expression by donor CD4 cells in spleens of the Ab-treated mice.

Treatment with anti–IL-6R Ab reduced the proportion of Th17 cells

To further analyze the role of IL-6 in promoting the generation of Th17 cells in our system, we injected LPS-treated recipients with anti-IL-6R Abs. Mice treated with this Ab had lower proportions of Th17 cells in their spleens, as compared with controls (Fig. 5C). These data thus confirm the crucial role IL-6 plays in the generation of Th17 cells.

Variations in the development of the Treg phenotype by donor CD4 cells in recipients treated with the different TLR ligands

The phenotype of Tregs is generally identified by the expression of the transcription factor Foxp3 (32). Naive CD4 cells of the TCR Tg 3A9 mice express a marginal level of Foxp3 (Fig. 6A). To monitor the proportion of Treg phenotype among donor CD4 cells that accumulate and proliferate in the spleen of recipients treated by the four TLR ligands, or PBS, we analyzed these cells for the Foxp3 expression. Data of a representative experiment are shown in Fig. 6B and demonstrate remarkably higher levels of Foxp3+ cells than those seen in the naive 3A9 cell preparation (Fig. 6A). The highest percentage of Foxp3+ was observed in the control mice treated with PBS (∼5%); lower levels were found in mice treated with the TLR ligands.

FIGURE 6.
FIGURE 6.

Frequency of Foxp3+ Tregs in the spleen of donor and recipient mice. (A) Flow cytometric analysis of spleen cells from naive 3A9 (donor) mice, showing the very low level of Foxp3 in these TCR Tg mice. (B) Flow cytometric analysis of spleen cells of TLR ligand–treated recipient mice (HEL-Tg), collected on day 3 after adoptive transfer of five million naive 3A9 CD4 cells. Recipients treated with the ligands have lower levels of Foxp3 cells than the PBS-treated controls. (C) Calculated mean percentages of Foxp3+ cells among donor and recipient CD4 cells. *p < 0.05.

Unlike the low values of Foxp3 among the donor (1G12+) cells, ∼20% of the host (1G12−) CD4 cells in these spleens expressed this marker, with slight increases in TLR ligand–treated mice as compared with the PBS-treated controls. The difference between donor and host cells in their expression of Foxp3, summarized in Fig. 6C, is in line with the well-known observation of relatively low proportions of Tregs in Tg mice expressing TCR against a neo-antigen (33, 34).

Variability in expression of Foxp3 by CD4 cells infiltrating the recipient mouse eyes

Next, we analyzed the Foxp3 expression by CD4 cells that infiltrated the inflamed eyes of recipients treated with the TLR ligands. The data of a representative experiment are shown in Fig. 7A, and data of this and two other experiments are summarized in Fig. 7B. The data indicate that the proportions of Foxp3+ cells in the affected eyes among both donor and recipient CD4 cells were higher than those observed in the spleens of the same mice, shown in Fig. 6. In addition, high levels of variability were noted among the mice treated with the different TLR ligands in the percentage of Foxp3-expressing cells, with generally an inverted correlation with the severity of pathological changes. Thus, the levels of Foxp3+ cells in eyes of mice treated with LPS and CpG were lower than those in mice treated with PGN and poly(I:C).

FIGURE 7.
FIGURE 7.

Frequency of Foxp3+ Tregs among donor and recipient CD4 cells in the eyes of recipient mice. (A) Flow cytometric analysis of CD4 cells collected from eyes of TLR ligand–treated recipient mice, collected on day 3 after adoptive transfer of five million naive 3A9 CD4 cells. An inverted correlation is seen in general between severity of inflammation and level of Foxp3+ cells among both donor and host-infiltrating CD4 cells. (B) Calculated mean percentages of Foxp3+ cells among donor and recipient CD4 cells. *p < 0.05.

Similar to the observation with recipient spleens, the proportions of cells expressing Foxp3 among the host cells in the recipient mouse eyes were remarkably higher than those among the donor cells. In addition, the Foxp3+ levels among host CD4 cells in eyes of recipients treated with LPS were lower than those in eyes of mice treated with other ligands (Fig. 7B).

Participation of non-CD4 cells in the ocular inflammatory process

The inflammatory process initiated by CD4 lymphocytes encompasses cells from other populations, including mainly CD8 cells, macrophages, and PMNs. To determine the involvement of these non-CD4 populations, we collected infiltrating cells from eyes with inflammation induced by the four TLR ligands and examined them by flow cytometry for the expression of Ags specific to these populations, namely, CD8, F4/80, and Gr-1, respectively. The added Abs did not react with the donor cells. Flow cytometric data of a representative experiment are shown in Fig. 8A, and data of this and two additional experiments are summarized in Fig. 8B. Noticeable differences were seen among the eye infiltrates in recipients treated with the four TLR ligands. Observations of interest include: 1) relatively low levels of macrophages and PMNs in eyes of LPS-treated recipient mice, 2) high levels of CD8+ cells in PGN-treated recipients, and 3) high level of macrophages in recipients treated with CpG. Also of interest are differences among the four groups in the proportions of PMNs expressing high and low levels of Gr-1, representing mature and immature neutrophils, respectively (35).

FIGURE 8.
FIGURE 8.

Involvement of non-CD4 lymphoid cells in the inflammatory process in eyes of recipients treated with the TLR ligands. (A) Infiltrating cells were collected from recipient eyes on day 7 after naive CD4 cell injection and were examined by flow cytometry, with gating on CD45+ cell population, for expression of surface markers specific for CD8, macrophages, and granulocytes. (B) Calculated percent values of cells expressing the markers among the CD45+ cell population.

Discussion

Data reported in this study extend our previous study (15) that established an experimental system in which TLR ligands trigger processes that convert naive CD4 cells specific to HEL into pathogenic Th cells capable of initiating inflammation in eyes expressing HEL (15). This system thus provides evidence to support the notion that pathogenic autoimmunity could be initiated by microbial products that function as TLR ligands (see beginning of article). In this study, we analyzed the different patterns of involvement of Th1 and Th17 in the pathogenic processes triggered by four different TLR ligands.

We compared the proportions of donor and host Th1 and Th17 cells in both the eyes and the spleen of the recipient mice, according to the expression of intracellular IFN-γ and IL-17, respectively. Tracing the donor cells in the recipients revealed that a large portion of the transferred naive cells home to the spleen, where they are activated by the Ag, HEL (15), and are polarized by the cytokine milieu they are exposed to. Differences among the four groups in the proportions of cells acquiring the phenotypes of Th1 or Th17 (Fig. 3) indicate that the four ligands we used varied in the cytokine milieu they created in the recipient mice.

The selective effects of the four TLR ligands on the polarization process of in vivo–activated naive CD4 cells, as shown in this article with the HEL system, were reproduced in another experimental system in which OVA-specific TCR Tg cells from OTII donor mice were adoptively transferred into naive B6 mice. The pattern of selective polarization of Th1 and Th17 cells in mice treated with the four ligands closely resembled that seen in this study with the HEL system (data not shown).

The next phase of the pathogenic process includes the migration of activated and polarized Th cells out of the spleen and other lymphoid organs into the target tissue, the eye in our system. Not much is known about this phase, but it is conceivable that it exerts a certain level of preferentiality, which determines the proportions of Th1 and Th17 that invade the recipient mouse eye.

Remarkable differences were found among the four recipient groups with regard to the proportions of donor cells expressing the Th1 or Th17 phenotypes in the recipient eyes (Fig. 2). Predominance of Th17 was seen in eyes of recipients treated with LPS, in contrast with eyes of CpG and poly(I:C) recipients, with higher proportions of Th1. An even distribution of the two phenotypes was observed in recipients of PGN. These differences among eye-infiltrating donor cells in the four recipient groups are more pronounced than those observed in the corresponding mouse spleens (Fig. 3). In addition to the possible phenotype selectivity during the migration of activated Th cells into the eye, mentioned earlier, we suggest that the eye-invading donor cells undergo an additional phase of activation/polarization in the recipient eye. These cells, with specificity toward HEL, are presumably reactivated by their target Ag in the recipient eye and simultaneously are polarized by the cytokine milieu of the recipient eye.

The TLR ligands are administered intravenously and trigger rapid production of polarizing cytokines by cells of various cell tissues. A portion of these cytokines are secreted and could be measured in serum samples (Fig. 4). An excessive production of IL-6 was observed in the serum of mice treated with LPS; profoundly lower levels of IL-6 were produced by mice treated with the other TLR ligands. IL-6 is a cytokine crucial for polarization toward the Th17 phenotype (36), and mice treated with LPS were also superior to mice of the other treatment groups in their preferentially high levels of Th17 cells in their spleen and eyes (Figs. 2, 3). Our data also indicate that the superiority of LPS in promoting Th17 differentiation is due, at least in part, to its preferential activation of B cells, resulting in higher levels of released IL-6 (Fig. 5A, 5B). This observation sheds new light on the mode of action of LPS and is in accord with a recent publication showing that B cells participate in the pathogenic processes of autoimmune diseases, mainly by producing high levels of IL-6 (37). LPS is a strong stimulant for a variety of other cytokines, and mice treated with this agent also produced high levels of IL-12p70, a major polarizing cytokine for Th1, and developed large populations of these cells (Figs. 2, 3).

The relationship between the serum cytokine profile and generation of Th1 by mice treated with the four TLR ligands was found, however, to be complicated, with different patterns of association between the injected ligands and the Th phenotype profiles (Figs. 2, 3). We suggest that the complexity is due to the involvement of two different cytokines in the polarization of the Th1 subpopulation, namely, IL-12 and IFN-α (22–25). The highest prevalence of Th1 cells was found in mice treated with CpG, which produced moderate serum levels of both IL-12p70 and IFN-α (Fig. 4). Unlike CpG, treatment with poly(I:C) stimulated only high levels of IFN-α, but undetectable levels of IL-12p70 (Fig. 4). The association between poly(I:C) and IFN-α was indicated by the study of Longhi et al. (23), and our observations suggest that generation of Th1 in poly(I:C)-treated mice could be entirely attributed to increased production of IFN-α, whereas both IL-12 and IFN-α are involved in the preferential generation of Th1 cells in CpG-treated mice.

Another issue we addressed in this study concerns the potential involvement of Tregs in the pathogenic processes triggered by the TLR ligands. The population of donor TCR-Tg CD4 cells contained a very low proportion of Foxp3+ Tregs (<1%), in line with published reports (33, 34). After their homing to the spleen and activation by HEL, the proportion of Foxp3+ cells increased to ∼5% (Fig. 6). We assume the majority of these Foxp3+ donor cells are induced Tregs. Interestingly, the proportion of donor Foxp3+ cells was lower in spleens of mice treated with the four tested TLR ligands, as compared with PBS-treated controls (Fig. 6), an observation similar to that made by Summers et al. (38) in a system model of autoimmune glomerulonephritis. More research is needed to unravel the effect of TLRs on development of Tregs. Donor Foxp3+ cells were also found among the eye-infiltrating cells, and significantly, the proportions of these cells were lower in severely inflamed eyes of mice treated with LPS and CpG, as compared with less affected eyes of poly(I:C)- or PGN-treated mice (Figs. 1, 7). This observation thus suggests that the Ag-specific induced Tregs exerted a moderating effect on the inflammatory process in our system.

In addition to differences among the infiltrating Th cells, inflamed eyes of mice treated with the four TLR ligands varied in their populations of non-Th cells, namely, CD8, macrophages, and PMNs (Fig. 8). These differences suggest that the milieus of cytokines, which determine the selective infiltration of Th subpopulations, are also responsible for the preferential migration of the non-CD4 cells into the inflamed eye. It is of note that a large portion of the inflammatory cell population expressed CD8, but no CD3. It is assumed that these CD8+ cells include DC and NK cells that express CD8. All these cells are affected by TLR ligands (39, 40). Our finding that PGN selectively promotes CD8 cells is in accord with Cottalorda et al.’s (41) report that treatment with another TLR2 ligand, Pam3CysSK4, enhances activation of CD8 cells.

The flow cytometric analysis recorded in Fig. 8 also shows a separation between two populations of cells, expressing low or medium levels of Gr-1. We assume the Gr-1 “low” cells are mainly macrophages, in line with the data of Zehntner et al. (42). We also suggest that the Gr-1 “medium” population consists of immature PMNs and some macrophages.

In summary, this study provides new data concerning the process whereby microbial products, which function as ligands for TLRs, trigger differentiation of naive CD4 cells into pathogenic Th cells. Analysis of the newly generated effector cells revealed that the different tested TLR ligands varied in their activity, generating different proportions of Th1 and Th17 among the cells that acquired pathogenic capacity and initiated inflammation. These findings shed new light on the mechanism of autoimmune inflammation triggered by microbial infection and provide new strategies for their prevention and treatment.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank R. Steven Lee for tail DNA analysis, the National Eye Institute Flow Cytometry Core for technical support, and the Histology Core for tissue section preparations.

Footnotes

  • This research was supported by the Intramural Research Program of the National Eye Institute, National Institutes of Health.

  • Abbreviations used in this article:

    DC
    dendritic cell
    HEL
    hen egg lysozyme
    IBD
    inflammatory bowel disease
    ODN
    oligodeoxynucleotide
    PGN
    peptidoglycan
    poly (I:C)
    polyinosinic:polycytidylic acid
    RA
    rheumatoid arthritis
    Tg
    transgenic
    Treg
    regulatory T cell.

  • Received June 22, 2012.
  • Accepted April 25, 2013.

References

  1. ↵
    1. Mills K. H.
    2011. TLR-dependent T cell activation in autoimmunity. Nat. Rev. Immunol. 11: 807–822.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Takeda K.,
    2. T. Kaisho,
    3. S. Akira
    . 2003. Toll-like receptors. Annu. Rev. Immunol. 21: 335–376.
    OpenUrlCrossRefPubMed
    1. Lang K. S.,
    2. M. Recher,
    3. T. Junt,
    4. A. A. Navarini,
    5. N. L. Harris,
    6. S. Freigang,
    7. B. Odermatt,
    8. C. Conrad,
    9. L. M. Ittner,
    10. S. Bauer,
    11. et al
    . 2005. Toll-like receptor engagement converts T-cell autoreactivity into overt autoimmune disease. Nat. Med. 11: 138–145.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Akira S.,
    2. K. Takeda
    . 2004. Toll-like receptor signalling. Nat. Rev. Immunol. 4: 499–511.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Liew F. Y.,
    2. D. Xu,
    3. E. K. Brint,
    4. L. A. O’Neill
    . 2005. Negative regulation of toll-like receptor-mediated immune responses. Nat. Rev. Immunol. 5: 446–458.
    OpenUrlCrossRefPubMed
  6. ↵
    1. van Heel D. A.,
    2. S. Ghosh,
    3. M. Butler,
    4. K. Hunt,
    5. B. M. Foxwell,
    6. D. Mengin-Lecreulx,
    7. R. J. Playford
    . 2005. Synergistic enhancement of Toll-like receptor responses by NOD1 activation. Eur. J. Immunol. 35: 2471–2476.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Radstake T. R.,
    2. M. F. Roelofs,
    3. Y. M. Jenniskens,
    4. B. Oppers-Walgreen,
    5. P. L. van Riel,
    6. P. Barrera,
    7. L. A. Joosten,
    8. W. B. van den Berg
    . 2004. Expression of toll-like receptors 2 and 4 in rheumatoid synovial tissue and regulation by proinflammatory cytokines interleukin-12 and interleukin-18 via interferon-γ. Arthritis Rheum. 50: 3856–3865.
    OpenUrlCrossRefPubMed
    1. Visser L.,
    2. M. J. Melief,
    3. D. van Riel,
    4. M. van Meurs,
    5. E. A. Sick,
    6. S. Inamura,
    7. J. J. Bajramovic,
    8. S. Amor,
    9. R. Q. Hintzen,
    10. L. A. Boven,
    11. et al
    . 2006. Phagocytes containing a disease-promoting Toll-like receptor/Nod ligand are present in the brain during demyelinating disease in primates. Am. J. Pathol. 169: 1671–1685.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Seibl R.,
    2. T. Birchler,
    3. S. Loeliger,
    4. J. P. Hossle,
    5. R. E. Gay,
    6. T. Saurenmann,
    7. B. A. Michel,
    8. R. A. Seger,
    9. S. Gay,
    10. R. P. Lauener
    . 2003. Expression and regulation of Toll-like receptor 2 in rheumatoid arthritis synovium. Am. J. Pathol. 162: 1221–1227.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Roelofs M. F.,
    2. L. A. Joosten,
    3. S. Abdollahi-Roodsaz,
    4. A. W. van Lieshout,
    5. T. Sprong,
    6. F. H. van den Hoogen,
    7. W. B. van den Berg,
    8. T. R. Radstake
    . 2005. The expression of toll-like receptors 3 and 7 in rheumatoid arthritis synovium is increased and costimulation of toll-like receptors 3, 4, and 7/8 results in synergistic cytokine production by dendritic cells. Arthritis Rheum. 52: 2313–2322.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Roep B. O.
    1996. T-cell responses to autoantigens in IDDM. The search for the Holy Grail. Diabetes 45: 1147–1156.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Török H. P.,
    2. J. Glas,
    3. L. Tonenchi,
    4. G. Bruennler,
    5. M. Folwaczny,
    6. C. Folwaczny
    . 2004. Crohn’s disease is associated with a toll-like receptor-9 polymorphism. Gastroenterology 127: 365–366.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Leadbetter E. A.,
    2. I. R. Rifkin,
    3. A. M. Hohlbaum,
    4. B. C. Beaudette,
    5. M. J. Shlomchik,
    6. A. Marshak-Rothstein
    . 2002. Chromatin-IgG complexes activate B cells by dual engagement of IgM and Toll-like receptors. Nature 416: 603–607.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Waldner H.,
    2. M. Collins,
    3. V. K. Kuchroo
    . 2004. Activation of antigen-presenting cells by microbial products breaks self tolerance and induces autoimmune disease. J. Clin. Invest. 113: 990–997.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Fujimoto C.,
    2. C. R. Yu,
    3. G. Shi,
    4. B. P. Vistica,
    5. E. F. Wawrousek,
    6. D. M. Klinman,
    7. C. C. Chan,
    8. C. E. Egwuagu,
    9. I. Gery
    . 2006. Pertussis toxin is superior to TLR ligands in enhancing pathogenic autoimmunity, targeted at a neo-self antigen, by triggering robust expansion of Th1 cells and their cytokine production. J. Immunol. 177: 6896–6903.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Goodnow C. C.
    2007. Multistep pathogenesis of autoimmune disease. Cell 130: 25–35.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Strober W.
    2004. Epithelial cells pay a Toll for protection. Nat. Med. 10: 898–900.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Reynolds J. M.,
    2. B. P. Pappu,
    3. J. Peng,
    4. G. J. Martinez,
    5. Y. Zhang,
    6. Y. Chung,
    7. L. Ma,
    8. X. O. Yang,
    9. R. I. Nurieva,
    10. Q. Tian,
    11. C. Dong
    . 2010. Toll-like receptor 2 signaling in CD4(+) T lymphocytes promotes T helper 17 responses and regulates the pathogenesis of autoimmune disease. Immunity 32: 692–702.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Sutmuller R. P.,
    2. M. H. den Brok,
    3. M. Kramer,
    4. E. J. Bennink,
    5. L. W. Toonen,
    6. B. J. Kullberg,
    7. L. A. Joosten,
    8. S. Akira,
    9. M. G. Netea,
    10. G. J. Adema
    . 2006. Toll-like receptor 2 controls expansion and function of regulatory T cells. J. Clin. Invest. 116: 485–494.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Caramalho I.,
    2. T. Lopes-Carvalho,
    3. D. Ostler,
    4. S. Zelenay,
    5. M. Haury,
    6. J. Demengeot
    . 2003. Regulatory T cells selectively express toll-like receptors and are activated by lipopolysaccharide. J. Exp. Med. 197: 403–411.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Veldhoen M.,
    2. R. J. Hocking,
    3. C. J. Atkins,
    4. R. M. Locksley,
    5. B. Stockinger
    . 2006. TGFbeta in the context of an inflammatory cytokine milieu supports de novo differentiation of IL-17-producing T cells. Immunity 24: 179–189.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Trinchieri G.
    1994. Interleukin-12: a cytokine produced by antigen-presenting cells with immunoregulatory functions in the generation of T-helper cells type 1 and cytotoxic lymphocytes. Blood 84: 4008–4027.
    OpenUrlFREE Full Text
  23. ↵
    1. Longhi M. P.,
    2. C. Trumpfheller,
    3. J. Idoyaga,
    4. M. Caskey,
    5. I. Matos,
    6. C. Kluger,
    7. A. M. Salazar,
    8. M. Colonna,
    9. R. M. Steinman
    . 2009. Dendritic cells require a systemic type I interferon response to mature and induce CD4+ Th1 immunity with poly IC as adjuvant. J. Exp. Med. 206: 1589–1602.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Pulendran B.
    2004. Modulating vaccine responses with dendritic cells and Toll-like receptors. Immunol. Rev. 199: 227–250.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Yang R.,
    2. F. M. Murillo,
    3. H. Cui,
    4. R. Blosser,
    5. S. Uematsu,
    6. K. Takeda,
    7. S. Akira,
    8. R. P. Viscidi,
    9. R. B. Roden
    . 2004. Papillomavirus-like particles stimulate murine bone marrow-derived dendritic cells to produce alpha interferon and Th1 immune responses via MyD88. J. Virol. 78: 11152–11160.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Barton G. M.,
    2. R. Medzhitov
    . 2002. Control of adaptive immune responses by Toll-like receptors. Curr. Opin. Immunol. 14: 380–383.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Shi G.,
    2. C. A. Cox,
    3. B. P. Vistica,
    4. C. Tan,
    5. E. F. Wawrousek,
    6. I. Gery
    . 2008. Phenotype switching by inflammation-inducing polarized Th17 cells, but not by Th1 cells. J. Immunol. 181: 7205–7213.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Shi G.,
    2. M. Ramaswamy,
    3. B. P. Vistica,
    4. C. A. Cox,
    5. C. Tan,
    6. E. F. Wawrousek,
    7. R. M. Siegel,
    8. I. Gery
    . 2009. Unlike Th1, Th17 cells mediate sustained autoimmune inflammation and are highly resistant to restimulation-induced cell death. J. Immunol. 183: 7547–7556.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Chen J.,
    2. B. P. Vistica,
    3. H. Takase,
    4. D. I. Ham,
    5. R. N. Fariss,
    6. E. F. Wawrousek,
    7. C. C. Chan,
    8. J. A. DeMartino,
    9. J. M. Farber,
    10. I. Gery
    . 2004. A unique pattern of up- and down-regulation of chemokine receptor CXCR3 on inflammation-inducing Th1 cells. Eur. J. Immunol. 34: 2885–2894.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Bettelli E.,
    2. T. Korn,
    3. M. Oukka,
    4. V. K. Kuchroo
    . 2008. Induction and effector functions of T(H)17 cells. Nature 453: 1051–1057.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Reiner S. L.
    2007. Development in motion: helper T cells at work. Cell 129: 33–36.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Hori S.,
    2. T. Nomura,
    3. S. Sakaguchi
    . 2003. Control of regulatory T cell development by the transcription factor Foxp3. Science 299: 1057–1061.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Jordan M. S.,
    2. A. Boesteanu,
    3. A. J. Reed,
    4. A. L. Petrone,
    5. A. E. Holenbeck,
    6. M. A. Lerman,
    7. A. Naji,
    8. A. J. Caton
    . 2001. Thymic selection of CD4+CD25+ regulatory T cells induced by an agonist self-peptide. Nat. Immunol. 2: 301–306.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Lambe T.,
    2. J. C. Leung,
    3. H. Ferry,
    4. T. Bouriez-Jones,
    5. K. Makinen,
    6. T. L. Crockford,
    7. H. R. Jiang,
    8. J. M. Nickerson,
    9. L. Peltonen,
    10. J. V. Forrester,
    11. R. J. Cornall
    . 2007. Limited peripheral T cell anergy predisposes to retinal autoimmunity. J. Immunol. 178: 4276–4283.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Panopoulos A. D.,
    2. L. Zhang,
    3. J. W. Snow,
    4. D. M. Jones,
    5. A. M. Smith,
    6. K. C. El Kasmi,
    7. F. Liu,
    8. M. A. Goldsmith,
    9. D. C. Link,
    10. P. J. Murray,
    11. S. S. Watowich
    . 2006. STAT3 governs distinct pathways in emergency granulopoiesis and mature neutrophils. Blood 108: 3682–3690.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Zhou L.,
    2. I. I. Ivanov,
    3. R. Spolski,
    4. R. Min,
    5. K. Shenderov,
    6. T. Egawa,
    7. D. E. Levy,
    8. W. J. Leonard,
    9. D. R. Littman
    . 2007. IL-6 programs T(H)-17 cell differentiation by promoting sequential engagement of the IL-21 and IL-23 pathways. Nat. Immunol. 8: 967–974.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Barr T. A.,
    2. P. Shen,
    3. S. Brown,
    4. V. Lampropoulou,
    5. T. Roch,
    6. S. Lawrie,
    7. B. Fan,
    8. R. A. O’Connor,
    9. S. M. Anderton,
    10. A. Bar-Or,
    11. et al
    . 2012. B cell depletion therapy ameliorates autoimmune disease through ablation of IL-6-producing B cells. J. Exp. Med. 209: 1001–1010.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Summers S. A.,
    2. O. M. Steinmetz,
    3. P. Y. Gan,
    4. J. D. Ooi,
    5. D. Odobasic,
    6. A. R. Kitching,
    7. S. R. Holdsworth
    . 2011. Toll-like receptor 2 induces Th17 myeloperoxidase autoimmunity while Toll-like receptor 9 drives Th1 autoimmunity in murine vasculitis. Arthritis Rheum. 63: 1124–1135.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Yang Q.,
    2. S. R. Goding,
    3. M. E. Hokland,
    4. P. H. Basse
    . 2006. Antitumor activity of NK cells. Immunol. Res. 36: 13–25.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Beauchamp N. M.,
    2. R. D. Yammani,
    3. M. A. Alexander-Miller
    . 2012. CD8 marks a subpopulation of lung-derived dendritic cells with differential responsiveness to viral infection and toll-like receptor stimulation. J. Virol. 86: 10640–10650.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Cottalorda A.,
    2. C. Verschelde,
    3. A. Marçais,
    4. M. Tomkowiak,
    5. P. Musette,
    6. S. Uematsu,
    7. S. Akira,
    8. J. Marvel,
    9. N. Bonnefoy-Berard
    . 2006. TLR2 engagement on CD8 T cells lowers the threshold for optimal antigen-induced T cell activation. Eur. J. Immunol. 36: 1684–1693.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Zehntner S. P.,
    2. C. Brickman,
    3. L. Bourbonnière,
    4. L. Remington,
    5. M. Caruso,
    6. T. Owens
    . 2005. Neutrophils that infiltrate the central nervous system regulate T cell responses. J. Immunol. 174: 5124–5131.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section