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


     
 

The Journal of Immunology, 2008, 181, 2220 -2226
Copyright © 2008 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Savage, N. D. L.
Right arrow Articles by Ottenhoff, T. H. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Savage, N. D. L.
Right arrow Articles by Ottenhoff, T. H. M.
Right arrowPubmed/NCBI databases
*Substance via MeSH

Human Anti-Inflammatory Macrophages Induce Foxp3+GITR+CD25+ Regulatory T Cells, Which Suppress via Membrane-Bound TGFβ-11

Nigel D. L. Savage2,*,{dagger}, Tjitske de Boer*,{dagger}, Kimberley V. Walburg{dagger}, Simone A. Joosten*,{dagger}, Krista van Meijgaarden*,{dagger}, Annemiek Geluk*,{dagger} and Tom H. M. Ottenhoff2,*,{dagger}

* Department of Immunohematology and Blood Transfusion and {dagger} Department of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
CD4+ T cell differentiation and function are critically dependent on the type of APC and the microenvironment in which Ag presentation occurs. Most studies have documented the effect of dendritic cells on effector and regulatory T cell differentiation; however, macrophages are the most abundant APCs in the periphery and can be found in virtually all organs and tissues. The effect of macrophages, and in particular their subsets, on T cell function has received little attention. Previously, we described distinct subsets of human macrophages (pro- and anti-inflammatory, m{varphi}1 and m{varphi}2, respectively) with highly divergent cell surface Ag expression and cytokine/chemokine production. We reported that human m{varphi}1 promote, whereas m{varphi}2 decrease, Th1 activation. Here, we demonstrate that m{varphi}2, but not m{varphi}1, induce regulatory T cells with a strong suppressive phenotype (Tm{varphi}2). Their mechanism of suppression is cell-cell contact dependent, mediated by membrane-bound TGFβ-1 expressed on the regulatory T cell (Treg) population since inhibition of TGFβ-1 signaling in target cells blocks the regulatory phenotype. Tm{varphi}2, in addition to mediating cell-cell contact-dependent suppression, express typical Treg markers such as CD25, glucocorticoid-induced TNF receptor (GITR), and Foxp3 and are actively induced by m{varphi}2 from CD25-depleted cells. These data identify m{varphi}2 cells as a novel APC subset capable of inducing Tregs. The ability of anti-inflammatory macrophages to induce Tregs in the periphery has important implications for understanding Treg dynamics in pathological conditions where macrophages play a key role in inflammatory disease control and exacerbation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Studies over the past 10 years have attempted to determine mechanisms of induction as well as modes of suppression by regulatory CD4+ T cell subsets. To date there are four widely accepted CD4+ T regulatory cell (Treg)3 types: 1) naturally occurring (nTreg), 2) inducible (Tr1), 3) anergic T cells, and 4) Th3 Tregs.

nTregs are thymic-derived CD4+CD25highFoxp3+ cells which require cell-cell contact for suppression. Although not yet fully elucidated, mouse models have shown that suppression may be mediated by either membrane-bound TGFβ-1 (1, 2), lymphocyte activation gene-3/CD223 (3), or glucocorticoid-induced TNF receptor (GITR) (4). A recent report showed that human CD4+CD25+Foxp3+ Tregs are not solely derived from the thymus, but rather from the CD4+ memory pool, and that they have limited proliferative ability, eliminating the possibility of self-renewal of the Treg population. These observations suggest a model whereby the stimulation of memory CD4+ cells in the periphery leads to the continuous generation and renewal of Foxp3+ Tregs (5).

Inducible (Tr1) Tregs are typically induced in vitro by the presence of IL-10 (6), and their suppressive ability is mediated by IL-10 secretion. Anergic T cells have been shown to have a suppressive phenotype that is cell-cell contact dependent and not IL-10 or TGFβ-1 mediated (7, 8). Typically, T cell anergy is induced by Ag stimulation of the TCR/CD3 complex in the absence of costimulatory signals. Interestingly, anergic T cells are also capable of rendering APC tolerogenic (9), which may be one mechanism of sustaining suppression.

The induction of Th3 cells is tightly linked with the induction of mucosal tolerance: mucosal administration of relevant Ag can prevent the onset of autoimmune diseases such as experimental allergic encephalitis, diabetes, and allergy in mice. The regulatory property of Th3 cells is thought to be mediated by soluble, secreted TGFβ-1, as the suppression can generally be reversed by administration of TGFβ-1-blocking Abs (reviewed in Ref. 10). Interestingly, the role that TGFβ-1 plays in both nTreg and Th3-mediated suppression is drawing the two cell types closer together than previously thought (11).

We and others have previously described diametrically opposed macrophage subsets, derived from the same CD14+ monocytic precursor population, which we designated m{varphi}1 and m{varphi}2 (12, 13). m{varphi}1 secrete high levels of IL-23 (IL-12p40/IL-23p19) upon stimulation, and in the presence of IFN-{gamma} initiate transcription of IL-12p35, leading to the secretion of IL12p70 as well. m{varphi}2 cells, on the other hand, do not secrete IL-12p70 or IL-23, but rather secrete IL-10 in response to LPS. The expression of HLA-DR at the cell surface of m{varphi}1 and m{varphi}2 cells is low compared with mature dendritic cells, and in the case of m{varphi}2 is further reduced upon TLR stimulation (12). m{varphi}2 cells express the scavenger receptor CD163 and are highly phagocytic, a trait ascribed to macrophages resident at mucosal surfaces (14) and to peritoneal macrophages (15). Microarray analysis of m{varphi}1 and m{varphi}2 cells revealed exclusive expression of another scavenger receptor, stabilin 1 (STAB1) in m{varphi}2 cells, and also found to be expressed in alternatively activated macrophages (Ref. 16 and our unpublished observations). Upon activation, m{varphi}2 secrete an abundance of chemokines but very few proinflammatory cytokines (17). The TLR-induced down-regulation of T cell stimulatory molecules such as HLA-DR, CD80, and CD86, the absence of proinflammatory cytokines (IL-23, IL-12), and the high levels of IL-10 secreted by m{varphi}2 cells prompted us to investigate whether m{varphi}2 cells, despite the absence of TLR stimulus, could induce the differentiation of Tregs and to discern which subset of Tregs these cells may belong to.

The data presented herein describe the induction of T cells into a strong Treg subset only after being stimulated once by m{varphi}2, but not by m{varphi}1, cells. The mechanism of suppression of the m{varphi}2-activated T cells (Tm{varphi}2) was found to require cell-cell contact and was mediated by membrane-bound TGFβ-1. In addition to surface-expressed TGFβ-1, these CD4+ Tregs were CD25+, GITR+, and Foxp3+ and were induced from CD25– cells. Given the abundance of macrophages in the peripheral immune system and in virtually all tissues (resident macrophages), it is exceedingly important to understand how macrophage subsets influence and shape the T cell repertoire in homeostatic and inflammatory situations.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Reagents

Abs for FACS analysis were purchased from BD Pharmingen (CD1a, CD4, CD14, CD25, CD80, CD86, CD163, CTLA-4, HLA-DR), R&D Systems (GITR, latency-associated peptide (LAP)), IQ Products (TB21-TGFβ-1), and from eBioscience (FoxP3). Anti-CD3 (OKT-3) and anti-CD28 (B-T3) were obtained from Leiden University Medical Center (LUMC) Pharmacy and Sanquin, respectively. Insect cell-derived, 97% pure, carrier-free recombinant LAP was purchased from R&D Systems. Goat polyclonal blocking Abs to TGFβRII and isotype control goat IgG were obtained from Abcam and used at 20 µg/ml. ALK inhibitor (TGFβRI kinase inhibitor SB431542) was purchased from Sigma-Aldrich and used at 0.1 µM. GITR agonist Ab (clone 110416) and isotype control were obtained from R&D Systems and used at 10 µg/ml. ELISA Ab kits were obtained from Invitrogen-BioSource (IL-12p40 and IL-10) and R&D Systems (MIP1-β) and the human 17-plex assay was performed using Bio-Rad Multiplex kits and equipment. CFSE and LPS were all obtained from Sigma-Aldrich and used according to the manufacturer’s recommendations.

Cell culture

All macrophages were generated as previously described (12). Briefly, monocytes from anonymous healthy blood donors’ buffy coats were enriched for CD14 expression with MACS microbeads (Miltenyi Biotec) and cultured in RPMI 1640 medium, L-glutamine, and 10% FCS in the presence of 5 ng/ml GM-CSF (Invitrogen-BioSource) or 50 ng/ml M-CSF (R&D Systems) for 6 days to generate m{varphi}1 and m{varphi}2, respectively. Hallmark cytokine secretion assays were performed by activating the macrophages with 100 ng/ml LPS for 16 h and performing ELISAs for IL-12p40 and IL-10 on supernatants. FACS analysis for CD1a, CD14, CD163, CD80, CD86, and HLA-DR was performed to validate the purity of the macrophage types. T cell lines were generated by adding 106 PBMC to 2 x 105 APC (either m{varphi}1 or m{varphi}2) in 24-well plates and adding exogenous IL-2 (25 U/ml) and IL-15 (10 ng/ml) to all cultures on day 3 and splitting as necessary until day 13 in Iscove’s medium supplemented with L-glutamine, penicillin/streptomycin, and 5% pooled human serum. Tm{varphi}1/2 cell lines were rested in IL-2-free medium for 48 h before FACS and 30 h before coculture assays. Tm{varphi}1/2 cultures were subjected to consecutive adhesion steps (24 h and 2 h) to remove any macrophages from the culture before the coculture assays. Coculture assays were performed by adding 5 x 103 or 104 rested Tm{varphi}1/2 live cells (donor A anti-donor B) to an "indicator" proliferation assay (104 T cells (donor A anti-mature dendritic cell donor B) + 5 x 104 irradiated PBMC donor B). Proliferation was determined by CFSE dilution. Briefly, cells (indicator responder T cells (TInd) or Tm{varphi}1/2) were labeled with 1 µM CFSE according to Sigma-Aldrich instructions on day 0, cocultured for 4 days as described above, stained for CD4, and analyzed on a FACSCalibur (BD Biosciences). Generation of supernatants for the supernatant transfer experiments was done by incubating Tm{varphi}1/2 cell lines with plate-bound anti-CD3 and anti-CD28 (1 µg/ml each) for 24 h. Cell-free supernatants were subsequently added to the indicator described above.

Live cell sorting was performed at the dedicated LUMC flow cytometry facility on a FACSAria (BD Biosciences) cell sorter.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
m{varphi}1 and m{varphi}2 induce proliferative allogeneic responses

To first determine whether m{varphi}1 and m{varphi}2 cells differed quantitatively in their capacity to present Ag, we tested and compared their ability to induce allogeneic T cell responses. Based on the assumption that allogeneic Ag density on m{varphi}1 and m{varphi}2 would be comparable and homogeneously expressed on all cells in each population tested in the absence of TLR stimulus as described previously (12), PBMC were cultured in the presence of HLA-mismatched m{varphi}1 or m{varphi}2 over a period of 7 days. Similar patterns of proliferation, both in intensity and kinetics, were induced by both subtypes of macrophages, which peaked at day 5 (Fig. 1).


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 1. m{varphi}1 and m{varphi}2 induce similar allogeneic T cell proliferation. PBMC (1 x 105) were incubated with 2 x 104 non-activated allogeneic m{varphi}1 or m{varphi}2 for a period of 7 days. Proliferation was assessed by tritiated thymidine incorporation of triplicate data points. There are no significant differences between proliferation of T cells responding to allogeneic m{varphi}1 and m{varphi}2 cells (two-way ANOVA). Results are representative of three individual experiments.

 
m{varphi}2 induce differentiation of allogeneic T cells with regulatory properties

Due to the phenotype of m{varphi}2 (TLR-mediated secretion of IL-10, low expression of HLA-DR, CD80, and CD86, poor Ag presentation capacity in Th1 recall responses) we investigated whether T cells (donor A) having encountered m{varphi}2 (donor B) could regulate the response of autologous TInd cells (donor A anti-donor B) to irradiated PBMCs (donor B). TInd cells were generated by incubating PBMC (donor A) to LPS-matured dendritic cells (donor B) for a period of 13 days. Fig. 2, B and C, shows that titration of T cells that have encountered Tm{varphi}2 into the coculture assay (containing TInd cells + irradiated PBMC (donor B)) are capable of reducing TInd cell proliferation to less than half maximal value as determined by CFSE dilution, even at the low ratio of 1:1. In contrast, the addition of similar numbers of T cells that had encountered allogeneic Tm{varphi}1 was incapable of reducing TInd cell proliferation (Fig. 2, A and C), showing that m{varphi}2, but not m{varphi}1, are capable of inducing T cells with a regulatory function. To determine the proliferative status of the Tm{varphi}1/2 cells in the coculture assays, bulk Tm{varphi}1/2 cultures were labeled with CFSE before addition to the coculture and assayed for dilution on day 4. Fig. 2D shows that in the absence of allogeneic PBMC (donor B) the Tm{varphi}1/2 cells do not proliferate; however, in the presence of Ag, Tm{varphi}1 proliferate more extensively than Tm{varphi}2.


Figure 2
View larger version (23K):
[in this window]
[in a new window]

  		FIGURE 2. m{varphi}2 induce differentiation of T cells (Tm{varphi}2) with regulatory properties. CFSE dilution coculture assay showing proliferation of 104 TInd cells in the presence of increasing numbers (5 x 103 and 5 x 104; 1:2 and 1:1 ratios, respectively) of Tm{varphi}1 (A) or Tm{varphi}2 (B) cells per well (multiple wells pooled for FACS analysis). Addition of Tm{varphi}1 cells did not reduce the proliferation of TInd cells, whereas addition of Tm{varphi}2 did. Numbers above histograms indicate percentage of CD4+ TInd cell population that divided. Histograms shown are representative of duplicate experiments. C, Graphical representation of data obtained in A and B, with TInd cell proliferation set to 100%. D, Histograms showing CFSE dilution of Tm{varphi}1 and Tm{varphi}2 cells in the coculture assay in the presence (+) or absence (–) of allogeneic PBMC (donor B). Numbers above histograms indicate percentage of CD4+ Tm{varphi}1/2 population that divided. All histograms were gated on CD4+ live cells.

 
Tm{varphi}2 regulate by cell-cell contact

To discern the mechanism of inhibition, we stimulated Tm{varphi}1 and Tm{varphi}2 cell lines with plate-bound anti-CD3 and anti-CD28 for 24 h. Additionally, we purified Tm{varphi}2 cells into CD25+ and CD25– populations (as in Fig. 6A) and stimulated them with immobilized Abs. The supernatants of these cultures were analyzed for various cytokines by Multiplex and ELISA. Interestingly, in bulk cultures, all cytokines, with the exception of TNF-{alpha}, were secreted at high levels by Tm{varphi}2 cells, including the immunosuppressive cytokines IL-10 and CCL4 (18) (Table I). Furthermore, in purified populations, we observed elevated levels of IL-6, IL-8, IL-13, and CCL4 in CD25+ Tm{varphi}2 cells (Table I). ELISAs for soluble TGFβ-1 on unmodified and acidified supernatants (acidification is necessary for TGFβ-1 detection) revealed no differences between the Tm{varphi}1 and Tm{varphi}2 cell subsets (data not shown). The bulk supernatants were transferred to CFSE-labeled TInd cells proliferating in response to exposure to irradiated stimulatory PBMC. Unexpectedly, elevated concentrations of cytokines including IL-10 or CCL4 were not inhibitory, while the cells producing them maintained a regulatory phenotype when given physical access to the proliferating T cells (Fig. 3). Furthermore, addition of neutralizing anti-IL-10 and/or anti-CCL4 and isotype Abs to the cell-cell contact experiments did not reverse the suppressive phenotype of the Tm{varphi}2 cells (data not shown). The lack of suppression from Tm{varphi}2 cell supernatants was observed in all experiments performed regardless of the TInd cell source. Testing several T cell lines and clones minimized the chances of encountering artifacts due to selective receptor expression by specific TInd cell lines.


View this table:
[in this window]
[in a new window]

  		Table I. Cytokine concentrations (pg/ml) secreted by Tm{varphi}1 and Tm{varphi}2a

 

Figure 3
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 3. Tm{varphi}2 cells regulate by cell-cell contact. A, CFSE dilution coculture assay showing proliferation of 104 TInd cells in the absence or presence of 104 Tm{varphi}1 or Tm{varphi}2 (1:1 ratio) per well (multiple wells pooled for FACS analysis). Addition of supernatants from the same Tm{varphi}1 or Tm{varphi}2 cells activated with anti-CD3/CD28 for 24 h shows no suppression of TInd cells. Numbers above histograms indicate percentage of CD4+ TInd population that divided. All histograms were gated on CD4+ live cells. B, Representation of data obtained in A, with TInd cell proliferation set to 100%.

 
Tm{varphi}2 cells express membrane bound TGFβ-1, which mediates their suppressive ability

Since the regulatory attribute of the Tm{varphi}2 cells was cell-cell contact dependent, we tested the expression of active TGFβ-1 (membrane TGFβ-1 (mTGFβ-1)) at the cell surface by flow cytometry. By using the TB21 Ab, we determined that a population of CD4+ Tm{varphi}2 cells expressed high levels of CD25 as well as active mTGFβ-1 long after initial stimulation (day 13). By contrast, Tm{varphi}1 cells showed few cells expressing CD25 and little mTGFβ-1 (Fig. 4C and see Fig. 6A). To determine whether TGFβ-1 was directly involved in Tm{varphi}2-mediated regulation, we titrated TGFβ-1 LAP into the cocultures. TGFβ-1 is typically expressed in a latent form due to its non-covalent association with LAP. Removal of LAP leads to presentation of bioactive TGFβ-1 (reviewed in Ref. 19). Titration of commercially available exogenous LAP dose dependently reversed the Tm{varphi}2 cells’ ability to suppress (Fig. 4, A and B). Staining Tm{varphi}1 and Tm{varphi}2 cells with anti-LAP Ab revealed similar high levels of inactive TGFβ-1 on both populations (data not shown). This observation, taken together with the lack of soluble factors accountable for the suppression (Fig. 3), supports the notion that active but not latent mTGFβ-1 is the mechanism of regulation used by Tm{varphi}2 cells.


Figure 4
View larger version (47K):
[in this window]
[in a new window]

  		FIGURE 4. CD4+ Tm{varphi}2 cells express CD25 and membrane-bound TGFβ-1, the inhibition of which reverses their suppressive ability. A, CFSE dilution coculture assay showing proliferation of CD4+ 104 TInd cells in the absence or presence of 104 Tm{varphi}2 (1:1 ratio) per well (multiple wells pooled for FACS analysis). LAP peptide titration from 5 to 250 ng/ml to the coculture reverses Tm{varphi}2 induced suppression, while addition of 250 ng/ml to TInd increased proliferation by only 14%. B, Representation of data obtained in A, with TInd cell proliferation set to 100%. C, FACS analysis of CD25 and mTGFβ-1 expression on live, CD4+, Tm{varphi}1, and Tm{varphi}2 cells 13 days after encountering m{varphi}1 or m{varphi}2 cells, respectively, and a further 48 h with IL-2 and IL-15 retraction to establish a resting population.

 
Next, we sorted the Tm{varphi}2 cells into CD25low and CD25high populations by flow cytometry (as shown in Fig. 6A). Subsequently, we determined the suppressive ability of these cells. Only the Tm{varphi}2 CD25+ cells were the population capable of inhibiting the proliferation of TInd cells at a ratio of 1:1 (Fig. 5). To further test the involvement of TGFβ-1 in suppression, we opted to target the TGFβ-1 signaling pathway at various levels. Typically, the TGFβ-1 signaling cascade is initiated by the binding of TGFβ-1 to TGFβ-receptor complex, which comprises two subunits, RI and RII. Ligand binding leads to heterodimerization and clustering of receptor subunits. In turn, this results in TGFβRI-mediated phosphorylation of secondary messengers called SMADs, namely SMAD2 and SMAD3. These then translocate to the nucleus to regulate transcription directly or interact with other transcription factor subunits. Addition of LAP, which inactivates TGFβ-1 as described before, reversed the inhibition induced by Tm{varphi}2 cells (Fig. 4, A and B). To further interfere with the ligand binding, we added neutralizing goat anti-human TGFβRII polyclonal Abs (ab10853) to our suppression assay. This inhibited the suppressive effects of CD25+ Tm{varphi}2 cells, while addition of isotype control polyclonal Abs had no effect. Anti-TGFβRII had no effect on proliferation of the controls where CD25– Tm{varphi}2 cells were added to the assay (Fig. 5A). GITR, a cell surface TNFR family member, is temporarily up-regulated on effector T cells upon activation, and triggering GITR signaling has been shown to abrogate suppressive effects mediated by TGFβ-1 (20). Indeed, addition of cross-linking GITR mAb (110416) to the same assays also abrogated the suppressive effect, whereas the isotype control and CD25– Tm{varphi}2 cells had no effect on TInd cell proliferation (Fig. 5B). Next, we targeted the TGFβ-1 signaling pathway farther downstream by interrupting the kinase activity of TGFβRI with SB431542-hydrate, known to inhibit phosphorylation of SMAD2 and SMAD3. Addition of SB431542-hydrate reversed the inhibition induced by CD25+ Tm{varphi}2 cells, while control proliferation remained unaffected (Fig. 5C). Taken together, these data show that m{varphi}2 cells induce a Tm{varphi}2 CD4+ population that expresses high levels of CD25 as well as mTGFβ-1, which is able to suppress in a TGFβ-1-dependent manner since blocking the TGFβ-1 signaling pathway reversed suppression.


Figure 5
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 5. Inhibition of TGFβ-1 signaling and GITR activation suppresses CD25+ Tm{varphi}2 cell suppression. Tm{varphi}2 cells were sorted based on CD4 and CD25 expression (as in Fig. 6) and added to a CFSE-labeled TInd cell coculture assay. Only CD4+CD25+ cells inhibit CD4+ TInd cell proliferation. A, Addition of anti-TGFβ-1RII Ab but not isotype controls inhibited suppression observed by CD4+CD25+ Tm{varphi}2 cells on CD4+ TInd cells. B, Addition of anti-GITR polyclonal Ab but not isotype control Ab reversed the observed CD4+CD25+ Tm{varphi}2-mediated suppression on CD4+ TInd cells. C, Addition of ALK5 inhibitor (SB431542), thus inhibiting TGFβ-1RI enzyme activity, reversed the suppression on CD4+ TInd cells induced by CD4+CD25+ Tm{varphi}2.

 
Tm{varphi}2 cells are CD25+, Foxp3+ and GITR+

More detailed FACS analysis on sorted populations (Fig. 6A) showed that Tm{varphi}2 CD4+CD25+ cells maintained very high levels of GITR at the cell surface long after stimulation as compared to the CD25– population (Fig. 6B). Expression of GITR has been associated with CD4+CD25+ Tregs (reviewed in Ref. 20). Tm{varphi}2 cells also expressed high levels of FoxP3 (Fig. 6C). FoxP3 has long been identified as a transcription factor associated with Tregs (21, 22). In general, the phenotypic analysis of the Tm{varphi}2 cells reveals expression of typical Treg markers that have previously been reported, with the exception of CTLA-4, which Tm{varphi}2 cells did not express at all (Fig. 6D).


Figure 6
View larger version (36K):
[in this window]
[in a new window]

  		FIGURE 6. Tm{varphi}2 cells are CD4+, CD25+, GITR+, FoxP3+, and CTLA-4–. A, Tm{varphi}2 cells were enriched to >95% by flow cytometry based on CD4 and CD25 expression for coculture assays (Fig. 5) and also subsequently stained with Treg-associated markers. Upper left panel, CD4 and CD25 expression on Tm{varphi}1 bulk cultures after 13 days protocol followed by a 48-h resting period in the absence of IL-2 and IL-15. Upper right panel, Same staining for cells exposed to m{varphi}2 cells following the same expansion protocol. B, GITR expression is maintained high on CD25+ Tm{varphi}2 cells. C, FoxP3 is expressed at higher levels in CD25+ Tm{varphi}2 cells. D, CTLA-4 was not detected in either CD25+ or CD25– Tm{varphi}2 cell populations by intracellular staining. Green histograms indicate CD4+CD25–; red histograms, CD4+CD25+; black histograms, isotype control. Plots are representative of three independent experiments.

 
m{varphi}2 cells induce Tm{varphi}2 cells from CD25– populations

To eliminate the possibility that m{varphi}2 cells were merely expanding pre-existing naturally occurring CD4+CD25+ Tregs present at the initiation of culture, we depleted the PBMC of CD25+ cells (donor A) before addition to macrophages (donor B) and expansion using the established protocol. Tm{varphi}2 cells established from a CD25– starter population retained the ability to suppress TInd cell proliferation whereas m{varphi}1 cells have no suppressive effect on TInd populations (Fig. 7). These data therefore strongly support the ability of m{varphi}2 cells to be able to induce Tregs from CD4+CD25– cells and may therefore have important implications in regulation of the immune system in the periphery.


Figure 7
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 7. m{varphi}2 cells induce Tm{varphi}2 suppressive cells from CD25-depleted PBMC. A, PBMCs from donor A were depleted of CD25+ cells by MACS-positive selection. Cells were subsequently put through the Tm{varphi}1/2 differentiation protocol and tested in a TInd cell CFSE coculture assay. Histograms show CFSE dilution coculture assay showing proliferation of CD4+ 104 TInd cells in the absence or presence of 104 Tm{varphi}1/2 (1:1 ratio) per well (multiple wells pooled for FACS analysis). B, Representation of data obtained in A, with TInd cell proliferation set to 100%.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Macrophages are the principal sentinel phagocytes resident in tissues and are distributed throughout the body. For years, laboratories have described the ability of macrophages to phagocytose microbial pathogens and present Ag to T cells, subsequently initiating a Th response. More recently, the discovery of various human macrophages subsets, namely m{varphi}1 and m{varphi}2 (pro- and anti-inflammatory, respectively), has prompted us to investigate the impact of these macrophage subsets on T cell activation. Herein we have studied the type of T cell responses these various subsets can elicit. We have previously reported that m{varphi}1 cells can support Th1 cell proliferation and function, whereas m{varphi}2 failed to do so (12). The characteristic features of TLR-activated m{varphi}2 (secretion of IL-10, down-regulation of HLA-DR and costimulatory molecules upon activation) prompted us to investigate whether m{varphi}2 cells could support Treg differentiation. Using an allogeneic model of T cell activation, we observed that non-TLR-stimulated m{varphi}1 and m{varphi}2 cells could initiate equally good allogeneic T cell responses. These results differ from those obtained by others, who reported absence of proliferation to allogeneic macrophages (23). The precise nature and phenotype of the macrophages used in that study, however, were not described, making it difficult to compare with results obtained in our laboratory. In any case, the Tm{varphi}1 and Tm{varphi}2 cells we generated did not display anergic phenotypes and were capable of proliferating to similar extents in subsequent stimulations (to anti-CD3 and anti-CD28) in the absence of exogenous IL-2 (data not shown). In the absence of TLR stimulus, HLA-DR and T cell costimulatory molecules such as CD80 and CD86 are comparable between m{varphi}1 and m{varphi}2 cells (12).

Addition of Tm{varphi}1/2 cells to autologous responder T cells showed that only Tm{varphi}2, but not Tm{varphi}1, cells have a strong suppressive functional phenotype. Analysis of cytokine production in the Tm{varphi} subsets indicated that despite anti-inflammatory cytokines being secreted by Tm{varphi}2 cells, the suppression of proliferation was most likely due to a cell-cell contact mechanism rather than to secreted cytokines. Although the presence of immunoregulatory cytokines such as CCL4 (18) and IL-10 did not account for the regulatory ability of T cells in our system, they were nonetheless a good correlate. Interestingly, Tm{varphi}2 cells also secreted elevated quantities of inflammatory cytokines IL-6, IL-8, IL-13, and IFN-{gamma}, indicating an ability to recruit additional inflammatory cells to the site of activation while maintaining an ability to suppress T cell responses via cell-cell contact. While IL-13 secretion has previously been ascribed proinflammatory properties in acute graft-vs-host disease (24), it has also recently been associated with anti-inflammatory processes (25). Significantly, IL-13 can increase TGFβ-1 secretion in vivo (26).

Our data show that Tm{varphi}2 cells express mTGFβ-1 at their cell surface and that inhibition of TGFβ-1 signaling could abrogate their suppressive abilities. The expression of mTGFβ-1 at the Tm{varphi}2 cell surface, however, appeared to be extremely labile since simple manipulations such as cell sorting (via CD25) led to a complete loss of distinguishable mTGFβ-1 at the end of the procedure (data not shown), whereas a distinct small population was observed before the sort (Fig. 4C). Typically, we detected only 3–16% of the Tm{varphi}2 CD4+CD25+ population to be mTGFβ-1+ by FACS, and yet strong suppression was always observed at 1:1 ratios, which could be reversed by targeting the TGFβ-1 pathway, and thus we suspect that our observations may be a gross underestimation of the real mTGFβ-1 expression due to the labile nature of the protein. Further experimentation, in particular bioassays, will need to be designed to determine the levels of mTGFβ-1 on the surface of resting Tm{varphi}2 cells with minimal interference from manipulations. The fickle properties of mTGFβ-1 detection therefore made it impossible for us to FACS sort sufficient cells based on active mTGFβ-1 expression. Staining Tm{varphi}1 and Tm{varphi}2 cells with anti-LAP (latent TGFβ-1) Ab revealed similar high levels of inactive TGFβ-1 on both populations (data not shown). We therefore conclude that the expression of active, but not latent, TGFβ-1 as depicted by TB21 staining (Fig. 4C) is the causative factor in suppressing bystander proliferation.

Interestingly, Oida et al. (33) reported that CD4+CD25+ cells are specifically capable of converting small amounts of latent TGFβ-1 into active form. Although expression of active mTGFβ-1 at the cell surface has been described before on T cells (1), there remain doubts as to the precise mechanism of signaling (27). In our case, however, we successfully reversed inhibition by three different TGFβ-1 targeting mechanisms, namely addition of exogenous LAP, neutralizing TGFβRII Ab, and inhibition of SMAD phosphorylation. Addition of anti-TGFβ-1 (TB21) Ab directly to suppression assays had little effect in our study, even at the high concentrations reported to be necessary for inhibition (50 µg/ml). Moreover, we constantly observed isotype effects at these Ab concentrations and therefore discounted these results. We suspect that the lack of TB21 blocking may be due to several factors, including the labile properties of surface expression of active TGFβ-1 in combination with requirement of tight apposition of membranes required for ligand/receptor signaling to occur, thus sterically inhibiting access of TB21 Ab to newly expressed TGFβ-1. TGFβRII expression may not fluctuate so readily and therefore may be more susceptible to Ab-mediated blocking. The possible mechanisms of reversal of inhibition by addition of GITR Ab remain unclear, as this could either directly play a role in TGFβ-1 signaling by inhibiting SMAD2/3 phosphorylation on T effector cells (reviewed in Ref. 28) or provide a secondary stimulus activating necessary MAPK in effector T cells necessary to overcome the suppressive effects of Tregs. Alternatively, GITR signaling could affect the Treg population directly as reported previously by Valzasina et al. (29 and reviewed in Ref. 20) since Tm{varphi}2 expressed elevated levels of GITR. Interestingly, interference of GITR signaling can abrogate suppression in naturally occurring Tregs (30). Collectively, these data clearly show that a subset of Tm{varphi}2 cells expresses mTGFβ-1; we unambiguously demonstrate that the mechanism of suppression to be TGFβ-1 dependent since interference of the TGFβ-1/TGFβR ligand interaction as well as inhibiting downstream phosphorylation events reversed the suppressive phenotype.

CD25, FoxP3, GITR, and CTLA-4 are all up-regulated on most T cells upon stimulation; therefore, our experiments we purposefully designed to assess expression of these markers long after stimulus (13 days) and after a 48-h retraction of exogenous IL-2/IL-15. This enabled us to distinguish activation-induced kinetic expression from sustained expression as reported on Tregs. Tm{varphi}2 cells expressed much higher levels of typical Treg markers than do Tm{varphi}1, including FoxP3 transcription factor, with the exception of CTLA-4. FoxP3 expression on CD4+CD25– appeared higher than expected, and this may be due to the presence of TGFβ-1 (originating from CD4+CD25+ Tm{varphi}2) cells, which has been shown to induce higher FoxP3 expression in CD4+CD25– cells (31).

Vukmanovic-Steijc et al. (5) recently described the rapid proliferative state of human CD4+CD45RO+Foxp3+CD25+ T cells in vivo, and they determined that a principal source of Tregs may be the CD4+ memory T cell pool. Stimulation of a memory CD4+ T cell population in peripheral tissues, where macrophages are the primary APC, leads to the generation of Tregs (5). Depletion of naturally occurring Tregs (CD4+CD25+ cells) from our starting culture material led to the expansion of functional Tm{varphi}2 cells capable of suppressing TInd cell proliferation, thus negating the possibility of expanding pre-existing Tregs. It would be interesting to establish whether m{varphi}2 play a role in the expansion of such cells in vivo. It is conceivable that given the large concentrations of M-CSF in human serum, the default differentiation path for circulating monocytes is to an anti-inflammatory m{varphi}2 phenotype. Only in the presence of additional stimuli will the default pathway then be skewed toward the differentiation of inflammatory cells. The presence of m{varphi}2 in humans has been described (14, 15), and Ags at mucosal surfaces are renowned for being capable of inducing a tolerogenic state despite constant antigenic challenge (32). Therefore, the presence of such macrophages may well be responsible for maintaining a Treg population in the periphery until an inflammatory stimulus occurs to shift the balance toward an inflammatory response.

In conclusion, we show that after encountering anti-inflammatory type 2 macrophages (m{varphi}2), CD4+ T cells adopt a CD25+FoxP3+mTGFβ-1+ functional suppressor phenotype. This, while widely accepted to be compatible with a thymic nTreg phenotype, has more recently been described as a population emerging from memory CD4+ T cells after encountering Ag in the periphery. Additional experiments are required to determine the mechanism of type 2 macrophage induction/maintenance of nTreg and Treg turnover in humans, including the induction mechanism as well as the effect on CD8+ T cells.

These studies shed new light on the complex mechanisms and function of Treg generation in humans, an intensively studied but yet unresolved area. These data have important implications in deciphering states of immunopathology and infection as well as tumor biology where the balance of proinflammatory and anti-inflammatory activities is thought to be critical.


   Acknowledgments
 
We thank Ferry Ossendorp and Frits Koning for critical review of the manuscript, and the Leiden University Medical Center flow cytometry unit for assistance with the cell sorting.


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


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

1 This work was supported by the Netherlands Organization for Scientific Research (NWO) and The Netherlands Leprosy Relief Foundation (NLR). Back

2 Address correspondence and reprint requests to Prof. dr T. H. M. Ottenhoff and Dr. N. D. L. Savage, Departments of Immunohematology and Blood Transfusion and Infectious Diseases, Leiden University Medical Center, E3-Q, 2 Albinusdreef, 2333ZA Leiden, The Netherlands. E-mail addresses: T.H.M.Ottenhoff{at}lumc.nl and N.D.L.Savage{at}lumc.nl Back

3 Abbreviations used in this paper: Treg, regulatory T cell; GITR, glucocorticoid-induced TNF receptor; LAP, latency-associated peptide; m{varphi}1/2, pro- and anti-inflammatory macrophages, respectively; mTGFβ-1, membrane TGFβ-1; nTreg, naturally occurring T regulatory cell; TInd, indicator responder T cell; Tm{varphi}1/2, allogeneic T cells raised against m{varphi}1/2 cells. Back

Received for publication August 9, 2007. Accepted for publication May 23, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Nakamura, K., A. Kitani, W. Strober. 2001. Cell contact-dependent immunosuppression by CD4+CD25+ regulatory T cells is mediated by cell surface-bound transforming growth factor β. J. Exp. Med. 194: 629-644. [Abstract/Free Full Text]
  2. Nakamura, K., A. Kitani, I. Fuss, A. Pedersen, N. Harada, H. Nawata, W. Strober. 2004. TGF-β1 plays an important role in the mechanism of CD4+CD25+ regulatory T cell activity in both humans and mice. J. Immunol. 172: 834-842. [Abstract/Free Full Text]
  3. Huang, C. T., C. J. Workman, D. Flies, X. Pan, A. L. Marson, G. Zhou, E. L. Hipkiss, S. Ravi, J. Kowalski, H. I. Levitsky, et al 2004. Role of LAG-3 in regulatory T cells. Immunity 21: 503-513. [Medline]
  4. Shimizu, J., S. Yamazaki, T. Takahashi, Y. Ishida, S. Sakaguchi. 2002. Stimulation of CD25+CD4+ regulatory T cells through GITR breaks immunological self-tolerance. Nat. Immunol. 3: 135-142. [Medline]
  5. Vukmanovic-Stejic, M., Y. Zhang, J. E. Cook, J. M. Fletcher, A. McQuaid, J. E. Masters, M. H. Rustin, L. S. Taams, P. C. Beverley, D. C. Macallan, A. N. Akbar. 2006. Human CD4+CD25hiFoxp3+ regulatory T cells are derived by rapid turnover of memory populations in vivo. J. Clin. Invest. 116: 2423-2433. [Medline]
  6. Levings, M. K., R. Sangregorio, F. Galbiati, S. Squadrone, M. R. de Waal, M. G. Roncarolo. 2001. IFN-{alpha} and IL-10 induce the differentiation of human type 1 T regulatory cells. J. Immunol. 166: 5530-5539. [Abstract/Free Full Text]
  7. Chai, J. G., I. Bartok, P. Chandler, S. Vendetti, A. Antoniou, J. Dyson, R. Lechler. 1999. Anergic T cells act as suppressor cells in vitro and in vivo. Eur. J. Immunol. 29: 686-692. [Medline]
  8. Lombardi, G., S. Sidhu, R. Batchelor, R. Lechler. 1994. Anergic T cells as suppressor cells in vitro. Science 264: 1587-1589. [Abstract/Free Full Text]
  9. Vendetti, S., J. G. Chai, J. Dyson, E. Simpson, G. Lombardi, R. Lechler. 2000. Anergic T cells inhibit the antigen-presenting function of dendritic cells. J. Immunol. 165: 1175-1181. [Abstract/Free Full Text]
  10. Faria, A. M., H. L. Weiner. 2006. Oral tolerance: therapeutic implications for autoimmune diseases. Clin. Dev. Immunol. 13: 143-157. [Medline]
  11. Jiang, S., R. I. Lechler, X. S. He, J. F. Huang. 2006. Regulatory T cells and transplantation tolerance. Hum. Immunol. 67: 765-776. [Medline]
  12. Verreck, F. A., T. de Boer, D. M. Langenberg, M. A. Hoeve, M. Kramer, E. Vaisberg, R. Kastelein, A. Kolk, R. Waal-Malefyt, T. H. Ottenhoff. 2004. Human IL-23-producing type 1 macrophages promote but IL-10-producing type 2 macrophages subvert immunity to (myco)bacteria. Proc. Natl. Acad. Sci. USA 101: 4560-4565. [Abstract/Free Full Text]
  13. Fleetwood, A. J., T. Lawrence, J. A. Hamilton, A. D. Cook. 2007. Granulocyte-macrophage colony-stimulating factor (CSF) and macrophage CSF-dependent macrophage phenotypes display differences in cytokine profiles and transcription factor activities: implications for CSF blockade in inflammation. J. Immunol. 178: 5245-5252. [Abstract/Free Full Text]
  14. Smythies, L. E., M. Sellers, R. H. Clements, M. Mosteller-Barnum, G. Meng, W. H. Benjamin, J. M. Orenstein, P. D. Smith. 2005. Human intestinal macrophages display profound inflammatory anergy despite avid phagocytic and bacteriocidal activity. J. Clin. Invest. 115: 66-75. [Medline]
  15. Xu, W., N. Schlagwein, A. Roos, T. K. van den Berg, M. R. Daha, C. van Kooten. 2007. Human peritoneal macrophages show functional characteristics of M-CSF-driven anti-inflammatory type 2 macrophages. Eur. J. Immunol. 37: 1594-1599. [Medline]
  16. Kzhyshkowska, J., G. Workman, M. Cardo-Vila, W. Arap, R. Pasqualini, A. Gratchev, L. Krusell, S. Goerdt, E. H. Sage. 2006. Novel function of alternatively activated macrophages: stabilin-1-mediated clearance of SPARC. J. Immunol. 176: 5825-5832. [Abstract/Free Full Text]
  17. Verreck, F. A., T. de Boer, D. M. Langenberg, L. van der Zanden, T. H. Ottenhoff. 2006. Phenotypic and functional profiling of human proinflammatory type-1 and anti-inflammatory type-2 macrophages in response to microbial antigens and IFN-{gamma}- and CD40L-mediated costimulation. J. Leukocyte Biol. 79: 285-293. [Abstract/Free Full Text]
  18. Joosten, S. A., K. E. van Meijgaarden, N. D. Savage, T. de Boer, F. Triebel, A. van der Wal, E. de Heer, M. R. Klein, A. Geluk, T. H. Ottenhoff. 2007. Identification of a human CD8+ regulatory T cell subset that mediates suppression through the chemokine CC chemokine ligand 4. Proc. Natl. Acad. Sci. USA 104: 8029-8034. [Abstract/Free Full Text]
  19. Saharinen, J., M. Hyytiainen, J. Taipale, J. Keski-Oja. 1999. Latent transforming growth factor-β binding proteins (LTBPs): structural extracellular matrix proteins for targeting TGF-β action. Cytokine Growth Factor Rev. 10: 99-117. [Medline]
  20. Esparza, E. M., R. H. Arch. 2006. Signaling triggered by glucocorticoid-induced tumor necrosis factor receptor family-related gene: regulation at the interface between regulatory T cells and immune effector cells. Front. Biosci. 11: 1448-1465. [Medline]
  21. Hori, S., T. Nomura, S. Sakaguchi. 2003. Control of regulatory T cell development by the transcription factor Foxp3. Science 299: 1057-1061. [Abstract/Free Full Text]
  22. Fontenot, J. D., M. A. Gavin, A. Y. Rudensky. 2003. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 4: 330-336. [Medline]
  23. Hoves, S., S. W. Krause, C. Schutz, D. Halbritter, J. Scholmerich, H. Herfarth, M. Fleck. 2006. Monocyte-derived human macrophages mediate anergy in allogeneic T cells and induce regulatory T cells. J. Immunol. 177: 2691-2698. [Abstract/Free Full Text]
  24. Jordan, W. J., P. A. Brookes, R. M. Szydlo, J. M. Goldman, R. I. Lechler, M. A. Ritter. 2004. IL-13 production by donor T cells is prognostic of acute graft-versus-host disease following unrelated donor stem cell transplantation. Blood 103: 717-724. [Abstract/Free Full Text]
  25. Hildebrandt, G. C., S. W. Choi, G. Mueller, K. M. Olkiewicz, B. B. Moore, K. R. Cooke. 2007. The absence of donor-derived IL-13 exacerbates the severity of acute graft-versus-host disease following allogeneic bone marrow transplantation. Pediatr. Blood Cancer 50: 911-914.
  26. Lee, C. G., R. J. Homer, Z. Zhu, S. Lanone, X. Wang, V. Koteliansky, J. M. Shipley, P. Gotwals, P. Noble, Q. Chen, et al 2001. Interleukin-13 induces tissue fibrosis by selectively stimulating and activating transforming growth factor β1. J. Exp. Med. 194: 809-821. [Abstract/Free Full Text]
  27. Ostroukhova, M., Z. Qi, T. B. Oriss, B. Dixon-McCarthy, P. Ray, A. Ray. 2006. Treg-mediated immunosuppression involves activation of the Notch-HES1 axis by membrane-bound TGF-β. J. Clin. Invest. 116: 996-1004. [Medline]
  28. Chen, W., S. M. Wahl. 2003. TGF-β: the missing link in CD4+CD25+ regulatory T cell-mediated immunosuppression. Cytokine Growth Factor Rev. 14: 85-89. [Medline]
  29. Valzasina, B., C. Guiducci, H. Dislich, N. Killeen, A. D. Weinberg, M. P. Colombo. 2005. Triggering of OX40 (CD134) on CD4+CD25+ T cells blocks their inhibitory activity: a novel regulatory role for OX40 and its comparison with GITR. Blood 105: 2845-2851.
  30. Shimizu, J., S. Yamazaki, T. Takahashi, Y. Ishida, S. Sakaguchi. 2002. Stimulation of CD25+CD4+ regulatory T cells through GITR breaks immunological self-tolerance. Nat. Immunol. 3: 135-142. [Medline]
  31. Tran, D. Q., H. Ramsey, E. M. Shevach. 2007. Induction of FOXP3 expression in naive human CD4+FOXP3 T cells by T-cell receptor stimulation is transforming growth factor-β dependent but does not confer a regulatory phenotype. Blood 110: 2983-2990. [Abstract/Free Full Text]
  32. Faria, A. M., H. L. Weiner. 2006. Oral tolerance and TGF-β-producing cells. Inflamm. Allergy Drug Targets 5: 179-190. [Medline]
  33. Oida, T., L. Xu, H. L. Weiner, A. Kitani, W. Strober. 2006. TGF-β-mediated suppression by CD4+CD25+ T cells is facilitated by CTLA-4 signaling. J. Immunol. 177: 2331-2339. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Respir. Crit. Care Med.Home page
R. E. O'Hehir, L. M. Gardner, M. P. de Leon, B. J. Hales, M. Biondo, J. A. Douglass, J. M. Rolland, and A. Sandrini
House Dust Mite Sublingual Immunotherapy: The Role for Transforming Growth Factor-{beta} and Functional Regulatory T Cells
Am. J. Respir. Crit. Care Med., November 15, 2009; 180(10): 936 - 947.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
A. Mantovani, C. Garlanda, and M. Locati
Macrophage Diversity and Polarization in Atherosclerosis: A Question of Balance
Arterioscler Thromb Vasc Biol, October 1, 2009; 29(10): 1419 - 1423.
[Abstract] [Full Text] [PDF]

Home page
 J Mol Cell BiolHome page
W. Chen and J. E. Konkel
TGF-{beta} and 'Adaptive' Foxp3+ Regulatory T cells
J Mol Cell Biol, July 31, 2009; (2009) mjp004v1.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
P. Verdijk, E. H.J.G. Aarntzen, W. J. Lesterhuis, A.C. I. Boullart, E. Kok, M. M. van Rossum, S. Strijk, F. Eijckeler, J. J. Bonenkamp, J. F.M. Jacobs, et al.
Limited Amounts of Dendritic Cells Migrate into the T-Cell Area of Lymph Nodes but Have High Immune Activating Potential in Melanoma Patients
Clin. Cancer Res., April 1, 2009; 15(7): 2531 - 2540.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Savage, N. D. L.
Right arrow Articles by Ottenhoff, T. H. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Savage, N. D. L.
Right arrow Articles by Ottenhoff, T. H. M.
Right arrowPubmed/NCBI databases
*Substance via MeSH

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS