Tolerogenic plasmacytoid DC

Abstract

Plasmacytoid (p)DC are type-I IFN-producing cells known for their capacity to promote anti-viral innate and adaptive immune responses. Despite their potent anti-viral function, when compared to conventional DC (cDC), pDC exhibit poor immunostimulatory ability and their interaction with T cells often favors the generation of Treg. pDC are activated primarily in response to single-stranded (ss) RNA and ss DNA through TLR7 and TLR9, respectively, but also through TLR-independent mechanisms. Non-lymphoid tissue pDC, such as those residing in the airways, gut, and liver, play a significant role in regulating mucosal immunity and are critical for the development of tolerance to inhaled or ingested Ags. Herein we discuss properties that define tolerogenic pDC and how their unique characteristics translate into an ability to regulate immunity and promote the development of tolerance. We cover the importance of pDC during intrathymic Treg development and the maintenance of peripheral tolerance, as well as their regulatory role in transplantation, autoimmunity, and cancer. We highlight recent findings regarding danger- and pathogen-associated molecular pattern (DAMP and PAMP, respectively) signaling in the regulation of pDC function, and how the ability of pDC to promote tolerance translates into the potential clinical applications of these cells as therapeutic targets to regulate immune reactivity.

1. Introduction

Plasmacytoid DC (pDC) are BM-derived APC that secrete type-I IFN in response to viruses and RNA/DNA or immune complexes. While present in tissues at low numbers in the healthy steady-state, pDC accumulate in lymphoid and non-lymphoid tissues under pathological conditions [1]. They are implicated both in protective immunity and in tolerance induction [2–6]. This review focuses on their tolerogenic properties and underlying mechanisms, including the induction of Treg and direct suppression of T cell responses, that appear to play important roles in pathological conditions, such as infectious disease and cancer, and in regulation of autoimmune disorders and allograft rejection. Better understanding of pDC tolerogenicity may lead to novel therapeutic strategies that target/utilize this important immune regulatory cell population.

2. pDC lineage commitment, development and characterization

The properties that define pDC may be partially explained by their unique developmental programming compared to other DC subsets. The majority of pDC arise from the same common DC progenitors as conventional (c)DC [7, 8], yet they express a genetic profile that more closely resembles lymphoid (T and B) cell development [9]. pDC lineage commitment is controlled by expression of the transcription factor E2-2, a member of the E protein family that plays a crucial role in lymphoid cell development [10]. E2-2 regulates expression of several critical proteins expressed at high levels by pDC relative to cDC, including IFN-regulatory factor (IRF) 7 and 8 [10, 11], and the transcription factor Spi-B [12], and may be the most important genetic determinant of pDC development.

Phenotypic characterization of pDC is not as clearly defined as that of other DC subsets. pDC have long been identified as a population of B220⁺CD11c^lowCD11b⁻Gr-1⁺ cells [13]. In humans, they express blood DC Ag (BDCA)-2 [14] and the IL-3 receptor (CD123) [15] and Ig-like transcript (ILT7) [16]. In mice, pDC express the murine pDC Ag (mPDCA)-1/BM stromal cell Ag (BST)-2/CD317 [17]. mPDCA-1 however, is upregulated on a variety of cells during inflammatory conditions, and therefore should be regarded as specific for pDC only in the steady state. Other murine pDC markers that have emerged include Siglec-H, Ly49Q, and CCR9 [18–20]. Additionally, mouse pDC express lymphocyte activation gene (LAG)-3 [21], a CD4-like molecule that is known to negatively regulate expansion of activated T cells [22].

3. Tolerogenic function of intrathymic pDC

Two recent reports suggest that, like conventional thymic myeloid (m)DC, human thymic pDC can drive natural (n) Treg development [23, 24]. Martin-Gayo et al [23] have shown that, following activation with CD40L + IL-3, mature thymic pDC efficiently promote the generation of CD4⁺CD25⁺Foxp3⁺ nTreg from autologous, positively-selected CD4⁺CD8⁺ thymocytes. The nTreg induced by pDC are more efficient IL-10 producers and inferior TGFβ producers compared with those induced by thymic mDC, reflecting the ability of the two major DC subsets to induce distinct Treg repertoires, based on their distinct Ag-presenting capacities [25]. Studies by Hanabuchi et al [24] have further shown that activated human thymic medullary pDC express the thymic stromal lymphopoietin (TSLP) receptor and become responsive to TSLP, inducing the generation of Foxp3⁺ Treg from CD4⁺CD8⁻CD25⁻ thymocytes. As in the former study, these Treg expressed more IL-10 but less TGFβ than those induced by TSLP-stimulated mDC. Taken together, these novel data suggest that thymic pDC play a critical role in the selection of Treg that can preferentially secrete IL-10 in response to self Ag in the periphery.

4. Regulation of pDC function by PAMPs and DAMPs

pDC in the periphery express multiple PRRs that play important roles in the detection, limitation and repair of tissue injury caused by bacteria and viruses through their release of PAMPs, such as LPS, single-stranded (ss)RNA, ssDNA, bacterial flagellin and peptidoglycan. The most studied PRRs are the TLRs and the nucleotide-oligomerization domain (NOD) proteins. pDC express high levels of TLR7 and TLR9, that are associated with their ability to produce large amounts of type-I IFNs [26].

In the mouse, spleen and liver pDC also express similar, or low-to-moderate levels of TLR4, as compared to cDC, although pDC behave differently when stimulated by LPS. Liver pDC secrete lower levels of IL-12, but higher levels of IL-10 and they skew allogeneic T cells towards Th2 differentiation, with an associated low Delta 4/Jagged 1 ratio [27]. Recently, we have reported that liver and spleen pDC express higher levels of NOD2 message than conventional mDC and that pDC are able to detect muramyl dipeptide (MDP), a breakdown product of bacterial peptidoglycans [28]. When compared to control liver pDC, freshly-isolated liver pDC from MDP-treated mice show weaker ability to induce allogeneic T cell proliferation, with associated reduction in IFNγ production, increased programmed death ligand-1 (PD-L1; =B7 homologue 1 [B7-H1]) expression, elevated cell surface PD-L1/CD86 ratio, and increased expression of the NF-κB signaling inhibitory molecules IRF4 and IkBα (Fig. 1). IRF4 has been shown to compete with IRF5 for binding to MyD88 and TRAF6 interfering with the ability of these factors to activate the NF-κB pathway [29]. Additionally, MDP stimulation in vivo significantly interferes with subsequent responses of liver pDC to TLR4 or TLR9 ligation. NOD2 ligation reduces IL-12, IL-6 and TNF-α production by pDC, in the presence or absence of either LPS or CpG stimulation. Furthermore, CpG-stimulated liver pDC from MDP-treated mice secrete lower levels of IFNα than their splenic counterparts. These findings demonstrate the importance of NOD2 signaling in liver pDC in controlling the responsiveness of pDC to gut-derived bacterial products.

Figure 1. Impact of TLR and NOD2 ligation on pDC function.

pDC function is markedly affected by recognition of PAMPs. Unlike conventional mDC, pDC produce large amounts of type-I IFN through a IRF-7-dependent mechanism [6]. TLR9 and TLR7 signaling has been shown to reduce the IL-3/IL-10-dependent release of Granzyme B by pDC, which under normal conditions, reduces T cell proliferation. pDC can sense, through TLR7, HCV-RNA physically transferred inside the pDC after cell-to-cell contact with infected cells, inducing type-I IFN secretion [78]. Mouse pDC also express TLR4, which senses LPS resulting in pro-inflammatory cytokine production (IL-6, IL-12, TNFα). In vivo exposure to the NOD2 ligand muramyl dipeptide (MDP) upregulates expression of IFN regulatory factor-4 (IRF-4), a negative regulator of TLR signaling, and impairs the ability of pDC to secrete type-I IFNs and pro-inflammatory cytokines in response to TLR ligands [28]. Moreover, NOD2 ligation upregulates PD-L1 on pDC and reduces their ability to induce allogeneic T cell proliferation.

The expression of different PRRs allows pDC to sense tissue injury that results in the release of damage-associated molecular patterns (DAMPs), such as self-DNA and high mobility group box 1 (HMGB1), which are recognized by TLR9 and TLR4, or by the receptor for advanced glycation end-products (RAGE) [30], respectively. In mice, HMGB1 suppresses pDC responses to TLR9 agonists, including viruses, and down-regulates costimulatory and adhesion molecules on pDC stimulated by CpG [31]. Moreover, pretreatment of pDC with HMGB1 inhibits their ability to promote Th1 responses [31]. The ability of HMGB1 to downregulate pDC responses suggests that this may be an intrinsic mechanism by which pDC contribute to the maintenance of peripheral tolerance to self Ags in the presence of necrotic cell death.

Thus, while ligation of PRRs on pDC allows them to initiate and promote innate and adaptive immune responses, this may also result in their ability to promote tolerance,- the outcome of PRR ligation on pDC appears to depend largely on the context in which pDC become activated. The distinct properties of liver pDC that are exposed continually to gut-derived PAMPs in the steady state appear to favor tolerance over immunity. This may contribute to the inherent tolerogenicity of the liver microenvironment [32, 33].

5. Molecular mechanisms utilized by tolerogenic pDC

The molecular mechanisms used by pDC to promote tolerance have been reviewed recently [2] and will not be discussed extensively herein. Our laboratory is currently investigating the role of several molecules expressed by pDC for their ability to regulate T cell-mediated immunity. Thus, splenic pDC from PD-L1-deficient (PD-L1^−/−) mice induce greater levels of CD4⁺ T cell proliferation (both syngeneic and allogeneic) in vitro compared to WT pDC [34]. Additionally, adoptive transfer of alloAg-pulsed PD-L1^−/− pDC into syngeneic hosts results in greater CD4⁺ T cell activation and reduced IL-10 production, but greater Foxp3 expression by T cells analyzed ex vivo, compared to mice given alloAg-pulsed WT pDC [34]. IL-27, an emerging IL-12 family member that induces IL-10 but not Foxp3 expression in CD4⁺ T cells, also exhibits immunoregulatory properties [35]. IL-27 and its receptor (WSX-1/TCCR) are elevated in murine liver pDC compared to splenic pDC [36]. Moreover, exposure to exogenous IL-27 upregulates PD-L1 on liver, but not spleen pDC. These findings suggest that PD-L1 function may underlie the ability of pDC to regulate immune responses, and that comparatively high levels of IL-27 produced by liver pDC may contribute to the inherent tolerogenic potential of the liver [32, 37] via upregulation of PD-L1.

Induction and support of Treg function by pDC has been reported in several models [38–40]. Recently, a unique property of pDC has been revealed, suggesting a previously unexplored mechanism through which they may promote tolerance via Treg. Young et al [25] demonstrated that, despite their activation, pDC fail to downregulate the MHC II ubiquitin E3 ligase membrane-associated RING-CH1 (MARCH I) and therefore, Ag:MHC II complexes are continuously ubiquitinated, internalized, and in a constant state of turnover. This contrasts with mDC that, through downregulation of MARCH I, form long-term, stable Ag:MHC II complexes on their surfaces following activation [41]. Based on studies showing that low levels of Ag can promote Treg development [42, 43], it is feasible that rapid Ag:MHC II turnover by pDC could translate into presentation of low levels of Ag (i.e. weaker TCR stimulation) and consequently their ability to promote Treg induction and function.

6. CCR9⁺ tolerogenic pDC regulate immune reactivity in vitro and in vivo

CCR9 has been identified as the receptor for thymus-expressed chemokine (TECK/CCL25) and directs T cell trafficking during thymic development and to the epithelium of the small intestine [44–46]. Recently, a subpopulation of CCR9⁺ DC in the small intestine has been shown to co-express classical pDC markers (B220, PDCA1, Ly6C), secrete IFN-α, migrate to CCL25, and home to the small intestine upon their adoptive transfer [47]. CCR9⁺ pDC express very low levels of Ag-presenting (MHC II) and co-stimulatory molecules,- a phenotype associated with DC immaturity or tolerogenicity [20]. Indeed, when LN B220⁺CD11c⁺ DC were sorted based on CCR9 expression, pulsed with OVA and cultured with CD4⁺ T cells, the CCR9⁺ pDC failed to induce CD4⁺ T cell proliferation in vitro and failed to prime T cells in vivo, suggesting that these OVA-loaded pDC could promote Ag-specific T cell tolerance [20]. The potent tolerogenic potential of CCR9⁺ pDC is evidenced by their capacity to suppress acute GVH disease when co-transferred into lethally-irradiated recipients with allogeneic CD4⁺CD25⁻ effector T cells [20]. One mechanism to account for these observations is an enhanced ability of CCR9⁺ pDC to induce highly suppressive Foxp3⁺ (CD4⁺CD25⁺) Treg and to inhibit the outgrowth of inflammatory IL-17-producing T cells [20]. Further investigation has revealed that B220⁺CCR9⁻ cells are precursors of conventional DC, suggesting that, in conjunction with other classical pDC markers, CCR9 may be a key marker for pDC and does not necessarily indicate a pDC subpopulation [48].

7. pDC and oral/mucosal tolerance: specific roles for liver and airway pDC

Several reports have implicated pDC in the induction and regulation of oral tolerance. The liver is a site of oral Ag presentation and liver pDC appear to rapidly induce anergy or deletion of Ag-specific T cells via a CD4⁺ T cell-independent mechanism. Goubier et al [33] have shown that pDC prevent oral Ag-induced T cell priming and are responsible for systemic tolerance to CD4⁺ and CD8⁺ T cell-mediated delayed-type hypersensitivity responses induced by Ag feeding. Also, systemic depletion of pDC prevents the induction of tolerance by Ag feeding. Moreover, transfer of oral Ag-loaded liver pDC to naive recipient mice induces Ag-specific suppression of CD4⁺ and CD8⁺ T cell responses. These data demonstrate that oral tolerance relies on Ag presentation by pDC to T cells, and suggest that pDC could represent a key therapeutic target for intestinal and systemic inflammatory diseases.

It has also been shown that systemic depletion of pDC during inhalation of inert Ag leads to classic features of asthma: IgE sensitization, airway eosinophilia, goblet cell hyperplasia, and Th2 cytokine production [49]. Adoptive transfer of pDC prior to sensitization abrogates this effect, likely by inhibiting the generation of effector T cells induced by Ag-presenting mDC. Together, the foregoing observations show that pDC can protect against inflammatory responses to harmless Ag and play a significant role in the development of mucosal tolerance.

8. pDC and hematopoietic stem cell and organ transplant tolerance

Although the concept that pDC have potential to promote transplant tolerance has emerged in recent years [39, 50–52], the role of pDC in experimental transplant tolerance remains poorly characterized. Reports of the influence of tolerogenic pDC on transplant outcome are summarized in Table 1. One of the first reports to highlight the importance of pDC in alloimmune tolerance identified these cells as a critical and necessary component of a population of facilitating cells (FC) in BM that could promote hematopoietic stem cell engraftment [53]. In experimental organ transplantation, pDC propagated from donor or recipient strain BM using fms-like tyrosine kinase 3 ligand (Flt3L) could prolong murine cardiac allograft survival [51]. This effect was not Ag-specific however, as third-party BM-pDC also significantly prolonged survival, suggesting a non-specific immunosuppressive effect. One potential explanation is the induction of IL-10-producing T cells by pDC via ICOS:ICOSL (B7RP-1) interaction [54, 55]. Flt3L-mobilized splenic pDC of donor origin also significantly prolong experimental cardiac allograft survival in the absence of immunosuppression, or in combination with anti-CD154 (CD40L) mAb therapy [56].

Table 1.

Tolerogenic pDC and Immune Regulation in Transplantation

Tolerogenic pDC	Model	Mechanism(s)	Reference
B220+CD11cloCD11b− (CD8+TCR−) Bone marrow or peripheral blood-derived, expanded using Flt3L, flow sorted	Hematopoietic Stem Cell Transplantation (Mouse)	Not reported; pDC may require interaction with effort from other cells within the CD8+TCR− population - pDC alone are not sufficient to promote engraftment	37
CCR9+B220+CD11cint (CD3−CD19−) Mobilized using Flt3L-secreting B16 melanoma, flow sorted from CD11c bead-purified cells from lymph nodes	GVH Disease (Mouse)	Expansion of Foxp3+ Treg; suppression of IL-17-producing Th17 pro-inflammatory cells; low T cell stimulatory capacity/T cell hyporesponsiveness	30
B220+CD11c+CD11b− Flow-sorted from Flt3L-expanded BM cells (Donor)	Cardiac Allograft (Mouse)	Induction of T cell hyporesponsiveness; regulation of T cell responses by expression of PD-L1	35
B220+CD11c+CD11b− Flow sorted from Flt3L-mobilized, splenic CD11c purified cells (Donor)	Cardiac Allograft (Mouse)	Not reported	40
PDCA-1+B220+CD11c+Gr-1+ (CD19−) Host pDC within the allograft, host lymph node and spleen	Cardiac Allograft (Mouse)	pDC acquire donor Ag from the graft and induce donor Ag-specific Foxp3+ Treg in the host LN	21
CD45R+/B220+CD4+ (TCR−CD45RA−) Flow sorted from gradient-enriched splenocytes	Cardiac Allograft (Rat)	Accumulation of pDC in allograft and spleen; suppression of CD4+ T cells through induction of CD8+ Treg and IDO	41

pDC promote tolerance in experimental transplant models. Due to low numbers of circulating and tissue-resident DC, Flt3L is utilized to mobilize and expand DC in vivo to numbers sufficient for experimentation. In experimental models, pDC are typically defined by co-expression of B220 and low levels of CD11c, and can also be classified as Gr-1⁺ or CD11b⁻. pDC also express pDC Ag (PDCA)-1 and a subset of pDC has been defined by expression of the intestinal homing chemokine receptor CCR9. A regulatory role for pDC in transplantation has been demonstrated in both hematopoietic stem cell and solid organ (cardiac allograft) transplantation, although the reported mechanisms for tolerance induction have varied. In both models, the induction of Foxp3⁺ Treg which have the ability to transfer tolerance has been identified as a key mechanism of the promotion of transplant tolerance by pDC. Other reported mechanisms include the ability to induce T cell hyporesponsiveness and suppression of pro-inflammatory Th17 responses.

It was shown recently that host-derived pDC are integral to the development of cardiac allograft tolerance induced by donor-specific transfusion and anti-CD154 mAb. Thus, PDCA-1⁺ pDC acquired donor MHC II-derived allopeptide Ag from the graft and migrated to LN (but not spleen), where they induced alloAg-specific CD4⁺Foxp3⁺ Treg necessary for tolerance induction [39]. In this model, depletion of pDC, or prevention of pDC LN homing, inhibited Treg development and tolerance induction. Interestingly, in a rat model of cardiac allograft tolerance induced by CD40Ig, pDC accumulated in the graft and the spleen, but not LN, and induced CD8⁺ Treg [57]. Regulation of alloreactive CD4⁺ T cells by pDC occurred either directly, through an IDO-dependent mechanism, or indirectly through CD8⁺ Treg, in a contact-dependent manner.

We have reported that analysis of peripheral blood DC, especially pDC, in liver transplant patients can be a helpful immune monitoring tool. Consistent with the more tolerogenic properties of pDC, higher incidences of circulating pDC relative to mDC are found in operationally tolerant pediatric liver allograft recipients, and in patients on low dose immunosuppressive therapy undergoing prospective drug weaning, compared with patients on maintenance immunosuppression [58]. In addition, high PD-L1/CD86 ratios on pDC correlate with elevated CD4⁺CD25^hiFoxp3⁺ Treg in patients who exhibit a state of immune tolerance to their transplant [59], which is consistent with evidence that the balance between expression of inhibitory PD-L1 and costimulatory B7-1 (CD80)/B7-2 (CD86) ligands regulates the outcome of their interaction with T cells [60]. Taken together, these findings suggest that pDC have potent regulatory capacity in transplantation, yet the mechanisms and site (graft vs. LN vs. spleen) of tolerance induction may depend upon the therapeutic regimen and model investigated.

9. pDC and regulation of autoimmunity

Type-I IFN production by pDC has been implicated in the pathogenesis of autoimmune diseases, such as psoriasis, systemic lupus erythematosus (SLE), arthritis and type-1 diabetes [2]. Under normal, healthy conditions, the subcellular localization of TLR9, high concentration of extracellular DNases and low levels of unmethylated CpG motifs within mammalian DNA, prevent pDC from sensing self DNA through TLR9 stimulation and the secretion of type-I IFN. However, several host factors can break down innate tolerance to self DNA, resulting in autoimmunity. For example, the anti-microbial peptide LL37 and HMGB1 released by damaged and/or dying cells binds self DNA fragments also released by dying cells to form aggregates that are protected from extracellular DNase degradation and delivered into early endosomes of pDC [61, 62]. Moreover, DNA-specific Abs bind self-DNA-LL37-HMGB1 complexes, increasing their translocation into TLR9-containing endosomes. The outcome of these responses is increased IFNα production, activation of mDC, and promotion of autoreactive T cell responses.

Although recent observations show that pDC are involved in the pathogenesis of EAE by promoting the induction of myelin oligodendrocyte glycoprotein (MOG)-specific Th17 cells [63], several reports suggest that tolerogenic pDC may regulate the severity of autoimmune disease (Table 2). Thus, mature pDC from rheumatoid arthritis patients with low disease activity express high levels of IDO and promote the differentiation of naïve CD4⁺CD25⁻ T cells into IL-10-secreting Treg or T regulatory type-1 (Tr1) cells [64]. Moreover, in vivo depletion of pDC exacerbates joint pathology and augments cell-mediated and humoral immunity to type II collagen in experimental arthritis [65]. In addition, in a murine model of lupus nephritis, low dose peptide therapy induces TGFβ production by pDC, expansion of autoAg-specific Treg and reduction of kidney-infiltrating inflammatory Th17 cells [42]. Moreover, a protective effect of pDC against TCRtg T cell-induced type-1 diabetes in NOD mice has been demonstrated [66]. Thus, pDC appear to play important roles in regulation of several experimental autoimmune disease states.

Table 2.

Tolerogenic pDC and the Regulation of Autoimmunity

Description of pDC	Disease/Model	Mechanism(s)	Reference
BDCA-4+ (CD304) Isolated from PBMC	Rheumatoid Arthritis (Human)	IDO-dependent induction of IL-10- producing Treg	48
120G8 mAb depletion of pDC	Experimental Arthritis (Mouse)	Systemic depletion of pDC enhanced cellular (T cell) and humoral (B cell) autoimmune responses;	49
Negatively selected mPDCA-1+Ly-6C+ cells	Experimental Lupus (Mouse)	Adoptive transfer of pDC reduced autoantibodies and T cell production of IFNγ and IL-17; T cells produced greater levels of TGFβ and pDC produced TGFβ but less IL-6; pDC enhanced function of CD4+CD25+ and CD8+ Treg	24
mPDCA-1 mAb depletion of pDC	NOD model of T1DM (Mouse)	Systemic depletion of pDC enhances disease progression and severity; IDO contributes to disease regulation	50

pDC can regulate autoimmune disease. pDC contribute to the regulation of autoimmune disease in both mice and humans and may serve as potential targets for therapeutic intervention. Reported mechanisms of autoimmune regulation include induction of IL-10-producing, CD4⁺CD25⁺ and CD8⁺ Treg, pDC production of TGFβ but reduced production of IL-6, promoting Treg induction and negative regulation of IL-17 production. The absence of pDC in experimental type I diabetes results in enhanced disease progression and severity, partially due to the loss of IDO production.

To reduce the secretion of IFNα during autoimmune disease, the use of statins, inhibitors of HMG-CoA reductase, has been suggested. Statins can profoundly reduce IFNα production by human pDC in response to TLR ligation, likely through the prevention of nuclear translocation of IRF7 [67]. A similar effect has been shown using the inhibitor of IκB kinase activity BAY 11-7082 [68]. Recently, it has been shown that IL-3/IL-10-stimulated human pDC can control the expansion of peripheral T cells through production of granzyme B (GrB) upon activation of the JAK/STAT pathway ([69] and Fig. 1). Thus, GrB⁺ pDC can suppress T-cell proliferation in a GrB-dependent, perforin-independent manner. Interestingly, this inhibitory mechanism can be broken by TLR7 and TLR9 signaling during infection, permitting maximal T cell responsiveness under these circumstances [70]. Thus despite their reported role in the pathogenesis of certain autoimmune diseases, pDC are also potential targets for restoration of tolerance to self Ags.

10. pDC and regulation of anti-tumor immunity

Due to their ability to secrete high levels of type-I IFN and TNF-α, pDC would appear to have potential to promote anti-tumor immunity through activation of NK cells and CD8⁺ T cells with potent cytolytic function. On the contrary however, pDC may regulate anti-tumor immunity and support immune evasion and tumor escape (Fig 2). pDC are recruited into the tumor microenvironment (Fig. 2A) via several receptor:ligand interactions, including CCL20:CCR6, stromal cell-derived factor (SDF)-1/CXCL12:CXCR4, and very late Ag (VLA)-5:vascular cell adhesion molecule (VCAM)-1 interactions [71, 72]. pDC exposed to the tumor microenvironment are protected from tumor macrophage-derived IL-10-induced apoptosis; they exhibit reduced IFN-α production upon TLR9 stimulation and can induce IL-10-producing CD4⁺ and CD8⁺ Treg [73, 74]. This suggests that anti-tumor immune responses can be regulated through both modulation of pDC function by the tumor, and by limiting anti-tumor cytolytic activity through induction of CD8⁺ Treg.

Figure 2. pDC regulate anti-tumor immunity and can promote tumor escape.

(A) pDC are recruited into the tumor microenvironment via multiple receptor-ligand interactions. Interaction of stromal cell-derived factor (SDF)-1/CCL12 with CXCR4 on pDC can upregulate very late Ag (VLA)-5 for transendothelial migration via VCAM-1 and can protect pDC from IL-10-induced apoptosis [71]. CCL20 binding to CCR6 on pDC can recruit pDC to the tumor site [72]. Factors produced within the tumor microenvironment can modulate IFNα production via TLR9 stimulation. pDC cultured with tumor cells can induce IL-10 production by CD4⁺ T cells and CD8⁺ Treg [73, 74]. (B) CD19⁺ pDC in tumor-draining LN (TDLN) produce high levels of IDO which can activate mature Treg via activation of the general control non-depressive kinase 2 (GCN2) pathway of protein synthesis inhibition [38]. Activated Treg can then activate other DC, leading to upregulation of PD-L1 which suppresses anti-tumor T cell responses. pDC-produced IDO and activated Treg can also convert naïve T cells into new Treg. IDO acts in an autocrine manner to suppress pDC production of IL-6, which prevents the conversion of Treg into IL-17-producing Th17 pro-inflammatory cells [76]. IDO also downregulates IFNα production by pDC [77].

In addition to intratumoral immune regulation, pDC also exhibit immunoregulatory properties in tumor-draining LN (TDLN) (Fig. 2B). A subpopulation of CD19⁺ pDC in TDLNs produces high levels of the tryptophan-catabolizing enzyme IDO [75]. Production of IDO by pDC has been linked directly to activation of naturally-occurring Foxp3⁺ Treg through modulation of the general control non-depressive kinase 2 (GCN2) pathway, which leads to inhibition of protein synthesis, and in Treg, results in their activation [38]. IDO-activated Treg can suppress effector T cell responses directly, or indirectly, through upregulation of PD-L1 on other DC that can then suppress effector T cells through the PD-L1:PD-1 pathway [38]. IDO plays a dual regulatory role by preventing conversion of these Treg into pro-inflammatory Th17 cells through autocrine inhibition of IL-6 production via upregulation of GCN2 in pDC [76]. IDO has also been reported to inhibit production of IFN-α, which may limit their ability to activate innate and adaptive anti-tumor immunity [77]. In relation to other tolerogenic pDC populations already described, it would be of interest to determine if the CD19⁺ IDO-producing subpopulation of pDC in TDLN corresponds to the tolerogenic CCR9⁺ pDC reported to induce Foxp3⁺ Treg, inhibit Th17 development and suppress GVH disease.

11. Prospects for targeting pDC to regulate immune reactivity

Collectively, these observations indicate that pDC exhibit functional dichotomy,- they can produce high levels of type-I IFNs and initiate the activation of adaptive immunity in infection and autoimmunity or they can play a pivotal role in immune regulation by inhibiting effector T cells and by inducing Treg. Therefore, manipulation of the immune system via targeting these two different aspects of pDC function offers considerable potential for therapeutic intervention. Attention should be afforded to targeting pDC for the induction of Treg, based on the multiple mediators expressed by pDC that induce and expand both CD4⁺ (Foxp3⁺ and IL-10-producing Foxp3⁻) and CD8⁺ Treg, and the reported role for this tolerogenic mechanism in transplantation, autoimmunity and cancer [38, 39, 42]. Therapeutic targeting should also consider regulation of IFNα and IL-6 production, combined with maintenance or augmentation of IDO and/or PD-L1 expression as key components of the tolerogenic capacity of pDC.

12. Concluding remarks

pDC are key regulators of immune responses against self and foreign Ags. They express a variety of cell surface and secrete cytokines with anti-inflammatory and immunoregulatory potential, many of which influence the potent ability of pDC to induce and/or expand both CD4⁺ and CD8⁺ Treg. Although there may not be a distinct marker to clearly identify pDC, both CD19 [38] and CCR9 [20] have been suggested as key markers of murine tolerogenic pDC populations that suppress anti-tumor immunity and prevent GVH disease, respectively, and that induce highly suppressive Foxp3⁺ Treg. Whether or not these markers correspond to the same pDC population and whether they promote tolerance through similar mechanisms remains unexplored. The tolerogenic potential of pDC is highlighted by loss of tolerance following their in vivo depletion and by the ability of pDC to suppress immune responses following adoptive transfer in DTH, organ transplantation, and GVH disease models. These findings support the view that pDC may play a significant role in the induction of tolerance and serve as potential targets for therapeutic intervention.

Abbreviations

Flt3L

fms-like tyrosine kinase 3 ligand

HMGB1

high mobility group box 1

pDC

plasmacytoid DC

PD-L1

programmed death ligand-1

TDLN

tumor-draining LN