Immunomodulating effects of the anti-viral agent Silibinin in liver transplant patients with HCV recurrence

Recently, several experimental and clinical studies have described the anti-oxidative, anti-fibrotic and anti-viral properties of Silibinin, a flavonolignan representing the main component (60 %) of silymarin extracted from the milk thistle (Silybum marianum gaertneri). In vivo and in vitro studies have shown that the anti-HCV effect of Silibinin involves stimulation of toll-like receptor (TLR) 7, interferon regulatory factor 3, and p38 mitogen-activated protein kinase pathways [1]. The safety and efficacy of Silibinin as anti-viral therapy has been demonstrated in chronic HCV-infected patients and in liver transplant patients with HCV recurrence unresponsive to standard combined therapy with interferon/ribavirin [2].

There is evidence that Silibinin has also immune-modulating properties and polarizes Th1/Th2 immune responses through functional modifications of dendritic cells (DCs) [3]. DCs are innate immune cells that are also important in the induction and regulation of adaptive immunity. Several subsets of DC have been identified, principally conventional myeloid (m) DC and also type-1 interferon (IFN)-producing plasmacytoid (p) DC, that may be specialized for modifying the quality, strength, and duration of immune responses [4, 5].

Silibinin has been shown to suppress the expression of MHC class I and II and costimulatory molecules (CD80, CD86) by murine bone marrow-derived DC, and this phenomenon has been associated with impairment of LPS-induced IL-12 secretion by these cells [3]. Moreover, Silibinin-treated DCs are highly proficient at antigen capture mediated via mannose receptor-mediated endocytosis and fail to induce Th1 responses and a normal cell-mediated immunity [3].

The outcome of HCV-related hepatitis depends on the balance between immune reactivity and immune suppression/tolerance. After viral infection, activated DCs induce the differentiation of naïve T cells into virus-specific CD4⁺ (IFN-γ; anti-viral Ab) and CD8⁺ T cells for adaptive immunity. It appears that via TLR7 plasmacytoid dendritic cell (pDC) sense HCV-RNA that is transfered inside the pDC after their contact with infected cells, thus inducing type-I IFN secretion. Despite these events, with potential to control and eliminate viral infection, the immune response fails to eliminate the virus in most HCV infections. During the past few years, several authors have demonstrated functional impairment of DC in HCV-infected patients. Thus, Nattermann et al. [6] have shown decreased numbers of BDCA1⁺ (myeloid dendritic cells (mDC)) and BDCA2⁺ DC (pDC) in peripheral blood of HCV-infected patients, together with a concomitant increase in their numbers in liver tissue. The authors suggested that the interaction of the HCV E2 Ag with CD81, a member of the tetraspanin family (considered the cellular receptor for entry of HCV expressed on DC), could result in impaired migration of DC towards the chemokine CCL21 that modulates DC trafficking from sites of infection to lymphoid tissue, with consequent impairment of T cell priming. In addition, decreased CD86 expression and IL-12 secretion by DC from chronically HCV-infected patients has been detected, along with impaired allogeneic T cell stimulatory activity of mDC and reduced IFN-γ and increased IL-10 secretion by T cells. It has been reported that the functional impairment of DC in HCV-infected patients may be related to the presence of HCV-RNA in both mDC and pDC [7]. By contrast, several observations show a normal ability of DC from chronically HCV-infected patients to prime T cells suggesting the need for improved understanding of the role of DC subsets in HCV pathogenesis [8‐10].

Recently, inhibition of effector T cells by regulatory T cells (Treg) and the PD-1/PD-L1 pathway has been described as a mechanism responsible for regulating adaptive T cell responses to HCV [11]. Treg are defined as a subset of CD4⁺T lymphocytes that highly express CD25 (IL-2 receptor α-chain) on their surface and the transcription factor Foxp3 intracellularly and constitute approximately 5 % of peripheral CD4⁺ T lymphocytes. Treg critically suppress HCV-specific lymphocyte proliferation, differentiation and cytokine production [11] and promote alloimmune tolerance [12‐15].

Regulation of T cell responses, including the induction of Treg, has been shown to be under the control of immune regulatory molecules expressed by DC subsets, including ICOSL (inducible costimulator ligand), PD-L1 (B7 homologue-1 = programmed death ligand-1; CD274), CD39 (ectonucleoside triphosphate diphosphohydrolase-1), and the non-classical HLA class I molecule HLA-G and its receptor, immunoglobulin-like transcript 4 (ILT4) [12]. However, enumeration of DC subsets and Treg and the expression of these immune regulatory molecules on DC subsets and PD-1 on Treg have not been examined in relation to HCV recurrence and Silibinin treatment after liver transplantation.

In this study, we examined peripheral blood DC subsets, Treg, and the expression by these cells of immune regulatory molecules in stable liver transplant patients with HCV recurrence treated for 14 consecutive days with iv infusion of Silibinin (iv-SIL). Our findings reveal changes in circulating pDC and mDC that have previously been associated with tolerogenic conditions (elevated pDC/mDC ratio, increased expression of ILT4, CD39, and HLA-G on pDC and increased expression by mDC of ICOSL) suggesting how iv-SIL might regulate anti-viral and alloimmunity. We have also observed multiple clinical correlations that could improve the clinical management of liver transplant patients and that deserve further analysis.

Both DC and Treg are believed to play important roles in the regulation of alloimune responses, host responses to HCV infection [21], and transplant outcome [13, 22]. In this study, we have monitored key immunological parameters previously recognized to be involved in the regulation of anti-viral immunity and tolerance/rejection for the first time in liver transplant patients with HCV recurrence treated for 14 consecutive days with iv-SIL. Recently, in a single-center, prospective, randomized, parallel-group, double-blind, placebo-controlled, phase 2 trial, we have shown that Silibinin monotherapy has significant anti-viral activity in patients with established HCV recurrence in the graft and who are unresponsive to standard therapy. We have also confirmed the safety and tolerability of iv-SIL without interaction with immunosuppressive drugs [2]. In the present study, we show that, associated with a significant decrease in HCV viral load, iv-SIL induces changes in circulating DC that have previously been associated with tolerogenic conditions [12, 23], providing new insight into how iv-SIL may regulate alloimmunity and anti-HCV immunity in liver transplant patients. After iv-SIL treatment, we observed a higher pDC/mDC ratio and pDC exhibited reduced HLA-DR and elevated ILT4, HLA-G, and CD39 expression compared with baseline values. In addition, after iv-SIL administration, blood mDC showed higher ICOSL expression. By contrast, no significant changes were detected in Treg frequency or PD-1 expression. Our study also reveals significant correlations between clinical parameters and DC and Treg markers which could prove helpful, especially if confirmed in larger, prospective studies, in the clinical management of liver transplant patients.

In humans, conventional mDC and “non-conventional” pDC exhibit inherent tolerogenic properties in the healthy steady-state, and each plays an important role in regulating innate and adaptative immunity. There is also evidence that pDC, more than mDC, might be predisposed to induce tolerance [24‐27]. We have shown that tolerant and prospective weaning liver transplant patients exhibit a significantly higher incidence of circulating pDC relative to mDC compared with those patients requiring maintenance immunosuppression [17]. Moreover, modulation of the pDC/mDC ratio has been proven to have a biological influence not only in the transplant setting but also in other diseases or conditions, such as chronic obstructive pulmonary disease and pregnancy [28, 29]. The findings of the current study now show that after 14 consecutive days of iv-SIL, the pDC/mDC ratio in peripheral blood is higher than at baseline and it does not correlate with any of clinical parameters evaluated, including the viral load. These data therefore suggest that iv-SIL in liver transplant patients may promote redistribution of peripheral blood DC subsets in favor of pDC. Whether or not this finding might be correlated with a regulatory/tolerogenic effect of iv-SIL will be addressed in further studies. Moreover, blood mDC after iv-SIL were modestly but not significantly reduced; their frequency was inversely correlated with serum AST level at the end of the treatment.

Since DC function and the outcome of DC-T cell interactions may depend on the net costimulatory and coregulatory signals delivered by the DC, we tested the expression of several costimulatory and coregulatory molecules on blood DC subsets. Notably, we found that the mean fluorescence intensity of HLA-II DR on pDC was slightly but significantly downregulated after iv-SIL, suggesting impaired ability of pDC to stimulate T cells and immune reactivity to the same extent as baseline pDC. Moreover, CD86 expression (a maturation marker which was not modified significantly by iv-SIL) on circulating pDC was positively correlated with total bilirubin level at baseline.

HLA-G and its receptor ILT4 are believed to play an important role in regulating both the maturation and function of DC. Thus, interaction between HLA-G and its receptors inhibits DC maturation [30]. Moreover, soluble HLA-G inhibits human DC-triggered allogeneic T cell proliferation [31] and HLA-G-expressing antigen-presenting cells induce CD4⁺ T cell Treg [32]. Although not directly demonstrated in the present study, these findings suggest that elevated HLA-G and ILT4 on peripheral blood pDC may be consistent with the ability of iv-SIL to promote a tolerogenic state. Moreover, expression of HLA-G on pDC after iv-SIL positively correlated with total bilirubin level.

We also observed for the first time that after iv-SIL, pDC, but not conventional mDC, exhibited significantly higher levels of CD39 (% and MFI) than at baseline, which was inversely correlated with viral load. It has been shown that maturation of DC, an important control point in the development of immune responses, can also be modulated by extracellular adenosine triphosphate (ATP) released by DC and also by damaged/necrotic cells. CD39 is an ectoenzyme that degrades ATP to AMP and can therefore modulate DC activation and inhibit their capacity to initiate Th1 responses [33]. In the liver transplant setting, it may temper alloimmunity and reduce the risk of graft injury and rejection [34].

The ICOS-ICOSL signaling pathway is involved in Th2 immune response enhancement and in germinal center formation, somatic hypermutation, and class switch recombination [35‐37]. Moreover, Treg are thought to develop as a consequence of signaling through costimulatory molecules [38], and some subsets of Treg appear to depend on ICOS-ICOSL signaling pathway for in vivo and in vitro suppression through IL-10 production [39, 40]. Enhanced ICOSL expression by DC has also been shown to be involved in the pathogenesis of infectious disease [41]. The present study demonstrates upregulation of ICOSL on peripheral blood mDC after iv-SIL which was inversely correlated with viral load. Several markers (HLA-DR, PD-L1, and PD-L1/CD86 ratio) on mDC that were not modified by iv-SIL treatment were positively correlated with transaminase levels.

There is evidence that circulating levels of Treg are elevated in tolerant liver transplant patients compared with non-tolerant patients, or healthy individuals [12, 16, 42, 43]. Moreover, the expression of PD-1 on total Treg as well as Treg subsets (naїve/central memory/effector memory/effector) is significantly higher in HCV-infected patients than in healthy controls. In addition, PD-1/PD-L1 signaling negatively regulates Treg by limiting STAT-5 phosphorylation in patients with chronic HCV infection, and PD-1 overexpression at the site of inflammation is involved in the establishment of a long-lasting inflammatory disease. The present study does not show any significant change in the incidence of peripheral blood Treg or Foxp3 expression in liver transplant patients after iv-SIL. Moreover, no change was also found in PD-1 expression by Treg. These findings suggest that despite a significant decrease in viral load by the end of treatment [2], iv-SIL does not affect a possible mechanism (PD-1/PD-L1 signaling on Treg cells) involved in the persistence of HCV infection. This is consistent with clinical findings showing that 16 days after the end of the treatment, the infection persisted with viral load mean values similar to baseline [2]. Interestingly, although not modified by iv-SIL, the expression of PD-L1 on Treg after treatment correlated positively with HCV viral load and serum AST.

Correlations observed in the present study between DC and Treg markers and liver functional tests, including HCV viral load, have not been described previously, and their biological/clinical significance is unclear. Thus, larger prospective studies are needed to validate their clinical significance. In order to discriminate the specific roles of iv-SIL, HCV viral load and patient immune suppression in affecting circulating DC and Treg, functional and mechanistic in vitro studies need to be undertaken.

In conclusion, we believe that this is the first study in liver transplant patients with HCV recurrence showing the impact of iv-SIL on circulating DC and Treg. The findings described reveal changes in circulating DC subsets that have been associated previously with regulatory/tolerogenic conditions and suggest how Silibinin might regulate alloimmunity and the anti-viral immune response. Moreover, we have observed multiple clinical correlations that could eventually improve the clinical management of liver transplant patients and that deserve further analysis.