Liver transplantation is the standard therapy for patients with end-stage liver disease or acute liver failure, who are then treated with immunosuppressive agents. Among these drugs, cyclosporin A (CsA) is widely used because of its ability to suppress cell-mediated immunity, interacting with T cell activation (1–3). CsA is notdevoid of several side effects, and can cause or worsen chronic renal insufficiency (4), neurological disorders (5), infectious diseases, cardiovascular diseases, and cancers (6).
To reduce the aforementioned side effects, several new drugs have been developed. Among these, Everolimus (Evr), a derivative of rapamycin belonging to the family of the mammalian target of rapamycin (mTOR) inhibitors, is currently used in different clinical settings. Evr acts at a later stage of the cell cycle, blocking cells in G1 stage through the inhibition of m-TOR (7), and also exerts several biological effects (8–12). Recently, different studies have focused their attention on the role of regulatory T cells (Tregs) in transplant tolerance. Acute rejection after liver transplantation was associated with reduced Treg number. Treg can inhibit recipient T lymphocytes recognizing the allogenic MHC of the transplanted organ (13). Therapy with the mTOR inhibitor rapamycin preferentially preserves Treg function (14–17) and, in the mouse model, natural Tregs are indeed expanded by this drug (18). Similarly, the same positive effect was detected in human Treg culture systems (19).
Nowadays, few data are available on the effects of Evr on main immune parameters, including the behavior of Treg, in liver-transplanted patients. Therefore, in a group of 29 individuals who received a liver transplant, we analyzed the effects of such drug and compared them with those induced by CsA. We focused our attention on the differentiation status of peripheral T cells, the amount and quality of Tregs, also evaluating those expressing CXCR3, a chemokine receptor responsible for the migration of T cells in inflamed tissue. Finally, we measured the polyfunctionality of peripheral CD4+ and CD8+ T cells, i.e., the simultaneous production of interferon (IFN)-γ and interleukin (IL)-2, along with the expression of the degranulation marker CD107a.
Materials and Methods
Patients and Study Design
We studied a total of 29 patients recruited by the Liver and Multivisceral Transplant Center of the University of Modena and Reggio Emilia in the period 2007–2010, compared with 20 age- and sex-matched donors. All subjects gave their informed consent, accordingly to the Italian laws.
Phenotypic analyses of T lymphocytes were performed at different times: the day before transplantation (T0), and 60 (T1), 90 (T2), 120 (T3) days after liver transplant. When possible, samples were also collected 130–220 days after transplant (T4).
Receivers were treated according to the “Kang” protocol [described in details by Masetti et al. (20)], that compares renal toxicity of two different regimens based upon Evr (Certican, Novartis Pharmaceutical, Basel, Switzerland) or CsA (Sandimmune, Novartis).
All patients were randomized on day 10 into one of the two groups on a 2:1 ratio for Evr group and CsA one, respectively.
Isolation and Freezing of Peripheral Blood Lymphocytes and Monocytes
Blood was collected through a venous drawing in ethylene diamine tetraacetic acid (EDTA), and peripheral blood mononuclear cells (PBMC) were isolated by Ficoll-Hypaque density gradient, according to standard procedures (21). Cells were then suspended in foetal bovine serum (FBS); the freezing solution [20% dimethyl sulfoxide (DMSO) in FBS], were cooled at 4°C for 30 min to reduce the DMSO toxicity (22). Then PBMC were transferred into the freezing solution and set for at least 24 h at −80°C inside a “thermal crib” (Nalgene, Rochester, NY) containing isopropanol, and subsequently transferred and stored in liquid nitrogen.
Staining of PBMC for Polychromatic Flow Cytometry
PBMC were thawed, washed twice with cold Hanks' balanced salt solution (Invitrogen, Carlsbad, CA, USA), resuspended in 300 μl phosphate-buffered saline (PBS), divided, and stained with one of the three different combinations of monoclonal antibodies (mAb) recognizing surface antigens for analyzing T cell differentiation, T cell activation, and Treg. The LIVE/DEAD Red Fixable Dead Cell Stain Kit (Molecular Probes, Eugene, OR), which is strongly recommended to exclude dead cells from the analysis, was added a few minutes before adding the antibodies (23,24).
The mAbs used for the differentiation panel included anti-CD3 conjugated with Pacific Blue, PB (clone UCHT1), anti-CD4 APC-H7 (clone RPA-T4), anti-CD8 AF488, anti-CD45RA PE-Cy7 (clone HI100), anti-CCR7 PE (clone 150503), and anti-CD127 AF647 (clone hIL-7R-M21). The activation panel contained anti-CD3 PB (clone UCHT1), anti-CD4 APC-H7 (clone RPA-T4), anti-CD8 AF647, anti-CD95 PE, anti-HLA-DR PE-Cy7 (clone G46-6), and anti-CXCR3 AF488 (clone 1C6/CXCR3); for the Treg panel, we used anti-CD3 PB (clone UCHT1), anti-CD4 APC-H7 (clone RPA-T4), anti-CD25 PE (clone 24212), anti-CD127 AF647 (clone hIL-7R-M21), anti-FoxP3 AF488 (clone 259D/C7), anti-CXCR3 AF488 (clone 1C6/CXCR3). Cells were stained with the different panels and treated as described (25). All the antibodies, except anti-CCR7 and anti-CD25 provided by R&D Systems (Minneapolis, MN), were from Becton Dickinson.
In Vitro PBMC Stimulation and Staining for Polychromatic Flow Cytometry
Before stimulation, thawed PBMC were rested at least 4 h at 37°C, in a 5% CO2 incubator, in a complete culture medium (RPMI 1640 supplemented with 10% FBS and 1% of each L-glutamine, sodium pyruvate, nonessential amino acids, and antibiotics; all from Invitrogen) containing 20 μg/ml DNAse (Sigma-Aldrich, St. Louis, MO, USA). At least 2 million PBMC were then washed and incubated overnight in the same medium, in the presence of 1 μg/ml staphylococcal enterotoxin B (SEB) (Sigma-Aldrich). All samples were incubated in the presence of monensin (2.5 μg/ml; Sigma-Aldrich) and brefeldin A (5 μg/ml; Sigma-Aldrich), of the costimulatory anti-CD28 mAb (1 μg/ml; R&D Systems) and of anti-CD107a mAb conjugated with PE-Cy5 (clone eBioH4A3, eBioscience, San Diego, CA) to evaluate degranulation in response to antigen stimulation (26). Directly conjugated antibodies obtained from eBioscience (anti-IL-2 PE clone MQ1-17H12, anti-IFN-γ PE-Cy7 clone 4S.B3, anti-CD4 APC-H7 clone RPA-T4), R&D Systems (anti-CD8 APC) and BD (anti-CD3 PB, clone UCHT1) were then used. Cells were stained in PBS with the LIVE/DEAD and with different antibodies for surface antigens (anti-CD3, anti-CD4, and anti-CD8), incubated for 20 min at room temperature and washed with PBS containing 5% FBS and 0.5 mM EDTA. Cells were fixed and permeabilized with the “Cytofix/Cytoperm buffer set” from Becton Dickinson prior to staining for intracellular antigens (anti-IL-2 and anti-IFN-γ) for 20 min at room temperature (27). Samples were finally fixed in PBS added with 1% paraformaldehyde, kept at 4°C and immediately analyzed.
Acquisition of Samples
Samples were analyzed using a CyFlow ML flow cytometer (Partec, Münster, Germany) equipped with a 488-nm blue solid-state laser (200 mW, kept at 50 mW for detection of FITC, PE, PE-Texas Red, PE-Cy5, and PE-Cy7), a 635-nm red diode laser (25 mW, for detection of APC, APC-Cy5.5, and APC-Cy7), a UV mercury lamp HBO (100 long life, 100 W), a 532-nm green solid state laser (100 mW, not used in this study), a 405-nm violet laser (50 mW), and a CCD camera. Data were acquired in list mode by using FloMax (Partec) software, and then analyzed by FlowJo 8.7 (Treestar Inc., Ashland, OR) under MacOS 10 (28,29). Samples were compensated by software, after acquisition.
Single staining and Fluorescence Minus One (FMO) controls were performed periodically for all antibody panels to set proper compensation and define positive signals. Simplified Presentation of Incredibly Complex Evaluation software (SPICE, version 5.1), was used to graphically analyze polychromatic flow cytometric data (30).
The trend analyses were performed by Skillings–Mack test and one-way Anova test were used to compare differences between the two groups, results are expressed as mean ± standard deviation; in the figures, significant differences are indicated by asterisks. The aforementioned analyses were performed using STATA 11, and plotted by GraphPad Prism 5 and SPICE 5.1 softwares. Pies were generated and compared by permutation test using SPICE 5.1. Differences were considered statistically significant when P < 0.05. All the parameters exposed in this report did not present significant differences between the two groups of treatment before liver transplantation.
Before Liver Transplantation, Patients Displayed Higher Levels of Activated CD4+ T Cells and Tregs Compared with Controls but Less CD4+ T Cells
The clinical characteristics of the enrolled patients are resumed in Table 1. The phenotypical analyses of T cells were performed before liver transplantation in a total of 29 patients (19 treated with Evr, 10 with CsA), and compared with those obtained from 20 donors.
Table 1. Clinical characteristics of the patients and causes of liver failure
Age (years, mean)
Sex (M; F)
Primary cause of end-stage liver disease (at T0):
Chronic active hepatitis/cirrhosis
Primary biliar cirrhosis
To identify CD4+ and CD8+ T lymphocytes, we first gated cells on the basis of their physical parameters (forward and side scatter, i.e., FSC and SSC, respectively), excluded dead cells by using the LIVE/DEAD staining (not shown) and then identified T cells on the basis of CD3 expression and SSC (Fig. 1A, left panel); a gate was then set on CD4+ or CD8+ T lymphocytes (Fig. 1A, right panel). As shown in Figure 1B (right panel), on the basis of the expression of two surface antigens, CCR7 and CD45RA, four different subpopulations of CD4+ or CD8+ T lymphocytes with a different maturation stage can be identified: naïve (N: CCR7+ and CD45RA+), central memory (CM: CCR7+ and CD45RA−), effector memory (EM: CCR7− and CD45RA−), and terminally differentiated cells expressing CD45RA (TEMRA: CCR7− and CD45RA+). In all these subpopulations the expression of the alpha-chain of IL-7 receptor (CD127) was also measured (Fig. 1B, right panel).
As shown in Figure 2A, at T0 the percentage of CD4+ T cells was higher in donors (66.7% ± 10.1%; mean ± standard deviation) compared with patients (57.3% ± 13.7%; P = 0.0108). As shown in Figure 2B, the same difference was found in naïve CD4+ T cells: donors had higher levels (46.6% ± 14.7%) than patients (24.8% ± 12.4%; P < 0.0001). We studied CD4+ and CD8+ T cell activation using CD95 and HLA-DR as biomarkers, along with the chemokine receptor CXCR3 (Fig. 1C). Patients showed a higher percentage of CD4+, CD95+, HLA-DR+ (7.6% ± 4.2%) and CD4+, CD95+, HLA-DR− (63.4% ± 12.8%) cells in comparison with donors (5.1% ± 2.9%; 44.1% ± 13.6%; P = 0.0113 and P < 0.0001, respectively). Also the frequency of CD95+, HLA-DR+, CXCR3+ (12.4 ± 8.9)% and CD4+, CD95+, HLA-DR−, CXCR3+ cells (10.4% ± 9.6%) was higher in patients undergoing liver transplantation when compared to donors (2.2% ± 1.6% or 1.6% ± 1.9%, respectively; P < 0.0001 in both cases) (Figs. 2C and 2D). No differences were found in the percentage of the same populations of activated CD8+ T cells (data not shown).
The frequency of Tregs, identified as CD3+, CD4+, CD25bright, FoxP3+, CD127− expressing or not CXCR3 (analyzed as described in Fig. 1D), was finally measured, and a higher percentage of these cells was observed in patients compared with donors (5.6% ± 2.7% vs. 2.9% ± 1.6%; P = 0.0004) (Fig. 2E). Moreover, also the level of Tregs expressing CXCR3+ was significantly higher in patients compared to donors (0.20% ± 0.20% vs. 0.01% ± 0.01%; P < 0.0001) (Fig. 2F).
After Liver Transplant, Patients Taking Evr Showed Higher Levels of Total CD4+ T Cells, Naïve CD4+, and Naïve CD8+ T Cells, but Lower Levels of CD8+ T Lymphocytes in Comparison with Those Assuming CsA
Phenotypical analyses of T cells performed 60 days after liver transplant (T1) in 19 patients treated with Evr and in six with CsA did not reveal any significant difference between the two groups; the same results were observed 90 days after transplant (T2) in 13 patients treated with Evr and in eight with CsA.
The evaluation of the trend of the cells after transplant within each group was performed keeping separate data obtained at T3 or at T4. However, in order to reach a number of patients sufficiently enough to compare the treatments with the two drugs, we analyzed together cells obtained from T3 and T4 (for Evr, n = 12 and, for CsA, n = 9).
As shown in Figure 3A, Evr group maintained stable levels of CD4+ T cells during the treatment; on the contrary, CsA group showed a decrease of these cells. Indeed, at T2 Evr group had a higher percentage of CD4+ T cells compared with CsA ones (52.1% ± 10.9% vs. 38.9% ± 10.8%; P = 0.0272). This trend was also maintained at T3 and T4. Also the frequency of naïve CD4+ T cells was higher in the Evr group: for example, at T2, the percentage of these cells was 24.3% ± 10.1% vs. 10.6% ± 9.3% (P = 0.0127).
Considering CD8+ T cells (Fig. 3B), in patients taking Evr we observed a significant increase from T0 to T2, whereas at T3 and T4 no differences were observed compared with T0; moreover, the P value for the trend was not significant. Also in patients taking CsA the percentage of CD8+ cells progressively and significantly increased (P for trend = 0.0021). Considering T3 and T4, i.e., >3 months after transplant, the percentage of CD8+ cells was lower in the Evr group compared to the CsA one (28.2% ± 9.7% vs. 42.4% ± 8.9%; P = 0.0127). On the contrary, the frequency of naïve CD8+ T cells was higher in Evr group (15.7% ± 14.8% vs. 3.1% ± 2.2%; P = 0.0027) (data not shown).
We measured the percentage of naïve CD8+ T lymphocytes expressing CD127, and found that in both groups a decrease of these cells was present at T1, followed by a tendency to the return to higher levels (Fig. 3C). Comparing the two groups, at T2 Evr group had a higher percentage of such cells (89.9% ± 11.3% vs. 80.8% ± 11.6%; P = 0.0266); the same difference was present in the following period, as Evr group displayed higher percentage of naïve CD8+, CD127+ T cells compared with pCsA (considering T3 and T4 together, 92.4% ± 6.8% vs. 86.9% ± 3.4%; P = 0.0346).
Patients Treated with Evr Showed a Higher Percentage of Regulatory T Cells Compared with Those Treated with CsA
The analysis of Tregs revealed the presence of an opposite trend between the two groups of patients, as those taking Evr showed a progressive increase of Tregs during the treatment, whereas those taking CsA a significant decrease (Fig. 4, upper panel; P for trend in the CsA group = 0.013, in the Evr group = 0.057). Starting from T3, patients in the Evr group showed a significantly higher percentage of Tregs compared with those assuming CsA (8.4% ± 2.9% vs. 4.9% ± 2.2%; P = 0.0066 at T3; 7.1% ± 2.5% vs. 3.2% ± 1.9%; P = 0.0032 at T4).
As far as the frequency of Tregs expressing CXCR3 was concerned, in the first period after transplant (i.e., T1 and T2) no significant differences were found between the groups. However, as shown in Figure 4 (lower panel), considering T3 and T4 the percentage of these cells was significantly higher in the Evr group (0.23% ± 0.19% vs. 0.07% ± 0.09%; P = 0.0276).
Functional Analysis Revealed a Higher Immunosuppressive Effect on CD8+ T Cells of Evr When Compared with CsA
To directly evaluate the immunosuppressive ability of the drugs by using a functional assay, we stimulated PBMC from patients assuming Evr or CsA with the superantigen SEB, which triggers T cells bearing different Vβ-T cell receptors. We measured the production of IL-2 and IFN-γ, along with the expression of CD107a. The simultaneous expression of the aforementioned functional markers was studied by polychromatic flow cytometry, as shown in Figure 1E.
As reported in Figure 5, Evr group showed a small but significant modification of the polyfunctionality among CD4+ cells (upper panels), as we observed a decrease in cells producing only IL-2 and an increase in cells producing only IFN-γ. No main changes were observed among CD8+ T cells (middle panels). However, as shown in the lower panel, the total response of CD8+ T cells (i.e., the sum of cells positive for any of the functional markers) was significantly higher in the CsA group (11.9% ± 9.1% vs. 4.5% ± 3.1% in the Evr group; P = 0.0312), as well as the total number of cells producing IFN-γ (9.1% ± 8.6% vs. 3.9% ± 3.2%; P = 0.011).
The main aims of the study were to analyze the immunosuppressive efficacy and the effects of Evr on different T cell subsets in patients who received a liver transplant, and to compare the efficacy of Evr with that of CsA. For this reason, we studied several phenotypic and functional parameters in PBMC from patients treated with different regimens.
First, it is to note that several differences between donors and patients were already present before liver transplantation. Regarding CD4+ T cells, patients displayed a lower percentage of both total CD4+ and naïve CD4+ T cells. Moreover, the percentage of activated T CD4+ cells, both CXCR3+ and CXCR3−, was higher if compared with healthy subjects. This could be due to the fact that most patients were suffering from a chronic viral infection, and thus their immune system was under a strong antigenic pressure. Furthermore, it is possible that their CD4+ T cells underwent a higher consumption, or that were selectively localized within the damaged organ. However, it remains to clarify why this phenomenon is observed only in CD4+ subpopulation, and why no main differences were found between patients and controls as far as CD8+ T cells were concerned.
Second, we found a higher percentage of Tregs in subjects undergoing liver transplantation than in controls. Again, this could be due to the pathologies responsible for their liver damages. Twenty patients out of 29 suffered from hepatocellular carcinoma (HCC), in most cases caused by chronic infection by HBV or HCV (7 and 10 cases, respectively). An increase in the frequency of Tregs in peripheral blood of HCC patients has been already described (31), and several authors have also found higher levels of these cells in different types of cancer (32,33). A chronic infection by HBV or HCV is able per se to induce an expansion of Tregs that likely play a protective role for the host, inhibiting the cytotoxic damage provoked by virus specific cells that attack the liver parenchyma (34).
Third, when we compared the effects of the treatments with Evr or CsA on the different T cell subsets studied, we observed that Evr patients were able to maintain constant the level of CD4+ T cells. Interestingly, also the percentage of naïve CD4+ T cells was higher in Evr patients. In patients who receive immunosuppression, maintaining a pool of naïve T cells, i.e., those able to respond to newly encountered foreign antigens can be of great importance for minimizing damages due to infective agents.
Fourth, in both groups, we observed initial increase in CD8+ T lymphocytes, which are known to play a major role in organ rejection. However, starting from 4 months after transplant, Evr patients displayed lower levels of such cells than CsA patients. As observed in CD4+ T cells, patients assuming Evr were also able to better preserve their pool of naïve CD8+ T lymphocytes. Among these cells, those expressing CD127 followed a similar trend in the two groups: an initial consistent decrease was followed by a recover of CD127+ naïve CD8+ T lymphocytes. However, the increase in naïve CD8+ T cells expressing CD127 was significantly higher in Evr group. The expression of IL-7 receptor on naïve T cells is necessary for an optimal utilization of IL-7, which promotes survival in the periphery (35). Therefore, a higher expression of CD127 would allow the maintenance of a higher pool of naïve cytotoxic lymphocytes, and guarantee the presence of cells in this differentiation status.
Fifth, regarding Tregs, we observed an opposite trend between the groups: Evr patients showed a progressive increase of the percentage of Tregs, whereas CsA group did not even maintain the levels present before liver transplantation. This phenomenon was well evident starting from 90 days after transplant. Animal models underline the importance of Tregs in preventing allograft rejection (36); some authors have also shown a decrease of peripheral circulating Tregs after liver transplantation, and a relative recovery of Tregs levels within the first year in patients who did not develop transplant rejection (35). It has been shown that rapamycin derivatives (such as Evr) are responsible for a selective expansion of Tregs because of the ability of these cells to survive rapamycin-induced apoptosis, occurring in other types of T lymphocytes, thanks to the constitutive expression of the serine-threonine kinase PIM2 (17). This resistance to apoptosis, together with the immunosuppressive effects, could contribute to better preserve the allograft function. Very few data exist on Evr and Tregs. A recent study reports that Evr decreases Treg population in renal transplant recipients (37). However, it is to note that in this study Tregs were identified by using three-color flow cytometry, without gating on CD3+ T cells, nor on living cells, and without excluding CD4+ monocytes from the analysis. Of note, patients under investigation were also treated with a different protocol, which included mycophenolate mofetil (MMF) in the first period after transplant, before the switch to Evr because of MMF side effects, including leukopenia.
Sixth, starting from 4 months after surgery, we observed a higher percentage of Tregs expressing the homing receptor CXCR3 in Evr patients when compared with CsA. CXCR3 is a chemokine receptor that allows T cells to reach inflamed tissues, such as the nonself liver (38). Thus, we could hypothesize that Treg-expressing CXCR3 cells are directed to this target, where they could exert their effects, and thus that Evr is more effective in preventing rejection also by allowing Tregs to exert an action in situ.
Seventh, we performed functional assays by stimulating PBMC in vitro with a superantigen like SEB, that activated several T cell Vβ families (39,40). At this regard, the main observation was that cells from Evr-treated patients display a minor total response by CD8+ T cells, and less production of IFN-γ. The polyfunctionality of either CD4+ or CD8+ T cells was well preserved in both groups. On the whole, these data suggest that Evr had higher immunosuppressive activity, at least on CD8+ T cells, which play a major role in the rejection.
In conclusion, taking into account all of the effects observed in Evr group such as: (i) a better preservation of naïve cells both in CD4+ and CD8+ T cell compartment; (ii) higher percentages of CD4+ but lower of CD8+ T cells; (iii) the presence of higher levels of Tregs and lower CD8+ response to SEB stimulation; (iv) the presence less severe side effects and a better preservation of renal function, as already shown by previous studies (41), our data suggest a wider use of this immunosuppressive drug after liver transplant. They also indicate the necessity to study the possible use of Evr in treating autoimmune diseases, where defects in the Treg compartment have been clearly demonstrated (42).