Low doses of the novel caspase-inhibitor GS-9450 leads to lower caspase-3 and -8 expression on peripheral CD4+ and CD8+ T-cells
- First Online:
- Cite this article as:
- Arends, J.E., Hoepelman, A.I.M., Nanlohy, N.M. et al. Apoptosis (2011) 16: 959. doi:10.1007/s10495-011-0620-2
- 543 Downloads
Chronic hepatitis C virus (HCV) infection is characterized by increased rates of apoptotic hepatocytes and activated caspases have been shown in HCV-infected patients. GS-9450, a novel caspase-inhibitor has demonstrated hepatoprotective activity in fibrosis/apoptosis animal models. This study evaluated the effects of GS-9450 on peripheral T-cell apoptosis in chronic HCV-infected patients. As sub study of the GS-US-227-0102, a double-blind, placebo-controlled phase 2a trial evaluating the safety and tolerability of GS-9450, apoptosis of peripheral CD4+ and CD8+ T-cells was measured using activated caspase-3, activated caspase-8 and CD95 (Fas). Blood samples were drawn at baseline, day 14 after therapy and at 5 weeks off-treatment follow-up in the first cohort of 10 mg. In contrast to the placebo-treated patients, GS-9450 caused a median of 46% decrease in ALT-values from baseline to day 14 in all treated patients (median of 118–64 U/l) rising again to a median of 140 U/l (19%) at 5 weeks off-treatment follow-up. In GS9450-treated patients, during treatment and follow-up, percentages of activated caspase-3+ and caspase-8 expression tended to decrease, in contrast to placebo-treated patients. Interestingly, compared to healthy controls, higher percentages of caspase-3 and caspase-8 positive CD4+ and CD8+ T-cells were demonstrated in HCV-infected patients at baseline. Decreased ALT-values were observed in all HCV-infected patients during treatment with low dose of the caspase-inhibitor GS-9450 accompanied by a lower expression of caspase-3 and -8 on peripheral T-cells. Furthermore, at baseline percentages of activated caspase-3, activated caspase-8 and CD95+ T-cells were higher in chronic HCV-infected patients compared to healthy controls.
KeywordsHepatitis C virus Apoptosis Caspase-3 Caspase-8 CD95
Persistence of the Hepatitis C Virus (HCV) in the liver leads to the development of liver fibrosis over the course of decades . This process of liver fibrosis progression is a complex interplay between activated T-cells, activated stellate cells and apoptosis of injured hepatocytes . Apoptosis or programmed cell death is a physiological phenomenon that occurs in most organs in the human body and is a key element in the host defense against viral pathogens . The main signaling pathway to induce apoptosis in hepatic diseases and normal healthy livers is the death receptor pathway consisting of different receptors of the Tumor Necrosis Factor family being TNF-related apoptosis-inducing ligand (TRAIL), TNF-alpha and Fas (CD95/APO-1) . Fas ligand (FasL) is expressed by natural killer cells and cytotoxic T lymphocytes (CTLs, i.e. CD8+ effector T-cells) and Fas–FasL interaction is the main effector mechanism of CTLs inducing apoptosis of virus infected hepatocytes [5, 6]. Recently, in a HCV genotype 1b transgenic mouse model an increased rate of apoptosis of peripheral CD4+ and CD8+ T-cells was noted when compared to healthy control mice . Furthermore, this was associated with an up-regulation of FasL on the hepatocytes suggesting that the hepatic microenvironment, with up-regulation of FasL, promotes increased T-cell apoptosis and thereby contributes to viral persistence.
Apoptosis is gradually being more recognized as an important factor in liver fibrosis development [8, 9, 10]. Caspases, proteolytic enzymes belonging to a family of intracellular cysteine proteases, play an important role in this apoptotic process. After interaction with the cell surface Fas-receptor, intra-cellular activation of death domains (FADD) result in proteolytic cleavage pro-caspase-8 into its active form caspase-8 . This process leads downstream to the cleavage of pro-caspase-3 into its active form caspase-3, the central protease in the apoptosis pathway. The role of apoptosis in chronic HCV is currently not well understood. Liver biopsy studies in patients with chronic HCV have shown an increased presence of apoptotic hepatocytes [12, 13]. Moreover, the percentage of apoptotic hepatocytes, demonstrated immunohistochemically as caspase-3 positive cells, has been shown to correlate with the amount of liver fibrosis [14, 15].
GS-9450 is an irreversible inhibitor of caspase-8, -9, and -1 and has demonstrated hepatoprotective activity in both fibrosis assays and apoptosis animal models (unpublished). Furthermore, a phase-1 trial dosing GS-9450 for 14 days in healthy volunteers proved to be safe and well tolerated . Recently, the results of a phase-2a study, evaluating the safety and tolerability of GS-9450 have been presented . However, since both a baseline liver biopsy was not mandatory for entering the phase-2 study and in the previous phase-1 study, the effect of GS-9450 on the T-cells was not evaluated, it was decided to analyze the effects of GS-9450 on peripheral T-cell apoptosis during GS-9450 therapy as a sub study during the phase-2a study.
Patients and methods
The GS-US-227-0102 was a phase 2a trial evaluating the safety and tolerability GS-9450, a potent irreversible inhibitor of caspase-8, -9, and -1. This novel drug is currently developed by Gilead Inc. (Durham, NC, USA). Ascending doses of GS-9450 were evaluated in this randomized placebo-controlled study . For the first cohort of 10 mg GS-9450, 8 patients were included from the Netherlands, 6 receiving GS-9450 and 2 receiving placebo. Inclusion was based on the presence of a chronic HCV infection and ALT or AST >1.5× the upper limit of normal (ULN; in our centre 35 U/l). Furthermore, patients had to have previously failed conventional anti-HCV therapy, were unable to tolerate it, or had contraindications for treatment with (peg)interferon-alfa/ribavirin. Important exclusion criteria were decompensated liver disease or evidence of hepatocellular carcinoma (based on liver biopsy within the previous 2 years), coinfection with hepatitis B virus (HBV) or human immunodeficiency virus (HIV) and current or near future pregnancy. A laboratory sub study of this trial was designed for this first cohort, evaluating the effects of GS-9450 on peripheral T-cells.
Peripheral blood (approximately 30 ml) was collected at baseline (i.e. day 0 of the study), at week 2 (i.e. day 14 of the study) and at week 7 (i.e. 5 weeks off-treatment follow-up). Within 24 h, peripheral blood mononuclear cells (PBMC) were isolated using a Ficoll-Hypaque density gradient centrifugation. Cells were re-suspended in RPMI 1640 (Gibco Life Technologies, Breda, The Netherlands) to which 20% fetal calf serum and 1% penicillin and streptomycin were added (hereafter called medium). Except for the caspase-8 experiment, which were performed on frozen PBMC, all assays were performed on fresh PBMC. As controls, blood of 5 healthy individuals (Healthy Controls, HC) was requested from the local blood bank (screened and negative for hepatitis B, HIV and hepatitis C).
Selection of apoptosis markers on peripheral T-cells
In vitro analysis of the caspase-inhibitor GS-9450 showed inhibition of activated caspase-8. Since caspase-3 (i.e. the main caspase protein in the apoptosis pathway) is a downstream caspase it was assumed that by inactivating caspase-8, the expression of caspase-3 would be reduced. Furthermore, CD95 of Fas is the main receptor initiating apoptosis in HCV-infected patients and upstream of caspase-8. Therefore, CD95, caspase-3 and -8 were chosen as markers for the measurement of apoptosis in peripheral T-cells. Although annexin-V/PI is a well-known assay to establish apoptosis, we choose to examine those markers that are directly involved in the working mechanism of GS9450.
Apoptosis analysis by FACS
PBMC in medium were washed and resuspended in PBA (phosphate buffered saline containing 0.16% BSA and 0.1% sodium-azide) to a concentration of 2 million PBMC/ml and divided in aliquots of 200 μl. Cells were incubated for 15 min at 4°C with 2 different sets of antibodies: (i) against activated caspase-3 (FITC, BD Biosciences, San José, CA, USA) and (ii) against activated caspase-8 (FITC, Biovision, Mountain View, CA, USA) Added to both these two different sets were antibodies against CD3 (pacific Blue), CD4 (PECy7, BD), CD8 (APCCy7, eBioscience, San Diego, USA) and CD95 (PE, BD). Cells for caspase-expression were washed and permeabilized with permalysis buffer (BD) and stained with antibodies for activated caspase-3 or activated caspase-8 for 15 min at 4°C. After washing, cells were fixed with Cellfix (BD) and at least 200,000 cells were acquired by Fluorescence Activated Cell Sorting (FACS, LRSII, BD).
Using the FACSDiva software (BD), we constructed a CD4 PECy7 and CD8 APCCy7 scatter plot, gated CD3+ T-cells, in order to identify all CD3+CD4+ and CD3+CD8+ T-cells. Within these populations we gated on activated caspase-3 and CD95 to obtain the percentage of caspase or CD95 expressing CD3+CD4+ or CD3+CD8+ T-cells.
Continuous data are presented as median values [with interquartile range (IQR)] and analyzed non-parametrically using a Mann–Whitney U-test for baseline values and a Friedman test for longitudinal analysis. Categorical variables are given as numbers of cases (percentage). A P value ≤0.05 was considered as statistically significant and all tests used were two-sided. All data were analyzed using GraphPad Prism (version 5.0 for Windows, GraphPad Software, San Diego, USA).
Characteristics of the patients treated with either GS-9450 or placebo
Dizziness and nausea
Fatigue, somnolence and feeling hot
Hyperhidrosis, affect lability, flank pain and tooth infection
Headache, palpitations, dry skin, flank pain and parosmia
Vomiting, affect lability, nausea, headache, sore throat, rhinitis and fatigue
Headache and gastro-enteritis
Nausea, headache, fatigue and back pain
Baseline expression of caspase-3, caspase-8 and CD95 on peripheral T-cells
Longitudinal analysis of caspase-3 and CD95 expression in peripheral T-cells
Since analysis of activated caspase-8 expression was evaluated on frozen PBMC, we investigated the effects of freezing/thawing on activated caspase-8 expression. Comparison of activated caspase-8 expression between frozen and fresh PBMC in healthy controls did not show significant differences between the two methods (P = 0.64; Wilcoxon signed rank test). Activated caspase-8 expression in CD4+ T-cells decreased in all but one patient, returning to baseline in 3 patients or declining further in another 4 patients at week 7 (P = 0.74 for GS9450 treated and P = 1.0 for placebo treated patients; Friedman test) (Fig. 4c). Similarly, expression of activated caspase-8 by CD8+ T-cells showed a decrease in 5 of 8 patients returning to baseline in 4 of them (P = 0.57 for GS9450 treated and P = 0.50 for placebo treated patients; Friedman test) (Fig. 4d). A mostly similar expression of activated caspase-3 and activated caspase-8 was noted in both placebo-treated and GS9450-treated patients demonstrating that inactivation of activated caspase-8 lowers the downstream expression of activated caspase-3. This is also demonstrated by the significant correlation between the expression of activated caspase-3 and activated caspase-8 (Spearman r = 0.40 and P = 0.008) when CD4+ and CD8+ T-cells at weeks 0, 2 and 7 were analysed together.
Differences in expression of CD95 by CD4+ T-cells during the study were less pronounced and not significant with 5 patients showing minimal changes while 3 patients showed a decrease in CD95 expression (Fig. 4e). In CD8+ T-cells CD95 expression decreased in 5 of 8 patients from baseline to week 2 and even further at week 7. A significant correlation between the expression of activated caspase-3 and CD95 was noted (Spearman r = 0.34 and P = 0.02).
In this study we evaluated the effects of the caspase-inhibitor GS-9450 on peripheral T-cells during a 14-day treatment period. Although treatment with GS-9450 resulted in lower ALT-values, it did not affect either the HCV viral load or the peripheral T-cells apoptosis rates. However, we demonstrate that at baseline, percentages of activated caspase-3, activated caspase-8 and CD95 T-cells are higher in chronic HCV-infected patients compared to healthy controls.
The effect of GS-9450 on lowering the ALT-value was strong (nearly 50% reduction) suggesting a decrease in apoptotic hepatocytes and thus prevention of release of aminotransferase enzymes. In addition, cessation of alcohol consumption was a study requirement and might explain in part the ALT-decrease in these patients. This is supported by the observation that the ALT-values of one placebo-treated patient decreased during therapy and even further declined after placebo-discontinuation at the time ALT-values in GS-9450 treated patients increased again. A competitor caspase-inhibitor, of which development recently has been stopped, showed a similar decline in ALT-values (in a dose-depended matter) in chronic HCV patients in both a 14-day and a 12-week dosing trial [18, 19]. In all these caspase-inhibitor studies an overshoot of aminotransferase activity occurred after discontinuation of the drug which might be caused by an abrupt release of blockade of apoptosis. This could be prevented by tapering instead of abrupt discontinuation and should be investigated in future studies.
In contrast to the decrease in ALT-values, treatment with GS-9450 did result only in a slight decrease in caspase-8 expression on peripheral T-cells. In rats it was found that the concentration of GS-9450 was about 30-fold higher in the liver compared to peripheral blood. If this concentration difference is true for human, than this hepatoselective effects of GS-9450 could be one explanation for small effect on caspase-8 expression in peripheral T-cell. Another possible explanation for this small effect on caspase-8 expression could be that the dose of 10 mg GS-9450 was too low to exert a larger effect on peripheral T-cell apoptosis while maximal concentrations were reached intra-hepatically. The relatively higher drop in activated caspase-3 levels could be explained from the working mechanism of GS9450. By inactivating activated caspase-8, the percentage of downstream cascadic activated caspase-3 expression is much more affected.
An interesting observation in this study was the significant differences in activated caspase-3, activated caspase-8 and CD95 expression by peripheral T-cells between chronic HCV patients and healthy controls. This could not be explained by a difference in T-cell activation markers (HLA-DR/CD38) or T-cell phenotypes (CD27/CD45RO) since no difference between HC and HCV-infected patients could be demonstrated for these T-cell features (data not shown). The reason for the higher rate of apoptosis in chronic HCV patients is currently unknown. It has been shown by Toubi et al.  that both spontaneous and dexamethasone-induced apoptosis was increased in chronic HCV-infected patients compared to healthy individuals. They also showed that dexamethasone-induced apoptosis of peripheral T-cells was correlated with increased ALT-values and decrease NF-κB levels. Furthermore, experimental HCV murine models have suggested that both HCV core and non-structural proteins enhance apoptosis of activated CD4+ and CD8+ T-cells via up-regulation of Fas–FasL interaction [7, 21]. This may suggest that HCV is capable of down-regulating specific immune responses by inducing T-cell apoptosis and therefore contributing to persistence of a chronic HCV infection.
This study has several limitations. First, as staging of liver fibrosis was not part of the study, correlation between liver fibrosis and apoptosis-markers could not be performed. This would have been of interest given the earlier publications demonstrating a relation between intra-hepatic caspase expression and the degree of fibrosis . When a correlation between the stage of liver fibrosis and peripheral T-cell apoptosis at baseline could be demonstrated, this might be a valuable marker for evaluation of liver fibrosis. However, no correlation was shown between ALT-values and either activated caspase-3, activated caspase-8 or CD95 expression (data not shown). Furthermore, the number of patients in this study was small possibly explaining the absence of significance in baseline CD95+CD8+ T-cells between HCV-infected patients and HC. Finally, the caspase-8 analyses were performed on frozen T-cells. Freezing of cells is known to induce apoptosis in cells [3, 14]. However, the differences in activated caspase-8 expression between fresh and frozen PBMC was limited (data not shown) and longitudinal changes in expression of activated caspase-3 and activated caspase-8 was similar. Therefore we think that the effects of freezing in this study is negligible. However, further studies of the role of activated caspase-3 and activated caspase-8 by peripheral T-cells is needed.
In conclusion, the caspase-inhibitor GS-9450 decreased ALT-values during a 14-day dosing period. Furthermore, the expression of activated caspase-8 and mainly of activated caspase-3 by peripheral T-cells decreased during the treatment period. Moreover, baseline expression of activated caspase-3, activated caspase-8 and CD95 were found to be higher in patients with chronic HCV compared to healthy controls.
Conflict of interest
Although this study was a laboratory sub study of the clinical trial GS-US-227-0102, the study was not sponsored by either company (Gilead and LG Life Sciences). K.R. Hirsch was employee of Gilead and J.G. Park is employee of LG Life Sciences. All other authors have no conflict of interest to report.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.