Abstract
The role of human papillomavirus (HPV) in human disease includes external genital and perianal warts (EGW), with some HPV genotypes having oncogenic potential (i.e., HPV-16 and -18). While green-tea extracts have antitumor and antiproliferative effects in vitro, the mechanism of action of sinecatechins in the treatment of EGW is not well understood. To investigate the role of immune-regulated genes further, an open-label, single institution, prospective study was conducted enrolling patients with clinically diagnosed EGW. Thirty subjects were enrolled, and 18 completed the trial. All patients applied sinecatechins 15% ointment to target lesions in the study. RNA expression microarrays were obtained from treated EGW lesions and analyzed for differential gene expression of immune-regulated genes. HPV types were analyzed and, based on copy number, were stratified into virological responders (VR) or nonresponders (VNR). Gene expression analysis of RNA samples was performed using TaqMan arrays for human T cell receptor and CD3 complex (TCR), Toll-like receptors (TLR) pathway, interferon (IFN) pathway, and antigen processing pathway. A total of 256 genes were analyzed across the four arrays. Genes that were significantly regulated between VRs and VNRs were CREB3L4, HIST1H3A, HIST1H3H, IFNA1, IFNA4, IFNA5, IFNA6, IFNA8, IFNA14, IFNG, IFNAR1, IL6, IRF9, MAPK4, MAPK5, MAPK14, NET1, and PIK3C2A in the IFN array. In the TCR array, HLA_B was found to be statistically significantly upregulated in both the VR and VNR groups; concomitantly, CD8A was found to be statistically significantly downregulated only in VRs. In the TLR array, only LBP and MAPK8 were found to be differentially regulated. In the antigen processing array, HLA-A, HLA-C, HLA-DMA, HLA-DMB, HLA-F, PSMA5, PSMB8, and PSMB9 were differentially downregulated. Based on these findings, it was determined that sinecatechins treatment modulates and downregulates genes involved in the pro-inflammatory response to HPV infection.
G
Green-tea extracts, and sinecatechins in particular, have been shown to modulate various cellular processes from apoptosis (14), inflammation signaling, and signal transduction pathways, including mitogen activated protein kinase (MAPK) signaling, Janus kinase (JAK) signaling, and PI3K/Akt signaling (5,27,34,35).
While the role of green-tea extracts has been well described in vitro, the mechanism of action of sinecatechins in the clinical setting are yet to be fully described. The authors recently published the first in vivo study, which focused on apoptotic gene expression of EGW after sinecatechins treatment (14). The study found that there was a downregulation of antiapoptotic genes verifying previous preclinical models. The study was the first to evaluate the in vivo alterations in pathways involved in EGW pathogenesis. To investigate the involvement of immune-regulated genes further, additional molecular studies were performed on sinecatechins-treated EGW as described previously.
Briefly, this was an open-label, single institution, prospective study enrolling patients with clinically diagnosed EGW. RNA expression microarrays on clinical EGW samples treated with sinecatechins were employed for differential gene expression analysis of immune-regulated genes. Thirty subjects (24 male) were enrolled, and 18 completed the trial. All patients received sinecatechins 15% ointment and were instructed to apply the ointment to target lesions for the duration of the study.
Biopsies were obtained at baseline (B1), at the first visit where 50% of target lesions had cleared (B2), and at the first visit with complete lesional clearance (B3). DNA and RNA were extracted from tissues using the Trizol reagent (Sigma-Aldrich). Clinical samples were next analyzed for HPV types as previously described (8), and the copy number was determined by real-time polymerase chain reaction (RT-PCR) with a custom primer set. Gene expression analysis of RNA samples was performed using TaqMan arrays for human T-cell receptor and CD3 complex, Toll-like receptors (TLR) pathway, interferon (IFN) pathway, and antigen processing pathway. Patients were then stratified into viral responders (VR) and nonresponders (VNR) as determined by the change in viral copy number between B1 and B3, as previously described (14).
Gene expression analysis was performed individually, and then group-based response values were generated using DataAssist software (Life Technologies). This group-based response was generated by determining fold changes between the following: biopsy 1 and biopsy 2 (B1_2), biopsy 2 to biopsy 3 (B2_3), and biopsy 1 to biopsy 3 (B1_3). These group-based responses were as follows: (B1_2) responses correlated between the start and the middle of the study, (B2_3) responses correlated between the middle and end of the study, and (B1_3) responses analyzed the response of the particular gene for the entire duration of the study. Statistical significance of these fold changes was determined by a nonparametric two-sided Wilcoxon signed-rank test with significance determined at the p<0.05 level.
A total of 256 genes were analyzed across the four arrays. In the interferon pathway associated gene array, genes that were significantly regulated between VRs and VNRs were: CREB3L4, HIST1H3A, HIST1H3H, IFNA1, IFNA4, IFNA5, IFNA6, IFNA8, IFNA14, IFNG, IFNAR1, IL6, IRF9, MAPK4, MAPK5, MAPK14, NET1, and PIK3C2A. Tables 1 and 2 show the results in the VRs and VNRs with associated p-values. The genes for interferon-α-4 and -5 were significantly downregulated in VRs between B1_2, and could represent an early response to sinecatechins treatment. The receptor for IFNAs (IFNAR1) showed significant downregulation during the treatment (B2_3) of the VRs group.
From the T cell receptor array, HLA_B was found to be statistically significantly upregulated in both the VR and VNR groups; concomitantly, the CD8A gene was found to be statistically significantly downregulated only in VRs. HLA-B belongs to the group of MHC class I molecules that function in antigen presentation to cytotoxic T cells, which are typically positive for the CD8 receptor, encoded by the CD8A gene. Interestingly, for the VRs, HLA-B was upregulated 2.76-fold (p=0.0469) only in VRs between B2_3 comparison, roughly slightly elevated later in the study. In the VNR group, HLA-B was actually upregulated 7.519-fold (p=0.0098) in the B1_3 comparison, translating to an overall upregulation in VNRs for the entirety of the study. In the VNR biopsy (2_3) comparison, this upregulation was noted to be 14.7961-fold (p=0.001). This correlates with a persistent upregulation of HLA-B gene throughout the duration of the study in VNRs treated with sinecatechins, and furthermore, toward the end of the study, there seemed to be a further upregulation. The CD8A gene was downregulated by 0.46978-fold (p=0.0461) in the VR group only between B1_3.
In the Toll-like receptor array, only two genes were found to be differentially regulated: LBP and MAPK8. Importantly, these two genes were found to be upregulated at least twofold only in the VNR group between both B1_2 and B1_3. No genes were differentially upregulated or downregulated in the VR group.
In the antigen-processing array, HLA-A, HLA-C, HLA-DMA, HLA-DMB, HLA-F, PSMA5, PSMB8, and PSMB9 were differentially downregulated, while no genes were upregulated. Importantly, HLA-A, HLA-DMA, and, HLA-F were downregulated in the VR group, whereas HLA-C, HLA-DMA, HLA-DMB, PSMA5, PSMB8, and PSMB9 were downregulated in the VNR group. Only HLA-DMA was downregulated in both the VR and VNRs. Of note, HLA-F was found to be downregulated to virtually undetectable levels (0.00773-fold, p=0.0313) in the biopsy (1_3) comparison, correlating to a persistent downregulation throughout the study. Similarly, HLA-C, HLA-DMA, HLA-DMB, PSMA5, PSMB8, and PSMB9 were also downregulated in their respective responder groups in the biopsy (1_3) comparisons.
Discussion
Significant receptor downregulation in response to decreased expression of IFNA substrates solely in the VR group may suggest a sinecatechins-induced negative feedback on the IFNA inflammatory response pathway. It is possible that the viral load in the VR group decreased by mechanisms that are independent of a direct interferon-mediated antiviral response and may be related to the previously described anti-inflammatory/antiproliferative activity of sinecatechins.
We observed a statistically significant downregulation of the IL6 gene, the gene for interleukin-6 seen only in the VR group. Importantly, this downregulation was 0.1974-fold (p=0.0469) between B1_2, and persisted between B2_3 by 0.30621-fold (p=0.0156). IL-6 has previously been shown in several cell culture models to be inhibited by green-tea extracts (1,23,33). This downregulation was exhibited at all points of the study. These results allow us for the first time to suggest that IL-6 expression decreases upon sinecatechin treatment occurring only in VR. Concomitantly, in those with increasing viral loads, IL-6 expression does not change in response to sinecatechins (Table 2). This correlation between sinecatechins treatment and IL-6 expression corroborates previously published preclinical models and suggests a mechanism of sinecatechins efficacy in the clinical setting (1,23,33). Additionally, MAPK14, which encodes p38 MAPK, was downregulated only in VRs by 0.42214-fold (p=0.0156) in the biopsy (2_3) comparison, corresponding to a late downregulation upon sinecatechins treatment, in direct contrast to previous in vitro studies (5,34). Additionally, a statistically significant downregulation of PIK3C2A was found, the gene for phosphatidyl-inositol-3-kinase (PI3K) in VNRs, by 0.47883-fold (p=0.0098) in the biopsy (1_3) comparison. Furthermore, it was found that MAP2KK, the gene for MAPK6/MEK6, was downregulated by 0.41898-fold (p=0.0313) for the study duration, corresponding to the biopsy (1_3) comparison. Importantly, MAPK6 has been shown to phosphorylate p38-MAPK (21). The finding that the genes for the effectors of the PI3K/Akt pathway and the p38-MAPK pathways were downregulated upon sinecatechins treatment is novel, as previously published reports of MAPK pathway attenuation in cells treated with green-tea extracts have demonstrated an increase in apoptosis from activation of p38 MAPK in ovarian cancer lines (2,12,31). How such activation can promote apoptosis or viral clearance in EGW is an area of growing interest.
It was observed in the T cell receptor array that the downregulation of T cell associated genes suggests an immune dysregulation in antigen presentation associated MHC class I molecules, namely HLA-B. A paucity of CD8A expression may be associated with lower viral clearance, since this correlates with the viral-load data. These findings additionally corroborate previously reported studies demonstrating that green-tea extracts decrease T cell activation in an IL-2 dependent manner (30,32). Whether the current findings represent a result of sinecatechins or an increased viral load in EGW requires further study.
The LBP gene encodes lipopolysaccharide binding protein, LBP, first described as an acute-phase reactant to gram-negative bacterial infection (9). LBP, along with the bactericidal permeability-increasing protein (BPI) binds CD14 and mediates monocyte activation (19). The MAPK8 gene encodes the c-Jun-N-terminal-Kinase 1 protein, JNK1, and participates in MAPK activation. Importantly, JNK1 is activated by TNF as a stress-response signaling to apoptosis pathways (10,29). In this context, both LBP and JNK1, protein products of LBP and MAPK8 respectively, are upregulated only in VNRs presumably as a result of persistent HPV infection. Previous reports, however, have shown that EGCG, the most active green-tea extract, promotes oxidative stress and activates JNK1 phosphorylation, although these studies only evaluated the JNK pathway and not specifically JNK1 mRNA expression (4,13). As such, this study is the first to report an in vivo downregulation of MAPK8 gene expression in sinecatechins-treated EGW. Whether this is a response to sinecatechins therapy or despite sinecatechins therapy remains to be elucidated.
The finding that the HLA genes are downregulated in sinecatechins treatment is novel, as previous studies have demonstrated an increase in T cell activation in vitro models. In fact, the current data set also found a decreased IFNG, the gene for IFN-g, in VNR. The current findings support previous preclinical in vitro studies of IFNG (16,22). Additionally, the finding of a downregulation of the interferon genes IFNA4, IFNA5, and IFNA14 in VR is the first to be reported and suggests novel mechanisms of immune regulation in VR that warrant further investigation.
As noted, the HLA-C, HLA-DMA, HLA-DMB, PSMA5, PSMB8, and PSMB9 genes were also downregulated in their respective responder groups in the biopsy (1_3) comparisons. The HLA genes are associated with the major histocompatibility complex group of proteins involved in antigen presentation and the PSMA and PSMB groups of genes encode proteasomal proteins. The observed overall downregulation in VNRs could be associated with downregulation of the immune response allowing for a permissive environment for viral replication. The mechanisms are not well understood, and recent studies highlight the importance of polymorphisms in antigen presentation in HPV-associated disease (6,28).
Conclusion
Based on the current findings, it was determined that sinecatechins treatment modulates and downregulates genes involved in pro-inflammatory response to HPV infection, which correlates closely with preclinical models of green-tea extracts (7,11,15). This observed overall downregulation of immune regulatory genes coupled with our previous finding of the upregulation of apoptosis-associated genes in virological responders suggests an overall mechanism of EGW clearance. This is novel and surprising, as other approved modalities for the treatment of warts and EGW, which include 5-fluoruracil, podophylin, and imiquimod, incite an immune response (3,18,24). While the prevalent HPV-6 type in VR lesions are not typically associated with malignancy, unregulated proliferation is a feature of EGW. The connection between inflammation and cancer cell proliferation is well known; in fact, several studies have shown that green-tea extracts have an antitumor effect. In the current study, a downregulation of inflammation-associated genes was found, including MAK8 and PIK2C3A, which are implicated in cancerous cell proliferation (20,31). Taken together with the authors' previous report, the current study provides sufficient findings to demonstrate that the mechanism of action and observed clinical efficacy of sinecatechins is, at least in part, due to promoting apoptosis and reducing inflammatory signaling. Given the above findings, it is likely that sinecatechins possibly dampen both the immune response and inflammatory signaling with a concomitant activation of apoptosis further allowing for EGW clearance. Thus, there is an overall tendency toward EGW clearance. Future studies will need to be directed toward identifying appropriate responders for specific therapy and to tease out further these immune-regulatory mechanisms observed in the current study.
Footnotes
Author Disclosure Statement
This study was funded by Fougera Pharmaceuticals, Inc., through study grants provided to Stephen K. Tyring. The remaining authors report no potential conflicts of interest.