Background
Human T-lymphotropic virus type-I (HTLV-I) is a retrovirus that infects 10 to 20 million people worldwide [
1]. There are areas in sub-Saharan Africa, the Caribbean, and South America where >1% of the general population is infected, [
2] and southwestern Japan including Nagasaki Prefecture is one of the endemic areas [
3]. Although the majority of infected people remain asymptomatic, HTLV-I is associated with severe diseases such as adult T-cell leukemia/lymphoma (ATL) and HTLV-I-associated myelopathy (HAM). Many strategies have been evaluated for the treatment of ATL and HAM, but no treatments have shown sufficient efficacy.
Tumor necrosis factor alpha (TNF-α) inhibitors (TNFi) are an important agent for a number of inflammatory conditions, including rheumatoid arthritis (RA), [
4] ankylosing spondylitis, [
5] and inflammatory bowel disease [
6]. However, multiple adverse effects of TNF-α inhibition have been identified, including infections, malignancies, and the induction of autoimmunity and demyelinating diseases. With respect to viral infection, hepatitis B virus (HBV) occasionally reactivates, and a flare of HBV disease may occur [
7].
However, it is unknown whether HTLV-I proliferates and whether HTLV-I-associated diseases worsen when biologics including TNFi are used. Answers to these questions are needed by clinicians who use biologics. In Japan, approximately one million individuals are carriers of HTLV-I, [
8] which means that one person per 100 individuals has an HTLV-I infection. In an RA cohort study, 21.3% of the RA patients were treated with a TNFi [
9]. Whenever possible, clinicians would prefer to avoid the use of TNFi to treat HTLV-I-infected patients, but in the case of patients with RA complicated by HTLV-I infection, the use of TNFi is unavoidable due to the high prevalence of both conditions. Because the use of biologics for such patients is relatively new, the problem of biologics–induced enhancement of HTLV-I in RA patients is also a fairly new concern. In addition, a significant increase in the standardized incidence ratio for malignant lymphoma was identified in a Japanese nationwide cohort of patients treated with biological disease-modifying anti-rheumatic drugs (DMARDs) including TNFi, [
10] but that study did not reveal whether the standardized incidence ratio for ATL increased. We searched for cases of ATL or HAM patients treated with a TNFi as an autoimmune disease treatment by conducting a PubMed search, but to the best of our knowledge, there were no such reports with the exception of one smoldering ATL case [
11].
For the above reasons, it is necessary establish whether a TNFi can be used safely to treat patients with inflammatory diseases such as RA complicated with HTLV-I infection. For this purpose, we plan to perform both in vitro and clinical investigations to ascertain the safety of TNFi treatment in patients with HTLV-I infection. To this end, we herein assessed changes in the cytokine profiles, associated proteins, proviral load (PVL), and apoptosis in an HTLV-I infected cell line treated with several different TNFi.
Methods
Cell lines
The HTLV-I-infected T-cell line HCT-5 derived from the cerebrospinal fluid cells of a HAM patient was used. This cell line is interleukin (IL)-2-dependent and was maintained in RPMI 1640 (Wako Pure Chemical Industries, Tokyo) containing 20% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, MA) and 1% penicillin/streptomycin (Thermo Fisher Scientific) supplemented with 100 U/ml of recombinant human IL-2 (kindly provided by Shionogi & Co., Osaka, Japan) and L-glutamine (Sigma-Aldrich, St. Louis, MO). Jurkat cells, a human T-cell lymphoblast-like cell line, were used as a control cell line. THP-1, a human monocytic leukemia cell line, was used to ascertain whether the TNFi worked as expected to inhibit TNF-α.
These cell lines were maintained in RPMI 1640 containing 10% FBS and 1% penicillin/streptomycin. All cell lines were incubated in a humidified incubator at 37 °C with an atmosphere of 5% carbon dioxide.
Reagents and TNFi
We purchased infliximab (IFX; Centocor, Malvern, PA), adalimumab (ADA; Abbott, Abbott Park, IL), etanercept (ETN; Amgen, Thousand Oaks, CA), golimumab (GLM; Janssen Biotech, Horsham, PA) and certolizumab pegol (CZP; UCB Pharma, Brussels, Belgium). We used lipopolysaccharide (LPS; Sigma-Aldrich) to stimulate the THP-1 cells.
Multiplex cytokine/chemokine bead assays
We performed multiplex cytokine/chemokine bead assays using culture supernatants and Milliplex MAP Human Cytokine/Chemokine Panel 1 Pre-mixed 41Plex (Merck Millipore, Darmstadt, Germany). We analyzed the results with a Bio-Plex® MAGPIX™ Multiplex Reader (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions. The cytokines/chemokines that could be measured by the Milliplex MAP Human Cytokine/Chemokine Panel 1 Pre-mixed 41Plex include vascular endothelial growth factor (VEGF), tumor necrosis factor-beta (TNF-β), TNF-α, transforming growth factor-α (TGF-α), regulated on activation, normal T cell expressed and secreted (RANTES), platelet-derived growth factor (PDGF)-AB/BB, PDGF-AA, macrophage inflammatory protein (MIP)-1α, macrophage inflammatory protein (MIP)-1β, macrophage-derived chemokine (MDC) (C-C motif chemokine [CCL]22), monocyte chemotactic protein-3 (MCP-3), monocyte chemotactic protein-1 (MCP-1), inducible protein-10 (IP-10), IL-2, -3, -4, -5, -6, -7, -8. -9, -10, -12, -13, -15, -17, IL-12 (p70), IL-12 (p40), IL-1ra, IL-1β, IL-1α, interferon-gamma (IFN-γ), interferon-α2 (IFN-α2), growth-related oncogene (GRO), granulocyte macrophage colony stimulating factor (GM-CSF), granulocyte colony stimulating factor (G-CSF), fractalkine, Fms-related tyrosine kinase 3 ligand (Flt-3 L), fibroblast growth factor (FGF)-2, eotaxin, epidermal growth factor (EGF), and soluble CD40L (sCD40L). We used the Bio-Plex Pro Human Cytokine Group II x-Plex Panel (Bio-Rad) to determine the serum levels of intercellular adhesion molecule (ICAM)-1, vascular cell adhesion molecule (VCAM)-1, IL-18, and stromal cell-derived factor (SDF)-1α.
Cytokine enzyme-linked immunosorbent assays (ELISAs)
We used a cytokine ELISA immune assay to measure the levels of IL-6, TNF-α, sICAM-1/CD54 and CCR5/RANTES in culture supernatants of HCT-5 and Jurkat cells to which we had added five types of TNFi. The IL-6, TNF-α, sICAM-1/CD54 and CCR5/RANTES Quantikine ELISA kits were purchased from R&D Systems (Minneapolis, MN). The level of each cytokine was determined according to the manufacturer’s instructions, and then the optical density at 450 nm was measured.
TNF-α receptor analysis
We performed fluorescence-activated cell sorting (FACS) analysis to examine the cell surface expression of TNF-R1 and -R2, using fluorescein isothiocyanate (FITC)-conjugated anti-CD120a (TNF-R1) and anti-CD120b (TNF-R2) human monoclonal antibodies (mAbs; MBL International, Woburn, MA). FITC-conjugated mouse IgG1 was used as an isotype control (BD Biosciences, San Jose, CA). Experiments were performed using a FACS Canto II Flow Cytometer and FACS Diva software (BD Biosciences).
RNA extraction and quantitative reverse transcription PCR for gene expression
We extracted total RNA from HCT-5 and Jurkat cells by using a Kingfisher Pure RNA Blood kit (Thermo Fisher Scientific) according to the manufacturer’s instructions, and the total RNA was applied for cDNA synthesis with a SuperScript VILO cDNA Synthesis Kit (Invitrogen, Carlsbad, CA). Reverse transcription was performed with 1 μg of total RNA at 25 °C for 10 min, 42 °C for 60 min, and 85 °C for 5 min, in a final volume of 20 μl.
The expressions of viral Tax (forward, 5′-CCCACTTCCCAGGGTTTGGA-3′; reverse, 5′-GGCCAGTAGGGCGTGA-3′; probe, 5′-FAM-CCAGTCTACGTGTTTGGAGACTGTGTACA-TAMRA-3′), and HTLV-I bZIP factor (HBZ) mRNAs (forward, 5′-CTCAGGGCTGTTTCGATGCT-3′; reverse, 5′-GCCCGTCCACCAATTCCT-3′; probe, 5′-FAM-CCTGTGTCATGCCCGGAGGACC-TAMRA-3′), and porphobilinogen deaminase (PBGD, forward, 5′-AACCAGCTCCCTGCGAAGA-3′; reverse, 5′-CCAGGATGATGGCACTGAACT-3′; probe 5′-FAM-ACTCCTGAACTCCAGATGCGGGAACT-TAMRA-3′) were measured using a cDNA template with LightCycler 480 probes Master Mix (Roche Diagnostics, Mannheim, Germany) and a LightCycler480 PCR System (Roche Diagnostics).
After 50 cycles, the absolute amounts of Tax, HBZ, and PBGD mRNA were interpolated from standard curves generated by the dilution method using plasmids derived from a clone transfected with a pCR2.1-TOPO Vector (Life Technologies, Tokyo, Japan) containing amplicons from the Tax, HBZ, and PBGD genes, respectively. To normalize the results for variability in the concentration and integrity of the RNA and cDNA, we used the PBGD gene as an internal control for each sample.
HTLV-I proviral load (PVL)
We extracted the genomic DNA of cells using Qiagen DNA Blood Mini kits (Qiagen, Crawley, UK). The quantitative polymerase chain reaction (qPCR) detection for HTLV-I was performed as described previously [
12‐
15]. Briefly, primers were set in the pX region, and the density of the template was 30 ng per reaction. PVL was quantified using the Tax primer and probe. The PVL was normalized using β-globin and is presented as a percentage.
Immunofluorescence
The HCT-5 cells were incubated for 10 min in phosphate-buffered saline (PBS) containing 4% paraformaldehyde at 4 °C and immersed in methanol at −20 °C for 10 min. After blocking in 5% normal horse serum in PBS, the cells were incubated in the diluted primary antibodies for 1 h at room temperature, followed by incubation in FITC-labeled and tetramethylrhodamine isothiocyanate (TRITC)-labeled secondary antibodies supplemented with Hoechst dye 33258 for nuclear staining. After being washed in PBS, the cells were mounted in Vectashield mounting medium (Vector Laboratories, Burlingame, CA) and scanned using a BIOREVO BZ-9000 fluorescence microscope (Keyence, Tokyo, Japan).
The following antibodies were used as primary antibodies: TNF Receptor 1 Polyclonal Antibody (Bioss Antibodies, Woburn, MA), TNFR2 Polyclonal Antibody (Proteintech, Chicago, IL), and Monoclonal Mouse anti-HTLV-I p19, HTLV-I p28 antibody (Chemicon, Hofheim, Germany).
Assessment of apoptosis
Apoptotic DNA ladders were detected by using an Apoptotic DNA Ladder Kit (Sigma-Aldrich). Briefly, the cultured HCT-5 cells in six-well plates were lysed and mixed with isopropanol. After the plates were washed with PBS, the pre-warmed elution buffer was applied to obtain DNA. For the DNA ladder detection, the samples were applied to a 1% agarose gel with ethidium bromide in 1× Tris, boric acid and EDTA-buffer. Apoptotic U937 cells in the kit were used as a positive control.
Annexin V staining
The evaluation of apoptosis and cell death was performed by staining with propidium iodide (PI) (MBL, Nagoya, Japan) and phycoerythrin (PE)-conjugated annexin V (MBL). After being washed with PBS, the cells were stained by PI and PE-conjugated annexin V for 15 min in ambient air. A FACS analysis was performed using a FlowSight Imaging Flow Cytometer (Merck-Millipore, Germany). PI-negative and annexin V-positive cells were defined as apoptotic cells. We calculated the percentage of apoptotic cells among all cells.
Statistical analysis
We used a Student’s t-test to determine the significance of differences in the levels of IL-6, TNF-α, sICAM-1/CD54, CCR5/RANTES, Tax, HBZ, and PVL, and the percentage of apoptotic cells. P-values <0.05 were considered significant. Statistical analyses were performed with JMP Statistical Software, ver. 11 (SAS Institute, Cary, NC).
Discussion
Our present findings demonstrated that none of the TNFi induced any significant change in the proliferation of HTLV-I-infected cells, the cytokine or chemokine production, the expression of TNFR1, the expressions of Tax or HBZ, the PVL, the level of GAG, or apoptosis, compared to the values in the HTLV-I-infected cells treated with PBS alone.
When a TNFi is necessary to treat RA patients complicated with HTLV-I infection, the occurrence and exacerbation of both ATL and HAM are a concern. Clinically, two RA cases complicated with HTLV-I infection were reported to show no increase in PVL by TNFi treatment [
21] In another report, a patient with smoldering ATL and HTLV-I-associated arthropathy that was refractory to corticosteroid, DMARDs and rituximab therapy was treated successfully with ETN [
11] These reports are valuable in that they show the safety of TNFi for HTLV-I infections, but at the same time their value is limited because they are case reports rather than studies on large numbers of patients.
The risk of TNFi-induced ATL must be considered. TNF-α was originally recognized in 1975 for its ability to lyse tumors in a variety of in vitro and mouse models, [
22] and TNFi were thus assumed to induce tumors. In contrast, a paradoxical tumor-promoting role of TNF has also become apparent [
23]. The tumorigenesis of TNFi is therefore more complex than original thought.
Tax is thought to induce ATL by promoting cellular proliferation through the enhancement of cellular survival and impairment of DNA damage-repair mechanisms [
24]. In the process of viral proliferation, HTLV-I regulates not only Tax [
25] but also HBZ [
26]. Our present results demonstrated that the levels of HBZ in HCT-5 cells were not significantly changed by treatment with various TNFi. Because Tax and HBZ were not increased following treatment with any of the TNFi in this study, TNFi may not induce ATL. In addition, PVL was not increased by the treatments in this study. Since there was no TNFi-induced increase in either Tax or HBZ, it seems reasonable that there was also no increase in PVL or GAG.
The inhibition of apoptosis is important for viral proliferation, and a functional inactivation of p53 is induced by Tax [
27]. In regard to the relationship between apoptosis and TNFi, it has been reported that the apoptosis of T lymphocytes was induced by IFX [
28] and ADA [
29]. in a human-mouse chimeric model. In addition, IFX activates the p53 gene, whereas ETN does not [
30]. In the present study, none of the TNFi appeared to enhance apoptosis, although we hypothesize that all TNFi except for ETN are capable of inducing the apoptosis of T lymphocytes, including HCT-5 cells, and inhibiting the occurrence of ATL.
Finally, we must consider the risk of HAM due to TNFi treatment. Several reports have examined the relation between inflammation and HTLV-I infection. Tax boosts the expression of the Th1 master regulator T box transcription factor and consequently promotes the production of IFN-γ [
31] HBZ has been shown to induce systemic inflammation by impairing the suppressive function of Treg cells via an interaction with Foxp3 and NFAT in HBZ-transgenic mice [
32] Clinically, patients with HAM have high circulating levels of TNFα- and IL-2-secreting HTLV-I-specific CD4+ T cells [
16] In the present study, we observed the elevated production of multiple cytokines and chemokines, including IL-6, RANTES, ICAM-1, TNF-α and IFN-γ, in the supernatant of HCT-5 cells.
In this vein, it should be noted that immunosuppressive and immunomodulatory therapies have been reported to have beneficial effects in patients with HAM, [
33] suggesting that the production of cytokines and chemokines is important in the pathophysiology of this disorder. Although we did not observe a reduction of cytokines or chemokines induced by any of the TNFi in the present study, on the basis of previous studies it appears that TNFi could have the effect of alleviating HAM.
With respect to our results showing that TNFR2 was decreased and internalized in the HCT-5 cells treated with ETN, it is apparent that among the TNFi, ETN had a different effect on TNFR2. ETN is a soluble p75 TNF receptor fusion protein, [
34] while the other TNFi are monoclonal antibodies directed against TNF-α [
35]. This difference may have affected the decreased expression of TNFR2 in the ETN-treated HCT-5 cells in the present study through a mechanism described by R.E. Kast in 2005 [
36].
This study has some limitations. Because this was an in vitro study using a cell line established from a HAM patient, the results do not necessarily reflect the clinical effects of TNFi in vivo. In addition, our findings have no validity in ATL cell lines because there are some etiological differences between HAM and ATL. For example, the expression of Tax is frequently disrupted in ATL, [
37] but HBZ is uniformly expressed in ATL [
38]. The questions of whether TNFi directly induce ATL and whether TNFi treatment presents a risk for the development of ATL in daily practice remain to be addressed. Tax has the ability to cause ATL through the induction of cell-cycle progression, DNA damage, impairment of DNA repair, and cellular transformation [
39]. The present study did not assess these factors. However, our findings showed no increase of Tax, and we thus speculate that Tax-mediated effects causing ATL did not increase. In this regard, we have begun clinical investigations in HTLV-I infected RA patients treated with biologics including TNFi.