Background
Human T-cell leukemia virus type 1 (HTLV-1) is the etiological agent of HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP), a chronic and slowly progressive neurodegenerative disease of the central nervous system (CNS) [
1,
2] and Adult T-cell leukemia/lymphoma (ATL) [
3]. Worldwide, an estimated 20 million people are infected with HTLV-1, thus placing this infection as a serious public health problem. Nevertheless, the majority of infected individuals remain asymptomatic carriers and less than 5 % develop HAM/TSP [
4‐
8].
Histopathological findings of CNS revealed that HAM/TSP affects mostly the lower and middle thoracic spinal cord, with marked degeneration of the corticospinal tracts and demyelination, accompanied by diffuse and symmetrical degeneration of the anterolateral and inner portion of the posterior columns [
9]. These findings are consistent with the HAM/TSP patient’s neurological symptoms, including spastic paraplegia of the lower extremities, loss of bladder control, and sexual dysfunction [
1,
10‐
12].
To date, the precise mechanisms by which HTLV-1 promotes these lesions remain poorly understood. Nevertheless, available data indicate that progression to HAM/TSP is characterized by presence of an exaggerated and chronic immune response [
13], accompanied by massive infiltration of mononuclear cells into the CNS, hyper secretion of pro-inflammatory cytokines and chemokines [
14] and spontaneous proliferation T lymphocytes [
13‐
24]. Although the lesions in the CNS have been primarily attributed to these infiltrating lymphocytes, growing evidence points that glial cells, particularly astrocytes, play a key role in this process [
14,
25]. Damage of astrocytes is structurally and functionally deleterious to the CNS, since these cells exert important functions, such as: maintenance of the integrity of the blood–brain barrier (BBB), neural cell survival and control of brain excitability [
26‐
28]. Histopathological findings from
post mortem tissues revealed that astrocytes from HAM/TSP lesions bear an activated phenotype and produce high amounts of pro-inflammatory cytokines, matrix metalloproteinases (MMPs) and chemokines [
14,
29,
30]. Additionally,
in vitro studies demonstrated that interactions with HTLV-1-infected lymphocytes resulted in morphological changes of astrocytes similarly to those found in
post mortem [
31,
32], being accompanied by metabolic deregulation [
33,
34].
However the participation of astrocytes in the pathophysiology of HAM/TSP remains poorly understood, particularly their role in the recruitment and trafficking of peripheral T cells into CNS. In this context, we conducted a study to investigate the morphological and functional alterations exerted by HTLV-1-infected T cell lines upon astrocytoma-derived cell lines. In particular, we used an in vitro model of T cell-astrocyte cell lines interaction to approach the potential the impact of HTLV-1-infected T cell lines in the integrity and gene expressing profile of migration-related genes of astrocytic cell lines. We also analyzed the migratory response of HTLV-1-T lymphocyte cell lines under the stimulation of astrocytic cell lines primed with supernatants derived from HTLV-1+ T cell lines. Our results indicate that under transient interactions with HTLV-1-infected T cell line cells, astrocytic cell lines undergo major morphological changes, together with modulation in the expression of a variety of cell-migration genes. In turn, such reactive astrocytic cell lines increase migratory responses of HTLV-1-infected lymphocytes, thus suggesting a role of these glial elements in the recruitment of additional T cells into CNS.
Discussion
The mechanisms underlying the pathogenesis of HAM/TSP are still poorly understood. Although previous evidence strongly suggests that astrocytes take part in this process through several mechanisms [
30‐
32], their participation in the migration of T cells into the CNS has not been studied in deep. Given the central role of astrocytes in the functional and structural homeostasis of the CNS [
28], any impairment of these cells would potentially compromise CNS functional and structural integrity. A better understanding of these processes is thus essential for designing future effective therapeutic and preventive strategies. In this context, we investigated the effects exerted by HTLV-1 on astrocytes using an
in vitro model of heterocellular interaction involving T cell lines and astrocytoma cell. Experimentally, we used a T cell line derived from a HAM/TSP patient (CIB), a T cell line originally infected by HTLV-1
in vitro (C91PL) and two astrocytoma cell lines (U251 and U87).
We first demonstrated that HTLV-1-infected T cell lines adhered significantly higher to the astrocytoma monolayers as compared to the non-infected T cell line. This finding suggests that astrocytes may represent a critical target of HTLV-1 infected lymphocytes. Previous studies also demonstrated that HTLV-1 infected cell lines adhered stronger to endothelial cell lines [
35]. We also observed that co-culture of astrocytoma cell lines with HTLV-1 infected, (but not with non-infected) T cell lines, resulted in rapid syncytium formation, seen as early as 6 h post co-culture. In addition, no syncytium was observed when a non-astrocyte cell line was used as a control, demonstrating that this effect is at least to some extent cell type specific. This result is in agreement with previous observations, in which rapid syncytium formation was detected when HTLV-1 infected lymphocyte cell lines were co-cultured with astrocyte cell lines [
32]. These findings allow us to hypothesize that HTLV-1 infected cells have an increased ability to cross the blood–brain barrier into the CNS, where they rapidly and strongly bind to the astrocytoma, leading to severe functional and morphological damage.
In a second set of experiments, we observed several and extensive morphological changes and cytopathic effect in 5-day culture of astrocytoma cell cultures that were transiently exposed to HTLV-1-infected T cell lines, but not to non-infected counterparts. The cytopathic effect was characterized by loss of cell-cell contact, cell shrinkage, accumulation of lysosomal vesicles and cell death. Notably, similar morphological alterations have been observed in endothelial cells after transient interaction with the MT-2 HTLV-1
+ T cell line [
35]. These alterations were rather cell type specific since co-culture of HTLV-1 infected T cell lines with the Hela cell line, a non-astrocytic lineage, did not result in any cytopathic effect (see Fig.
4). We also detected an enhancement of astrocytoma apoptosis secondary to the interaction with HTLV-1-infected cells as shown in Fig.
5, a finding corroborated by similar results previously reported [
36‐
38]. Several mechanisms has been proposed to explain the apoptotic effect of HTLV-1 infected T cells on astrocytes, such as the direct effect of secreted Tax-1 protein in inducing apoptotic pathways in the target cell through the down regulation in the cellular expression of
bcl-2 [
36,
37] and susceptibility of astrocytes to apoptosis secondary to increase in the production of TNF-α [
38].
Of note, we observed the presence of HTLV-1 viral particles in the cytoplasm of astrocytoma cells after transient interaction with HTLV-1
+ T cells line, but not with the non-infected T cell line. Infection of astrocytes with HTLV-1 has been reported in previous studies [
39‐
41], and suggests that direct infection may also be involved in the functional and morphological damage of human astrocytes in HAM/TSP.
We also investigated the role played by the soluble factors secreted by HTLV-1
+ T cell lines on cultured astrocyte cell lines. We used qPCR arrays to quantitatively evaluate the expression levels of a large numbers of cell migration-related genes, including cytokines/chemokines and extracellular matrix/cell adhesion proteins. A short-term (1 h) exposure of astrocytoma cell lines to fractionated HTLV-1-infected T cell lines-derived supernatants, resulted in statistically significant increase in the mRNA of TNF-α, various chemokines, as well as VCAM-1 and MMP-8, as compared with supernatants derived from the non-infected T cell line, as shown in Fig.
6. These results strongly corroborate the findings reported by Ando and cols. (2013), who found an increased CXCL10 production by astrocytes that were transiently co-cultured with CIB cells [
42] and also with results reported by Tomoo and cols. (2013), who found that level of CXCL10 in the CSF correlated with progression to HAM/TSP [
43], suggesting that CXCL10 might play a critical role in stimulating migration of HTLV-1 infected T cells in the transwell chamber and ultimately reflecting what happens
in vivo, when T-cells migrate into the CNS. Our results are also in line with previous histopathological findings from post-mortem tissues, in which it has been found that astrocytes from HAM/TSP lesions exhibit higher contents of proinflammatory cytokines, MMP and chemokines [
14,
29,
30].
Moreover, our data corroborate those described by Akaoka (2011) and Szymocha (2000), who demonstrated that infiltrating HTLV-1-infected T lymphocytes in the CNS continuously induce significant changes in astrocytes, not only through cell-cell interaction [
44], but also through soluble elements, particularly the 42 kDa Tax-1 viral protein [
31]. Tax is a transactivator/oncoprotein thought to play an important role in the course of HAM/TSP, and soluble Tax1 has been shown to induce functional changes in human astrocytes [
36] and to modulate gene expression on target cells [
15].
Conjointly, these findings strongly point to an important paracrine effect of HTLV-1 infected lymphocytes on astrocytes and highlight that the functional impairment of astrocytes occurring in HAM/TSP, would be much more complex than ever described.
Our findings suggest that the functional impairment of astrocytes caused by interactions with HTLV-1-infected T cells contributes to the perpetuation and amplification of the CNS damage. Considering that we applied a 30–50 kDa supernatant fraction to stimulate the astrocytoma cells, it is conceivable that Tax1 protein is involved in these effects. Yet, additional studies with blocking antibodies are required to confirm this hypothesis.
Since astrocytes primed with HTVL-1-infectd T cells expressed higher levels of various cell adhesion and migration-related mRNA molecules, it seemed plausible to hypothesize that supernatants from such primary cultures could enhance T cell migration. Actually, this was the case and supernatants from HTLV-1 primed astrocytoma cell lines were able increase the migration of HTLV-1 infected T cell lines. Conceptually, these data suggest that activated astrocytes would enhance the recruitment of additional T cells into the CNS. Since main focus of this study was to understand the impact of a HAM/TSP derived T cell line (CIB), the in vitro derived HTLV-1 infected T cell line (C91PL) was used as a control only in the first set of experiments when we observed that the effects were similar when using both cell lines.
Although we have not performed immune staining to confirm the presence of HTLV-1 inside the cells, we believe that the viral particle detection by electron microscopy is a convincing evidence of astrocyte infection. Similarly, we also think that the enhanced T cell migration after being primed by supernatants from infected astrocytes suggests an increased concentration of chemotactic factors, as indicated by the elevated expression of the corresponding mRNAs by HTLV-1-infected astrocytes.
Of remark, our results are of utmost importance, since most of the previous work that investigated HTLV-1-induced changes on astrocytes used
in vitro derived HTLV-1-infected T cell lines, such as MT-2 or C91PL [
29,
31‐
34,
45]. In our study we used a HAM/TSP patient-derived lymphocyte T cell line. Indeed, to our knowledge, only Ando and co-workers conducted functional assays on astrocytes using same cell line [
42].
Methods
Cell lines
The human malignant astrocytoma cell lines U251 and U87, both derived from malignant gliobastoma multiforme (gently provided by Dr Maria do Socorro Pombo de Oliveira, National Cancer Institute, Rio de Janeiro) [
46], were grown in high glucose (4500 mg/l) Dulbecco Modified Eagle Medium (DMEM; Gibco-BRL, Grand Island, NY) supplemented with 10 % fetal calf serum (FCS) (Cultilab, Campinas, Brasil), 2 mM L-glutamine (Gibco, Scotland, UK), 100 U/ml penicillin and 100 g/ml streptomycin (Gibco, Scotland, UK) at 37 °C in a humidified incubator with 5 % CO
2. Hela cells, an epithelial cell line derived from cervical cancer cells [
47], were used as a negative control in some experiments. The HTLV-1-infected T cell line CIB (gently provided by Dr. Olivier Hermine, Necker Hospital, Paris) was derived from a HAM/TSP patient [
48]. The HTLV-1-infected T cell line C91PL was derived from human cord blood T lymphocytes immortalized by HTLV-I infection [
49] (used in several assays as a positive control and gently provided by Dr. Olivier Hermine, Necker Hospital, Paris). The non-infected T-cell line CEM, derived from a patient with acute lymphoblastic leukemia, [
50] (used as a negative control and gently provided by Dr Maria do Socorro Pombo de Oliveira, National Cancer Institute, Rio de Janeiro) was cultured in RPMI 1640 supplemented with 10 % fetal calf serum, 100 U/ml penicillin and 100 g/ml streptomycin at 37 °C, 5 % CO
2. Cultures of the HTLV-1-infected T cell lines were further supplemented with 100 U/ml of interleukin 2 (IL-2) (Roche, France).
Heterocellular adhesion assay
For adhesion assays, astrocytoma and Hela cells were first plated in 25 cm
2 flasks until sub-confluence. Then, HTLV-1-infected or non-infected T lymphocytes were added to the astrocyte monolayer at a ratio astrocyte: lymphocyte of 1:4 and allowed to adhere for 30 min at 37 °C, 5 % CO
2 in RPMI 1640 supplemented with 2 mM L-glutamine and antibiotics in absence of FCS and IL-2. Supernatants containing floating lymphocytes were then discarded and the contents of each flask (astrocytes + adhered lymphocytes) were fixed in absolute ethanol, stained with Giemsa and counted to determine the adhesion index (AI), as previously described [
51] and validated for this type of analysis [
52], using the following formula:
$$ \mathrm{AI} = \frac{\mathrm{Astrocytes}\ \mathrm{with}\ \mathrm{adhered}\ \mathrm{lymphocytes}}{\mathrm{Total}\ \mathrm{number}\ \mathrm{o}\mathrm{f}\ \mathrm{astrocytes}}\times \frac{\mathrm{Lymphocytes}\ \mathrm{adhered}\ \mathrm{t}\mathrm{o}\ \mathrm{astrocytes}}{\mathrm{Total}\ \mathrm{number}\ \mathrm{o}\mathrm{f}\ \mathrm{astrocytes}}\times 100 $$
To assess syncytium formation, astrocytoma cells were initially grown in 25 cm2 flasks until sub-confluence. HTLV-1-infected T lymphocytes (CIB or C91PL) or the non-infected T cell line (CEM) were added to the monolayer of astrocytoma at astrocyte/lymphocyte ratio of 1:4 and left to adhere for up to 20 h in culture conditions similar to those described above. Supernatant was then discarded, washed briefly with PBS, fixed in absolute ethanol, stained with Giemsa and observed by light microscopy. Similar assays were performed using a non-astrocytic cell line (Hela), to assess whether changes were specific for astrocytes.
Morphological assessment of cultured astrocytoma cells
Astrocytoma cells were initially grown in 25 cm
2 flasks until sub-confluence. Transient co-cultures of astrocytes with either HTLV-1-infected T lymphocytes (CIB or C91PL) or the non-infected T cell line (CEM) were performed by replacing the medium in the astrocytoma cultures with those from cultures of HTLV-1-infected or non-infected T cell lines in the absence of IL-2 at 37 °C, 5 % CO
2, at astrocyte/lymphocyte ratio of 1:4. After 3 h, lymphocytes were completely removed from the cultures by vigorous washing using cold PBS. Duration of co-culture was set at 3 h, since the study conducted by Mor-Vaknin et al., [
32], demonstrated that syncytium was formed as early as 4 h post co-culture and in our experiments syncytium was formed as early as 6 h post co-culture.
Hela cells were used as controls for nonspecific interactions between astrocytes and HTLV-1-infected cells. For Mock controls, astrocytes were culture alone with RPMI.
After lymphocyte removal by vigorous washings with cold PBS, astrocytoma and Hela cells were kept in culture for up to 5 days, replenishing fresh medium every two days. At days 5 post co-culture, astrocytes were processed for optical or transmission electron microscopy, as previously described [
53], using current 2.5 % glutaraldehyde 1 % paraformaldehyde fixation, followed by uranyl acetate plus lead citrate staining, with the analysis of ultrathin sections being done under a transmission electron microscope (Jeol JEM-1011) on the Rudolf Barth Electron Microscopy Platform at the Oswaldo Cruz Institute (Fiocruz, Rio de Janeiro).
Apoptosis of astrocytoma cells induced by HTLV-1-infected lymphocytes
The ability of HTLV-1 or non-infected T cells to induce apoptosis on the astrocytoma cells was measured by cytofluorometry using FITC-coupled Annexin V Apoptosis Detection Kit I (Becton-Dickinson/Pharmingen, San Diego). After transient co-cultures during 3 h, followed by lymphocyte removal by vigorous washings with cold PBS, astrocytes were kept in culture for additional 2 h, trypsinized and stained with FITC-Annexin V and Propidium Iodide according to the manufacturer’s instructions. Events were acquired using a FACSCalibur Flow Cytometer (Becton-Dickinson, San Diego, USA).
Cell adhesion and cell migration-related gene expression of astrocytoma cells primed with supernatants from HTLV-1-infected T lymphocytes
We further evaluated the effect exerted by HTLV-1-infected cells on the expression of cell adhesion and cell migration-related genes in astrocytoma cells. Astrocytes were primed using supernatants from HTLV-1-infected T cell cultures, using as control counterpart supernatants from non-infected T cell cultures. For this purpose, lymphocytes were grown in 75 cm2 culture flasks at an initial concentration of 1x107 cells/mL during 24hs. Supernatants were then collected and concentrated using the centripep YM-50 and YM-30 filters (Millipore, corporation, County Cork, Ireland), according to the manufacturer’s instructions. The fraction was added to the astrocyte monolayer for 1 h at 37 °C and then, cells were then trypsinized and washed. As a mock control, growing astrocytes were treated with RPMI alone.
Total RNA was extracted using RNeasy Micro Kit (Qiagen, Hilden, Germany), according to the manufacturers instructions, including DNAse treatment. Total RNA was quantified by spectrophotometry using Nanodrop 2000 (Thermo Scientific, USA). RNA integrity was assessed for the presence of ribosomal RNA 28S and 18S, using denaturating agarose gel electrophoresis as previously described [
54]. RNA was used for the RT-PCR if the ratio A
260/A
230 ≈ 2.0.
Gene expressing profiles for cytokines/chemokines and extracellular matrix (ECM) proteins/adhesion molecules were ascertained by quantitative PCR. cDNAs were prepared from 1 μg RNA of each sample using “RT2 First Strand” (SABioscience, Maryland, USA), according to the manufacturers instructions. Then, RT2 SYBR Green/ROX PCR Master Mix (SABioscience) and’ nuclease-free water were added to the cDNA. Twenty-five μL of this mixture were added to 96 microplates containing primers for each of the 84 target genes for ECM proteins/adhesion molecules using corresponding qPCR arrays (SABioscience/Qiagen, USA), as seen in supplementary data in S1 Table and S2 Table. Amplification signals were captured using the ABI Prism 7000 PCR device (Applied Biosystems, Foster City, CA, USA). Melting curve analysis was performed for each sample at the end of the PCR.
Migration profile of HTLV-1-T lymphocytes cell lines under the stimulation of supernatants from primed astrocytoma cell lines
Cell migration studies were conducted in
transwell chambers (Corning Costar, Cambridge, MA, USA), as described elsewhere [
55], using the 8-μm pore size inserts coated with 10 μg/ml of BSA, for 1 h at 37 °C. Supernatants from astrocytoma cultures previously primed with HTLV-1-T cell lines
(3 h co-culture
) were harvested and added into the lower chamber, whereas 10
6 CIB cells/100 μl (of RPMI/1 % BSA) were added into the upper chambers. After 4 h incubation at 37 °C in 5 % CO
2, we counted the cells that migrated into the lower chambers using trypan blue cell viability analysis.
Statistical analysis
All statistical analyses were performed using the GraphPad Prism 5.01 software package. Quantitative data were expressed as the mean ± standard error. The Student’s t-test and the One Way ANOVA test were used to compare the differences among groups. Differences were considered statistically significant when the p values were ≤ 0.05.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
ESG designed the study, collected and analyzed data and wrote the manuscript; LLL designed the study, performed experiments and wrote the manuscript; MRA designed and analyzed qPCR assays; SCR processed samples for and analyzed electron microscopy data; DCBH designed experiments, analyzed data and wrote the manuscript; SDSB and WS designed study, analyzed data and wrote the manuscript. Authors declare no financial or personal interest. All authors have read and approved the final version of the manuscript.