JOURNAL TRANSCRIPT
Original Paper Intervirology 2015;58:106–114 DOI: 10.1159/000371766
Received: June 17, 2014 Accepted after revision: December 26, 2014 Published online: March 27, 2015
High IFN-γ/IL-10 Expression Ratio and Increased Frequency of Persistent Human T-Cell Lymphotropic Virus Type 1-Infected Clones Are Associated with Human T-Cell Lymphotropic Virus Type 1-Associated Myelopathy/Tropical Spastic Paraparesis Development Otávio M. Espíndola a Luã C. Oliveira a Priscilla M.S. Ferreira a Ana Claudia C.B. Leite b Marco Antonio S.D. Lima b, c Maria José Andrada-Serpa a a Laboratory for Research on Viral Pathogenesis, and b Laboratory for Clinical Research on Neuroinfections, Evandro Chagas National Institute of Infectious Diseases – FIOCRUZ, and c Neurology Department, The Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
Key Words Human T-cell lymphotropic virus type 1 · Human T-cell lymphotropic virus type 1-associated myelopathy/tropical spastic paraparesis · IFN-γ · IL-10 · Clonality · Biological markers
Abstract Background/Aims: Human T-cell lymphotropic virus type 1 (HTLV-1) is a retrovirus that causes a persistent infection, and only 0.5–5% of infected individuals will develop HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/ TSP). Therefore, we investigated parameters to discriminate HTLV-1 asymptomatic carriers (ACs) with an increased chance to develop HAM/TSP. Methods: We evaluated integration patterns of HTLV-1 provirus, the relative expression of HTLV-1 tax and HBZ mRNAs and of IFN-γ and IL-10 mRNAs, in addition to proviral load (PVL) levels. Results: HAM/TSP patients presented a higher number of large persistent HTLV-1-carrying clones compared to ACs, and the expression of the HTLV-1 tax and HBZ genes by infected cells was
© 2015 S. Karger AG, Basel 0300–5526/15/0582–0106$39.50/0 E-Mail
[email protected] www.karger.com/int
detected at low levels and correlated positively with PVL. In addition, HAM/TSP patients and ACs with high PVL expressed higher levels of IFN-γ mRNA in comparison to IL-10, while ACs with low PVL presented an equilibrate IFN-γ/IL-10 ratio. Conclusions: The presence of large persistent HTLV1-infected clones in association with viral gene expression, even at small levels, could stimulate the intense inflammatory response in HTLV-1-infected individuals. This was supported by a high ratio of IFN-γ/IL-10 relative expression in HAM/TSP patients and ACs with high PVL, indicating that these parameters could aid the identification of ACs with a high risk to develop HAM/TSP. © 2015 S. Karger AG, Basel
Introduction
Human T-cell lymphotropic virus type 1 (HTLV-1) is a retrovirus that causes a persistent infection, and 0.5–5% of infected individuals will develop the associated diseases: adult T-cell leukemia/lymphoma [1] and HTLV-1-associOtávio de Melo Espíndola, PhD Laboratório de Pesquisa em Patogenia Viral Instituto Nacional de Infectologia Evandro Chagas – INI-FIOCRUZ Avenida Brasil 4365, Manguinhos – Rio de Janeiro, RJ 21040-900 (Brazil) E-Mail otavio.espindola @ ini.fiocruz.br
ated myelopathy/tropical spastic paraparesis (HAM/TSP) [2]. In contrast, most HTLV-1-infected individuals will remain asymptomatic carriers (ACs) throughout life [2]. HAM/TSP is a demyelinating disease of the spinal cord that initiates as an inflammatory reaction after blood-brain barrier disruption. The immune response is mediated by cytotoxic T lymphocytes (CTLs) against HTLV-1-infected T cells in the central nervous system, resulting in exacerbated production of Th1 cytokines by activated cells, such as IFN-γ and TNF-α, which induces the death of central nervous system cells [2, 3]. However, the events or mechanisms that disrupt the asymptomatic condition in HTLV-1 infection and triggers HAM/TSP development are still unknown. It has been shown that high HTLV-1 proviral load (PVL) in peripheral blood is a risk factor for neurological involvement [4]. It has been demonstrated that HTLV-1-infected T-cell clones can persist for many years in patients [5, 6], and infected individuals generally carry 500–5,000 distinct infected T-cell clones defined by proviral integration sites [7]. Indeed, clonal expansion of infected cells is responsible for maintaining HTLV-1 PVL levels [8]. In contrast, infected individuals show a chronic and intense response of activated HTLV-1-specific CTLs, which is elicited by persistent expression of viral antigens [9, 10]. Therefore, expression of HTLV-1 proteins, notably Tax protein, influences both the proliferation of infected cells as their susceptibility to lysis by CTLs [8]. HTLV-1 Tax transactivator protein interacts with different transcription factors, including CREB/ATF, SRF, NFAT, and NF-κB [11, 12], thereby regulating the expression of viral genes and the function of different cellular processes, such as cell proliferation, differentiation and death, cytoskeleton reorganization, and DNA repair [13– 16]. HTLV-1 provirus also presents an antisense gene coding for a negative regulator of Tax functions, denominated HBZ (HTLV-1 basic leucine zipper factor) [17]. HBZ partially represses Tax activity through the competition for association with transcription factors such as CREB [18], CREB-2 [17], and AP-1 [19, 20], and it is important in survival of HTLV-1-transformed cells [21] and in enhancement of in vivo infectivity [22]. Usually, a long interval elapses from the time of infection to disease onset in HTLV-1 carriers, ranging from 20 to 30 years [2]. The development of markers to assist the early identification of HTLV-1 ACs at a higher risk to develop HAM/TSP would allow better clinical monitoring and implementation of therapeutic interventions to delay disease progression. Therefore, we examined the profiles of IFN-γ and IL-10 mRNA expression and parameters of
HTLV-1 infection, such as PVL levels, the relative expression of viral tax and HBZ genes, and the pattern of provirus integration sites, in ACs and HAM/TSP patients, to determine biological markers for development of HAM/TSP.
High IFN-γ/IL-10 Ratio and HAM/TSP
Intervirology 2015;58:106–114 DOI: 10.1159/000371766
Materials and Methods Patients and Samples The study was conducted at the Instituto Nacional de Infectologia Evandro Chagas (INI/FIOCRUZ), with approval of the Research Ethics Committee of the INI/FIOCRUZ (CAAE 0013.0.009.000-07) from 2009 to 2013. Forty-six HTLV-1-infected individuals (21 ACs and 25 HAM/TSP patients) and 12 healthy donors (HDs) were enrolled. After informed consent, peripheral blood was collected in tubes with EDTA or heparin. PBMCs were obtained by centrifugation of heparinized blood on Histopaque 1077 (Sigma-Aldrich), and cells were counted in automatic cell counter (XT1800i, Symex). Cell suspensions were adjusted to 106 PBMCs/ml in RPMI-1640 supplemented with 10% FBS, and 107 cells were incubated without stimulation at 37° for 20 h in 5% CO2 atmosphere. Fresh and cultured PBMCs were stored at –20° in Trizol reagent (Invitrogen) until use. DNA Extraction and HTLV-1 PVL DNA from EDTA peripheral blood was obtained using the Puregene Blood Core kit (Qiagen), following the manufacturer’s instructions. HTLV-1 PVL was determined by real-time PCR using TaqMan probes in Smart Cycler II equipment (Cepheid), as previously described [23]. The gene for β-globin was used as a reference, and infected cells were detected by amplification of the HTLV-1 tax gene. Standard curves to determine the total number of leukocytes and the number of infected cells in reactions were constructed, respectively, with twofold dilutions of human genomic DNA (Promega) and DNA from the HTLV-1-infected TARL-2 cell line. PVL as the frequency of HTLV-1-infected cells in peripheral blood leukocytes was calculated as [(copy number of tax)/(copy number of β-globin/2)] ×100. Amplification of HTLV-1 Integration Sites by Inverse Long-PCR Integration sites of HTLV-1 provirus in host-cell DNA was evaluated by inverse long-PCR (IL-PCR), as described by Etoh et al. [5]. Initially, 1 μg of DNA from peripheral blood was digested with 5U EcoRI (New England Biolabs) at 37° for 2 h. Afterwards, DNA fragments (300 ng) were autocircularized overnight at 4° with 2 U T4 DNA ligase, in duplicate (New England Biolabs). After purification, circularized DNA was incubated overnight at 37° with 10 U MluI (New England Biolabs). After further purification, the samples were submitted to PCR with a GeneAmp XL PCR kit (Applied Biosystems), using 40 pmol of U3 and U5 LTR-directed primers, under the following amplification cycle: denaturation at 94° for 1 min, 37 cycles of denaturation at 94° for 1 min and annealing/ extension at 68° for 7 min, and final extension at 72° for 10 min. Amplicons were electrophoresed in a 1% agarose gel in 1× TBE buffer at 80 V for 3 h and DNA was stained with GelRed (Biotium). Oligoclonal integration was considered when 2–10 well-defined bands were observed and polyclonal integration was defined when only smear background or more than 10 bands were present.
107
P
1
1’
2
AC
HAM/TSP 2 2’
1
1’
2
2’
3
3’
P
1
1’
2
2’
p = 0.046
10 Persistent HTLV-1-infected cell clones (n)
HAM/TSP 1
8 6 4 2 0
a
Fig. 1. Persistence of HTLV-1-infected cell clones. a The pattern
of HTLV-1 provirus integration was evaluated in peripheral blood leukocytes from HTLV-1 ACs and HAM/TSP patients. DNA samples obtained over an average interval of 36.4 months (95% CI: 23.1–49.7) were submitted to IL-PCR [23]. Bands visualized on agarose gel after electrophoresis represent large HTLV-1-infected clonal populations when present in both duplicates, while amplification background represent proviral integration sites from mi-
RNA Extraction and Quantitative Reverse Transcription PCR for Gene Expression Total RNA was extracted from fresh or cultured PBMCs using Trizol reagent, according to manufacturer’s instructions, and was readily used for cDNA synthesis with SuperScript III (Invitrogen). Reverse transcription was performed with 2 μg of total RNA and random hexamers (Invitrogen) at 50° for 50 min, in a final volume of 20 μl. Relative expression of IFN-γ and IL-10 mRNAs and of viral tax and HBZ mRNAs was performed by quantitative PCR in Rotor-Gene Q equipment (Qiagen) using the Rotor Gene Probe PCR kit (Qiagen), with 100 ng of cDNA in a final volume of 25 μl, and GAPDH expression as a reference. The expression of GAPDH, IFN-γ, and IL-10 mRNAs was evaluated with predesigned TaqMan gene expression assays (Applied Biosystems), HTLV-1 tax cDNAs were amplified as described for PVL quantification, and HBZ expression was detected as previously described [24]. Duplicate reactions were submitted to: activation/denaturation at 95° for 3 min, 50 cycles of denaturation at 95° for 5 s, and annealing/extension at 60° for 15 s; fluorescence was detected at the end of each cycle. Relative gene expression was evaluated by the 2−ΔΔCt method [25]. Statistical Analysis Graphs and statistical analysis were performed using GraphPad Prism 5. After defining the normal distribution of data by Kolmogorov-Smirnov test, parametric data were tested by Pearson correlation, ANOVA, and Student’s t test, with Welch’s correction when indicated. Nonparametric data were evaluated by Kruskal-Wallis and Mann-Whitney tests and Spearman correlation, and association between qualitative variables was tested by χ2 or Fisher’s exact test. Results with p < 0.05 were considered statistically significant.
108
AC HAM/TSP Neurological condition
b
Intervirology 2015;58:106–114 DOI: 10.1159/000371766
nority clones. Results are representative of 6 HAM/TSP patients and 5 ACs. Arrowheads indicate large persistent HTLV-1 provirus-carrying clones. P, 500-bp DNA ladder. b The frequency of these large persistent HTLV-1-infected clones was determined by visualization of the same band size in consecutive samples in at least one of duplicates. Differences in mean number of persistent clones between ACs and HAM/TSP patients were calculated by Student’s t test.
Results
HTLV-1 Provirus Integration Pattern and Persistence of Infected Cell Clones DNA samples from HAM/TSP patients (n = 6) and ACs (n = 5) with similar high PVL levels (HAM/TSP: 20.60 ± 15.31% and ACs: 12.35 ± 1.25%; p = 0.246, Student’s t test with Welch’s correction) obtained in two or three distinct periods with a mean interval of 36.4 months (95% CI: 23.1–49.7) between the first and last sample were used to evaluate HTLV-1 provirus integration by ILPCR. Both HAM/TSP patients and ACs showed predominantly stochastic patterns of proviral integration. In ILPCR, the amplification background in agarose gel electrophoresis represents minor clones, while clear bands are related to large HTLV-1-infected clonal populations (fig. 1a). The integration profile was oligoclonal in all 6 HAM/TSP patients and in 3 of 5 ACs. Therefore, no association was observed between the integration pattern of HTLV-1 and the clinical condition of the evaluated subjects (p = 0.189, Fisher’s exact test). On the other hand, chronic HAM/TSP patients (mean disease duration of 14 ± 8.1 years) presented higher frequency of large persistent HTLV-1-infected clones (5.82 ± 1.71 clones) compared to ACs (3.25 ± 1.72 clones; p = 0.046; fig. 1b). Espíndola et al.
p < 0.001
a
0.8 0.6 0.4 0.2 0 AC
HAM/TSP
b
105 10 4 103 102 101 10 0
AC
100
50
0 HD
AC
c
103 102 101 10 0 AC
HAM/TSP
p = 0.161 p < 0.001
p = 0.005 p = 0.717
10 8 Relative expression ©Ct (IFN-Dž-GAPDH) mRNA
Relative expression ©Ct (IL-10-GAPDH) mRNA
d
HAM/TSP
10 4
p = 0.001
p = 0.006 150
p = 0.474 105 Relative expression ©Ct (HBZ-GAPDH) mRNA
Relative expression ©Ct (tax-GAPDH) mRNA
Tax mRNA expression (tax/GAPDH ratio × 10,000)
1.0
p = 0.115
10 6
HAM/TSP
e
107 10 6 105 10 4 103 102 101 10 0
HD
Fig. 2. Relative expression of HTLV-1 tax, and IFN-γ and IL-10 mRNAs. a Expression of tax mRNAs was evaluated in fresh
AC
HAM/TSP
PBMCs from HTLV-1 ACs (n = 21) and HAM/TSP patients (n = 25), and values were arbitrarily multiplied by 10,000. b The expression of tax mRNAs was also further evaluated in PBMCs from ACs (n = 18) and HAM/TSP patients (n = 24) after 20 h of incubation without stimulation. c The relative expression of HBZ mRNAs was performed in PBMCs from ACs (n = 15) and HAM/TSP patients
(n = 20), while the expression of IL-10 (d) and IFN-γ (e) mRNAs was evaluated in PBMCs from ACs (n = 16), HAM/TSP patients (n = 18), and HDs (n = 7) after 20 h of incubation without stimulation. GAPDH expression rates were used to normalize values. In b–e, after conversion of log(2) data to linear scale, the lowest value obtained in each analysis was arbitrated as equal to 1. Statistical analysis was performed with Mann-Whitney test, and results with p < 0.05 were considered statistically significant.
Expression of HTLV-1 tax and HBZ mRNAs in PBMCs from Infected Individuals In the same way that chronic HAM/TSP patients exhibited higher PVL median than ACs [6.28% (95% CI: 7.24–13.11) and 4.31% (95% CI: 3.38–7.27), respectively, Mann-Whitney, p = 0.002] and higher frequency of large persistent HTLV-1-infected clonal populations (fig. 1b), these individuals also showed higher relative expression of tax mRNAs by fresh PBMCs in comparison to ACs (p < 0.001; fig. 2a). However, the HTLV-1
tax gene was weakly expressed ex vivo. Expression of tax mRNAs was 10,000–100,000 times lower than basal cellular transcription levels defined by GAPDH expression. The expression of tax mRNAs was upregulated after overnight incubation of PBMCs without stimulation. However, statistically significant differences between ACs and HAM/TSP patients disappeared (fig. 2b). On the other side, HBZ mRNA expression was evaluated only in PBMC incubated for 20 h without stimulation, because it is expressed at low levels. Although HBZ
High IFN-γ/IL-10 Ratio and HAM/TSP
Intervirology 2015;58:106–114 DOI: 10.1159/000371766
109
p < 0.001
Table 1. Relative expression of HBZ mRNAs in ACs and HAM/ TSP patients
p < 0.001
–©Ct (IFN-Dž – IL-10) mRNA PBMC 20 h (log2)
10
–10
HD (n = 7)
AC HAM/TSP (n = 16) (n = 18)
Fig. 3. Difference between IFN-γ and IL-10 mRNAs expression levels in PBMCs from HTLV-1 ACs, HAM/TSP patients, and HDs. Variation in relative expression of these genes was calculated as –ΔCt = (mean CtIFN-γ – mean CtIL-10). Statistical significance was determined by a Mann-Whitney test.
mRNA expression in PBMCs were 3.01 times higher in HAM/TSP patients than ACs (table 1), this difference was not significant (p = 0.474, Mann-Whitney; fig. 2c), even when it was corrected with PVL levels (p = 0.594, Mann-Whitney). IL-10 and IFN-γ mRNA Expression in PBMCs from HTLV-1-Infected Individuals Differences in relative expression of IL-10 mRNAs were not observed between ACs and HAM/TSP patients (p = 0.717; fig. 2d). However, both ACs and HAM/TSP patients presented reduced IL-10 mRNA expression in comparison to uninfected individuals (p = 0.005 and p = 0.006, respectively). On the other hand, HAM/TSP patients showed increased levels of IFN-γ expression compared to ACs (p < 0.001) and HDs (p = 0.001; fig. 2e), but no differences were observed between ACs and HDs (p = 0.161). The ratio between IFN-γ and IL-10 mRNA relative expression levels was also investigated by calculating the –ΔCt (mean CtIFN-γ – mean CtIL-10) (fig. 3), taking that commercial realtime PCR for these genes presented similar efficiency rates. On average, HAM/TSP patients expressed more IFN-γ than IL-10 (–ΔCt = 5.10 or 34.30 times), while ACs presented balanced levels of expression of both cytokines (−ΔCt = 0.27). In contrast, HDs expressed more IL-10 instead of IFN-γ (–ΔCt = –6.61 or 97.68 times). However, differences were statistically significant only in HAM/TSP patients (fig. 3). Intervirology 2015;58:106–114 DOI: 10.1159/000371766
32.48 ± 2.75 18.08 ± 0.75 14.45 ± 2.72 –1.59 ± 2.72 3.01 (2.19 – 19.84)
Relative expression of HBZ mRNAs in PBMCs incubated for 20 h without stimulation, from HTLV-1 ACs and HAM/TSP patients was determined by the 2–ΔΔCt method [25]. Values are presented as means ± SD. Ct = Cycle threshold. a Mean Ct HBZ – mean CtGAPDH. b Mean ΔCt HAM/TSP − mean ΔCtAC.
Table 2. Correlation between tax, HBZ, IL-10, and IFN-γ gene expression and PVL
Genes
110
HAM/TSP
CtHBZ 34.03 ± 4.47 CtGAPDH 18.18 ± 0.81 16.04 ± 4.63 ΔCta ΔΔCtb 0.00 ± 4.63 Normalized expression (2−ΔΔCt) 1.00 (0.04 – 24.76)
0
–20
AC
p = 0.181
Samples, n
Correlation coefficient, Ra
p
21 25 46
0.817 0.433 0.730