ORIGINAL RESEARCH published: 09 January 2019 doi: 10.3389/fimmu.2018.03094 Study of CD27 and CCR4 Markers on Specific CD4+ T-Cells as Immune Tools for Active and Latent Tuberculosis Management Irene Latorre 1,2,3*, Marco A. Fernández-Sanmartín 4 , Beatriz Muriel-Moreno 1,2,3 , Raquel Villar-Hernández 1,2,3 , Sergi Vila 1 , Maria L. De Souza-Galvão 5 , Zoran Stojanovic 6 , María Á. Jiménez-Fuentes 5 , Carmen Centeno 6 , Juan Ruiz-Manzano 2,6 , Joan-Pau Millet 7,8 , Israel Molina-Pinargote 7 , Yoel D. González-Díaz 7 , Alicia Lacoma 1,2,3 , Edited by: Lydia Luque-Chacón 9 , Josefina Sabriá 9 , Cristina Prat 1,2,3† and Jose Domínguez 1,2,3† Jesús Gonzalo-Asensio, 1 Servei de Microbiologia, Hospital Universitari Germans Trias i Pujol, Institut d’Investigació Germans Trias i Pujol, Barcelona, Universidad de Zaragoza, Spain Spain, 2 CIBER Enfermedades Respiratorias, CIBERES, Instituto de Salud Carlos III, Madrid, Spain, 3 Departament de Reviewed by: Genètica i Microbiologia, Universitat Autònoma de Barcelona, Barcelona, Spain, 4 Plataforma de Citometría, Institut Damien Portevin, d’Investigació Germans Trias i Pujol, Barcelona, Spain, 5 Unitat de Tuberculosi de Drassanes, Hospital Universitari Vall Swiss Tropical and Public Health d’Hebron, Barcelona Spain, 6 Servei de Pneumologia, Hospital Universitari Germans Trias i Pujol, Barcelona, Spain, 7 Serveis Institute, Switzerland Clínics, Unitat Clínica de Tractament Directament Observat de la Tuberculosi, Barcelona, Spain, 8 CIBER de Epidemiología y Paul Fisch, Salud Pública, CIBERESP, Instituto de Salud Carlos III, Madrid, Spain, 9 Servei de Pneumologia, Hospital Sant Joan Despí University Hospital Freiburg, Germany Moises Broggi, Sant Joan Despí, Barcelona, Spain *Correspondence: Irene Latorre ilatorre@igtp.cat The immunological characterization of different cell markers has opened the possibility of considering them as immune tools for tuberculosis (TB) management, as they could † These authors have contributed equally to this work and as co-senior correlate with TB latency/disease status and outcome. CD4+ T-cells producing IFN-γ+ authorship with a low expression of CD27 have been described as an active TB marker. In addition, there are unknown homing receptors related to TB, such as CCR4, which might be useful Specialty section: for understanding TB pathogenesis. The aim of our study is focused on the assessment This article was submitted to Microbial Immunology, of several T-cell subsets to understand immune-mechanisms in TB. This phenotypic a section of the journal immune characterization is based on the study of the specific immune responses of Frontiers in Immunology T-cells expressing CD27 and/or CCR4 homing markers. Subjects enrolled in the study Received: 08 July 2018 Accepted: 13 December 2018 were: (i) 22 adult patients with active TB, and (ii) 26 individuals with latent TB infection Published: 09 January 2019 (LTBI). Blood samples were drawn from each patient. The expression of CD27 and/or Citation: CCR4 markers were analyzed within CD4+ T-cells producing: (i) IFN-γ+ , (ii) TNF-α+ , Latorre I, Fernández-Sanmartín MA, (iii) TNF-α+ IFN-γ+ , and (iv) IFN-γ+ and/or TNF-α+ . The percentage of CD27− within Muriel-Moreno B, Villar-Hernández R, Vila S, Souza-Galvão MLD, all CD4+ T-cell populations analyzed was significantly higher on active TB compared Stojanovic Z, Jiménez-Fuentes MÁ, to LTBI after PPD or ESAT-6/CFP-10 stimulation. As previously reported, a ratio based Centeno C, Ruiz-Manzano J, Millet J-P, Molina-Pinargote I, on the CD27 median fluorescence intensity (MFI) was also explored (MFI of CD27 in González-Díaz YD, Lacoma A, CD4+ T-cells over MFI of CD27 in IFN-γ+ CD4+ T-cells), being significantly increased Luque-Chacón L, Sabriá J, Prat C and during disease (p < 0.0001 after PPD or ESAT-6/CFP-10 stimulation). This ratio was Domínguez J (2019) Study of CD27 and CCR4 Markers on Specific CD4+ also assessed on the other CD4+ T-cells functional profiles after specific stimulation, T-Cells as Immune Tools for Active being significantly associated with active TB. Highest diagnostic accuracies for active and Latent Tuberculosis Management. Front. Immunol. 9:3094. TB (AUC ≥ 0.91) were achieved for: (i) CD27 within IFN-γ+ TNF-α+ CD4+ T-cells in doi: 10.3389/fimmu.2018.03094 response to ESAT-6/CFP-10, (ii) CD27 and CCR4 markers together within IFN-γ+ CD4+ Frontiers in Immunology | www.frontiersin.org 1 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management T-cells in response to PPD, and (iii) CD27 MFI ratio performed on IFN-γ+ TNF- α+ CD4+ T-cells after ESAT-6/CFP-10 stimulation. The lowest diagnostic accuracy was observed when CCR4 marker was evaluated alone (AUC ≤ 0.77). CD27 and CCR4 expression detection could serve as a good method for immunodiagnosis. Moreover, the immunological characterization of markers/subset populations could be a promising tool for understanding the biological basis of the disease. Keywords: CD4 T-cells, CD27, CCR4, flow cytometry, immunity, latent tuberculosis, tuberculosis INTRODUCTION cytokines upon antigen encounter. This subset phenotype (CD27− CD4+ or CD27low CD4+ ) is specifically increased in Given the limited knowledge on tuberculosis (TB) biomarkers, whole blood and at the site of infection during active disease (12– the study of different T-cell subsets, as well as Mycobacterium 15). This is due to the process known as homing, which refers tuberculosis specific antigens and cytokines, are attractive options to the migration of specific cell subsets to the infected tissue. to follow in order to understand TB pathogenesis as well as Recently, a new strategy based on CD27 marker detection has the interplay between infection and disease (1–3). Usually, TB been developed for active TB diagnosis (16, 17). This assay is outcome is understood as a bimodal model between active TB able to discriminate between active TB and LTBI by analyzing and latent TB infection (LTBI). However, in the past years, CD27 expression on specific M. tuberculosis CD4+ T-cells that infection has been associated with a dynamic and wide spectrum respond secreting IFN-γ. This new approach, assesses the ratio containing different latency phases (4). Furthermore, active TB of the median fluorescence intensity (MFI) between CD27 on is known to be a heterogeneous disease which comprises a wide CD4+ T-cells and CD27 on specific T-cells in response to PPD range of manifestations and forms. The key to control the spread or ESAT-6/CFP-10 antigens. There are still other novel and of TB is a rapid diagnosis in an early stage. However, active TB potential homing markers to explore that might be useful tools confirmation can be difficult and the commonly used test systems for understanding TB pathogenesis and improving diagnosis. For are still insufficient. Due to all these reasons, the development of example, the chemokine receptor CCR4, which is considered alternative diagnostic methods remains a challenge for improving a homing marker that could be expressed on several cells of TB control. In this aspect, the immunological characterization of the immune system including T helper type 1 (Th1) cells. It is different cell markers has opened the possibility of considering known that T-cells expressing CCR4 surface marker are recruited them as immune tools for TB management, as they could in inflammatory sites (18). Some evidence suggest that the correlate with cell differentiation, latency/disease status, and induction of CCR4 expression is associated with the migration outcome (1, 5). of CD4+ T-cells into the lungs, indicating that this homing M. tuberculosis sensitization can be detected by tuberculin marker could play a protective role in the immunity against skin test (TST), classically used in LTBI diagnosis. However, this some respiratory pathogens (19, 20). Together, these findings assay presents cross-reactive antigens and other proteins which open the possibility for new studies on the development of novel could lead to false-positive results. More than a decade ago, the strategies for TB management and understanding the different in vitro interferon (IFN)-gamma (γ) release assays (IGRAs) were mechanisms against the disease. In the present study, we focus introduced. These tests have been proven to be more specific on the assessment of several T-cell subsets in order to characterize and useful for LTBI diagnosis and sensitization detection than different TB latency/disease immune-mechanisms. This immune the TST. They are not affected by cross-reaction caused by characterization could allow the development of new strategies BCG vaccination and/or non-tuberculous mycobacteria (NTM) for TB management based on the study of the immune response infection (6–8). These techniques are based on the detection of of T-cells expressing CD27 and/or CCR4 markers in patients with IFN-γ released by sensitized T-cells after stimulation with specific active TB and LTBI individuals. M. tuberculosis antigens (ESAT-6 and CFP-10) (9). Nevertheless, both TST and IGRAs identify sensitization to the bacilli, which is translated into a detectable immune response. Therefore, they cannot discriminate between active TB and infection, nor MATERIALS AND METHODS identify those individuals with high risk of developing active TB Study Population and Inclusion Criteria when infected (10, 11). New biomarkers are therefore needed For this study we enrolled subjects with active TB or LTBI to improve disease immune diagnosis, providing prognostic suspicion, who attended the four following centers located information, assessing risk-stratification in LTBI individuals, and in Barcelona (Spain): Hospital Germans Trias i Pujol, Unitat revealing general biological mechanisms of the pathogenesis. de Tuberculosi Vall d’Hebron-Drassanes, Serveis Clínics-Unitat CD4+ T-cells with a low expression of CD27 have been Clínica de Tractament Directament Observat de la Tuberculosi described as an immune biomarker of active TB disease and lung and Hospital Sant Joan Despí Moises Broggi. A total of 16 mL tissue destruction. The decrease of CD27 expression indicates of blood per patient were drawn in CPT tubes (BD Biosciences, the existence of differentiated effector T-cells which produce San Jose, CA, USA). Blood was directly sent to the Institut Frontiers in Immunology | www.frontiersin.org 2 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management d’Investigació Germans Trias i Pujol for peripheral blood APC (BD Bioscience; clone B27) and anti-TNF-α PE-Cy7 (BD mononuclear cells (PBMCs) isolation and cytometry testing. Bioscience; clone Mab11). Markers detection was performed in Subjects enrolled in the study were classified as: (i) adult a BD LSRFortessa flow cytometer (BD Bioscience). A total of patients with active TB (pulmonary or extrapulmonary) with a 100.000 alive CD3+ T-cells were acquired within 2–3 h after positive culture and/or PCR for M. tuberculosis. Patients were staining. enrolled within the first 4 weeks of starting anti-TB therapy; and (ii) individuals with LTBI enrolled during contact tracing Flow Cytometry and Data Analysis studies or LTBI screenings. In this group, LTBI was defined Aggregated cells were excluded by gating on the diagonal that based on a positive TST and/or IGRAs in the absence of clinical appears with forward scatter (FSC)-H and FSC-A characteristics. symptoms and radiological signs compatible with active TB. Then, lymphocytes were selected according to their FSC-A and Chemoprophylaxis was prescribed in all of these subsets, being all side scatter (SSC)-A. CD4+ and CD8+ T-cells were gated based of them enrolled during the first 4 weeks of preventive therapy. on alive CD3+ T-cells. The gating strategy is represented in TST was performed according the Mantoux technique Figure S1 in Supplementary Material. Specific IFN-γ and/or using two tuberculin units of PPD RT23 (Statens Serum TNF-α secretion was analyzed on CD4+ and CD8+ T-cells. Institut, Copenhagen, Denmark), and was evaluated within The expression of CD27 and/or CCR4 markers were studied 48–72 h. According to the Spanish Pulmonology and Thoracic within the following populations after PPD or ESAT-6/CFP- Surgery Society guidelines, a TST ≥ 5 mm was considered 10 stimulation: (i) IFN-γ+ CD4+ T-cells, (ii) TNF-α+ CD4+ positive (21, 22). T-SPOT.TB (Oxford Immunotec, Abingdon, T-cells, and (iii) IFN-γ+ TNF-α+ CD4+ T-cells. In addition, a UK) and QuantiFERON-TB Gold In-Tube (QFN-G-IT; boolean analysis was performed on CD4+ T-cells producing Qiagen, Düsseldorf, Germany) were performed and interpreted IFN-γ+ and/orTNF-α+ in order to characterize the expression according to the manufacturer’s instructions provided in the kits. of CD27 and/or CCR4 on T-cells producing any cytokine. To assess the expression of CD27 and/or CCR4 homing markers, the PBMCs Isolation, Preservation, and frequency of cytokine production after specific stimulation was Stimulation defined as positive when it was twice the amount when compared PBMCs were isolated using CPT tubes (BD Biosciences). to its negative control (unstimulated sample). Fluorescence Afterwards, cells were cryopreserved and stored in liquid Minus One (FMO) controls were included in each experiment to nitrogen for later flow cytometry analyses. Cryopreserved set up the gates. A ratio based on CD27 MFI was also calculated PBMCs were thawed and rested during 2 h in a humidified as suggested by Portevin et al. (16). This ratio is performed incubator at 37◦ C with 5% CO2 in RPMI 1,640 medium (Biowest, measuring the MFI of CD27 marker on CD4+ T-cells over the Nuaillé, France) containing 10% of heat-inactivated fetal calf MFI of CD27 marker on IFN-γ+ CD4+ specific T-cells. This serum (FCS) with benzonase (Sigma, St. Louis, MO, USA; ratio based on CD27 MFI was also studied in the other T-cells final concentration 10 U/mL). PBMCs from each patient were phenotypes (TNF-α+ CD4+ ; IFN-γ+ TNF-α+ CD4+ ; and IFN- stimulated overnight at 37◦ C with 5% CO2 with the recombinant γ+ and/orTNF-α+ CD4+ T-cells). Results comparing percentages proteins ESAT-6/CFP-10 (Lionex Diagnostics and Therapeutics, of CD27− and/or CCR4+ T-cells, as well as CD27 MFI ratios Braunschweig, Germany; final concentration 2 µg/mL for between groups were performed using the two-tailed Mann- each antigen) and PPD (Statens Serum Institut, Copenhagen, Whitney U-test for pairwise comparisons. Differences were Denmark; final concentration 10 µg/mL). The staphylococcal considered statistically significant when a p-value was <0.05. enterotoxin B (SEB; Sigma; final concentration 2.5 µg/mL) was Correlations between the percentage of CD27− T-cells and CD27 used as a positive control. A negative control without stimulation MFI ratio were calculated using the two-tailed non-parametric was also included. Cells were also co-stimulated with anti-CD28 Spearman test. Receiver operating characteristic (ROC) analysis and anti-CD49d monoclonal antibodies (BD Bioscience; final and areas under the curve (AUC) were calculated in order concentration 2 µg/mL each). After 2 h of incubation, Brefeldin to assess the accuracy of the different biomarkers for TB A (BFA; Sigma; final concentration 3 µg/mL) was added into diagnosis. Flow cytometry data was analyzed using BD FACSDiva the culture media to inhibit the intracellular vesicular transport. software (BD Bioscience). Graphical representation is based on Then, PBMCs were left in the incubator overnight. GraphPad Prism version 4 (GraphPad Software, Inc, San Diego, CA). Surface and Intracellular Staining After stimulation, PBMCs were stained with the following surface RESULTS antibodies: anti-CD4 BV786 (BD Bioscience; clone SK3), anti- CD3 PerCP (BioLegend, San Diego, CA, USA; clone SK7), Patient Characteristics anti-CD27 BV605 (BD Bioscience; clone L128), anti-CCR4 PE- A total of 48 subjects were enrolled in the study: (i) 22 CF594 (BD Bioscience; clone 1G1), and anti-CD8 BV510 (BD active pulmonary and extrapulmonary TB patients with M. Bioscience; clone SK1). A viability marker was also used to tuberculosis culture confirmation, and (ii) 26 individuals with exclude dead cells (LIVE/DEAD, Near-IR fluorescent reactive LTBI. Demographical and clinical characteristics are detailed dye; Thermo Fisher, Waltham, MA, USA). For intracellular in Table 1. Overall, 62.5 (30/48) were men and 37.5% (18/48) staining, PBMCs were fixed/permeabilized (IntraStain; Dako, women. The mean age (years) ± standard deviation (SD) was Santa Clara, CA, USA) and then stained with anti-IFN-γ 43.54 ± 16.06. Frontiers in Immunology | www.frontiersin.org 3 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management TABLE 1 | Demographic and clinical characteristics of the participants regarding This was also observed after ESAT-6/CFP-10 stimulation in the study group. disease patients (p < 0.0001 for IFN-γ+ vs. TNF-α+ and p VARIABLES Active TB LTBI = 0.035 for IFN-γ+ vs. IFN-γ+ /TNF-α+ ) and LTBI (p < 0.0001 for IFN-γ+ vs. TNF-α+ and p = 0.043 for IFN-γ+ vs. Participants, n 22 26 IFN-γ+ /TNF-α+ ). Interestingly, cells that only produced TNF-α Mean age, years ± SD 38.68 ± 15.35 47.65 ± 15.77 after PPD stimulation were significantly increased in active TB Male gender, n(%) 18 (81.8) 12 (46.2) in comparison with LTBI (p < 0.0001). This difference was not POSITIVE PCR, n(%) observed after ESAT-6/CFP-10 specific stimulation (Figure 1B). Positive 10 (45.4) – Regarding CD8+ T-cells, specific cytokines responses were Negative 4 (18.2) – detectable in active TB patients and LTBI individuals, indicating Unknown 8 (36.4) – that this T-cell population is also abundant in the immune TYPE of TB response against M. tuberculosis. After PPD stimulation, cells Pulmonary 18 (81.8) – which produced only TNF-α were predominant in active TB Extrapulmonary a 4 (18.2) – patients (p = 0.009 for TNF-α+ vs. IFN-γ+ /TNF-α+ ). This ANTI-TB TREATMENT, n(%) phenotype was not observed for LTBI individuals. In addition, Before starting treatment 3 (13.6) – although differences were not significant, cytokine response After starting treatment (<1month) 19 (86.4) – frequency after ESAT-6/CFP-10 stimulation was higher in active Not prescribed – 26 (100) TB vs. LTBI (Figure 1C). Mean time of anti-TB treatment, days ± SD 15.14 ± 9.77 – CHEMOPROPHYLAXIS, n(%) Expression of CD27 and/or CCR4 Markers Before starting treatment – 2 (7.7) After starting treatment (<1month) – 24 (92.3) Regarding the Clinical Status The percentage of CD27− and/or CCR4+ surface homing Not prescribed 22 (100) – markers was studied on active TB and LTBI within the IFN- Mean time of chemoprophylaxis, days ± SD – 18.6 ± 8.06 γ+ CD4+ T-cells subset after M. tuberculosis specific stimulation PPD CYTOMETRY RESPONDERS, n(%)b (Figure 2A). When the expression of CD27 and CCR4 was IFNγ + CD4+ T cells 22 (100) 26 (100) analyzed separately, the proportion of CD27− or CCR4+ within TNF-α + CD4+ T-cells 22 (100) 26 (100) IFN-γ+ CD4+ T-cells was significantly higher in active TB when TNF-α + IFN-γ + CD4+ T-cells 22 (100) 26 (100) compared with LTBI in response to PPD (p < 0.0001 for CD27− IFNγ + and/orTNF-α + CD4+ T-cells 22 (100) 26 (100) and p = 0.006 for CCR4+ ) or ESAT-6/CFP-10 recombinant ESAT-6/CFP-10 CYTOMETRY RESPONDERS, n(%)b proteins (p < 0.0001 for CD27− ), with the exception of CCR4 IFNγ + CD4+ T cells 21 (95.5) 21 (80.8) marker in response to ESAT-6/CFP-10, where no statistical TNF-α + CD4+ T-cells 22 (100) 25 (96.2) significance was obtained. In addition, both surface T-cell TNF-α + IFN- γ + CD4+ T-cells 21 (95.5) 21 (80.8) markers were analyzed together (CD27− CCR4+ phenotype IFNγ + and/orTNF-α + CD4+ T-cells 22 (100) 25 (96.2) within IFN-γ+ CD4+ T-cell compartment). The proportion of a PleuralTB (n = 2), ganglionar TB (n = 1), and pericardical TB (n = 1). CD27− CCR4+ IFN-γ+ CD4+ T-cells was significantly associated b Number of individuals with a positive CD4+ T-cell response to the specified antigen. The with active TB (p < 0.0001 after PPD or ESAT-6/CFP-10 frequency of the response to any cytokine after specific stimulation was defined as positive stimulation; Figures 2B,C) and reduced the overlapping between when it was twice the amount when compared to its negative control (unstimulated sample). the two clinical status after PPD stimulation. These findings could TB, tuberculosis; LTBI, latent tuberculosis infection; SD, standard deviation; PPD, purified indicate that the loss of CD27 and the increase of CCR4 markers protein derivative. could be associated with M. tuberculosis uncontrolled replication. In our study, active TB patients were recruited within the 4 weeks of starting therapy. In order to explore if these firsts weeks after Cytokine Profile of M. tuberculosis Specific treatment initiation influenced the expression of CD27 and/or CD4+ and CD8+ T-Cell Response in Active CCR4 markers, we performed a Spearman test correlation. No TB and LTBI Individuals significant correlation was observed between days of treatment To better define the specific M. tuberculosis responses in active (within the 4 weeks of starting therapy) and the percentage of TB and LTBI individuals, IFN-γ and/or TNF-α cytokines were CD27− and/or CCR4+ within IFN-γ+ CD4+ T-cells in response measured on CD4+ /CD8+ T-cells after PPD or ESAT-6/CFP-10 to PPD (Figure S2A in Supplementary Material) or ESAT- stimulation (Figure 1A). All individuals included in this study 6/CFP-10 (Figure S2B in Supplementary Material). were responsive to SEB positive control in the cytometry assays. CD27 and/or CCR4 markers were further characterized Regarding CD4+ T-cells functional profile after PPD on CD4+ T-cells producing: (i) TNF-α+ , (ii) TNF-α+ IFN- stimulation, cells producing only IFN-γ were significantly lower γ+ , and (iii) IFN-γ+ and/or TNF-α+ (Boolean analysis) compared to cells producing only TNF-α+ or IFN-γ+ /TNF-α+ after M. tuberculosis specific stimulation. The proportion simultaneously in active TB patients (p < 0.0001 for IFN-γ+ vs. of CD27− , CCR4+ , and CD27− CCR4+ T-cells within TNF-α+ and p < 0.0001 for IFN-γ+ vs. IFN-γ+ /TNF-α+ ) and these three subsets was significantly higher in active TB LTBI individuals (p = 0.0001 for IFN-γ+ vs. IFN-γ+ /TNF-α+ ). patients compared to LTBI individuals in response to Frontiers in Immunology | www.frontiersin.org 4 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management FIGURE 1 | CD4+ /CD8+ T-cells specific cytokine secretion phenotype regarding active TB or LTBI. (A) Dot plots from one active TB representative donor showing the expression of CD4+ and CD8+ T-cells producing IFN-γ and/or TNF-α after PPD stimulation. The frequency of the different cytokines production profile from the total CD4+ and CD8+ T-cells is indicated in each dot plot. (B) CD4+ T-cells and (C) CD8+ T-cells cytokine secretion after PPD or ESAT6/CFP10 stimulation. Black bars represent T-cells which only produce IFN-γ cytokine. Gray bars correspond to T-cells which only produce TNF-α. Dotted bars represent T-cells producing both IFN-γ and TNF-α cytokines. Bars depict medians with interquartile ranges. Differences between conditions were calculated using the two-tailed Mann-Whitney U-test. Only significant differences are represented in the graphs. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. aTB, active TB; LTBI, latent tuberculosis infection. PPD or ESAT-6/CFP-10 specific stimulation (Figure S3 in was observed in T-cells responding to PPD (Figure 3C) or Supplementary Material), with the exception of CCR4 marker ESAT-6/CFP-10 (Figure 3D), which is supported by a significant in response to ESAT-6/CFP-10, where no statistical significance correlation coefficient (for PPD: Spearman’s rho = 0.869, p < was obtained. Furthermore, a significant positive correlation 0.0001; for ESAT-6/CFP-10: Spearman’s rho = 0.892, p < 0.0001). was observed on CD27− expression between antigen-specific The approach based on CD27 MFI was also assessed on (i) IFN-γ+ CD4+ and TNF-α+ CD4+ T-cells (Figure S4A in TNF-α+ CD4+ T-cells, (ii) IFN-γ+ TNF-α+ CD4+ T-cells, and Supplementary Material) or TNF-α+ IFN-γ+ CD4+ T-cells (iii) IFN-γ+ and/or TNF-α+ CD4+ T-cells after M. tuberculosis (Figure S4B in Supplementary Material). specific stimulation. These ratios were significantly higher in active TB patients than in LTBI individuals after PPD (Figure S5A in Supplementary Material) or ESAT-6/CFP-10 CD27 MFI Ratio Analysis (Figure S5B in Supplementary Material) stimulation. An approach based on CD27 MFI on CD4+ T-cells was assessed as suggested by Portevin et al. (16). This method consists on evaluating the ratio between CD27 MFI in CD4+ T-cells and the MFI of CD27 in specific IFN-γ+ CD4+ T-cells. Therefore, a Diagnostic Accuracy of the Different low CD27 MFI on specific CD4+ T-cells which produce IFN-γ+ Biomarkers implies a high CD27 ratio (this is a consequence of an increase To asses TB diagnostic accuracy of the different approaches of the CD27− IFN-γ+ CD4+ T-cells phenotype). In this study, analyzed in this study, we performed a ROC curve analysis a high ratio was significantly associated with active TB when (Table 2). Highest AUC values (AUC > 0.90) for discriminating T-cells were stimulated with PPD (Figure 3A; p < 0.0001) or active TB from LTBI were achieved when evaluating: (i) CD27 ESAT-6/CFP-10 (Figure 3B; p < 0.0001). within IFN-γ+ CD4+ and IFN-γ+ TNF-α+ CD4+ T-cells in In order to explore whether a high CD27 MFI ratio response to ESAT-6/CFP-10 [AUC (95% confidence interval, CI) was associated with an increase of the percentage of CD27− 0.90 (0.79–1.01) and 0.92 (0.84–1.01) respectively], (ii) CD27 and within IFN-γ+ CD4+ T-cells, a Spearman test correlation was CCR4 markers together within IFN-γ+ CD4+ , TNF-α+ CD4+ performed. A positive correlation between these two variables and IFN-γ+ TNF-α+ CD4+ T-cells in response to PPD [AUC Frontiers in Immunology | www.frontiersin.org 5 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management FIGURE 2 | CD27− and/or CCR4+ phenotype from CD4+ IFN-γ+ specific T-cells in patients with active TB and LTBI individuals. (A) Representative example from an active TB and a LTBI individual showing the strategy for determining CD27− or CCR4+ T-cells after PPD stimulation. CD27 and CCR4 expression was analyzed within the specific IFN-γ+ CD4+ T-cells population after PPD or ESAT-6/CFP-10 stimulation. A negative control without stimulation was also included in the analysis for each patient enrolled in the study. Fluorescence Minus One (FMO) controls were included in each experiment for setting up gates. (B) Percentage of PPD or (C) ESAT6/CFP10 specific CD27− , CCR4+ , and CD27− CCR4+ within IFN-γ+ CD4+ specific T-cells. Horizontal lines represent medians. Differences between conditions were calculated using the two-tailed Mann-Whitney U-test. **p < 0.01, ****p < 0.0001. ns, non-significant. aTB, active TB; LTBI, latent tuberculosis infection. (95% CI) 0.91 (0.83–0.99), 0.90 (0.82–0.99), and 0.90 (0.81– results confirm previous reports on CD27 modulation in 0.99) respectively], and (iii) CD27 MFI ratio performed on IFN- specific CD4+ T-cells producing IFN-γ, which is downregulated γ+ CD4+ and IFN-γ+ TNF-α+ CD4+ T-cells after ESAT-6/CFP- during active disease. Furthermore, a high CD27 MFI ratio 10 specific stimulation [AUC (95% CI) 0.90 (0.79–1.01) and 0.91 proposed as a disease biomarker by other authors (16, 17) (0.82–1.01), respectively]. The lowest diagnostic accuracy was was associated with active TB patients compared to LTBI observed when CCR4 marker was evaluated alone (Figure 4). individuals. This study adds novel information on other potential homing biomarkers such as CCR4, showing that in combination DISCUSSION with CD27 (CD27− CCR4+ IFN-γ+ CD4+ T-cells) could be a phenotype to discriminate between disease and infection. Approaches based on the study of the host immune response Furthermore, we also characterized CD27 on CD4+ T-cells have emerged as potential tools for TB management, studying the producing TNF-α, observing that this marker has a high power interplay between the host and M. tuberculosis, and discovering of discrimination when analyzed within IFN-γ+ TNF-α+ CD4+ suitable disease biomarkers. Here we have analyzed specific T-cells population. immune-mechanisms based on the characterization of different The study of cytokine profiles on specific M. tuberculosis T-cell subsets and the expression of surface receptors such T-cell responses has suggested that specific subsets may serve as CD27 and/or CCR4 involved in the migration of certain as disease biomarkers associated with bacterial load, treatment lymphocytes to the disease inflammatory sites. Briefly, our response or disease outcome. This data is still limited and Frontiers in Immunology | www.frontiersin.org 6 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management FIGURE 3 | CD27 MFI ratio according to the study group. A ratio based on CD27 MFI was calculated after specific stimulation as suggested by Portevin et al. (16). This ratio is based on the MFI of CD27 in CD4+ T-cells over MFI of CD27 in IFN-γ+ CD4+ T-cells. (A) CD27 MFI ratio after PPD or (B) ESAT-6/CFP-10 stimulation in patients with active TB and LTBI individuals. Horizontal lines represent medians. Differences between conditions were calculated using the two-tailed Mann-Whitney U-test. ****p < 0.0001. (C,D) Correlation of the percentage of CD27− marker within IFN-γ+ CD4+ specific T-cells with the CD27 MFI ratio after PPD or ESAT-6/CFP-10 stimulation. Correlation was calculated using the two-tailed non-parametric Spearman test. aTB, active TB; LTBI, latent tuberculosis infection. need to be further evaluated. In this context, a previous study the accuracy of the assay and to correlate T-cell responses with indicated that CD4+ T-cells that only produce TNF-α could be antigen load or high risk of TB progression (29). However, associated with active disease (23). Others have reinforced this data about its accuracy over classical IGRAs or correlation with data suggesting that PPD specific TNF-α+ CD4+ T-cells with an disease state is still limited. effector phenotype can accurately discriminate active TB from In this study, we confirm that the evaluation of the frequency LTBI, or even recently acquired from remote LTBI (24, 25). of CD27− within functional CD4+ T-cells, together with the The results we obtained on M. tuberculosis T-cells functional CD27 MFI ratio, were suitable biomarkers for TB which profile were also in agreement with those obtained in previous could discriminate disease from infection with acceptable AUC studies, as we found that CD4+ T-cell only producing TNF-α values between 0.82 and 0.92 depending on the stimuli used in response to PPD were increased in active TB patients. While (PPD or ESAT-6/CFP-10). The CD27 marker is a member it is widely accepted that CD4+ T-cells play an essential role of the TNF-receptor superfamily, expressed by lymphocytes, against the bacilli’s immune response, protective immunity to M. which is downregulated during effector differentiated T-cells tuberculosis by CD8+ T-cells still remains controversial. In the able to produce cytokines (14). Thus, due to the persistent last years, CD8+ T-cells have emerged as a possible population antigenic stimulation during active TB, it has been proposed actively involved in the immunopathology (26–28). Our study as an immune biomarker of the disease (16, 17, 30, 31). The also corroborates the importance of CD8+ T-cells as TB control recently developed immune assay based on the detection of players, showing detectable cytokine responses in this population CD27 MFI ratio (TAM-TB assay) has also been proposed as which tend to be higher during disease in response to ESAT-6 and an alternative way for measuring this receptor (16). Here, we CFP-10 antigens. One important challenge for TB management show that the percentage of CD27− significantly correlated and diagnosis is to find specific antigens capable to elicit CD8+ with CD27 MFI quantification, indicating that both immune T-cells responses. In this context, a new generation of QFN strategies are accurate enough for TB diagnosis. However, the called QFN-Plus has incorporated new peptides able to induce calculation of a ratio based on MFI allows normalization of IFN-γ responses on CD4+ and CD8+ T-cells, trying to increase the results avoiding subjectivity and discrepancies on CD27 Frontiers in Immunology | www.frontiersin.org 7 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management positive or negative gating. In addition, our results also add new infection), found that in an active TB context T-cells presented information about the CCR4 homing marker. We have observed a CD27 marker downregulation. In contrast, no difference on that overexpression of CCR4 receptor within IFN-γ+ CD4+ T- CCR4 expression was found regarding the clinical status of the cells is a poor immune biomarker of disease when evaluated individuals (irrespective of HIV infection) when this marker alone, however, when combined together with CD27− IFN- was detected alone (31). In addition, other possible disease γ+ CD4+ T-cells, its discriminatory capacity increased (0.91 immune markers have been studied by others in order to improve after PPD stimulation). CCR4 has been suggested as a lung TB diagnosis accuracy. For example, the expression of the homing receptor expressed on T-cells. A recent study focused activation marker HLA-DR on specific CD4+ T-cells has shown on the detection of CD27 or CCR4 markers (among others) a good discriminatory capacity between active TB and LTBI on active TB and LTBI individuals (with and without HIV (30, 31). This study also adds new data on CD27 and/or CCR4 characterization within T-cells secreting IFN-γ and/or TNF-α. TABLE 2 | ROC curve analysis of the different approaches. IFN-γ cytokine does not fully represent the response against M. tuberculosis, having TNF-α an important role during active TB PHENOTYPE AUC (95% CI) disease. In this context, we found that CD27− and/or CCR4+ PPD ESAT-6/CFP-10 expression within: (i) TNF-α+ CD4+ ; (ii) IFN-γ+ TNF-α+ CD4+ T-cells, and (iii) IFNγ+ and/orTNF-α+ CD4+ T-cells, as well CD27− as CD27 MFI ratio measured in these functional populations, IFNγ + CD4+ 0.85 (0.72–0.97) 0.90 (0.79–1.01) were increased in active TB patients in comparison with LTBI TNF-α + CD4+ 0.85 (0.74–0.97) 0.86 (0.75–0.97) individuals. Highest discriminatory capacities were achieved TNF-α + IFN-γ + CD4+ 0.88 (0.77–098) 0.92 (0.84–1.01) when measuring CD27− or CD27 MFI ratio within IFN- IFNγ + and/orTNF-α + CD4+ 0.82 (0.69–0.95) 0.87 (0.76–0.98) γ+ TNF-α+ CD4+ T-cells (AUC 0.92 and 0.91 after ESAT- CCR4+ 6/CFP-10 stimulation). This indicates that CD4+ T-cells lacking IFNγ + CD4+ 0.73 (0.59–0.87) 0.60 (0.43–0.78) CD27 marker are able to differentiate into effector T-cells TNF-α + CD4+ 0.70 (0.55–0.85) 0.50 (0.33–0.67) and increment their capacity to secrete IFN-γ and/or TNF-α TNF-α + IFN-γ + CD4+ 0.77 (0.64–0.91) 0.65 (0.49–0.83) cytokines. In addition, the study of CD27 on TNF-α producing IFNγ + and/orTNF-α + CD4+ 0.72 (0.58–0.87) 0.59 (0.43–0.75) T-cells increased the detection of positive responses by flow CD27− CCR4+ cytometry after ESAT-6/CFP-10 stimulation, especially in the IFNγ + CD4+ 0.91 (0.83–0.99) 0.86 (0.73–0.98) LTBI group. In mice it has been shown that IFN-γ+ CD4+ T-cells TNF-α + CD4+ 0.90 (0.82–0.99) 0.84 (0.71–0.96) which have CD27 receptor downregulated are accumulated TNF-α + IFN-γ + CD4+ 0.90 (0.81–0.99) 0.89 (0.77–0.99) preferentially in the lungs during mycobacterial infection (32). IFNγ + and/orTNF-α + CD4+ 0.88 (0.78–0.97) 0.84 (0.71–0.96) Furthermore, CD27low specific CD4+ T-cells are increased in MFI CD27 lungs of patients with active TB, and percentages of this subset IFNγ + CD4+ 0.84 (0.71–0.97) 0.90 (0.79–1.01) are higher in lung tissue than in blood. Interestingly, when this TNF-α + CD4+ 0.84 (0.71–0.97) 0.87 (0.76–0.97) T-cell subset was detected in blood, it correlated with tissue TNF-α + IFN-γ + CD4+ 0.84 (0.71–0.97) 0.91 (0.82–1.01) destruction and TB severity. This correlation was not observed IFNγ + and/orTNF-α + CD4+ 0.84 (0.71–0.97) 0.88 (0.77–0.99) when the T-cell subset was detected in the lungs. The reasons AUC, area under the curve; CI, confidence interval; PPD, purified protein derivative. of this discordance are still unclear, but could be explained by FIGURE 4 | Heatmap depicting Areas Under the Curve (AUC) values for the different approaches. A ROC curve analysis was performed to determine the diagnostic accuracy for TB diagnosis. AUC values are represented for: (i) percentage of CD27− marker, (ii) percentage of CCR4+ marker, (iii) percentage of CD27− CCR4+ signature, and (iv) CD27 MFI ratio. Values are shown for the different CD4+ T-cells functional populations analyzed after PPD or ESAT-6/CFP-10 specific stimulation. High AUC values are indicated by intensity of blue color. Frontiers in Immunology | www.frontiersin.org 8 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management its generation from different precursors. CD27low specific CD4+ All enrolled subjects gave a written informed consent for T-cells can be generated in the lymph nodes and then migrate participating in this study. to peripheral blood, while those located in the lungs can be generated locally from other precursors (14). In the same context, AUTHOR CONTRIBUTIONS severe TB induced the upregulation of CCR4 gene (among others) in pulmonary compartments of infected rhesus monkeys IL and JD designed the study. IL, MF-S, and RV-H designed (33). According to these findings, it would be interesting to study the experiments. BM-M, SV, RV-H, and IL performed the CD27 and/or CCR4 in samples from active TB patients coming experiments. MDS-G, ZS, MJ-F, CC, JR-M, J-PM, IM-P, YG-D, from the site of infection in order to understand better the LL-C, JS, and CP contributed with resources. IL, MF-S, and RV-H mechanisms and pathogenesis of the disease. Further studies in analyzed the data. IL, JD, AL, and CP supervised the study. IL this direction need to be addressed. and JD wrote the paper. All authors revised and approved the The monitoring of anti-TB therapy efficacy is a key point manuscript. for TB control. Petruccioli E. et al. showed that the expression of CD27 increases on specific T-cells in cured active TB patients after 1 year of therapy completion (17). These findings FUNDING suggest that CD27 might serve as a tool for following-up This research was supported by: (i) a grant from the active TB patients, detecting efficacy of treatment, and exploring Instituto de Salud Carlos III (PI 13/01546, PI16/01912, and inflammatory status. In this line, the study of a differential PI18/00411), integrated in the Plan Nacional de I+D+I phenotype on T-cells expressing CD27 and/or CCR4 homing and cofunded by the ISCIII Subdirección General de markers during the treatment follow-up of active TB patients Evaluación and the Fondo Europeo de Desarrollo Regional is needed for validating these findings. Some data support that (FEDER); and (ii) a grant from the Sociedad Española de only after 2 months of TB therapy the expression of CD27 Neumología y Cirugía Torácica (project 25/2016; SEPAR; starts its modulation (14). Thus, according to these data, the Barcelona, Spain). JD is a researcher from the Miguel Servet detection of CD27− CD4+ T-cells during the first month of programme. treatment on patients recruited in our study should not be altered. This hypothesis is also reinforced by our results, as no significant correlation between days of treatment (within the ACKNOWLEDGMENTS firsts 4 weeks of therapy) and the CD27/CCR4 expression was We thank G. Requena from IGTP Cytometry Core Facility found. for his contribution to this publication. The authors would Limitations of this study need to be addressed. First, although also like to thank: (i) M. Pérez from Hospital Germans results obtained on CD27 are robust among different studies, Trias i Pujol; (ii) J. Santiago, N. Forcada, D. Romero, it is important to uniformly validate this immune assay for J. Soteras, N. Altet, and X. Casas from Serveis Clínics- routine purposes and results reproducibility, choosing common Unitat Clínica de Tractament Directament Observat starting material (PBMCs or whole blood), same specific stimuli de la Tuberculosi; and (iii) S. Damoun from Institut and fluorochromes, as well as standardizing protocols. And d’Investigació Germans Trias i Pujol for their technical second, the triggering of host immune responses and disease assistance. outcome does not depend only on a single factor. Immune status of the host depends on a troika of multiple parameters covering host genetics, the pathogen and extrinsic elements SUPPLEMENTARY MATERIAL (34). Therefore, it is necessary to study and combine all these variables together in other to translate possible host immune The Supplementary Material for this article can be found TB biomarkers into potential immune assays with clinical online at: https://www.frontiersin.org/articles/10.3389/fimmu. applications. 2018.03094/full#supplementary-material In summary, our findings on surface homing markers such Figure S1 | Gating strategy used for CD4+ /CD8+ T-cells cytokine secretion and CD27/CCR4 analysis. Aggregated cells were taken off by gating on the diagonal as CD27 and CCR4 on M. tuberculosis specific CD4+ T-cells that appears with Forward-Scatter (height; FSC-H) vs. Forward-Scatter (area; gather the required features for using them as potential TB FSC-A) dot plot. For CD4+ and CD8+ T-cells gating, alive CD3+ T-cells were first biomarkers. Therefore, it would be crucial to further evaluate selected. Cytokine profiles on T-cells were analyzed on CD4+ /CD8+ T-cells. these receptors in a larger cohort of patients in order to develop CD27/CCR4 expression was studied within IFN-γ+ CD4+ T-cells gated from alive possible and simplified routine immune assays for TB diagnosis, CD3+ T-cells. assessment of therapy efficacy/relapse, and risk-stratification of Figure S2 | Correlation of homing markers expression with days of treatment. LTBI individuals. Correlation of days after starting anti-TB therapy in active TB patients (within the 4 weeks of starting treatment) with (A) percentage of CD27− and/or CCR4+ within IFN-γ+ CD4+ T-cells in response to PPD or (B) ESAT-6/CFP-10 antigens. Correlation was calculated using the two-tailed non-parametric Spearman test. ETHICS STATEMENT Figure S3 | CD27− and/or CCR4+ phenotype within functional CD4+ T-cells producing IFN-γ and/or TNF-α in patients with active TB and LTBI individuals. The study was approved by the Ethics Committee of the Percentage of PPD or ESAT-6/CFP-10 specific CD27− , CCR4+ , and Hospital Germans Trias i Pujol (Reference number PI-17-134). CD27− CCR4+ within (A) TNF-α+ CD4+ T-cells, (B) IFN-γ+ TNF-α+ CD4+ T-cells, Frontiers in Immunology | www.frontiersin.org 9 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management and (C) IFN-γ+ and/or TNF-α+ CD4+ T-cells. Horizontal lines represent medians. Figure S5 | CD27 MFI ratio calculated on functional CD4+ T-cells producing Differences between conditions were calculated using the two-tailed IFN-γ and/or TNF-α. A ratio based on CD27 MFI was calculated after specific Mann-Whitney U-test. ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001, ∗∗∗∗ p < 0.0001. ns, stimulation in active TB patients and LTBI individuals. This ratio is based on the non-significant; aTB, active TB; LTBI, latent tuberculosis infection. MFI of CD27 in CD4+ T-cells over: (i) MFI of CD27 in TNF-α+ CD4+ T-cells, (ii) MFI Figure S4 | Relationship of the CD27− expression on the different of CD27 in IFN-γ+ TNF-α+ CD4+ T-cells, and (iii) MFI of CD27 in IFN-γ+ and/or antigen-specific T-cells populations analyzed. Correlation of the CD27− TNF-α+ CD4+ T-cells after (A) PPD or (B) ESAT-6/CFP-10 antigen stimulation. expression on IFN-γ+ CD4+ T-cells with (A) TNF-α+ CD4+ or (B) TNF-α+ IFN-γ+ Horizontal lines represent medians. Differences between conditions were CD4+ T-cells after PPD or ESAT-6/CFP-10 antigen stimulation. Correlation was calculated using the two-tailed Mann-Whitney U-test. ∗∗∗∗ p < 0.0001. aTB, active calculated using the two-tailed non-parametric Spearman test. TB; LTBI, latent tuberculosis infection. REFERENCES Mycobacterium tuberculosis-specific CD4 T cells after HIV-1 infection. J Exp Med. (2010) 207:2869–81. doi: 10.1084/jem.20100090 1. Petruccioli E, Scriba TJ, Petrone L, Hatherill M, Cirillo DM, Joosten 16. Portevin D, Moukambi F, Clowes P, Bauer A, Chachage M, Ntinginya SA, et al. Correlates of tuberculosis risk: predictive biomarkers for NE, et al. Assessment of the novel T-cell activation marker- progression to active tuberculosis. Eur Respir J. (2016) 48:175–63. tuberculosis assay for diagnosis of active tuberculosis in children: a doi: 10.1183/13993003.01012-2016 prospective proof-of-concept study. Lancet Infect Dis. (2014) 14:931–8. 2. Goletti D, Lee MR, Wang JY, Walter N, Ottenhoff THM. Update doi: 10.1016/S1473-3099(14)70884-9 on tuberculosis biomarkers: from correlates of risk, to correlates of 17. Petruccioli E, Petrone L, Vanini V, Cuzzi G, Navarra A, Gualano G, et al. active disease and of cure from disease. Respirology. (2018) 23:455–66. Assessment of CD27 expression as a tool for active and latent tuberculosis doi: 10.1111/resp.13272 diagnosis. J Infect. (2015) 71:526–33. doi: 10.1016/j.jinf.2015.07.009 3. Serra-Vidal MM, Latorre I, Franken KL, Diaz J, de Souza-Galvao ML, Casas I, 18. Slight SR, Khader SA. Chemokines shape the immune responses et al. Immunogenicity of 60 novel latency-related antigens of Mycobacterium to tuberculosis. Cytokine Growth Factor Rev. (2013) 24:105–13. tuberculosis. Front Microbiol. (2014) 5:517. doi: 10.3389/fmicb.2014.00517 doi: 10.1016/j.cytogfr.2012.10.002 4. Barry CE III, Boshoff HI, Dartois V, Dick T, Ehrt S, Flynn J, et al. The spectrum 19. Mikhak Z, Strassner JP, Luster AD. Lung dendritic cells imprint T cell lung of latent tuberculosis: rethinking the biology and intervention strategies. Nat homing and promote lung immunity through the chemokine receptor CCR4. Rev Microbiol. (2009) 7:845–55. doi: 10.1038/nrmicro2236 J Exp Med. (2013) 210:1855–69. doi: 10.1084/jem.20130091 5. Walzl G, Ronacher K, Hanekom W, Scriba TJ, Zumla A. Immunological 20. Bordon Y. Mucosal immunology: air miles for T cells. Nat Rev Immunol. biomarkers of tuberculosis. Nat Rev Immunol. (2011) 11:343–54. (2013) 13:705. doi: 10.1038/nri3537 doi: 10.1038/nri2960 21. Gonzalez-Martin J, Garcia-Garcia JM, Anibarro L, Vidal R, Esteban J, 6. Ferrara G, Losi M, D’Amico R, Roversi P, Piro R, Meacci M, et al. Use Blanquer R, et al. Consensus document on the diagnosis, treatment in routine clinical practice of two commercial blood tests for diagnosis of and prevention of tuberculosis. Arch Bronconeumol. (2010) 46:255–74. infection with Mycobacterium tuberculosis: a prospective study. Lancet (2006) doi: 10.1016/S1579-2129(10)70061-6 367:1328–34. doi: 10.1016/S0140-6736(06)68579-6 22. Ruiz-Manzano J, Blanquer R, Calpe JL, Caminero JA, Cayla J, Dominguez 7. Dominguez J, Ruiz-Manzano J, De Souza-Galvao M, Latorre I, Mila C, Blanco JA, et al. Diagnosis and treatment of tuberculosis. Arch Bronconeumol. (2008) S, et al. Comparison of two commercially available gamma interferon blood 44:551–66. doi: 10.1157/13126836 tests for immunodiagnosis of tuberculosis. Clin Vaccine Immunol. (2008) 23. Harari A, Rozot V, Enders FB, Perreau M, Stalder JM, Nicod LP, et al. 15:168–71. doi: 10.1128/CVI.00364-07 Dominant TNF-alpha+ Mycobacterium tuberculosis-specific CD4+ T cell 8. Latorre I, De Souza-Galvao M, Ruiz-Manzano J, Lacoma A, Prat C, responses discriminate between latent infection and active disease. Nat Med. Altet N, et al. Evaluating the non-tuberculous mycobacteria effect in (2011) 17:37–26. doi: 10.1038/nm.2299 the tuberculosis infection diagnosis. Eur Respir J. (2010) 35:338–42. 24. Pollock KM, Whitworth HS, Montamat-Sicotte DJ, Grass L, Cooke GS, doi: 10.1183/09031936.00196608 Kapembwa MS, et al. T-cell immunophenotyping distinguishes active from 9. Andersen P, Munk ME, Pollock JM, Doherty TM. Specific immune- latent tuberculosis. J Infect Dis. (2013) 208:952–68. doi: 10.1093/infdis/jit265 based diagnosis of tuberculosis. Lancet (2000) 356:1099–104. 25. Halliday A, Whitworth H, Kottoor SH, Niazi U, Menzies S, Kunst H, doi: 10.1016/S0140-6736(00)02742-2 et al. Stratification of latent Mycobacterium tuberculosis infection by cellular 10. Sester M, Sotgiu G, Lange C, Giehl C, Girardi E, Migliori GB, et al. immune profiling. J Infect Dis. (2017) 215:1480–7.doi: 10.1093/infdis/jix107 Interferon-gamma release assays for the diagnosis of active tuberculosis: 26. Rozot V, Vigano S, Mazza-Stalder J, Idrizi E, Day CL, Perreau M, et al. a systematic review and meta-analysis. Eur Respir J. (2011) 37:100–11. Mycobacterium tuberculosis-specific CD8+ T cells are functionally and doi: 10.1183/09031936.00114810 phenotypically different between latent infection and active disease. Eur J 11. Latorre I, De Souza-Galvao M, Ruiz-Manzano J, Lacoma A, Prat C, Immunol. (2013) 43:1568–77. doi: 10.1002/eji.201243262 Fuenzalida L, et al. Quantitative evaluation of T-cell response after 27. Lin PL, Flynn JL. CD8 T cells and Mycobacterium tuberculosis infection. Semin specific antigen stimulation in active and latent tuberculosis infection Immunopathol. (2015) 37:239–49. doi: 10.1007/s00281-015-0490-8 in adults and children. Diagn Microbiol Infect Dis. (2009) 65:236–46. 28. Behar SM. Antigen-specific CD8(+) T cells and protective doi: 10.1016/j.diagmicrobio.2009.07.015 immunity to tuberculosis. Adv Exp Med Biol. (2013) 783:141–63. 12. Jiang J, Wang X, Wang X, Cao Z, Liu Y, Dong M, et al. Reduced doi: 10.1007/978-1-4614-6111-1_8 CD27 expression on antigen-specific CD4+ T cells correlates with 29. Petruccioli E, Chiacchio T, Pepponi I, Vanini V, Urso R, Cuzzi G, et al. First persistent active tuberculosis. J Clin Immunol. (2010) 30:566–73. characterization of the CD4 and CD8 T-cell responses to QuantiFERON-TB doi: 10.1007/s10875-010-9418-1 Plus. J Infect. (2016) 73:588–97. doi: 10.1016/j.jinf.2016.09.008 13. Streitz M, Tesfa L, Yildirim V, Yahyazadeh A, Ulrichs T, Lenkei R, et al. Loss of 30. Musvosvi M, Duffy D, Filander E, Africa H, Mabwe S, Jaxa L, et al. T-cell receptor on tuberculin-reactive T-cells marks active pulmonary tuberculosis. biomarkers for diagnosis of tuberculosis: candidate evaluation by a simple PLoS ONE (2007) 2:e735. doi: 10.1371/journal.pone.0000735 whole blood assay for clinical translation. Eur Respir J. (2018) 51:1800153. 14. Nikitina IY, Kondratuk NA, Kosmiadi GA, Amansahedov RB, Vasilyeva IA, doi: 10.1183/13993003.00153-2018 Ganusov VV, et al. Mtb-specific CD27low CD4 T cells as markers of lung tissue 31. Riou C, Berkowitz N, Goliath R, Burgers WA, Wilkinson RJ. Analysis destruction during pulmonary tuberculosis in humans. PLoS ONE (2012) of the phenotype of Mycobacterium tuberculosis-specific CD4+ 7:e43733. doi: 10.1371/journal.pone.0043733 T cells to discriminate latent from active tuberculosis in HIV- 15. Geldmacher C, Ngwenyama N, Schuetz A, Petrovas C, Reither uninfected and HIV-infected individuals. Front Immunol. (2017) 8:968. K, Heeregrave EJ, et al. Preferential infection and depletion of doi: 10.3389/fimmu.2017.00968 Frontiers in Immunology | www.frontiersin.org 10 January 2019 | Volume 9 | Article 3094 Latorre et al. Homing Markers in Tuberculosis Management 32. Kapina MA, Shepelkova GS, Mischenko VV, Sayles P, Bogacheva P, Conflict of Interest Statement: The authors declare that the research was Winslow G, et al. CD27low CD4 T lymphocytes that accumulate in the mouse conducted in the absence of any commercial or financial relationships that could lungs during mycobacterial infection differentiate from CD27high precursors be construed as a potential conflict of interest. in situ, produce IFN-gamma, and protect the host against tuberculosis infection. J Immunol. (2007) 178:976–85. doi: 10.4049/jimmunol.178.2.976 Copyright © 2019 Latorre, Fernández-Sanmartín, Muriel-Moreno, Villar- 33. Qiu L, Huang D, Chen CY, Wang R, Shen L, Shen Y, et al. Severe tuberculosis Hernández, Vila, Souza-Galvão, Stojanovic, Jiménez-Fuentes, Centeno, induces unbalanced up-regulation of gene networks and overexpression of Ruiz-Manzano, Millet, Molina-Pinargote, González-Díaz, Lacoma, Luque- IL-22, MIP-1alpha, CCL27, IP-10, CCR4, CCR5, CXCR3, PD1, PDL2, IL-3, Chacón, Sabriá, Prat and Domínguez. This is an open-access article distributed IFN-beta, TIM1, and TLR2 but low antigen-specific cellular responses. J Infect under the terms of the Creative Commons Attribution License (CC BY). The use, Dis. (2008) 198:1514–9. doi: 10.1086/592448 distribution or reproduction in other forums is permitted, provided the original 34. Bastos HN, Osorio NS, Gagneux S, Comas I, Saraiva M. The author(s) and the copyright owner(s) are credited and that the original publication troika host-pathogen-extrinsic factors in tuberculosis: modulating in this journal is cited, in accordance with accepted academic practice. No use, inflammation and clinical outcomes. Front Immunol. (2017) 8:1948. distribution or reproduction is permitted which does not comply with these doi: 10.3389/fimmu.2017.01948 terms. Frontiers in Immunology | www.frontiersin.org 11 January 2019 | Volume 9 | Article 3094