YCLIM-07461; No. of pages: 10; 4C: Clinical Immunology (2015) xx, xxx–xxx
available at www.sciencedirect.com
Clinical Immunology www.elsevier.com/locate/yclim
4Q4 5
F
O
3
R O
2
Differential expression of CD57 in antigen-reactive CD4 + T cells between active and latent tuberculosis infection
Ji Yeon Lee a , Ina Jeong a , Joon-Sung Joh a , Young Won Jung b , Soo Yeon Sim c , Boram Choi c , Hyeon-Gun Jee c , Dong-Gyun Lim c,⁎
6
a
7
b
Division of Respiratory Diseases, Department of Internal Medicine, National Medical Center, Seoul 100-799, South Korea Jung-gu Community Health Center, Seoul 132-713, South Korea c Center for Chronic Diseases, Research Institute, National Medical Center, Seoul 100-799, South Korea
D
8
P
1Q3
10
E
9
Received 16 December 2014; accepted with revision 8 April 2015
17 22
Abstract The development of diagnostic tests that predict the progression of latent tuberculosis infection to active disease is pivotal for the eradication of tuberculosis. As an initial step to achieve this goal, our study's aim was to identify biomarkers that differentiate active from latent tuberculosis infection. We compared active and latent tuberculosis infection groups in terms of the precursor frequency, functional subset differentiation, and senescence/exhaustion surface marker expression of antigen-specific CD4+ T cells, which were defined as dividing cells upon their encountering with Mycobacterium (M.) tuberculosis antigens. Among several parameters shown to have statistically significant differences between the two groups, the frequency of CD57-expressing cells could differentiate effectively between active disease and latent infection. Our results suggest that the expression of CD57 in M. tuberculosis-reactive CD4+ T cells could be a promising candidate biomarker with which to identify individuals with latent tuberculosis infection prone to progression to active disease. © 2015 Published by Elsevier Inc.
C
15 20 16 21
KEYWORDS Tuberculosis; CD57; CD4+ T cell; Biomarker
E
13 18 14 19
T
11
R
23 24
R
25 26 27
O
28 29
C
30 31
34
U
33
N
32
Abbreviations: ATB, active tuberculosis; AUC, area under the curve; CFSE, carboxyfluorescein succinimidyl ester; IGRA, interferon gamma release assay; LTBI, latent tuberculosis infection; Mtb, Mycobacterium tuberculosis; ROC, receiver operating characteristic; TB, tuberculosis. ⁎ Corresponding author at: Center for Chronic Diseases, Research Institute, National Medical Center, 245, Euljiro, Jung-gu, Seoul 100-799, South Korea. Fax: +82 2 2276 2319. E-mail address: [email protected] (D.-G. Lim).
1. Introduction
35
Tuberculosis (TB) remains a highly prevalent and lifethreatening disease to humans. When infected with Mycobacterium tuberculosis (Mtb), only a small proportion of the infected population develop primary active tuberculosis (ATB) disease, while over 90% carry the infection in a latent or subclinical stage [1]. It is estimated that the risk of active disease is approximately 5% in the first 18 months following
36
http://dx.doi.org/10.1016/j.clim.2015.04.011 1521-6616 © 2015 Published by Elsevier Inc. Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
37 38 39 40 41 42
2
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104
107
2.2. Mycobacterial antigens
122
F
55
A total of 24 patients with active pulmonary TB were recruited at either the National Medical Center or Jung-gu Community Health Center (Seoul, Korea) (Table 1). These patients were diagnosed, based on either a positive culture for Mtb or positive acid-fast bacilli in the sputum, or both, and their blood samples were obtained before or within the first 2 weeks of anti-TB medication. Moreover, a second study group of 24 healthy subjects with LTBI, which was defined by a positive tuberculin skin test or IGRA, in the absence of clinical symptoms, were recruited. HIV infection was negative in all tested individuals. Age and gender data of the study groups are shown in Table 1. The study was approved by the Ethics Committee of National Medical Center (IRB no. H-1208/022-002; 23 August 2012), and written informed consent was obtained from all participants.
O
54
106
R O
53
2.1. Study population
P
52
105
Recombinant ESAT-6 and CFP-10 antigens were provided by Dr. Sang-Nae Cho (International Tuberculosis Research Center, Seoul, Korea). PPD (RT50) was obtained from Statens Serum Institute (Copenhagen, Denmark).
D
51
2. Materials and methods
E
50
T
49
C
48
E
47
R
46
R
45
initial infection, and 5% for the remaining lifetime [2]. Due to the large number of infected individuals in the latent stage worldwide, reactivation of latent infection is a significant contributing factor to the prevalence rate of TB, especially in developed countries. Thus, in addition to the treatment of active cases, prophylactic treatment of latent infection represents an important option for the eradication of TB. Latent infection is known to be cleared effectively with standard drug regimens [3]. However, at least 20 patients with latent TB infection (LTBI) need to be administered with chemotherapeutic agents, which carry side effects, for each potential LTBI patient who progresses to ATB and is cured, which is neither cost-effective nor efficient in terms of a public health strategy to tackle TB. Therefore, an availability of prognostic biomarkers stratifying the risk of progression from LTBI to ATB would allow the targeting of treatment to those most at risk. So far, several measures, such as interferon gamma (IFN-γ) release assay (IGRA) [4], plasma cytokine levels [5,6], and even whole-blood gene expression signatures [7], have been evaluated for their ability to predict progression from LTBI to active disease. However, to date, no reliable method with suitable clinical applicability is available. Given that the outcome of microbial infection depends on the body's defense mechanisms, candidate prognostic biomarkers are likely to be associated with host immune components. The protective immune response to Mtb infection is currently considered to be mediated mainly by T helper (Th) 1-type CD4+ T cells via the secretion of IFN-γ and tumor necrosis factor alpha (TNF-α), which, in turn, recruit monocytes and granulocytes and promote their antimicrobial activities [8]. High incidence of TB in HIV-infected patients, IFN-γ receptor mutant individuals, or anti-TNF-α-treated patients with rheumatoid arthritis provides strong support to this theory [9–11]. It has been revealed recently that another functional subset of CD4+ T cells—so-called interleukin (IL)-17-producing Th17 cells—also participates in protective immunity against Mtb infection [12]. Furthermore, IL-4producing Th2 cells, IL-10-producing Tr1 cells, or Foxp3+ regulatory T cells (Treg) are also involved in the immune response to Mtb infection, either by inhibiting the antimicrobial activity of Th1 cells or regulating the excessive destructive reaction of Th1 cells [13–15]. In addition to their functional differentiation, CD4+ T cell activity also depends on their functional integrity. When T cells are stimulated repeatedly by chronic viral infection, they enter a state of dysfunction called senescence and/or exhaustion [16,17]. As TB is also considered a chronic microbial infection, it is likely that such chronic infection leads eventually to the functional impairment of Mtb-reactive CD4+ T cells, resulting in a failure to control Mtb infection. Based on these considerations, as an initial step in the development of prognostic biomarkers for Mtb infection, we investigated whether there are any differences in the phenotypes and functional signatures of Mtb-specific CD4+ T cells that allow a distinction between LTBI and ATB. The antigen (Ag)-specific CD4+ T cell population was defined as dividing cells upon their encountering with their cognate antigens. Our results show that the increased frequency of CD57-expressing cells in the Agreactive CD4+ T cell population is a differential biomarker for ATB from LTBI.
N C O
44
U
43
J.Y. Lee et al.
108 109 110 111 112 113 114 115 116 117 118 119 120 121
123 124 125 126
2.3. Isolation and stimulation culture of cells
127
Peripheral blood was collected in heparin tubes, and peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation using Ficoll-Paque solution (GE Healthcare, Uppsala, Sweden). CD14+ monocytes were purified by positive selection, using CD14 magnetic beads
128
Table 1 Demographic and clinical characteristics of the study population.
t1:1 Q2 t1:2
Total number Age (years) Sex
Mean Range Male Female
Clinical characteristics Acid-fast bacilli on microscopy (%) Mtb culture positive (%) Radiologic evidence of pulmonary TB (%) Median length of medication TST-positive subjects IGRA-positive subjects
129 130 131 132
ATB
LTBI
t1:3
24 48 28–75 15 9
24 48 23–59 14 10
18/23 (75)
n/a
t1:4 t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11
23/23 (95.8) 24/24 (100)
n/a n/a
t1:12 t1:13
2.96 days
n/a
t1:14
n/a n/a
15 9
t1:15 t1:16
Data represent the number of patients or subjects, unless otherwise indicated. ATB, active tuberculosis; LTBI, latent tuberculosis infection; Mtb, Mycobacterium tuberculosis; TB, tuberculosis; TST, tuberculin skin test; IGRA, IFN-γ release assay; n/a, not applicable.
Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
t1:17 t1:18 t1:19 t1:20 t1:21
Differential expression of CD57 in antigen-reactive CD4+ T cells between active and LTBI
186
2.5. Statistical analysis
187
Statistical analysis was performed using Prism V5.04 software (GraphPad, San Diego, USA). The nonparametric Mann–Whitney U test was used to compare the two groups. A receiver
146 147 148 149 150 151 152 153 154 155 156
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184
188 189
3. Results
194
3.1. Mycobacterial antigen-specific CD4+ T cell population
195
To identify the immune response parameters that allow differentiation between patients with active TB and those with latent infection, we analyzed the phenotypic characteristics and functional differentiation status of Mtb-reactive CD4+ T cells that are known to play a major role in protective and/or pathogenic reactions to the microbial agent. Ag-specific CD4+ T cells were defined as the dividing cells upon encountering with their cognate antigens. When CFSE-labeled CD4+ T cells were cultured with autologous monocytes, a negligible number of dividing CD4+ T cells were observed in the ‘nonstimulated’ culture wells, compared to a markedly higher number in those wells nonspecifically stimulated with anti-CD3/anti-CD28 mAbs (Fig. 1A). Moreover, a distinct population of dividing cells was detected in those cultures stimulated with mycobacterial Ags. There was no difference in the mean percentage of dividing CD4+ T cells between the LTBI and ATB groups (22.7% ± 11.8% vs 18.2% ± 13.6%; mean ± SD) in response to Mtb-specific Ags ESAT-6 and CFP-10. In contrast, the mean percentages were statistically higher in the LTBI group (44.7% ± 17.1%), compared to the ATB group (32.9% ± 17.2%), when CD4+ T cells were stimulated with PPD (Fig. 1B). However, as the overlap in individual values of the percentage cell division between these two groups was substantial, it was not possible to assign the patients' clinical status correctly using this parameter.
197
3.2. Functional differentiation of Mtb-specific CD4+ T cells
221
CD4+ Th cells are classified into several functional subsets, depending on the type of cytokine produced, such as IFN-γ-producing Th1, IL-17-producing Th17, IL-4-producing Th2, and IL-10-producing Tr1 [18]. Another subset that exhibits an immunosuppressive function—the Foxp3+ Treg subset—is defined by the expression of Foxp3 in the cell nucleus [19]. As the functional differentiation status of memory/effector T cells could considerably affect the outcome of an immune response to mycobacterial infection, we investigated whether there was any difference in CD4+ T cell subset differentiation between LTBI and ATB. Intracellular cytokine and Foxp3 staining was performed after CD4+ T cell stimulation cultures and analyzed on viable and divided cell fractions (Figs. 2A, B). Of the CD4+ T cells which had undergone cell division by stimulation with anti-CD3 and anti-CD28, approximately 20% was identified as an IFN-γ-producing Th1 subset: whereas upon stimulation with the mycobacterial Ags, ESAT-6/CFP-10 or PPD, the majority (almost 80%) of dividing cells were a Th1 subset. There was no difference, however, in the proportion of the Th1 subsets between the LTBI and ATB groups, irrespective of the stimulating agent used (Fig. 2C). All other functional
223
T
145
C
144
E
143
R
142
R
141
N C O
140
U
139
F
185
For phenotypic analyses, cultured cells were harvested and stained for 30 min on ice with an optimized antibody cocktail: CD4-eFlour780 (RPA-T4; eBioscience), CD196-PE (R6H1; eBioscience), CD57-APC (NK-1; BD Biosciences), CD183-PerCP/Cy5.5 (1C6/CXCR3; BD Biosciences), and PD1-BV421 (EH12.2H7; Biolegend, San Diego, CA). In some experiments, CD196-PE was replaced with Tim-3-PE (344823; R&D Systems, Minneapolis, MN). For the identification of Foxp3+ Treg cells, cultured cells were surface-stained with anti-CD4-eFlour780 and anti-CD25-APC (M-A251; BD Biosciences). After washing, intracellular Foxp3 was detected with a phycoerythrin (PE)-conjugated anti-human Foxp3 staining set (eBioscience), according to the manufacturer's instructions. For intracellular cytokine staining, first, the cells were surface-stained with anti-CD4-eFlour780. After fixation and permeabilization, using Intracellular Fixation & Permeabilization Buffer (eBioscience), the cells were incubated with the following antibody cocktail, prepared in the permeabilization buffer, for 30 min: IFN-γ-eFlour450 (4S.B3), IL-17A-PerCP-Cy5.5 (eBio64DEC17), IL-4-PE (8D4-8), and IL-10-eFlour660 (JES3-9D7; all from eBioscience). After washing with permeabilization buffer, cells were resuspended in PBS containing 1% fetal bovine serum (FBS; Invitrogen). Isotype-controlled staining was performed for all the staining procedures. The stained cells were analyzed using a FACSverse flow cytometer and FACSuite software V1.0.5 (Becton Dickinson, San Jose, CA).
138
O
159
137
R O
2.4. Flow cytometry
136
190
P
158
135
operating characteristic (ROC) curve defined the sensitivity and specificity of the diagnostic approach. Correlation analysis was carried out using the Spearman's test. Differences were considered statistically significant at P ≤ 0.05.
D
157
(Miltenyi Biotec, Bergisch Gladbach, Germany), according to the manufacturer's instructions. From the remaining PBMCs, following CD14+ monocyte isolation, CD4+ T cells were purified by negative selection using a magnetic activated cell sorting (MACS) CD4+ T cell isolation kit (Miltenyi Biotec). Cells were cultured in RPMI 1640 medium, supplemented with 2 mM glutamine, 1% (v/v) non-essential amino acids, 1% (v/v) sodium pyruvate, 1% (v/v) HEPES, penicillin (50 U/mL), streptomycin (50 μg/mL) (medium and culture supplements from Invitrogen, Grand Island, NY), and 10% (v/v) human serum (Biowest, Nuaille, France). Monocytes were pre-incubated for 3 h with ESAT-6/CFP-10 (1 μg/mL of each), PPD (10 μg/mL), anti-CD3/anti-CD28 monoclonal antibodies (mAbs) (1 μg/mL of each; both from BD Biosciences, San Diego, CA), or without any stimulus. CD4+ T cells were labeled with carboxyfluorescein succinimidyl ester (CFSE) (Molecular Probes, Leiden, The Netherlands) by incubating for 5 min at 37 °C in phosphate-buffered saline (PBS) containing 5 μM of CFSE, according to standard protocols, and then co-cultured with autologous monocytes, treated as above, at a ratio of 4:1 for 5 days. For intracellular cytokine staining experiments, cells were further stimulated for 5 h with phorbol 12-myristate 13-acetate (PMA) and ionomycin (Cell Stimulation Cocktail) in the presence of brefeldin A (all from eBioscience, San Diego, CA) for the final 4 h of culture.
134
E
133
3
Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
191 192 Q5 193
196
198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220
222
224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
4
J.Y. Lee et al.
A
LTBI 0.11
None
B
ATB
ESAT-6 / CFP-10
0.25
PPD
Anti-CD3 & -CD28
64.69
PPD
Anti-CD3 & -CD28
33.73
76.80
75.08
*
80 60 40 20
F
37.08
31.29
0
Count
LTBI ATB
LTBI ATB
LTBI ATB
R O
CFSE
O
ESAT-6 / CFP-10
% Divided CD4+ T cells
100
251 252 253 254
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276
3.3. Exhaustion/senescence phenotype of Mtb-specific CD4+ T cells +
We examined the functional integrity of Mtb-specific CD4 T cells by measuring CD57 and PD-1 expression on cell membranes. CD57 and PD-1 are known to be representative senescence and exhaustion markers in human T cells, respectively [17,20]. Ag-stimulated CD4+ T cells were surface-stained with fluorescence-labeled mAbs against CD57 and PD-1. Fig. 3A illustrates the gating and analyzing strategy used to identify the functional integrity of Ag-reactive CD4+ T cells. Interestingly, the percentage of CD57-expressing cells was significantly higher in the ATB, compared with LTBI, group, when cells were stimulated with ESAT-6/CFP-10 and PPD (Fig. 3B). In contrast, the percentage of PD-1-expressing cells was lower in the ATB, compared with LTBI, group (Fig. 3B). The latter finding was unexpected, considering that PD-1 is an exhaustion marker of CD8+ T cells associated with chronic viral infection [17]. Therefore, we examined the expression of Tim-3, another exhaustion marker [21], on the same cultured cells. As shown in Supplementary Fig. 1, the majority of Ag-reactive CD4+ T cells expressed Tim-3, but the percentage of Tim-3+ cells did not differ between the LTBI and ATB groups.
These data suggest that PD-1 might not be an exhaustion marker in human T cells that is specific to Mtb infection. Due to the great difference in the proportions of CD57- and PD-1-expressing cells between the two groups, we used ROC curve analysis to evaluate the differentiating power of these two markers. Compared to CD57-expressing cells in response to PPD (AUC: 0.7917), and PD-1-expressing cells in response to ESAT-6/CFP-10 (AUC: 0.7659), the frequency of CD57expressing cells in response to ESAT-6/CFP-10 was associated with superior differentiating power to differentiate LTBI from ATB (AUC: 0.9023) (Fig. 3C). Given that a higher and lower percentage of Ag-specific CD4+ T cells in ATB expressed CD57 and PD-1, respectively, compared with LTBI, we next addressed the potential of a combined expression pattern of CD57 and PD-1 on these cells as a clinical biomarker. From analyses of the frequency of CD57+PD-1− cells among proliferating CD4+ T cells, induced by stimulation with ESAT-6/CFP-10, we found a slightly greater difference between the LTBI and ATB groups (P ≤ 0.0001), compared to the difference observed in CD57− or PD-1-single positive cells (Fig. 4). Correspondingly, a marginally better differentiating ROC curve was generated for CD57+PD-1− cell frequency in response to ESAT-6/CFP-10 (AUC: 0.9182). Thus, these data demonstrate that the expression pattern of CD57 alone, or combined with PD-1, on Ag-driven dividing CD4+ T cells, effectively differentiates ATB from LTBI.
277
4. Discussion
304
Identification of biomarkers differentiating active from latent TB disease could facilitate the development of diagnostic tests that can predict the progression of latent infection to active disease. By comparing phenotypes and
305
E
T
C
250
E
249
R
248
R
247
subsets were detected, each representing less than 10% of Ag-reactive CD4+ T cells (Fig. 2C). Although there was no difference between LTBI and ATB in the proportion of Th17, Th2, Tr1, and Treg subsets upon stimulation with ESAT-6/CFP-10, the proportion of these subsets was found to be higher in ATB, compared to LTBI, upon stimulation with PPD. Nevertheless, it was again impossible to assign the patients' clinical status correctly using these parameters, because there was a large overlap in individuals, with extremely low percent values between these two groups.
N C O
246
U
245
D
P
Figure 1 Detection of Mycobacterium tuberculosis-specific CD4+ T cell population. CFSE-labeled CD4+ T cells were stimulated with mycobacterial antigens or anti-CD3/anti-CD28 mAbs for 5 days, and the percentages of divided cells were evaluated by flow cytometry. Representative histograms of responding CD4+ T cells from one latent tuberculosis infection (LTBI) subject and one active tuberculosis (ATB) patient are shown in (A). The percentages of divided CD4+ T cells, in response to stimulation with mycobacterial antigens are shown in (B). Each dot represents the values obtained from individual subjects, and horizontal bars indicate the average values. Groups were compared, using the Mann–Whitney U test. * P b 0.05.
Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 Q6 294 295 296 297 Q7 298 299 300 301 302 303
306 307 308
Differential expression of CD57 in antigen-reactive CD4+ T cells between active and LTBI
CFSE
Anti-CD3 & CD28
ESAT-6 / CFP-10
PPD
IFN-γ
Anti-CD3 & CD28
O P
IL-10
CD25
R O
IL-17
Foxp3
PPD
CFSE
FSC-A
F
FSC-A
ESAT-6 / CFP-10
CD4
CD4
SSC-A
B SSC-A
A
5
C
D
IL-4
IFN-γ (Th1) 30
C
60 20
E
40 10
LTBI ATB
IL-10 (Tr1)
20 10 5 1
N C O
15
LTBI ATB
R
LTBI ATB
**
0
U
LTBI ATB
ESAT-6 / CFP-10
LTBI ATB
LTBI ATB
PPD
Anti-CD3 & CD28
LTBI ATB
**
20 15 10
*
**
5
0
0
% Positive cells
R
20
25
T
80
IL-4 (Th2)
E
IL-17 (Th17) 40
100
0 LTBI ATB
LTBI ATB
LTBI ATB
LTBI ATB
LTBI ATB
Foxp3 (Treg)
25 20 15 10
*
5 0 LTBI ATB
LTBI ATB
LTBI ATB
Figure 2 Functional differentiation of Mycobacterium tuberculosis-specific CD4+ T cells. (A) Gating strategy and representative intracellular cytokine staining profiles of divided CD4+ T cell populations for the determination of the frequency of functional CD4+ T cell subsets. CD4+ T cells were stimulated for 5 days, as described in Fig. 1, and then re-stimulated with PMA and ionomycin for further 5 h, in the presence of brefeldin A. (B) Gating strategy and representative intracellular Foxp3 staining profiles of divided CD4+ T cell population for the determination of Foxp3+ Treg cell frequency. CD4+ T cells stimulated for 5 days were stained intracellularly with anti-Foxp3 monoclonal antibody without re-stimulation. Viable cells were gated, based on FSC and SSC profiles, and then CD4+ T cells with CFSE dilution were gated further. (C) Percentages of Th1, Th17, Th2, Tr1, and Foxp3+ Treg cells in divided CD4+ T cell populations from LTBI subjects (n = 20 for ESAT/CFP-10 stimulation; n = 24 for PPD and anti-CD3/anti-CD28 stimulation) and ATB patients (n = 22 for ESAT/CFP-10 stimulation; n = 23 for PPD and anti-CD3/anti-CD28 stimulation). Groups were compared, using the Mann–Whitney U test. * P b 0.05, ** P b 0.005. Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
6
J.Y. Lee et al. ESAT-6 / CFP-10
PPD
Anti-CD3 & CD28
PD-1
A
CD57
C
****
CD57 : ESAT-6/CFP-10
p =
available at www.sciencedirect.com
Clinical Immunology www.elsevier.com/locate/yclim
4Q4 5
F
O
3
R O
2
Differential expression of CD57 in antigen-reactive CD4 + T cells between active and latent tuberculosis infection
Ji Yeon Lee a , Ina Jeong a , Joon-Sung Joh a , Young Won Jung b , Soo Yeon Sim c , Boram Choi c , Hyeon-Gun Jee c , Dong-Gyun Lim c,⁎
6
a
7
b
Division of Respiratory Diseases, Department of Internal Medicine, National Medical Center, Seoul 100-799, South Korea Jung-gu Community Health Center, Seoul 132-713, South Korea c Center for Chronic Diseases, Research Institute, National Medical Center, Seoul 100-799, South Korea
D
8
P
1Q3
10
E
9
Received 16 December 2014; accepted with revision 8 April 2015
17 22
Abstract The development of diagnostic tests that predict the progression of latent tuberculosis infection to active disease is pivotal for the eradication of tuberculosis. As an initial step to achieve this goal, our study's aim was to identify biomarkers that differentiate active from latent tuberculosis infection. We compared active and latent tuberculosis infection groups in terms of the precursor frequency, functional subset differentiation, and senescence/exhaustion surface marker expression of antigen-specific CD4+ T cells, which were defined as dividing cells upon their encountering with Mycobacterium (M.) tuberculosis antigens. Among several parameters shown to have statistically significant differences between the two groups, the frequency of CD57-expressing cells could differentiate effectively between active disease and latent infection. Our results suggest that the expression of CD57 in M. tuberculosis-reactive CD4+ T cells could be a promising candidate biomarker with which to identify individuals with latent tuberculosis infection prone to progression to active disease. © 2015 Published by Elsevier Inc.
C
15 20 16 21
KEYWORDS Tuberculosis; CD57; CD4+ T cell; Biomarker
E
13 18 14 19
T
11
R
23 24
R
25 26 27
O
28 29
C
30 31
34
U
33
N
32
Abbreviations: ATB, active tuberculosis; AUC, area under the curve; CFSE, carboxyfluorescein succinimidyl ester; IGRA, interferon gamma release assay; LTBI, latent tuberculosis infection; Mtb, Mycobacterium tuberculosis; ROC, receiver operating characteristic; TB, tuberculosis. ⁎ Corresponding author at: Center for Chronic Diseases, Research Institute, National Medical Center, 245, Euljiro, Jung-gu, Seoul 100-799, South Korea. Fax: +82 2 2276 2319. E-mail address: [email protected] (D.-G. Lim).
1. Introduction
35
Tuberculosis (TB) remains a highly prevalent and lifethreatening disease to humans. When infected with Mycobacterium tuberculosis (Mtb), only a small proportion of the infected population develop primary active tuberculosis (ATB) disease, while over 90% carry the infection in a latent or subclinical stage [1]. It is estimated that the risk of active disease is approximately 5% in the first 18 months following
36
http://dx.doi.org/10.1016/j.clim.2015.04.011 1521-6616 © 2015 Published by Elsevier Inc. Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
37 38 39 40 41 42
2
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104
107
2.2. Mycobacterial antigens
122
F
55
A total of 24 patients with active pulmonary TB were recruited at either the National Medical Center or Jung-gu Community Health Center (Seoul, Korea) (Table 1). These patients were diagnosed, based on either a positive culture for Mtb or positive acid-fast bacilli in the sputum, or both, and their blood samples were obtained before or within the first 2 weeks of anti-TB medication. Moreover, a second study group of 24 healthy subjects with LTBI, which was defined by a positive tuberculin skin test or IGRA, in the absence of clinical symptoms, were recruited. HIV infection was negative in all tested individuals. Age and gender data of the study groups are shown in Table 1. The study was approved by the Ethics Committee of National Medical Center (IRB no. H-1208/022-002; 23 August 2012), and written informed consent was obtained from all participants.
O
54
106
R O
53
2.1. Study population
P
52
105
Recombinant ESAT-6 and CFP-10 antigens were provided by Dr. Sang-Nae Cho (International Tuberculosis Research Center, Seoul, Korea). PPD (RT50) was obtained from Statens Serum Institute (Copenhagen, Denmark).
D
51
2. Materials and methods
E
50
T
49
C
48
E
47
R
46
R
45
initial infection, and 5% for the remaining lifetime [2]. Due to the large number of infected individuals in the latent stage worldwide, reactivation of latent infection is a significant contributing factor to the prevalence rate of TB, especially in developed countries. Thus, in addition to the treatment of active cases, prophylactic treatment of latent infection represents an important option for the eradication of TB. Latent infection is known to be cleared effectively with standard drug regimens [3]. However, at least 20 patients with latent TB infection (LTBI) need to be administered with chemotherapeutic agents, which carry side effects, for each potential LTBI patient who progresses to ATB and is cured, which is neither cost-effective nor efficient in terms of a public health strategy to tackle TB. Therefore, an availability of prognostic biomarkers stratifying the risk of progression from LTBI to ATB would allow the targeting of treatment to those most at risk. So far, several measures, such as interferon gamma (IFN-γ) release assay (IGRA) [4], plasma cytokine levels [5,6], and even whole-blood gene expression signatures [7], have been evaluated for their ability to predict progression from LTBI to active disease. However, to date, no reliable method with suitable clinical applicability is available. Given that the outcome of microbial infection depends on the body's defense mechanisms, candidate prognostic biomarkers are likely to be associated with host immune components. The protective immune response to Mtb infection is currently considered to be mediated mainly by T helper (Th) 1-type CD4+ T cells via the secretion of IFN-γ and tumor necrosis factor alpha (TNF-α), which, in turn, recruit monocytes and granulocytes and promote their antimicrobial activities [8]. High incidence of TB in HIV-infected patients, IFN-γ receptor mutant individuals, or anti-TNF-α-treated patients with rheumatoid arthritis provides strong support to this theory [9–11]. It has been revealed recently that another functional subset of CD4+ T cells—so-called interleukin (IL)-17-producing Th17 cells—also participates in protective immunity against Mtb infection [12]. Furthermore, IL-4producing Th2 cells, IL-10-producing Tr1 cells, or Foxp3+ regulatory T cells (Treg) are also involved in the immune response to Mtb infection, either by inhibiting the antimicrobial activity of Th1 cells or regulating the excessive destructive reaction of Th1 cells [13–15]. In addition to their functional differentiation, CD4+ T cell activity also depends on their functional integrity. When T cells are stimulated repeatedly by chronic viral infection, they enter a state of dysfunction called senescence and/or exhaustion [16,17]. As TB is also considered a chronic microbial infection, it is likely that such chronic infection leads eventually to the functional impairment of Mtb-reactive CD4+ T cells, resulting in a failure to control Mtb infection. Based on these considerations, as an initial step in the development of prognostic biomarkers for Mtb infection, we investigated whether there are any differences in the phenotypes and functional signatures of Mtb-specific CD4+ T cells that allow a distinction between LTBI and ATB. The antigen (Ag)-specific CD4+ T cell population was defined as dividing cells upon their encountering with their cognate antigens. Our results show that the increased frequency of CD57-expressing cells in the Agreactive CD4+ T cell population is a differential biomarker for ATB from LTBI.
N C O
44
U
43
J.Y. Lee et al.
108 109 110 111 112 113 114 115 116 117 118 119 120 121
123 124 125 126
2.3. Isolation and stimulation culture of cells
127
Peripheral blood was collected in heparin tubes, and peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation using Ficoll-Paque solution (GE Healthcare, Uppsala, Sweden). CD14+ monocytes were purified by positive selection, using CD14 magnetic beads
128
Table 1 Demographic and clinical characteristics of the study population.
t1:1 Q2 t1:2
Total number Age (years) Sex
Mean Range Male Female
Clinical characteristics Acid-fast bacilli on microscopy (%) Mtb culture positive (%) Radiologic evidence of pulmonary TB (%) Median length of medication TST-positive subjects IGRA-positive subjects
129 130 131 132
ATB
LTBI
t1:3
24 48 28–75 15 9
24 48 23–59 14 10
18/23 (75)
n/a
t1:4 t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11
23/23 (95.8) 24/24 (100)
n/a n/a
t1:12 t1:13
2.96 days
n/a
t1:14
n/a n/a
15 9
t1:15 t1:16
Data represent the number of patients or subjects, unless otherwise indicated. ATB, active tuberculosis; LTBI, latent tuberculosis infection; Mtb, Mycobacterium tuberculosis; TB, tuberculosis; TST, tuberculin skin test; IGRA, IFN-γ release assay; n/a, not applicable.
Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
t1:17 t1:18 t1:19 t1:20 t1:21
Differential expression of CD57 in antigen-reactive CD4+ T cells between active and LTBI
186
2.5. Statistical analysis
187
Statistical analysis was performed using Prism V5.04 software (GraphPad, San Diego, USA). The nonparametric Mann–Whitney U test was used to compare the two groups. A receiver
146 147 148 149 150 151 152 153 154 155 156
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184
188 189
3. Results
194
3.1. Mycobacterial antigen-specific CD4+ T cell population
195
To identify the immune response parameters that allow differentiation between patients with active TB and those with latent infection, we analyzed the phenotypic characteristics and functional differentiation status of Mtb-reactive CD4+ T cells that are known to play a major role in protective and/or pathogenic reactions to the microbial agent. Ag-specific CD4+ T cells were defined as the dividing cells upon encountering with their cognate antigens. When CFSE-labeled CD4+ T cells were cultured with autologous monocytes, a negligible number of dividing CD4+ T cells were observed in the ‘nonstimulated’ culture wells, compared to a markedly higher number in those wells nonspecifically stimulated with anti-CD3/anti-CD28 mAbs (Fig. 1A). Moreover, a distinct population of dividing cells was detected in those cultures stimulated with mycobacterial Ags. There was no difference in the mean percentage of dividing CD4+ T cells between the LTBI and ATB groups (22.7% ± 11.8% vs 18.2% ± 13.6%; mean ± SD) in response to Mtb-specific Ags ESAT-6 and CFP-10. In contrast, the mean percentages were statistically higher in the LTBI group (44.7% ± 17.1%), compared to the ATB group (32.9% ± 17.2%), when CD4+ T cells were stimulated with PPD (Fig. 1B). However, as the overlap in individual values of the percentage cell division between these two groups was substantial, it was not possible to assign the patients' clinical status correctly using this parameter.
197
3.2. Functional differentiation of Mtb-specific CD4+ T cells
221
CD4+ Th cells are classified into several functional subsets, depending on the type of cytokine produced, such as IFN-γ-producing Th1, IL-17-producing Th17, IL-4-producing Th2, and IL-10-producing Tr1 [18]. Another subset that exhibits an immunosuppressive function—the Foxp3+ Treg subset—is defined by the expression of Foxp3 in the cell nucleus [19]. As the functional differentiation status of memory/effector T cells could considerably affect the outcome of an immune response to mycobacterial infection, we investigated whether there was any difference in CD4+ T cell subset differentiation between LTBI and ATB. Intracellular cytokine and Foxp3 staining was performed after CD4+ T cell stimulation cultures and analyzed on viable and divided cell fractions (Figs. 2A, B). Of the CD4+ T cells which had undergone cell division by stimulation with anti-CD3 and anti-CD28, approximately 20% was identified as an IFN-γ-producing Th1 subset: whereas upon stimulation with the mycobacterial Ags, ESAT-6/CFP-10 or PPD, the majority (almost 80%) of dividing cells were a Th1 subset. There was no difference, however, in the proportion of the Th1 subsets between the LTBI and ATB groups, irrespective of the stimulating agent used (Fig. 2C). All other functional
223
T
145
C
144
E
143
R
142
R
141
N C O
140
U
139
F
185
For phenotypic analyses, cultured cells were harvested and stained for 30 min on ice with an optimized antibody cocktail: CD4-eFlour780 (RPA-T4; eBioscience), CD196-PE (R6H1; eBioscience), CD57-APC (NK-1; BD Biosciences), CD183-PerCP/Cy5.5 (1C6/CXCR3; BD Biosciences), and PD1-BV421 (EH12.2H7; Biolegend, San Diego, CA). In some experiments, CD196-PE was replaced with Tim-3-PE (344823; R&D Systems, Minneapolis, MN). For the identification of Foxp3+ Treg cells, cultured cells were surface-stained with anti-CD4-eFlour780 and anti-CD25-APC (M-A251; BD Biosciences). After washing, intracellular Foxp3 was detected with a phycoerythrin (PE)-conjugated anti-human Foxp3 staining set (eBioscience), according to the manufacturer's instructions. For intracellular cytokine staining, first, the cells were surface-stained with anti-CD4-eFlour780. After fixation and permeabilization, using Intracellular Fixation & Permeabilization Buffer (eBioscience), the cells were incubated with the following antibody cocktail, prepared in the permeabilization buffer, for 30 min: IFN-γ-eFlour450 (4S.B3), IL-17A-PerCP-Cy5.5 (eBio64DEC17), IL-4-PE (8D4-8), and IL-10-eFlour660 (JES3-9D7; all from eBioscience). After washing with permeabilization buffer, cells were resuspended in PBS containing 1% fetal bovine serum (FBS; Invitrogen). Isotype-controlled staining was performed for all the staining procedures. The stained cells were analyzed using a FACSverse flow cytometer and FACSuite software V1.0.5 (Becton Dickinson, San Jose, CA).
138
O
159
137
R O
2.4. Flow cytometry
136
190
P
158
135
operating characteristic (ROC) curve defined the sensitivity and specificity of the diagnostic approach. Correlation analysis was carried out using the Spearman's test. Differences were considered statistically significant at P ≤ 0.05.
D
157
(Miltenyi Biotec, Bergisch Gladbach, Germany), according to the manufacturer's instructions. From the remaining PBMCs, following CD14+ monocyte isolation, CD4+ T cells were purified by negative selection using a magnetic activated cell sorting (MACS) CD4+ T cell isolation kit (Miltenyi Biotec). Cells were cultured in RPMI 1640 medium, supplemented with 2 mM glutamine, 1% (v/v) non-essential amino acids, 1% (v/v) sodium pyruvate, 1% (v/v) HEPES, penicillin (50 U/mL), streptomycin (50 μg/mL) (medium and culture supplements from Invitrogen, Grand Island, NY), and 10% (v/v) human serum (Biowest, Nuaille, France). Monocytes were pre-incubated for 3 h with ESAT-6/CFP-10 (1 μg/mL of each), PPD (10 μg/mL), anti-CD3/anti-CD28 monoclonal antibodies (mAbs) (1 μg/mL of each; both from BD Biosciences, San Diego, CA), or without any stimulus. CD4+ T cells were labeled with carboxyfluorescein succinimidyl ester (CFSE) (Molecular Probes, Leiden, The Netherlands) by incubating for 5 min at 37 °C in phosphate-buffered saline (PBS) containing 5 μM of CFSE, according to standard protocols, and then co-cultured with autologous monocytes, treated as above, at a ratio of 4:1 for 5 days. For intracellular cytokine staining experiments, cells were further stimulated for 5 h with phorbol 12-myristate 13-acetate (PMA) and ionomycin (Cell Stimulation Cocktail) in the presence of brefeldin A (all from eBioscience, San Diego, CA) for the final 4 h of culture.
134
E
133
3
Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
191 192 Q5 193
196
198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220
222
224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
4
J.Y. Lee et al.
A
LTBI 0.11
None
B
ATB
ESAT-6 / CFP-10
0.25
PPD
Anti-CD3 & -CD28
64.69
PPD
Anti-CD3 & -CD28
33.73
76.80
75.08
*
80 60 40 20
F
37.08
31.29
0
Count
LTBI ATB
LTBI ATB
LTBI ATB
R O
CFSE
O
ESAT-6 / CFP-10
% Divided CD4+ T cells
100
251 252 253 254
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276
3.3. Exhaustion/senescence phenotype of Mtb-specific CD4+ T cells +
We examined the functional integrity of Mtb-specific CD4 T cells by measuring CD57 and PD-1 expression on cell membranes. CD57 and PD-1 are known to be representative senescence and exhaustion markers in human T cells, respectively [17,20]. Ag-stimulated CD4+ T cells were surface-stained with fluorescence-labeled mAbs against CD57 and PD-1. Fig. 3A illustrates the gating and analyzing strategy used to identify the functional integrity of Ag-reactive CD4+ T cells. Interestingly, the percentage of CD57-expressing cells was significantly higher in the ATB, compared with LTBI, group, when cells were stimulated with ESAT-6/CFP-10 and PPD (Fig. 3B). In contrast, the percentage of PD-1-expressing cells was lower in the ATB, compared with LTBI, group (Fig. 3B). The latter finding was unexpected, considering that PD-1 is an exhaustion marker of CD8+ T cells associated with chronic viral infection [17]. Therefore, we examined the expression of Tim-3, another exhaustion marker [21], on the same cultured cells. As shown in Supplementary Fig. 1, the majority of Ag-reactive CD4+ T cells expressed Tim-3, but the percentage of Tim-3+ cells did not differ between the LTBI and ATB groups.
These data suggest that PD-1 might not be an exhaustion marker in human T cells that is specific to Mtb infection. Due to the great difference in the proportions of CD57- and PD-1-expressing cells between the two groups, we used ROC curve analysis to evaluate the differentiating power of these two markers. Compared to CD57-expressing cells in response to PPD (AUC: 0.7917), and PD-1-expressing cells in response to ESAT-6/CFP-10 (AUC: 0.7659), the frequency of CD57expressing cells in response to ESAT-6/CFP-10 was associated with superior differentiating power to differentiate LTBI from ATB (AUC: 0.9023) (Fig. 3C). Given that a higher and lower percentage of Ag-specific CD4+ T cells in ATB expressed CD57 and PD-1, respectively, compared with LTBI, we next addressed the potential of a combined expression pattern of CD57 and PD-1 on these cells as a clinical biomarker. From analyses of the frequency of CD57+PD-1− cells among proliferating CD4+ T cells, induced by stimulation with ESAT-6/CFP-10, we found a slightly greater difference between the LTBI and ATB groups (P ≤ 0.0001), compared to the difference observed in CD57− or PD-1-single positive cells (Fig. 4). Correspondingly, a marginally better differentiating ROC curve was generated for CD57+PD-1− cell frequency in response to ESAT-6/CFP-10 (AUC: 0.9182). Thus, these data demonstrate that the expression pattern of CD57 alone, or combined with PD-1, on Ag-driven dividing CD4+ T cells, effectively differentiates ATB from LTBI.
277
4. Discussion
304
Identification of biomarkers differentiating active from latent TB disease could facilitate the development of diagnostic tests that can predict the progression of latent infection to active disease. By comparing phenotypes and
305
E
T
C
250
E
249
R
248
R
247
subsets were detected, each representing less than 10% of Ag-reactive CD4+ T cells (Fig. 2C). Although there was no difference between LTBI and ATB in the proportion of Th17, Th2, Tr1, and Treg subsets upon stimulation with ESAT-6/CFP-10, the proportion of these subsets was found to be higher in ATB, compared to LTBI, upon stimulation with PPD. Nevertheless, it was again impossible to assign the patients' clinical status correctly using these parameters, because there was a large overlap in individuals, with extremely low percent values between these two groups.
N C O
246
U
245
D
P
Figure 1 Detection of Mycobacterium tuberculosis-specific CD4+ T cell population. CFSE-labeled CD4+ T cells were stimulated with mycobacterial antigens or anti-CD3/anti-CD28 mAbs for 5 days, and the percentages of divided cells were evaluated by flow cytometry. Representative histograms of responding CD4+ T cells from one latent tuberculosis infection (LTBI) subject and one active tuberculosis (ATB) patient are shown in (A). The percentages of divided CD4+ T cells, in response to stimulation with mycobacterial antigens are shown in (B). Each dot represents the values obtained from individual subjects, and horizontal bars indicate the average values. Groups were compared, using the Mann–Whitney U test. * P b 0.05.
Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 Q6 294 295 296 297 Q7 298 299 300 301 302 303
306 307 308
Differential expression of CD57 in antigen-reactive CD4+ T cells between active and LTBI
CFSE
Anti-CD3 & CD28
ESAT-6 / CFP-10
PPD
IFN-γ
Anti-CD3 & CD28
O P
IL-10
CD25
R O
IL-17
Foxp3
PPD
CFSE
FSC-A
F
FSC-A
ESAT-6 / CFP-10
CD4
CD4
SSC-A
B SSC-A
A
5
C
D
IL-4
IFN-γ (Th1) 30
C
60 20
E
40 10
LTBI ATB
IL-10 (Tr1)
20 10 5 1
N C O
15
LTBI ATB
R
LTBI ATB
**
0
U
LTBI ATB
ESAT-6 / CFP-10
LTBI ATB
LTBI ATB
PPD
Anti-CD3 & CD28
LTBI ATB
**
20 15 10
*
**
5
0
0
% Positive cells
R
20
25
T
80
IL-4 (Th2)
E
IL-17 (Th17) 40
100
0 LTBI ATB
LTBI ATB
LTBI ATB
LTBI ATB
LTBI ATB
Foxp3 (Treg)
25 20 15 10
*
5 0 LTBI ATB
LTBI ATB
LTBI ATB
Figure 2 Functional differentiation of Mycobacterium tuberculosis-specific CD4+ T cells. (A) Gating strategy and representative intracellular cytokine staining profiles of divided CD4+ T cell populations for the determination of the frequency of functional CD4+ T cell subsets. CD4+ T cells were stimulated for 5 days, as described in Fig. 1, and then re-stimulated with PMA and ionomycin for further 5 h, in the presence of brefeldin A. (B) Gating strategy and representative intracellular Foxp3 staining profiles of divided CD4+ T cell population for the determination of Foxp3+ Treg cell frequency. CD4+ T cells stimulated for 5 days were stained intracellularly with anti-Foxp3 monoclonal antibody without re-stimulation. Viable cells were gated, based on FSC and SSC profiles, and then CD4+ T cells with CFSE dilution were gated further. (C) Percentages of Th1, Th17, Th2, Tr1, and Foxp3+ Treg cells in divided CD4+ T cell populations from LTBI subjects (n = 20 for ESAT/CFP-10 stimulation; n = 24 for PPD and anti-CD3/anti-CD28 stimulation) and ATB patients (n = 22 for ESAT/CFP-10 stimulation; n = 23 for PPD and anti-CD3/anti-CD28 stimulation). Groups were compared, using the Mann–Whitney U test. * P b 0.05, ** P b 0.005. Please cite this article as: J.Y. Lee, et al., Differential expression of CD57 in antigen-reactive CD4+ T cells between active and latent tuberculosis infection, Clin. Immunol. (2015), http://dx.doi.org/10.1016/j.clim.2015.04.011
6
J.Y. Lee et al. ESAT-6 / CFP-10
PPD
Anti-CD3 & CD28
PD-1
A
CD57
C
****
CD57 : ESAT-6/CFP-10
p =
