Rheumatol Int (2012) 32:2731–2736 DOI 10.1007/s00296-011-1984-x
ORIGINAL ARTICLE
Disturbed Th17/Treg balance in patients with rheumatoid arthritis Qian Niu • Bei Cai • Zhuo-chun Huang Yun-ying Shi • Lan-lan Wang
•
Received: 7 September 2010 / Accepted: 3 July 2011 / Published online: 2 August 2011 Ó Springer-Verlag 2011
Abstract Proinflammatory Th17 cells and CD4?CD25? regulatory T (Treg) cells are two newly identified T lymphocyte subsets, which have opposite effects on autoimmunity and inflammation. To assess the Th17/Treg pattern and cytokine microenvironment in peripheral blood of patients with RA, we included 66 RA patients and 20 healthy volunteers. Of all these subjects, peripheral Th17 and Treg frequencies were analyzed by flow cytometry (FCM) and the plasma levels of interleukin (IL)-17, 23, 6, tumor necrosis factor (TNF)-a, transforming growth factor (TGF)-b1 were detected by ELISA. The results demonstrated that RA patients revealed an obvious increase in peripheral Th17 frequencies and levels of Th17-related cytokines (IL-17, IL-23, IL-6, TNF-a) while a significant decrease in Treg frequencies and Treg-related cytokine (TGF-b1) levels when compared with healthy people. Our study indicated that development of RA is associated with peripheral Th17/Treg imbalance and characterized by a proinflammatory cytokine microenvironment, which supports continuing generation of Th17 cells. Keywords Rheumatoid arthritis Th17 cells Regulatory T cells Autoimmunity Inflammation Inflammatory cytokine microenvironment
Q. Niu B. Cai Z. Huang L. Wang (&) Department of Laboratory Medicine, West China Hospital, Sichuan University, No. 37, Guo Xue Xiang, Chengdu 610041, The People’s Republic of China e-mail:
[email protected] Y. Shi (&) Department of Nephrology, West China Hospital, Sichuan University, No. 37, Guo Xue Xiang, Chengdu 610041, The People’s Republic of China e-mail:
[email protected]
Introduction Rheumatoid arthritis (RA) is a chronic autoimmune disease with persistent inflammation of multiple synovial joints, which results in progressive tissue destruction of bone and cartilage [1]. The pathogenesis of this destructive disease was classically viewed as involving two hierarchical systems, governing inflammation and autoimmunity, respectively. The key players in this paradigm were the proinflammatory cytokine tumor necrosis factor (TNF)-a and the Th1 subset of helper T cells [2, 3], suggesting that the Th1/Th2 imbalance characterized by enhanced Th1 response played a crucial role in driving RA. Knowledge recently acquired in the field of immunopathology has challenged the classical paradigm for RA. Th17 cells and Foxp3?CD4?CD25? regulatory T cells are newly described as two distinct T cell subsets from Th1 and Th2 cells, having opposite effects on autoimmunity and inflammation. Th17 cells is a key effector in the immune response and play critical roles in the development of autoimmunity by producing IL-17 and, to a lesser extent, TNF-a and IL-6 [4], while Treg cells orchestrate the overall immune response and play a role in maintaining peripheral immune tolerance by regulating the activity of the effector T cells [5]. Therefore, the Th17/Treg balance may control the development of autoimmunity and inflammation. In the present study, we aim to evaluate the Th17/Treg balance pattern in patients with RA.
Methods Patients This study was approved by the Ethics Committee of the Chinese Human Genome and the Ethics Committee of
123
2732
Rheumatol Int (2012) 32:2731–2736
Table 1 Demographics of RA patients and healthy controls Group
n
Male/female
Age (years)
P value [0.05
Active RA
36
30/6
48 ± 12
Inactive RA
30
25/5
44 ± 9
Healthy control
20
17/3
49 ± 10
Values are expressed as mean ± SD or number
West China Hospital, and written informed consent was obtained from all participants. A total of 66 patients diagnosed as RA based on the 1987 criteria of the American College of Rheumatology [6] and 20 healthy controls were included. Disease activity was assessed with the 28-joint disease activity score (DAS28) [7]. None of the patient had advanced liver diseases, renal failure, malignant diseases, infectious diseases or other inflammatory diseases (such as septicemia, pneumonia.). Demographic features of all the patients and healthy volunteers were provided in Table 1. Blood samples For all patients and control subjects, a 3-ml fasting blood sample was drawn into a BD Vacutainer tube containing sodium heparin at 8:00–9:00 a.m. Whole blood was used for flow cytometric analysis. Plasma was obtained after centrifugation and stored at -80°C for the measurement of the cytokines. Flow cytometric analysis of Th17 and Treg cells Cell preparation For analysis of Th17, 500 ll whole blood of every sample was cultured in complete culture medium (RPMI 1640 supplemented with 10% heat-inactivated fetal calf serum) for 5 h, in the presence of phorbol myristate acetate (PMA, 50 ng/ml) plus ionomycin (1 lg/ml) and monensin (1 lg/ ml). The incubators were set at 37°C under a 5% CO2 environment. For analysis of Treg, 50 ll whole blood of every sample was aliquoted into a tube for further staining. Surface and intracellular staining For Th17 analysis, 50 ll stimulated whole blood was incubated with PerCP-labeled anti-human CD3 (SK7) (BD Bioscience, San Diego, USA) and fluorescein isothiocyanate (FITC)-labeled anti-human CD8 (SK2) (BD Bioscience) at 4°C for 30 min. (Because CD4 positive cells would turn into CD4 negative cells after stimulation with PMA due to the endocytosis of CD4 molecules caused by PMA, so we used anti-human CD3 and CD8 for the surface staining of Th17
123
and analyzed CD3?CD8- cells instead of CD3?CD4? cells.) For Treg analysis, 50 ll whole blood without stimulation was incubated with FITC/allophycoerythrin (APC)labeled anti-human CD4/CD25 cocktail (RPA-T4/BC96) (eBioscience, San Diego, California, USA) at 4°C for 30 min. After the surface staining, the whole blood were stained with phycoerythrin (PE)-labeled anti-human IL-17A (SCPL1362) (eBioscience) for Th17 detection or PE-labeled anti-human Foxp3 (PCH101) (eBioscience) for Treg detection after fixation and permeabilization according to the manufacturer’s instructions (eBioscience). Isotype controls were given to enable correct compensation and confirm antibody specificity. Stained cells were run on a FACSCanto cytometer (BD Bioscience), and the data were analyzed using FACSDiva software (BD Bioscience). ELISA detection of plasma IL-17, IL-23, IL-6, TNF-a and TGF-b1 The plasma concentrations of IL-17, IL-23, IL-6, TNF-a and TGF-b1 were measured by enzyme-linked immunosorbent assay (ELISA), following the manufacturer’s instructions (all ELISA kits from Bender MedSystems, Burlingame, USA). All samples were measured in duplicate. Statistical analysis Values are expressed as mean ± SD or median (range) in the tables and figures. Data were analyzed using SPSS 16.0 software (Chicago, IL). Firstly, the data were analyzed by ANOVA or Kruskal–Wallis H test. If significance was found, then Student–Newman–Keuls or Mann–Whitney test was performed to detect the difference among groups. Spearman’s correlation was used as a test of correlation between two continuous variables. P Values less than 0.05 were considered significant.
Results Basic clinical characteristics of patients There was no significant difference in age and gender among RA patients and healthy controls (Table 1). Frequencies of circulating Th17 in patients with RA As shown in Fig. 1, the frequencies of Th17 (CD8-IL17?/ CD3?CD8- T) cells were remarkably higher in patients with active RA (2.31 ± 1.38%) and inactive RA (1.91 ± 1.18%) than that of healthy subjects (0.71 ± 0.45%) (P \ 0.01), while there was no obvious difference between two RA groups (P [ 0.05).
Rheumatol Int (2012) 32:2731–2736
2733
Fig. 1 Frequencies of circulating Th17 (CD8-IL17?) cells in patients with RA. Whole blood from RA patients and healthy subjects were stimulated with PMA, ionomycin and monensin for 5 h, and then stained with labeled anti-human antibodies as described in
methods. a Representative FACS pictures from a single subject in each group. b Collective analysis of results from all three groups. *P \ 0.01 versus healthy control
Table 2 Frequencies of circulating Foxp3?CD4?CD25? Treg, Foxp3-CD4?CD25? Teff cells in RA patients and healthy subjects Group
n
Foxp3?CD4?CD25? Treg cells (%) (1)
Foxp3-CD4?CD25? Teff cells (%) (2)
(2)/(1) ratio
Active RA
36
1.54 ± 0.23*
24.59 ± 10.86*
39.99 ± 17.94*
Inactive RA
30
1.59 ± 0.24*
20.05 ± 8.14
21.02 ± 4.64
Healthy control
20
4.33 ± 0.35
16.02 ± 6.66
4.95 ± 0.66
Values are expressed as mean ± SD. * P \ 0.01 versus healthy control
Fig. 2 Frequencies of circulating Tregs in patients with RA. Whole blood from RA patients and healthy subjects were stained with labeled anti-human antibodies as described in methods. a CD4?CD25? T subsets were gated by flow cytometry. Plots in
intern box represented CD4?CD25? T cells. b Representative pictures of Foxp3 expression in CD4?CD25? T subsets from a single subject in each group
Frequencies of circulating Treg and Teff in patients with RA
all these differences between two RA groups were not statistically significant (P [ 0.05) (Table 2, Fig. 2).
The frequencies of Treg (Foxp3?CD4?CD25?/CD4? T cells) in both active RA (1.54 ± 0.23%) and inactive RA groups (1.59 ± 0.24%) decreased significantly when compared with healthy control group (4.33 ± 0.35%) (P \ 0.01), while the frequencies of Teff (Foxp3-CD4? CD25?/CD4? T cells) and the Teff/Treg ratios were all significantly higher in active RA group (24.59 ± 10.86% and 39.99 ± 7.94%) than those in inactive RA (20.05 ± 8.14% and 21.02 ± 4.64%) and healthy control groups (16.02 ± 6.66% and 4.95 ± 0.66%) (P \ 0.01). However,
Plasma concentrations of cytokines in patients with RA As shown in Table 3, the plasma concentrations of IL-17, IL-23, IL-6 and TNF-a in patients with active RA [IL-17: 2.85(1.84-4.91) pg/ml; IL-23: 54.33(49.49–65.37) pg/ml; IL-6: 32.98(8.90–176.88) pg/ml; TNF-a: 1,158.10(338.63– 2,303.12) pg/ml] and inactive RA [IL-17: 2.69(1.91–5.11) pg/ml; IL-23: 65.84(51.88–76.69) pg/ml; IL-6: 14.59(3.83– 62.44) pg/ml; TNF-a: 1,417.81(707.15–2,289.46) pg/ml] increased significantly when compared with healthy
123
2734
Rheumatol Int (2012) 32:2731–2736
Table 3 Plasma concentrations of cytokines in RA patients and healthy subjects Group
n
IL-17 (pg/ml)
IL-23 (pg/ml)
IL-6 (pg/ml)
TNF-a (pg/ml)
Active RA
36
2.85 (1.84–4.91)*
54.33 (49.49–65.37)*
32.98 (8.90–176.88)*
1,158.10 (338.63–2,303.12)*
Inactive RA
30
2.69 (1.91–5.11)*
65.84 (51.88–76.69)*
14.59 (3.83–62.44)*
1,417.81 (707.15–2,289.46)*
Health control
20
1.43 (0.96–2.57)
36.07 (32.99–48.82)
1.55 (1.27–1.78)
Values are expressed as median (range). * P \ 0.001 versus healthy control;
#
13.50 (9.73–19.08)
TGF-b1 (ng/ml) 4.36 (2.70–7.09)*, # 6.90 (4.78–8.70)* 24.19 (16.58–27.07)
P \ 0.05 versus inactive RA
Fig. 3 Spearman correlations between circulating Th17 cells and plasma cytokines. a Frequencies of circulating Th17 cells positively correlate with plasma IL-23 concentrations (r = 0.302, P \ 0.01). b Frequencies of circulating Th17 cells negatively correlate with plasma TGF-b1 concentrations (r = -0.436, P \ 0.001)
subjects [IL-17: 1.43(0.96–2.57) pg/ml; IL-23: 36.07 (32.99–48.82) pg/ml; IL-6: 1.55(1.27–1.78) pg/ml; TNF-a: 13.50(9.73–19.08) pg/ml] (P \ 0.001), while there was no obvious difference between two RA groups (P [ 0.05). Plasma concentrations of TGF-b1 in active RA [4.36 (2.70–7.09) ng/ml] and inactive RA groups [6.90(4.78– 8.70) ng/ml] decreased significantly when compared with healthy control group [24.19(16.58–27.07) ng/ml] (P \ 0.001), and the TGF-b1 concentrations in the active RA group was significantly lower than that in the inactive RA groups. In addition, frequencies of Th17 cells were positively correlated with plasma concentrations of IL-23 (r = 0.302, P \ 0.01, Fig. 3a) and negatively correlated with plasma concentrations of TGF-b1 (r = -0.436, P \ 0.001, Fig. 3b).
Discussion Although the Th1/Th2 imbalance paradigm allowed researchers to achieve breakthroughs in the treatment of RA, it fails to explain some incontrovertible facts, such as the paradoxical effects of IFN-c, which alleviates inflammation in murine models of arthritis but promotes arthritis flares in healthy mice, and the lack of efficacy of monoclonal antibodies to IFN-c in most patients with RA [8, 9]. Recent knowledge about the pathogenesis of RA has led to a new paradigm for this destructive disease, in which the Th17/Treg imbalance is the focus of concern in those studies. To explore the shift of Th17/Treg balance in detail
123
in patients with RA, we analyzed Th17/Treg on different levels including cell frequencies and related cytokines secretion. The results demonstrated that balance of peripheral Th17/Treg was disturbed in RA patients, and the Th17/Treg imbalance might act synergistically with microinflammation on the development of RA. Th17 cells, an exciting newcomer to the T helper (TH) cell field in the last few years, distinct from Th1 and Th2 cells, have been shown to play a crucial role in several autoimmune and inflammatory diseases, such as inflammatory bowel disease (IBD), childhood arthritis [juvenile idiopathic arthritis (JIA)] and acute coronary syndrome (ACS) [10–12]. Th17 cells are characterized by production of IL-17, an important inflammatory cytokine which exerts robust proinflammatory effect by, alone and also in combination with TNF-a and IL-1, inducing the synthesis of proinflammatory cytokines (such as IL-6 and IL-8) and chemokines [such as macrophage chemoattractant protein (MCP)-1 and macrophage inflammatory chemokine-2 (MIP-2)], which mediate tissue infiltration and inflammation [13]. IL-17 has been found at high levels in synovial fluid from RA patients [14], suggesting the potential role of IL-17 in the development of RA. The inflammatory symptoms of collagen induced arthritis (CIA), the animal model for RA, were alleviated when blocked the endogenous IL-17 [15]. In addition to IL-17, Th17 cells also produce other proinflammatory cytokines to a lesser extent, including IL-6 and TNF-a, which were two well-known inflammatory mediators and critical players in regulating tissue inflammation. Therefore, the roles of Th17 cells and
Rheumatol Int (2012) 32:2731–2736
their proinflammatory cytokines in mediating autoimmune and inflammation are worthy of deeper research. In this study, results demonstrated that patients with RA exhibited a significant increase in peripheral Th17 frequency as compared with healthy controls, and this frequency was higher in active RA group than that in inactive RA group, suggesting that Th17 cells may be the key player in the pathogenesis and progression of RA. Due to the increase of Th17 cells, plasma concentrations of Th17related cytokines (IL-17, IL-6 and TNF-a) increased accordingly, which may play crucial roles in the inflammatory process of RA. Upregulated IL-17 and TNF-a can exert synergistic effects in stimulating synovial fibroblasts and epithelial cells to secrete IL-6, IL-8, and PGE2, which contribute to the formation of local tissue inflammation, and also inducing the receptor activator of NF-jB ligand (RANKL), which regulate osteoclastogenesis and the bone resorption process [13, 16]. Contrary to Th17 cells, Foxp3?CD4?CD25? Treg cells have been identified as a key part of the immune system’s apparatus for controlling inflammatory processes [17]. Treg cells exert their regulatory function, in part, by secretion of anti-inflammatory cytokines including TGF-b1, and so they are important in the induction and maintenance of immune homeostasis. Defects in Treg function or reduced numbers have been put forward as a cause of failed tolerance in several human autoimmune diseases, including RA, multiple sclerosis (MS) and IBD [10, 18, 19]. In a series of murine experiments, ablation of Treg led to multiple organ autoimmunity, which could be reversed by adoptive transfer of these Foxp3?CD4?CD25? T cells [20, 21]. Our results verified that peripheral Foxp3?CD4?CD25? Treg frequency and Treg-related cytokine (TGF-b1) all significantly decreased in patients with RA, suggesting again that Treg cells and the anti-inflammatory cytokine TGF-b1 have a potentially protective effect in the progression and stability of RA. While, the frequencies of peripheral Foxp3-CD4?CD25? Teff cells increased markedly in active RA patients, which in combination with the decrease of Treg cells directly led to the increase of peripheral Teff/Treg ratios in active RA patients. Results suggested that the imbalance between immune regulators and responsive T cells might be the major cause of failed peripheral tolerance, which contributed to autoimmune inflammation in RA patients. Meanwhile, our data showed a reciprocal relationship between peripheral Th17 and Treg cells numbers, as well as between Th17 and Treg-related cytokines, indicating that the peripheral Th17/Treg balance was disturbed in RA patients and the numerical and functional imbalance of Th17/Treg might have a potential impact on the disease progression. Several studies have independently demonstrated that the development of Th17 and Treg cells may be
2735
interrelated, and reciprocally influenced by cytokines, such as TGF-b1, IL-6 and IL-23 [22–24]. TGF-b1 and IL-6 are sufficient for the differentiation of murine Th17 cells in contrast to TGF-b1 alone, which promotes the development of Treg cells that express Foxp3, a transcription factor important for maintaining the regulatory function of Treg [25]. However, human Th17 cells generation appears to be driven by IL-6 or IL-1 and enhanced by IL-23, but not TGF-b1 [26]. Therefore, the cytokine milieu of the local environment plays a pivotal role in the differentiation from naive CD4? T cells to Treg or Th17 cells. In the present study, results showed that the concentrations of IL-6 and IL-23, both of which promote the differentiation of Th17 cells, were all significantly higher in RA patients, and IL-23 concentrations were positively correlated with Th17 cells frequencies. While, TGF-b1 concentrations were obvious lower in RA patients and negatively correlated with Th17 cells frequencies. These suggested that the proinflammatory cytokine microenvironment, characterized by elevated IL-6 and IL-23 levels and decreased TGF-b1 concentrations, could potentially support the continued generation of Th17 cells but meanwhile suppress the development of Treg cells, which led to the Th17/Treg numerical imbalance in RA patients. This numerical imbalance might consequently result in the functional imbalance of Th17/Treg which contributed to enhancing the formation of the inflammatory cytokine microenvironment, and eventually formed a positive feedback mechanism to amplify proinflammatory immune responses.
Conclusions Our results demonstrated that numerical and functional imbalance of Th17/Treg existed in the peripheral blood of patients with RA. Inflammatory cytokine microenvironment led to the enhancement of Th17 cell differentiation at the expense of Tregs, and these combined changes contributed to triggering autoimmunity and inducing inflammation. All these suggested a potential role for Th17/Treg imbalance in the pathogenesis and progression of RA. Therefore, blocking critical cytokines in vivo, notably IL-6 and IL-23, or giving exogenous TGF-b1, may promote recovery of RA patients though restoring the Th17/Treg balance. Acknowledgments We thank the patients for donation of samples to the study, as well as the hospital staff who made this study possible. We also thank Yi Li and Bing Yang for helping process patient samples. We are very grateful to Prof. B. Ying for critical reading of the manuscript. This research was sponsored by National Natural Science Foundation of China (No. 30670819, 30772051 and 30950010).
123
2736 Conflict of interest of interest.
Rheumatol Int (2012) 32:2731–2736 The authors declare that they have no conflict
References 1. Firestein GS (2003) Evolving concepts of rheumatoid arthritis. Nature 423:356–361 2. McInnes IB, Schett G (2007) Cytokines in the pathogenesis of rheumatoid arthritis. Nat Rev Immunol 7:429–442 3. Falgarone G, Duclos M, Boissier MC (2007) TNF-a antagonists in rheumatoid arthritis patients seen in everyday practice. Joint Bone Spine 74:523–526 4. Bettelli E, Oukka M, Kuchroo VK (2007) T(H)-17 cells in the circle of immunity and autoimmunity. Nat Immunol 8:345–350 5. Marie-Christophe B, Eric A, Geraldine F et al (2008) Shifting the imbalance from Th1/Th2 to Th17/treg: The changing rheumatoid arthritis paradigm. Joint Bone Spine 75:373–375 6. Arnett FC, Edworthy SM, Bloch DA et al (1988) The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 31:315–324 7. Prevoo ML, van’t Hof MA, Kuper HH et al (1995) Modified disease activity scores that include twenty-eight-joint counts. Development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum 38:44–48 8. Manoury-Schwartz B, Chiocchia G, Bessis N et al (1997) High susceptibility to collagen-induced arthritis in mice lacking IFNgamma receptors. J Immunol 158:5501–5506 9. Boissier MC, Chiocchia G, Bessis N et al (1995) Biphasic effect of interferongamma in murine collagen-induced arthritis. Eur J Immunol 25:1184–1190 10. Eastaff-Leung N, Mabarrack N, Barbour A et al (2010) Foxp3? regulatory T cells, Th17 effector cells, and cytokine environment in inflammatory bowel disease. J Clin Immunol 30:80–89 11. Nistala K, Moncrieffe H, Newton KR et al (2008) Interleukin17—producing T cells are enriched in the joints of children with arthritis, but have a reciprocal relationship to regulatory T cell numbers. Arthritis Rheum 58:875–887 12. Chengl X, Yu X, Ding Y-J et al (2008) The Th17/Treg imbalance in patients with acute coronary syndrome. Clin Immunol 127: 89–97 13. Paradowska A, Mas´lin´ski W, Grzybowska-Kowalczyk A et al (2007) The function of interleukin 17 in the pathogenesis of rheumatoid arthritis. Arch Immunol Ther Exp 55:329–334
123
14. Kim KW, Cho ML, Park MK et al (2005) Increased interleukin17 production via a phosphoinositide 3-kinase/Akt and nuclear factor jB-dependent pathway in patients with rheumatoid arthritis. Arthritis Res Ther 7:R139–R148 15. Berg WB, Miossec P (2009) IL-17 as a future therapeutic target for rheumatoid arthritis. Nat Rev Rheumatol 5:549–553 16. Lubberts E (2008) IL-17/Th17 targeting: on the road to prevent chronic destructive arthritis? Cytokine 41:84–91 17. Hori S, Nomura T, Sakaguchi S (2003) Control of regulatory T cell development by the transcription factor Foxp3. Science 299:1057–1061 18. Abdulahad WH, Stegeman CA, van der Geld YM et al (2007) Functional defect of circulating regulatory CD4 T cells in patients with Wegener’s granulomatosis in remission. Arthritis Rheum 56:2080–2091 19. Venken K, Hellings N, Liblau R et al (2010) Disturbed regulatory T cell homeostasis in multiple sclerosis. Trends Mol Med 16:58–68 20. Wright GP, Notley CA, Xue SA et al (2009) Adoptive therapy with redirected primary regulatory T cells results in antigenspecific suppression of arthritis. Proc Natl Acad Sci USA 106:19078–19083 21. Morgan ME, Flierman R, van Duivenvoorde LM et al (2005) Effective treatment of collagen-induced arthritis by adoptive transfer of CD25? regulatory T cells. Arthritis Rheum 52:2212–2221 22. Bettelli E, Carrier Y, Gao W et al (2006) Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature 441:235–238 23. Kurts C (2008) Th17 cells: a third subset of CD4? T effector cells involved in organ-specific autoimmunity. Nephrol Dial Transplant 23:816–819 24. Wilson NJ, Boniface K, Chan JR et al (2007) Development, cytokine profile and function of human interleukin 17-producing helper T cells. Nat Immunol 8:950–957 25. Fontenot JD, Gavin MA, Rudensky AY (2003) Foxp3 programs the development and function of CD4? CD25? regulatory T cells. Nat Immunol 4:330–336 26. Acosta-Rodriguez EV, Napoletani G, Lanzavecchia A et al (2007) Interleukins 1b and 6 but not transforming growth factor-b are essential for the differentiation of interleukin 17-producing human T helper cells. Nat Immunol 8:942
