|Home | About | Journals | Submit | Contact Us | Français|
Regulating the differentiation and persistence of encephalitogenic T-cells is critical for the development of experimental autoimmune encephalomyelitis (EAE). We recently reported that CD5 has an engagement-dependent prosurvival activity in T-cells which had a direct role in the induction and progression EAE. We predicted that CD5 regulates T-cell apoptosis/survival through the activation of CK2, a prosurvival serine/threonine kinase that associates with the receptor. To test this hypothesis we generated mice expressing CD5 with the inability to bind and activate CK2 and assessed their susceptibility to EAE. We found mice deficient in CD5-CK2 signaling pathway were mostly resistant to development of EAE. Resistance to EAE was associated with a dramatic decrease in a population of effector TH-cells that co-express γIFN and IL-17 and to a lesser extent cells that express γIFN or IL-17 in draining lymph nodes and spinal cords. We further show that T-cells deficient in CD5-CK2 signaling hyperproliferate following primary stimulation, however following restimulation, they rapidly develop non-responsiveness and exhibit elevated activation-induced cell death. Our results provide a direct role for CD5-CK2 pathway in T-cell activation and persistence of effector T-cells in neuroinflammatory disease. This study predicts that targeting of γIFN+/IL-17+ TH cells will be useful for the treatment of multiple sclerosis and other systemic autoimmune diseases.
Experimental autoimmune encephalomyelitis (EAE) is a CD4 T-cell mediated disease model of Multiple Sclerosis (MS) (1, 2). The generation of effector CD4+ T-cells is a critical event for the progression of EAE. Until recently the disease was primarily considered TH1 dependent, however now the newly described TH17 lineage of CD4+ T-cells has emerged to be the major effector cell for the development of EAE (3). It is currently thought that IL-17 up-regulates the expression of chemokines, such as CCL2, CCL20 and MIP-2, to mediate inflammation in the CNS to promote disease (4, 5). In contrast, the TH2 and regulatory T-cell subsets have a protective role in EAE (6-8).
CD5 is primarily considered to be a negative co-stimulator of T-cell activation signals (9-13), and we recently showed that it also plays a critical role in promoting prosurvival signals in T-cells (14). T-cells can adjust their threshold for activation by altering the level CD5 on the cell surface (10, 15-18). The upregulation of CD5 can also lead to the development of antigen specific unresponsiveness. This was demonstrated by Hawiger et al (2004), when they reported the generation of a population T-cells with elevated levels of CD5 that developed antigen-specific nonresponsiveness (19). The non-responsiveness was dependent on CD5, since such a population could not be generated in CD5-/- mice or mice treated with anti-CD5. Conversely, by lowering CD5 expression T-cells can develop responsiveness as recently demonstrated in tumor infiltrating lymphocytes isolated from a lung carcinoma (15).
We recently showed that CD5-/- mice had decreased severity of EAE following myelin oligodendrocyte glycoprotein (MOG35-553) peptide immunization (14). Consistent with the inhibitory role of CD5, the CD5-/- mice exhibited an enhanced T-cell response to MOG-immunization. This initial response was followed by elevated activation-induced cell death (AICD) which conferred resistance to disease. In addition we found the pro-survival activity in T-cells provided by CD5 can be abrogated by blockade of CD5 engagement with soluble CD5-Ig in mice and is therapeutically beneficial for treatment of EAE. Another recent study demonstrated that myelin specific T-cells that escape superagonistic induced AICD, in fact, had elevated levels of CD5 and were unresponsive to re-exposure to antigen (16). Together these data suggest that CD5 levels are critical in determining both the responsiveness as well as the survival of T-cells after antigenic stimulation.
We and others previously established that CK2, a serine/threonine kinase, is associated with the cytoplasmic tail of CD5 and upon engagement of CD5 the kinase is activated (20-22). CK2 activity has been shown to promote cell survival of lymphocytes both by directly inhibiting apoptosis and by activating or inducing expression of prosurvival molecules (23, 24). We predicted that the CD5 engagement would deliver prosurvival/anti-apoptotic signals to T-cells by triggering CK2 activity; a process that would have a direct effect on the development of EAE. In order to address this hypothesis, we evaluated EAE susceptibility in mice expressing a CK2 binding/activation deficient CD5. In this study we report two important observations. First, we found that CD5-CK2 binding/activation deficient mice are resistant to EAE which was associated with a dramatically reduced population of infiltrating T helper (TH) cells that express both γIFN and IL-17 as well as a reduction in classical TH1 (γIFN+) and the newly described TH17 (IL-17+). Second, our studies also revealed a novel mechanism for regulating T-cell activation through CK2.
C57BL/6 mice were purchased from Jackson Laboratory (Bar Harbor, ME). CD5WT transgenic mice were provided by Dr. Paul Love (10). CD5-/- mice were crossed into C57BL/6 for 10 generations (14, 25). The CK2-binding deficient CD5 transgene was generated by introducing a micro-deletion of amino acid residues S458-S461 (CD5Δ458-461) by site directed mutation (Stratagene) of the coding region of CD5. The resultant cDNA was sequenced in entirety to confirm the presence of desired mutation and absence of non-desired mutations. The mutated cDNA was then cloned into the CD2 minigene construct generously given to us by Dr. Love (10). Founder lines were generated in the C57BL/6 mice and bred into CD5-/- mice and screened using forward primer (5′atggactcccacgaagtgctg3′) and reverse primer (5′cttgtagaggatggtgcca3′). This same set of primers was also used to genotype the CD5WT transgenic mice. All animals were housed and treated in accordance to National Institutes of Health and UAB IACUC guidelines.
Mice were immunized with 150 μg of MOG35-55 peptide (Biosynthesis) which was modified from a previously described protocol and EAE symptoms were monitored daily for 30-35 days using a standard clinical score ranging from 0-6 (14). Briefly, mice received a single immunization of peptide in CFA on day 0, followed by pertussis toxin (LIST Biologicals) on day 0 and day 2. Animals with a clinical score <2 as defined by loss of tail tone are considered free of clinical disease. Infiltrating cells were isolated from spinal cord as previously described (26). Briefly, spinal cord homogenates were obtained from 8-10 perfused animals and incubated with collagenase (2mg/ml) and DNAse (5units/ml) for 1h at 37°C; mononuclear cells were purified by two step Percoll gradient centrifugation.
Single cell suspensions of axillary and brachial draining lymph node (DLN) and spinal cord cells were incubated with anti-CD16/32 (2.4G2, FcR block) before staining with anti-CD8 (53-6.7) or anti-CD4 (L3T4), CD11b (M1/70) and anti-CD69 (H1.2F3) (Ebiosciences) conjugated to various fluorochromes as needed. For intracellular cytokine staining, cells were stimulated with 50 ng/ml PMA and 500 ng/ml ionomycin (Sigma-Aldrich) in the presence of Golgi-stop (BD Biosciences) for 4 h, before surface staining with anti-CD4-FITC and intracellular cytokine staining with anti-γIFN-Cy5 and anti-IL-17-PE (BD-Biosciences). For the measurement of cell cycle in vivo, mice were injected with 1.5 mg of BrDU 12 h before tissues were harvested. Cells were stained with anti-CD4-APC before intracellular staining with anti-BrDU-FITC and analyzed by flow cytometry as described in the manufacturer’s instructions (BD-bioscience).
To assess primary T-cell responses, spleen cells were cultured in 1μg/ml anti-CD3 (145-2C11) for 72 h. Frequency of CD4 and CD8 cells in cycle was determined by pulse labeling cells with BrDU (10μM) for 90 min prior to flow cytometric analysis of BrDU incorporation as describe above. Cell divisions were assessed by tracking the dilution of FL1-fluorescence of CD4 or CD8 cells labeled with CFSE (0.2μM) (Invitrogen) by flow cytometry. The kinetics of proliferation was assessed in triplicate at 24, 42 and 72 h of stimulation and addition of [3H]thymidine for the last 18 h of incubation. Incorporation of radioactivity was measured using a liquid scintillation counter (Packard). To assay for early apoptotic events in T-cells, we measured loss of mitochondrial membrane potential by incubating cells with dihexyloxacarbocyanine (DiOC6; Invitrogen) at a final concentration of 20nM for 15 mins at 37°C prior to staining for with anti-CD4-APC, anti-CD8-PE and 7-AAD (Invitrogen) (14, 27).
T-cell recall response was assessed by culturing spleen cells for 24 h with 1μg/ml of anti-CD3. Cells were washed and rested by culturing in fresh media for an additional 72 h. Live cells were separated from dead cells by FICOLL density gradient centrifugation, restimulated with 1μg/ml of anti-CD3 and assayed and BrDU incorporation as described above. Early apoptosis was assayed as described above.
Results were analyzed for statistical significance using a two-tailed Student’s t test.
We recently demonstrated that CD5-/- mice had a delay in onset and decrease in severity of EAE, which was due to the inability of activated T-cells to persist (14). Since, CD5 engagement activates CK2, a kinase known to regulate cell survival (21, 24), we predicted that the attenuation of EAE in CD5-/- mice was due to the loss of CD5-CK2 signaling. In order to test this hypothesis, we reconstituted CD5-/- mice with a CD2 promoter regulated transgene encoding a CK2 binding/activation deficient CD5 (CD5Δ458-461) with restricted expression in T-cells (Fig 1A). This microdeletion of four amino acids (SSDS) completely abrogates the interaction of CK2 with CD5 which is necessary for the activation of the kinase following engagement of the receptor (21). As controls, we used CD5-/- mice reconstituted with wild type CD5 transgene (CD5WT) (10). In both of these transgenic lines CD5 expression is under the control of the same regulatory elements. Both CD5WT and CD5Δ458-461 mice contained similar frequencies and numbers of CD4 and CD8 T-cell populations in the thymus and peripheral lymphoid organs (Fig 1B and C and data not shown). The expression level of CD5 on mature T-cells in the thymus and in the periphery was similar in both lines of CD5 transgenic mice. However, the expression of CD5 was lower on immature thymocytes (DN and DP) in CD5Δ458-461 mice compared to CD5WT mice (Fig 1D and E). The significance for this finding is currently unclear and is being explored.
To address the role of CD5-CK2 signaling in susceptibility to EAE, we compared the progression of disease in CD5+/+, CD5-/-, CD5WT and CD5Δ458-461 mice following immunization with MOG35-55 peptide. As we reported previously, CD5-/- mice exhibited a delayed onset and decreased severity of EAE (Fig 2). Reconstitution of CD5-/- mice with CD5WT transgene restored the disease phenotype to that of CD5+/+ mice. Most remarkable was that CD5Δ458-461 mice were almost completely resistant to the development of EAE and in fact disease severity in these mutant mice was significantly lower than in CD5-/- mice (Fig 2A). The cumulative disease index and maximum clinical score was CD5WT=CD5+/+>CD5-/-> CD5Δ458-461 mice (Fig 2B and C). While incidence of disease was greater than 90% in the CD5+/+, CD5WT and CD5-/- mice, only 20% (3 of 15) of the CD5Δ458-461 mice developed disease (Fig 2D).
In order to begin defining the underlying mechanism of resistance to EAE in CD5-CK2 binding/activation deficient (CD5Δ458-461) mice, we analyzed draining lymph nodes (DLN) and spinal cords for changes in the CD4+ T-cells populations at various time points during the acute phase of the disease and compared it to that of CD5+/+, CD5WT and CD5-/- mice.
We found no significant difference in frequency and absolute numbers of CD4+ T-cells in the DLN between the four groups of mice at all time points during the acute phase of EAE (data not shown). Furthermore, the percentage of activated (CD69+) CD4+ T-cells in the DLN from the CD5Δ458-461 was similar to that found in the DLN from the CD5+/+ and CD5WT mice (Fig 3A). The results indicate that initial T-cell priming is not compromised in CD5-CK2 activation deficient mice. As we reported previously, DLN from CD5-/- mice had elevated frequencies of activated (CD69+) CD4+ T-cells on day 4 day and were significantly decreased by day 12.
In contrast to DLN, we found differences in CD4+ T-cell populations between CD5WT and CD5Δ458-461 mice when we analyzed the spinal cord. On day 9 after induction, the frequency of CD4+ T-cells in the spinal cords of the CD5Δ458-461 mice was similar to CD5+/+, CD5WT and CD5-/- animals (9.9%, 8.0%, 9.2%, 7.0% respectively). However, the proportion of the CD4 T-cells in CD5+/+, CD5WT and CD5-/- mice sharply increased at later time points but not in the CD5Δ458-461 mice (Fig 3B). Histological analyses of spinal cord sections on day 17, the peak of the acute phase of the disease, showed that both the gray matter and myelin-rich white matter was infiltrated with inflammatory mononuclear cells in the CD5WT mice. In contrast, CD5Δ458-461 mice had mononuclear cells present only in the gray matter (Fig 4A). Staining of sections with Luxol fast blue revealed that white matter infiltration correlated with demyelination in the CD5WT cords; such demyelination was not observed in the CD5Δ458-461 mice (Fig 4B).
Two recent reports now establish that in EAE the CNS is a site for encephalitogenic T-cell amplification following re-exposure to antigen as well as recruitment and activation of new T-cell clones by a process known as epitope spreading (28-30). The lack of expansion of CD4 T cells in spinal cords of CD5Δ458-461 mice may reflect an inability of cells to proliferate in the inflamed tissue following re-exposure to CNS antigens (19, 31, 32). To address this possibility we injected mice with BrDU 12h before sacrifice on day 17 and assessed BrDU incorporation as a measure of cells in cycle. We found a greater proportion of CD4+ T-cells from the spinal cords of CD5Δ458-461 mice had incorporated BrDU compared to CD5WT mice (Fig 5). The greater numbers of BrDU positive cells may reflect differences in peripheral activation followed by recruitment into the CNS rather than in situ proliferation. However, we do not believe this is the case because by day 17, DLN in both groups of mice contained very small percentage of CD4+ T-cells that were BrDU positive (data not shown). This indicated to us that the low frequency of CD4+ T-cells in the spinal cords of CD5Δ458-461 mice in the presence of greater proliferation represents a lower capacity to persist rather than an inability to respond to antigen in the CNS.
The presence of infiltrating of CD4+ T-cells in the CNS at day 9 seen in CD5Δ458-461 mice is similar to that observed in IL-23 p19-/- mice immunized with MOG35-55 peptide (Fig 3C) (3, 4, 33). These recent studies led to the discovery that the IL-23 induced differentiation of IL-17 expressing TH cells (TH17) is essential for the immunopathogenesis of EAE. TH17 cells are described as a population of TH cells that express IL-17 and not γIFN (34). To address the possibility that resistance of CD5Δ458-461 mice to EAE may be associated with altered populations of TH effector cells, we analyzed DLN for the presence of γIFN (TH1) and IL-17 (TH17) T-cell populations on day 4, day 9 and day 17 following induction of acute EAE. On day 4, DLN of CD5Δ458-461 mice contained a greater frequency of CD4+ T-cells that expressed γIFN or IL-17 than of CD5+/+, CD5WT mice, whereas only γIFN expessing CD4+ T-cells were elevated in the CD5-/- mice (Fig 6). A small but detectable population of cells that co-expressed both γIFN and IL-17 was detected in DLN of CD5Δ458-461 and CD5-/- mice but not in CD5+/+ or CD5WT DLN. This indicates that early in the immune response a greater number of effector cells are generated in mice unable to activate CK2 through CD5. Analysis of day 9 DLN revealed that the proportion of TH17 cells was similar in both CD5+/+, CD5WT and CD5Δ458-461 mice and is slightly elevated in the CD5-/- mice. In contrast, the frequencies of TH1 and TH cells that co-express γIFN and IL-17 in the DLN of CD5+/+ and CD5WT mice were nearly 2 fold than in CD5-/- mice and almost 6 fold greater than in the DLN from CD5Δ458-461 mice. The frequency of effector TH cell populations in DLN, especially on day 9, is greater than one would expect if only MOG-reactive T-cells are activated. Since we use CFA to prepare the immunogen, we believe a large proportion of the reactive T-cells are in response to mycobacterial antigens.
We next assessed the TH effector cell populations in spinal cord on day 9, before the onset of clinical symptoms, and day 17, at the peak of disease. On day 9, the TH17 frequencies were similar in CD5+/+, CD5WT and CD5Δ458-461 but were lower in the cords of CD5-/- mice. In contrast the TH1 cells were significantly reduced in the cords from CD5Δ458-461 mice compared CD5+/+, CD5WT, CD5Δ458-461 (Fig 6). Compared to both CD5+/+ and CD5WT mice, TH cells expressing both γIFN and IL-17 were reduced in cords from CD5Δ458-461 mice as well as the CD5-/- mice, albeit to a lesser extent. On day 17, cords of CD5+/+, CD5WT and CD5-/- mice had greater proportion of TH1, TH17 and T cells expressing γIFN and IL-17 (Fig 6). Overall the results show that the resistance of CD5-CK2 binding/activation deficient mice to EAE is associated with decrease in generation and/or persistence of TH effector cells. An intriguing finding suggested by our results is that it is the IL-17 and γIFN co-expressing CD4+ cell which seems to be primarily associated with resistance to EAE and not the TH17 population, an aspect that will be addressed in Discussion.
The pseudo ITAM domain of CD5 that includes Y429 and Y441 is considered to be primarily involved with the inhibitory activity of CD5 (10-12). Since this domain is intact in CD5Δ458-461 mice, we tested the prediction that T-cells from these mice will respond to TCR stimulation to the same extent as T-cells expressing wildtype CD5. We assayed T-cell response by assessing BrDU incorporation after in vitro anti-CD3 stimulation. Surprisingly we found the CD4+ and CD8+ T-cells from CD5Δ458-461 had a frequency of BrDU+ cells comparible to that of T-cells from CD5-/- mice, which were both significantly higher than the frequency of BrDU+ T-cells from CD5WT and CD5+/+ mice (Fig 7A). This data indicates that greater numbers of T-cells from CD5Δ458-461 mice enter into cell cycle after activation but does not address the possibility of S phase arrest, a mechanism that would explain the decrease in amount of CD4+ T-cells observed in the CNS of MOG-immunized CD5Δ458-461 mice. To address this possibility, we assessed cell division in CFSE-labeled T-cells from CD5+/+, CD5WT, CD5-/- and CD5Δ458-461 mice after anti-CD3 stimulation. The number of undivided CD4+ and CD8+ T-cells from both the CD5-/- and CD5Δ458-461 mice were much less than the T-cells from the CD5+/+ and CD5WT mice (Fig 7B). In contrast, CD5+/+ and CD5WT T-cells had a greater frequency of cell that had undergone 5 or more divisions compared to T-cells from CD5-/- and CD5Δ458-461 mice. We also determined the kinetics of T-cell proliferation by [3H]thymidine incorporation. At 24 and 42 h following anti-CD3 stimulation, we found that the CD5-/- and CD5Δ458-461 T-cells proliferated at a greater rate than CD5+/+ and CD5WT T-cells (Fig. 8A). In contrast, thymidine uptake in T-cells from CD5-/- and CD5Δ458-461 mice at 72h had reduced from that observed at 42h and the proliferation of the CD5Δ458-461 T-cells was significantly lower than that of T-cells from CD5+/+ and CD5WT mice, a characteristic associated with loss in viability. To confirm the loss of decrease in the viability of the CD5-/- and CD5Δ458-461 T-cells was due to increased cell death, we assessed the frequency of cells poised for apoptosis by measuring mitochondrial depolarization using the dye DiOC6. After 24 h of stimulation, CD4+ T-cells from both the CD5-/- and CD5Δ458-461 cells exhibited a greater proportion of T-cells undergoing early stages of apoptosis, characterized by the greater frequency of a DiOC6lo population compared to CD5+/+ and CD5WT cells (Fig. 8B). Overall these data indicate 1) CD5-CK2 binding/activation deficient T-cells are hyperproliferative, and 2) have decreased ability to respond to prolonged stimulation.
We next examined the role of CD5-dependent CK2 activation pathway has in regulating T-cell recall response. We determined the effect of anti-CD3 restimulation by measuring the frequency of cycling and apoptotic T-cells from CD5+/+, CD5WT, CD5-/- and CD5Δ458-461 mice in a recall response assay. After 48 h of re-stimulation, CD5Δ458-461 T-cells had a significantly lower amount of cells in cycle compared to CD5+/+, CD5WT and CD5-/- T-cells (Fig 9A). We also assessed apoptosis by measuring mitochondrial depolarization using the dye DiOC6. We found that CD5Δ458-461 cells exhibited a greater proportion of T-cells undergoing early stages of apoptosis compared to CD5+/+, CD5WT and CD5-/- cells (Fig 9B). These results show that CD5Δ458-461 T-cells exhibit low threshold for the development of anergy and also are more susceptible to AICD.
We recently reported that EAE was less severe in CD5-/- mice compared to CD5+/+ mice (14). This was due to an enhanced activation induced cell death of CD5-/- T-cells leading to the discovery that CD5 had a prosurvival role in activated T-cells. We predicted that CD5-dependent CK2 activation was the mechanism by which CD5 promoted survival. Consistent with our prediction, we report here that CD5-CK2 binding/activation deficient mice, CD5Δ458-461, are very resistant to MOG-induced EAE (Fig. 2). An unexpected finding was that both disease severity and incidence of EAE in CD5Δ458-461 mice was also significantly lower than CD5-/- mice. The lower disease penetrance may reflect differences in TCR activation thresholds between T cells lacking only the CD5-dependent CK2 activation pathway and T-cells with no CD5 expression. CD5-/- T-cells lack both ITIM-associated and CK2 binding domain associated regulatory pathways resulting in lower threshold for activation than CD5Δ458-461 T-cells. In fact, immunization of CD5-/- mice with MOG35-55 peptide leads to upregulation of CD69 on greater number of DLN T-cells than in CD5WT and CD5Δ458-461 mice (14) and data not shown). Since CD5 is involved in thymic selection, another possibility is that the MOG-specific repertoire in CD5Δ458-461 mice is altered (9, 10). The fact that MOG immunization in CD5Δ458-461 mice leads to activated T-cells in DLN and differentiation into all the effector TH populations seen in CD5+/+ and CD5WT mice argues against altered repertoire. Treg cells have an important role in controlling encephalitogenic T-cell activity and can contribute to differences in disease susceptibility and/or severity (35). We found no differences in numbers of CD4+CD25+ Treg cells between CD5+/+, CD5WT, CD5-/- and CD5Δ458-461 mice (data not shown); however, additional functional experiments need to be performed to address this question. One might suggest that enhanced resistance of CD5Δ458-461 mice to EAE is a transgene-integration site induced artifact. We do not believe this is the case as we have just generated “knock-in” CD5Δ458-461 mice in which the endogenous CD5 gene was altered by gene targeting. In this model, where integration artifacts are not an issue, severity and incidence of EAE recapitulates disease progression in the transgenic CD5Δ458-461 mice (unpublished observation).
CK2 activation promotes survival by directly inhibiting pro-apoptotic pathways as well as by enhancing pro-survival signaling cascades (23, 24, 36-38). We therefore predicted that resistance of CD5Δ458-461 mice to EAE will be associated with decrease and/or absence of infiltrating cells in the spinal cord. However our results revealed that this was not the case. Spinal cords of CD5Δ458-461 mice had substantial infiltrating mononuclear cells localized in the grey matter with similar numbers of CD4+ T-cells early in the disease (day 9) compared to CD5WT mice (Fig3B and and4A).4A). However, CD4+ T-cell numbers in spinal cords of CD5+/+, CD5WT and CD5-/- mice increased as disease progressed compared to very little change in CD5Δ458-461 mice. A similar phenomenon was reported in IL-23p19-/- mice, where resistance to EAE occurred in the presence of T-cell infiltration in the CNS (33). A subsequent report by the same group revealed that IL-23 was necessary for generation of IL-17 expressing T cells now called TH17, and this population was absent in spinal cords of mice resistant to EAE (3). The TH17 cells are now considered to be the critical effector population in the pathogenesis of EAE and collagen induced arthritis (3, 4, 39). Recent reports describe the TH17 population as a subset of TH cells that express IL-17 and not γIFN (4, 34, 40). Early in the acute phase of EAE (day 9) we observed that the frequency of total IL-17 producing CD4+ T-cells in DLN and spinal cords was significantly lower in CD5Δ458-461 mice than both CD5+/+ and CD5WT mice; however the reduction was primarily limited to a TH population that coexpressed both IL-17 and γIFN and not the TH17 subset (Fig 6). We also observed a decrease in TH1 cells (γIFN). By day 17, the peak of acute EAE in CD5WT mice, spinal cords of CD5Δ458-461 mice contained fewer TH1, TH γIFN+IL-17+ and TH17 cells than CD5+/+, CD5WT and CD5-/- mice. Because BrDU uptake was greater in CD5Δ458-461 cords, we believe the lower numbers of these TH cell subsets in CD5Δ458-461 mice represents a decreased ability to persist rather than failure to proliferate (Fig. 5).
The γIFN+IL17+ TH cells probably represent a population distinct from TH17 cells and were first identified in synovial cells obtained patients with lyme arthritis (41). This population of TH has also been observed in vitro CD4+ T-cell polarization experiments in which IL-23 is added to the culture (40) and proteolipid protein peptide (PLP) stimulated CD4+ T-cell cultures obtained from DLN of PLP primed mice (3). However, its potential importance in pathogenesis has not been considered. The TH17 cells, thought to be the pathogenic population in EAE, develop directly from TH0 cells and represent a distinct lineage of T helper cells whose differentiation from TH0 cells is not related to the TH1 lineage (4, 40). The generation TH17 cells is dependent on IL-23 and does not require T-bet or STAT4 (3, 4, 40), two transcription factors necessary for generation of TH1 T-cells. Remarkably, both T-bet deficient and STAT4 deficient mice are resistant to EAE (6, 42), indicating that pathogenic T-cells in EAE are derived from the initial stages of TH1 differentiation which, when exposed to IL-23, express IL-17. Our finding that resistance of CD5Δ458-461 mice to EAE is associated with decrease in γIFN+/IL-17+ CD4 T-cells and not TH17 T-cells is consistent with this paradigm. It is unclear if the γIFN+/IL17+ T-cells differentiate from γIFN+/IL17- T-cells, which are also reduced in CD5Δ458-461 mice; a subject of ongoing investigation.
The pseudo-ITAM/ITIM motif in CD5 cytoplasmic tail is considered to be necessary for its ability to negatively regulate T-cell activation (10-13). Our studies indicate that the CD5-CK2 pathway is involved in regulating survival (Fig. 8 and and9).9). T-cells in CD5-/- mice lack both inhibitory and pro-survival activity of CD5 that may compensate for each other in response to antigen stimulation. This property could explain the intermediate phenotype of CD5-/- mice in susceptibility to EAE compared to CD5+/+, CD5WT and CD5Δ458-461 mice. By this paradigm, the lack of increase in CD4+ T-cell numbers in cords of CD5Δ458-461 mice compared to CD5+/+, CD5WT and CD5-/- mice can be explained by the loss of CD5-dependent survival signals in the presence of normal CD5-inhibitory activity (Fig. 3B). However, the presence of greater proportion of different TH effector cells in DLN early in the immune response (day 4) and the results from the BrDU pulse labeling experiment indicating that T-cells in cords of CD5Δ458-461 hyperproliferate provide the first indication that the CD5-dependent CK2 activation pathway may also be involved in the negative regulation of T-cell activation and differentiation (Fig. 5 and and6).6). Results from in vitro stimulation experiments support this finding (Fig. 7 and and8).8). The absence of any difference in upregulation of CD69 between CD4+ T-cells from CD5+/+, CD5WT and CD5Δ458-461 mice indicate that negative regulatory activity mediated by CD5-dependent CK2 activation targets pathways distal to TCR/CD3 signaling (Fig. 3A). The CD5-ITIM-associated pathway is likely to be involved in attenuating TCR proximal signaling pathways (12). We now propose that CD5-ITIM-dependent and CD5-CK2 activation dependent signals cooperatively regulate T-cell proliferation and differentiation. Mice carrying targeted mutation of the ITIM domain will be needed to conclusively test this model.
Following MOG immunization CD4+ T-cells in CD5Δ458-461 mice are primed in DLN, infiltrate the CNS and enter into cell cycle equal to or with greater efficiency than that in CD5WT mice. This leaves open the question of mechanism of resistance of CD5Δ458-461 mice to EAE. In this disease model, naive CD4+ T-cells, after being primed in secondary lymphoid tissues migrate to the CNS where they are re-exposed to antigen presented by microglia and dendritic cells in the perivascular space (28, 30). The re-encounter of antigen leads to local amplification of encephalitogenic T-cells, damage to nervous tissue, expression of “new” antigens followed by recruitment of other T-cell clones by a process known as epitope spreading (29). Encephalitogenic T-cells that have poor ability to persist or those easily induced to become non-responsive will fail to induce the pathogenic cascade in the CNS (43, 44). In experiments that to some extent recapitulates responses following re-exposure to TCR stimulation, we observed that both CD4+ and CD8+ T-cells from CD5Δ458-461 mice readily develop non-responsiveness and exhibit much increase sensitivity to apoptosis upon restimulation than T-cells from CD5+/+,CD5WT and CD5-/- mice (Fig. 9). Overall from these data, we suggest that the mechanism underlying the decreased severity to EAE in CD5Δ458-461 mice was not due to the failure of CD4+ T-cell priming in the secondary lymphoid organs, but rather due to low threshold for development of anergy on restimulation and a decrease in the ability of effector CD4+ T-cells to persist.
Autoreactive T-cells play a fundamental role in the development, disease severity and perpetuation of multiple sclerosis and EAE. In the course of this study we made two significant findings. First, we report that the CD5-dependent CK2 activation pathway is an important mechanism by which CD5 regulates both T-cell activation and persistence in an inflammatory disease. Second, we have determined that γIFN and IL-17 co-expressing CD4+ T-cells are likely to be play key role in the pathogenesis of EAE. Targeting strategies that alter the generation and or the persistence of TH-γIFN+IL-17+ cells through manipulation the CD5-CK2 activation pathway may be a useful therapeutic approach for MS and other inflammatory autoimmune diseases (45-47).
1Address correspondence to Chander Raman, Department of Medicine, Division of Clinical Immunology and Rheumatology, University of Alabama at Birmingham, LHRB 463, 1530 3rd Avenue South, Birmingham AL 35294-0007. Phone: 205-934-2472; FAX: 205-934-2542; E-mail Address: ude.bau@namarc
2This work was supported by grants from the National Institutes of Health (AG16221) and from the Lupus Research Institute to C.R., NIH-T32 (AR07450-23) to R.C.A. and a grant from the National Multiple Sclerosis Society (PP0675) to S.R.B.