CD45RB Ligation Inhibits Allergic Pulmonary Inflammation by Inducing CTLA4 Transcription¹

Immune- mediated disorders, including allergic asthma, are increasing in severity and prevalence (1). Current understanding of the pathogenesis of allergy, autoimmune diseases, and transplantation rejection emphasizes their inflammatory nature. Central to these processes is T cell activation, which is essential in promoting allergic inflammatory disorders, including asthma (2).

CD45 is a type I transmembrane protein tyrosine phosphatase expressed on nucleated hemopoietic cells (3). Alternative mRNA splicing gives rise to multiple CD45 isoforms that differ only in the length and glycosylation of their extracellular domains (4). Although the exact function of each isoform remains unclear, mAbs recognizing CD45RB isoforms differentiate two populations of CD4⁺ T cells: CD4⁺CD45RB^high (5) and CD4⁺CD45RB^low (6) that secrete different cytokines and have distinct functional properties (7). The CD4⁺CD45RB^high population contains naive effector T cells that promote autoimmunity (8) and inflammatory bowel disease (9), while the CD4⁺CD45RB^low population contains previously activated cells as well as T regulatory (Treg)⁴ cells, which inhibit T cell effector activity (9, 10).

Previous studies have shown that CD45RB ligation prolongs allograft survival and can induce tolerance in murine models (11, 12, 13). Anti-CD45 mAb (aCD45) increases the expression of CTLA4, a negative regulator of T cell activation (7, 14). Interestingly, a recent study of 167 Danish families mapped a functional gene SFRS8 (splicing factor, arginine/serine-rich 8) associated with asthma. This gene regulates the splicing of CD45 (15), suggesting a possible role of CD45RB in allergic asthma. In this study, we asked whether CD45RB signals would impact immune responses in a murine model of allergic inflammation. We also examined the role of CTLA4, known to decrease T cell activation, in CD45RB-mediated inhibition of allergic inflammation. We suggest mechanisms by which aCD45 augments CTLA4 expression and decreases allergic inflammation.

Six- to 8-wk-old BALB/c male mice were purchased from The Jackson Laboratory. The mice were maintained according to the guidelines of the Committee on Animals of the Harvard Medical School and Animal Welfare Program from School of Medicine, University of California, San Diego.

Mouse anti-CD45RB mAb (MB23G2) (aCD45), described previously (16), and mouse anti-CTLA4 (UC10-4F10-11) (aCTLA4) were purchased from Bio Express Culture Services. Rat IgG2a (LO-DNP-16; Serotec) was used as a negative control.

Murine splenocytes and EL4 cells (murine thymoma cell line) were grown in RPMI 1640 medium supplemented with 10% heat inactivated FCS, 2 mM l-glutamine, 50 U of penicillin/ml, and 50 μg of streptomycin/ml. EL4 cells were treated with aCD45 (3 μg) for 24 and 48 h (h) at 2 × 10⁶ cells/ml. Murine spleen cells were treated with aCD45 (3 μg) for 24 and 48 h at 5 × 10⁶ cells/ml. As a positive control, EL4 cells or spleen cells were also treated with 5 μg/ml Con A (Sigma-Aldrich) and 100 ng/ml PMA (Sigma-Aldrich). As a negative control, both EL4 cells and murine splenocytes remained untreated.

Mice were sensitized and challenged with the allergen OVA as previously described (17, 18, 19, 20). OVA mice were sensitized via i.p. injection with 10 μg of chicken OVA (Sigma-Aldrich) and 1 mg of Al(OH)₂ (alum; Sigma-Aldrich) in 0.2 ml of PBS (Sigma-Aldrich), followed by an injection on day 7 with identical reagents. PBS mice received 1 mg of alum in 0.2 ml of PBS without OVA. On days 14–20, mice received aerolized challenges with 6% OVA or PBS, respectively, for 20 min/day via an ultrasonic nebulizer (model 5000; DeVillbiss). All groups were sacrificed at day 21 and analyzed for the allergic parameters described below.

Each mouse underwent BAL as previously described (21, 22). BAL cells were pelleted and the supernatant was stored at –80°C. Cells were resuspended in RPMI 1640 (5 × 10⁵ cells/ml). Slides for differential cell counts were prepared with Cytospin (Shandon) and fixed and stained with Diff-Quik (Dade Behringm). For each sample, an investigator blinded to the treatment groups performed two counts of 100 cells.

BAL IL-13 was measured by ELISA according to the manufacturer’s specifications (R&D Systems). Briefly, samples of BAL fluid were aliquoted in duplicate into 96-well plates (50 μl/well) precoated with Ab to specific cytokines and assayed according to the manufacturer’s instructions. OD was measured at 450 nm. Cytokine concentrations were determined by comparison with known standards.

Total serum IgE levels were determined by ELISA as previously described (17, 22). Total serum IgE concentrations were calculated by using standard curve generated with commercial IgE standard (BD PharMingen).

Airway measurements were assessed using whole-body plethysmography 4 h after the final aerosol OVA challenge. Mice were placed in individual chambers and increasing doses of methacholine were nebulized into the chambers via an inlet at concentrations of 20, 40, 80, and 100 mg/ml for 2 min as previously described (17, 20). Readings were averaged over 5 min after the nebulization. The whole-body plethysmography system measures changes in box pressure during expiration and inspiration; peak expiratory and peak inspiratory pressures, respectively; inspiratory time; expiratory time; and relaxation time (Tr = time of the pressure decay to 36% of total box pressure during expiration) and generates a value called enhanced pause (enhanced pause = peak expiratory pressure/peak inhibitory pressure [(expiratory time − Tr)/Tr] that relates to airway measurements in BALB/c mice (23).

BALB/c mice received 100 μg of aCD45 via i.p. injection 1 day before sensitization (day –1) and/or 100 μg of aCTLA4 i.p. 2 days before sensitization (day –2). Control BALB/c mice received Ig 2 days before sensitization (day −2).

Total RNA was isolated with TRI reagent (Sigma-Aldrich). Isolated RNA was reverse transcribed with SuperScript II RNase reverse transcriptase (Invitrogen Life Technologies). Specific primer pairs for GAPDH (housekeeping gene), CTLA4, IFN-γ, IL-2, IL-4, IL-5, IL-6, and IL-13 were designed with Primer Express software (Applied Biosystems). Direct detection of the PCR product was monitored by measuring the increase in fluorescence caused by the binding of SYBR Green to dsDNA. Using 5 μl of cDNA, 5 μl of primer, and 10 μl of SYBR Green Master Mix (Applied Biosystems) per well, the gene-specific PCR products were measured continuously by means of a GeneAmp 5700 Sequence Detection System (Applied Biosystems) during 40 cycles. The threshold cycle of each target product was determined and set in relation to the amplification plot of GAPDH. The difference in threshold cycle values of two genes was used to calculate the fold difference as previously described (24).

The mouse CTLA4 promoter containing –1221 bp from the transcription start site was cloned into the pXP2 basic vector, which contains the luciferase reporter gene, as described previously (25). The luciferase construct that contains the –1221-bp fragment (p1221) was used as a template for further 5′ deletion constructs –335 (p335), −238 (p238), and –62 (p63). All constructs were sequenced in both directions by dideoxy sequencing (25).

CTLA4 promoter luciferase reporter constructs (described above; 2 μg) and β-galactosidase reporter gene (pGK; 1 μg) were added to 2 × 10⁶ EL4 cells resuspended in 100 μl of Nucleofector solution (Amaxa Biosystems) and electroporated using the C-9 program of the Nucleofector. After 24 h, cells were lysed in reporter lysis buffer (Promega). Then, 10 μl of the cell lysate was mixed with 100 μl of luciferase assay reagent (Promega) and luciferase activity was measured by a luminometer (Turner BioSystems). Luciferase activity was normalized for transfection efficiency by β-galactosidase activity measured with Galacto-light systems according to the manufacturer’s instructions (Applied Biosystems). Fold activation was calculated as the ratio of luciferase vs β-galactosidase activity.

Transcription factor binding sites from CTLA4 promoter region were identified using TESS (www.cbil.upenn.edu/cgi-bin/tess/tess33).

ANOVA was performed for dose-response curves (see Fig. 5 c). Statistical analysis of normally distributed data was performed by t test, and nonparametric data were analyzed by a Mann-Whitney U test (Sigma Stat 3.0 software). Bonferroni correction for statistical adjustment of the p value for multiple comparisons was applied as a post hoc analysis. Data are reported as means ± SEM. Statistical significance was defined by p < 0.05.

We analyzed allergic parameters using a model of allergen OVA-induced pulmonary inflammation (17, 18, 26, 27). OVA- sensitized and -challenged mice showed a significant increase in BAL eosinophilia, IL-13, and total serum IgE when compared with PBS control mice (Figs. 1, a and b, and 2). CTLA4 is a negative regulator of T cell activation and administration of a blocking mAb to CTLA4 (aCTLA4) to OVA-sensitized and -challenged mice increased allergen-induced eosinophils (Fig. 1,a), total serum IgE (Fig. 1,b), and BAL IL-13 (Fig. 2). These data indicate that CTLA4-mediated influences are manifested in allergic pulmonary inflammation.

Previous studies in transplantation models indicate that aCD45 prolongs allograft survival through induction of CTLA4 expression in T cells (7, 14). In concert, incubation of murine splenocytes for 24 h with aCD45 resulted in an ∼5-fold increase in CTLA4 mRNA levels (Fig. 3).

The role of CD45RB signals has not previously been examined in allergic immune responses. Although aCD45 induces CTLA4, its effectiveness in allograft models may also relate to deviation toward a CD45RB^low phenotype and induction of Th2 cytokine expression (28, 29). Although in allergic responses, Th2 (as well as Th1) cytokines have been shown to promote allergic inflammation, it is not clear whether aCD45 would be effective in this setting (30, 31). We therefore examined the effect of aCD45, which was administered 2 days before OVA sensitization and challenge. Administration of aCD45 in an established OVA murine model of allergic inflammation markedly decreases allergen-induced BAL eosinophilia, total serum IgE, and BAL IL-13, such that they resemble those observed in control mice “immunized” with PBS (Fig. 4). These changes were not evident when mice were treated with control Ig rather than aCD45 (data not shown).

We next examined the effect of CTLA4 blockade on protection from allergic responses mediated by administration of aCD45 to OVA-sensitized plus -challenged mice. In mice that also received aCTLA4, aCD45 no longer decreased OVA-induced pulmonary eosinophilia and BAL IL-13 secretion or airway measurements (Fig. 5). The abrogation of the ability of aCD45 to inhibit allergic responses suggests that aCD45 decreases allergic inflammation through CTLA4 pathways.

aCD45 increased CTLA4 in vivo and this pathway appears critical for CD45RB-mediated suppression of allergic inflammation. To determine potential mechanisms by which aCD45 increases CTLA4 expression in T cells, murine T cells (EL4) were treated with aCD45 in vitro. Untreated EL4 cells did not express detectable levels of CTLA4 mRNA expression. Increased CTLA4 levels were observed 24 and 48 h after treatment with aCD45 (Fig. 6). Treatment with Con A plus PMA (similar to TCR signals) was used as a positive control and showed an increase of 6 ± fold at 24 h (data not shown). Thus, aCD45 may directly augment CTLA4 transcription.

To define the regulatory regions involved in aCD45-inducible CTLA4 gene expression in T cells, we analyzed a panel of CTLA4 luciferase reporter gene constructs. These luciferase constructs were defined at positions: −335 (p335), −238 (p238), and –63 (p63) bp relative to the start site for CTLA4 transcription. aCD45 induces p335 and p238 promoter region activation, as determined by an increase in luciferase activity (Fig. 7,a). A larger construct located within 1.2 kb (p1221) of the transcription start site also increased luciferase activity (5-fold relative to medium; data not shown). Deletion from positions –335 (p335) to –62 (p63) in the CTLA4 promoter abrogates luciferase activity. The p335 and p238 constructs showed a 5- and 4-fold increase, respectively, in luciferase activation, indicating the presence of positive regulatory elements for CTLA4-inducible expression by aCD45. In addition, we analyzed this CTLA4 promoter region (–335 to –1) by TESS (see Materials and Methods). Binding sites for AP-1, NFAT-1, IL-6.RE-BP, and NF-IL2A were identified, in addition to others (Fig. 7 b). Notably, these data indicate that aCD45 may increase CTLA4 by directly augmenting its transcription.

Positive regulatory elements for induction of CTLA4 by aCD45 may include putative sites for a number of transcription factors involved in cytokine production. We next examined cytokine transcript levels in EL4 cells that were or were not treated with aCD45 (Fig. 8). IL-2, IL-4, IL-5, IL-6, and IL-13 mRNA levels were increased after 24 h of stimulation with aCD45 as analyzed by real-time PCR. aCD45 does not increase the mRNA levels of IFN-γ.

Murine spleen cells treated with aCD45 and untreated cells were examined for cytokine transcript levels at two different time points (24 and 48 h; Fig. 9). IL-2 and IL-13 mRNA levels were increased at 24, but not at 48 h (data not shown), while IL-4 and IL-5 mRNA levels were increased at 48 h, but not at 24 h (data not shown). IFN-γ levels increased 3-fold relative to untreated at both time points.

A recent study of asthmatic Danish families mapped a functional gene associated with asthma responsible for the regulation of the splicing of the CD45 molecule (15). Although the functional consequences are unknown, aCD45 alters CD45 splicing (28), suggesting that aCD45 might impact allergic responses. Indeed, we found that CD45RB signals suppress allergic responses in a murine model of allergic inflammation. Our results showed that aCD45 decreases allergen-induced pulmonary eosinophilia, BAL IL-13, total serum IgE, and airway responses. Furthermore, CD45RB signals can no longer suppress allergic inflammation when CTLA4 is inhibited. These data support a role for CTLA4 in CD45RB-mediated decreases of allergic inflammation. In this regard, we found that aCD45 may directly augment CTLA4 transcription. To determine the regulatory elements involved, we performed a reporter gene analysis with CTLA4 promoter constructs. We identified an aCD45-inducible regulatory region localized from −335 to –62 bp relative to the CTLA4 transcription start site. Together, these findings suggest that CD45RB signals may play a novel role in the modulation of allergic inflammation.

Previous studies have shown that aCD45 is a potent tolerogenic agent that prolongs allograft survival in murine models (7, 11, 14, 28, 29, 32). The immunomodulatory function of aCD45 may occur through increased CTLA4 on T cells, as evidenced in transplantation models (7, 14). To address the mechanisms by which CD45RB decreases allergic immune responses, we also examined the role of CTLA4 in allergic inflammation. Consistent with previous models, blockade of CTLA4 maintained or increased allergic responses (33, 34). Also, in the presence of aCTLA4, aCD45 no longer suppressed IL-13 production. aCD45 also enhances cytokines in vitro. One interpretation is that in the presence of allergen stimulation in vivo, aCD45 is capable of enhancing Th2 cytokines, including IL-13, but it appears that CTLA4-mediated inhibition predominates. Overall, CD45RB-mediated modulation of CTLA4 is a crucial immune regulatory pathway.

To elucidate the molecular mechanisms by which CD45RB signals modulate CTLA4 expression, T cells were stimulated with aCD45 and CTLA4 mRNA levels were analyzed. Our data indicate that aCD45 increases transcription of the CTLA4 promoter region in T cells, a finding supporting direct induction of CTLA4 by aCD45. It is already known that T cell activation leads to CTLA4 induction. However, this generally requires full T cell activation and entry into the cell cycle (25). Remarkably, aCD45 appears able to induce CTLA4 even though this mAb is not mitogenic and does not induce activation markers such as CD69 or CD44 (Refs. 7 and 35 and D. Rothstein, unpublished data). Our reporter gene analysis indicates that the regulatory region for CD45RB-inducible CTLA4 expression is located within –335 to –62 bp relative to the transcription start site. Interestingly, CD45RB signals induced the CTLA4 promoter region that we previously ascertained to be TCR signal inducible (25, 36). Thus, although aCD45 does not induce overt T cell activation, it clearly results in signals that partially overlap with those mediated by the TCR. The importance of this finding is that T cells may be induced to express CTLA4 without undergoing normal activation, proliferation, and development of their full-fledged effector repertoire.

Computer analysis of the CTLA4 upstream region (−335 to –62) also revealed potential DNA binding sites for NF-IL2A (Oct-1), IL6REBP as well as AP-1 and NFAT. We focus on NF-IL2A and IL6REBP because they have been shown to regulate the gene expression of IL-2 and IL-6 (37, 38). Interestingly, in T cells stimulated with aCD45, we found an increase of IL-2 and IL-6 levels. Whether CD45RB signals may regulate these cytokines by acting on specific CTLA4 promoter regions remain to be determined. In this scenario, CD45RB signals may influence NF-IL2A and IL-6RE-BP DNA binding sites localized within this CTLA4 promoter upstream region. We hypothesize that induction of CTLA4 before TCR stimulation inhibits subsequent activation of allergic immune responsiveness. This CTLA4 induction may specifically dampen proinflammatory cytokines such as IL-4, IL-5, IL-6, and IL-13 that are enhanced by TCR ligation under the influence of aCD45.

The signaling pathways triggered by CD45 signals in T cells remain unknown. Analysis of the 335-bp regulatory region of the CTLA4 promoter reveals potential binding sites for NFAT (25). Cyclosporin A treatment abrogates CTLA4 induction and allograft survival induced by aCD45 (7), suggesting the possibility that NFAT signaling may be involved in CTLA4 induction. The CD45 PTPase is directly involved in T cell activation in part through its ability to regulate the activity of Src tyrosine kinases such as Fyn and Lck (39, 40). However, we showed that CD45 has additional substrates including the TCR ζ-chain and JAKs (41, 42). Moreover, depending on the site of dephosphorylation, CD45 has the potential to either activate or inactivate CD45 substrates such as src kinases (39, 40). Finally, the role of individual isoforms and identity of specific substrates remain obscure. Presumably, mAb-mediated ligation of CD45RB alters PTPase activity, substrate interactions, or both. However, only a limited set of pathways leading to NFAT must be activated, since aCD45 is not mitogenic and does not lead to overt T cell activation (7).

Our analyses indicate that aCD45 augments CTLA4 expression, which may be instrumental in inhibiting some allergic pulmonary responses. aCD45 induces a shift in CD45 isoform expression from CD45RB^high to CD45RB^low. This molecular weight. shift is specific to CD4 cells in vivo and a proportion of CD45RB^low cells coexpress CD25 and CTLA4 (7, 28). We envision that cells up-regulating CTLA4 are derived from those that were also induced to shift their CD45 isoform expression after aCD45 treatment. However, this remains unproven and CTLA4 could be induced on a subpopulation of CD4 cells originally within the CD45^low population. Luke et al. (43) have suggested that aCD45 may simply deplete CD45RB^high T cells, leaving CD45RB^low T cells enriched for those expressing CTLA4. Although we and Lazarovitz et al. (11) have both shown that leukocytes are partially and temporarily depleted from the periphery during aCD45 therapy, we previously reported that there is no depletion of CD4 cells in spleen or lymph node, where the shift in CD45 isoforms and induction of CTLA4 are readily visualized (7, 28). Moreover, there is no depletion of the CD44^low subset as would be expected if CD45RB^high cells were selectively depleted (7). In addition, we have previously shown that aCD45 treatment increases CD25 expression on most of the CTLA4-positive T cells. These CD25-positive cells from aCD45-treated mice exhibit Treg activity, suggesting that aCD45 induces Tregs (44). Approximately 70% of CTLA4-positive cells from naive mice treated with aCD45 also coexpressed Foxp3, suggesting that aCD45 may induce Tregs in vivo (D. Rothstein, unpublished data).

Our data indicate that CD45RB ligation markedly inhibits allergen-induced pulmonary eosinophilia, IL-13 production, and systemic induction of serum IgE. Moreover, we showed that aCD45 can no longer decrease allergic inflammation when CTLA4 is blocked. Finally, aCD45 can increase CTLA4 mRNA levels and promoter activation in T cells. Taken together, these results suggest that CD45RB signals decrease allergic inflammation through regulation of CTLA4. Future investigations of the specific lymphocyte populations and signaling pathways that may be targets for the effects of CD45RB signals will be crucial in the understanding of how to manipulate this critical pathway to decrease allergic responses.

We thank Dr. Xin Lu for statistical analysis, Dr. Vipul Jain for his critique of this manuscript, and Dr. Timothy Bigby for his assistance.

The authors have no financial conflict of interest.