Nasal vaccination with troponin reduces troponin specific T-cell responses and improves heart function in myocardial ischemia–reperfusion injury

Abstract

Myocardial ischemia with subsequent reperfusion (MI/R) can lead to significant myocardial damage. Ischemia initiates inflammation at the blood–microvascular endothelial cell interface and contributes significantly to both acute injury and repair of the damaged tissue. We have found that MI/R injury in mice is associated with a cellular immune response to troponin. Myocardial cells exclusively synthesize troponin and release the troponin into the bloodstream following injury. Mucosally administered proteins induce T cells that secrete anti-inflammatory cytokines such as IL-10 and transforming growth factor β at the anatomical site where the protein localizes. We found that nasal administration of the three subunits of troponin (C, I and T isoforms), given prior to or 1 h following MI/R, decreased infarct size by 40% measured 24 h later. At 1.5 months following MI/R, there was a 50% reduction in infarct size and improvement in cardiac function as measured by echocardiography. Protection was associated with a reduction of cellular immunity to troponin. Immunohistochemistry demonstrated increased IL-10 and reduced IFN-γ in the area surrounding the ischemic infarct following nasal troponin. Adoptive transfer of CD4+ T cells to mice from nasally troponin-treated mice 1 h after the MI/R decreased infarct size by 72%, whereas CD4+ T cells from IL-10−/− mice or nasally BSA-treated mice had no effect. Our results demonstrate that IL-10-secreting CD4+ T cells induced by nasal troponin reduce injury following MI/R. Modulation of cardiac inflammation by nasal troponin provides a novel treatment to decrease myocardial damage and enhance recovery after myocardial ischemia.

Introduction

Myocardial ischemia–reperfusion (MI/R) injury is a potent stimulus for tissue destruction and subsequent cardiac failure. Myocardial infarction linked to MI/R is one of the most common causes of death in both the Western and the developing worlds (1, 2). During the course of the disease, different cardiac proteins are released into the bloodstream. One such protein is troponin, which is synthesized exclusively in myocardial cells (3, 4). The cardiac troponins form part of the regulatory mechanism for muscle contraction, and a high concentration of troponin in the bloodstream reflects not only cardiac ischemia but also the mass of injured left ventricle (LV), which is a strong predictor of cardiovascular death. Thus, the level of troponin in the bloodstream is a very sensitive marker of myocardial injury (5). In addition, anti-troponin antibodies develop following myocardial injury (6). To our knowledge, there are no reports of a cellular immune response to troponin following myocardial injury.

MI is also associated with an inflammatory response resulting in an accumulation of monocytes/macrophages with reperfusion injury after the MI/R (7–11). Most of those monocytes stain with CD68 or CD11b+, which typically mark macrophages. After a reintroduction and activation of macrophages in post-ischemic myocardium, they adhere to the endothelial surface and migrate into tissues. Activated macrophages can also mechanically block blood flow, ultimately leading to direct tissue injury, such as nitric oxide (NO) formation. They may release pro-inflammatory mediators such as IL-6 and tumor necrosis factor-α (TNF-α), which can directly induce cell death as well as contribute to vessel wall injury, hemorrhage, edema and tissue necrosis, thus amplifying the local inflammatory reaction (8, 12).

Mucosal tolerance is a well-established method whereby, regulatory T cells can be induced to a specific antigen by administering the antigen nasally or orally (13). Upon re-stimulation with the mucosally administered antigen, T cells in mucosally tolerized animals secrete cytokines such as transforming growth factor-β1 (TGF-β1) or IL-10, which are potent anti-inflammatory cytokines with tissue-protective properties (13, 14). Nasal administration of antigen preferentially induces regulatory T cells that secrete IL-10 and is associated with a reduction of CD11b(+) cells (macrophages and macrophage-like cells) at the sites of inflammation (13,15–17). Macrophages may contribute to secondary myocardial infarct expansion by enhancing NO synthesis, which may be reduced by elevated IL-10 levels (8, 18).

The experiments described in this work demonstrate that nasal administration of a cardiac autoantigen troponin affects cellular immune responses to troponin and results in a decrease in myocardial inflammation and an improved outcome after MI/R injury.

Methods

Animals

Female C57BL/6 and C57BL/IL-10−/− mice, 8–10 weeks of age, were purchased from Jackson Laboratory (Bar Harbor, ME, USA) and housed at the Harvard Medical School Animal Care Facilities according to the institutional guidelines. All protocols for animal experiments were approved by local committee and were conducted according to the NIH Guide for Care and Use of Laboratory Animals.

Nasal troponin administration

Nasal administration

Mice were treated nasally with 40 μg porcine troponin together with Emulsome (19) three times every other day. Control mice received BSA protein (Sigma) together with Emulsome or myelin oligodendrocyte glycoprotein (MOG) 35–55 or PBS. For in vitro studies, mice were immunized in the footpad with 100 μg troponin mixed 1:1 with CFA 2 days after the last administration and mice underwent MI/R surgery (as described below) 2 days following the last treatment.

Cell culture of lymphocytes

For proliferation and cytokine assays, spleen cells from mice were pooled and cultured in 96-well plates at 5 × 10⁵ and 106 cells ml⁻¹ (respectively) in serum-free medium X-VIVO 20 (Biowhittaker, Walkersville, MD, USA). To measure cytokines, culture supernatants were collected at 24 h for IL-2 and IL-4; 40 h for IL-6, IL-10 and IFN-γ and at 72 h for TGF-β. For proliferation, cells were pulsed with thymidine at 72 h and radioactivity determined 16 h later (15).

ELISA for cytokines

Quantitative ELISA for IL-2, IL-4, IL-10 and IFN-γ was performed using paired mAb specific for corresponding cytokines as per manufacturer's recommendations (Pharmingen). TGF-β was determined as previously described (15).

MI/R model

A midsternal thoracotomy was performed to expose the anterior surface of the heart. The left anterior descending coronary artery (LAD) was identified and an 8.0 suture (Ethicon) was placed around the artery and surrounding myocardium. Regional left ventricular ischemia was induced for 60 min by ligation of LAD and confirmed by discoloration of myocardium and changes in cardiac rhythm. Sham-operated animals served as surgical controls and were subjected to the same surgical procedures as the experimental animals, with the exception that the LAD was not ligated. At the end of the ischemic period, the ligature was loosened to allow reperfusion. The incision was closed and the animals were allowed to recover (20). Blinding was performed both when the initial LAD ligation was performed and when the re-ligation was performed after 24 h.

Morphometric determination of infarct size

Twenty-four hours after reperfusion, the LAD was re-ligated and 0.3–0.4 ml of 1% Evans Blue in PBS (pH 7.4) was retrogradely injected into the heart to delineate the non-ischemic area. The heart was excised and rinsed in ice-cold PBS. Five to six biventricular sections of similar thickness were made perpendicular to the long axis of the heart and incubated in 1% 2,3,5-triphenyl tetrazolium chloride (TTC, Sigma Chemicals) in PBS (pH 7.4) for 15 min at 37°C and photographed on both sides. Area at risk (AAR) and infarct area were delineated and calculated for both sides of the section. AAR was calculated as the left ventricular area excluding Evans Blue dye after ligation of the LAD. Infarct area was calculated as the risk area that becomes necrotic as distinguished by TTC staining or Masson Trichrome staining. The cumulative areas of ischemic infarct for all sections for each heart were used for comparison. Infarct size was expressed as the ratio of infarct area to AAR (20–22).

Histology and immunohistology

Histology was performed on animals sacrificed 24 or 72 h after ischemia. Frozen tissue sections from mice before and after MI/R injury were fixed in 4% paraformaldehyde over night (O/N) followed by 4.5% sucrose for 4 h and then 20% sucrose for O/N at 4°C. Tissue was frozen in the presence of OCT and stored at −70°C until use. The staining included immunological markers for T cells (CD4), macrophages and neutrophils (CD11b) plus pro-inflammatory (IFN-γ) and anti-inflammatory (IL-10 and TGF-β) cytokines as previously described (15). Sections were evaluated in a blinded manner, and controls included use of isotype-matched mAbs and demonstrated that pre-absorption of anti-cytokine antibodies with respective cytokines (5 μg ml⁻¹, 16 h, 4°C) either blocked or left unchanged the results of antibody staining (15).

Real-time PCR analysis

Real-time PCR [reverse transcription (RT)–PCR] was carried out essentially as described by Zhang et al. (23). Briefly, total RNA was extracted from the LV using an RNeasy kit (Qiagen) according to the manufacturer's instructions. cDNA was synthesized using the Applied Biosystems kit. The primers and probes for the TaqMan RT–PCRs were Mm00443258_m1 (TNF-α), Mm00434165_m1 (IL-12) and Mm00446190_m1 (IL-6). Reactions were performed according to the manufacturer's directions using an Applied Biosystems PRISM 7700 thermal cycler. The mice Actin gene, a housekeeping gene, was used to normalize each sample and each gene.

MTT assay of living cells in cardiac tissue

Cardiac reductase activity of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) to formazan was determined using an MTT assay (24). Cardiac myocyte cultures were established from newborn C57BL/6 mice. To induce oxidative stress, cultures were treated with 200 mM H₂O₂ for 4 h before co-culture with IL-10 (5 ng ml⁻¹) or IFN-γ (5 ng ml⁻¹) O/N and then incubated with the MTT solution. The amount of purple formazan formed was detected and quantified by measuring the absorbance of the solution at 550 nm using a Perkin-Elmer spectrophotometer.

Adoptive transfer of CD4+ T cells

Mice were nasally treated with 40 μg of troponin on days 1, 3 and 5. On day 7, they were immunized with troponin in CFA. Ten days after immunization, lymph nodes and spleens were removed and stimulated in vitro with 40 μg of troponin in 24-well plates (1 ml in each well, containing 5 × 10⁶ cells) in T-cell medium buffer (25) for 2 days. On day 3, the cells were split into two wells and incubated for one more day with IL-2. On day 4, the cells were harvested and the CD4+ T cells were purified by negative selection using a mouse CD4+ T-cell column (R&D Systems, Cat. MCD43). Leukocyte suspensions were incubated with a mixture of mAbs and then loaded onto T-Cell Subset Columns. B cells, CD4-negative T cells and monocytes bind to glass beads coated with anti-immunoglobulin via both F(ab) and Fc interactions. The resulting column eluate contains a highly enriched T-cell subset population with virtually no B cells, monocytes or non-selected T cells. A total of 106 CD4+ T cells were injected in 0.2 ml DMEM intravenously. Naive mice were subjected to adoptive transfer of CD4+ T cells from troponin-tolerized mice 1 h after MI/R surgery. For controls, cells from animals immunized with BSA were treated in an identical fashion.

Carboxyfluorescein succinimidyl ester labeling of CD4+ T cells

CD4+ T cells were isolated from the spleens of naive C57BL/6 mice and labeled with 5 mM carboxyfluorescein succinimidyl ester (CFSE) for 5 min and then cultured in vitro with different doses of troponin in 96-well plates [100 μl in each well, containing 5 × 10⁵ labeled T cells and 1 × 10⁵ radiated antigen-presenting cell (APC) isolated from the same animal] in T-cell medium buffer (25) for 3 days (26, 27). The cells were then stained with anti-CD4-APC and analyzed by flow cytometry. Gates were set to exclude dead cells and 5000 CD4+ CFSE events collected the percentage of CD4+ CFSE dim events. Cells that proliferated in response to troponin, with resultant reduction in CFSE intensity, were measured directly by flow cytometry. The control consisted of foreign proteins contained in the FCS used for in vitro culture.

Echocardiography

Animals underwent echocardiography 6 weeks after the last I/R injury as described (20). Mice were anesthetized with isofluorane for echocardiographic examination. Mice heart rates ranged from 600 to 650 beats min⁻¹ during echocardiography. Echocardiography was performed using a 6–15 MHz Ultraband Intraoperative Linear Array probe and a Sonos 5500 ultrasound imaging system (Philips Ultrasound, Andover, MA, USA). Images were obtained from both the parasternal short and long axes, saved to cine-loop using native frame rates of >60 Hz and analyzed using an offline analysis program. Endocardial borders were traced and end-systolic and end-diastolic areas and volumes were calculated. We used the methodology described by Perrino et al. (28) for echo image analysis. The formula we used to calculate left ventricular ejection fraction (LVEF) was as follows: LVEF = [(LVIDd)3 − (LVIDs)3/(LVIDd)3]. Echocardiographic acquisition and analysis were performed by a technician who was blinded to treatment groups.

Data analysis

All continuous and ordinal data are expressed as mean + SEM. Signification was determined by one-way analysis of variance. Values of P < 0.05 were considered statistically significant; r values were calculated using an Excel statistical program, version 7 (Microsoft Corp.).

Results

Cytokine profile following nasal administration of troponin

Prior to studies in the MI/R model, we investigated cytokine profiles in C57BL/6 mice nasally administered a mixture of three heart-specific porcine troponins (cTnT, cTnI and cTnC), with or without a mucosal adjuvant three times every other day. Control mice received BSA together with adjuvant. Two days after the last treatment, the mice were immunized with 100 μg troponin polypeptide mixed 1:1 in CFA, and splenocytes were taken 10 days later for in vitro assays. As shown in Fig. 1, mice nasally treated with troponin showed a reduction in proliferation (Fig. 1A) and secretion of the pro-inflammatory cytokines IFN-γ and IL-6 (Fig. 1A) (P < 0.03). The optimal dose per mouse was found to be 40 μg (data not shown). Administration of troponin in combination with an emulsome-based mucosal adjuvant (19) resulted in a greater decrease in proliferation (Fig. 1A) and a greater reduction of IL-6 compared with troponin alone (P < 0.05). No active TGF-β was detected.

Fig. 1.

Effect of nasal troponin on heart infarct size 24 h following MI/R injury. (A) C57BL/6 mice were nasally administered 40 μg troponin with or without a nasal adjuvant every other day for three treatments. Control animals received BSA. Ten days after immunization, popliteal lymph node cells were cultured with 100 μg troponin for 72 h. Proliferation was measured by pulsing with (26) thymidine. Cytokine profile of splenocytes from animals treated nasally with troponin and measured by ELISA taken 24–40 h after culture. (B) Mice treated three times with nasal troponin and subjected to MI/R injury 2 days later. At 24 h after MI/R injury, TTC staining was performed to assess infarct size. The pale regions are outlined and denote the infarcted tissue. The sections presented are from the LV of representative individual mice. Infarct volume is significantly less in nasal troponin-treated mice (n = 12) compared with BSA-treated mice (n = 20, P < 0.005). (C) For proliferation, splenocytes from treated animals were pulsed with (26) thymidine at 72 h and radioactivity determined 16 h later. There was a significant reduction of proliferation against troponin in the treated mice versus control (P < 0.03).

In order to investigate the specificity of the immune modulation following nasal troponin, we measured responses to troponin in cervical lymph nodes in animals nasally administered troponin, BSA or MOG peptide 35–55 and then immunized with troponin. As shown in Fig. 1(A), only nasal troponin resulted in a reduction of IFN-γ and IL-6 and an increase in IL-10 (P = 0.003). If, however, animals were nasally treated with BSA, immunized and challenged in vitro with BSA, IL-10 increased from 118 + 9 to 198 + 11 (P = 0.005). If nasal MOG was used, IL-10 increased from 158 + 12 to 248 + 20 (P = 0.007). These results demonstrate that the specific induction of IL-10 related to antigen administered nasally and that all antigens tested were able to induce IL-10 responses when given nasally.

Having shown the immunologic effects of mucosally administered troponin in C57BL/6 mice, we then proceeded to test mucosal troponin in the MI/R model. For all the experiments in the MI/R model, we administrated troponin in combination with the emulsome-based mucosal adjuvant as there was a greater reduction of IL-6 when troponin was given with adjuvant. IL-6 is a pro-inflammatory cytokine, which plays an important role in ischemia-induced myocardial damage (8, 29).

To test for immunologic cross-reactivity between porcine and murine troponin, mice were nasally treated with 40 μg of either murine or porcine troponin or BSA on days 1, 3 and 5. On day 7, they were immunized with porcine troponin in CFA. Ten days after immunization, spleens were removed and stimulated in vitro with 50 μg of porcine troponin. Mice nasally treated with porcine or murine troponin showed a similar reduction in: (i) proliferation: porcine = 8898.7 + 711 counts per minute (c.p.m.), murine = 7395 + 433 c.p.m. versus control = 14 001 + 250 c.p.m. (P < 0.005 for both porcine and murine versus control); (ii) IFN-γ secretion: porcine = 2866.7 + 145 pg ml⁻¹, murine = 2216 + 116 c.p.m. versus control = 3783 + 117 c.p.m. (P < 0.05 for both porcine and murine versus control) and (iii) IL-6 production: porcine = 407 + 7 pg ml⁻¹, murine = 403 + 13 c.p.m. versus control = 536 + 23 c.p.m. (P < 0.05 for both porcine and murine versus control). This demonstrates an immunologic cross-reactivity between porcine and murine troponin.

Nasal vaccination with troponin reduces tissue damage after MI/R

To investigate whether nasal administration of troponin had an effect on myocardial tissue injury, we administered troponin nasally three times every other day for 1 week. C57Bl6 mice were randomly divided into two groups and were nasally administered with either troponin or BSA (control) as described above. Two days after the last treatment, animals underwent 60 min of myocardial ischemia followed by 24 h of reperfusion. The myocardial infarct sizes were determined by TTC staining. As shown in Fig. 1(B), nasal administration of troponin reduced ischemic infarct size by 49%, 24 h following the MI/R surgery (from 13.4 ± 1.6%, n = 20 to 6.1 ± 0.9%, n = 12, P < 0.005 versus BSA-treated animals). To further investigate the specificity of the protection, we nasally treated animals (n = 11) with a non-cardiac autoantigen (MOG 35–55 peptide) and induced myocardial ischemia. No protection was observed with nasal MOG (MOG: 10.1 ± 1.4% versus BSA: 13.4 ± 1.6%, non-significant; troponin 6.1 + 0.9% versus MOG, P = 0.023). These results demonstrate the cardiac specificity of the autoantigen required for protection following myocardial tissue injury.

To investigate the immunologic effects of nasal troponin, we measured the reactivity of splenocytic cells to troponin 24 h after surgery. The level of troponin is known to correlate with the level of myocardial disease in vivo both in mice and humans (21). Troponin-specific cellular immune responses occur in association with MI/R injury and as shown in Fig. 1(C), there was a significant reduction in proliferation against troponin in nasally troponin-treated mice versus nasal BSA controls (5658 ± 282 c.p.m. versus 1386 ± 84 c.p.m., P < 0.03). Following nasal troponin adminstration, the level of proliferation against troponin was similar to basal levels (842 ± 43 c.p.m.). This data demonstrate that nasal administration of troponin prior to MI/R induction modulates the cellular immune response to this cardiac antigen, an effect which is correlated with a reduction in infarct size (Fig. 1B and C). The effect was not due to an antigen non-specific effect following nasal therapy, as it was not observed with nasal BSA treatment. Furthermore, this data support the use of reactivity against troponin as an indicator of inflammation following MI/R.

Nasal vaccination with troponin improves survival and heart function after MI/R injury

Having shown an acute effect of nasal troponin in MI/R injury, we then evaluated the long-term consequences of prior nasal troponin administration on the myocardium following MI/R injury. Animals were divided into two groups and were nasally administered with either PBS or troponin three times over a period of 1 week. Following nasal troponin, animals underwent MI/R injury. Six weeks after injury, animals underwent echocardiographic evaluation to assess wall thickness, motion abnormalities and overall cardiac function. The results are presented in Table 1. During the 6-week period, four mice died in the control group (n = 9), whereas only one died in the treated group (n = 7). As shown in Table 1, mice subjected to MI/R injury showed severe ventricular wall thinning, as demonstrated by reduced posterior wall thickness and intraventricular septal wall thickness, reduced ejection fraction %) and mean velocity of fractional shortening corrected for heart rate. However, mice that were treated nasally with troponin showed little damage as measured by echocardiographic function. M-mode images from the myocardium are shown in Fig. 2(A). These results were further confirmed by histopathology as morphological analysis by hematoxylin–eosin and masson trichrome staining demonstrated that nasal troponin-treated animals had significantly reduced fibrosis and collagen deposition (Fig. 2B; normal myocardium is red, whereas fibrosed tissue appears blue). Nasal administration of troponin reduced ischemic infarct size as measured by collagen deposition by 78% at 6 weeks following the MI/R injury (from 30 ± 4.5% to 9.6 ± 3.5%, P < 0.005) (Fig. 2C). Thus, nasal troponin protects against structural and functional myocardial damage.

Table 1.

Effect of nasal troponin on cardiac function as measured by echocardiography

	Sham (n = 6)	PBS (n = 5)	Troponin (n = 6)
LVDd (mm)	3.1 ± 0.08	3.5 ± 0.11a	3.2 ± 0.07b
PWT (mm)	1.1 ± 0.03	0.79 ± 0.03a	0.87 ± 0.02b
IVSW (mm)	0.861 ± 0.02	0.78 ± 0.02a	0.89 ± 0.02b
mVcfc	5.6 ± 0.2	4.6 ± 0.1a	5.4 ± 0.2b
EF%	91 ± 3	51 ± 2.5a	63 ± 1.5b
HR (bpm)	666 ± 50	662 ± 55	596 ± 60

LVDd, left ventricular diameter (diastole); PWT, posterior wall thickness; IVSW, intraventricular septal wall thickness; mVcfc, mean velocity of fractional shortening corrected for heart rate; EF, ejection fraction; HR, heart rate; mVcfc, mean velocity of fractional shortening corrected for heart rate.

^aVersus sham, P < 0.05.

^bVersus PBS, P < 0.05.

Fig. 2.

Nasal vaccination with troponin protects against tissue injury. (A) M-mode images from the myocardium. The arrow represents the increased gap in the LV due to myocardium damage compared with the troponin-treated animals. (B) Masson trichrome staining was performed to assess collagen deposition (blue regions) that occurs after tissue death (magnification ×5). There is less myocardial tissue and more collagen deposition in the control group. Hematoxylin–eosin (H&E) staining (magnification ×20) was performed to assess tissue integrity and demonstrates less myocardial injury in treated animals. Arrows identify collagen deposition. (C) Quantification of percentage of collagen in the LV shows significant (P < 0.05) anterior wall thinning and collagen deposition 1.5 months after injury in control mice (n = 5) versus troponin-treated mice (n = 6), in which the myocardium appeared normal with decreased collagen deposition.

Immunohistochemistry of the heart following nasal vaccination with troponin

Surrounding the lethally damaged core of the myocardial infarct is an area of ischemic damage with partially preserved energy metabolism (30). With time, and in absence of any treatment, this area progresses to infarction owing to ongoing excitotoxicity, post-ischemic inflammation and apoptosis (30). Thus, a prime goal of therapy is to protect the myocardium from ischemia-induced inflammation (18). To further investigate the effect of nasal troponin, immunohistochemical analysis of cytokines was performed. CD4+ and CD11b+ cells, and cells expressing IFN-γ and IL-10, were quantified at 24 h after MI/R injury (Fig. 3A and B). We found a significant reduction in macrophage-type cells (CD11b+) as detected by immunostaining of heart sections at 24 h after surgery (from 389 + 36 to 115 + 12, P < 0.002) (Fig. 3B). In addition, there was a decrease in the number of CD4+ cells expressing IFN-γ in the nasal troponin group (from 55 + 7% to 17 + 3%, P < 0.01) and an increased in the percentage of CD4+ cells expressing intracellular IL-10 was increased in the nasal troponin group (from 17 + 2% to 78 + 8%, P < 0.002). No difference in CD4+ cells expressing of TGF-β was observed. Thus, animals nasally vaccinated with troponin had enhanced expression of the anti-inflammatory cytokine IL-10 and reduced expression of the pro-inflammatory cytokine IFN-γ in the ischemic region.

Fig. 3.

Immunohistology of LV 24 h after ischemic cardiac injury. (A) Serial sections of hearts from control or nasal troponin-treated mice 24 h after MI/R injury were labeled using mAbs directed against CD4, CD11b, IFN-γ or IL-10. For CD4+ cells, staining shows immunoperoxidase (magnification ×5) on a background of hematoxylin staining; arrows identify regions of CD4+ infiltration. For immunofluorescence (magnification ×20), arrows identify co-localization (yellow) of CD4+ cells (red) with either IFN-γ or IL-10 (green). (B) Quantification of cell populations. Results represent blinded analysis of three samples per group stained in the area of the LV. Results are in standard error of the mean troponin versus control: for CD11b (P < 0.002), CD4 (P < 0.01); for CD4+IFN-γ+( P < 0.01) and CD4+IL-10+ (P < 0.002). (C) RT–PCR from the area of the infarct in LV showed a significant reduction in the pro-inflammatory cytokines IL-6 and TNF-α (P < 0.05).

We further investigated the inflammatory milieu in cardiac tissue by performing gene expression analysis using semi-quantitative RT–PCR on RNA isolated from the infarcted left ventricular region. As shown in Fig. 3(C), 24 h after injury, the nasal troponin-treated group showed a significant reduction in the pro-inflammatory cytokines IL-6 and TNF-α (P < 0.05) that was associated with the reduction in macrophages (CD11b+) in the nasal troponin-treated group (Fig. 3B). Taken together, this data demonstrate that troponin-treated animals had significantly reduced fibrosis and normalized cardiac function compared with control animals and thus, nasal troponin protects against structural and functional myocardial damage.

IL-10 protects cardiac myocytes from death following oxidative stress. To investigate potential mechanisms by which IL-10 exerts its protective effect on cardiac myocytes following nasal troponin, we measured the effect of IL-10 on oxidative stress in a model of oxidative stress using hydrogen peroxide in myocardial cell cultures. Oxidative stress can lead to myocardial damage caused by reactive oxidative species (ROS), such as superoxide anion and hydrogen peroxide. Mitochondrial oxidative phosphorylation is a major source of energy to the cell and may thus be subject to direct attack by ROS (31). Mitochondrial membrane lipids, proteins and nucleic acids are all subject to ROS-mediated damage. Measurement of MTT has been used to assess myocyte cell function (24). The MTT assay measures only mitochondrial reductase enzymes, and thus is directly related to the number of viable cells. As shown in Table 2, exposure of myocytes to hydrogen peroxide (200 μM) for 4 h caused a significant decrease in cardiac mitochondrial function, as measured by the MTT assay (24). The percent of cell death following H₂O₂ treatment was 39.4 + 3.1% and this was reduced to 13.3 + 3.6% following treatment with IL-10 (P < 0.01). No effect was observed with IFN-γ. Thus, in addition with being a potent inhibitor of monocyte/macrophage activation, IL-10 has a protective effect on myocytes.

Table 2.

IL-10 protects cardiac myocytes from death following oxidative stress

Condition	% Cell death (MTT)
No stress	0
H2O2	39.4 ± 3.1
H2O2 + IL-10	13.3 ± 3.6a
H2O2 + IFN-γ	33.2 ± 5.7

Cardiac myocyte cultures were established from newborn C57BL/6 mice as described (24). To induce oxidative stress, cultures were treated with 200 mM H₂O₂ for 4 h before co-culture with 5 ng ml⁻¹ IL-10 or 5 ng ml⁻¹ IFN-γ. Results are expressed as standard error of the mean.

^aIL-10 versus H₂O₂ (P = 0.005); IL-10 versus IFN-γ (P = 0.02).

Nasal vaccination with troponin after MI/R reduces tissue damage

To investigate whether nasal troponin affects myocardial tissue damage if administered after the ischemic insult, troponin was administered nasally 1 h after MI/R injury. As shown in Fig. 4(A), in two independent experiments, troponin administration following ischemic injury reduced infarct size by 46.9% (from 31.3 ± 1.3% to 16.6 ± 2.6%, P <0.004) and by 39.8% (from 15.1 ± 1% to 9.1 ± 0.7%, P <0.01) at 24 h following the MI/R injury. The difference in infarct size in the control groups relates to the variability in the placement of sutures. To investigate whether the reduction in infarct size was associated with reduction in splenocyte proliferation, we measured proliferation to troponin. As shown in Fig. 4(B), there was significant reduction of proliferation against troponin in the treated mice versus control (6330 ± 430 c.p.m. versus 2929 ± 120 c.p.m., P < 0.001). RT–PCR was also carried out using splenocytes obtained 24 h after surgery. The troponin-treated group showed a significant (P < 0.02) reduction in proinflammatory cytokines IL-12 and TNF-α (Fig. 4C) that correlated with the observed reduction in splenocyte proliferation (Fig. 4B). These results demonstrate a correlation between the reduction in the cellular peripheral immune response to troponin with the reduction of ischemic infarct size (Fig. 4A). We then examined the level of pro-inflammatory cytokines in the heart when animals were treated 1 h after MI/R injury. We performed RT–PCR for IL-6 and TNF-α from the left ventricular infarct region 24 h following surgery. The treated group showed a significant reduction in pro-inflammatory cytokines IL-6 and TNF-α (P < 0.05) (Fig. 4C) that correlated with the reduction in ischemic infarct size and reduction in the peripheral immune response to troponin. To further investigate the effect of nasal troponin, immunohistochemical analysis was performed. CD4+ cells expressing IFN-γ and IL-10 were quantified at 24 h after MI/R injury (Fig. 5A and B). We found a significant decrease in the number of CD4+ cells expressing IFN-γ in the nasal troponin group (from 65 + 8% to 15 + 6%, P < 0.01) and an increase in the percentage of CD4+ cells expressing intracellular IL-10 in the nasal troponin group (from 14 + 2% to 72 + 7%, P < 0.02). Thus, animals nasally vaccinated with troponin had enhanced expression of the anti-inflammatory cytokine IL-10 and reduced expression of the pro-inflammatory cytokine IFN-γ in the ischemic region of the myocardium.

Fig. 4.

Nasal vaccination with troponin 1 h after MI/R injury reduces tissue damage. (A) Infarct volume is significantly less in mice that received troponin 1 h following MI/R injury. Experiment I: nasal troponin (n = 7, P < 0.004) compared with (n = 4) mice that received PBS; Experiment II: nasal troponin (n = 6, P < 0.05) compared with (n = 5) mice that received PBS. The difference of infarct size in the control groups in Experiment I versus Experiment II may relate to the variability in the placement of sutures. (B) Twenty-four hours after administration of troponin, mice were sacrificed, splenocytes were cultured with troponin and proliferation was measured. (C) Correlation between troponin proliferation and infarct volume in individual animals (r = 0.86). (D) RT–PCR analysis of the expression of TNF-α and IL-12 in splenocytes and TNF-α and IL-6 in the LV from control versus troponin-treated animals (P < 0.05).

Fig. 5.

Immunohistology of LV 24 h after ischemic cardiac injury following 1-h treatment with troponin. (A) Immunofluorescence arrows identify co-localization (yellow) of CD4+ cells with either IFN-γ or IL-10 (magnification ×40). (B) For quantification of cell populations, results represent blinded analysis of three samples per groups stained in the area of the LV. Results are in standard error of the mean troponin versus control: for IFN-γ+( P < 0.01), IL-10 (P < 0.02).

The cellular immune response to troponin that we observed after MI/R injury and the effect of nasal troponin given 1 h after MI/R suggested that underlying endogenous reactivity to troponin existed and was being boosted both by myocardial damage and by nasal troponin. To demonstrate endogenous reactivity to troponin, we isolated CD4+ T cells from naive C57BL/6 mice and measured their reactivity to troponin using CFSE labeling. CFSE labeling is a more sensitive measure than simple thymidine incorporation and is able to detect and quantify low levels of autoantigen-specific T cells (26, 27). As shown in Fig. 6, naive splenic CD4+ T cells labeled with CFSE and cultured in vitro with troponin responded to troponin as measured by reduction in CFSE intensity by flow cytometry. Figure 6 (A) shows flow cytometry and Fig. 7(B) shows mean fluorescence intensity (MFI). We found significant cell proliferation in response to troponin as measured by percentage of dividing cells (troponin: 13.5 + 0.8% versus control: 6.6 + 0.4%, P < 0.002) and reduction in CFSE MFI in response to troponin (130 + 8 versus control: 71 + 5, P = 0.004). This was seen at the 100-μg ml⁻¹ dose, the dose used to demonstrate proliferation to troponin after MI/R and its reduction following nasal troponin (Figs 1C, 4B and 7D). The control consisted of foreign proteins contained in the FCS used for in vitro culture. These results thus demonstrate the presence of underlying endogenous reactivity to troponin in naive animals.

Fig. 6.

Endogenous reactivity to troponin in C57BL/6 mice. CD4+ T cells were isolated from the spleen of naive C57BL/6 mice, labeled with CFSE and cultured in vitro with 10 or 100 μg ml⁻¹ troponin in the presence of radiated APC isolated from the same animal for 3 days. The cells were then stained with anti-CD4-APC and analyzed by flow cytometry. Gates were set to exclude dead cells. Five thousand CD4+ CFSE events were collected and the percentage of CD4+ CFSE dim events were determined. (A) Flow cytometry of CD4+ CFSE dim T cells expressed as a percentage. (B) MFI of CSFE-labeled cells.

Fig. 7.

Adoptive transfer of CD4+ T cells from nasal troponin-treated mice reduces tissue damage following MI/R injury. (A) Scheme for adoptive transfer of CD4+ T cells from troponin-tolerized mice to naive mice 1 h after MI/R surgery. (B) FACS analysis of CD3+/4+ cells following column enrichment. (C) Mice that received 1 × 10⁶ CD4+ T cells from nasal troponin-treated mice wild type (n = 5) had a lower infarct volume (P < 0.05) than mice (n = 5) that received 1 × 10⁶ CD4+ T cell from nasal BSA-treated mice or 1 × 10⁶ CD4+ T cell from nasal troponin-treated IL-10−/− mice (n = 7) or BSA-treated IL-10−/− mice (n = 7). (D) Twenty-four hours after intravenous administration of CD4+ T cells, mice were sacrificed and their splenocytes cultured with troponin and proliferation was measured.

Adoptive transfer of CD4+ T cells from mice nasally treated with troponin reduces heart ischemic infarct size when given 1 h post-MI/R injury

To investigate the role of CD4+ cells in reduction of myocardial infarct size following treatment with nasal troponin, adoptive transfer experiments were performed. As shown in Fig. 7(A), mice were nasally treated with 40 μg of troponin or BSA (control mice) on days 1, 3 and 5. On day 7, the mice were immunized with troponin in CFA or BSA in CFA. Ten days after immunization, both the lymph nodes and spleens were removed and stimulated in vitro with 40 μg of troponin or BSA in 24-well plates. After 4 days of in vitro culture, CD4+ T cells were purified by negative selection and adoptively transferred to mice 1 h following MI/R injury. The purity of recovered CD3+CD4+ cells was 90.8% with no detectable CD3+CD8+ cells (Fig. 7B). Adoptive transfer of CD4+ T cells taken from nasal BSA-treated mice that were immunized with BSA in CFA served as control. As shown in Fig. 7(C), heart infarct size was reduced by 72% (from 18 ± 6% to 5 ± 1.2%, P < 0.05) in animals that received CD4+ T cells from mice nasally treated with troponin, as compared with animals that received CD4+ T cells from nasal BSA-treated mice.

To establish that IL-10 was also crucial in these adoptive transfer experiments, CD4+ T cells were adoptively transferred from nasally treated IL-10−/− animals. As shown in Fig. 7(C), no reduction of ischemic infarct size was observed when CD4+ T cells from nasal troponin-treated IL-10−/− animals were transferred. Thus, nasal troponin reduces ischemic size via IL-10-dependent CD4+ T cells.

We also tested immune responses to troponin following adoptive transfer. As shown in Fig. 7(D), there was a significant reduction of proliferation against troponin in the mice that received CD4+ T cells from nasal troponin-treated animals versus controls (3912 ± 340 c.p.m. BSA versus 1693 ± 120 c.p.m. control, P < 0.03). The level of proliferation against troponin in treated mice was similar to the basal level proliferation. These results demonstrate that nasal troponin induces troponin-specific anti-inflammatory CD4+ T cells, which function to protect the myocardium following MI/R injury, and like nasal troponin, given 1 h post-infarct, these cells are effective when given 1 h post-MI/R injury.

Discussion

Ischemia and subsequent reperfusion leads to myocardial injury through a variety of mechanisms, one of which involves an inflammatory response in the damaged myocardium. Myocardial ischemia is a common phenomenon in patients with coronary artery disease and experimental models of myocardial ischemia provide a platform for investigating new modes of therapy. We show here that nasal vaccination with troponin is effective in reducing MI/R injury by inducing CD4+ T cells that act by secreting IL-10 and reducing inflammation at the site of myocardial injury.

Currently, therapies for ischemic heart disease are directed at the rapid restoration of blood flow to the ischemic region (7). However, during reperfusion, the heart undergoes further damage due in large part to the generation of ROS, e.g. superoxide anion, elevated levels of which can be detected within minutes after the reintroduction of oxygen to ischemic tissues. ROS have been shown to be key mediators of cellular and myocardial injury, causing lipid peroxidation and apoptosis (7, 8). The innate immune response to ischemia–reperfusion injury is the most common cause of myocardial inflammation (8). Although innate immune responses may preserve myocardial function in the short term, they may be maladaptive in chronic states leading to production of ROS (8). In addition as demonstrated here, there is an adaptive immune response directed against specific myocardial antigens such as troponin that occurs after MI/R injury. Thus, modulation of the immune response following MI/R injury appears important in reducing of tissue damage.

Because we used porcine troponin for nasal administration to mice, we established that there was functional immunologic cross-reactivity between porcine and murine troponin by demonstrating that both murine and porcine troponin given nasally affected immune responses in animals immunized with porcine troponin. Furthermore, BLAST comparison of the sequences of porcine and murine troponin-I showed 95% homology.

The presence of recruited leukocytes at the site of inflammation is dependent upon the coordinated expression of adhesion molecules on inflammatory cells and the activated capillary endothelium. T cells are re-stimulated upon encounter with the target immunogen presented by local APCs (7, 9). Thus, several types of APCs are activated following heart injury, including dendritic and macrophage-like cells, which express MHC molecules and produce pro-inflammatory cytokines such as TNF-α and IL-12, which may enhance the appearance of adhesion molecules (32, 33). Furthermore, pro-inflammatory cytokines play a direct role in tissue damage following myocardial damage: IFN-γ halts collagen synthesis by smooth muscle cells; TNF-α contributes to post-ischemic myocardial dysfunction via induction of myocyte apoptosis (34, 35). In line with this, following MI/R injury, we found macrophage-type cells (CD11b) by immunostaining of heart sections at 24 h, which serve to enhance the destructive effect of infiltrating CD4+ T cells. In animals treated with nasal troponin, there was decrease of these CD11b cells.

Both cardiac and skeletal muscles are exquisitely controlled by changes in the intracellular calcium concentration. Troponin is part of thin filament (along with actin and tropomyosin) and is the protein to which calcium binds to accomplish this regulation. Troponin has three subunits, TnC, TnI and TnT. When calcium is bound to specific sites on TnC, the structure of the thin filament changes in such a manner that myosin attaches to thin filaments and produces force and/or movement. Troponin is synthesized exclusively in myocardial cells. While normal levels of cardiac troponin-I are ∼10 ng ml⁻¹, in patients with acute myocardial infarction, serum cardiac troponin-I is elevated within 4–6 h, reaches a mean peak level of 112 ng ml⁻¹ at 18 h and remains above normal for up to 6–8 days following infarction (36). Moreover, it has been shown that an autoimmune response to troponin induces severe inflammation in the myocardium followed by fibrosis and heart failure with increased mortality in mice (37). We observed that increased cellular proliferative responses to troponin occur in the spleen 24 h after MI/R injury and that reduction in proliferation to troponin was associated with protection by nasal vaccination with troponin. Thus, high concentrations of troponin following cardiac ischemia may lead to an injurious adaptive inflammatory response directed at the ischemic site following ischemic reperfusion injury and measurement of cellular immune responses to troponin can serve as an indicator of protection by nasal troponin.

An important question is why does troponin given nasally 1 h after I/R induce protection whereas troponin released into the circulation at the same time after injury afford no protection? We believe that different immune mechanisms are induced following mucosal versus intravenous delivery of antigen. We previously found that mucosal (oral) but not intravenous alloantigen resulted in increased T_h2 cell activation in cardiac allografts (38). Thus, we believe than troponin released from the circulation does not afford protection as it does not induce troponin-specific IL-10-secreting T cells whereas nasal troponin does. Troponin released into the circulation induces anergy of troponin-reactive T cells, whereas mucosal antigen induces IL-10-secreting T cells due to the unique immunologic properties of the nasal mucosa in inducing IL-10-specific immune responses (14). Because such regulatory T cells are triggered in an antigen-specific fashion but suppress via cytokine release in an antigen-non-specific fashion, they mediate ‘bystander suppression’ when they encounter the nasal autoantigen at the target organ. Thus, mucosal tolerance can be used to treat inflammatory processes that are not autoimmune in nature via the secretion of cytokines such as IL-10 after antigen-specific triggering (16). Furthermore, using CFSE labeling, we found endogenous reactivity of CD4+ T cells to troponin in naive animals. These findings, together with reports (18) that showed elevation of IL-10 secreting T cells following MI/R injury in dog, suggest that underlying cellular endogenous reactivity to troponin is being boosted both by myocaridal damage and by nasal troponin.

Although we have studied troponin, it is known that immune responses have been reported to other contractile proteins such as myosin. Murine myocarditis may be induced by immunization with cardiac myosin (39) and cellular responses to both mysosin and troponin have been observed in autoimmune myocarditis (37, 40). In addition, auto-antibodies to both troponin and myosin may induce myocarditis (41) and cardiac auto-antibodies may play a role in dilated cardiomyopathy which has led to initial clinical testing of Ig-adsorption therapy (42).

Anti-inflammatory cytokines may have a protective effect following the inflammation that occurs after MI/R injury. It has been suggested that IL-10 inhibits inducible nitric oxide synthase (iNOS) activity after MI/R and consequently exerts cardioprotective effects (18). Studies have also shown that TGF-β can attenuate myocardial injury induced by I/R, though we did not find increased TGF-β in our studies (43). In animals nasally vaccinated with troponin, we found enhanced expression of the anti-inflammatory cytokine IL-10 and reduced expression of the pro-inflammatory cytokine IFN-γ in the ischemic region. It has recently been shown that IFN-γ parallels iNOS activity during the course of ischemia following reduction in the blood flow (8). IL-10 is the mediator of the protection we observed following nasal vaccination with troponin. Furthermore, it has been shown that IL-10−/− mice have larger ischemic injury as compared with wild type animal following MI/R injury (44). We have previously shown that IL-10-secreting CD4+ T cells induced by nasal MOG (35–55) reduce injury ischemic injury following stroke (15). In addition, we observed a dramatic reduction of CD11b(+) cells (macrophage) in nasal MOG-treated animals. CD11b(+) cells may contribute to secondary infarct expansion by enhancing NO synthesis that may be reduced by IL-10 levels. IL-10 has also been shown to reduce inflammation in autoimmune animal models including experimental autoimmune encephalomyelitis (25). IL-10 may deactivate macrophage-like cells and thus limit their involvement in a secondary inflammatory process. Moreover, IL-10 targets the interface between heart and periphery by preventing adhesion and extravasation of leukocytes. Finally, we have demonstrated that IL-10 protects cardiac myocytes from oxidative stress in vitro, using a myocardial cell culture system.

Administration of antigen by the mucosal route is known to induce regulatory T cells (16). We have found that Tr1 type Tregs that primarily act via IL-10 are induced by the nasal route and it appears that this is the type of regulatory T cell induced by nasal troponin. In a recent study in which we induced Tregs by nasal anti-CD3 to suppress models of lupus, we induced an IL-10-secreting CD4+ CD25- LAP+ regulatory T cell which also shared the properties of Tr1 cells. (45) We did not observe an increase in Foxp3 expression in these cells.

Given our finding that IL-10 is responsible for the effect observed, it is theoretically possible that intravenous infusion of IL-10 post-infarction might have a beneficial effect. However, large amounts of systemic IL-10 may be required for IL-10 to be active in the heart and there may be systemic side-effects. CD4+ T cells from nasal troponin-treated mice secrete IL-10 when they encounter troponin in the myocardium. Thus, it is targeted local delivery of IL-10. As discussed above, because regulatory T cells are triggered in an antigen-specific fashion but suppress in an antigen-non-specific fashion, they mediate bystander suppression when they encounter the nasal autoantigen at the target organ (16). Nasal troponin also has the advantage that it could be given prophylactically to patients at risk for cardiac damage. Whether intravenous IL-10 would be a better therapeutic agent than nasal troponin given immediately after myocardial damage both in terms of efficacy and toxicity is not known.

We were surprised to find that nasal immunization with troponin 1 h after MI/R could lead to an effective immune response within a 24-h time period. Nonetheless, we found immune responses in the spleen to troponin 24 h after injury as measured by proliferation. Moreover, both in human and mouse, there is detectable level of troponin in the blood (36) and since we found endogenous troponin-specific CD4+ T cells in naive animal, we believe this explain the protective effect of nasal immunization with troponin 1 h after MI/R. Consistent with this, we found immune changes as measured by RT–PCR both in the spleen and in the heart 24 h after MI/R injury and these changes were modulated by nasal troponin. Furthermore, this modulatory effect was lost in IL-10-deficient mice, clearly showing the role of IL-10 in the immunologic effects seen at 24 h and is supported by transfer of protection that is also dependent on IL-10. We believe that these findings suggest that there may be underlying endogenous reactivity to troponin that is being boosted both by myocardial damage and by nasal troponin even though we did not find measurable proliferative responses to troponin in naive animals.

In line with our results, other anti-inflammatory approaches are being investigated for cardioprotection in acute myocardial infarction. It has been shown that activation of macrophages by complement can exacerbate cardiac injury following ischemic insult. C-reactive protein (CRP) enhances heart failure by activation of complement and increasing myocardial and infarct size (46, 47), and administration of CRP inhibitors reduces heart infarct size and improves cardiac function (46).

Our results not only demonstrate the presence of a cellular immune response to troponin in response to damage after MI/R injury but they also have clear clinical implications. First, our results suggest prophylactic treatment with nasal troponin may be of benefit to those at a risk for ischemic cardiac injury including subjects undergoing cardiac bypass surgery or coronary angioplasty. Nasal troponin treatment significantly improved heart function after MI/R injury, and this effect remained significant up to 6 weeks following heart ischemia. Second, we have shown that nasal vaccination 1 h following MI/R can significantly reduce ischemic damage, making this approach clinical applicable for patients in the immediate period following myocardial infarction. Thus, nasal vaccination with troponin is a novel therapeutic intervention for treatment of cardiac ischemia both in those in risk and in the period immediately following cardiac injury.

Funding

National Institutes of Health (AI-43458) to H.L.W.; Human Frontier Science Program Organization to D.F.

Glossary

Abbreviations

AAR

area at risk

APC

antigen-presenting cell

CFSE

carboxyfluorescein succinimidyl ester

c.p.m.

counts per minute

CRP

C-reactive protein

iNOS

nitric oxide synthase

LAD

left anterior descending coronary artery

LV

left ventricle

LVEF

left ventricular ejection fraction

MFI

mean fluorescence intensity

MI/R

myocardial ischemia–reperfusion

MOG

myelin oligodendrocyte glycoprotein

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide

NO

nitric oxide

O/N

over night

ROS

reactive oxidative species

RT

reverse transcription

TGF-β

transforming growth factor-β

TNF-α

tumor necrosis factor-α

TTC

2,3,5-triphenyl tetrazolium chloride