In Vivo Visualization of Cardiac Allograft Rejection and Trafficking Passenger Leukocytes Using Bioluminescence Imaging
Background— We investigated the feasibility of bioluminescence imaging (BLI) for the in vivo assessment of cardiac allograft viability and visualization of passenger leukocytes during the course of acute rejection.
Methods and Results— Hearts of FVB (H-2q) luciferase-green fluorescent protein transgenic mice (β-actin promoter) or FVB luciferase transgenic mice (CD5 promoter) were heterotopically transplanted into either BALB/c (H-2d) or FVB recipients. Light intensity emitting from the recipient animals was measured daily by in vivo BLI until 12 days after transplantation. Graft beating score (0 to 4) was assessed by daily abdominal palpation until 12 days after transplantation. Inflammatory cell infiltration (CD45 stain) and structural changes of green fluorescent protein-positive cardiomyocytes were followed by immunohistochemistry. All cardiac allografts were acutely rejected by 12 days after transplantation. The intensity of light emitting from cardiac allografts declined 4 days after transplantation and correlated with graft beating scores (R2=0.91, P=0.02). Immunohistochemistry confirmed these results by showing an increase of CD45+ inflammatory cell infiltration and destruction of green fluorescent protein-positive cardiomyocytes in the cardiac allografts during acute rejection. In vivo BLI visualized migration and proliferation of CD5+ passenger leukocytes in both syngeneic and allogeneic recipients. In the allograft recipients, light signal from CD5+ passenger leukocytes peaked at 6 hours and diminished by 12 hours, whereas in the syngeneic recipients, the signal remained high until 10 days after transplantation.
Conclusions— BLI is a useful modality for the quantitative assessment of in vivo cardiac graft viability and tracking of passenger leukocytes in vivo during the course of acute rejection.
Allogeneic heart transplantation has become a commonly used therapy for end-stage heart disease. Early graft loss due to acute rejection, though reduced by current immunosuppressive strategies, remains a significant problem in clinical heart transplantation.1 In addition, many cardiac allograft recipients experience episodes of acute rejection, and these episodes are a critical risk factor for the subsequent development of graft coronary artery disease, which predisposes a patient to late graft loss.2 Acute rejection is an immune response mediated by the coordinated infiltration and effector functions of host alloantigen-specific T cells in the allograft.3 In addition, donor-derived passenger leukocyte migration contributes to acute rejection.4
Cell migration is a crucial element during the development of the immune system and it mediates the immune response during acute cardiac rejection. There is extensive and continual redistribution of cells to different anatomic sites throughout the body. These trafficking patterns control immune function and host responses to the transplanted heart. The ability to monitor the fate and function of migrating cells is therefore imperative to both understanding the role of lymphocytes in acute cardiac rejection and to devising rational therapeutic strategies. Determining the fate of immune cells and understanding the functional changes associated with migration and proliferation requires effective means of obtaining in vivo measurements in the context of intact organ systems.
We have developed in vivo bioluminescence imaging (BLI) on the basis of the observations that light passes through mammalian tissues and that luciferase can serve as an internal biological source of light in the living body.5 This method is a rapid and noninvasive functional imaging method that uses light-emitting reporters and external photon detection to follow biological processes in living animals in real time. Using this approach, we have elucidated the spatiotemporal trafficking patterns of malignant cells, lymphocytes, and other mature immune cells within living animal models of human biology and disease.5–8 In addition, to enable the in vivo study of biological processes that involve movement of cells within the 3-dimensional organism, eg, developmental cell migration, immune cell trafficking, and engraftment of bone marrow and other tissues, we have previously generated a transgenic donor mouse line in which a cytomegalovirus β-actin promoter drives the expression of 2 reporters, luciferase and green fluorescent protein (GFP). Furthermore, we have also developed a transgenic mouse line in which a CD5 promoter drives the expression of the luciferase gene.
Using these transgenic mice and in vivo BLI technology, we tested the hypothesis that in vivo BLI can be used for the in vivo assessment of allograft viability and in vivo visualization of donor-derived CD5+ cells during the course of acute rejection.
Previously, we created a transgenic mouse line (FVB-L2G85) by pronuclear injection of a gene construct expressing 2 reporters, firefly luciferase and GFP, under the control of the widely expressed β-actin promoter (Dr Cao, unpublished data). The overexpression of luciferase and GFP in the heart was confirmed by bioluminescent and fluorescent microscopy (Figure 1A and 1B). We have also developed a transgenic mouse line in which a CD5 promoter drives the expression of the luciferase gene.9 Female BALB/c and FVB mice (4 to 6 weeks old) were obtained from The Jackson Laboratory (Bar Harbor, Me) and used as recipients. All procedures were approved by the Animal Care and Use Committee of Stanford University.
Mouse Heterotopic Heart Transplantation
Hearts of transgenic FVB-L2G85 mice (H-2q) were heterotopically transplanted into the abdomen of wild-type BALB/c (H-2d) or syngeneic FVB mice as an acute rejection model. Heterotopic cardiac transplantation was performed according to the method of Corry et al10 with some modifications. Anesthesia was induced with 3% inhaled isoflurane (Halocarbon Laboratories). During surgery, the animals were maintained on 2.5% inhaled isoflurane. The total ischemic time of cardiac allograft was 40 minutes.
Graft Survival and Allograft Functional Analyses
Graft viability was assessed by direct abdominal palpation of the heterotopically transplanted heart, as previously described.11 Cardiac graft function was expressed as the beating score, assessed by the Stanford cardiac surgery laboratory graft scoring system (0: no contraction; 1: contraction barely palpable; 2: obvious decrease in contraction strength, but still contracting in a coordinated manner, rhythm disturbance; 3: strong, coordinated beat but noticeable decrease in strength or rate, distention/stiffness; or 4: strong contraction of both ventricles, regular rate, no enlargement or stiffness).
In Vivo Bioluminescent Imaging
Mice were anesthetized with 2% inhaled isoflurane, and luciferin was administered at a dose of 150 mg/kg intraperitoneally. At the time of imaging, animals were placed in a light-tight chamber. With an in vivo imaging system using a cooled charge couple device camera (IVIS 100, Xenogen Corp), photons transmitted through the tissue emitted from intracellular luciferase were collected for 10 seconds to 10 minutes (as indicated in the figure legends), depending on the intensity of the bioluminescence emissions.
Because bioluminescence is dependent on tissue penetration, the intensity from the cardiac grafts may vary depending on the location of the grafts in the abdomen (ie, cardiac grafts can change position and bowel may cover the grafts); therefore, we chose the 3-dimensional imaging system, which allowed us to obtain signal intensity from 8 different directions (Figure 1C), thus allowing us to calculate the summation of light intensities obtained from 8 different direction scans, which then was used to represent cardiac allograft viability.
Using applications in LivingImage software (Xenogen Corp), an overlay on Igor image analysis software (Wavemetrics), gray scale reference images were collected under low light, and the intensity of the bioluminescent signals from the animal was measured in complete darkness (blue indicates least intense and red most). The 2 images were then superimposed and annotated using Canvas (Deneba). To quantify allograft viability, the light emitting from the allograft was outlined as the region of interest, and region of interest photon intensity was measured using LivingImage software. In vivo photon intensity measures were correlated to ex vivo tissue sections by dissecting the tissues, incubating fresh tissues in D-luciferin, and imaging these tissues without the overlying tissues.
Tissue Collection and Immunofluorescent Histology
To evaluate inflammatory cell infiltration, cardiomyocyte destruction, and proliferation of donor derived fibroblasts, cardiac grafts were perfused with saline and rapidly excised at different days after transplantation. They were fixed in 2% paraformaldehyde for 2 hours and cryoprotected in 30% sucrose overnight. Tissue was then frozen in optimal cutting temperature compound (OCT Compound, Sakura Finetek USA, Inc) and sectioned at 5 μm on a cryostat. Sections were blocked and incubated with either rat anti-CD45, rat anti-CD5, or mouse anti-vimentin (all BD Pharmingen) primary antibodies for 1 hour at room temperature. After washing in phosphate buffered saline, sections were incubated with either goat anti-rat Alexa Fluor 594 (red), goat anti-rat Alexa Fluor 488 (green), or goat anti-mouse Texas Red (all Molecular probes) secondary antibodies for 30 minutes at room temperature. Sections were then washed in phosphate buffered salien, counterstained with 4,6-diamidino-2-phenylindole (DAPI; Molecular Probes) and examined with a Leica DMRB fluorescent microscope (Leica Microsystems).
Values are expressed as mean±SE. Differences in cardiac graft beating score and light intensity of cardiac grafts were analyzed by a 2-way repeated-measures ANOVA. Correlation between cardiac graft beating score and light intensity emitted from cardiac grafts were analyzed by regression analysis (StatView 5.0; SAS Institute). Significance was accepted at P<0.05.
In Vivo Visualization of Acute Cardiac Rejection and Quantitative Analyses of Cardiac Allograft Viability Using Bioluminescence Imaging
Bioluminescent and fluorescent microscopic analysis of the heart from luciferase-GFP transgenic mice confirmed expression of both luciferase and GFP (Figure 1A and 1B). There was no bioluminescent signal from the nontransgenic littermates (data not shown). Using the hearts of luciferase-GFP transgenic mice as donor organs, we investigated the feasibility of BLI for quantitative assessment of cardiac allograft viability in the course of acute rejection. As the luciferase-GFP transgenic mice have a FVB background, we used BALB/c mice as recipients to induce acute rejection. In the course of acute rejection, we measured cardiac graft viability in vivo using 3-dimensional BLI as described earlier.
All FVB cardiac allografts transplanted into BALB/c recipients were acutely rejected by 12 days after transplantation. In contrast, all FVB cardiac isografts transplanted into FVB recipients survived until 12 days after transplantation. Cardiac allograft beating score assessed by daily abdominal palpation showed significant difference between isografts and allografts (P<0.001, Figure 2A). The intensity of light emitting from cardiac allografts declined 4 days after transplantation and correlated with graft beating scores (R2=0.91, P=0.02; Figure 2B and 2C). To confirm correlation between cardiac graft viability and light intensity, cardiac allografts and isografts were procured at days 2, 4, 6, 8, 10, and 12 after transplantation and stained with an antibody against CD45, a marker expressed on all inflammatory cells. Immunohistochemistry showed an increase of inflammatory cell infiltration and destruction of GFP+ cardiomyocytes in the cardiac allografts in the course of acute rejection (Figure 2D). In contrast, only mild inflammatory cell infiltration and preserved GFP+ cardiomyocyte structure were observed in cardiac isografts by 12 days after transplantation (Figure 2D). Taken together, changes of photon signal intensity of the cardiac graft correlate with cardiac graft viability and histological findings in the course of acute cardiac rejection. Therefore, our data suggest that in vivo BLI may be a useful tool for the quantitative assessment of cardiac graft viability in vivo.
In Vivo Visualization of Donor-Derived Passenger CD5+ Cell Response in the Cardiac Allograft Recipients in the Course of Acute Rejection
Solid organ grafts contain bone marrow-derived hematopoietic cells, passenger leukocytes, of donor origin. These donor-derived hematopoietic cells are transferred to the recipient at the time of transplantation.4 We next investigated the possibility of BLI to visualize passenger leukocytes and to track their destination in vivo using CD5 promoter luciferase transgenic mice as a donor of syngeneic and allogeneic cardiac transplantation. As shown in Figure 3, the presence of CD5+ donor-derived passenger leukocytes in the donor heart was confirmed by immunohistochemistry. In vivo BLI of donor heart, however, did not show light emitting from CD5+ donor-derived passenger leukocytes, probably because of the small number of CD5+ donor-derived passenger leukocytes in donor hearts. Interestingly, in vivo BLI successfully visualized the migration and proliferation of CD5+ donor-derived passenger leukocytes shortly after transplantation (30 minutes) in both syngeneic and allogeneic recipients, indicating that these cells proliferated immediately (Figure 4). In the cardiac allograft recipients, bioluminescence signal from CD5+ donor-derived passenger leukocytes increased with time. The signal peaked at 6 and 8 hours after transplantation, rapidly decreased at 12 hours, and had diminished at 1 day after transplantation (Figure 4). In contrast, in vivo BLI of the syngeneic recipients showed that signal intensity from CD5+ donor-derived passenger leukocytes increased at 30 minutes after transplantation, peaked at 1 and 2 days, and was measurable until 10 days after transplantation (Figure 4). CD5+ donor-derived passenger leukocytes migrated from the cardiac allograft in the abdomen through the recipient body and were observed at anatomic sites corresponding to the location of regional abdominal lymph nodes (1 hour), liver (2 hours), thoracic lymph nodes (4 hours), and inguinal lymph nodes (6 hours). In comparison, in the syngeneic recipients, CD5+ donor-derived passenger leukocytes were found in the neck (30 minutes), chest, and liver (2 hours). Confirmation of tissue origin was obtained by ex vivo imaging of dissected tissue incubated with D-luciferin (data not shown). Because all FVB cardiac allografts transplanted into BALB/c recipients were acutely rejected at 12 days after transplantation, our data suggest that the contribution of the donor-derived CD5+ passenger leukocytes to acute rejection is limited to the early phase of acute rejection.
This study was designed to show the feasibility of using in vivo BLI to visualize the changes of cardiac allograft viability and donor-derived passenger CD5+ cells in response to cardiac allografting during the course of acute rejection. Using in vivo BLI, we have successfully shown that light intensity emitted from cardiac allografts decreases after 4 days in the course of acute rejection. In addition, the light intensity emitted from donor-derived passenger CD5+ cells diminished within 1 day after allo-transplantation. In vivo BLI of different promoter luciferase transgenic donor heart recipients allowed us to quantify the kinetics of the viability of the cardiac allografts and the location of the donor-derived passenger CD5+ cells longitudinally.
Graft beating score is nonquantitative and a subjective classification system. It is useful to determine cardiac graft survival as used elsewhere; however, it is not suitable to assess cardiac graft viability because this method assess only whether the cardiac allograft is beating or not. In vivo BLI of the cardiac graft provides detailed information about cardiac allograft viability in the course of acute rejection by calculating light emitting from each cardiac cell within the cardiac graft with 3-dimensional imaging. Therefore, this modality may be useful tool to assess cardiac graft viability in the clinical heart transplantation.
We observed a temporal increase of light intensity 3 to 5 days after transplantation. We attempted to prove donor-derived cell proliferation, including passenger leukocytes and fibroblasts, by staining with anti-CD45 antibody and anti-vimentin antibody. We did not see any GFP and CD45 double-positive cells at days 3 to 5, however, and we did not see an increase of GFP- or vimentin-positive cells from day 0 to day 6. This temporal increase of light intensity may be due to an increase of luciferase gene expression by ischemia-reperfused injured myocardium. We have observed a similar phenomenon in other animal models of BLI exposed to stress (unpublished data).
By using this in vivo imaging approach, we found that donor-derived passenger CD5+ cells in cardiac allograft proliferated immediately after transplantation and diminished by 24 hours after transplantation in the course of acute rejection. In addition, these cells migrated to liver and thoracic lymph nodes, but not the spleen. In contrast, donor-derived passenger CD5+ cells in a syngeneic graft stayed approximately 10 days after transplantation and diminished possibly because of their life span (around 7 days). This study is the first study to visualize donor-derived passenger leukocytes in the recipient. Traditionally, the evaluation of donor-derived passenger leukocyte migration from the transplanted organ to the other organs and destruction of cardiomyocyte structure by acute rejection have been performed post-mortem on histological specimens or other biological experiments. Therefore, the question of the in vivo sequence of these events has been controversial. BLI is a useful modality for tracking immune cells in vivo and reducing the number of animals per experiment, as changes in a given population can be studied over time.
We observed that CD5+ donor-derived passenger leukocytes also proliferated immediately after transplantation in syngeneic recipients. This data suggests that alloantigen-dependent response is not necessary to drive CD5+ donor-derived passenger leukocytes’ proliferation. One can surmise that alloantigen-independent response, such as ischemia-reperfusion of CD5+ donor-derived passenger leukocytes, might cause their proliferation in the recipient after transplantation.
Donor-derived passenger leukocytes that are transplanted with the graft have the capacity to present donor alloantigens as intact molecules to the responding T cells by means of the so-called direct pathway of allorecognition.12 CD5 is expressed at relatively high levels on all T lineage cells, at low levels on B-1a cells, and below detectable levels on B-2 cells.9 Therefore, the light signal we observed in this CD5 promoter luciferase mice donor study is emitted from donor-derived passenger T cells. In the present study, we demonstrated that donor-derived passenger T cells diminished within 24 hours after transplantation in cardiac allograft recipients. Further studies using luciferase genes driven by other cell surface markers such as CD4 (T helper cell), CD8 (cytotoxic T cell), CD11c (dendritic cell), CD11b (macrophage), and B220 (B cell) for the in vivo visualization of immune response will increase our basic understanding of not only the passenger leukocyte function and destination in allograft recipient, but also the mechanisms of immune response in solid organ and cellular transplantation.
We have assessed CD5+ cell location in the cardiac transplant recipient mice in the course of acute rejection using BLI. We did not assess the functional and immunological changes of these cells. Nevertheless, this is the first study to visualize CD5+ cells in vivo in the course of acute rejection.
The main limitation of this study is the lack of confirmation of the sensitivity and specificity of in vivo BLI. This study is intended to be a preliminary study to show proof of concept of a new technique and to show the ability of in vivo BLI to visualize the change of transplanted hearts and behavior of transplanted immune cells. A large-volume study to confirm the sensitivity and specificity of this technique are now being done in our laboratory.
The results of this study provide the foundation for refinement of BLI in a non-human primate study. Clinically, this novel method of noninvasive diagnosis of cardiac allograft rejection could potentially eliminate the need for routine surveillance endomyocardial biopsy. The transplanted heart could be visualized by BLI after transplantation by luciferase gene transfection (viral or other methods) to the explanted heart from donor at the time of preservation. There is possibly the ability to assess viability of the transplanted heart noninvasively using this technology. BLI is a novel technique, and problems with scattering of the signal and transmission through thicker layers are real and known in the research society. There is much to be improved on concerning these issues. At this moment, the importance of BLI is the ability of the technique to provide information about the physiology of living cells in vivo in small animal models. This study is intended to be a preliminary study to show proof of concept of a new technique. Application of in vivo BLI to the study of cardiovascular disease as well as transplant immunobiology will greatly accelerate and refine preclinical analyses and lead to the development of tools with clinical utility. A number of advances have already been described and suggest that there is considerable growth yet to be realized in the nascent field of molecular imaging.
This work was supported by National Institutes of Health grant HL65669 (Dr Robbins).
Vassalli G, Gallino A, Weis M, von Scheidt W, Kappenberger L, von Segesser LK, Goy JJ. Alloimmunity and nonimmunologic risk factors in cardiac allograft vasculopathy. Eur Heart J. 2003; 24: 1180–1188.
Miura M, Morita K, Kobayashi H, Hamilton TA, Burdick MD, Strieter RM, Fairchild RL. Monokine induced by IFN-gamma is a dominant factor directing T cells into murine cardiac allografts during acute rejection. J Immunol. 2001; 167: 3494–3504.
Sweeney TJ, Mailander V, Tucker AA, Olomu AB, Zhang W, Cao Y, Negrin RS, Contag CH. Visualizing the kinetics of tumor-cell clearance in living animals. Proc Natl Acad Sci U S A. 1999; 96: 12044–12049.
Edinger M, Cao YA, Verneris MR, Bachmann MH, Contag CH, Negrin RS. Revealing lymphoma growth and the efficacy of immune cell therapies using in vivo bioluminescence imaging. Blood. 2003; 101: 640–648.
Yang Y, Contag CH, Felsher D, Shachaf CM, Cao Y, Herzenberg LA, Tung JW. The E47 transcription factor negatively regulates CD5 expression during thymocyte development. Proc Natl Acad Sci U S A. 2004; 101: 3898–3902.
Tanaka M, Terry RD, Mokhtari GK, Inagaki K, Koyanagi T, Kofidis T, Mochly-Rosen D, Robbins RC. Suppression of graft coronary artery disease by a brief treatment with a selective epsilonPKC activator and a deltaPKC inhibitor in murine cardiac allografts. Circulation. 2004; 110 (suppl II): II-194–II-199.