Background
T cell migration to sites of inflammation and infection is essential for adaptive immunity and host protection. Current methods of analyzing T cell migration and homing primarily rely on obtaining mouse end-organ tissues and subsequent detection of cells within these organs, either through in situ analysis by immunohistologic methods or by isolation and analysis of recovered T cells from disrupted tissues. While each of these approached offers unique advantages (and disadvantages), neither permits "real-time" analysis of T cell dynamics following antigenic exposure and is limited to specific time points and tissues chosen for study. Bioluminescent imaging represents a powerful alternative for the monitoring immune cell homing and migration in vivo that does not require animal sacrifice for analysis.
Luciferase proteins produce light in the visible spectrum (approximately 560 nm for firefly luciferase) following interaction with luciferin substrate molecules. This reaction only requires ATP and oxygen and thus can occur in any actively metabolic cell. Luciferin, a small water-soluble molecule, readily crosses cell membranes and can penetrate into virtually all tissues. Light produced by the luciferase-luciferin reaction is detectable by low-light detection devices, such as charge-coupled device (CCD)
4cameras. Bioluminescence imaging is a good model for
in vivo imaging based on the low background signal, since it only detects chemical reactions between the enzyme and substrate. Bioluminescence also requires relatively short imaging times (seconds to minutes), is easy to use, and the instrumentation is relatively inexpensive [
1,
2].
Luciferase imaging techniques have previously been used to track localization of bacterial and viral pathogens, monitor therapeutic responses in tumor xenografts, and allograft cell survival [
3‐
6]. Recent studies have utilized bioluminescence to track T cell homing patterns following adoptive transfer of luciferase-expressing allogeneic T cells into MHC-mismatched recipient mice [
7,
8]. We have also reported use of luciferase-expressing CD8 T cells for analysis of population dynamics in response to viral-based antigenic-challenge [
9]. These studies reveal that T cell proliferation, in the context of graft versus host disease or viral infection, can be visualized in 'real-time' within recipient mice. The luciferase-expressing T cells can also be identified within specific tissue sites in order to determine homing and migration kinetics over time [
7,
8].
In the current study, we describe the generation of a transgenic mouse (T-Lux) model in which the luciferase gene is specifically expressed by T cells, thereby permitting analysis of T cell population dynamics in living mice in real-time. By crossing the T-lux transgenic mice with OVA-specific CD4 TCR transgenic mice (OT-II), we have generated a reporter model with which to track antigen-specific CD4 T cells in vivo, in real-time. Our studies reveal the population dynamics of the CD4 clonal response within draining lymphoid tissues and antigen injection sites in individual mice and demonstrate the utility of this model for non-invasive analyses of T cell clonal responses.
Methods
Mice
C57BL/6 and C57BL/6.Ly5.2 (CD45.1+) mice were obtained from Jackson Laboratory (Bar Harbor, ME). Thy1.1-expressing C57BL/6 mice were obtained from Charles River Breeding Laboratories. Mice used for imaging were placed on a low light diet (Harlan Tekland) to reduce non-specific luminescence. All mice were housed or bred in our specific pathogen-free facility, and were initially used at 6–10 weeks of age. All mice were housed and treated according to National Institutes of Health guidelines under the auspices of the UAB Institutional Animal Care and Use Committee (IACUC) of the University of Alabama at Birmingham.
Generation of T-lux transgenic mouse
The luciferase coding sequence was excised from the pGL3 plasmid (Promega) by Xba I/Nco I digestion, blunted with Klenow and ligated into a Sma I-linearized hCD2 minigene cassette plasmid (the kind gift of Dr. Dmitri Kioussis; [
10]). The T-lux transgene was restriction mapped for correct orientation and sequenced across the integration joints to confirm the correct reading frame and terminus of the luciferase gene (using primers OCW938 and OCW939). The reporter gene construct was introduced into single-cell embryos of donor C57BL/6 mice by pronuclear injection in the UAB Transgenic Mouse Facility. Potential founder lines were screened by luminescence analysis of peripheral blood. Red blood cells were lysed from 30 μL of blood using ACK buffer for 2 minutes. Cells were then plated into 96-well plates and luciferin (100 μg/ml) added to each well 10 minutes prior to imaging. Bioluminescence imaging of the plates was performed using the IVIS
® Imaging System with stage at 15 cm height for 600 seconds. One founder line (T-lux 9) with readily detectable luminescence that transmitted the transgene in the germline was selected for subsequent studies.
Antibodies
The following antibodies were used for fluorescence labelling of cells for flow cytometric analyses: fluorescein isothiocyanate (FITC)-labelled anti-mouse CD8 (clone 53-6.7); phycoerythrin (PE)-labelled anti-mouse CD4 (RM4-5); PE-labelled anti-mouse CD45.2 (104); biotin-labelled anti-mouse CD8 (H35-17.2) and CD45.1 (A20). All were purchased from eBiosciences (San Diego, CA). PerCP-labelled anti-mouse CD3 (145-2C11) was purchased from BD Pharmingen (San Jose, CA).
Flow cytometric analysis
Single cell suspensions from specified target organs were analyzed by FACS at the indicated time points. Cells were stained for flow cytometry with FITC-, PE-, PerCP-, or biotin-conjugated mAbs indicated in the appropriate figure legends. Cells labelled with biotinylated primary Ab were detected with allophycocyanin-conjugated streptavidin. For analytical flow cytometry, at least 100,000 events with forward and side scatter properties of lymphocytes were collected on a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) and analyzed using CellQuest software (BD Biosciences, San Jose, CA). Some analyses were performed using FlowJo software (Treestar, Inc, Ashland, OR).
Isolation and Transfer of T-lux Cells
Splenocytes were harvested from T-lux mice and CD4+ T lymphocytes isolated using Dynal Bead Separation (Dynal Biotech LLC), according to manufacturer's instructions. Briefly, T-lux splenocytes were incubated with CD4+ specific magnetic beads for 20 minutes at 4°C. The cell-attached beads were then incubated with Detach-a-bead at room temperature for 1 hour. Thoroughly washed cells were counted, resuspended in RPMI medium without serum, and transferred to recipient mice via tail vein injection.
In vivo Bioluminescence Imaging of Mice
Mice were anesthetized with isofluorane gas and placed in a light-tight chamber. A photographic (gray-scale) reference image was obtained at 10 minutes after D-luciferin injection (2.5 mg intraperitoneal); bioluminescent images were collected immediately thereafter. Images were obtained with a CCD camera cooled to -120°C, using the IVIS® 100 Imaging System (Xenogen Corp., Alameda, CA) with the field of view set at 10 cm height. The photographic images used a 0.2 second exposure, 8f/stop, 2 binning (resolution), and an open filter. The bioluminescent images used exposures ranging from 120 to 600 seconds, 1f/stop, 8 binning and open filter. The bioluminescent and gray-scale images were overlaid using Living Image software (Xenogen Corp.). Igor image analyses software (Wavemetrics, Lake Oswego, OR) was also used to obtain a pseudocolor image representing bioluminescence intensity (blue, least intense, and red, most intense). Regions of interest were drawn around organs and the total counts (photons) were summed in the entire organ areas. The total counts in each region of interest were normalized to total acquisition time to obtain counts/sec.
Discussion
T cells undergo a complex pattern of clonal proliferation and expansion following primary and secondary antigenic challenges [
32‐
35]. Using luciferase-expressing OT-II T cells, we were able to monitor antigen-specific CD4
+ T cell dynamics serially in the same animal. In accord with previous reports, our results demonstrated initial expansion within the draining lymph node followed rapidly by migration and accumulation within a footpad injection site [
20,
30]. Notably, and in agreement with a previous study [
30], we find that overwhelmingly the T cells that leave the lymph node draining the site of antigen injection migrate to the antigen injection site, although some enhanced accumulation of cells in other secondary lymphoid tissues is evident. Despite early increases in T cells numbers in the draining lymph nodes, T cell numbers within both the draining lymph node and the injection site demonstrated a peak 7 days after injection, followed by a rapid decline in signal intensity both in the draining lymph nodes and injection site that was only marginally delayed kinetically in the latter. The concordance of the peak of T cell expansion in the draining lymph node and injection site in our study is somewhat different from previous studies, which demonstrated a more prolonged time course [
20] and kinetic delay between peak T cell numbers in draining lymph node and injection site. This likely reflects the more rapid clearance of the antigen-adjuvant depot in our study, as the ME adjuvant is far less viscous and is more rapidly cleared from tissue sites than the oil-based emulsions used in previous studies (unpublished observations). This highlights the importance of sustained antigen in the injection site to prolong recruitment and/or survival of immigrating T cells attracted to the site of antigen delivery [
30].
In the current study, there was no detectable redistribution of CD4 T cell signal to non-lymphoid tissues following clonal contraction in the draining LNs and injection site, which is in contrast to previous studies wherein antigen and adjuvant were delivered systemically, or in studies of the CD8 response to viral infection. Again, this likely reflects differences in the distribution and kinetics of antigen clearance in the different studies. This may also reflect characteristics of the OT-II TCR transgenic model, clonotypic T cells from which have a relatively low avidity for the MHC-OVAp complex that may allow endogenous clones reactive to OVA peptide to out-compete the OT-II response over time. Future studies, following crosses of the T-lux reporter transgene with other TCR transgenic specificities should help address this. In any case, given the relative technical ease with which T cell clonal dynamics can be monitored using the T-lux model, efforts to implement these types of single animal longitudinal studies represents a significant advance for conducting studies of adjuvant type and delivery for future studies, with attendant advantages for optimizing future vaccine therapies for infectious agents and cancer.
Homeostatic proliferation of mature T cells occurs within the lymphopenic host and has been shown to be associated with a concomitant shift toward memory phenotypes (reviewed in [
36]). T cell expansion in this setting has been shown to be due to IL-7 [
23,
24] and IL-15/IL-2-dependent mechanisms [
37]. IL-15, in particular, has been demonstrated to induce a memory phenotype in CD8
+ T cells that is associated with an alteration of cell homing receptors [
38,
39]. It is not surprising; therefore, that we observed T cell expansion and localization of T-lux cells within non-lymphoid tissues following transfer into the RAG-deficient mice. Localization within non-lymphoid tissues as well as the activation profile of these cells has important implications in immune reconstitution following infection and bone marrow transplantation, and implementation of the T-lux model for efficient characterization of this process is demonstrated herein.
Bioluminescent imaging technology provides a mechanism to observe T cell homing and specific CD4
+ T cell immune responses
in vivo, in real-time in individual animals. This technique eliminates the need for large groups of mice terminated at varying time points, reducing animal usage and cost. In addition, this provides a method to efficiently observe and quantitate T cell responses in all tissue sites simultaneously, providing a window on T cell response dynamics not easily achieved using approaches that require tissue disruption for localization and quantitation of T cells. Also, given the lack of alloreactivity of the luciferase reporter molecule in the C57BL/6 strain, the T-lux mouse provides an excellent method for long-term immune studies following adoptive transfer. We have monitored T-lux T cell survival up to 75 days following adoptive transfer and found no evidence of immune clearance in re-transfer studies (Figure
6C, and data not shown). Previous studies in rats have demonstrated the absence of immunogenicity of firefly luciferase using transgenic tissue allografts [
40]. In contrast, studies using green fluorescence protein (GFP), another bioluminescent tracking molecule, have demonstrated an immune response to the molecule and decreased lifespan for GFP-expressing cells [
41]. Thus, the T-lux mouse is ideal for model systems investigating long-term immune recognition, such as mouse bone marrow transplant models. Other groups have utilized luciferase expression for analysis of lymphocyte homing following adoptive transfer [
1‐
4]. These authors have utilized bioluminescence to perform elegant studies on the role of regulatory T cells in graft versus host disease [
7,
8]. We feel that our T-Lux model improves on previous luciferase models through the generation of T-cell-specific bioluminescence. Limiting luciferase expressing to the T cell compartment enables the specific study of T cell dynamics within the intact mouse and eliminates potential confounding factors, such as an alteration of the normal TCR clonal precursor frequency that can affect T cell dynamics in adoptive transfer studies [
26].
A current limitation of the T-lux model is the attenuation of bioluminescent signal from T cells in deeper tissues. This is due to light absorption by superficial tissues situated between the T cell source and the detector when imaging cells within deeper organs. Decreased or attenuated luciferase signal is particularly apparent in the case of interference within blood rich organs, such as spleen, as hemoglobin absorbs light in the wavelength emitted by the luciferase enzyme [
42]. However, this limitation only affects absolute quantitation of T cells within deep tissues, while relative quantitation and dynamic changes over time are unaffected, as demonstrated herein. Further, the orientation of the animal can offset much of this limitation, depending on the tissue of interest (e.g., spleen). Finally, recent advances in the development of tomographic luminescent imaging instruments that can generate 3D images promise to enhance both sensitivity and spatial resolution of the T cell signal using this model, and should permit finer mapping of the real-time dynamics of T cell distribution to all tissue sites in future studies.
Acknowledgements
This work was supported by grants from the National Institutes of Health (AI035783 to C.T.W., 5P30CA013148 to K.R.Z.) and Children's Center for Research and Innovation of the Alabama Children's Hospital Foundation (to J.H.C.).
The authors thank I.N. Crispe and members of the Weaver lab for helpful discussions, and acknowledge James Oliver, Michael Blake, Karen Janowski and Henrietta Turner for technical assistance. We also acknowledge the UAB Transgenic Facility for embryo injections, the UAB Digestive Diseases Research Developmental Center (DDRDC) for generation and phenotyping of transgenic mice and the UAB Epitope Recognition and Immunoreagent Core Facility for antibody preparations.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
The authors contributed to the work as following: JC analyzed data and wrote manuscript; KD performed research; TC contributed design and analytical expertise; KZ designed research and contributed to the preparation of the manuscript; and CW designed research, analyzed data and wrote manuscript. All authors read and approved the final version of the manuscript.