Dendritic cell elimination as an assay of cytotoxic T lymphocyte activity in vivo

https://doi.org/10.1016/S0022-1759(00)00300-8Get rights and content

Abstract

We show in this paper that the survival of antigen-loaded dendritic cells in vivo may be used as a sensitive readout of CTL activity. We have previously shown that dendritic cells labeled with the fluorescent dye CFSE and injected sub-cutaneously into mice migrate spontaneously to the draining lymph node where they persist for several days. In the presence of effector CTL responses, dendritic cells loaded with specific antigen rapidly disappear from the draining lymph node. In this paper we extend the above observations and set up a simple and sensitive method to reveal CTL activity in individual mice in vivo. Dendritic cells were labeled with two different fluorochromes, loaded with antigen or left untreated, and mixed together before injection into mice. We show that only the dendritic cells loaded with specific antigen were cleared from the draining lymph node, while dendritic cells not loaded with antigen remained unaffected. Cytotoxic responses generated by immunization with peptide-loaded dendritic cells, or by infection with influenza virus, could be revealed using this method. Comparison of the differential survival of dendritic cells populations mixed together also allowed us to accurately evaluate the disappearance of dendritic cells, irrespective of variability in the injection site and other parameters. Given the ability of dendritic cells to efficiently take up and present complex antigens, nucleic acids and apoptotic bodies, this method may also allow the evaluation of cytotoxic activity against antigens that are not characterized in terms of peptide epitopes.

Introduction

The study of the size, functional activity, and kinetics of CTL responses is essential to the understanding of the effects of immunotherapy. The demonstration of CTL responses is typically achieved in vitro, by measuring the cytotoxic activity of T cells freshly isolated ex vivo against antigen-loaded targets labeled with radioactive isotopes (Brunner et al., 1968, Matzinger, 1991). This method accurately reflects the presence of effector CTL in vivo, but its sensitivity is often low. In addition, it is limited by the number of effector cells that can be recovered ex vivo, thus is often only useful at the peak of strong virus-induced CTL responses. In vitro re-stimulation of cells ex vivo in the presence of antigen, either in bulk cultures or in limiting dilution conditions, greatly increases the sensitivity of the assay. However, this method does not distinguish between effector T cells that are already cytotoxic in vivo, memory cells that re-acquire cytotoxic activity during culture in vitro, or even naı̈ve T cells that may first become activated in response to high concentrations of antigen present during the in vitro culture. In addition, this method may be confounded by non-specific target cell cytotoxicity. Even in these improved conditions, the size of the cytotoxic response is possibly widely underestimated (Flynn et al., 1998).

Additional assays have recently become available to estimate more directly CD8+ T cell responses in vivo. For example, the expansion and/or activation of T cell receptor transgenic T cells can be evaluated by FACS staining (Butz and Bevan, 1998, Hermans et al., 1999), and antigen-specific T cells can be directly labeled by staining with fluorescent tetrameric MHC–peptide complexes (Altman et al., 1993, Murali-Krishna et al., 1998). In combination with intracellular staining for cytokines or for cytotoxic mediators such as perforin, these methods can also provide some information on the functional activity of the detected CD8+ T cells. Again, potential limitations are the availability of appropriate reagents such as TCR transgenics, and suitable tetramers and peptides to detect the response of interest.

With these considerations in mind, we wished to explore an alternative means by which CTL generation may be detected. In this report we demonstrate that DC labeled with fluorescent dye can be tracked to the DLN, and that DC loaded with MHC class I-binding peptide do indeed serve as targets for CTL. Antigen-specific killing of antigen-loaded DC could be detected in individual mice, whether CTL were generated by immunization with antigen-pulsed DC, or in response to a viral infection. We propose that this technique may be used to directly measure T cell-mediated cytotoxic activity in vivo.

Section snippets

Mice

C57BL/6 mice were from breeding pairs originally obtained from The Jackson Laboratories, Bar Harbor, ME. Strain 318 mice (Pircher et al., 1989), transgenic for a T cell receptor (TCR) specific for H-2 Db+fragment 33–41 of the LCMV glycoprotein (LCMV33–41), were kindly provided by Dr. H. Pircher, Institute of Medical Microbiology and Hygiene, University of Freiburg, Germany. All mice were maintained at the Biomedical Research Unit of the Wellington School of Medicine by brother×sister mating. In

The elimination of antigen-loaded DC from the draining lymph nodes serves as an indicator of CTL activity in vivo

We have shown in previous experiments that DC loaded with an MHC class I-binding peptide antigen are capable of eliciting a CD8+ T cell response upon in vivo injection (Hermans et al., 1997, Hermans et al., 1999). In the course of those experiments we also observed that DC disappear from the DLN of immunized mice, with kinetics that parallel the known kinetics of CD8+ T cell activation to effector function (Hermans et al., 2000). In this manuscript we explore the potential of DC elimination as

Discussion

The demonstration of effector CTL activity in vivo is often difficult to achieve. Only in some cases, such as in some viral infections, are the numbers of effector CTL generated sufficient to allow direct demonstration of CTL activity ex vivo (Gardner et al., 1974, Koszinowski and Ertl, 1975, Zinkernagel and Doherty, 1974). More often, responses can only be demonstrated after in vitro re-stimulation to allow the differentiation of CTL precursors into effector cells and their enrichment through

Acknowledgements

We thank the personnel of the Wellington School of Medicine Biomedical Research Unit for animal husbandry. This work was supported by grants from the NZ Cancer Institute, the Marsden Fund, the Cancer Society of New Zealand and an equipment grant from the New Zealand Lotteries Board. F.R is a recipient of a Wellington Medical Research Foundation Malaghan Senior Fellowship. D.R. was supported by a generous donation from Sir Roy McKenzie.

References (17)

There are more references available in the full text version of this article.

Cited by (0)

View full text