Tracking of leukocyte recruitment into tissues of mice by in situ labeling of blood cells with the fluorescent dye CFDA SE

Abstract

Models of inflammatory and immune diseases are extensively studied in mice with engineered genetic mutations, and tracking the recruitment of blood leukocytes into tissues is an important component of these studies. A direct in situ method for labeling the total pool of blood cells in mice by a single intravenous injection of the fluorescent dye CFDA SE (CFSE) is described. The fluorescence intensity of labeled cells initially declines, but remains stable after 4 h, enabling detection weeks after labeling. Labeled leukocytes can be tracked as they accumulate in lymphoid tissues and sites of inflammation and then be immunophenotyped for analysis by flow cytometry. This method is rapid, reproducible and simple to perform.

Introduction

The emigration of leukocytes from blood into tissues is critical for protection of the host during pathological inflammatory processes, as well as physiological recirculation of lymphocytes that participate in immune surveillance. Historically, leukocyte recruitment and lymphocyte recirculation have been investigated by isolating and labeling of leukocytes in vitro with subsequent reinjection of labeled cells into animals and tracking their distribution or accumulation into tissues and compartments (Parish, 1999). In vitro labeling and reinjection of cells is time consuming, laborious and results in only a small cohort of emigrating cells being labeled. Moreover the maintenance of physiological integrity has been a concern due to in vitro handling of leukocytes, particularly myeloid cells that are readily activated by pyrogens and mechanical forces.

In early studies, leukocytes were labeled in vitro with radioisotopes Gowans and Knight, 1964, Cybulsky and Movat, 1982, Issekutz and Movat, 1982, Chin and Hay, 1984. Although convenient, radioisotopes are a potential biohazard and have a number of experimental drawbacks. The precise localization of cells within tissues can only be determined by time consuming autoradiography of histological sections and that cell division cannot be tracked. Gamma emitting and high-energy β isotopes may affect cell viability, while isotopes with lower emission energies are difficult to follow in vivo and are subject to quenching. Reutilization of radioisotopes is a documented phenomenon, thus bystander cells may inadvertently be labeled (Ford, 1978). Finally, due to decay of high-energy radioisotopes, only short-term studies are possible (Parish, 1999). Nevertheless, these important studies established the concept of the continuous migration of lymphocytes between blood and lymph and identified some of the cytokines and inflammatory mediators that govern leukocyte recruitment and migration in vivo.

Over the last 20 years, fluorescent labels have become widely available and are now the preferred method for labeling of cells, avoiding some of the pitfalls of radioactive isotopes. Fluorescent labels include DNA binding dyes, membrane inserting dyes and dyes localizing to the cytoplasm (Parish, 1999). Fluorescent labeling in combination with flow cytometry has permitted the comparative analysis of many samples, used large numbers of individual cells for analysis and, using multiple labels, permitted simultaneous subset analyses (Young, 1999). DNA binding dyes such as bisbenzimide (Hoechst 33258) or Thiazole Orange can be used for labeling of leukocytes ex vivo or in situ. Their limitations include transient labeling of cells by some dyes or alternatively, interference with cell proliferation, presumably by affecting DNA replication (Samlowski et al., 1991). Membrane dyes, such as PKH2 or PKH26, on the other hand, can persist for weeks or in some cases even months. However, ex vivo processing of cells is still required and a less uniform staining is observed relative to some other dyes (Parish, 1999). Cytoplasmic dyes (with some unavoidable membrane binding) such as FITC or TRITC have relatively low fluorescence are toxic at relatively low doses and require ex vivo processing. Moreover, impaired ability of over-labeled lymphocytes to enter lymphoid organs has been observed (Butcher et al., 1980). Nevertheless, long-term studies have been done using ex vivo labeling and have contributed much to our understanding of lymphocyte recirculation and cellular turnover rates (Young and Hay, 1999).

An approach to label blood leukocytes in situ with a fluorescent tracer would have obvious advantages including a simple and rapid protocol, the ability to label a large pool of leukocytes, and avoidance of ex vivo handling. Ristevski et al. (2003) recently developed a method for direct in situ leukocyte labeling in sheep. This has made it possible to achieve a labeling index (percent of labeled cells) that is 20-fold higher than with ex vivo methods, making it feasible to track minor leukocyte subpopulations in vivo (Ristevski et al., 2003). Given the advantages of in situ labeling, we endeavored to use this method in mice, but attempts at using the same protocol were unsuccessful. Mice provide extremely powerful experimental models for elucidating molecular mechanisms of pathophysiological processes, although their relatively small size and small blood volume pose technical challenges. There is a wide array of available knockouts, conventional transgenics and antibody reagents. In this manuscript, we present a method for direct in situ labeling of the total blood leukocyte pool in mice, and demonstrate the utility of this approach for tracking both short-term emigration and long-term trafficking of individual blood leukocyte subtypes.