Lymphocyte kinetics in health and disease Becca Asquith 1* , Jose ´ A.M. Borghans 2, 3* , Vitaly V. Ganusov 3, 4, 6* and Derek C. Macallan 5* 1 Department of Immunology, Wright-Fleming Institute, Imperial College London, London W2 1PG, UK 2 Department of Immunology, University Medical Center Utrecht, 3584EA Utrecht, the Netherlands 3 Theoretical Biology, Padualaan 8, Utrecht University, 3584CH Utrecht, the Netherlands 4 Los Alamos National Laboratory, Los Alamos, NM 87545, USA 5 Centre for Infection, Division of Cellular & Molecular Medicine, St George’s, University of London, Cranmer Terrace, London SW17 0RE, UK 6 Institute of Biophysics and Krasnoyarsk Science Centre, Krasnoyarsk, Russia 660036 Quantitative understanding of immunology requires the development of experimental and mathematical tech- niques for estimation of rates of division and death of lymphocytes under different conditions. Here, we review the advantages and limitations of several labelling methods that are currently used to quantify turnover of lymphocytes in vivo. In addition to highlighting insights into lymphocyte kinetics which have recently been gained thanks to the development of novel techniques, we dis- cuss important directions for future experimental and theoretical work in the field of lymphocyte turnover. Introduction Our understanding of cellular immunology has advanced on many fronts over the last few decades. As descriptive and mechanistic models have been developed, the need for complementary quantitative approaches has increased. Nowhere is this more true than in understanding lympho- cyte turnover and the maintenance of lymphocyte homeo- stasis. The importance of understanding lymphocyte turnover lies not just in being able to make quantitative observations but also in being able to predict responses, determine constraints and develop new paradigms for vaccinology and therapeutics. It also gives insight into the impact of factors driving or modifying immune responses, and addresses key questions in understanding immunological processes such as:  Why do T-cell immunological memory and protective levels of antibody sometimes persist for decades, but in other cases decline rapidly? Can we influence their duration?  Does the survival of clones determine persistence of memory or does continued recruitment have a more important role? Do memory clones, which divide relatively frequently, become exhausted or are they protected?  How are cell division and cell death linked?  ow does ageing influence lymphocyte kinetics? Does exhaustion of lymphocyte precursors have a role in immunosenescence?  What determines whether cells live or die? What are the roles of stimulation (not enough or too much), failure to compete and ‘survival niches’? Lymphocyte turnover can be influenced by many immunological or cellular factors (Table 1) and a large body of experimental data describing lymphocyte kinetics in different settings has emerged over the last few years. Such data are, however, only one part of the descriptive process; the other component is quantitative interpret- ation. For this, mathematical models are required to enable expression of data in meaningful, comparable parameters. This review highlights some of the key topics that were presented at a recent meeting on ‘Lymphocyte Kinetics in Health and Disease’ held in London, UK y . Although other methods, including analyses of Ki67 or Annexin V expression, T-cell receptor excision circles (TREC) and measurement of cells with chromosome damage or during T-cell reconstitution have provided valu- able insights into T-cell dynamics, the primary focus of this article is on lymphocyte labelling studies. Our aim is to provide an overview of the ways in which labelling studies of lymphocyte kinetics are currently being used to increase our understanding of immunology, in addition to present- ing some of the important unanswered questions and future directions for both experimental and theoretical work. Basic immunology: memory and homeostasis Investigation of lymphocyte kinetics has been applied to several aspects of normal in vivo human lymphocyte beha- viour. The attribute of ‘memory’, for example, has been addressed using deuterium labelling in humans [1–5] (Box 1 and Figure 1), and confirmed previous observations [6] that memory T cells have higher turnover rates than naı ¨ve T cells. In mice using BrdU (bromodeoxyuridine) labelling, a similar turnover pattern is seen [7]. Memory T cells, however, are quite heterogeneous both phenotypically (i.e. surface markers expressed) and kinetically (i.e. rates of proliferation and death). For example, it has been shown Review Corresponding author: Ganusov, V.V. (vitaly.ganusov@gmail.com) * In alphabetical order: all authors contributed equally to this report.. y The meeting website, which includes abstracts and presentations can be found at: http://s106175301.websitehome.co.uk/idrn_new/events/previous/lymphocyte.php.. 182 1471-4906/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.it.2009.01.003 Available online 13 March 2009