Lymphocyte Function Group


Werner HELD
Full Professor
Ludwig Center for Cancer Research
of the University of Lausanne

Web site:
Phone # : +41 (0)21 692 5958
Fax # : +41 (0)21 692 5995



Werner Held studied Microbiology at the University of Bern and obtained a Ph.D. in Immunology 1990 with Ch. Muller and M. Hess in the Dept. of Pathology at the University of Bern. He did his postdoctoral training in the labs of H. Acha-Orbea and H.R. MacDonald at the Lausanne Branch of the Ludwig Institute (1990-93) and in the lab of D.H. Raulet at the University of California at Berkeley (1993-96). After receiving a career development award (START fellowship) from the Swiss National Science Foundation he became an Assistant Member at the Lausanne Branch of the Ludwig Institute in 1996. He was promoted to Associate Member of the Ludwig Institute in 2002 and to Associate Professor (ad personam) at the University of Lausanne in 2006. Since 2011 he is a Full Professor at the Ludwig Center for Cancer Research of the University of Lausanne.



Cytotoxic CD8 T cells and Natural Killer (NK) cells can kill other cells. They play important protective roles against infected and cancerous cells. We aim at discovering factors that influence the ability of cytotoxic lymphocytes to detect and eliminate diseased cells.



NK cells and CD8 T have the ability to determine whether host cells are healthy or diseased and to kill the latter. The recognition strategies used by NK cells and CD8 T cells are fundamentally different. Whereas T cells assemble almost indefinite numbers of T cell receptors (TCR) for almost indefinite numbers of non-self ligands, NK cells determine whether host cells express specific self-ligands at a normal surface density. To do so NK cells use arrays of two types of receptors, one that stimulates the NK cell, and another one that inhibits. Upon the interaction with a host cell, it is the balance between inhibitory and activation signals that determines whether the NK cell will kill. Besides the specific recognition of diseased host cells a multitude of additional developmental, extrinsic and intrinsic factors influence the effectiveness of cytotoxic effector cells. We aim at the discovery of such factors to understand their role in disease and to improve the protective effect of cytotoxic effectors against disease.



Lymphocyte development

The maintenance of self-renewing tissues in the adult such as the skin or the gut depends on conserved signaling pathways such as the Wnt (wingless/int-1) pathway. Wnt signaling via the canonical pathway is characterized by the intracellular stabilization of β-catenin upon the binding of extracellular Wnt proteins to cells. Beta-catenin induces transient transcriptional responses via the binding to transcription factors of the Tcf/Lef family (T cell factor/Lymphoid enhancer factor). Uncontrolled Wnt signaling due to mutations in pathway components is implicated in a variety of cancers.

The role of the canonical Wnt pathway for lymphocyte development and/or haematopoiesis in general has been unclear. We have found that NK cell development was impaired in mice lacking Tcf-1 (Tcf7) (generated and provided by Hans Clevers, Utrecht), one of the nuclear effectors of the canonical Wnt signaling pathway. In addition we found that Tcf-1 deficiency resulted in the accelerated death of CD4+8+ thymocytes, a critical developmental intermediate of T cell development. A genetic complementation approach revealed that the N-terminal domain in Tcf-1 was needed for thymocyte development (Ioannidis et al. 2001). However, we have further found that T cell development was normal in the combined absence of β-catenin and γ-catenin (plakoglobin), the only known alternative transmitter of canonical Wnt signals. In addition to T cell development, NK cell development and haematopoiesis was also normal in the combined absence of β- and γ-catenin (Jeannet et al. 2008). These data suggest that Wnt signal transduction by haematopoietic cells is not essential for steady-state haematopoiesis and stem cell maintenance in vivo.


The fact that steady-state haematopoiesis occurs in the absence of β− and γ-catenin does not exclude the possibility that aberrant Wnt signaling perturbs haematopoiesis. Along these lines we are currently addressing the role canonical Wnt pathway for the induction and/or the maintenance of leukemia.

CD8 T cell function

Even though lymphocyte development does not require β− and γ-catenin it was possible that the Wnt signaling pathway plays roles for lymphocyte function. To begin to address this issue we infected mice with Lymphocytic Choriomeningitis Virus (LCMV), which is normally controlled by CD8+ T cells. Mice lacking Tcf-1 mounted a normal primary CD8 T cell response to LCMV infection. In addition the effector and the effector memory CD8 T cell responses were unimpaired. However, Tcf-1-deficient CD8 T cells had a cell-intrinsic defect to expand upon secondary virus challenge and to protect from recurrent virus infection. Functional CD8 T cell memory was dependent on the catenin-binding domain in Tcf-1 and on the Wnt signaling intermediates β-catenin and/or γ-catenin (Jeannet et al 2010). These findings demonstrated that the canonical Wnt signaling pathway plays an essential role for memory CD8 T cell differentiation under physiological conditions in vivo. They raise the possibility that modulation of Wnt signaling may be exploited to improve the generation of CD8 memory T cells during vaccination or for therapies designed to promote sustained cytotoxic CD8 T cell responses against tumors.

NK cell effector response

NK cell effector functions are under the control of inhibitory and activating cell surface receptors, which interact with ligands expressed on other cells. For example the mouse NK cell receptor Ly49A interacts with Major Histocompatibility Complex class I (MHC-I) molecules on other cells and inhibits the NK cells' effector functions. When host cells lose MHC-I expression, which is a feature of infected or transformed cells, they become susceptible to NK cell mediated attack (this has been termed "missing-self" recognition.
Unexpectedly, we found that Ly49A not only interacts with MHC-I molecules expressed on other cells (in trans), but also with the MHC-I molecules expressed by the NK cell (in cis). While trans interaction mediates the classical inhibition of NK cell effector function, there was no evidence that cis interaction resulted in inhibitory signaling. Cis association and trans interaction occurs via the same binding site and is thus mutually exclusive. Indeed, cis association sequesters Ly49A and restricts the number of inhibitory receptors available to bind MHC-I on target cells. This reduces NK cell inhibition via Ly49A. By lowering the threshold at which NK cell activation exceeds NK cell inhibition, cis interaction allows an optimal discrimination of normal from diseased host cells (Doucey et al., 2004). Subsequent to Ly49A, additional cell surface receptors have been reported to have the unusual property to bind the same ligand both in trans as well in cis (for review see (Held and Mariuzza, 2008).

NK cell education

NK cells, which express inhibitory MHC-I receptors, express activation receptors that respond efficiently to stimulation, while NK cells that lack MHC-I receptors respond poorly. The basis for this MHC-I guided “education” is not well understood but the MHC-I-dependent effect allows efficient NK cell reactions to diseased host cells that have lost MHC-I expression. Current models generally imply that NK cell education is dependent on an interaction of inhibitory receptors with MHC-I expressed on other cells. However, since the inhibitory Ly49A receptor can also bind MHC-I on the NK cell itself (in cis) we tested whether cis interaction played a role for education. To address this issue we designed a Ly49A variant, which can bind MHC-I molecules expressed on other cells but not those expressed in cis. While this Ly49A receptor variant readily inhibited NK cell effector function, it failed to educate NK cells (Chalifour et al. 2009). These data dissociate the classical inhibitory from an educating function of Ly49A and they suggest that cis interaction of Ly49A is necessary for NK cell education.

Structural basis of MHC-I recognition in cis versus trans

It was unclear how cell surface receptors can bind ligand expressed in cis and trans, and why the two types of interactions can have distinct functional outcomes. In close collaboration with Roy Mariuzza (University of Maryland) we have found that cis and trans interactions of Ly49 receptors are based on two distinct receptor conformations. Ly49 receptors are homodimeric type II glycoproteins, with each chain composed of a ligand-binding C-type lectin-like domain, connected by a stalk of approximately 70 residues to the transmembrane and cytoplasmic domain. Ly49 receptors were found to bind two MHC-I molecules in trans when the two ligand-binding domains are back-folded onto the long stalk region. In contrast, dissociation of the ligand-binding domains from the stalk and their reorientation relative to the NK cell membrane allow monovalent binding of MHC-I in cis (Back et al., 2009). The distinct conformations (back-folded and extended) define the structural basis for cistrans binding by Ly49 receptors and they represent the likely explanation for the divergent functional consequences of cis versus trans interactions.



Models for cis and trans interactions of Ly49 receptors with MHC-I. The α1, α2 and α3 domains of the MHC-I heavy chain are cyan; β2-microglobulin (β2m) is blue; Ly49 is red. The Ly49 homodimer on the natural killer (NK) cell (bottom) binds two MHC-I molecules on the target cell (top). The stalks (green) are drawn arbitrarily. To bind in trans, the stalks may adopt a back-folded conformation (left). For cis interaction of Ly49 with MHC-I, the Ly49 homodimer binds one MHC-I molecule on the NK cell itself. In this case, the stalks may assume an extended conformation (right). Adapted from Back et al. 2009. Immunity 8: 269.

NK cell inactivation

In addition to detecting cells with MHC-I loss, NK cells can also react to host cells, which up-regulate specific cell surface molecules due to stress, infection or transformation (termed "induced-self" recognition). Such events can be recognized by the activating NKG2D receptor, which is expressed by NK cells as well as activated CD8 T cells.
While the transient encounter with NKG2D ligand-expressing cells activates NK cells, we have found that the chronic exposure of NK cells to cell-bound NKG2D ligand profoundly impairs NKG2D function. We have further shown that the sustained engagement not only impacted the function of the NKG2D receptor but also that of heterologous NK cell activation pathways, which use signal transduction pathways that are distinct from those used by NKG2D. We have termed this phenomenon “cross-tolerance" induction (Coudert et al., 2008). These experiments revealed that the persistent engagement of a single NK cell activation receptors can broadly but reversibly render mature NK cells dysfunctional. The suppression of the NK cells’ cytolytic function likely reduces the NK cells’ efficacy to control endogenous and exogenous stress. However this may be needed to limit tissue damage.


See text for details, adapted from (Held 2008 Eur. J. Immunol. 38 4)

NK cell mediated control of leukemia

NK cells utilize several distinct recognition strategies to detect diseased and/or foreign cells. Based on the retrospective analysis of patient data, many, if not all of these strategies have been proposed to contribute to the recognition of leukemic cells following bone marrow transplantation. However, it not known whether NK cells exert a direct anti-leukemia effect in vivo and, if so, which NK cell recognition strategy is best suited to protect against leukemia. To address this issue we are currently testing, using defined mouse models, whether NK cells have the capacity to control specific types of primary leukemia in vivo.



Selected Publications

Back, J., Malchiodi, E.L., Cho, S., Scarpellino, S., Schneider, P., Melissa C. Kerzic, M.C., Mariuzza, R.A., Held, W. 2009. Distinct conformations of Ly49 Natural Killer cell receptors mediate MHC class I recognition in trans and cis. Immunity. 31: 598-608

Chalifour, A., Scarpellino, L., Back, J., Brodin, P., Devèvre, E., Gros, F., Lévy, F., Leclercq, G., Höglund, P., Beermann, F., Held, W. 2009. A role for Ly49A-MHC class I cis interaction for NK cell education. Immunity. 30: 337-347.

Coudert, J.D., Scarpellino, L., Gros, F., Vivier, E. and Held, W. 2008. Sustained NKG2D engagement induces cross-tolerance of multiple distinct NK cell activation pathways. Blood. 111: 3571-3578.

Doucey, M.-A., Scarpellino, L., Zimmer, J., Guillaume, P., Luescher, I.F., Bron, C., Held, W. 2004. Cis-association of Ly49A with MHC class I restricts natural killer cell inhibition. Nature Immunology 5: 328-336

Held,, W and Mariuzza, R.A. 2008. Cis interactions of immunoreceptors with MHC and non-MHC ligands. Nature Reviews Immunology. 8: 269-278

Jeannet, G., Scheller, M., Scarpellino, L, Duboux, S., Gardiol, N., Back, J, Kuttler, F., Malanchi, I., Birchmeier, W., Leutz, A., Huelsken, J. and Held, W. 2008 Long-term, multilineage hematopoiesis occurs in the combined absence of β-catenin and γ-catenin. Blood. 111: 142-149.

Jeannet, G., Boudousquié, C., Gardiol, N., Kang, J., Huelsken J. and Held, W. 2010. Essential role of the Wnt pathway effector Tcf-1 (T cell factor 1) for the establishment of functional CD8 T cell memory. Proc. Natl. Acad. Sci. USA. 107: 9777-9782.

List of Publications

pdf   Publications-Held.pdf  (163 Kb)



Elisenda Alari Pahissa, Postdoctoral Fellow

Maxime Danilo, Ph.D. Student

Camille Grandclement, Postdoctoral Fellow

Beena Jeevan-Raj, Postdoctoral Fellow

Jasmine Gehrig, Ph.D. Student

Irene Pizzitola, Postdoctoral Fellow

Laurène Pousse, Ph.D. Student


 in this site:

Ch. des Boveresses 155 - CH-1066 Epalinges  - Switzerland  -  Phone +41 21 692 59 66  -  Fax  +41 21 692 59 95