Skip to main content

Hedgehog signalling in T cells

Hedgehog (Hh) is a fascinating signalling pathway that has baffled researchers up to this day. Hh was discovered by Christiane Nüsslein-Volhard and Eric Wieschaus in a genetic screen carried out in Drosophila melanogaster1 and named based on the appearance of the mutant larvae which had denticle ‘spikes’ arranged on its back, reminiscent of a hedgehog. Later on, three vertebrate homologues of Hh – Sonic (Shh), Indian (Ihh) and Desert Hh (Dhh) – were identified and the pathway initiated by these ligands turned out to be crucial for embryonic development and tissue maintenance in the adult.2

Hedgehog signalling in vertebrates – it all happens at the cilium

A major scientific breakthrough in our understanding of the Hh pathway was the observation that, in vertebrates, Hh signalling is tied to the primary cilium, an ancient sensory organelle that projects from the surface of cells.3 In canonical Hh signalling, Hh ligands are secreted by Hh-producing cells, which generates an extracellular ligand gradient. Hh-responsive cells respond to this signalling cue; Hh ligands bind to the transmembrane receptor, Patched (Ptch), at the base of the primary cilium of the responding cell. Upon ligand binding, Ptch exits the cilium and releases its inhibition of the key signal transducer, Smoothened (Smo). Smo moves into the cilium and activates Gli transcription factors. These leave the cilium and enter the nucleus to initiate a Hh-specific target gene programme (Figure 1a). 

The immunological synapse as a ‘modified cilium’

While primary cilia are found on nearly all cells in our body, the haematopoietic lineage, including lymphocytes, was thought to be unable to form a primary cilium.4 Gillian Griffiths and Jane Stinchcombe were the first to discover that the immunological synapse formed between a cytotoxic T cell (CTL) and a target cell is structurally very similar to sites of primary cilia formation. In both structures, the centrosome is ‘docked’ at the plasma membrane via distal appendage proteins, and the Golgi apparatus and endocytic recycling compartment are polarised towards this point making the area a focus for endo- and exocytosis.5–7 This has led to the notion that the immunological synapse may represent a ‘modified cilium’.

Investigating the Hedgehog pathway in mature CD8

T cells When I joined Gillian Griffiths’ lab as a postdoc, I wanted to know whether the structural similarities between the immunological synapse and the cilium would extend to functional properties. Pioneering work by Tessa Crompton and co-workers over the last 15 years had already identified the importance of Hh signalling for multiple steps of T-cell development in the thymus8 and more recently in CD4 Th1-Th2 differentiation.9 The role of Hh signalling in mature CD8 T cells, however, was unknown (Figure 1b). In my research, I asked whether ciliary signalling is active in CD8 T cells and started with looking at the Hh pathway that had been shown to rely on a functional primary cilium. CD8 T cells are crucial for the body’s defence against infection and tumours through their ability to differentiate into CTLs and kill infected and tumour cells. Upon target cell recognition, CTLs form an immunological synapse: cortical actin clears away from the centre of this synapse and allows the centrosome to dock at the plasma membrane – precisely at the point where target cell recognition has occurred (Figure 1c). The cytotoxic granules move along the microtubules towards the centrosome and are secreted at the site where the centrosome is docked making killing very efficient and specific to the target cell.

Figure 1

Figure 1. a. Canonical Hedgehog (Hh) signalling in ciliated cells; b. Hh signalling in thymocytes and mature T cells in the periphery: Hh signalling initiated by an extracellular Hh ligand (produced by the thymic stroma, thymocytes or peripheral tissues) is indicated in dark green, and intracellular Hh signalling in light green; c. Migrating CTL (left) and synapsing with a target cell (right); d. Intracellular Hh signalling in cytotoxic T lymphocytes (CTL). Image credit: M de la Roche

Hedgehog signalling targets Rac1 required for CTL killing

During my stay in the Griffiths lab, I showed that Hh signalling plays a major role in CTL killing and discovered a novel signalling mode of the pathway.10 Naïve CD8 T cells and CTLs readily upregulate Hh signalling after T-cell receptor (TCR) triggering. Importantly, when Hh signalling was inhibited genetically or via small molecule Hh inhibitors, CTL killing was diminished. Analysing conjugates between CTL and target cells using immunohistochemistry I found that Hhinhibited CTLs were unable to clear actin from the centre of the synapse and to dock the centrosome at the plasma membrane. This combined defect in actin and microtubule reorganisation prompted me to investigate the small Rho GTPase Rac1, which had been shown to regulate actin remodelling and microtubule dynamics at the leading edge of non-immune cells.  I identified Rac1 as a novel Hh target gene in T cells that provides the CTL with the cytoskeletal machinery needed for centrosome polarisation and cytotoxic granule secretion. I next asked whether Hh ligands were involved and found that CD8 T cells do not produce Shh or Dhh, but express Ihh. Upon TCR stimulation, CD8 T cells upregulate their production of Ihh but this ligand is not processed for secretion. Subcellular localisation of Hh components revealed that Ihh colocalizes with the receptor Ptch on intracellular vesicles ready for signalling. The finding that CD8 T cells produce their own Hh ligand makes biological sense, since CTL have to eliminate infected and tumour cells throughout the body in many different environments and therefore cannot rely on an exogenous Hh gradient for function (Figure 1d). Taken together, my work has uncovered a novel role for Hh signalling in immune cell function and identified a new cell autonomous intracellular signalling mode. This has opened a new field of research and might pave the way for future therapeutic interventions to modulate Hh signalling in T cells. Right now we have only scratched the surface of Hh signalling in T cells. The role of the pathway during an immune response in vivo as well as its molecular make-up is likely to keep us busy for many more years to come. Furthermore, we know that other immune cells also polarise their centrosome to form synapses and so may form a modified cilium. It will be interesting to investigate whether Hh plays a role in these cells as well. 

Hedgehog inhibition in the clinic – a double-edged sword? Hh inhibitors have attracted attention in the clinic due to the role of amplified Hh signalling in the development of many cancers. Different Hh inhibitors are currently in trials as therapeutics for various cancers.11 However, in the majority of cancers, Hh inhibitors have been unsuccessful and alarmingly a trial in pancreatic cancer had to be prematurely stopped since the placebo group of patients was doing better than those treated with the Hh inhibitor.12 Our in vitro work suggests that Hh inhibitors also inhibit CD8 T cell killing and thereby diminish our body’s very own anti-tumour response. It is thus very timely and urgent to find out how Hh signalling works in T cells in vivo and whether we can modulate the pathway to improve T cell function – not only in patients with tumours, but also during infection and vaccination.

Maike de la Roche
Sir Henry Dale Fellow and Junior Group Leader
Cancer Research UK Cambridge Institute & University of Cambridge


  1. Nusslein-Volhard et al . 1980 Nature 287 795–801
  2. Briscoe & Therond 2013 Nature Reviews Molecular Cell Biology 14 416–429
  3. Singla & Reiter 2006 Science 313 629–633
  4. Wheatley 1995 Pathobiology 63 222–238
  5. Stinchcombe et al . 2006 Nature 443 462–465
  6. Stinchcombe et al. 2015 Current Biology  25 3239–3244
  7. Griffiths et al. 2010 Journal of Cell Biology 189 399–406
  8. Crompton et al. 2007 Nature Reviews Immunology  7 726–735
  9. Furmanski et al. 2013 Journal of Immunology 190 2641–2649
  10. de la Roche et al . 2013 Science 342 1247–1250
  11. Amakye et al . 2013 Nature Medicine 19 1410–1422
  12. Infinity Pharmaceuticals 2012