In this issue of Blood, Nair et al describe a new population of type II natural killer T (NKT) cells with follicular helper phenotype (TFH), which is more abundant in patients and mice with Gaucher disease (GD) and is capable of regulating B-cell activity.1
It is now well established that αβ T lymphocytes recognize not only peptide epitopes but also lipid and glycolipid antigens presented by nonpolymorphic CD1 molecules.2 Of the 5 CD1 family members expressed in humans, CD1d molecules are highly conserved in mammalian species and present endogenous and microbial glycolipids to a subset of T cells known as NKT cells, so called because they express natural killer cell markers in addition to a T-cell receptor (TCR).2 Two subsets of CD1d-restricted NKT cells have been described: (1) invariant NKT (iNKT) or type I NKT cells, expressing a semi-invariant TCR; and (2) type II NKT cells, with a much broader TCR repertoire. The prototype lipid antigen for type I iNKT cells is α-galactosylceramide (α-GalCer), a marine-sponge–derived glycosphingolipid, and the availability of CD1d-α-GalCer tetramers has allowed a detailed understanding of the role of iNKT cells in several disease settings.2 Conversely, there is no such prototypic lipid antigen recognized by type II NKT cells, and because of the lack of specific tools to identify them, this population has been less characterized. The best-studied antigen for type II NKT cells to date is sulfatide, a myelin-derived glycolipid.3 In addition, reactivity to lysophospholipids, which are generated during inflammatory responses following hepatitis B infection or in multiple myeloma, has been reported.3
Both subsets of NKT cells are innate-like lymphocytes that rapidly produce large amounts of cytokines upon TCR engagement and play an important immune-regulatory role in inflammatory conditions, autoimmunity, and cancer.2 Hence, understanding the identity of antigens that trigger NKT cell activation in health and disease is of importance, as harnessing these cells in vivo may provide therapeutic opportunities to either enhance or suppress immune responses.
Dysregulation of lipid metabolism occurring in obesity and congenital metabolic disorders has the potential to affect the development and/or function of NKT cells and the concomitant chronic inflammation.4,5 Gaucher disease (GD) is a disorder of glycosphingolipid (GSL) metabolism due to an inherited deficiency of the acidic β-glucosidase enzyme, resulting in progressive lysosomal accumulation of β-glucosylceramide (βGL1) and its deacylated product, glucosylsphingosine (Lyso-GL1; LGL1). Accumulation of these lipids in GD patients is associated with chronic inflammation and B-cell activation, often manifested by polyclonal and monoclonal gammopathy.6 To gain insights into mechanisms underlying lipid-associated inflammation in GD patients, Nair and al set out to analyze βGL1- and LGL1-specific T-cell responses.
Using βGL1- and LGL1-loaded CD1d tetramers, the authors convincingly demonstrated a lipid-specific CD1d-restricted NKT cell population in the blood of healthy donors and the spleen and liver of wild-type mice. Unlike type I NKT cells, human βGL1- and LGL1-specific NKT cells have a naive phenotype and a higher proportion of CD8 expression. Furthermore, although a detailed analysis of the functional specificity and affinity of individual Vβ families for βGL1- and LGL1-CD1d complexes remains to be done, analysis of TCRβ usage following cell sorting and lipid-specific expansion revealed a much broader T-cell repertoire than anticipated from previously published data on sulfatide-specific T cells.3,7 Results obtained with in vitro–expanded βGL1- and LGL1-specific NKT cells revealed that, despite a similar TCRβ usage, the 2 populations are not cross-reactive, as they can specifically recognize βGL1 and LGL1 pulsed target cells, respectively. From the broad pattern of tetramer staining observed, a wide range of binding affinities is expected, suggesting that combined biophysical and structural data will eventually elucidate the fine details of this molecular recognition.
Despite expression of the transcription factor PLZF, a distinct transcriptional profile marks GL1- and LGL1-specific NKT cells in comparison with type I NKT cells, with a prominent Th-17 and a TFH signature. Interestingly, the TFH signature is present also at steady state, unlike type I NKT cells, a fraction of which acquire it only upon antigen stimulation. Experiments performed in wild-type mice revealed that in vivo activation of βGL1- and LGL1-specific T cells with their cognate antigens led to the induction of a germinal center B-cell response and lipid-specific antibodies. Likewise, in vitro activation of human βGL1- and LGL1-specific T cells induced plasmablast differentiation from cocultured autologous B cells. However, it remains to be determined whether the B memory responses elicited by type II NKT cells are short lived, as seen in the case of type I NKT TFH.8
These results are of great interest, as the authors report an increased frequency of LGL1-specific T cells in a mouse model of GD and in patients with GD (over 20-fold in mice and 3-fold in humans). Notably, the frequency of type I NKT in GD mice is significantly reduced, in agreement with data showing impairment in their selection in mouse models of lysosomal storage disorders.5,9 The mechanisms by which the frequency of LGL1-specific T cells is selectively enhanced in GD patients and mice remain unclear. However, it is tempting to speculate that enhanced availability of LGL1, and/or factors regulating its loading or intracellular trafficking (in the presence of lysosomal GSL accumulation), may affect the density of LGL1-CD1d complexes presented by antigen-presenting cells. Determination of the TCR usage of LGL1-specific T cells in GD patients and their affinity of binding to LGL1-CD1d complexes will also provide important insights into the understanding of their selective expansion and their potential role in GD. Interestingly, LGL1-specific T cells in GD patients display a memory phenotype, consistent with in vivo antigen exposure, and their increase correlates with clinical disease activity and serum levels of inflammatory cytokines. However, it remains to be determined whether antibodies to βGL1 and LGL1 are detectable in the serum of mice and humans with GD and whether their titers correlate with frequencies of lipid-specific NKT cells and disease activity. Selective deletion of CD1d expression on B cells or myeloid cells in murine models of GD will be of importance to provide conclusive evidence in support of the hypothesis that βGL1- and LGL1-specific T cells might modulate B-cell activation and chronic inflammation (see figure). The imbalance between type I and type II NKT cells, together with chronic inflammation, may also be contributing to the onset of hematologic malignancies, often associated with the progression of GD.10 Further elucidation of the role of type I and II NKT cells will advance our understanding of the pathophysiology of GD and associated disorders and possibly open new therapeutic strategies, in association with the currently available enzyme-replacement therapy and substrate-reduction therapy.
Conflict-of-interest disclosure: The authors declare no competing financial interests.