No antigen-presentation defect in Unc93b13d/3d (3d) mice

Abstract

To the editor: Unc93b1 is a critical regulator of Toll-Like Receptor (TLR) signaling1. The original description of mice with an inactivating mutation in Unc93b1 (H412R, also termed 3d) reported that these mice have defective TLR3, TLR7 and TLR9 signaling as well as major histocompatibility complex (MHC) class II presentation and cross-presentation defects1. Mutations in Unc93b1 have also been identified as a genetic etiological factor in herpes simplex virus-1 encephalitis, where susceptibility is associated with defects in TLR signaling2. Interestingly, in vitro studies of human cells have not found defects in MHC class II presentation after Unc93b1 knock-down3, raising the possibility that the role of Unc93b1 in antigen presentation, at least for MHC class II, is limited to mice. Further studies have shown that Unc93b1 controls TLR signaling by acting as a trafficking chaperone, associating with TLRs (such as TLR3, 7, 8, 9, 11 and 13) and delivering them from the endoplasmic reticulum to endosomes4, 5, 6. On the other hand, despite ample studies of Unc93b13d/3d mice, a mechanistic understanding for the role of Unc93b1 in antigen presentation is still lacking. Because of the discrepancies between human and mouse studies and the lack of any mechanistic basis for an antigen presentation defect in Unc93b13d/3d mice, we sought to confirm the antigen presentation defect reported in Tabeta et al.’s original description of Unc93b13d/3d mice1. First, we measured proliferation of ovalbumin (OVA)-specific OT-II T cells transferred into either wild-type or Unc93b13d/3d recipients, after injection of soluble OVA. In contrast to the findings reported by Tabeta et al., cells transferred into an Unc93b13d/3d host proliferated to a similar extent as in a wild-type host, both in terms of frequency of proliferating cells and average number of divisions (Fig.1a,b). Of note, the activation state of cells, followed by CD44 upregulation, was also identical in both wild-type and Unc93b1-deficient hosts (not shown). Figure 1 Antigen presentation is unaffected in Unc93b13d/3d mice Next, we tested the ability of Unc93b13d/3d hosts to cross-present antigens to CD8+ T cells, using Act-OVA cells as an antigen source, as described in the initial characterization of Unc93b13d/3d mice1. To ensure that we only measured cross-presentation, we injected OVA-expressing H-2Kbm1 cells in wild-type and Unc93b13d/3d mice. As H-2Kbm1 cells cannot directly present peptides derived from OVA on MHC class I molecules7, CD8+ T cell activation must be the result of cross-presentation by host cells. In this setting, we also failed to detect any significant differences between the proliferation profiles of OVA-specific OT-I T cells transferred in wild-type or Unc93b1-deficient hosts (Fig.1c,d). Moreover, the activation status of CD8+ T cells also appeared similar in both strains. Indeed, upregulation of CD44 and production of interferon-γ (IFNγ) after restimulation with either peptide or PMA-ionomycin (not shown) were not affected by the genotype of the host (Fig.1e, f). We also studied CD4+ T cell proliferation in this setting and found similar proliferation 7 days after activation in wild-type and Unc93b13d/3d mice (not shown). Taken together, our data do not support a direct role for Unc93b1 in antigen presentation – either MHC class II presentation or cross-presentation. We do observe defective TLR3, TLR7, and TLR9 function as originally reported for Unc93b13d/3d mice5. Given the well-recognized role of TLRs in activating antigen-presenting cells to enhance T cell priming8, it is certainly likely that Unc93b1 plays an indirect role in T cell activation in some models through its control of TLR trafficking. It is possible that differences in the level of contaminating TLR ligands may account for the discrepancy between our findings and those published by Tabeta et al. Regardless of the explanation, it is important to establish that there is no evidence to support a function for Unc93b1 beyond control of TLR localization. This clarification is particularly relevant to studies that employ Unc93b13d/3d mice as a model of multiple TLR deficiencies9 because any additional functions for Unc93b1 would complicate interpretation of such studies. Thus, Unc93b13d/3d mice, especially when crossed with other TLR-deficient strains, may offer a more straightforward model of multiple TLR deficiencies than the commonly used Myd88−/−Trif−/− mice, which also lack signaling downstream of the Interleukin-1 receptor family.