Polarity proteins have recently been implicated as key regulators of T cell migration, signaling, effector function, and fate 
. Dlg1, a member of the ancestral Scribble polarity network, directly binds and regulates the activity of the proximal TCR-associated kinases and has been implicated in regulating TCR-mediated signal transduction and cell polarity in the context of IS formation and cell migration. Although studies using Dlg1 knockdown or over-expression technologies in primary T cells and transformed cell lines have demonstrated a role for Dlg1 in T cell signaling, activation, and the establishment of synaptic polarity during antigen recognition 
, its function in the context of T cell development, immune homeostasis or immune response in vivo
remains unclear. This is due in part to seemingly conflicting reports which have alternatively suggested that Dlg1 is a negative regulator 
, or a positive regulator 
of T cell function.
In this study, three independent dlg1
-deficient mouse models were generated and examined in an effort to unravel the functional significance of Dlg1 in T cell development and effector function in a null background. In general, we found that germline and conditional dlg1
knockout mice had no apparent defects in CD4+ or CD8+ T cell polarization-dependent events such as development, migration, activation, signaling or proliferation. These data were surprising considering previous reports demonstrating a role for Dlg1 in regulating TCR-mediated actin polymerization, signal specificity and function, in the context of acute knockdown or over-expression in mature T cells. While we observed diminished cytoskeletal reorganization in primary mouse CD8+ 
and human Jurkat T cells () with acute Dlg1 knockdown, we did not observe defects in mature T cells from conditional or germline dlg1
knockout mouse models (). These data suggest that the timing and/or duration of dlg1
ablation may greatly affect the phenotype of Dlg1-deficient cells and impact the ability to examine the role Dlg1 on T cell function. This hypothesis is supported by studies of Th1 and Th2-type cytokine secretion, which demonstrated differential cytokine secretion in T cells with an acute or conditional loss of Dlg1, but not a germline loss of dlg1
(). Because the conditional dlg1
knockout leads to Dlg1 ablation later in T cell development, we hypothesize that a shorter developmental window was available for selection of cells that have compensated for Dlg1 loss. Indeed, while all three models demonstrated modest, if any, defects, it is noteworthy that the most significant defects observed (i.e. in Th1 and Th2 development) occurred under the conditions where dlg1
was ablated later in development or in experiments using acute knockdown strategies.
The phenomenon of compensatory factors that mask the effects of Dlg1 deficiency is not unprecedented. In neuronal cells, evidence suggests a key role for Dlg1 in regulating synaptic AMPA receptor trafficking. However neurons from mouse embryos in which the dlg1
gene has been ablated develop normally and form synapses with normal levels of AMPA receptors and no detectable abnormalities 
. Investigators have hypothesized that the lack of a phenotype in Dlg1-mutant cells could be caused by the compensation of other Dlg-family members (Dlg2, Dlg3, Dlg4) or other adaptive processes during neuronal development and synapse maturation that could compensate for the normal function of Dlg1 
. Based on the results of our collaborative study, we hypothesize that similar event(s) could be occurring in developing T lymphocytes, making it difficult to appreciate the functional role of Dlg1 in developing and peripheral lymphocyte populations. Further support for this working hypothesis comes not only from comparative investigation within this study, but also from comparison of previously published works demonstrating a critical role for Dlg1 in cytoskeletal organization 
, alternative p38 activation 
, NFAT activation 
and effector function 
, in the context of acute knockdown in T cells ( and ). An important area for future investigation will be to identify the changes in Dlg family protein expression and/or the regulation of other proteins that may compensate for the loss of Dlg1; this will lay the groundwork to more fully assess the significance of this pathway for T cell development and function.
Notably, our studies also uncovered novel contextual roles for Dlg1 as both a positive and negative regulator of T cell function, as Dlg1 ablation was found to inhibit Th1, while enhancing Th2, cytokine production in CD4+ T cells. Initial studies of Dlg1 by Xavier et al. found that the overexpression of Dlg1 in conjunction with Vav1, attenuated Vav1-induced NFAT activity in Jurkat T cells. These studies also found that the long term diminution of Dlg1 in Jurkat T cells via stable siRNA-based knockdown resulted in impaired NFAT reporter activation 
. In a follow-up study, Stephenson et al provided data indicating that Dlg1 may negatively regulate T cell proliferation utilizing mouse dlg1
-deficient T cells generated by recombination-activating gene 2 (rag2
)-deficient complementation 
. Both reports supported a role for Dlg1 as a negative regulator of T cell activation and function by suppressing NFAT-mediated transcription and cycle entry, respectively. However, two studies by Round et al., utilizing acute siRNA-based knockdown in mouse TCR transgenic CD8+ T cells found that Dlg1 knockdown attenuated NFAT-mediated transcription of endogenously regulated NFAT genes (NFATc1 and IFNγ), while complementary studies of Dlg1 overexpression resulted in a tightly regulated, dose-dependent enhancement of NFAT-mediated transcription. Moreover, Round et al provided clear biochemical evidence for a direct interaction between the MAP kinase p38 and Dlg1, which allows for nucleation of a signaling complex that results in alternative p38 activation and downstream NFAT phosphorylation 
Collectively, these results may appear conflicting and contradictory, regarding a role for Dlg1 as a positive or negative regulator of T cell activation and function. However, our data supports a view where Dlg1 may specify TCR signals that enhance or attenuate particular T cell responses in a context dependent manner. Indeed, the differential assembly of membrane microdomains, signaling molecules, and co-polarization of cytokine receptors at the IS has been implicated in controlling memory versus effector cell development 
, as well as Th1/Th2 lineage commitment and effector function 
. In this study, both conditional knockout and acute knockdown approaches demonstrated that Dlg1 positively regulates Th1 cytokine production, while negatively regulating Th2 cytokine production. Specifically, acute siRNA-mediated knockdown of Dlg1in differentiated Th1 and Th2 cells inhibited IFNγ and TNFα production in Th1 cells, while enhancing IL-4 production in Th2 cells (). While similar results were observed in the conditional knockout system, we did not observe differences in cytokine production in T cells derived from dlg1
germline knockout mice.
More recently, two groups have shown experimental data which support a model in which Dlg1 couples p38 to NFAT activation and T cell function in primary human T cells to promote distinct signaling pathways in distinct T cell subsets 
. Work by Zanin-Zhorov et al. has demonstrated that in primary human CD4+CD25+ FoxP3+ T regulatory (Treg) cells, Dlg1 accumulation at the IS is significantly higher than in effector CD4+CD25− T cells. In addition, diminution of Dlg1 expression impaired Treg cell suppression activity, caused a reduction in the amount of Foxp3 per cell, and led to diminished alternative p38 phosphorylation and NFATc1 activation, while enhancing Akt phosphorylation in response to TCR stimulation. These data support our working hypothesis by demonstrating that in human Treg cells, Dlg1 functions as both a positive and negative regulator of discrete signal transduction pathways; promoting alternative p38 activation, while inhibiting Akt activation. Similarly, opposing effects of Dlg1 activity have been observed between TCR-induced p38 and ERK activation 
. Zanin-Zhorov et al. also elucidate a relationship between Dlg1, NFAT, and Foxp3, which may be of significance since both Foxp3 and NFAT are critical for Treg cell function, commitment, and maintenance 
. In this study our data, although subtle, hints at a possible relationship between Foxp3 and Dlg1 in Treg cell commitment or maintenance; we observed a trend towards a decreased percentage of CD4+Foxp3+ cells in dlg1ko;BG
germline deficient mice (). It is interesting to speculate that while compensatory mechanisms may have allowed for T cell development in the mouse models investigated here, Dlg1 may be indispensable for the maintenance as well as function of Treg cell populations. Zanin-Zhorov et al. indicated that Dlg1 recruitment to the IS was diminished in patients with rheumatoid arthritis, suggesting that Dlg1 function and the regulation of the alternative p38 pathway may contribute to dysregulated Treg cell function in rheumatoid arthritis or human autoimmune conditions 
While addressing the entire list of discrepancies observed among all the groups investigating the role of Dlg1 in T lymphocytes is beyond the scope of this paper, it seems clear from our and others’ results reported to date that Dlg1 can facilitate or attenuate discrete TCR signals and that its role in regulating T cell functionality can vary within T cell subsets or at particular stages of T cell development 
If there are compensatory mechanisms in play, then many of our studies of the role of Dlg1 in T cell function, including surface receptor regulation during primary activation, uropod formation and localization of DPC and IS constituents, and proliferation should be revisited in experimental systems where Dlg1 is acutely knocked out rather than stably deleted during T cell development. Because these data preclude further analysis of Dlg1 using long-term genetic approaches, future efforts to characterize the role of dlg1 in T cell development and function should make use of genetic systems that allow one to acutely delete dlg1. For example, an estrogen-receptor Cre-recombinase (ER-Cre) system where “floxed” alleles can be induced to recombine following exposure to tamoxifen permits targeted and controlled acute Dlg1 ablation might work well. Until the ideal model systems are developed and validated, however, acute knock-down of Dlg1 remains a potent strategy by which to continue exploring the functional significance of Dlg1.