Lymphocyte signaling and immune cell response

National Institute Of Arthritis And Musculoskeletal And Skin Diseases

Investigators

Abstract

The study of the chromatin landscape has shown us how genomic architecture can control immune response. We established that genomic accessibility plays a factor in the differential rate of response of innate versus adaptive immune cells. Super-enhancers (SEs) are another distinct feature of the chromatin landscape where multiple enhancers, transcription factors (TFs), RNA polymerase II and transcriptional co-activators conjugate to regulate the expression of specific genes. Genes regulated by SEs are highly linked to the cell identity, but also denote functional hubs. It has been shown that SEs in adaptive immune cells are linked to genes that encode cytokines, cytokine receptors and transcription factors, which are vital for their normal development and function. In addition, SE architecture is also enriched for loci encoding long noncoding RNAs, which often flank key protein coding genes. The goal of our research is to understand how the chromatin landscape regulates immune cell development and function. To do so, we identified SEs in natural killer (NK) cells to determine if there are genes within these regions with unknown relevance to NK cell biology. We hypothesize that genes transcribed from SEs in NK cells are likely to play critical roles in NK cell development and effector function, and may provide clues to mechanism of removal of virally-infected cells and cancer cells. We anticipated that this SE-based approach has the potential to identify key NK cell genes that were previously overlooked but likely to have critical functions. Within the top-ranking SE of NK cells was Ugcg, a gene that encodes UDP-glucose ceramide glucosyltransferase (UGCG), which is an enzyme that catalyzes the first glycosylation step in the synthesis of glycosphingolipids (GSLs). GSLs are important components of the cell membrane that regulate lipid raft organization, signal transduction and cell migration. Adjacent to Ugcg is a putative long noncoding RNA (lncRNA), Gm12596, which also lies within this SE. LncRNAs are noncoding RNAs that are at least 200 bp in length and control multiple aspects of cell biology. Both Ugcg and Gm12596 are highly expressed in NK cells and less so in other hematopoietic cells—hematopoietic stem cells, early progenitors, and innate and adaptive lymphoid cells. To determine if there is any influence on immune cell development, we developed a conditional mouse model that deletes Ugcg in all hematopoietic cells (Vav1cre-Ugcgloxp). We found that most hematopoietic cells in the bone marrow and spleen developed, however there was a total ablation of NK cells. Furthermore, the use of the drug ibiglustat, which inhibits UGCG activity, allowed us to study viable NK cells with reduced GSL synthesis. With this model, we determined that not only is NK cell survival impacted, but function as well. We found that NK cell cytolytic function and response to infection was greatly reduced with inhibition of UGCG activity. Taken together, we found that Ugcg is critical for NK cell survival and function. Over the past year, we have focused our research on understanding which enzyme and GSL downstream of UGCG is required for NK cell survival and function. The loss of cytotoxic granules with inhibition of UGCG activity suggest an intracellular role for GSLs in NK cells. UGCG generates GlcCer, the precursor for about 400 different species of GSLs. Enzymes downstream of UGCG process this precursor further to create the various classes of GSLs. B4GALNT1 is one of those enzymes that utilizes simple GSLs to generate complex GSLs. One of the major differences between simple and complex GSLs is the cellular location. Simple GSLs can be found on the cell surface or within the cell, while complex GSLs are only located on the cell surface. The intracellular loss of cytotoxic granules with inhibition of UGCG activity suggest that it is simple GSLs that are required to maintain cytotoxic granules. Therefore, we studied mice with loss of B4galnt1 and found that the quantity of cytotoxic granules within these mice were normal to wild type littermates. Additionally, we found that there was no change in NK cell frequency. This data strongly suggest that simple GSLs are required for both NK cell survival and maintenance of cytotoxic granules. There are five simple GSLs—GlcCer, LacCer, GM3, GD3 and GT3. To determine which simple GSL is required for NK cell survival, we utilized CRISPR-Cas9 technology to reduce synthesis of GlcCer, LacCer and GM3 by deleting the enzymes that generate these GSLs—Ugcg (to inhibit synthesis of GlcCer), B4galt5 and B4galt6 (for LacCer), and St3gal5 (for GM3). We found that along with deletion of Ugcg, loss of B4galt5 and B4galt6 also reduced survival of NK cells. However, the loss of St3gal5 did not reduce cell survival. Furthermore, we identified B4galt5 (but not B4galt6) to be highly expressed in NK cells and located within an SE. In all, this suggest that B4galt5 and LacCer are required for survival of NK cells. We imaged LacCer within NK cells to determine its location and whether it colocalizes with cytotoxic granules. We found LacCer on the plasma membrane and clustered with cytotoxic granules. With inhibition of UGCG activity, we found that both LacCer and cytotoxic granules are more dispersed than clustered within the cell, and suggest that the integrity of cytotoxic granules is impaired. Additionally, when we treat NK cells with LacCer prior to ibiglustat treatment, cell death induced by ibiglustat treatment is repressed. Altogether, our work show that NK cells require simple GSLs, specifically LacCer, for survival and cytolytic function. This work has been published in Cell (doi: 10.1016/j.cell.2025.04.007). Based on RNA-seq and lipidomic studies, NK cells express higher levels of B4galt5 and LacCer than both CD4+ and CD8+ T cells at steady state. However, after MCMV (mouse cytomegalovirus) infection, CD8+ T cells upregulate both B4galt5 and LacCer. CD8+ T cells are adaptive lymphocytes that become activated and cytotoxic after infection. With conditional loss of Ugcg, the frequency of naive or non-activated CD8+ T are not impacted. However, after MCMV infection, expansion of effector memory and antigen-specific CD8+ T cells, which are activated and cytotoxic, is impaired. This suggests a shared requirement of GSLs for NK and CD8+ T cells in their cytotoxic state - also published in Cell (doi: 10.1016/j.cell.2025.04.007). Another interesting observation we see in CD8+ T cells is upregulation of CCR7, which is a chemokine receptor that restricts T cells to lymphoid tissues. Constitutively high expression of CCR7 can reduce mobility of T cells to sites of infection and cancer. To determine if T cell trafficking is impaired, we injected B16-F10 melanoma cells into CD4cre-Ugcg mice, which lack Ugcg only in T cells for 20 days. We found that mice without Ugcg in T cells have larger tumors than their wild-type littermates. This corresponded with fewer T cells identified in the tumor by flow cytometry. This data suggests that the high level of CCR7 impedes trafficking of T cells into the tumor microenvironment. Further investigation is underway to understand the mechanism of how GSLs regulate chemokine receptors. As mentioned, Gm12596 is a lncRNA that is adjacent to Ugcg and lies within the same SE. LncRNAs play critical roles in regulating important cellular processes, including transcription. LncRNAs can regulate transcription of genes adjacent to them. Therefore, we hypothesize that Gm12596 regulates the expression of Ugcg. We have created Gm12596 knockout mice, and preliminary studies show a decrease in Ugcg expression and a decline in NK cell frequency. Further studies are underway to determine how Gm12596 regulates Ugcg expression, including whether it is the SE locus or the Gm12596 transcript that is required for Ugcg expression.