E proteins have been shown to play an important role during various stages of B and T cell development. E proteins execute this regulatory function mainly via binding to the E-box sequence around the target genes. However, the activity of E proteins can be inhibited by Id proteins, such as Id2 and Id3 in the immune system.

Previous studies have suggested crucial roles played by Id2 and Id3 in the differentiation process of certain T cell lineages. Id2 was required for CD8 T cell differentiation and activation, while Id3 was required to inhibit E2A activity upon pre-TCR and TCR signaling. During my PhD study, I performed a wide array of experiments to probe into the functions of Id2 and Id3 during the development of regulatory T cells (Treg) and innate-like follicular helper T (Tfh) cells. My work helped unveil that Id2 and Id3 expression was required for Treg cells to suppress the onset of fatal Th2-mediated inflammation. Meanwhile, I also discovered that thymic Id2 and Id3 expression was indispensable for inhibiting the expansion of innate-like Tfh cells and the pathogenesis of αβ T cell lymphoma.

Besides T cells, B cells are also a target population to be regulated by Id proteins like Id3. Ablating Id3 expression specifically in the B cells, I found that these Id3-deficient B cells were clearly impaired in their ability to differentiate into germinal centers upon antigen stimulation. In addition, without Id3 expression, these B cells can no longer elicit any IgG1 response during the memory phase upon encountering a secondary challenge.

To summarize, my PhD study helped to reveal the important functions played by Id2 and Id3 during the differentiation and activation of different lymphocytes.

Each type of cell in the immune system performs critical function to protect the body and maintain health. In the past few decades, the development and activation of T, B and NK cells have been thoroughly studied. However, it’s only until recently that we developed high throughput technology to study how genome folding structure is regulated in the immune cell development process. Here, I exploited the new techniques to study the development of B cell, T cell, and the activation process of neutrophils.

In B cell activation, the transition from the follicular B to the plasma cell stage is orchestrated by an ensemble of transcriptional regulators. We examined how changes in expression patterns of key developmental regulators relate to alterations in nuclear positioning and found that gene activation at the onset of plasma cell development was concomitant with a gain in euchromatic character for an ensemble of genomic regions that dictate plasma cell fate, including the Prdm1 and Atf4 loci. We found that to permit the onset of plasma cell development the Ebf1 locus repositioned to peri-centromeric heterochromatin. Plasma cell differentiation was also associated with increased inter-chromosomal associations concurrent with alterations in gene expression and co-localization of Prdm1, Xbp1 and Atf4 transcripts in nuclear bodies. These data indicate that Ebf1 enforces the follicular B cell fate and that the onset of plasma cell fate is orchestrated by elaborate changes in compartmentalization and inter-chromosomal associations.

In T cell differentiation, it is well known that Bcl11b specifies T cell fate. Here, we show that in developing T cells, the Bcl11b enhancer repositioned from the lamina to the nuclear interior. Our identified a non-coding RNA named ThymoD (thymocyte differentiation factor) controls the re-localization of Bcl11b enhancers. ThymoD transcription promoted demethylation at CTCF bound sites and activated cohesin-dependent looping to reposition the Bcl11b enhancer from the lamina to the nuclear interior and to juxtapose the Bcl11b enhancer and promoter into a single-loop domain. These large-scale changes in nuclear architecture were associated with the deposition of activating epigenetic marks across the loop domain, plausibly facilitating phase separation. These finding indicate how noncoding transcription orchestrates chromatin folding and compartmentalization to direct with high precision enhancer-promoter communication.

During neutrophil activation, the cell undergoes large-scale changes in nuclear morphology. We found that neutrophils rapidly (<3 hours) repositioned the ensemble of pro-inflammatory genes towards the transcriptionally permissive compartment upon microbial exposure. The repositioning of genes was closely lined with recruitment of cohesin, which permits immediate enhancer-promoter interaction and activation of genes. In the meantime, the activated enhancers were highly associated with cistromic elements occupied by PU.1, CEBPB, TFE3, JUN and FOSL2. The study revealed, upon microbial challenge, how the cohesin machinery is recruited to an activated enhancer repertoire to activate an inflammatory gene program through regulating 3D genome structure.

T cells are critical players in the adaptive immune system with broad roles that include activating other immune cells and mediating the response to threats such as viruses and cancer. A recent study demonstrated that ThymoD, a long non-coding RNA, directs T cell lineage determination by instructing changes in chromatin folding across a genomic region encoding for Bcl11b, a critical transcription factor in immune development. ThymoD facilitated binding of CTCF, an architecture protein, to bring the Bcl11b enhancer and promoter region close together and promote its transcription. Here, we examined the role of CTCF in regulating Bcl11b by ablating CTCF binding sites found within the ThymoD and Bcl11b genomic regions. We identified five sites in which CTCF binding was lost in ThymoD mutants. When we deleted them in a DN3-like cell line, each of these mutations significantly reduced Bcl11b expression. Bcl11b expression further diminished when pairs of CTCF sites were deleted together, demonstrating that partially redundant CTCF sites cooperatively orchestrate Bcl11b expression.We then asked whether CTCF binding sites were necessary for immune development by knocking out two of these sites individually in mice: CR4 and B4. Mild but significant changes in the CR4 -/- mice were observed in thymic double negative T cell populations. In order to control for age and size differences between mice, we employed a system of culturing bone marrow progenitors on stromal cells that present Notch ligands. There were dramatic changes to all double negative T cell compartments derived from CR4 -/- progenitors, with an increase of DN1/2 and a loss of DN3/4 cells. We further saw an increase in developmental markers for alternative immune lineages in progenitors derived from B4 -/- mice. Our genomics analysis of the CR4 -/- mice demonstrated that CTCF binding was only lost at the targeted region. This loss was associated with a significant reduction in Bcl11b expression and an overlapping set of differentially regulated genes to those found in ThymoD p(A)/p(A) and Bcl11b -/- mice. Altogether, this study establishes a requirement for CTCF binding in activating Bcl11b expression and directing T cell fate determination.

Neutrophils are characterized by a unique nuclear morphology, where the nucleus possesses a segmented multi-lobed form. During inflammation, neutrophils become activated and utilize their anti-microbial abilities. Neutrophil activation and extracellular trap formation are associated with the cell undergoing a change in nuclear morphology via delobing of the nucleus. While much work has been done concerning which parts of the activation pathway are required for NETosis, little is known about which portions of the activation pathway are involved with these morphology dynamics. Using quantitative image analysis, we established the promyelocyte HL-60 cell-line as a model system in which to study human neutrophil nuclear morphology dynamics and differentiated HL-60s as a model to investigate activation-induced nuclear morphological changes. We find that ATRA differentiated HL-60s and DMF differentiated HL-60s can be used to observe changes in the measured circularity of the nucleus induced by PMA and A23187 stimulation respectively. Using chemical inhibitors in differentiated HL-60 stimulation, it was discovered that NOX and NE activity are required for PMA-induced nuclear morphological change while MPO and PAD4 activity are not, and that PAD4 activity is required for A23187-induced nuclear morphological change. These findings demonstrate how factors that are known to contribute to NETosis also contribute to pre-NETosis neutrophil nuclear dynamics.

Enhancers harbor instructions encoded for the interactions between cis-elements and transcription factors to orchestrate lineage specific gene programs. Here we developed a modified method for chromosome conformation capture (3C), named MID Hi-C, to reveal how in mouse embryonic stem cells differential cooperation of enhancers and the chromatin remodeler BAF, as instructed by the underlying transcription factor motifs, modulate enhancer-promoter communication. We show that BAF-dependent enhancers permit genomic interactions beyond enhancer boundaries. BAF-dependent enhancers do not dictate genomic interactions within enhancer-promoter loop domains but rather act to instruct remote enhancer-promoter communication. In contrast, BAF-independent enhancers interact with promoter regions within tightly insulated enhancer-promoter loop domains that are marked by promoter and enhancer boundary elements. In addition, enhancer activeness modulated by BAF enforces compartment segregation. Based on these observations, we propose that enhancer cis elements instruct with great precision BAF-induced enhancer-promoter communication and compartmental segregation. We also characterize the gene programs and enhancer landscapes that instructed the cell fate bifurcation in the extraembryonic endoderm lineage orchestrated by lineage specific transcription factors. The simplicity of this developmental system will be valuable for dissecting how enhancer cis elements exploit the 3D genome principles to inscribe proper cell differentiation.

Recently studies have shown that in subjects with acute Kawasaki disease (KD), an acute pediatric vasculitis of the coronary artery, there exists a population of myeloid dendritic cells (mDCs) that have the ability to secrete the suppressive cytokine interleukin (IL)-10. Enumeration of the mDC and the regulatory T cell (Treg) populations from 3 cohorts of pediatric subjects: acute KD, acute pediatric febrile controls, and healthy children have been analyzed with an expanded panel of surface markers. From these cohorts of subjects, it was found that a newly described population of tolerogenic mDC is present in circulation and is defined as being CD11c+CD11b+CD14+ILT-4+CD4+ and by the expression of differential amounts of HLA-G. This newly described population of mDCs express the DCs maturation marker CD86, which is a T cell co-receptor. An in vitro assay plating with these newly found tolerogenic mDC population produced IL-10 in a dose dependent fashion when stimulated with scalar doses of purified fragments of the heavy chain constant region of immunoglobulin (Fc). These tolerogenic mDC did not exhibit properties that enable them to polarize T cells into peripherally induced regulatory T cells (iTreg), but the Fc stimulated tolerogenic mDC demonstrated the ability to suppress T helper 1 (Th1) differentiation as demonstrated by reduced levels of interferon (IFN)γ mRNA transcript in naïve T cell co-culture in vitro. In pediatric subjects these mDC express the adenosine 2A receptor (A2AR) which induces intracellular cAMP production, which in turn can cause production of IL-10 in these tolerogenic cells. Expression of CD31 on these tolerogenic mDC can offer a possibility for these cells to directly suppress inflamed tissue by being able to bind to CD38 on the inflamed endothelial cells.

This study identifies a new population of tolerogenic mDC in pediatrics who appears to compensate for the lack of a fully developed adaptive immune system in these young children.

DNA demethylation is promoted by TET enzymes, a family of three proteins -TET1, TET2 and TET3- which oxidize methyl group of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC). TET proteins have been previously associated to cancer. In particular, Tet2 loss of function is strongly associated to haematopoietic malignancies in humans, such as B cell and T cell lymphoma. In this dissertation we explore the role of TET proteins in B and Treg cells. To do this we rely in the use of in vivo and in vitro models. We generated a mouse model with Tet2/3 deletion, CD19 Tet2/3 DKO mice. The TET deficient mice showed an expansion of Germinal Center B cells and developed B cell lymphoma, as early as 6 weeks of age; coupled to this, we observed an increase of G-quaruplexes and R-loops and increase of DNA damage. Mice generated with the adiddtional deletion of DNMT1, the enzyme responsible to maintain DNA methylation and also a G-quadruplex binding protein, on top of the Tet2/3 deletion, showed an inhibition of Germinal Center B cells and a significant delay on B cell lymphoma. We also explored the role of TET proteins in Treg cells using genomic and epigenomic approaches in in vivo and in vitro models. The Tet2/3 deficient Treg cells showed a loss of Treg-specific molecular features, and an impairement of IL-2/STAT5 signaling pathway. In our in vitro system, we induced Treg (iTreg) cells with TGF and treated them with Vitamin C, a potent activator of TET proteins and other dioxygenases. Oposite to what we observed in vivo, boosting TET activity with Vitamin C enhanced IL-2 responsiveness in iTreg cells by increasing IL2R expression and STAT5 activity.

B-Cell precursor acute lymphoblastic leukemia (B-ALL) is the most common malignancy in children. Recently, we and others described a new subtype of the disease, affecting 60% of children with Down Syndrome (DS) and about 10% of patients with sporadic ALL, in which chromosomal rearrangements result in over-expression of the cytokine receptor-like factor 2 (CRLF2) receptor1-4. This over-expression is often accompanied by mutations in additional proteins in the CRLF2 pathway, such as JAK2, a downstream effector in the pathway5-9, and IL7RA, the second subunit in the TSLP receptor10. Based on mutation analyses, aberrant CRLF2 expression was thought to play a causal role in the development of B-ALL. While some data obtained in mouse systems support this assertion2,3,6, no studies have been performed in human cells to ascertain whether or not CRLF2 contributes to B-ALL pathogenesis. Due to the prominent difference between mouse and human B lymphoid development, particularly in the TSLP/IL7 pathways, it is important to study the contribution of activation of the TSLP pathway to the development of B-ALL in human cells. In the research described here, I hypothesized that aberrant expression of CRLF2 in cooperation with secondary mutations in the TSLP pathways contributes to B-ALL initiation.

This hypothesis was tested primarily by utilizing cord-blood (CB) hematopoietic-progenitors transduced with a set of lentiviral vectors carrying CRLF2 alone or in combination with JAK2 or IL7RA mutations. Outcome of forced TSLP pathway activation was cell context specific.

Expression of CRLF2 in CB hematopoietic-progenitors from a ubiquitous promoter resulted in skewed differentiation towards the myeloid lineage while transcription of the same genes from a B-cell-specific promoter accelerated B-lymphoid differentiation in vitro, underscoring the importance of expressing the genes of interest in the right cellular context for B-ALL pathogenesis.

Transduced CB cells were transplanted in NOD/LtSz-scid IL2Rγnull (NSG) mice, which are known to support human B-Cell differentiation. Transplanted cells expressing CRLF2 with mutant IL7RA exhibited population expansion, enhanced B-cell differentiation, and a significant block in differentiation at the pro-pre B-cell stage, resembling the stage of differentiation of leukemic blast cells.