The zinc finger DNA binding protein Ikaros is expressed in hematopoietic precursors, where it regulates genes involved in antigen receptor recombination, hemoglobin synthesis, and genome stability by recruiting chromatin remodeling complexes (Georgopoulos, 2002). Germline deletion of Ikzf1 in mice results in arrested erythroid and lymphoid development, leading to immunodeficiency and immature B and T cell leukemia. In conventional CD4+ and CD8+ T cells, Ikaros functions as a transcriptional repressor of inflammatory cytokine genes (Thomas et al., 2007; Bandyopadhyay et al., 2012; Thomas et al., 2010; O’Brien et al., 2014). Conventional CD4+ T cells with loss of Ikaros function are unable to differentiate into peripherally-induced regulatory T cells (iTreg) (Shin et al., 2014), and are resistant to suppression by thymic regulatory T cells (Shin et al., 2014). However, the role of Ikaros in thymic Treg development and function has not been addressed.

The Ikaros family members Helios and Eos have each been deleted or knocked down in human or mouse regulatory T cells (Getnet et al., 2010; Pan et al., 2009; Gokhale et al., 2019). Loss of Eos function in Treg is associated with increased expression of inflammatory cytokines like IL-2 and IFNg, and an inability to control pathogenic conventional T helper cell responses in an IBD model (Pan et al., 2009; Gokhale et al., 2019), although a separate study found that Eos-deficient Treg had normal function (Rieder et al., 2015). Helios also contributes to the control of Treg activation and cytokine production (Sebastian et al., 2016), but this may be secondary to its role in promoting stable expression of the Foxp3 gene (Getnet et al., 2010; Sebastian et al., 2016; Kim et al., 2015).

In this study, we conduct a genome-scale multi-omic analysis of open chromatin, active histone marks, Ikaros occupancy, Foxp3 occupancy, and gene expression in wild-type and Ikzf1-deficient regulatory T cells. We find that Ikaros plays a crucial role in establishing the normal landscape of enhancer activity, Foxp3 binding, and gene expression in Treg that cannot be filled by other Ikaros family members. Loss of Ikaros function in Treg results in uncontrolled T cell-dependent inflammatory responses in vivo.

To address the role of Ikaros in the Treg lineage, we crossed mice with a floxed Ikzf1 allele (Schwickert et al., 2014) with mice carrying a Foxp3-YFP-Cre reporter/driver. This strain generates an Ikzf1-null allele and the Cre neither generates a Foxp3 fusion protein nor affects Foxp3 function (Rubtsov et al., 2008). Male Ikzf1-fl-Foxp3-YFP-Cre mice (B6 background) do not develop overt autoimmune pathology under specific pathogen-free housing conditions over an 8-week timeframe, and basic aspects of T cell and Treg thymic development are unaltered (Figure 1—figure supplement 1a). CD4+CD25+Foxp3+peripheral Treg in these mice exhibit a nearly complete loss of Ikaros protein (Figure 1a and b) and full DNA demethylation at the Foxp3 CNS2-TSDR (Figure 1—figure supplement 1b), indicating that they are of thymic origin (Floess et al., 2007). The expression of Ikaros in conventional CD4+ T cells from Ikzf1-fl-Foxp3-YFP-Cre mice is indistinguishable from that of control Foxp3-YFP-Cre mice (Figure 1b). Ikzf1-fl-Foxp3-YFP-Cre mice exhibit a statistically significant increase in total and effector Treg pools, with a concomitant decrease in the naive Treg pool in the spleen and lymph nodes (Figure 1—figure supplement 1c). Ikzf1-deficient Treg express normal levels of Foxp3, GITR, and PD1 (Figure 1c and d), and higher levels of the high-affinity IL-2 receptor CD25 and the costimulatory receptor ICOS (Figure 1c). Ikaros-deficient Treg maintained Eos, Aiolos, and Helios protein expression, exhibiting a mild increase in the expression of Eos and Aiolos particularly in the lymph nodes (Figure 1d).

Figure 1 with 1 supplement see all

Download asset Open asset

Ikaros contributes significantly to the Treg gene expression program

Ikaros is a transcription factor, so to assess how loss of Ikaros function impacts the Treg gene expression program, we compared the transcriptomes of wild-type and Ikaros-deficient Treg isolated directly ex vivo from 6 to 8 week old aged-matched mice (n=3). A total of 661 genes were differentially expressed in Ikzf1 cko Treg, 149 of which greater than twofold (FDR <0.05, Supplementary file 1). Some of these genes were downregulated compared to wild-type Treg (Figure 1e, clusters 1, 4, and 5), but 80% were upregulated (Figure 1e, clusters 7 and 8), indicating that Ikaros functions primarily as a transcriptional repressor in Treg. Loss of Ikaros results in up-regulation of at least 12 factors that negatively regulate Treg function, e.g., multiple genes involved in Wnt signaling (Wisp1, Ctnnd1, Ctnna1), Ox40, Tlr2, Lag3, Tnf, and Ifng (Figure 2a and Supplementary file 2). Increased Ifng expression correlated with decreased DNA methylation at the Ifng locus compared to wild-type Treg (Figure 1—figure supplement 1d). Ikzf1 cko Treg also exhibited down-regulation of at least 10 factors that are required for full Treg function, e.g., the activin receptor Acvr1b, Nr4a1/Nur77, Tet1, and perforin (Figure 2b and Supplementary file 2). Bcl6, which is required for follicular Treg function (Nurieva et al., 2009), is also down-regulated in Ikaros-deficient Treg. However, Ikzf1 cko Treg up-regulated at least 24 factors known to promote Treg function (Supplementary file 2), suggesting that the loss-of-function program may be counteracted by a gain-of-function program in the absence of inflammation.

Figure 2

Download asset Open asset

To simulate antigen encounters by Treg during an immune response, we stimulated wild-type and Ikaros-deficient Treg through the TCR and CD28 in vitro. Previous studies established a role for Ikaros in restricting IL-2 and IFNg production by conventional CD4 + and CD8+ T cells (Thomas et al., 2007; Bandyopadhyay et al., 2012; Thomas et al., 2010; O’Brien et al., 2014). We find that Ikaros plays a similar role in Treg, as unlike wild-type Treg, Ikzf1 cko Treg are capable of secreting IL-2 and IFNg protein upon stimulation (Figure 3a). Treg from mice expressing a dominant-negative form of Ikaros likewise ectopically express IL-2, IFNg, and TNFa at the protein level upon stimulation (Figure 3b). Consistent with their increased expression of IL-2 and IL-2R, Ikaros-deficient Treg exhibit enhanced IL-2-induced STAT5 phosphorylation (Figure 3c) and increased proliferative capacity (Figure 3d) compared to wild-type Treg.

Figure 3

Download asset Open asset

At genome scale, stimulation through the TCR and CD28 led to differential expression of 895 genes (FDR <0.05, Supplementary file 1), 533 down-regulated (Figure 1e, clusters 2, 3, and 5), and 362 up-regulated in Ikzf1 cko Treg compared to wild-type Treg (Figure 1e, clusters 6, 7, and 8). Consistent with its role in antigen receptor selection during lymphocyte development (Georgopoulos et al., 1994), gene ontology analysis (Supplementary file 3) of genes down-regulated in Ikaros-deficient Treg shows enrichment for immune cell development and TCR/VDJ recombination and diversification (Figure 3e). Genes upregulated in Ikaros-deficient Treg are enriched for networks involved in interferon and cytokine production and responses (Figure 3e). The set of upregulated genes includes at least 20 factors known to promote Treg function like Foxp3, Il2ra, Icos, Ezh2, and Gpr83 (Supplementary file 2). However, loss of Ikaros results in up-regulation of at least 11 factors that negatively regulate Treg function like Irf1, Il12rb2 (IL-12 receptor), Il3, and several genes in the Notch and Wnt pathways like Notch2, Maml3, Rbpj, Wisp, and Ctnnd1 (Figure 2c), and down-regulation of at least 14 factors that are required for full Treg function like Tcf7, Satb1, Foxp1, Id3, Smad3, Il27ra, Tlr7, and follicular Treg (Tfreg) genes like Bcl6, Cxcr5, and S1pr1 (Figure 2d).

The Wnt pathway is normally repressed in Treg, and ectopic Wnt signaling in Treg has been associated with ectopic IFNg production and reduced suppressive function (Sumida et al., 2018). We show that the Wnt signaling factor B-catenin is elevated in Ikzf1 cko Treg (Figure 4a), and that ectopic IFNg production by Ikaros-deficient Treg is Wnt-dependent, while IFNg production by conventional CD4+ T cells does not depend on Wnt (Figure 4b and c). Together, these transcriptomic analyses indicate that Ikaros normally regulates a large proportion of the core Treg program (Shevyrev and Tereshchenko, 2019; Figure 3f, genes in red), and is required to restrain Wnt, Notch, and inflammatory cytokine gene expression in the Treg lineage. The impact of the loss of Ikaros function on the Treg transcriptome likely stem from both direct, cell-intrinsic effects and from indirect, cell-extrinsic effects.

Figure 4

Download asset Open asset

Ikaros is required for establishing the Treg open chromatin and enhancer landscape

To gain mechanistic insight into how Ikaros regulates the Treg gene expression program, we measured Ikaros binding, Foxp3 binding, chromatin accessibility, and H3K27ac enhancer marks in wild-type and Ikaros-deficient Treg using ATAC-seq and ChIP-seq. Loss of Ikaros function induced remodeling of 1431 genomic regions (n=3, FDR <0.05, Figure 5—figure supplements 1 and 2 and Supplementary file 4), one-third of which (513) exhibiting reduced accessibility and two-thirds (918) of which become more accessible (Figure 5a). Regions with reduced accessibility in Ikaros-deficient Treg were enriched for nearby genes involved in leukocyte development and differentiation, while regions showing increased accessibility were enriched for genes involved in cytokine signaling and response to interferon-gamma (Figure 5—figure supplement 1d). At genome scale, increased accessibility at genomic elements after deletion of Ikaros correlated significantly with increased expression of nearby genes, while decreased accessibility correlated significantly with reduced gene expression (Figure 5—figure supplement 1e).

Figure 5 with 3 supplements see all

Download asset Open asset

To explore how Ikaros regulates the Treg enhancer landscape, we measured histone acetylation at nucleosomes flanking open chromatin regions in wild-type and Ikzf1 cko Treg (n=3, Figure 5—figure supplement 2a and b). Out of approximately 21,000 H3K27ac peaks called in both cell populations, 40% were affected by the loss of Ikaros function (Supplementary file 5), with 323 regions showing a >two-fold reduction acetylation and 1206 regions exhibiting a >two-fold increase in H3K27ac (Figure 5b). Differential analysis also showed that Ikaros controls histone acetylation at ~25% of Treg open chromatin regions (OCR, 9937 out of ~40,000, FDR <0.05), with the vast majority (78%, 7788) showing increased acetylation upon loss of Ikaros function (Supplementary file 5 and Figure 5—figure supplement 2c). At genome scale, enrichment of the H3K27ac mark correlates with regional accessibility (Figure 5—figure supplement 2d), and strength of the enhancer signature at a given element correlates with the level of nearby gene expression (Figure 5—figure supplement 2d). Dense collections of multiple enhancers that tend to drive expression of genes involved in cell identity are called ‘super-enhancers.’ We defined 1700 Treg super-enhancers based on H3K27ac density (Supplementary file 6), 20% of which (324) are regulated by Ikaros (Figure 5—figure supplement 2e).

Ikaros ChIP-seq analysis identified 7642 Ikaros binding sites in WT Tconv, 83% of which (6361) are located in open chromatin, and 7061 Ikaros binding sites in WT Treg, 76% of which (5477) are located in open chromatin (Figure 5c and Supplementary file 7). Of all accessible Ikaros binding sites detected, 39% (3567) were Tconv-specific, 31% (2794) were shared between Tconv and Treg, and 29% (2683) were only detected in Treg (Figure 5d). The Ikaros ChIP-seq signal is enriched at accessible Ikaros motifs in the Treg and Tconv genome (Figure 5e), and the set of genes with accessible Ikaros binding motifs showed increased expression in Ikaros-deficient compared to wild-type Treg (Figure 5f). Motif analysis (Supplementary file 8) at regions that exhibit increased accessibility in Ikzf1 cko Treg shows enrichment of the Ikaros GGGAA core binding sequence that is shared with immune trans-activators like NFkB, NFAT, Notch, and Stat1/4, and these regions are also enriched for AP-1 (Fos/Jun) and T-bet (Tbx21) motifs (Figure 5g inset). This suggests that Ikaros can directly repress inflammatory gene expression in Treg by competing with NFkB, NFAT, Notch, and Stats for binding to enhancers and recruiting epigenetic factors that silence these elements (Molnár and Georgopoulos, 1994; Trinh et al., 2001; Kleinmann et al., 2008; Katerndahl et al., 2017; Heizmann et al., 2020). Loci under direct repressive control of Ikaros in Treg include Bcl6, Notch2, Irf4, and Ifng (Figure 5h and Figure 5—figure supplement 3a–c). However, the majority of genomic regions exhibiting increased accessibility in Ikzf1 cko Treg are not bound by Ikaros in wild-type cells (873 of 918, Figure 5—figure supplement 3d), suggesting that indirect gene regulation due to observed alterations in the expression of other transcription factors is another mechanism by which Ikaros establishes the Treg gene expression program.

Foxp3 cooperates with Ikaros for DNA binding across the Treg genome

Regions that exhibit reduced accessibility in Ikzf1 cko compared to wild-type Treg are enriched for the binding motif for Ikaros and the motif for TCF1 (Figure 5g), a factor that cooperates with Foxp3 to enforce Treg function (Xing et al., 2019) and is down-regulated in Ikzf1 cko Treg (Figure 3f). These regions are likewise enriched for the GTAAACA Foxp3/forkhead motif (Figure 5i inset), suggesting that Foxp3 may cooperate with Ikaros at these sites. To test this, we compared Foxp3 genome occupancy in wild-type vs. Ikzf1 cko Treg by ChIP-seq (Figure 6a, Figure 6—figure supplement 1a–c). A total of 4423 Foxp3 binding sites were detected in the open chromatin landscape of wild-type Treg (Supplementary file 9), and this ChIP-seq signal was enriched at accessible Foxp3 motifs. Consistent with the motif analyses (Figure 5h), we find a remarkable 74% of all Foxp3 binding sites are co-bound by Ikaros (3255 of 4423 sites, Figure 6b). Loss of Ikaros in Ikzf1 cko Treg results in reduced Foxp3 binding affinity at 70% of Ikaros-Foxp3 co-bound sites (2254 of 3256, Figure 6b inset), and reduced Foxp3 binding at 80% of all sites strongly bound by Foxp3 in wild-type Treg (3543 of 4422, Figure 6a and c). As a result, the set of all direct target genes with accessible Foxp3 binding motifs showed significantly increased expression in Ikaros-deficient compared to wild-type Treg (Figure 6d).

Figure 6 with 1 supplement see all

Download asset Open asset

Figure 6—source data 1

Immunoblot analysis of Foxp3-Ikaros co-precipitation in transfected HEK293T cells.

https://cdn.elifesciences.org/articles/91392/elife-91392-fig6-data1-v1.pdf

Download elife-91392-fig6-data1-v1.pdf

Consistent with these observations, we find that Ikaros and Foxp3 exist in a complex in the nuclei of cells (Figure 6e), confirming prior proteomic data suggesting a physical interaction between Foxp3 and Ikaros in transfected cells (Rudra et al., 2012). At the Foxp3 and Ikaros co-bound target genes Il2 and Ifng, retroviral expression of Foxp3 in CD4 + T cells results in direct promoter occupancy and silencing of both genes (Figure 6f). Co-expression of dominant-negative Ikaros (IkDN or Ik7) abrogates the binding of Foxp3 to the Il2 promoter and inhibits the repressive activity of Foxp3 (Figure 6f). At genome scale, a significant number of genes in addition to Il2 are concordantly regulated by both dominant-negative Ikaros and Ikzf1 gene deletion (Figure 6g). In addition to the promoter, we also observe Ikaros-Foxp3 co-binding at a defined distal enhancer of Il2 located 83 kb upstream of the promoter (Mehra and Wells, 2015) in wild-type Treg (Figure 6h). Loss of Ikaros function in Ikzf1 cko Treg results in loss of Foxp3 binding at this enhancer, which is accompanied by increased histone acetylation, chromatin accessibility (Figure 6h), and Il2 expression (Figures 1 and 2). Other examples of Foxp3-regulated genes that exhibit Ikaros-dependent Foxp3 binding, enhancer activity, and expression are Tcf7 (Figure 6i), Il2ra, Rbpj, and Maml3 (Figure 6—figure supplement 1d and e). Together, these results indicate that a large portion of the Treg epigenome and transcriptome, including two-thirds of the core Treg program and the majority of the Foxp3 gene regulatory program, is dependent on Ikaros.

Ikaros is required for Treg control of conventional T cell differentiation

The large-scale dysregulation of gene expression in Ikzf1-cko Treg, especially upon stimulation, suggests that extrinsic control of inflammatory immune responses may be dysregulated in mice lacking Ikaros in the Treg lineage. We observed no clear signs of frank autoimmunity in aged (1-year-old) Ikzf1-fl-Foxp3-YFP-Cre mice (Figure 7—figure supplement 1a). The conventional T cell pool in 6–8 week-old (Figure 7) and 10-month-old (Figure 7—figure supplement 1) Ikzf1-fl-Foxp3-YFP-Cre mice showed a statistically significant accumulation of CD4+ T cells (Figure 7a, Figure 7—figure supplement 1b) with a memory phenotype (Figure 7b and c), and a concomitant reduction in naive phenotype CD4+ T cells (Figure 7d, Figure 7—figure supplement 1b). Ikaros-deficient Treg maintained comparable suppressive activity against wild-type Tconv in vitro (Figure 7e), however, Ikaros-sufficient, conventional CD4+ T cells from mice lacking Ikaros in the Treg lineage were resistant to suppression by both Ikzf1-cko and wild-type Treg (Figure 7f and g), likely due increased frequency of suppression-resistant memory cells (Yang et al., 2007; Afzali, 2011). At 6 months, Ikzf1-fl-Foxp3-YFP-Cre mice also exhibited increased frequencies of follicular helper CD4 + T cells in the lymph nodes and spleen (Figure 7h), which was associated with elevated levels of total IgM, IgG, and especially IgA in the serum (Figure 7i). The elevated IgA was accompanied by increased frequencies of IgA-positive B cells in the spleen and mesenteric lymph nodes (Figure 7j and k). These results suggest perturbed immune homeostasis in mice that lack Ikaros function in Treg owing to a defect in the control of conventional CD4 + T cell differentiation.

Figure 7 with 1 supplement see all

Download asset Open asset

Ikaros is required for Treg control of pathogenic T cell-mediated mucosal inflammation and acquired immune tolerance

To address this, we tested the ability of Ikaros-deficient Treg to control inflammatory colitis in an in vivo adoptive transfer model of IBD. Rag-deficient mice that received conventional CD4+CD25-negative T cells alone (n=5) developed severe disease as evidenced by progressive weight loss (Figure 8a), gross and histological intestinal pathology (Figure 8b and Figure 8—figure supplement 1a), extensive cellular infiltration, and tissue damage in the inner mucosal epithelial layer of the colon (Figure 8c), and high numbers of activated Tconv in the colon (Figure 8c), spleen and mesenteric lymph nodes of these animals (Figure 8—figure supplement 1d–g). Upon co-transfer into Rag-deficient mice (n=5), wild-type Treg accumulated in the mesenteric lymph nodes (Figure 8—figure supplement 1i) and colon (Figure 8d), were able to control the activation and expansion of conventional helper T cells in lymphoid tissues (Figure 8—figure supplement 1e and g), and limit CD4+ T cell infiltration into the intestinal epithelium (Figure 8d). Recipients of wild-type Treg exhibited only mild intestinal pathology at the gross and microscopic levels (Figure 8b and d and Figure 8—figure supplement 1b), and lost little weight over the course of the experiment (Figure 8a). Co-transferred Ikaros-deficient Treg accumulated in the spleen (Figure 8—figure supplement 1h and i) and, despite reduced expression of the alpha4-beta7 integrin (Itga4, Itgb7) and Ccr9 genes involved in homing to the intestine, accumulated in the intestinal epithelium to numbers 5–10-fold higher than in the recipients of wild-type Treg (Figure 8e). Despite the presence of large numbers in the intestinal mucosa, Ikaros-deficient Treg were completely unable to protect RAG-deficient mice (n=5) from infiltration and colitis mediated by wild-type conventional T cells at the level of weight loss (Figure 8a) and intestinal pathology (Figure 8b and e and Figure 8—figure supplement 1c). Similar to mice that received no Treg, recipients of Ikaros-deficient Treg exhibited extensive inflammatory infiltrates in the colon, with thickening and detachment of the epithelial layer from underlying tissues (Figure 8e). Cell necrosis and mononuclear lymphocyte infiltration were also more pronounced in recipients of Ikaros-deficient Treg. These results indicate that Treg depend on Ikaros to control mucosal inflammation during a conventional T cell response.

Figure 8 with 1 supplement see all

Download asset Open asset

Regulatory T cells are required for the induction of peripheral alloimmune tolerance. To determine whether Treg-intrinsic Ikaros function is required for acquired tolerance to organ transplants, we transplanted fully mismatched cardiac allografts into Ikzf1-fl-Foxp3-YFP-Cre or control Foxp3-YFP-Cre mice under combined blockade of the CD28 and CD40 costimulatory pathways (n=5 per group). While costimulatory blockade induced long-term allograft tolerance in wild-type recipients, this treatment failed to induce tolerance in mice lacking Ikaros in the Treg lineage (Figure 9a). Similar results were obtained when anti-CD40L plus donor-specific transfusion was used as a tolerizing regimen (Figure 8—figure supplement 1j and k). Intragraft analysis of gene expression showed elevated levels of multiple Th1-related transcripts in rejecting grafts from Ikzf1-Treg-cko recipients, despite elevation of Foxp3 (Figure 9b). Histopathological analysis of cardiac graft tissue from Ikzf1-Treg-cko recipients showed extensive myocardial necrosis (Figure 9c) associated with increased CD4 + T cell infiltration despite numbers of Foxp3+ Treg comparable to that in tolerant recipients (Figure 9d). These results reveal an important role for Ikaros in the ability of Treg to control inflammation and establish acquired immune tolerance.

Figure 9

Download asset Open asset

The studies reported here establish a crucial role for Ikaros in regulatory T cells that cannot be replaced by the Ikaros family members Helios, Eos, or Aiolos. While Ikaros is required for induction of Foxp3 by TGF-B in conventional T cells (Heller et al., 2014), thymic-derived Ikaros-deficient Treg exhibited normal Foxp3 expression. Instead, loss of Ikaros activity results in significant dysregulation of the Treg gene expression program, including pro-inflammatory cytokine and chemokine genes normally not expressed by Treg (e.g. Ifng, Tnf, Il3, Il12rb, Tlr2), and genes required for normal Treg function (e.g. Tcf7, Lef1, Satb1, Nr4a1). Many of these dysregulated genes are Foxp3 targets, and we show that Foxp3 cooperates with Ikaros to occupy the majority of Foxp3 binding sites in Treg.

Mice with Treg lineage-specific loss of Ikaros occupancy showed an accumulation of activated regulatory and helper T cells in the secondary lymphoid tissues, but no evidence for increased T cell infiltration into organs or frank autoimmunity. This might be explained by the fact that Treg isolated directly ex vivo showed elevated expression of many genes that promote Treg function, potentially balancing the dysregulated program predicted to have deleterious effects on Treg homeostasis and function. However, upon TCR stimulation, Ikzf1-deficient Treg induce a set of inflammatory Th1, Notch, and Wnt pathway genes normally repressed in Treg, and fail to control in vivo cellular and humoral immune responses mediated by conventional T cells. Importantly, our results show that the loss of suppressive function in vivo is not due to failure of Treg to home to sites of tissue inflammation. The lack of spontaneous inflammation or autoimmunity in Ikzf1-fl-Foxp3-Cre mice is similar to mice with Treg-specific deletion of Prdm1 (Bankoti et al., 2017; Cretney et al., 2018), Icos (Guo et al., 2008), Il10 (Rubtsov et al., 2008) and Mef2d (Di Giorgio et al., 2020). Mice lacking Il10 in Treg exhibit mild colitis, but no autoimmunity, while mice with deletion of Prdm1 in Treg show signs of autoimmunity only in aged mice. Moreover, deletion of Mef2d, Blimp1, Icos, or Il10 in Treg does not impact their suppressive activity in vitro, but impairs their function in vivo in the context of inflammation.

Although the literature is not in complete agreement, Eos and Helios are considered to be necessary for Treg function through their contribution to the Treg gene expression program. Eos, like Ikaros, is required to repress inflammatory gene expression by Treg (Pan et al., 2009; Gokhale et al., 2019), and Eos can cooperate with Foxp3 to strengthen a core Treg gene expression program when ectopically co-expressed in conventional T cells (Fu et al., 2012). Loss of Helios function in Treg has the primary effect of destabilizing Foxp3 expression (Getnet et al., 2010; Sebastian et al., 2016; Kim et al., 2015), and also contributes to core Treg gene expression when ectopically co-expressed with Foxp3 (Fu et al., 2012). Despite these functions, neither Eos, Helios, nor Aiolos were able to compensate for the loss of Ikaros, despite the fact that Ikzf1-deficient Treg express all these proteins at comparable or higher levels as compared to wild-type Treg. The availability of Helios ChIP-seq (Kim et al., 2015) and Helios-dependent gene expression (Yates et al., 2018) data allowed us to compare Ikaros and Helios genome occupancy and gene regulatory programs (Figure 10). Of the 1838 Helios binding sites and 5477 Ikaros binding sites detected in Treg open chromatin, 64 are shared, representing only ~3.5% of Helios binding sites and ~1% of Ikaros binding sites. Similarly, we found a statistically significant overlap between the Ikaros-dependent set of 660 genes and the Helios-dependent set of 147 genes, but this consisted of only nine genes (Figure 10). For five of these genes, Ikaros and Helios have the same effect on expression, while Ikaros and Helios have the opposite effect on expression of the other four genes. This level of discordance at the level of both genome occupancy and gene regulation likely explains whey Helios, for example, cannot compensate for the loss of Ikaros in Treg.

Figure 10

Download asset Open asset

Coding mutations in IKZF1 in humans are a cause of common variable immune deficiency (CVID) and autoimmunity (Boast et al., 2021). To date, studies have focused on the impact of these mutations on T and B lymphocyte function, and our results here suggest Treg defects could also contribute to the immune dysregulation observed in these patients. In addition, common genetic polymorphism at the IKZF1 locus have been associated with SLE susceptibility by GWAS (Cunninghame Graham et al., 2011), and one mechanism for this is an SLE-associated distal regulatory element required for normal expression of Ikaros in human T cells (Su et al., 2020). Given the important roles for Ikaros in conventional B and T cell function, and its role newly defined here in regulatory T cell function, Ikaros is a relevant target for novel therapies for autoimmunity, organ transplant rejection, and cancer.

All data supporting the manuscript are contained in the supplementary files. Sequencing data have been deposited in GEO under accession codes GSE200176, GSE200177, GSE200178 and GSE200179.

1. 1. Thomas RM
   2. Pahl MC
   3. Wang L
   4. Grant SFA
   5. Hancock WW

   (2024) NCBI Gene Expression Omnibus

   ID GSE200179. Foxp3 depends on Ikaros for control of regulatory T cell gene expression and function.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE200179
2. 1. Pahl MC
   2. Thomas R
   3. Wells AD

   (2024) NCBI Gene Expression Omnibus

   ID GSE200178. Foxp3 depends on Ikaros for control of regulatory T cell gene expression and function [RNA-Seq].

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE200178
3. 1. Pahl MC
   2. Thomas RM
   3. Wells AD

   (2024) NCBI Gene Expression Omnibus

   ID GSE200176. Foxp3 depends on Ikaros for control of regulatory T cell gene expression and function [ATAC-Seq].

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE200176
4. 1. Pahl MC
   2. Thomas R
   3. Wells AD

   (2024) NCBI Gene Expression Omnibus

   ID GSE200177. Foxp3 depends on Ikaros for control of regulatory T cell gene expression and function [ChIP-Seq].

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE200177

1. 1. Afzali B

   (2011) Relative resistance of human cd4+ memory t cells to suppression by cd4+cd25+ regulatory t cells: memory cells as barriers to treg-cell therapy

   Am J Transplant 11:1734–1742.

   https://doi.org/10.1111/j.1600-6143.2011.03635.x

   • Google Scholar
2. 1. Anders S
   2. Pyl PT
   3. Huber W

   (2015) Htseq--a python framework to work with high-throughput sequencing data

   Bioinformatics 31:166–169.

   https://doi.org/10.1093/bioinformatics/btu638

   • PubMed
   • Google Scholar
3. 1. Bandyopadhyay S
   2. Montagna C
   3. Macian F

   (2012) Silencing of the Il2 gene transcription is regulated by epigenetic changes in anergic T cells

   European Journal of Immunology 42:2471–2483.

   https://doi.org/10.1002/eji.201142307

   • PubMed
   • Google Scholar
4. 1. Bankoti R
   2. Ogawa C
   3. Nguyen T
   4. Emadi L
   5. Couse M
   6. Salehi S
   7. Fan X
   8. Dhall D
   9. Wang Y
   10. Brown J
   11. Funari V
   12. Tang J
   13. Martins GA

   (2017) Differential regulation of effector and regulatory t cell function by blimp1

   Scientific Reports 7:12078.

   https://doi.org/10.1038/s41598-017-12171-3

   • PubMed
   • Google Scholar
5. 1. Boast B
   2. de Nunes-Santos CJ
   3. Kuehn HS
   4. Rosenzweig SD

   (2021) Ikaros-associated diseases: from mice to humans and back again

   Frontiers in Pediatrics 9:705497.

   https://doi.org/10.3389/fped.2021.705497

   • PubMed
   • Google Scholar
6. 1. Chen C
   2. Rowell EA
   3. Thomas RM
   4. Hancock WW
   5. Wells AD

   (2006) Transcriptional regulation by Foxp3 is associated with direct promoter occupancy and modulation of histone acetylation

   The Journal of Biological Chemistry 281:36828–36834.

   https://doi.org/10.1074/jbc.M608848200

   • PubMed
   • Google Scholar
7. 1. Cretney E
   2. Leung PS
   3. Trezise S
   4. Newman DM
   5. Rankin LC
   6. Teh CE
   7. Putoczki TL
   8. Gray DH
   9. Belz GT
   10. Mielke LA
   11. Dias S
   12. Nutt SL

   (2018) Characterization of Blimp-1 function in effector regulatory T cells

   Journal of Autoimmunity 91:73–82.

   https://doi.org/10.1016/j.jaut.2018.04.003

   • PubMed
   • Google Scholar
8. 1. Cunninghame Graham DS
   2. Morris DL
   3. Bhangale TR
   4. Criswell LA
   5. Syvänen AC
   6. Rönnblom L
   7. Behrens TW
   8. Graham RR
   9. Vyse TJ

   (2011) Association of NCF2, IKZF1, IRF8, IFIH1, and TYK2 with systemic lupus erythematosus

   PLOS Genetics 7:e1002341.

   https://doi.org/10.1371/journal.pgen.1002341

   • PubMed
   • Google Scholar
9. 1. Di Giorgio E
   2. Wang L
   3. Xiong Y
   4. Akimova T
   5. Christensen LM
   6. Han R
   7. Samanta A
   8. Trevisanut M
   9. Bhatti TR
   10. Beier UH
   11. Hancock WW

   (2020) MEF2D sustains activation of effector Foxp3+ Tregs during transplant survival and anticancer immunity

   The Journal of Clinical Investigation 130:6242–6260.

   https://doi.org/10.1172/JCI135486

   • PubMed
   • Google Scholar
10. 1. Dobin A
    2. Davis CA
    3. Schlesinger F
    4. Drenkow J
    5. Zaleski C
    6. Jha S
    7. Batut P
    8. Chaisson M
    9. Gingeras TR

    (2013) STAR: ultrafast universal RNA-seq aligner

    Bioinformatics 29:15–21.

    https://doi.org/10.1093/bioinformatics/bts635

    • PubMed
    • Google Scholar
11. 1. Floess S
    2. Freyer J
    3. Siewert C
    4. Baron U
    5. Olek S
    6. Polansky J
    7. Schlawe K
    8. Chang HD
    9. Bopp T
    10. Schmitt E
    11. Klein-Hessling S
    12. Serfling E
    13. Hamann A
    14. Huehn J

    (2007) Epigenetic control of the foxp3 locus in regulatory T cells

    PLOS Biology 5:e38.

    https://doi.org/10.1371/journal.pbio.0050038

    • PubMed
    • Google Scholar
12. 1. Fornes O
    2. Castro-Mondragon JA
    3. Khan A
    4. van der Lee R
    5. Zhang X
    6. Richmond PA
    7. Modi BP
    8. Correard S
    9. Gheorghe M
    10. Baranašić D
    11. Santana-Garcia W
    12. Tan G
    13. Chèneby J
    14. Ballester B
    15. Parcy F
    16. Sandelin A
    17. Lenhard B
    18. Wasserman WW
    19. Mathelier A

    (2020) JASPAR 2020: update of the open-access database of transcription factor binding profiles

    Nucleic Acids Research 48:D87–D92.

    https://doi.org/10.1093/nar/gkz1001

    • PubMed
    • Google Scholar
13. 1. Fu W
    2. Ergun A
    3. Lu T
    4. Hill JA
    5. Haxhinasto S
    6. Fassett MS
    7. Gazit R
    8. Adoro S
    9. Glimcher L
    10. Chan S
    11. Kastner P
    12. Rossi D
    13. Collins JJ
    14. Mathis D
    15. Benoist C

    (2012) A multiply redundant genetic switch “locks in” the transcriptional signature of regulatory T cells

    Nature Immunology 13:972–980.

    https://doi.org/10.1038/ni.2420

    • PubMed
    • Google Scholar
14. 1. Georgopoulos K
    2. Bigby M
    3. Wang JH
    4. Molnar A
    5. Wu P
    6. Winandy S
    7. Sharpe A

    (1994) The Ikaros gene is required for the development of all lymphoid lineages

    Cell 79:143–156.

    https://doi.org/10.1016/0092-8674(94)90407-3

    • PubMed
    • Google Scholar
15. 1. Georgopoulos K

    (2002) Haematopoietic cell-fate decisions, chromatin regulation and ikaros

    Nature Reviews. Immunology 2:162–174.

    https://doi.org/10.1038/nri747

    • PubMed
    • Google Scholar
16. 1. Getnet D
    2. Grosso JF
    3. Goldberg MV
    4. Harris TJ
    5. Yen HR
    6. Bruno TC
    7. Durham NM
    8. Hipkiss EL
    9. Pyle KJ
    10. Wada S
    11. Pan F
    12. Pardoll DM
    13. Drake CG

    (2010) A role for the transcription factor helios in human cd4(+)cd25(+) regulatory t cells

    Molecular Immunology 47:1595–1600.

    https://doi.org/10.1016/j.molimm.2010.02.001

    • PubMed
    • Google Scholar
17. 1. Gokhale AS
    2. Gangaplara A
    3. Lopez-Occasio M
    4. Thornton AM
    5. Shevach EM

    (2019) Selective deletion of Eos (Ikzf4) in T-regulatory cells leads to loss of suppressive function and development of systemic autoimmunity

    Journal of Autoimmunity 105:102300.

    https://doi.org/10.1016/j.jaut.2019.06.011

    • PubMed
    • Google Scholar
18. 1. Guo F
    2. Iclozan C
    3. Suh WK
    4. Anasetti C
    5. Yu XZ

    (2008) CD28 controls differentiation of regulatory T cells from naive CD4 T cells

    Journal of Immunology 181:2285–2291.

    https://doi.org/10.4049/jimmunol.181.4.2285

    • PubMed
    • Google Scholar
19. 1. Heizmann B
    2. Le Gras S
    3. Simand C
    4. Marchal P
    5. Chan S
    6. Kastner P

    (2020) Ikaros antagonizes DNA binding by STAT5 in pre-B cells

    PLOS ONE 15:e0242211.

    https://doi.org/10.1371/journal.pone.0242211

    • PubMed
    • Google Scholar
20. 1. Heller JJ
    2. Schjerven H
    3. Li S
    4. Lee A
    5. Qiu J
    6. Chen ZME
    7. Smale ST
    8. Zhou L

    (2014) Restriction of IL-22-producing T cell responses and differential regulation of regulatory T cell compartments by zinc finger transcription factor Ikaros

    Journal of Immunology 193:3934–3946.

    https://doi.org/10.4049/jimmunol.1401234

    • PubMed
    • Google Scholar
21. 1. Katerndahl CDS
    2. Heltemes-Harris LM
    3. Willette MJL
    4. Henzler CM
    5. Frietze S
    6. Yang R
    7. Schjerven H
    8. Silverstein KAT
    9. Ramsey LB
    10. Hubbard G
    11. Wells AD
    12. Kuiper RP
    13. Scheijen B
    14. van Leeuwen FN
    15. Müschen M
    16. Kornblau SM
    17. Farrar MA

    (2017) Antagonism of B cell enhancer networks by STAT5 drives leukemia and poor patient survival

    Nature Immunology 18:694–704.

    https://doi.org/10.1038/ni.3716

    • PubMed
    • Google Scholar
22. 1. Kharchenko PV
    2. Tolstorukov MY
    3. Park PJ

    (2008) Design and analysis of ChIP-seq experiments for DNA-binding proteins

    Nature Biotechnology 26:1351–1359.

    https://doi.org/10.1038/nbt.1508

    • PubMed
    • Google Scholar
23. 1. Kim HJ
    2. Barnitz RA
    3. Kreslavsky T
    4. Brown FD
    5. Moffett H
    6. Lemieux ME
    7. Kaygusuz Y
    8. Meissner T
    9. Holderried TAW
    10. Chan S
    11. Kastner P
    12. Haining WN
    13. Cantor H

    (2015) Stable inhibitory activity of regulatory T cells requires the transcription factor Helios

    Science 350:334–339.

    https://doi.org/10.1126/science.aad0616

    • PubMed
    • Google Scholar
24. 1. Kleinmann E
    2. Geimer Le Lay AS
    3. Sellars M
    4. Kastner P
    5. Chan S

    (2008) Ikaros represses the transcriptional response to notch signaling in t-cell development

    Molecular and Cellular Biology 28:7465–7475.

    https://doi.org/10.1128/MCB.00715-08

    • PubMed
    • Google Scholar
25. 1. Mehra P
    2. Wells AD

    (2015) Long-range transcriptional control of the il2 gene by an intergenic enhancer

    Molecular and Cellular Biology 35:3880–3891.

    https://doi.org/10.1128/MCB.00592-15

    • PubMed
    • Google Scholar
26. 1. Molnár A
    2. Georgopoulos K

    (1994) The Ikaros gene encodes a family of functionally diverse zinc finger DNA-binding proteins

    Molecular and Cellular Biology 14:8292–8303.

    https://doi.org/10.1128/mcb.14.12.8292-8303.1994

    • PubMed
    • Google Scholar
27. 1. Northrop JK
    2. Thomas RM
    3. Wells AD
    4. Shen H

    (2006) Epigenetic remodeling of the il-2 and ifn-gamma loci in memory cd8 t cells is influenced by cd4 t cells

    Journal of Immunology 177:1062–1069.

    https://doi.org/10.4049/jimmunol.177.2.1062

    • PubMed
    • Google Scholar
28. 1. Nurieva RI
    2. Chung Y
    3. Martinez GJ
    4. Yang XO
    5. Tanaka S
    6. Matskevitch TD
    7. Wang YH
    8. Dong C

    (2009) Bcl6 mediates the development of T follicular helper cells

    Science 325:1001–1005.

    https://doi.org/10.1126/science.1176676

    • PubMed
    • Google Scholar
29. 1. O’Brien S
    2. Thomas RM
    3. Wertheim GB
    4. Zhang F
    5. Shen H
    6. Wells AD

    (2014) Ikaros imposes a barrier to CD8+ T cell differentiation by restricting autocrine IL-2 production

    Journal of Immunology 192:5118–5129.

    https://doi.org/10.4049/jimmunol.1301992

    • PubMed
    • Google Scholar
30. 1. Pan F
    2. Yu H
    3. Dang EV
    4. Barbi J
    5. Pan X
    6. Grosso JF
    7. Jinasena D
    8. Sharma SM
    9. McCadden EM
    10. Getnet D
    11. Drake CG
    12. Liu JO
    13. Ostrowski MC
    14. Pardoll DM

    (2009) Eos mediates Foxp3-dependent gene silencing in CD4+ regulatory T cells

    Science 325:1142–1146.

    https://doi.org/10.1126/science.1176077

    • PubMed
    • Google Scholar
31. 1. Rieder SA
    2. Metidji A
    3. Glass DD
    4. Thornton AM
    5. Ikeda T
    6. Morgan BA
    7. Shevach EM

    (2015) Eos is redundant for regulatory t cell function but plays an important role in il-2 and th17 production by cd4+ conventional t cells

    Journal of Immunology 195:553–563.

    https://doi.org/10.4049/jimmunol.1500627

    • PubMed
    • Google Scholar
32. 1. Robinson MD
    2. McCarthy DJ
    3. Smyth GK

    (2010) edgeR: a Bioconductor package for differential expression analysis of digital gene expression data

    Bioinformatics 26:139–140.

    https://doi.org/10.1093/bioinformatics/btp616

    • PubMed
    • Google Scholar
33. 1. Rubtsov YP
    2. Rasmussen JP
    3. Chi EY
    4. Fontenot J
    5. Castelli L
    6. Ye X
    7. Treuting P
    8. Siewe L
    9. Roers A
    10. Henderson WR
    11. Muller W
    12. Rudensky AY

    (2008) Regulatory T cell-derived interleukin-10 limits inflammation at environmental interfaces

    Immunity 28:546–558.

    https://doi.org/10.1016/j.immuni.2008.02.017

    • PubMed
    • Google Scholar
34. 1. Rudra D
    2. deRoos P
    3. Chaudhry A
    4. Niec RE
    5. Arvey A
    6. Samstein RM
    7. Leslie C
    8. Shaffer SA
    9. Goodlett DR
    10. Rudensky AY

    (2012) Transcription factor Foxp3 and its protein partners form a complex regulatory network

    Nature Immunology 13:1010–1019.

    https://doi.org/10.1038/ni.2402

    • PubMed
    • Google Scholar
35. 1. Schwickert TA
    2. Tagoh H
    3. Gültekin S
    4. Dakic A
    5. Axelsson E
    6. Minnich M
    7. Ebert A
    8. Werner B
    9. Roth M
    10. Cimmino L
    11. Dickins RA
    12. Zuber J
    13. Jaritz M
    14. Busslinger M

    (2014) Stage-specific control of early B cell development by the transcription factor Ikaros

    Nature Immunology 15:283–293.

    https://doi.org/10.1038/ni.2828

    • PubMed
    • Google Scholar
36. 1. Sebastian M
    2. Lopez-Ocasio M
    3. Metidji A
    4. Rieder SA
    5. Shevach EM
    6. Thornton AM

    (2016) Helios controls a limited subset of regulatory t cell functions

    Journal of Immunology 196:144–155.

    https://doi.org/10.4049/jimmunol.1501704

    • PubMed
    • Google Scholar
37. 1. Shevyrev D
    2. Tereshchenko V

    (2019) Treg heterogeneity, function, and homeostasis

    Frontiers in Immunology 10:3100.

    https://doi.org/10.3389/fimmu.2019.03100

    • PubMed
    • Google Scholar
38. 1. Shin DS
    2. Jordan A
    3. Basu S
    4. Thomas RM
    5. Bandyopadhyay S
    6. de Zoeten EF
    7. Wells AD
    8. Macian F

    (2014) Regulatory T cells suppress CD4+ T cells through NFAT-dependent transcriptional mechanisms

    EMBO Reports 15:991–999.

    https://doi.org/10.15252/embr.201338233

    • PubMed
    • Google Scholar
39. 1. Su C
    2. Johnson ME
    3. Torres A
    4. Thomas RM
    5. Manduchi E
    6. Sharma P
    7. Mehra P
    8. Le Coz C
    9. Leonard ME
    10. Lu S
    11. Hodge KM
    12. Chesi A
    13. Pippin J
    14. Romberg N
    15. Grant SFA
    16. Wells AD

    (2020) Mapping effector genes at lupus GWAS loci using promoter Capture-C in follicular helper T cells

    Nature Communications 11:3294.

    https://doi.org/10.1038/s41467-020-17089-5

    • PubMed
    • Google Scholar
40. 1. Sumida T
    2. Lincoln MR
    3. Ukeje CM
    4. Rodriguez DM
    5. Akazawa H
    6. Noda T
    7. Naito AT
    8. Komuro I
    9. Dominguez-Villar M
    10. Hafler DA

    (2018) Activated β-catenin in Foxp3⁺ regulatory T cells links inflammatory environments to autoimmunity

    Nature Immunology 19:1391–1402.

    https://doi.org/10.1038/s41590-018-0236-6

    • PubMed
    • Google Scholar
41. 1. Thomas RM
    2. Gao L
    3. Wells AD

    (2005) Signals from CD28 induce stable epigenetic modification of the IL-2 promoter

    Journal of Immunology 174:4639–4646.

    https://doi.org/10.4049/jimmunol.174.8.4639

    • PubMed
    • Google Scholar
42. 1. Thomas RM
    2. Chunder N
    3. Chen C
    4. Umetsu SE
    5. Winandy S
    6. Wells AD

    (2007) Ikaros enforces the costimulatory requirement for IL2 gene expression and is required for anergy induction in CD4+ T lymphocytes

    Journal of Immunology 179:7305–7315.

    https://doi.org/10.4049/jimmunol.179.11.7305

    • PubMed
    • Google Scholar
43. 1. Thomas RM
    2. Chen C
    3. Chunder N
    4. Ma L
    5. Taylor J
    6. Pearce EJ
    7. Wells AD

    (2010) Ikaros silences T-bet expression and interferon-gamma production during T helper 2 differentiation

    The Journal of Biological Chemistry 285:2545–2553.

    https://doi.org/10.1074/jbc.M109.038794

    • PubMed
    • Google Scholar
44. 1. Trinh LA
    2. Ferrini R
    3. Cobb BS
    4. Weinmann AS
    5. Hahm K
    6. Ernst P
    7. Garraway IP
    8. Merkenschlager M
    9. Smale ST

    (2001) Down-regulation of TDT transcription in CD4(+)CD8(+) thymocytes by Ikaros proteins in direct competition with an Ets activator

    Genes & Development 15:1817–1832.

    https://doi.org/10.1101/gad.905601

    • PubMed
    • Google Scholar
45. 1. Wells AD
    2. Gudmundsdottir H
    3. Turka LA

    (1997) Following the fate of individual T cells throughout activation and clonal expansion. Signals from T cell receptor and CD28 differentially regulate the induction and duration of a proliferative response

    The Journal of Clinical Investigation 100:3173–3183.

    https://doi.org/10.1172/JCI119873

    • PubMed
    • Google Scholar
46. 1. Whyte WA
    2. Orlando DA
    3. Hnisz D
    4. Abraham BJ
    5. Lin CY
    6. Kagey MH
    7. Rahl PB
    8. Lee TI
    9. Young RA

    (2013) Master transcription factors and mediator establish super-enhancers at key cell identity genes

    Cell 153:307–319.

    https://doi.org/10.1016/j.cell.2013.03.035

    • PubMed
    • Google Scholar
47. 1. Xing S
    2. Gai K
    3. Li X
    4. Shao P
    5. Zeng Z
    6. Zhao X
    7. Zhao X
    8. Chen X
    9. Paradee WJ
    10. Meyerholz DK
    11. Peng W
    12. Xue HH

    (2019) Tcf1 and Lef1 are required for the immunosuppressive function of regulatory T cells

    The Journal of Experimental Medicine 216:847–866.

    https://doi.org/10.1084/jem.20182010

    • PubMed
    • Google Scholar
48. 1. Yang J
    2. Brook MO
    3. Carvalho-Gaspar M
    4. Zhang J
    5. Ramon HE
    6. Sayegh MH
    7. Wood KJ
    8. Turka LA
    9. Jones ND

    (2007) Allograft rejection mediated by memory T cells is resistant to regulation

    PNAS 104:19954–19959.

    https://doi.org/10.1073/pnas.0704397104

    • PubMed
    • Google Scholar
49. 1. Yates K
    2. Bi K
    3. Haining WN
    4. Cantor H
    5. Kim HJ

    (2018) Comparative transcriptome analysis reveals distinct genetic modules associated with Helios expression in intratumoral regulatory T cells

    PNAS 115:2162–2167.

    https://doi.org/10.1073/pnas.1720447115

    • PubMed
    • Google Scholar

Author details

1. Rajan M Thomas

   Center for Spatial and Functional Genomics, The Children’s Hospital of Philadelphia, Philadelphia, United States

   Contribution

   Conceptualization, Supervision, Visualization, Project administration, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-2652-9992

2. Matthew C Pahl

   Center for Spatial and Functional Genomics, The Children’s Hospital of Philadelphia, Philadelphia, United States

   Contribution

   Resources, Software, Supervision, Visualization, Project administration, Writing – review and editing

   Competing interests

   No competing interests declared

3. Liqing Wang

   Department of Pathology, Perelman School of Medicine at the University of Pennsylvania and The Children’s Hospital of Philadelphia, Philadelphia, United States

   Contribution

   Visualization, Project administration

   Competing interests

   No competing interests declared

4. Struan FA Grant

   1. Center for Spatial and Functional Genomics, The Children’s Hospital of Philadelphia, Philadelphia, United States
   2. Department of Pediatrics, Perelman School of Medicine at the University of Pennsylvania and The Children’s Hospital of Philadelphia, Philadelphia, United States

   Contribution

   Supervision, Funding acquisition, Project administration, Writing – review and editing

   Competing interests

   No competing interests declared

5. Wayne W Hancock

   Department of Pathology, Perelman School of Medicine at the University of Pennsylvania and The Children’s Hospital of Philadelphia, Philadelphia, United States

   Contribution

   Supervision, Funding acquisition

   Competing interests

   No competing interests declared

6. Andrew D Wells

   1. Center for Spatial and Functional Genomics, The Children’s Hospital of Philadelphia, Philadelphia, United States
   2. Department of Pathology, Perelman School of Medicine at the University of Pennsylvania and The Children’s Hospital of Philadelphia, Philadelphia, United States

   Contribution

   Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Visualization, Writing – review and editing

   For correspondence

   adwells@pennmedicine.upenn.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3630-2145

• 1,350

  views

• 159

  downloads

• 1

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.