Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Dec 1.
Published in final edited form as: Nat Genet. 2015 May 4;47(6):607–614. doi: 10.1038/ng.3283

NALP3 inflammasome up-regulation and CASP1 cleavage of the glucocorticoid receptor causes glucocorticoid resistance in leukemia cells

Steven W Paugh 1,2, Erik J Bonten 1,2, Daniel Savic 3, Laura B Ramsey 1,2, William E Thierfelder 1,2, Prajwal Gurung 4, R K Subbarao Malireddi 4, Marcelo Actis 5, Anand Mayasundari 5, Jaeki Min 5, David R Coss 6, Lucas T Laudermilk 1,2, John C Panetta 2, J Robert McCorkle 1,2, Yiping Fan 7, Kristine R Crews 1,2, Gabriele Stocco 1,2,15, Mark R Wilkinson 1,2, Antonio M Ferreira 6, Cheng Cheng 8, Wenjian Yang 1,2, Seth E Karol 1,2,9, Christian A Fernandez 1,2, Barthelemy Diouf 1,2, Colton Smith 1,2, J Kevin Hicks 1,2, Alessandra Zanut 1,2, Audrey Giordanengo 1,2, Daniel Crona 1,2, Joy J Bianchi 1,2, Linda Holmfeldt 1,10, Charles G Mullighan 1,10, Monique L den Boer 11, Rob Pieters 11,12, Sima Jeha 1,9, Thomas L Dunwell 13, Farida Latif 13, Deepa Bhojwani 1,9, William L Carroll 14, Ching-Hon Pui 1,9, Richard M Myers 3, R Kiplin Guy 5, Thirumala-Devi Kanneganti 4, Mary V Relling 1,2, William E Evans 1,2
PMCID: PMC4449308  NIHMSID: NIHMS675373  PMID: 25938942

Abstract

Glucocorticoids are universally used in the treatment of acute lymphoblastic leukemia (ALL), and leukemia cell resistant to glucocorticoids confers a poor prognosis. To elucidate mechanisms of glucocorticoid resistance, we determined the sensitivity to prednisolone of primary leukemia cells from 444 newly diagnosed ALL patients, revealing significantly higher expression of caspase 1 (CASP1) and its activator NLRP3 in glucocorticoid resistant leukemia cells, due to significantly lower somatic methylation of CASP1 and NLRP3 promoters. Over-expression of CASP1 resulted in cleavage of the glucocorticoid receptor, diminished glucocorticoid-induced transcriptional response and increased glucocorticoid resistance. Knockdown or inhibition of CASP1 significantly increased glucocorticoid receptor levels and mitigated glucocorticoid resistance in CASP1 overexpressing ALL. Our findings establish a new mechanism by which the NLRP3/CASP1 inflammasome modulates cellular levels of the glucocorticoid receptor and diminishes cell sensitivity to glucocorticoids. The broad impact on glucocorticoid transcriptional response suggests this mechanism could also modify glucocorticoid effects in other diseases.

Glucocorticoids are steroid hormones that regulate multiple physiological processes involved in inflammation, immunity, metabolism and other homeostatic functions. They exert their effects by binding to the glucocorticoid receptor (GR, NR3C1), triggering its activation and translocation to the nucleus, leading to transcriptional changes responsible for diminished proliferative capacity and apoptosis of acute lymphoblastic leukemia (ALL) cells1. Synthetic glucocorticoids are widely prescribed medications used to treat a variety of human diseases with an inflammatory component (e.g., asthma and autoimmune disorders), and are an essential component of curative therapy of ALL and lymphomas. Advances in the treatment of children with ALL have led to five-year disease-free survival rates exceeding 85%2. However, children whose ALL cells show in vitro resistance to glucocorticoids have a significantly worse treatment outcome (disease-free survival) than patients whose ALL cells are sensitive to glucocorticoids3–6. Yet, relatively little is known about the mechanisms causing leukemia cells from some patients to exhibit de novo resistance to glucocorticoids or why leukemia cells are more resistant to glucocorticoids at the time of disease recurrence5.

Here we report higher expression of two pro-inflammatory genes, CASP1 and its activator NLRP3 (NLR family, pyrin domain containing 3) in primary ALL cells that exhibited de novo resistance to glucocorticoids. We found that leukemia cells exhibiting higher expression of CASP1 and NLRP3 had significantly lower methylation of their promoter regions compared to glucocorticoid sensitive ALL. We show that CASP1/NLRP3 overexpression induces glucocorticoid resistance via CASP1 cleavage of the glucocorticoid receptor in its transactivation domain, reducing cellular levels of functional glucocorticoid receptor and diminishing glucocorticoid transcriptional effects. We further show that enforced expression of a glucocorticoid receptor that has been mutated to eliminate CASP1 cleavage sites mitigates glucocorticoid resistance due to CASP1 overexpression. Finally, we show that stably knocking down CASP1 expression with shRNA or reducing CASP1 activity with an inhibitory protein (CrmA) in CASP1-overexpressing leukemia cells increases cellular glucocorticoid receptor levels and markedly increases sensitivity to glucocorticoids.

Results

Higher CASP1, NLRP3 in glucocorticoid resistant leukemia

The de novo sensitivity of primary leukemia cells to prednisolone differed widely (>1000-fold) among patients in three independent cohorts of newly diagnosed children with ALL (Fig. 1A–C). We found that CASP1 and NLRP3, both members of the NALP3 inflammasome, were the two most highly over-expressed genes sharing a common pathway in steroid resistant ALL cells (Fig. 1D–E, Supplementary Fig. 1). The mean expression of CASP1 in steroid resistant leukemia was 1.6-fold higher than in sensitive leukemia cells (p = 3.2 × 10−7; Fig. 1D), whereas the mean expression of NLRP3 was 2.4-fold higher in prednisolone-resistant leukemia cells across all three cohorts of patients (p = 3.5 × 10−7; Fig. 1E).

Figure 1. Glucocorticoid resistant leukemia cells have higher expression and hypo-methylation of CASP1 and NLRP3 genes.

Figure 1

Open in a new tab

Primary leukemia cells were obtained from 444 patients (B and T cell leukemia) with newly diagnosed acute lymphoblastic leukemia and analyzed for their sensitivity to prednisolone using the MTT assay (see Methods)28. Distributions of measured LC50 values are shown for the three independent cohorts of patients; sensitive and resistant leukemias are highlighted in blue and orange, respectively (panels A–C). CASP1 (panel D) and NLRP3 (panel E) expression was significantly higher in glucocorticoid resistant leukemia cells from these three cohorts of newly diagnosed patients with B-lineage leukemia. In both patient cohorts for whom DNA was available for DNA methylation analysis (St. Jude Protocols XV and XVI), significantly lower levels of CASP1 (panel F) and NLRP3 (panel G) methylation were found in leukemia cells (from patients with B lineage leukemia) with higher expression of CASP1 and NLRP3. k-means clustering analysis (▲ represents k-means identified group A, ● represents k-means identified group B, pink and red squares represent k-means identified centers for group A and B respectively) utilizing only CASP1 and NLRP3 methylation status significantly discriminated sensitive leukemias (blue symbols, higher methylation) from resistant leukemias (orange symbols, lower methylation) in both St. Jude Protocol XV and XVI (panel H and supplementary Fig. 2) patients. Welch’s t-test p-values are shown for panels 1D–G and Fisher's Exact test p-value is shown for panel H. Boxes and whiskers are as defined in Online Methods.

Methylation of CASP1 and NLRP3 regulates their expression

To understand the basis for higher CASP1 and NLRP3 expression in glucocorticoid resistant leukemia cells, we assessed the relationship between CASP1 and NLRP3 mRNA expression and methylation of their promoter regions in leukemia cells. This revealed a highly significant relationship between the level of methylation of the CASP1 promoter and CASP1 mRNA expression in ALL cells (p = 1.4 × 10−22; Fig. 1F, Supplementary Fig. 2 panels A–C). In a subset of patients enrolled on St. Jude Protocol XVI where matching germline DNA from normal lymphocytes was available for methylation analysis (n = 55), CASP1 promoter methylation did not differ significantly (Paired t-test p = 0.495, Supplementary Fig. 3) in lymphocyte germline DNA and leukemia cell DNA across the entire population. In contrast, of 10 patients with significantly lower CASP1 promoter methylation in their ALL cells than their normal leukocytes, 70% were glucocorticoid resistant (n = 7), consistent with somatic CASP1 demethylation in glucocorticoid resistant ALL cells. Methylation of the promoter region of NLRP3 was significantly greater in leukemia cell DNA than in germline leukocyte DNA (Paired t-test p = 8.8×10−11, Supplementary Fig. 3), and the level of NLRP3 promoter methylation in leukemia cells correlated significantly with NLRP3 expression in leukemia cells (p = 6.7 × 10−4; Fig. 1G, Supplementary Fig. 2 panels D–F). Categorization of patients using k-means clustering of NLRP3 methylation and CASP1 methylation levels significantly distinguished prednisolone-sensitive (blue symbols) from prednisolone-resistant (orange symbols) leukemia cells (Fisher's Exact test for count data p = 4.7 × 10−7; Fig. 1H).

CASP1 cleaves the glucocorticoid receptor

Based on a prior report of CASP1 cleavage of the androgen receptor7, we identified a putative CASP1 cleavage site (LLID) in the transactivation domain of the glucocorticoid receptor (Fig. 2A, upper panel), and we performed enzymatic assays that confirmed glucocorticoid receptor cleavage by recombinant CASP1 (Fig. 2A, lower panel). Site directed mutagenesis of the LLID motif ablated cleavage of the glucocorticoid receptor at this location (Fig. 2B) and led to the identification of a secondary CASP1 cleavage site (IKQE) (Fig. 2A, top). CASP1-induced cleavage of the glucocorticoid receptor was inhibited by both tetrapeptide (Fig. 2A, lower panel) and small molecule inhibitors (VX765 and VRT043198) of CASP18 (Fig. 2C).

Figure 2. CASP1 cleaves the glucocorticoid receptor and increases resistance to glucocorticoids.

Figure 2

Open in a new tab

Bioinformatic analysis of the glucocorticoid receptor amino acid sequence (panel A, top) revealed a putative CASP1 cleavage site (LLID) in the glucocorticoid receptor (NR3C1) transactivation domain that is similar to a previously reported CASP1 cleavage site in the androgen receptor. Enzymatic assays revealed that recombinant CASP1 cleaves the glucocorticoid receptor and that this cleavage was inhibited by a CASP1 tetrapeptide inhibitor (panel A, bottom). Site directed mutagenesis of NR3C1 at the putative cleavage site (LLID motif) blocked CASP1 cleavage at this location and revealed a secondary CASP1 cleavage site more proximal to the carboxy terminus of the protein, as evidenced by a smaller enzymatic product (panel B). Further inhibition studies showed that small molecule inhibitors VX-765 and VRT-043198 (panel C) can also inhibit recombinant CASP1 cleavage of NR3C1, with VRT-043198 having higher inhibitory activity. Enforced expression of CASP1 in a human leukemia cell line (NALM-6) increased resistance to prednisolone and dexamethasone after activation of the NALP3 inflammasome (by the addition of LPS and ATP). NALM-6 cells were transduced with a lentivirus containing full length CASP1 and puromycin N-acetyl-transferase or puromycin N-acetyl-transferase alone (Control). Cells were puromycin selected and then their sensitivity to prednisolone and dexamethasone (panel E, data is representative of four independent experiments with standard deviation error bars) was measured using the MTT assay, in the presence(+) and absence(−) of inflammasome activation (LPS/ATP). The panel E inset is a western blot depicting the lower glucocorticoid receptor protein levels in cells over expressing CASP1 (24 hours after activation of CASP1).

CASP1 over-expression increases glucocorticoid resistance

We expressed full length CASP1 in a human leukemia cell line (NALM-6) to determine whether CASP1 over-expression increases resistance to glucocorticoids (Fig. 2D. Under standard conditions for activation of the NALP3 inflammasome and CASP19–13, cells transduced with empty vector control virus did not show differences in their sensitivity to prednisolone or dexamethasone, whereas cells transduced with CASP1-expressing virus had markedly higher LC50 for prednisolone and dexamethasone (Fig. 2E). On average, CASP1 over-expression increased resistance to prednisolone by 16.9-fold (Student's t-test p = 5.7×10−6) and dexamethasone by 5.3-fold (Student's t-test p = 1.2×10−9), with a 35% reduction in glucocorticoid receptor levels by 24 hours (Fig 2E). This was validated in a different human ALL cell line (697), revealing a similar increase in resistance to glucocorticoids (Supplementary Fig. 4).

CASP1 blunts transcriptional response to glucocorticoids

Because glucocorticoids regulate expression of target genes by binding to the glucocorticoid receptor, triggering its translocation into the nucleus and association with a glucocorticoid response element (GRE) to increase or decrease gene transcription, we assessed the ability of glucocorticoids to influence global gene expression in the presence or absence of CASP1 enforced expression. This revealed markedly diminished glucocorticoid-induced changes in gene expression levels in cells with CASP1 enforced expression compared to controls (Fig. 3A, 3B). Likewise, induction of Bim protein levels, encoded by BCL2L11, a known glucocorticoid response gene, was markedly blunted in CASP1 overexpressing cells (Fig. 3A inset). BCL2L11 upregulation was also significantly blunted at the mRNA level in CASP1 overexpressing cells as compared to control (fold change = 1.39 ± 0.12 versus fold change = 1.78 ± 0.17, p = 0.024). Genome wide assessment of changes in gene expression after glucocorticoid treatment (results of 4 replicate experiments) identified 93 genes that were induced by at least 3-fold following glucocorticoid treatment of control cells, 64 (68%) of which had lower (by at least 25%) induction in CASP1 overexpressing cells. Among these genes, the mean induction in control cells was 8.4-fold, compared to only 4.4-fold in CASP1 overexpressing cells (Paired t-test p = 0.01). Analysis of the sequences of these genes (including 5 kilobases up and downstream of each), showed a significant enrichment (46/93, 49%) of either known GREs (glucocorticoid receptor binding motif, Supplementary Table 1, motifs 1–9,14–16) or a genomic region shown by ChIP-seq experiments to bind the glucocorticoid receptor14, as compared to only 28% (1409/5000) of the randomly selected unchanged genes (Fisher’s Exact Test p = 2.2 × 10−5). Inclusion of an additional motif containing the top two most frequent bases at each position of a previously reported motif16 (motif 10, Supplementary Table 1) increased the percentage of upregulated genes with a GRE to 62% (58/93) of the transactivated genes.

Figure 3. Transcriptional modulation induced by glucocorticoids is diminished by CASP1.

Figure 3

Open in a new tab

Genome-wide gene expression was measured (n = 4) with median values shown (panels A and B, error bars depict median absolute deviation) and fold-differences (with or without prednisolone treatment) were compared between control cells treated with prednisolone and the same cells not treated with prednisolone or in cells with enforced expression and activation of CASP1 treated with prednisolone or not. The top 50 genes activated (blue bars) and repressed (green bars) by prednisolone in untreated control cells are shown, with the fold-change in CASP1 overexpressing cells shown by the adjacent gold bars. Additionally, control and CASP1 overexpressing cells were treated with and without prednisolone and LPS/ATP and protein lysates were collected for determination of Bim protein levels by western blotting, a known glucocorticoid response gene (panel A inset). The histogram (panel C) depicts the distribution of average normalized read count ratios at all reproducible glucocorticoid receptor binding sites identified across control and CASP1 overexpressing cells after prednisolone treatment. Binding sites with values above 0 highlight stronger enrichment in control cells while ratio values less than 0 depict sites with stronger enrichment in CASP1 overexpressing cells. The top binding motif identified (Panel D) is consistent with the canonical glucocorticoid receptor motif in both control and CASP1 overexpressing cells. The number and the extent of overlap of glucocorticoid receptor binding sites found is given in Panel E. ChIP enrichments near prednisolone-responsive genes are shown for 4 genes in Panel F. The images display ChIP enrichment in control cells (upper) and CASP1 overexpressing cells (lower).

Of the 119 genes whose expression was down-regulated by at least 50% following glucocorticoid treatment of control cells, 98 (82%) of these 119 genes were down-regulated to a lesser extent (by at least 25%) in CASP1 overexpressing cells. Among these genes, the mean reduction in control cells was 58%, compared to only 33% in CASP1 overexpressing cells (Paired t-test p < 2.2 × 10−16). Analysis of the sequences of these genes, including 5 kilobases up and downstream, revealed a significant enrichment (29/119, 24%) of NF-κB binding motifs (a transcription factor known to interact with an activated glucocorticoid receptor, Supplementary Table 1, motif 11) or a genomic region shown by ChIP-seq to bind to the glucocorticoid receptor14, as compared to randomly selected unchanged genes (849/5000, 17%; Fisher’s Exact Test p = 0.048). The top 50 up-regulated and down-regulated genes in control cells and their changes in cells with CASP1 enforced expression are depicted in Fig. 3, with the complete gene lists provided in the supplement (Supplementary Tables 2–4).

Additionally, CASP1 expression blunted genome-wide glucocorticoid receptor binding to DNA in response to prednisolone, as evidenced by ChIP-seq. The distribution of normalized read count ratios at all reproducible glucocorticoid receptor binding sites identified across control and CASP1 overexpressing cells showed a more than 4 fold enrichment of reads in control cells and 96% of all glucocorticoid receptor binding events displayed stronger read enrichments in control cells (Fig. 3C), in line with a lower level of functional glucocorticoid receptor from CASP1 over-expression. In support of these analyses, an assessment of identified binding sites revealed a marked overlap in glucocorticoid receptor binding events (Fig. 3E) between control and CASP1 overexpressing cells, and a 2.9 fold enrichment in the total number of binding sites was observed in control cells (10024 versus 3467 sites in control and CASP1 overexpressing cells, respectively). Notably, the top DNA-binding motif identified for control and CASP1 overexpressing cells (Fig. 3D) is consistent with the canonical glucocorticoid receptor binding motif. The extent of ChIP enrichment in control cells compared to CASP1 overexpressing cells is exemplified for four genes in Fig. 3F. Glucocorticoid receptor binding in CASP1 overexpressing cells consistently exhibit less ChIP enrichment compared to glucocorticoid receptor binding in control cells. Collectively, these data document a lower level of genome-wide glucocorticoid receptor occupancy in CASP1 overexpressing cells.

CASP1 and NLRP3 expression is higher in relapsed leukemia

To determine whether CASP1 and NLRP3 expression differed in leukemia cells at the time of disease recurrence (Fig. 4), we examined the mRNA expression levels of CASP1 and NLRP3 in paired ALL cells obtained at diagnosis and relapse from forty-nine patients17. This showed significantly higher expression of both CASP1 and NLRP3 in ALL cells at the time of relapse when compared to the corresponding samples obtained from the same patients at diagnosis (Paired t-test p = 3.2 × 10−4 and 4.2 × 10−3 respectively). Likewise, in a separate cohort of patients, we found that ALL cells at the time of disease recurrence (relapse) were more resistant to glucocorticoids when compared to sensitivity at the time of initial diagnosis (Fig. 4C; p = 9.72 × 10−7).

Figure 4. CASP1 and NLRP3 expression is higher and glucocorticoid resistance is greater in leukemia cells at the time of disease relapse.

Figure 4

Open in a new tab

Expression levels of CASP1 (panel A) and NLRP3 (panel B) in leukemia cells obtained from 49 patients at diagnosis and at the time of disease relapse. Quantile normalized measures of gene expression revealed significantly higher expression of NLRP3 and CASP1 at relapse. Paired t-test values are shown for panels A and B. Panel C depicts dexamethasone sensitivity (LC50) for 7 paired leukemia (ALL) cells obtained at diagnosis and at relapse from the same patients (connecting lines). Boxes and whiskers are as defined in Online Methods.

CASP1 inhibition reverses glucocorticoid resistance

To determine whether inhibition of CASP1 could diminish CASP1-induced glucocorticoid receptor cleavage and increase sensitivity to glucocorticoid treatment, we first used shRNA to knockdown over-expressed CASP1. This revealed that knockdown of CASP1 expression by ~50% (Fig. 5A, shCASP1) reduced CASP1-induced glucocorticoid receptor cleavage (Fig. 5A, shCASP1), and markedly enhanced glucocorticoid sensitivity (prednisolone LC50 = 12.8 ± 4.7 µM vs. 570 ± 423.2 µM, mean ± S.E.M., 44-fold reduction, t-test p = 0.028) (Fig. 5C, shCASP1) in CASP1 overexpressing ALL cells when compared to cells transduced with scrambled non-targeting shRNA (Fig. 5C, shNT).

Figure 5. Knockdown or inhibition of CASP1 reverses glucocorticoid receptor cleavage and prednisolone resistance.

Figure 5

Open in a new tab

NALM-6 cells overexpressing CASP1 were transduced with lentiviral vectors containing non-targeting scrambled hairpin (shNT) or shRNA hairpins targeting CASP1 (shCASP1). The two cell lines were cultured for 48 hours in the presence and absence of LPS/ATP, and analyzed on Western blot for the levels of CASP1 and glucocorticoid receptor (Panel A). The intensities of the glucocorticoid receptor and active CASP1 normalized against HSP90, are shown in Panel A. After activation of CASP1 (Panel A, lane 2 and 4), glucocorticoid receptor is approximately 2-fold lower in scrambled hairpin cells (Panel A, lane 4) whereas shCASP1 cells showed minimal degradation of the glucocorticoid receptor (Panel A, lane 2).

The prednisolone sensitivity (LC50) of shCASP1 and shNT cells was determined by MTT assays (Panel C). In the absence of CASP1 activation (− LPS/ATP), the PRED-LC50 of shCASP1 and shNT cells were comparable (Panel B). Upon activation of CASP1, shCASP1 cells were 44-fold more sensitive to prednisolone compared to shNT cells (Panel C). Error bars represent S.E.M, n = 4. Transduction of CrmA in CASP1 overexpressing cells showed effects similar to knockdown of CASP1 by shRNA. Without CASP1 activation, the PRED-LC50 for CrmA-expressing cells and GFP-expressing cells was comparable (Panel D). However, upon activation of CASP1, CrmA blocked CASP1 induced glucocorticoid receptor cleavage (Panel B) and increased glucocorticoid sensitivity by 43-fold compared to the GFP expressing control cells (Panel D). Error bars are as defined for Panel C.

Likewise when we transduced CASP1 overexpressing NALM-6 cells with the gene encoding CrmA (cytokine response modifier A), a known inhibitor of CASP1 catalytic activity18–20, this blocked CASP1 induced glucocorticoid receptor cleavage (Fig. 5B) and markedly increased sensitivity to glucocorticoids (3.2 ± 0.2 µM vs. 137.3 ± 24.3 µM, mean ± S.E.M., 43-fold reduction in LC50, t-test p = 0.011, Fig. 5D) when compared to GFP transfected controls.

We also over-expressed CrmA + RFP or RFP alone in glucocorticoid resistant primary leukemia cells that had a high level of CASP1 expression compared to several other glucocorticoid sensitive primary leukemia xenografts (not shown). This led to a diminution of the de novo glucocorticoid resistance (Supplementary Fig. 6A, dexamethasone LC50 = 0.14 µM (95% confidence interval = 0.1426 × 10−2 – 0.2773 µM versus > 10 µM, p < 1 × 10−14) and partial restoration of glucocorticoid receptor protein levels (Supplementary Fig. 6B). The levels of CASP1 expression in glucocorticoid-resistant primary ALL cells isolated from two patients (prednisolone LC50 = 1387 µM and 206.4 µM) were comparable to that of the glucocorticoid-resistant NALM-6 ALL cells in which heterologous CASP1 was over-expressed to recapitulate this mechanism of glucocorticoid-resistance (Supplementary Fig. 7).

Removal of CASP1 cleavage sites mitigates resistance

We overexpressed full length wild-type glucocorticoid receptor or a glucocorticoid receptor that had been mutated to eliminate the CASP1 cleavage sites (i.e., alanines substituted for the LLID and IKQE motifs identified in in vitro enzymatic assays), in leukemia cells overexpressing CASP1 in combination with either shRNA-based knockdown of overexpressed CASP1 or non-targeting shRNA control. Enforced expression of wild-type glucocorticoid receptor was unable to reverse CASP1 induced glucocorticoid resistance, whereas expression of a glucocorticoid receptor without the CASP1 cleavage sites markedly attenuated CASP1 induced glucocorticoid resistance (9.4-fold reduction in LC50), restoring sensitivity to levels similar to cells with shRNA-based knockdown of overexpressed CASP1 (Fig. 6).

Figure 6. Expression of glucocorticoid receptor with both CASP1 cleavage sites mutated to AAAA mitigates CASP1 induced prednisolone resistance.

Figure 6

Open in a new tab

NALM-6 cells overexpressing CASP1 co-transfected with either non-targeting scrambled hairpin (shNT) or shRNA hairpins targeting CASP1 (shCASP1) were stably transduced with lentiviral vectors containing a wild-type glucocorticoid receptor or a glucocorticoid receptor in which both CASP1 cleavage sites had been eliminated (double mutagenized glucocorticoid receptor, DM-GR). The sensitivity (LC50) of these cells to prednisolone was determined by MTT assays, revealing that cells overexpressing a glucocorticoid receptor without CASP1 cleavage sites remained sensitive to glucocorticoids when CASP1 was over-expressed. The inset is a western blot depicting recombinant glucocorticoid receptor protein levels (wild-type, or double mutagenized), in cells at time of MTT drug sensitivity assays. Data are representative of three independent experiments with standard deviation bars.

Discussion

The present study has revealed a novel mechanism by which CASP1 and its activator NLRP3 modulate the biological and pharmacological effects of glucocorticoids via cleavage of the glucocorticoid receptor. Glucocorticoids mediate their effects by binding to the glucocorticoid receptor, causing it to translocate into the nucleus where it modulates the expression of genes that contain a GRE. Prior work has shown that low cellular levels of functional glucocorticoid receptor, due either to siRNA knockdown14,21, rare mutations in the human NR3C1 gene encoding the glucocorticoid receptor22 or heterogeneity in the cellular levels of glucocorticoid receptor in leukemia cells by undefined mechanisms23, decrease sensitivity to glucocorticoids. However, it was previously not known that CASP1 can cleave the glucocorticoid receptor and thereby reduce functional receptor levels and modulate cellular response to glucocorticoids.

We have demonstrated that recombinant CASP1 cleaves the glucocorticoid receptor in its transactivation domain, and that forced over-expression of CASP1 coupled with its activation via the NLRP3 inflammasome causes human leukemia cells to become more resistant to glucocorticoids. Overexpression of CASP1 without its activation via the NLRP3 inflammasome did not alter glucocorticoid sensitivity of leukemia cells, demonstrating that catalytically active CASP1 is required for loss of glucocorticoid sensitivity. Further, overexpression of a glucocorticoid receptor lacking CASP1 cleavage sites mitigated the effects of CASP1 overexpression on leukemia cell sensitivity to glucocorticoids. Activation of the NALP3 inflammasome in the absence of CASP1 overexpression had only a small effect on glucocorticoid resistance, possibly due to activation of endogenous CASP1. Furthermore, our finding of higher expression of CASP1 and NLRP3 in leukemia cells at the time of disease relapse provides a plausible mechanism for prior observations that leukemia cells at relapse are more resistant to prednisolone compared to leukemia cells at initial diagnosis19. In support of this, we observed markedly greater resistance to dexamethasone in ALL cells obtained from patients at the time of relapse. Likewise, using publically available data, we found that many glucocorticoid-resistant ALL cell lines24,25 and primary ALL xenografts26,27 express relatively high levels of CASP1 (supplemental Figure 8). Collectively, this indicates that although there are multiple mechanisms via which ALL cells can become resistant to glucocorticoids28–30, CASP1 over-expression in glucocorticoid resistant ALL is not uncommon.

Importantly, when we knocked down CASP1 expression in CASP1 overexpressing cells, we were able to reverse CASP1-induced glucocorticoid receptor cleavage and markedly increase sensitivity to glucocorticoids, suggesting CASP1 inhibition as a potential new therapeutic strategy to mitigate glucocorticoid resistance. There have been small molecule CASP1 inhibitors in clinical development31, although phase 2 clinical trials for epilepsy failed to show therapeutic benefit and were terminated31. The lack of clinical efficacy of this class of agents is consistent with their inability to reverse glucocorticoid resistance in our CASP1 overexpressing cells, which may be due to their poor penetration into cells (not shown). Our experiments showing that CrmA inhibition of CASP1 catalytic activity restores sensitivity to glucocorticoids in CASP1 overexpressing leukemia cells provide a proof of principle for this strategy and a strong rationale for future studies to identify small molecule CASP1 inhibitors that can achieve sufficient cellular concentrations to reverse glucocorticoid resistance in CASP1 overexpressing leukemia cells.

CASP1 belongs to a family of cysteine proteases that cleave proteins following an aspartic acid residue. Produced as a pro-enzyme, CASP1 requires removal of its caspase activation and recruitment domain (CARD) before it becomes an active enzyme32. CARD cleavage is mediated by the formation of inflammasomes, of which the most extensively characterized is the NALP3 inflammasome. NALP3 (encoded by NLRP3) can be activated by exposure to pathogen associated molecular pattern or damage associated molecular pattern molecules, or by whole pathogens or environmental irritants32. There is also emerging evidence that the NALP3 inflammasome can form in response to host-derived molecules, including extracellular ATP, glucose or monosodium urate crystals10,32–34. In vivo induction of the NALP3 inflammasome typically results in self-oligomerization, recruitment of the ASC (PYCARD) adaptor protein, and clustering and auto-activation of CASP1. CASP1 is known to have pro-inflammatory effects, including the activation of inflammatory cytokines (e.g. interleukin 1β and interleukin 18). Our findings raise the possibility that during inflammatory processes, CASP1 negatively regulates anti-inflammatory glucocorticoid signaling to further amplify the pro-inflammatory effects of the NALP3 inflammasome.

The significant effects we observed on glucocorticoid sensitivity by partial reduction of cellular glucocorticoid receptor levels is consistent with prior studies showing that glucocorticoid receptor (NR3C1) haploinsufficiency produces a number of phenotypes including familial glucocorticoid resistance35 and adrenal hyperplasia36, alters blood pressure, glucose and lipid homeostasis in glucocorticoid receptor heterozygous mice37 and causes glucocorticoid resistance in Jurkat leukemia cells38.

The fact that not all glucocorticoid responsive genes were affected by CASP1 overexpression likely reflects differences among genes in their sensitivity to activated glucocorticoid receptor as a transcription factor. Whether glucocorticoids cause transactivation or transrepression of a given gene is determined in part by the presence or absence of either a positive GRE, leading to transactivation or a negative GRE leading to transrepression. Glucocorticoids may also repress gene expression via binding of the activated glucocorticoid receptor to transcription factors such as NF-κB and AP139,40. Although we found a significant enrichment of positive GREs among transactivated genes and negative GREs in transrepressed genes, failure to find GREs in all glucocorticoid-responsive genes likely reflects the current incomplete definition of GREs. Our ChIP-seq experiments showed that DNA binding by the glucocorticoid receptor is markedly diminished in CASP1 overexpressing cells. The broad spectrum of genes affected in CASP1 overexpressing cells indicates that our findings likely have broader implications, beyond the modulation of the anti-leukemic effects of glucocorticoids.

Taken together, our findings demonstrate a novel mechanism whereby leukemia cells develop resistance to glucocorticoids via somatic epigenetic changes that cause overexpression of CASP1 and NLRP3, leading to enhanced CASP1-mediated cleavage of the glucocorticoid receptor and diminished cellular response to glucocorticoids. These findings provide the foundation for future efforts to identify small molecule inhibitors of CASP1 that can reverse this mechanism of glucocorticoid resistance and thereby improve the treatment of acute lymphoblastic leukemia and potentially other illnesses for which glucocorticoids are used therapeutically.

Online Methods

Patients

A total of 444 children aged 18 years or younger with newly diagnosed ALL were included in this study. Two hundred seventy one were enrolled on the St. Jude Total Therapy XV or XVI protocols; 173 were enrolled on the 9th ALL Dutch Childhood Oncology Group protocol at Erasmus Medical Center, Sophia Children's Hospital, in Rotterdam, The Netherlands or on treatment protocols 92 and 97 of the German Cooperative Study Group for Childhood Acute Lymphoblastic Leukemia in Hamburg, Germany. Patients were enrolled in these studies if they had a diagnosis of ALL, were younger than 18 years of age, and had not been previously treated for ALL. In an additional cohort of 49 pediatric patients with relapsed B-precursor ALL, cryopreserved bone marrow samples from diagnosis and relapse from the Children's Oncology Group (COG) were subjected to gene expression analysis as previously described.17 All patients in the relapsed cohort had a bone marrow or bone marrow/central nervous system combined relapse and had been initially treated on COG protocols for newly diagnosed ALL17. In a separate cohort of patients enrolled at St. Jude Children’s Research Hospital, dexamethasone LC50 was measured at time of diagnosis and relapse. Written informed consent was obtained from all patients or their parents/guardians, and assent from the patients, as appropriate. The research and use of these samples were approved by the institutional review board at the host institution.

Gene expression analyses

Total RNA was extracted with TriReagent (Molecular Research Center, Inc.) from freshly isolated or cryopreserved mononuclear cell suspensions from patient bone marrow aspirates obtained at diagnosis. In the COG relapsed cohort, an RNA sample was extracted from leukemia cells from patient bone marrow aspirates obtained at the time of original diagnosis and again at the time of disease recurrence. All gene expression microarrays were performed by the St. Jude Children’s Research Hospital, Hartwell Center for Bioinformatics & Biotechnology. High-quality RNA was hybridized to the HG-U133A (GPL96) or HG-U133 Plus 2.0 (GPL570) oligonucleotide microarrays in accordance with the manufacturer’s protocol (Affymetrix). These microarrays contain 22,283 or 54,675 gene probe sets, representing approximately 18,400 or 47,400 human transcripts, respectively. Gene expression data were MAS541 processed using the affy42 Bioconductor43 R-project package or using Affymetrix Microarray Suite version 5.044,45 as previously described28.

DNA methylation analyses

DNA was isolated at the time of diagnosis from leukemia cells obtained from patients via bone marrow aspirates. Genome-wide DNA methylation status was determined using either an Infinium HumanMethylation27 BeadChip Kit or Infinium HumanMethylation450 BeadChip Kit in accordance with the manufacturer’s protocol (Illumina). HumanMethylation27 BeadChip experiments were performed at either Emory Integrated Genomics Core (EIGC) or Wellcome Trust Centre for Human Genetics Genomics Lab, Oxford, UK. HumanMethylation450 BeadChip experiments were performed at the Heflin Center for Genomic Science at the University of Alabama at Birmingham. DNA methylation status was classified as low if the Beta value46 was less than or equal to 0.25 and high if greater than 0.25.

Statistical analyses

Analyses were performed using R software unless otherwise specified. Exact Wilcoxon Mann-Whitney Rank Sum tests were used for analyses of differential gene expression and differential DNA methylation and Stouffer’s Z-score method was used for meta-analysis47. For boxplots, the upper and lower value of each box depicts the 75th and 25th percentile respectively, the solid line depicts the group median and the top and bottom of each dashed vertical line depicts the most extreme data point which is no more than 1.5 times the interquartile range (75th percentile – 25th percentile) from the box. K-means clustering analysis with a k = 2 was used for DNA methylation data and either untransformed or log-transformed gene expression data, with Fisher’s exact test for determining clustering significance. Fisher’s exact test was used to assess the significance of enrichment of known GREs.

CASP1 enzymatic assays

Recombinant human CASP1 (100–200 U, where U = 1 pmol/min at 30°C, 200 µM YVAD-pNA) from a CASP1 assay kit for drug discovery (Enzo Life Sciences, catalog number BML-AK701-0001) was incubated at 30°C with wild-type or mutated glucocorticoid receptor in the presence (10 µM unless otherwise indicated) or absence of inhibitors, (Ac-YVAD-CHO, Enzo Life Sciences, catalog number BML-P403-9090), VX765, VRT-043198. Substrates for enzyme assays were prepared in CASP1 assay buffer (Enzo Life Sciences, catalog number KI-111) consisting of 50 mM HEPES, pH7.4, 100 mM NaCl, 0.1% CHAPS, 10 mM DTT, 1 mM EDTA and 10% glycerol.

Site-directed mutagenesis

Site-directed mutagenesis was performed on NR3C1 using a QuikChange Site-Directed Mutagenesis Kit (Stratagene). Mutagenesis of the LLID motif was performed in 4 sequential mutagenesis reactions from Myc-DDK tagged NR3C1 (Origene, Rockville, MD, catalog # RC220189). The following mutagenesis primers were used: MutF1-4-NR3C1 and MutR1-4-NR3C1 (supplemental table 5). Mutagenesis of the IKQE motif was performed in 2 sequential mutagenesis reactions from the LLID mutated NR3C1. The following mutagenesis primers were used: MutF5-6-NR3C1 and MutR5-6-NR3C1 (supplemental table 5).

Successful mutagenesis was confirmed by Sanger sequencing.

Sensitivity of leukemia cells to glucocorticoids

Leukemia cells were isolated at diagnosis from patient bone marrow aspirates. If the leukemia cell percentage from diagnostic bone marrow samples was less than ninety percent, magnetic activated cell sorting (Miltenyl Biotec) was performed to further enrich leukemia cells. If red cell contamination was greater than thirty percent, red blood cell lysis was performed. Cells were centrifuged at 300g for five minutes and resuspended in RPMI-1640 with 2 mM L-glutamine, 20% heat-inactivated fetal bovine serum, 1X Antibiotic-Antimycotic solution (Life Technologies) and 1X ITS solution (Life Technologies) at a concentration of two million cells per milliliter. Eighty microliters of this leukemia cell suspension was then plated into each experimental well of round-bottom 96-well plates. Twenty microliters of decreasing concentrations of prednisolone were added and the plates were incubated for ninety-six hours in a humidified incubator containing 5% CO2 at 37°C. For the final six hours, ten microliters of 5 milligram per milliliter MTT (3-4,5-dimethylthiazol-2,5-diphenyl tetrazolium bromide) was added to each experimental well. Drug resistance assays in NALM-6 and 697 cell lines (Leibniz-Institut DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH) were performed using between 0.25 and 2.5 million cells per milliliter. Assays were developed and performed as described for primary ALL cells. LC50 values were determined for patients at diagnosis or replapse as previously described28, and consistent with prior analyses, glucocorticoid resistant ALL was defined as having an LC50 of 64 µM or greater, whereas glucocorticoid sensitive cases were defined as having an LC50 less than 0.1 µM. Among patients treated on the Total XV protocol (the only uniformly treated cohort with sufficient follow-up to assess event free survival), those patients whose samples were classified as sensitive to prednisolone (LC50 < 0.1 µM) had a significantly better treatment outcome than the other patients, (5 year event free survival 95.6% vs. 82.7%, p = 0.01 logrank test).

CASP1 enforced expression

Full length CASP1 cDNA (Origene, catalog # RC218364) was subcloned into a lentiviral backbone (System Biosciences, catalog number CD527A-1), in frame with a T2A linked puromycin resistance gene. Lentivirus was produced in 293T cells and the NALM-6 leukemia cell line was transduced. Seventy-two hours post transduction, cells were selected with 2.5 micrograms per milliliter puromycin. Where indicated, cells were treated with 10 µg/mL LPS (InvivoGen) followed 2 hours later by 5 mM ATP (Roche). NALM-6 and 697 cell lines were routinely authenticated and screened for mycoplasma.

Expression of mutant glucocorticoid receptor without CASP1 cleavage sites

Doubly mutated NR3C1 (LLID and IKQE) was subcloned into the pLX30448 lentiviral backbone (Addgene plasmid 25890). Lentivirus was produced in 293T cells and NALM-6 leukemia cell lines with CASP1 overexpression and shRNA directed towards CASP1 or a non-targeting shRNA were transduced. Seventy-two hours post transduction, cells were selected with blasticidin.

Expression of GFP and CrmA

Green fluorescent protein (Addgene plasmid 15301)49 or CrmA encoding cDNAs (Addgene plasmid 11832)50 was subcloned into the pLX30448 lentiviral backbone (Addgene plasmid 25890). Lentivirus was produced in 293T cells and NALM-6 leukemia cell lines with CASP1 overexpression and shRNA directed towards CASP1 or a non-targeting shRNA were transduced. Seventy-two hours post transduction, cells were selected with blasticidin.

Western blotting

In brief, cells were pelleted by centrifugation, washed once with PBS and lysed with RIPA buffer or CASP1 assay buffer (Enzo Life Sciences, catalog number KI-111), equal amounts of proteins (1–20 µg) separated by 4–12% Novex Bis-Tris gels (Life Technologies), and then transblotted to PVDF membranes (Life Technologies). Anti-glucocorticoid receptor (1:1000–1:10,000, BD catalog number 611227), anti-CASP1 p20 (Adipogen catalog number AG-20B-0048), anti-DDK tubulin (1:1000, Origene catalog number TA50011), anti-Bim (1:1000, Cell Signaling catalog number 2819) anti-Tubulin (1:1000, Santa Cruz, sc-8035), or anti-HSP90 (1:1000, Cell Signaling catalog number 4874) were used as primary antibodies followed by appropriate secondary HRP-conjugated IgG (1:1000, Dako) and immunocomplexes were visualized by chemiluminescence using a Chemidoc Imager (BioRad). Signal intensities of CASP1 and glucocorticoid receptor where quantified using Quantity One Software (BioRad) and normalized for the signal intensity of the loading control in the same lane.

Glucocorticoid induced changes in gene expression

NALM-6 transduced with either empty vector (control) or CASP1 containing lentiviral particles were treated with or without LPS/ATP, with or without 0.3 mM prednisolone, for 24 hours. RNA was extracted and hybridized to Affymetrix PrimeView oligonucleotide microarrays and RMA processed.

To be included as a gene that was transactivated by glucocorticoids, the median level of mRNA expression from 4 replicate experiments had to increase in control cells by at least 3-fold. To be included as a gene that was transrepressed by glucocorticoids, the median level of mRNA expression had to decrease by at least 50%.

Chromatin immunoprecipitation sequencing

NALM-6 transduced with either empty vector (control) or CASP1 containing lentiviral particles were treated with or without LPS/ATP, and with or without 0.3 mM prednisolone for 24 hours. Chromatin immunoprecipation sequencing (ChIP-seq) was performed with an antibody to the glucocorticoid receptor (Santa Cruz sc-1003), as previously described51. Briefly, cells were cross-linked with formaldehyde (1% final concentration) for 10 minutes at room temperature and then incubated with glycine (0.125 M final concentration) for 5 minutes to stop the cross-linking reaction. Cells were pelleted and washed with 1X PBS containing protease inhibitors (Roche Protease Inhibitor Cocktail Tablets, 11836153001) and stored at −80C prior to further experimentation. Crosslinked cell pellets were resuspended in cold lysis buffer (5 mM PIPES pH 8.0, 85 mM KCl and 0.5% NP-40 solution) containing protease inhibitors (Roche Protease Inhibitor Cocktail Tablets, 11836153001), and pelleted to isolate nuclei. The pellet was resuspended in cold RIPA lysis buffer (1X PBS, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS solution). Sonication was performed using a Bioruptor sonication system (Diagenode). Sonicated chromatin was added to Dynabeads (Invitrogen) coated with anti-glucocorticoid receptor antibodies (sc-1003) and incubated overnight at 4°C. The solution was washed with a LiCl wash solution (100 mM Tris pH 7.5, 500 mM LiCl, 1% NP-40 and 1% sodium deoxycholate) at 4°C, and a final wash using 1X TE buffer. Dynabeads were resuspended in an elution buffer (1% SDS and 0.1 M NaHCO3) and placed at 65°C to reverse crosslinks. DNA was purified and a standard library preparation was performed prior to sequencing on an Illumina HiSeq 2000 sequencer. Glucocorticoid receptor binding sites were identified using the MACS peak caller52.

ChIP-seq analyses

Reproducible glucocorticoid receptor ChIP-seq binding events were identified by a comparison of glucocorticoid receptor genome binding coordinates across replicate ChIP-seq experiments. Normalized read counts were assessed by compiling all reproducible binding events from control and CASP1 overexpressing cells (unique and common across both cells) and determining the number of reads, normalized to the total number of aligned reads per ChIP experiment, mapping to a 100-bp fragment of DNA centered on ChIP-seq binding site summits for all ChIP-seq replicate experiments. The average normalized read counts for each cell was determined and used to determine the ratio of read counts between control and CASP1 overexpressing cells. The MEME program was used to identify enriched motifs at glucocorticoid receptor binding sites53. Normalized read count ratios are defined as the control divided by CASP1 overexpressing cell next-generation sequencing reads and a reproducible peak was defined as being identified by both replicate ChIP-seq experiments in each cell.

Expression of CASP1 and NLRP3 in leukemia cells at the time of disease relapse versus at diagnosis

Gene expression was assessed at diagnosis and relapse as previously described17. This dataset was MAS541 processed using the affy42 Bioconductor43 R-project package and then quantile normalized. A paired t-test was used for comparison of the matching diagnosis and relapse data. In a separate cohort of patients enrolled at St. Jude Children’s Research Hospital, dexamethasone LC50 was measured at time of diagnosis or relapse.

Expression analysis of glucocorticoid sensitive and resistant xenograft ALLs

Affmetrix U133Plus2 expression data of primary ALL cells expanded as xenografts in mice were downloaded from the Pediatric Preclinical Testing Program, and MAS5 processed using the affy42 Bioconductor43 R-project package. Expression data was log2 transformed, and these ALL xenografts were categorized as either dexamethasone sensitive or dexamethasome resistant on the basis of their in vivo response to glucocorticoids26,27. Those xenografts determined to exhibit partial (intermediate) or complete resistance in vivo were categorized as resistant, and those that were sensitive in vivo were categorized as sensitive26,27.

Expression analysis of glucocorticoid-sensitive and resistant NCI60 ALL cell lines

Expression data of the NCI60 ALL cell lines (Chiron Pharmaceuticals) were constant translated (add 0.1) to eliminate zero values prior to log2transformation and combined with expression data from xenografts (see Supplementary Fig. 7) measured from RNA isolated at diagnosis. Additionally, expression data from NALM-6 cells were obtained and combined with this dataset. Data were median centered and scaled by dividing by the intrasample standard deviation, followed by quantile normalization. NCI60 drug sensitivity measurements showed that all cells lines in the leukemia panel, except CCRF−CEM, were prednisolone resistant (accession number: NSC 19987, September 2014 release).

Supplementary Material

1
2
3
4
5

Acknowledgements

We gratefully acknowledge the patients and parents who participated in this research. We are appreciative of the expert technical assistance of Michael Roberts, Yaqin Chu, Yan Wang, Monique A. Payton, John Stukenborg, Siamac Salehy, Margaret Needham, May Chung, Natalya Lenchik, Melanie Loyd and Emily Walker. We thank Julie Groff and Elizabeth Stevens for figure preparation assistance and Caleb Simmons for assistance with manuscript preparation. We thank Doug Green for his scientific advice and discussion of the manuscript. We thank Clinton Stewart, Geoff Neale, John Morris and Karen Rakestraw for their technical advice and expertise. This work was supported in part by NIH National Cancer Institute grant R37CA36401 (W.E.E., M.V.R., C.H.P.), NIH National Institute of General Medical Sciences Pharmacogenomics Research Network grant U01 GM92666 (M.V.R. and W.E.E.), NIH grant F32CA141762 (S.W.P), and an American Recovery and Reinvestment Act supplement 3R37CA036401-26S1 (W.E.E). This work was also supported by Cancer Center Support Grant CA 21765 from the National Cancer Institute and by the American Lebanese Syrian Associated Charities (ALSAC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Footnotes

URLs

William E. Evans lab website: http://www.stjuderesearch.org/site/lab/evans/

Pediatric Preclinical Testing Program website: http://pptp.nchresearch.org/

NCI60 ALL cell line expression data from Chiron Pharmaceuticals: https://wiki.nci.nih.gov/download/attachments/155845004/WEB_DATA_CHIRON.zip

Accession codes

DNA methylation, gene expression and ChiP-seq data available at GEO accession: GSE66708. Primary and relapse expression data available at GEO accession: GSE28460. NALM-6 expression data available at GEO accession: GSM852997.

Competing Interests

Authors William E. Evans, Steven W. Paugh and Erik J. Bonten are named as co-inventors on a pending patent application that relates to the subject matter of the article, which was filed by St. Jude Children’s Research Hospital.

Author Contributions

S.W.P., E.J.B., D.S., L.B.R., W.E.T., P.G., R.K.S.M., M.A., A.M., J.M., D.R.C., L.T.L., Y.F., R.K.G., T.D.K., M.V.R. and W.E.E. designed experiments. C.H.P., S.J., M.V.R. and W.E.E. designed clinical trials. S.W.P., E.J.B., D.S., L.B.R., W.E.T., P.G., R.K.S.M., M.A., A.M., D.R.C., L.T.L., Y.F., A.Z., A.G., D.C., J.J.B., L.H. performed experiments. S.W.P., E.J.B., D.S., L.B.R., W.E.T. and W.E.E wrote the manuscript (reviewed by all authors). S.W.P., E.J.B., D.S., L.B.R., W.E.T., D.R.C., L.T.L., J.C.P., J.R.M., Y.F., K.R.C., G.S., M.R.W., A.M.F., C.C., W.Y., S.E.K., C.A.F., B.D., C.S., J.K.H., A.Z., A.G., D.C., J.J.B. L.H., C.G.M., M.L.d.B., R.P., S.J., T.L.D., F.L., D.B., W.L.C., C.H.P., R.M.M., R.K.G., T.D.K., M.V.R. and W.E.E. analyzed data.

References

  • 1.Yudt MR, Cidlowski JA. The glucocorticoid receptor: coding a diversity of proteins and responses through a single gene. Mol Endocrinol. 2002;16:1719–1726. doi: 10.1210/me.2002-0106. [DOI] [PubMed] [Google Scholar]
  • 2.Pui CH, et al. Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med. 2009;360:2730–2741. doi: 10.1056/NEJMoa0900386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Den Boer ML, et al. Patient stratification based on prednisolone-vincristine-asparaginase resistance profiles in children with acute lymphoblastic leukemia. J Clin Oncol. 2003;21:3262–3268. doi: 10.1200/JCO.2003.11.031. [DOI] [PubMed] [Google Scholar]
  • 4.Kaspers GJ, et al. In vitro cellular drug resistance and prognosis in newly diagnosed childhood acute lymphoblastic leukemia. Blood. 1997;90:2723–2729. [PubMed] [Google Scholar]
  • 5.Pieters R, et al. Relation of cellular drug resistance to long-term clinical outcome in childhood acute lymphoblastic leukaemia. Lancet. 1991;338:399–403. doi: 10.1016/0140-6736(91)91029-t. [DOI] [PubMed] [Google Scholar]
  • 6.Dordelmann M, et al. Prednisone response is the strongest predictor of treatment outcome in infant acute lymphoblastic leukemia. Blood. 1999;94:1209–1217. [PubMed] [Google Scholar]
  • 7.Wellington CL, et al. Caspase cleavage of gene products associated with triplet expansion disorders generates truncated fragments containing the polyglutamine tract. J Biol Chem. 1998;273:9158–9167. doi: 10.1074/jbc.273.15.9158. [DOI] [PubMed] [Google Scholar]
  • 8.Boxer MB, Shen M, Auld DS, Wells JA, Thomas CJ. Probe Reports from the NIH Molecular Libraries Program. Bethesda (MD): 2010. A small molecule inhibitor of Caspase 1. [Google Scholar]
  • 9.Ogura Y, Sutterwala FS, Flavell RA. The inflammasome: first line of the immune response to cell stress. Cell. 2006;126:659–662. doi: 10.1016/j.cell.2006.08.002. [DOI] [PubMed] [Google Scholar]
  • 10.Mariathasan S, et al. Cryopyrin activates the inflammasome in response to toxins and ATP. Nature. 2006;440:228–232. doi: 10.1038/nature04515. [DOI] [PubMed] [Google Scholar]
  • 11.Shenoy AR, et al. GBP5 promotes NLRP3 inflammasome assembly and immunity in mammals. Science. 2012;336:481–485. doi: 10.1126/science.1217141. [DOI] [PubMed] [Google Scholar]
  • 12.Lu B, et al. Novel role of PKR in inflammasome activation and HMGB1 release. Nature. 2012;488:670–674. doi: 10.1038/nature11290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Subramanian N, Natarajan K, Clatworthy MR, Wang Z, Germain RN. The adaptor MAVS promotes NLRP3 mitochondrial localization and inflammasome activation. Cell. 2013;153:348–361. doi: 10.1016/j.cell.2013.02.054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Reddy TE, et al. Genomic determination of the glucocorticoid response reveals unexpected mechanisms of gene regulation. Genome Res. 2009;19:2163–2171. doi: 10.1101/gr.097022.109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Meijsing SH, et al. DNA binding site sequence directs glucocorticoid receptor structure and activity. Science. 2009;324:407–410. doi: 10.1126/science.1164265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Dunham I, et al. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489:57–74. doi: 10.1038/nature11247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hogan LE, et al. Integrated genomic analysis of relapsed childhood acute lymphoblastic leukemia reveals therapeutic strategies. Blood. 2011;118:5218–5226. doi: 10.1182/blood-2011-04-345595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Ray CA, et al. Viral inhibition of inflammation: cowpox virus encodes an inhibitor of the interleukin-1 beta converting enzyme. Cell. 1992;69:597–604. doi: 10.1016/0092-8674(92)90223-y. [DOI] [PubMed] [Google Scholar]
  • 19.Komiyama T, et al. Inhibition of interleukin-1 beta converting enzyme by the cowpox virus serpin CrmA. An example of cross-class inhibition. J Biol Chem. 1994;269:19331–19337. [PubMed] [Google Scholar]
  • 20.Garcia-Calvo M, et al. Inhibition of human caspases by peptide-based and macromolecular inhibitors. J Biol Chem. 1998;273:32608–32613. doi: 10.1074/jbc.273.49.32608. [DOI] [PubMed] [Google Scholar]
  • 21.Wang JC, et al. Chromatin immunoprecipitation (ChIP) scanning identifies primary glucocorticoid receptor target genes. Proc Natl Acad Sci U S A. 2004;101:15603–15608. doi: 10.1073/pnas.0407008101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Charmandari E, et al. A novel point mutation in helix 11 of the ligand-binding domain of the human glucocorticoid receptor gene causing generalized glucocorticoid resistance. J Clin Endocrinol Metab. 2007;92:3986–3990. doi: 10.1210/jc.2006-2830. [DOI] [PubMed] [Google Scholar]
  • 23.Pui CH, Dahl GV, Rivera G, Murphy SB, Costlow ME. The relationship of blast cell glucocorticoid receptor levels to response to single-agent steroid trial and remission response in children with acute lymphoblastic leukemia. Leuk Res. 1984;8:579–585. doi: 10.1016/0145-2126(84)90006-7. [DOI] [PubMed] [Google Scholar]
  • 24.Houghton PJ, et al. The pediatric preclinical testing program: description of models and early testing results. Pediatr Blood Cancer. 2007;49:928–940. doi: 10.1002/pbc.21078. [DOI] [PubMed] [Google Scholar]
  • 25.Shoemaker RH. The NCI60 human tumour cell line anticancer drug screen. Nat Rev Cancer. 2006;6:813–823. doi: 10.1038/nrc1951. [DOI] [PubMed] [Google Scholar]
  • 26.Bachmann PS, Gorman R, Mackenzie KL, Lutze-Mann L, Lock RB. Dexamethasone resistance in B-cell precursor childhood acute lymphoblastic leukemia occurs downstream of ligand-induced nuclear translocation of the glucocorticoid receptor. Blood. 2005;105:2519–2526. doi: 10.1182/blood-2004-05-2023. [DOI] [PubMed] [Google Scholar]
  • 27.Bachmann PS, et al. Divergent mechanisms of glucocorticoid resistance in experimental models of pediatric acute lymphoblastic leukemia. Cancer Res. 2007;67:4482–4490. doi: 10.1158/0008-5472.CAN-06-4244. [DOI] [PubMed] [Google Scholar]
  • 28.Holleman A, et al. Gene-expression patterns in drug-resistant acute lymphoblastic leukemia cells and response to treatment. N Engl J Med. 2004;351:533–542. doi: 10.1056/NEJMoa033513. [DOI] [PubMed] [Google Scholar]
  • 29.Pottier N, et al. The SWI/SNF chromatin-remodeling complex and glucocorticoid resistance in acute lymphoblastic leukemia. J Natl Cancer Inst. 2008;100:1792–1803. doi: 10.1093/jnci/djn416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Jones CL, et al. Loss of TBL1XR1 disrupts glucocorticoid receptor recruitment to chromatin and results in glucocorticoid resistance in a B-lymphoblastic leukemia model. J Biol Chem. 2014;289:20502–20515. doi: 10.1074/jbc.M114.569889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bialer M, et al. Progress report on new antiepileptic drugs: a summary of the Eleventh Eilat Conference (EILAT XI) Epilepsy Res. 2013;103:2–30. doi: 10.1016/j.eplepsyres.2012.10.001. [DOI] [PubMed] [Google Scholar]
  • 32.Schroder K, Tschopp J. The inflammasomes. Cell. 2010;140:821–832. doi: 10.1016/j.cell.2010.01.040. [DOI] [PubMed] [Google Scholar]
  • 33.Zhou R, Tardivel A, Thorens B, Choi I, Tschopp J. Thioredoxin-interacting protein links oxidative stress to inflammasome activation. Nat Immunol. 2010;11:136–140. doi: 10.1038/ni.1831. [DOI] [PubMed] [Google Scholar]
  • 34.Martinon F, Petrilli V, Mayor A, Tardivel A, Tschopp J. Gout-associated uric acid crystals activate the NALP3 inflammasome. Nature. 2006;440:237–241. doi: 10.1038/nature04516. [DOI] [PubMed] [Google Scholar]
  • 35.Karl M, et al. Familial glucocorticoid resistance caused by a splice site deletion in the human glucocorticoid receptor gene. J Clin Endocrinol Metab. 1993;76:683–689. doi: 10.1210/jcem.76.3.8445027. [DOI] [PubMed] [Google Scholar]
  • 36.Bouligand J, et al. Familial glucocorticoid receptor haploinsufficiency by non-sense mediated mRNA decay, adrenal hyperplasia and apparent mineralocorticoid excess. PLoS One. 2010;5:e13563. doi: 10.1371/journal.pone.0013563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Michailidou Z, et al. Glucocorticoid receptor haploinsufficiency causes hypertension and attenuates hypothalamic-pituitary-adrenal axis and blood pressure adaptions to high-fat diet. FASEB J. 2008;22:3896–3907. doi: 10.1096/fj.08-111914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Riml S, Schmidt S, Ausserlechner MJ, Geley S, Kofler R. Glucocorticoid receptor heterozygosity combined with lack of receptor auto-induction causes glucocorticoid resistance in Jurkat acute lymphoblastic leukemia cells. Cell Death Differ. 2004;11(Suppl 1):S65–S72. doi: 10.1038/sj.cdd.4401413. [DOI] [PubMed] [Google Scholar]
  • 39.McKay LI, Cidlowski JA. Cross-talk between nuclear factor-kappa B and the steroid hormone receptors: mechanisms of mutual antagonism. Mol Endocrinol. 1998;12:45–56. doi: 10.1210/mend.12.1.0044. [DOI] [PubMed] [Google Scholar]
  • 40.Teurich S, Angel P. The glucocorticoid receptor synergizes with Jun homodimers to activate AP-1-regulated promoters lacking GR binding sites. Chem Senses. 1995;20:251–255. doi: 10.1093/chemse/20.2.251. [DOI] [PubMed] [Google Scholar]

Methods-only references

  • 41.Hubbell E, Liu WM, Mei R. Robust estimators for expression analysis. Bioinformatics. 2002;18:1585–1592. doi: 10.1093/bioinformatics/18.12.1585. [DOI] [PubMed] [Google Scholar]
  • 42.Gautier L, Cope L, Bolstad BM, Irizarry RA. affy--analysis of Affymetrix GeneChip data at the probe level. Bioinformatics. 2004;20:307–315. doi: 10.1093/bioinformatics/btg405. [DOI] [PubMed] [Google Scholar]
  • 43.Gentleman RC, et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 2004;5:R80. doi: 10.1186/gb-2004-5-10-r80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Cheok MH, et al. Treatment-specific changes in gene expression discriminate in vivo drug response in human leukemia cells. Nat Genet. 2003;34:85–90. doi: 10.1038/ng1151. [DOI] [PubMed] [Google Scholar]
  • 45.Yeoh EJ, et al. Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling. Cancer Cell. 2002;1:133–143. doi: 10.1016/s1535-6108(02)00032-6. [DOI] [PubMed] [Google Scholar]
  • 46.Kuan PF, Wang S, Zhou X, Chu H. A statistical framework for Illumina DNA methylation arrays. Bioinformatics. 2010;26:2849–2855. doi: 10.1093/bioinformatics/btq553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Stouffer SA, Suchman EA, DeVinney LC, Star SA, Williams RM., Jr The American soldier: adjustment during army life. 1949 (Studies in social psychology in World War II, Vol. 1.). [Google Scholar]
  • 48.Yang X, et al. A public genome-scale lentiviral expression library of human ORFs. Nat Methods. 2011;8:659–661. doi: 10.1038/nmeth.1638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Boehm JS, et al. Integrative genomic approaches identify IKBKE as a breast cancer oncogene. Cell. 2007;129:1065–1079. doi: 10.1016/j.cell.2007.03.052. [DOI] [PubMed] [Google Scholar]
  • 50.Muzio M, Salvesen GS, Dixit VM. FLICE induced apoptosis in a cell-free system. Cleavage of caspase zymogens. J Biol Chem. 1997;272:2952–2956. doi: 10.1074/jbc.272.5.2952. [DOI] [PubMed] [Google Scholar]
  • 51.Savic D, Gertz J, Jain P, Cooper GM, Myers RM. Mapping genome-wide transcription factor binding sites in frozen tissues. Epigenetics Chromatin. 2013;6:30. doi: 10.1186/1756-8935-6-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Zhang Y, et al. Model-based analysis of ChIP-Seq (MACS) Genome Biol. 2008;9:R137. doi: 10.1186/gb-2008-9-9-r137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Bailey TL, Williams N, Misleh C, Li WW. MEME: discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res. 2006;34:W369–W373. doi: 10.1093/nar/gkl198. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS675373-supplement-1.pdf (174.4KB, pdf)
2
NIHMS675373-supplement-2.xls (60.5KB, xls)
3
NIHMS675373-supplement-3.xls (67KB, xls)
4
NIHMS675373-supplement-4.xls (488.5KB, xls)
5
NIHMS675373-supplement-5.doc (1.9MB, doc)

RESOURCES