Analysis and therapeutic targeting of the IL-1R pathway in anaplastic large cell lymphoma

Abstract

Anaplastic large cell lymphoma (ALCL), a subgroup of mature T-cell neoplasms with an aggressive clinical course, is characterized by elevated expression of CD30 and anaplastic cytology. To achieve a comprehensive understanding of the molecular characteristics of ALCL pathology and to identify therapeutic vulnerabilities, we applied genome-wide CRISPR library screenings to both anaplastic lymphoma kinase positive (ALK+) and primary cutaneous (pC) ALK- ALCLs and identified an unexpected role of the interleukin-1R (IL-1R) inflammatory pathway in supporting the viability of pC ALK- ALCL. Importantly, this pathway is activated by IL-1α in an autocrine manner, which is essential for the induction and maintenance of protumorigenic inflammatory responses in pC-ALCL cell lines and primary cases. Hyperactivation of the IL-1R pathway is promoted by the A20 loss-of-function mutation in the pC-ALCL lines we analyze and is regulated by the nonproteolytic protein ubiquitination network. Furthermore, the IL-1R pathway promotes JAK-STAT3 signaling activation in ALCLs lacking STAT3 gain-of-function mutation or ALK translocation and enhances the sensitivity of JAK inhibitors in these tumors in vitro and in vivo. Finally, the JAK2/IRAK1 dual inhibitor, pacritinib, exhibited strong activities against pC ALK- ALCL, where the IL-1R pathway is hyperactivated in the cell line and xenograft mouse model. Thus, our studies revealed critical insights into the essential roles of the IL-1R pathway in pC-ALCL and provided opportunities for developing novel therapeutic strategies.

Graphical abstract

Figure 1.

Genome-wide CRISPR library screens identify the IL-1R pathway as a novel oncogenic driver in pC-ALCL. (A) Outline of the workflow of the depletion CRISPR library screens in lymphoma cell lines (upper). Overview of the genome-wide CRISPR screen results (lower). Shown are the volcano plots of all genes in 2 lines. y-axis indicates the significance (−log₁₀p); x-axis indicates the log₂ fold change (sgRNA on/off). The dashed lines indicate P = .05 and log2(FC) = ±4. (B) Top significantly enriched cellular pathways identified among the oncogenic hits identified from MAC1 or Karpas299 genome-wide CRISPR screens [log₂(fold change) < −4 and P < .05] in panel A. (C) Left, diagrammatic representation of the workflow for the sgRNA toxicity assay. Right, ALCL lines were transduced with IRAK1, MYD88, IL1R1, IL1RAP, or control sgRNAs along with green fluorescent protein (GFP). The fraction of viable GFP+/sgRNA+ cells relative to the live cell fraction is plotted at the indicated times after sgRNA induction, normalized to day 0 values. Error bars denote standard deviation (SD) of triplicates. P was calculated comparing day 0 to each time point of indicated sgRNA induction in MAC1 and MAC2A lines. ∗∗P < .01.

Figure 2.

IL-1α production in ALK⁻ ALCL cell lines and primary cases. (A) Relative IL-1α and IL-1β mRNA expressions, measured by real-time polymerase chain reaction (PCR) and normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) values, in indicated lymphoma cell lines. Error bars denote standard error of the mean (SEM) of triplicates. (B) IL-1α production, measured by in-cell ELISA (cell lysate) and normalized to cell numbers, in indicated lymphoma cell lines. Error bars denote SEM of triplicates. (C) MAC1 and MAC2A lines were transduced with IL-1α or ctrl sgRNAs, selected, and expression induced. IL-1α production was measured by in-cell ELISA as in panel B. P was calculated comparing the ctrl sgRNA (sgCTRL) and sgIL-1α groups; ∗∗P < .01. Error bars denote SEM of triplicates. (D) Immunofluorescence confocal microscopy analysis of the distribution of endogenous IL1R1 and IL-1α in the indicated lines. Antibodies used are indicated on the top. (E) Indicated ALCL lines were transduced with Ctrl or IL-1α sgRNAs and selected. Relative proliferations upon sgRNA induction were measured by the Promega CellTiter Cell Proliferation Assay (MTS) and normalized to the Ctrl sgRNA groups. P was calculated comparing sgCTRL and sgIL-1α groups; ∗P < .05; ∗∗P < .01. Error bars denote SEM of triplicates. (F) Immunohistochemistry (IHC) of IL-1α, p-IRAK4, and CD30 expression of 2 pC ALK⁻ ALCL cases. Sections were examined microscopically using 100× original magnification. The depicted images are representative of the 19 ALK⁻ ALCL cases examined. (G) IL-1α and p-IRAK4 IHC scores (detailed at supplemental Table 2) in 3 subgroups of primary ALK⁻ ALCL cases. ∗P < .05; ∗∗P < .01.

Figure 2.

IL-1α production in ALK⁻ ALCL cell lines and primary cases. (A) Relative IL-1α and IL-1β mRNA expressions, measured by real-time polymerase chain reaction (PCR) and normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) values, in indicated lymphoma cell lines. Error bars denote standard error of the mean (SEM) of triplicates. (B) IL-1α production, measured by in-cell ELISA (cell lysate) and normalized to cell numbers, in indicated lymphoma cell lines. Error bars denote SEM of triplicates. (C) MAC1 and MAC2A lines were transduced with IL-1α or ctrl sgRNAs, selected, and expression induced. IL-1α production was measured by in-cell ELISA as in panel B. P was calculated comparing the ctrl sgRNA (sgCTRL) and sgIL-1α groups; ∗∗P < .01. Error bars denote SEM of triplicates. (D) Immunofluorescence confocal microscopy analysis of the distribution of endogenous IL1R1 and IL-1α in the indicated lines. Antibodies used are indicated on the top. (E) Indicated ALCL lines were transduced with Ctrl or IL-1α sgRNAs and selected. Relative proliferations upon sgRNA induction were measured by the Promega CellTiter Cell Proliferation Assay (MTS) and normalized to the Ctrl sgRNA groups. P was calculated comparing sgCTRL and sgIL-1α groups; ∗P < .05; ∗∗P < .01. Error bars denote SEM of triplicates. (F) Immunohistochemistry (IHC) of IL-1α, p-IRAK4, and CD30 expression of 2 pC ALK⁻ ALCL cases. Sections were examined microscopically using 100× original magnification. The depicted images are representative of the 19 ALK⁻ ALCL cases examined. (G) IL-1α and p-IRAK4 IHC scores (detailed at supplemental Table 2) in 3 subgroups of primary ALK⁻ ALCL cases. ∗P < .05; ∗∗P < .01.

Figure 3.

IL-1R pathway mediates NF-κB activation in ALK⁻ ALCL with higher IL-1α. (A) The workflow of using RNA-seq analysis to profile gene expression changes upon sgRNAs depletion of IL-1α in ALCL lines (upper). IL-1α–upregulated genes grouped according to gene expression signatures, in both MAC1 and MAC2A cell lines (lower). (B) Gene set enrichment analysis (GSEA) of gene expression signatures that were enriched among IL-1α upregulated genes in both MAC1 (upper) and MAC2A (lower) lines. Only the overlapped and shared signatures among all sgRNA transduced samples in both cell lines were shown. (C) NF-κB-driven luciferase reporter–engineered MAC1 (upper) or Karpas299 (lower) lines were transduced with indicated sgRNAs. Relative NF-κB reporter activities were measured after 4 days of induction. P was calculated comparing sgCTRL and the indicated sgRNA groups; ∗P < .05; ∗∗P < .01. (D) Indicated ALCL lines were transduced with MYD88, IL-1α, or Ctrl sgRNAs, selected, and expression induced. Lysates were analyzed by immunoblotting for the indicated proteins. (E) Indicated ALCL lines were transduced with MYD88 or Ctrl shRNAs, selected, and expression induced. Lysates were analyzed by immunoblotting for the indicated proteins. (F) MAC1 line was stable engineered with IKKβ WT, IKKβ S176/180E, or empty control, then transduced with MYD88 or Ctrl shRNAs along with GFP. The fraction of viable, GFP+/shRNA+ cells relative to the live cell fraction is plotted at the indicated times after shRNA induction, normalized to day 0 values. Error bars denote SD of triplicates. P was calculated by comparing the IKKβ S176/180E and IKKβ WT engineered lines with MYD88 shRNA transduction. ∗∗P < .01.

Figure 4.

The nonproteolytic protein ubiquitination network in ALCL. (A) Distribution of mutations in genes related to the inflammatory pathways of ALK⁻ALCL cell lines, obtained from RNA-seq analysis. (B) A20 protein expression in indicated ALCL lines. (C) TNFAIP3 (A20) mRNA expression in 417 T-cell lymphoma samples (144 peripheral T-cell lymphomas [PTCL] NOS, 127 AITL, 69 ALK⁻ ALCL, 56 ALK⁺ ALCL, 21 ATLL) and 61 T cells from healthy donors from publicly available microarray data set. t tests were performed between the control normal T-cell group and the other PTCL subtypes, and a P < .05 is considered statistically significant. (D) Overall survival in ALK⁻ ALCL with higher or lower expression values of A20. The threshold (cut-point value for high or low-expression group) was set using maximally selected rank statistics with the “survminer” R package. The cut-point value has been determined as 8.29134 (log2 expression), and the number of patients in the high-expression and low-expression groups is 29 and 8, respectively. (E) NF-κB–driven luciferase reporter–engineered MAC1 line was transduced with inducible lentiviral constructs encoding A20 WT, C103A, and C779/782A complementary DNAs or empty control. Relative NF-κB reporter activities were measured after 1 day of induction. (F) NF-κB–driven luciferase reporter–engineered MAC1 line was transduced with indicated sgRNAs. Relative NF-κB reporter activities were measured after 4 days of induction. (G) MAC1 cells were transduced with RNF31 or ctrl sgRNAs, selected, and expression induced. Cell lysates were subjected to biotin-labeled M1-specific TUBE binding and streptavidin purification and analyzed by immunoblotting. (H) MAC1 cells were transduced with RNF31 or ctrl sgRNAs, selected, and expression induced. IRAK1 immunoprecipitations or total lysates from these lines were immunoblotted for the indicated proteins. In panels E-F, error bars denote SEM of triplicates. P was calculated comparing sgCTRL and the indicated sgRNA groups or comparing empty vector and A20 expression groups; ∗∗P < .01.

Figure 5.

Cooperation of the JAK- STAT3 mutation and the IL-1R pathway in ALCL. (A) Indicated ALCL lines were transduced with STAT3 or JAK2 sgRNAs along with GFP. The fraction of viable GFP+/sgRNA+ cells relative to the live cell fraction is plotted at the indicated times after sgRNA induction, normalized to day 0 values. Error bars denote SD of triplicates. P was calculated comparing day 0 to each time point of indicated sgRNA induction in the MAC1 and MAC2A lines. ∗∗P < .01. (B-C) Indicated ALCL lines were transduced with MYD88, IL-1α, or ctrl sgRNAs, selected, and expression induced. Lysates were analyzed by immunoblotting for the indicated proteins. (D) Using CRISPR site-specific knockin to generate a MAC1 line carrying the STAT3 D661Y mutation. The 661st amino acid in WT and mutated STAT3 sequences and their coding sequences are marked red. To prevent the continuous DNA cutting by the Cas9 protein, the guiding RNA targeting protospacer adjacent motif (PAM) was also switched from “TGG” to “TCG” to produce a silent mutation. (E) MAC1 parental or STAT3 D661Y knockin lines were transduced with MYD88 or ctrl sgRNAs along with GFP. The fraction of viable GFP+/sgRNA+ cells relative to the live cell fraction is plotted at the indicated times after sgRNA induction, normalized to day 0 values. Error bars denote SD of triplicates. P was calculated by comparing the parental and STAT3 D661Y knockin lines with the indicated MYD88 sgRNA transduction. ∗∗P < .01. (F) MAC1 parental or STAT3 D661Y knockin lines were transduced with MYD88 or ctrl sgRNAs, selected, and expression induced. Lysates were analyzed by immunoblotting for the indicated proteins.

Figure 6.

IL-1R pathway contributes to the JAK inhibitor treatment sensitivity in pC-ALCL. (A) Chromatin IPs from the indicated antibodies were subjected to real-time PCR analysis for candidate STAT3-binding regions in the IL1R1 and IL1RAP loci in the indicated ALCL lines. (B-C) Indicated ALCL lines were transduced with STAT3, JAK2, or Ctrl sgRNAs, selected, and expression induced, IL1R1 and IL1RAP expressions were measured by real-time PCR. (D) Indicated ALCL lines were treated with the JAK inhibitor ruxolitinib at the indicated concentrations for 24 hours, and IL1R1 and IL1RAP expressions were measured. (E) Viability of indicated cell lines after treatment with AS2444697, ruxolitinib, or both (left). Formal calculation of synergism between AS2444697 and ruxolitinib (right). Positive highest-single-agent (HSA) synergy score values indicate synergy. (F) NSG mice bearing MAC1 xenografts were treated with AS2444697, ruxolitinib, the combination of AS2444697 and ruxolitinib, as well as vehicle controls. Tumor growth was measured as a function of tumor volume. Error bars denote SEM. (G-H) Tumor weight (G) and size (H) at the treatment end point. In panels A-D, error bars denote SEM of triplicates. P was calculated comparing sgCTRL and the indicated sgRNA groups or comparing untreated and treated groups; ∗P < .05; ∗∗P < .01. In panels F-G, error bars denote SEM. ∗P < .05; ∗∗P < .01.

Figure 7.

Targeting IRAK1/JAK2 using pacritinib in pC ALK⁻ ALCL cell lines and xenograft mouse models. (A) Indicated ALCL lines were treated with pacritinib at the indicated concentrations for 24 hours. Lysates were analyzed by immunoblotting for the indicated proteins. (B) Indicated ALCL lines were treated with pacritinib at the indicated concentrations for 3 days. Viability was measured by an MTS assay and normalized to DMSO-treated cells. Error bars denote SEM of triplicates. (C) The half-maximal inhibitory concentration (IC50) of pacritinib in ALCL lines. The IC50 was calculated using GraphPad prism (right). (D) NSG mice bearing MAC1 xenografts were treated with pacritinib or vehicle control. Tumor growth was measured as a function of tumor volume. Error bars denote SEM. ∗P < .05; ∗∗P < .01. (E-F) Tumor weight (E) and size (F) in pacritinib and vehicle treatment groups at the treatment end point. ∗∗P < .01. (G) Model of the action of the IL-1R pathway in pC ALK⁻ ALCL presented in this study.