IRF4 mediates non-enzymatic dependency on IRE1 in Multiple Myeloma cells

doi:10.21203/rs.3.rs-4494530/v1

Download PDF

Article

IRF4 mediates non-enzymatic dependency on IRE1 in Multiple Myeloma cells

https://doi.org/10.21203/rs.3.rs-4494530/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Multiple Myeloma (MM) arises through oncogenic transformation of immunoglobulin-secreting plasma cells. MM often co-opts the endoplasmic-reticulum (ER) stress mitigator, inositol-requiring enzyme 1 (IRE1) to sustain malignant growth. While certain MMs require enzymatic IRE1-dependent activation of the ER-homeostatic transcription factor XBP1s, others display a non-enzymatic dependency on IRE1 that is not yet mechanistically understood. Interferon regulatory factor 4 (IRF4) stimulates gene programs that promote immune-cell proliferation and plays an essential role in MM. Here we show that the non-enzymatic reliance on IRE1 in MM entails IRF4. IRE1 silencing increased inhibitory phosphorylation of IRF4, disrupting its chromatin-binding activity and mRNA transcription. IRF4 knockdown recapitulated, whereas IRF4 re-expression reversed the anti-proliferative phenotype of IRE1 silencing. Mechanistic studies revealed that IRF4 engages the E2F1 and CDC25A genes and promotes CDK2 activation to drive cell cycle progression. Our results uncover an unexpected functional link between IRE1 and IRF4 in MM.

Biological sciences/Cancer/Haematological cancer/Myeloma

Health sciences/Diseases/Haematological diseases/Haematological cancer/Myeloma

Genome instability, mutational burden, and metabolic restrictions induce endoplasmic-reticulum (ER) stress in tumor cells^1–4. In response, ER-resident inositol-requiring enzyme 1 (IRE1) functions via a conserved cytoplasmic kinase-endoribonuclease (RNase) module¹ to restore homeostasis. The IRE1 kinase mainly acts to regulate IRE1’s RNase activity^5–7. The IRE1 RNase cleaves the mRNA encoding X-Box binding protein 1 (XBP1), leading to the generation of “XBP1 spliced” (XBP1s) via non-conventional splicing¹. XBP1s then acts as a potent transcription factor to expand ER capacity and mitigate ER stress. Furthermore, IRE1’s RNase cleaves multiple additional mRNAs via a stringent, endomotif-based mechanism called regulated IRE1-dependent decay (RIDD)⁸, and through a more promiscuous process of RIDD lacking endomotif (RIDDLE)⁶.

Multiple Myeloma (MM) is a plasma-cell cancer with malignant characteristics that have rendered it practically incurable to date. MM cells retain a clonal immunoglobulin (Ig)-secretory phenotype. Hence, they display constitutive ER stress and increased levels of IRE1 and XBP1s⁹. Indeed, the IRE1-XBP1 pathway has emerged as a key player in MM pathogenesis^9–11. Experimental IRE1 loss-of-function attenuates in vitro growth of certain MM cell lines^9,11 and blocks their in vivo tumor growth⁹. Dependency on IRE1 in MM has been attributed primarily to XBP1s, because B-lineage-specific transgenic overexpression of XBP1s in mice induces a MM-like phenotype¹⁰. Moreover, in certain MM models XBP1 silencing mimics IRE1 disruption^9,11. However, recent work by our laboratory has identified several MM cell lines that depend on IRE1 but not on XBP1s itself or more generally on IRE1’s enzymatic activity¹². How IRE1 controls MM growth in such models remains unknown.

The transcriptional regulator Interferon response factor 4 (IRF4) orchestrates B-cell activation as well as plasma-cell differentiation and survival^13–17. During plasma-cell differentiation, IRF4 cooperates non-redundantly with the IRE1-XBP1s pathway, most prominently by inducing transcription of PRDM1/BLIMP1^18,19. BLIMP1 in turn facilitates—along with IRE1’s RNase activity—the de-repression of XBP1s^18,20,21, which is essential for Ig secretion^20,22,23. IRF4 is regarded as central to MM pathogenesis owing to its involvement in controlling transcriptional programs that promote cell cycle progression and cell survival^14,24. IRF4 gene amplification provides an important indicator of poor prognosis in MM²⁵. While some MMs have chromosomal translocations that juxtapose the IgH enhancer to the IRF4 locus, IRF4 gene alterations are relatively infrequent, altogether accounting for less than a quarter of MM cases^26–28. Additional molecular mechanisms that may drive IRF4 overexpression in MM include transcriptional self-induction¹³ and activation by Myc¹⁴. Loss of XBP1s indirectly upregulates both IRF4 and BLIMP1²⁹, highlighting an interconnected regulatory network between these genes. Here we report the surprising finding that IRF4 plays a key role in mediating non-enzymatic, XBP1-independent reliance on IRE1 in MM. Our study advances current mechanistic understanding of both IRE1 and IRF4 in this disease.

Gene expression and analysis

Total RNA was extracted from cells using RNA Plus Mini kit (Qiagen) and analyzed by qRT-PCR with RNA-to-Ct kit (Applied Biosystems) and the Taqman probes provided in the supplemental.

Nuclear Run-On

Click-iT Nascent RNA Capture Kit (Life Technologies) was used. Cells were pulsed with 0.5mM EU and total RNA was extracted (Qiagen). Nascent RNA was then biotinylated and captured according to the manufacturer's instructions. Biotinylated RNA was reverse-transcribed and PCR followed immediately (Thermo-Fisher).

Confluence

5000 cells/well were plated and centrifuged in ULA 96-well plates (Corning) forming single spheroids⁹. Spheroids were live-imaged by an IncuCyte instrument (EssenBioscience) with a 4x objective.

TUNEL assay

Cells were stained with “In Situ Cell Death Detection” (Roche) kit, and visualized by FACS after staining with LIVE/DEAD fixable dyes (Life Technologies), according to protocol.

Subcellular Fractionation

We used the “Subcellular Protein Fractionation Kit for Cultured Cells” (Thermo). Cell pellets were sequentially lysed with different lysis buffers according to manufacturer’s instructions. Each post-lysis supernatant was a different cellular fraction (Cytoplasmic, Membrane, Soluble nuclear, Chromatin-bound nuclear). Lysates were subjected to immunoblot analysis (IB).

Co-immunoprecipitations

Cells were lysed in a buffer containing NP-40. Post-nuclear supernatants were pre-cleared and CDK2 complexes were captured on CDK2-conjugated magnetic beads. Precipitates were then washed and eluted in sample buffer for IB.

Cell proliferation

Cells were incubated with Cell Trace Violet (Thermo-Fisher). After a 20-min incubation in a 37°C water-bath, Serum-Free RPMI 1640 was added to the cells and incubated for 5min to absorb unbound dye. Cells were plated and treated for 72h, harvested and stained with “LIVE/DEAD Fixable Dead Cell Stain” (Life Technologies), and immediately analyzed by FACS.

DNA replication

BrdU incorporation was measured using the “BrdU Staining Kit for Flow Cytometry” (eBioscience/Invitrogen). Cells treated in triplicate were pulsed with BrdU (10uM) for 45min. Cells were then stained with “Zombie Fixable Viability Kit” (Biolegend) and treated with BrdU staining working solution and DNase I as per manufacturer’s instructions. Finally, anti-BrdU antibody or isotype control were added and samples were analyzed by FACS.

ChIP-sequencing

Nuclei from double crosslinked cells were isolated, sonicated, and the fragmented chromatin was immunoprecipitated overnight. Antibody-chromatin complexes were pulled-down with Protein A magnetic beads. Eluted DNA was purified with MinElute PCR purification kit (Qiagen), and sequencing libraries were constructed with the Ovation Ultralow Library System V2 (Tecan). Libraries were sequenced with Illumina Nextseq, using paired-end 50bp read configuration. For further information, see supplemental methods.

DNA methylation

Total DNA was extracted from cells using the AllPrep DNA/RNA Mini Kit (Qiagen) and 100ng DNA was analyzed by MethylFlash Global DNA methylation (5-mC) ELISA Easy kit (Epigentek), following manufacturer’s protocol. Absorbance was measured at 450nm and total 5mc-DNA was calculated based on a standard curve.

Silencing of IRE1 but not XBP1 downregulates IRF4

We recently discovered an unexpected nonenzymatic dependency on IRE1 in a number of MM cell lines, including AMO1, KMS27, L363, and JJN3¹². While some of these lines displayed weaker IRE1 and/or XBP1 dependency in massive-scale genome-wide functional screens (DepMap.Org Portal; ref. ³⁰), an in-depth analysis based on inducible shRNA silencing revealed a much greater dependency of these lines on IRE1 as compared to XBP1 under exponential-phase culture in vitro and during tumor-xenograft growth in vivo¹². Furthermore, while IRE1 silencing in these models caused tumor regression, pharmacologic inhibition of IRE1 did not alter tumor growth despite its effective blockade of IRE1’s enzymatic activity. To identify genes that might contribute to this nonenzymatic dependency on IRE1, we combined data from the DepMap.Org Portal (Chronos model, which assigns cell-fitness-effect scores of 0 or -1, respectively, to non-essential or essential genes) with our previously obtained transcriptomic and proteomic data from IRE1-silenced MM cells¹². We first examined a set of genes that showed unique essentiality to MM versus other cancers (DepMap.Org Portal; ref. ³⁰), while at the same time being required in IRE1-dependent MM cell lines (corrected gene-effect: Supp.Table 1). We then filtered these genes further based upon their reliance on IRE1 for mRNA expression in AMO1 cells (Fig. 1A). IRF4 had the lowest Chronos score (Fig. 1B; Supp.Table 1; ref. ³⁰), highlighting it as the most crucial among these genes for MM cell survival. The RNAseq data further showed that Doxycycline (Dox)-induced shRNA silencing of IRE1 but not XBP1 significantly downregulated the IRF4 mRNA (Fig. 1C). Corresponding proteomics data¹² confirmed the depletion of IRF4 protein (Fig. 1D). Further interrogation of the AMO1, KMS27, and L363 cell lines verified the selective depletion of IRF4 mRNA and protein upon IRE1 but not XBP1 knockdown (Fig. 1E-F; Fig. S1A-B). Thus, in IRE1-dependent XBP1-independent MM models, IRE1 upregulates IRF4 without requirement for XBP1s. To assess this further, we used an IRE1 RNase inhibitor, which successfully blocked both XBP1 splicing and RIDD (Fig. S1C-D). IRE1 inhibition did not decrease IRF4 levels; rather, in AMO1 cells it increased IRF4 levels (Fig. 1G-1J), in keeping with previous evidence that IRF4 can be targeted by RIDD³¹. These results identified IRF4 as a potential mediator of non-enzymatic IRE1 dependency in MM cells and revealed that IRE1 can promote IRF4 expression distinctly from its known IRF4 suppression via XBP1s and RIDD.

IRE1 silencing attenuates IRF4 activity

The evidence that IRE1 knockdown depleted both the IRF4 mRNA and protein suggested that IRE1 may primarily control IRF4 at the mRNA level. Indeed, an actinomycin D chase experiment revealed that IRE1 knockdown slightly accelerated IRF4 mRNA degradation (Fig. S2A), but this effect seemed too modest to fully explain the overall reduction in IRF4 mRNA. Importantly, nascent-mRNA analysis showed that IRE1 silencing significantly reduced de novo IRF4 transcription, affecting this particular feature more substantially than mRNA turnover (Fig. 2A; Fig. S2A). Our transcriptomic data¹² indicated that IRE1 silencing in AMO1 cells upregulated the IRF4-inducing NFκB pathway³². Therefore, we interrogated several other transcription factors previously shown to either suppress (MITF) or induce (Ikaros, Aiolos, Myc) IRF4³³. However, IRE1 knockdown downregulated MITF mRNA and protein (Fig. 2B). Furthermore, IRE1 silencing did not decrease Ikaros/IKZF1 and Aiolos/IKZF3 mRNA levels (Fig. 2C), and although it did deplete the Aiolos protein (Fig. 2D), this occurred later than the downregulation of the IRF4 protein. Moreover, while IRE1 silencing decreased MYC mRNA and protein abundance (Fig. 2D; Fig. S2B), MYC knockdown did not reduce IRF4 mRNA or protein expression (Fig. 2E; Fig. S2C). Taken together, this data excluded these previously known IRF4-regulating transcription factors from instigating transcriptional control of IRF4 downstream to IRE1.

Cycloheximide-chase experiments further revealed that IRE1 knockdown did not increase turnover of the IRF4 protein; rather, it augmented IRF4 protein stability (Fig. 2F). This contrasted with a destabilization of IRF4 upon XBP1 knockdown (Fig. S2D). Moreover, a specific E1 ubiquitin-ligase inhibitor did not restore IRF4 levels during IRE1 silencing (Fig. S2E). These data suggested that IRE1 silencing does not downregulate IRF4 by reducing its protein stability. Since IRF4 is an auto-stimulatory transcription factor^13,34, we next interrogated the activity of IRF4 itself. Subcellular fractionation revealed that IRE1 silencing markedly decreased the amount of IRF4 bound to chromatin (Fig. 2G), indicating a functional disruption of IRF4 activity. Furthermore, IRE1 silencing but not enzymatic blockade of IRE1 with kinase or RNase inhibitors significantly increased IRF4 phosphorylation on Ser114 as well as Ser270 (Fig. 2H). To examine the functional consequence of these phosphorylation sites, we ectopically expressed in AMO1 cells a non phosphorylatable IRF4 mutant in which we substituted the targeted serines with alanines (S114A_S270A). This mutant displayed a 5-fold increase in its chromatin-binding capacity as compared to WT IRF4 (Fig. 2I; Fig. S2F), indicating that phosphorylation on these sites inhibits IRF4 activity. Together, these findings suggest that IRE1 controls IRF4 transcription mainly by regulating IRF4’s phosphorylation state, chromatin binding and self-inducing activity.

IRF4 silencing recapitulates the anti-proliferative phenotype of IRE1 knockdown

To validate DepMap’s evidence for IRF4’s essentiality, we generated AMO1 and KMS27 lines expressing Dox-inducible shRNA against IRF4 (shIRF4) and confirmed efficient IRF4 depletion upon Dox treatment (Fig. S3A). Consistent with DepMap data, IRF4 knockdown substantially inhibited spheroid growth of these cells (Fig. 3A); notably, IRF4 knockdown disrupted cell growth with kinetics and to an extent comparable to those of IRE1 silencing.

Previous studies indicate that IRF4 disruption has multifaceted outcomes, including cell cycle arrest and/or cell death^35,36. In keeping with this, depletion of either IRE1 or IRF4 not only inhibited proliferation (Fig. 3A) but also induced apoptosis, evident by increased cleavage of apoptotic caspase substrates PARP1 and Lamin A/C, as well as a robust increase in cytosolic histone H3, H2AX, and γH2AX–a marker of DNA-damage-induced apoptosis (Fig. S3B). Caspase Glo assays (Fig. S3C) and Annexin V/PI staining (Fig. S3D) confirmed the induction of apoptosis in response to IRE1 or IRF4 silencing. We previously found that growth deficiency upon IRE1 knockdown stems from loss of proliferation rather than increased cell death¹². Notably, IRF4 silencing appeared to confer growth inhibition by 48 h, while inducing clear apoptosis activation later, at 72 h (Fig. 3A; Fig. S3B-D). Furthermore, the pan-caspase inhibitor Q-VD–used at a concentration that effectively prevented multiple apoptotic features (Fig. S3E-F)–failed to rescue growth upon silencing of either IRE1 or IRF4 (Fig. S3G-H). Thus, similar to IRE1 silencing, IRF4 knockdown inhibits proliferation independently of apoptosis.

To further characterize the attenuated proliferation, we quantified cell divisions by staining AMO1 cells with carboxyfluorescein succinimidyl ester (CFSE) and tracking dye dilution through flow cytometry. Etoposide, used as an anti-proliferative control, confirmed the lack of CFSE dilution in the absence of cell division (Fig. 3B; Fig. S3I-J). Both IRE1- and IRF4-silenced cells underwent fewer mitotic divisions than did their non-silenced counterparts, whereas XBP1-silenced cells showed no difference versus controls (Fig. 3B; Fig. S3I-J). Additional staining to monitor cell death indicated that cells that failed to divide proceeded to die (Fig. S3J), suggesting that proliferative defects precede the apoptotic phenotype in both backgrounds. Bromo-deoxy-uridine (BrdU) incorporation further showed that knockdown of IRE1 or IRF4 progressively attenuated DNA replication with similar kinetics (Fig. 3C). To characterize specific cell cycle phases, we tracked DNA content by propidium iodide (PI) staining of AMO1 cells. Flow cytometry showed that knockdown of either IRE1 or IRF4 led to G1 phase accumulation at the expense of S-phase by 16 h post-Dox treatment (Fig. 3D), suggesting rapid G1 arrest. KMS27 cells, which have a longer doubling time, nevertheless displayed G1-arrest by 24 h of IRE1 or IRF4 knockdown (Fig. S3K). We next followed a cohort of G2-synchronized AMO1 cells transitioning first into G1- and then into S-phase. Not only did these cells undergo a G1 arrest, but also their G1 proportion further increased due to a faster transition from G2 to the G1 phase of the next cell cycle (Fig. 3E). Thus, the proliferative disruption conferred by IRE1 or IRF4 silencing may be due to defects in early as well as late phases of the cell cycle. Substantiating this notion, already at 24 h after Dox addition, IRE1 or IRF4 silencing inactivated CDK2–a kinase that temporally coordinates not only G1/S transition but also G2/M progression and mitosis^37–41 (Fig. 3F; Fig. S3L-M). This CDK2 inactivation was evident from the altered migration profile of the CDK2 protein⁴² (Fig. 3F); as well as from its increased binding to p21 and decreased binding to Rb (Fig. 3F). Furthermore, IRE1 or IRF4 knockdown attenuated stimulatory CDK2 phosphorylation on Thr160 (Fig. S3L-M) and IRF4 knockdown augmented inhibitory CDK2 phosphorylation on Thr14 (Fig. S3M). One factor that may contribute to the altered CDK2 phosphorylation state is the phosphatase Cdc25A, which was significantly downregulated upon knockdown of IRF4 (Fig. S3N). In keeping with the known nuclear localization of CDK2 in proliferating cells⁴³, we detected CDK2 in the soluble nuclear fraction of control cells; strikingly, silencing of IRE1 or IRF4 eliminated CDK2 from this compartment (Fig. 3G), providing further evidence for the disruption of CDK2. Taken together, these results show that knockdown of either IRE1 or IRF4 leads to a remarkably similar anti-proliferative phenotype in IRE1-dependent MM cells, characterized by cell cycle arrest in conjunction with CDK2 inactivation.

IRE1 and IRF4 regulate a highly overlapping set of cell cycle genes

Previous transcriptomic data revealed marked changes in the expression of multiple cell cycle genes already by 24 h of IRE1 silencing in AMO1 cells¹². To assess if IRF4 exerts similar cell cycle controls, we performed bulk RNA sequencing (RNA-seq) of AMO1 shIRF4 cells treated with Dox for 0, 8, 16 and 24 h and KMS27 shIRF4 cells treated with Dox for 0, 20 and 30 h (to accommodate the slower KMS27 growth rate). For comparison, we used cells expressing non-targeting control shRNA. As expected, IRF4 silencing depleted mRNAs encoding IRF4 itself and other previously characterized IRF4 targets such as MYC^14,44, PRDM1^13,14,17,19 and XBP1^17,19 (Fig. S4A). For consistency, we re-analyzed the published IRE1 RNA-seq alongside the IRF4 data. Gene-set enrichment analysis (GSEA) revealed a strong and significant decrease in the Hallmark pathways “G2/M Checkpoint”, “E2F Targets”, “Myc targets” and “Unfolded Protein Response” for both datasets (Fig. 4A; Fig. S4B), corroborating the observed cell cycle defects (Fig. 3) and the previous IRE1 RNA-seq results¹². To gauge the overlap between IRE1- and IRF4-dependent differentially expressed genes (DEGs) further, we leveraged the GSEA Leading-Edge analysis feature (Supp.Table 2). DEGs in the top cell cycle-related molecular signatures (G2M Checkpoint and E2F Targets) overlapped by approximately 84%, with 91% of the IRF4 DEGs being shared by IRE1 (Fig. 4B). Analysis with the more granular Wiki Pathway “G1/S Cell Cycle Control” indicated a similar overlap (Fig. S4C), consistent with a mechanistic linkage between IRE1 and IRF4 in cell cycle control. By contrast, while both IRE1 and IRF4 knockdown depleted the ‘Unfolded Protein Response’ gene-set and led to a similar decline in PERK and ATF6 (Fig. 4A; Fig. S4B), the two backgrounds showed only 45% overlap (Fig. S4C), suggesting different entry points into this pathway distinct from their strong convergence on cell cycle pathways.

In keeping with the GSEA, multiple cell cycle genes including E2Fs were perturbed also at the protein level (Fig. 4D; Fig. S3N). Of note, MYC protein was more refractory to changes under either of the two silenced backgrounds (Fig. 4D), suggesting a plausible explanation for why MYC targets were not prominent among enriched gene-sets (Fig. 4A). Rb transcription remained essentially unaffected upon silencing of IRE1 or IRF4 (Fig. 4C). If anything, the Rb protein appeared to be degraded, subsequent to its dephosphorylation in both silenced backgrounds (Fig. 4D; Fig. S4E-G). These results exclude Rb as a key mediator for the attenuation of E2F and of cell cycle progression. On the other hand, inactivation of CDK2 (Fig. 3F; Fig. S3L-M; Fig. 4D) likely accounted for the cessation of Rb phosphorylation. The loss in E2F function, as evident by the downregulation of multiple E2F targets and concomitant modulation of several cell cycle regulators (Fig. 4C-D), could decrease CDK2 activity through a number of mechanisms, including de-repression of the CDK inhibitor CDKN1A/p21 and downregulation of Cyclins, UHRF1, and Cdc25A (Fig. 4C-D; Fig. S3N).

IRF4 directly controls E2F1 gene transcription

E2F1 was one of the earliest G2/M checkpoint gene transcripts that IRF4 silencing decreased, and its downregulation preceded and persisted beyond the onset of cell cycle arrest (Fig. 4E). Additionally, in IRF4-depleted KMS27 cells, ‘E2F Targets’ was the most downregulated pathway (Fig. S4H). Notably, IRF4 knockdown induced DEGs that displayed only partial overlap between AMO1 and KMS27 cells (Fig. S4I), likely reflecting their distinct mutational background. Nevertheless, in both cell lines, IRF4 silencing specifically depleted E2F1 and some of its downstream targets, such as Cdc25A (data not shown and Fig. 4C-D).

E2F1/2²⁴ as well as E2F5¹⁴ have been identified as direct IRF4 target genes in lymphoid cancers including MM. To determine how IRF4 controls the E2F1 gene in IRE1-dependent MM cells, we performed a chromatin immunoprecipitation and sequencing (ChIP-seq) analysis for IRF4 and in parallel for Serine 2-phosphorylated, transcriptionally-engaged RNA-polymerase II (RNAPIIpS2) in AMO1 cells. As expected⁴⁵, IRF4 bound to intragenic (promoter) and intergenic (enhancer) regions, aligning with RNAPIIpS2 (Fig. 5A-B). IRF4 bound such regions within the PRDM1 locus (Fig. S5A), validating data quality. IRF4 occupancy was negligible at the RB1 locus (Fig. S5B), the transcription of which is not IRF4-dependent (Fig. 4C; Fig. S5B-RNAPIIpS2). In contrast, IRF4 silencing clearly attenuated RNAPIIpS2 binding to the E2F1 locus (Fig. 5A), indicating reduced E2F1 transcription. Interestingly, while IRF4 knockdown did not alter intragenic IRF4 occupancy, it significantly diminished both IRF4 and RNAPIIpS2 binding at the E2F1 gene-cluster proximal enhancers (Fig. 5A). A targeted transcription factor motif search using JASPAR identified not only IRF4 (and overlapping IRF5) motifs in these distal peaks, but also an E2F1 consensus sequence (Fig. 5A-bottom). Given that E2F1 exerts positive autoregulation^46–48, this finding supports the possibility that the non-coding region proximal to the E2F1 gene-cluster is an enhancer region that is IRF-dense and controls E2F1 transcription. IRF4 also bound intragenic regions of E2F2 and E2F3 (Fig. 5B), suggesting that it may control the wider E2F family with varying intensity. Similarly, IRF4 likely regulated Cdc25A gene expression via an IRF4-motif-containing intergenic region (Fig. S5C), suggesting an additional mode for IRF4 control of CDK2 activity. In stark contrast, IRF4 did not directly regulate UHRF1 (Fig. S5D), suggesting that the modulation of UHRF1 by IRF4 (Fig. 4C-D; Fig. S5D) may be indirectly relayed via E2F1, as previously shown^49–51.

IRF4 repletion in IRE1-deficient cells rescues E2F1 expression and cell proliferation

The regulation of IRF4 activity by IRE1 and the similarity between the cell cycle phenotypes of IRE1 and IRF4 silencing supported the possibility of a functional linkage between IRE1, IRF4 and cell cycle control. To more specifically affirm IRF4’s ability to mediate the non-enzymatic dependency on IRE1, we attempted to ectopically express IRF4 in AMO1 cells harboring inducible IRE1 silencing. However, constitutive IRF4 over-expression proved to be cytotoxic (not shown). To circumvent this, we opted for Dox-inducible expression of IRF4, which was better tolerated–provided that the ectopic IRF4 levels did not greatly exceed the endogenous amount (Fig. S6A-D). Dox-induced IRF4 re-expression successfully rescued shRNA silencing of endogenous IRF4 (Fig. S6A), providing functional validation of this strategy. Underscoring the importance of stoichiometry for IRF4 reconstitution, higher IRF4 expression did not improve the IRF4 rescue efficiency; rather, it was more lethal (Fig. S6A-B). Importantly, IRF4 repletion during IRE1 silencing was sufficient to rescue the proliferation of AMO1 cells (Fig. 6A-B). Similar to the dynamics in IRF4-deficient cells, small increases in IRF4 levels had a major impact on the extent of proliferative rescue in IRE1-deficient cells (Fig. S6C-D). Although IRF4 re-expression effectively restored growth for up to 72 h, a modest growth decline of up to 15% occurred beyond this time point (Fig. 6B). Of relevance, IRF4 re-expression delayed yet did not prevent caspase-3/7 activation during IRE1 silencing (Fig. S6E). Residual apoptosis was also evident from Annexin V/PI staining (Fig. S6F) and TUNEL assay (Fig. 6C). Therefore, it is likely that the non-rescued apoptosis limited the restoration of growth by re-expressed IRF4 after the 72 h timepoint. IRF4 repletion was also sufficient to rescue the previously documented decrease in DNA methylation due to IRE1 silencing¹² (Fig. 6D). Importantly, IRF4 re-expression enabled a substantial recovery of G1/S transition in IRE1-depleted cells (Fig. 6E; Fig. S6G). Furthermore, it rescued expression of E2F and the downstream targets UHRF1 and CDC25A and restored CDK2 activity (Fig. 6F-G; Fig. S6H). Together, these results demonstrate the ability of IRF4 to reverse all the primary features of the growth inhibition caused by IRE1 silencing, affirming IRF4’s pivotal functional importance in mediating non-enzymatic dependency on IRE1 in MM cells.

Our study identifies IRF4 as a crucial mechanistic conduit of the recently discovered non-enzymatic dependency on IRE1 in MM cells (Fig. 7). The regulation of IRF4 by IRE1, albeit multifaceted, did not involve other IRF4-targeting transcription factors. Although depletion of XBP1s during plasma-cell differentiation induces IRF4 expression²⁹, we found, surprisingly, that IRE1 can sustain IRF4 transcription independently of XBP1s. Notably, IRE1 is known to exert RIDD activity not only against IRF4 itself but also towards IRF4-suppressing miRNAs (-17-α and − 125b-5p)^52–54, which may explain why IRE1 silencing decreased IRF4’s mRNA stability alongside with its transcription. On the other hand, considering that IRF4 repletion was sufficient to rescue the decrease in DNA methylation conferred by IRE1 silencing, it is unlikely that IRE1 promotes IRF4 activity through this latter chromatin modification. Importantly, our results indicate that IRE1 non-enzymatically supports IRF4 mRNA expression by modulating IRF4’s chromatin-binding activity: IRE1 silencing significantly increased IRF4 phosphorylation on Ser114 and Ser270—a modification which we showed to be markedly inhibitory to chromatin binding by IRF4. Further mechanistic investigation revealed that IRF4 engages the E2F1 and CDC25A genes and promotes CDK2 activation to drive cell cycle progression (Fig. 7). Our observations lay the groundwork for future investigations aiming to identify the specific structural scaffolding role of IRE1 that controls IRF4’s phosphorylation state and thereby cell cycle progression. Phosphatases or kinases that have been previously linked to IRE1⁵⁵ and exert modulation of cell cycle progression or IRF4 expression could serve as interesting candidates.

Our transcriptomics studies demonstrated an extensive overlap of DEGs upon silencing of IRE1 or IRF4, explaining the remarkable similarity in most phenotypic features of the two silenced backgrounds. Moreover, our ChIP-seq results identified E2F1 as a novel, direct node of cell cycle regulation downstream of IRF4. Additionally, CDK2 activity proved to be a reliable functional marker for IRE1 or IRF4-mediated cell cycle regulation. The inactivation of CDK2 upon IRE1 or IRF4 silencing may be a culmination of extended E2F loss-of-function, which depletes drivers of DNA replication such as UHRF1, the loss of which causes hypersensitivity to DNA damage⁵⁶, as well as other regulators of G1/S transition. Additionally, IRF4 controls the central CDK2-phosphatase CDC25A⁵⁷. In turn, CDK2 inactivation facilitates G1 arrest or faster G2/M resolution by regulating the Anaphase Promoting Complex.

Crucially, IRF4 repletion in IRE1-silenced cells was sufficient to rescue the anti-proliferative phenotype of IRE1 disruption and reversed most of the consequences of IRE1 depletion, including E2F1 expression, CDK2 activity, and cell cycle progression. A notable exception was apoptosis; however, this difference may be expected, considering that IRE1 can suppress apoptosis via RIDD against DR5⁵⁸. Regardless, apoptosis is dispensable for the non-enzymatic cell cycle regulation by IRE1. Our results also indicate that the stoichiometry of IRF4 plays a critical role in its ability to restore its own function as well as the antiproliferative effects of IRE1 loss, in keeping with the ‘kinetic control’ model¹⁶, which suggests that IRF4 levels fine-tune diverse transcriptional programs by shifting the balance of IRF4 homo- and hetero-dimers.

The identification of IRF4 as a critical conduit of non-enzymatic IRE1-dependent cell cycle control in MM has important potential translational implications for the development of novel diagnostic and therapeutic strategies for both targets. It would also be valuable to explore the interplay between IRE1 and IRF4 in the context of immune cells and anti-cancer immunity. Recent evidence supports disrupting IRE1 as a potential strategy to enhance anti-tumor immunity^4,59, while cytotoxic immune responses become pivotal in refractory MM, where IRF4 becomes dispensable⁶⁰. To our knowledge, this study represents the first mechanistic investigation of cell cycle regulation by IRF4, thereby advancing the current functional understanding not only of IRE1 but also IRF4 in MM.

ACKNOWLEDGEMENTS

We thank Tao Sun for plasmid design, Rachana Pradhan, Ximo Pechuan-Jorge, Anatoly Belov, and Meena Choi for bioinformatic analyses, Yuxin Liang and Manching Ku for coordinating bulk RNA-sequencing, and Steffan Vartanian for methodological guidance regarding CDK2.

AUTHOR CONTRIBUTIONS

I.O. and A.A. conceived the study, formulated hypotheses, conceptualized, designed and/or performed experiments, interpreted results and wrote/edited the manuscript. B.D. and V.K. designed and performed ChIP-seq experiments as well as visualized and interpreted the data. V.C.P. performed proteomics studies and analyzed data. I.Z-G. and L.M.G. provided insightful data that supported the main study hypothesis. I.O. and S.M. designed and implemented genetic tools and generated cell lines. A.A., J.L. and Z.M, supervised the work.

CONFLICT OF INTEREST DISCLOSURES

All authors were employees of Genentech, Inc. during the performance of this work.

Walter, P. & Ron, D. The Unfolded Protein Response: From Stress Pathway to Homeostatic Regulation. Science 334, 1081–1086 (2011).
Wang, M. & Kaufman, R. J. The impact of the endoplasmic reticulum protein-folding environment on cancer development. Nat. Rev. Cancer 14, 581–597 (2014).
Hetz, C., Chevet, E. & Oakes, S. A. Proteostasis control by the unfolded protein response. Nat. Cell Biol. 17, 829–838 (2015).
Cubillos-Ruiz, J. R., Bettigole, S. E. & Glimcher, L. H. Tumorigenic and Immunosuppressive Effects of Endoplasmic Reticulum Stress in Cancer. Cell 168, 692–706 (2017).
Prischi, F., Nowak, P. R., Carrara, M. & Ali, M. M. U. Phosphoregulation of Ire1 RNase splicing activity. Nat Commun 5, 3554 (2014).
Thomas, A. L. et al. Decoding non-canonical mRNA decay by the endoplasmic-reticulum stress sensor IRE1α. Nat Commun 12, 7310 (2021).
Belyy, V., Zuazo-Gaztelu, I., Alamban, A., Ashkenazi, A. & Walter, P. Endoplasmic reticulum stress activates human IRE1α through reversible assembly of inactive dimers into small oligomers. eLife 11, e74342 (2022).
Maurel, M., Chevet, E., Tavernier, J. & Gerlo, S. Getting RIDD of RNA: IRE1 in cell fate regulation. Trends Biochem. Sci. 39, 245–254 (2014).
Harnoss, J. M. et al. Disruption of IRE1α through its kinase domain attenuates multiple myeloma. Proc. Natl. Acad. Sci. 116, 16420–16429 (2019).
Carrasco, D. R. et al. The Differentiation and Stress Response Factor XBP-1 Drives Multiple Myeloma Pathogenesis. Cancer Cell 11, 349–360 (2007).
Mimura, N. et al. Blockade of XBP1 splicing by inhibition of IRE1α is a promising therapeutic option in multiple myeloma. Blood 119, 5772–5781 (2012).
Zuazo-Gaztelu, I. et al. A nonenzymatic dependency on inositol-requiring enzyme 1 controls cancer cell cycle progression and tumor growth. bioRxiv 2023.11.22.567905 (2023) doi:10.1101/2023.11.22.567905.
Sciammas, R. et al. Graded Expression of Interferon Regulatory Factor-4 Coordinates Isotype Switching with Plasma Cell Differentiation. Immunity 25, 225–236 (2006).
Shaffer, A. L. et al. IRF4 addiction in multiple myeloma. Nature 454, 226–231 (2008).
Silva, N. S. D., Simonetti, G., Heise, N. & Klein, U. The diverse roles of IRF4 in late germinal center B‐cell differentiation. Immunol Rev 247, 73–92 (2012).
Ochiai, K. et al. Transcriptional Regulation of Germinal Center B and Plasma Cell Fates by Dynamical Control of IRF4. Immunity 38, 918–929 (2013).
Tang, T. F. et al. Regulatory network of BLIMP1, IRF4, and XBP1 triad in plasmacytic differentiation and multiple myeloma pathogenesis. Cell Immunol 380, 104594 (2022).
Shaffer, A. L. et al. Blimp-1 Orchestrates Plasma Cell Differentiation by Extinguishing the Mature B Cell Gene Expression Program. Immunity 17, 51–62 (2002).
Wu, J. et al. A p38α-BLIMP1 signalling pathway is essential for plasma cell differentiation. Nat. Commun. 13, 7321 (2022).
Shaffer, A. L. et al. XBP1, Downstream of Blimp-1, Expands the Secretory Apparatus and Other Organelles, and Increases Protein Synthesis in Plasma Cell Differentiation. Immunity 21, 81–93 (2004).
Tellier, J. et al. Blimp-1 controls plasma cell function through the regulation of immunoglobulin secretion and the unfolded protein response. Nat. Immunol. 17, 323–330 (2016).
Reimold, A. M. et al. Plasma cell differentiation requires the transcription factor XBP-1. Nature 412, 300–307 (2001).
Shapiro-Shelef, M. et al. Blimp-1 Is Required for the Formation of Immunoglobulin Secreting Plasma Cells and Pre-Plasma Memory B Cells. Immunity 19, 607–620 (2003).
Fedele, P. L. et al. The transcription factor IRF4 represses proapoptotic BMF and BIM to licence multiple myeloma survival. Leukemia 35, 2114–2118 (2021).
Heintel, D. et al. High Expression of MUM1/IRF4 mRNA Is Associated with a Shorter Overall Survival in Patients with Multiple Myeloma. Blood 106, 1550 (2005).
Iida, S. et al. Deregulation of MUM1/IRF4 by chromosomal translocation in multiple myeloma. Nat. Genet. 17, 226–230 (1997).
Yoshida, S. et al. Detection of MUM1/IRF4-IgH fusion in multiple myeloma. Leukemia 13, 1812–1816 (1999).
Agnarelli, A., Chevassut, T. & Mancini, E. J. IRF4 in multiple myeloma—Biology, disease and therapeutic target. Leuk. Res. 72, 52–58 (2018).
Hu, C. A., Dougan, S. K., McGehee, A. M., Love, J. C. & Ploegh, H. L. XBP‐1 regulates signal transduction, transcription factors and bone marrow colonization in B cells. EMBO J. 28, 1624–1636 (2009).
Simoes, R. de M. et al. Genome-scale functional genomics identify genes preferentially essential for multiple myeloma cells compared to other neoplasias. Nat. Cancer 4, 754–773 (2023).
Quwaider, D. et al. RNA sequencing identifies novel regulated IRE1-dependent decay targets that affect multiple myeloma survival and proliferation. Exp Hematology Oncol 11, 18 (2022).
Boddicker, R. L. et al. The oncogenic transcription factor IRF4 is regulated by a novel CD30/NF-κB positive feedback loop in peripheral T-cell lymphoma. Blood 125, 3118–3127 (2015).
Bjorklund, C. C. et al. Rate of CRL4CRBN substrate Ikaros and Aiolos degradation underlies differential activity of lenalidomide and pomalidomide in multiple myeloma cells by regulation of c-Myc and IRF4. Blood Cancer J. 5, e354–e354 (2015).
Lehtonen, A. et al. Differential Expression of IFN Regulatory Factor 4 Gene in Human Monocyte-Derived Dendritic Cells and Macrophages. J. Immunol. 175, 6570–6579 (2005).
Low, M. S. Y. et al. IRF4 Activity Is Required in Established Plasma Cells to Regulate Gene Transcription and Mitochondrial Homeostasis. Cell Reports 29, 2634-2645.e5 (2019).
Fedele, P. L. et al. Loss of IRF4 Results in Multiple Myeloma Cell Apoptosis through the Transcriptional up-Regulation of the BH3-Only Proteins Bmf and BIM. Blood 134, 3103–3103 (2019).
Sørensen, C. S. et al. A Conserved Cyclin-Binding Domain Determines Functional Interplay between Anaphase-Promoting Complex–Cdh1 and Cyclin A-Cdk2 during Cell Cycle Progression. Mol Cell Biol 21, 3692–3703 (2001).
Mitra, J. & Enders, G. H. Cyclin A/Cdk2 complexes regulate activation of Cdk1 and Cdc25 phosphatases in human cells. Oncogene 23, 3361–3367 (2004).
Boer, L. D. et al. Cyclin A/cdk2 coordinates centrosomal and nuclear mitotic events. Oncogene 27, 4261–4268 (2008).
Chung, J. H. & Bunz, F. Cdk2 Is Required for p53-Independent G2/M Checkpoint Control. PLoS Genet. 6, e1000863 (2010).
Oakes, V. et al. Cyclin A/Cdk2 regulates Cdh1 and claspin during late S/G2 phase of the cell cycle. Cell Cycle 13, 3302–3311 (2014).
Gu, Y., Rosenblatt, J. & Morgan, D. O. Cell cycle regulation of CDK2 activity by phosphorylation of Thr160 and Tyr15. EMBO J. 11, 3995–4005 (1992).
Hiromura, K., Pippin, J. W., Blonski, M. J., Roberts, J. M. & Shankland, S. J. The subcellular localization of cyclin dependent kinase 2 determines the fate of mesangial cells: role in apoptosis and proliferation. Oncogene 21, 1750–1758 (2002).
Mondala, P. K. et al. Selective antisense oligonucleotide inhibition of human IRF4 prevents malignant myeloma regeneration via cell cycle disruption. Cell Stem Cell 28, 623-636.e9 (2021).
Li, N. et al. Multiple myeloma risk variant at 7p15.3 creates an IRF4-binding site and interferes with CDCA7L expression. Nat. Commun. 7, 13656 (2016).
Xu, J. et al. The Role of Crosstalk between AR3 and E2F1 in Drug Resistance in Prostate Cancer Cells. Cells 9, 1094 (2020).
Neuman, E., Flemington, E. K., Sellers, W. R. & Kaelin, W. G. Transcription of the E2F-1 gene is rendered cell cycle dependent by E2F DNA-binding sites within its promoter. Mol. Cell. Biol. 14, 6607–6615 (1994).
Gao, S. et al. Androgen Receptor Tumor Suppressor Function Is Mediated by Recruitment of Retinoblastoma Protein. Cell Rep. 17, 966–976 (2016).
Abbady, A.-Q. et al. TCR pathway involves ICBP90 gene down-regulation via E2F binding sites. Biochem. Pharmacol. 70, 570–579 (2005).
Unoki, M., Nishidate, T. & Nakamura, Y. ICBP90, an E2F-1 target, recruits HDAC1 and binds to methyl-CpG through its SRA domain. Oncogene 23, 7601–7610 (2004).
Mousli, M. et al. ICBP90 belongs to a new family of proteins with an expression that is deregulated in cancer cells. Br. J. Cancer 89, 120–127 (2003).
Lerner, A. G. et al. IRE1α Induces Thioredoxin-Interacting Protein to Activate the NLRP3 Inflammasome and Promote Programmed Cell Death under Irremediable ER Stress. Cell Metab. 16, 250–264 (2012).
Ahmadiany, M., Alavi-Samani, M., Hashemi, Z., Moosavi, M. A. & Rahmati, M. The Increased RNase Activity of IRE1α in PBMCs from Patients with Rheumatoid Arthritis. Adv. Pharm. Bull. 9, 505–509 (2019).
Huang, J. et al. IRE1α inhibition attenuates neuronal pyroptosis via miR-125/NLRP1 pathway in a neonatal hypoxic-ischemic encephalopathy rat model. J. Neuroinflammation 17, 152 (2020).
Goupil, S. L., Laprade, H., Aubry, M. & Chevet, E. Exploring the IRE1 interactome: From canonical signaling functions to unexpected roles. J. Biol. Chem. 300, 107169 (2024).
Mancini, M., Magnani, E., Macchi, F. & Bonapace, I. M. The multi-functionality of UHRF1: epigenome maintenance and preservation of genome integrity. Nucleic Acids Res. 49, gkab293- (2021).
Sexl, V. et al. A rate limiting function of cdc25A for S phase entry inversely correlates with tyrosine dephosphorylation of Cdk2. Oncogene 18, 573–582 (1999).
Lu, M. et al. Opposing unfolded-protein-response signals converge on death receptor 5 to control apoptosis. Science 345, 98–101 (2014).
Guttman, O. et al. Antigen-derived peptides engage the ER stress sensor IRE1α to curb dendritic cell cross-presentation. J Cell Biol 221, e202111068 (2022).
Davis, L. N. et al. MYC Inhibition Potentiates CD8+ T Cells Against Multiple Myeloma and Overcomes Immunomodulatory Drug Resistance. Clin. Cancer Res. (2024) doi:10.1158/1078-0432.ccr-24-0256.

Yes there is potential conflict of interest. All authors were employees of Genentech, Inc. during the performance of this work

Download PDF

Editorial decision: revise
20 Aug, 2024
Review #1 received at journal
17 Aug, 2024
Review #2 received at journal
07 Aug, 2024
Reviewer #2 agreed at journal
18 Jul, 2024
Reviewer #1 agreed at journal
12 Jul, 2024
Reviewers invited by journal
08 Jul, 2024
Editor assigned by journal
29 May, 2024
Submission checks completed at journal
29 May, 2024
First submitted to journal
29 May, 2024

You are reading this latest preprint version

IRF4 mediates non-enzymatic dependency on IRE1 in Multiple Myeloma cells

Status:

Version 1

Abstract

Figures

INTRODUCTION

METHODS

Gene expression and analysis

Nuclear Run-On

Confluence

TUNEL assay

Subcellular Fractionation

Co-immunoprecipitations

Cell proliferation

DNA replication

ChIP-sequencing

DNA methylation

RESULTS

Silencing of IRE1 but not XBP1 downregulates IRF4

IRE1 silencing attenuates IRF4 activity

IRF4 silencing recapitulates the anti-proliferative phenotype of IRE1 knockdown

IRE1 and IRF4 regulate a highly overlapping set of cell cycle genes

IRF4 directly controls E2F1 gene transcription

IRF4 repletion in IRE1-deficient cells rescues E2F1 expression and cell proliferation

DISCUSSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1