BRIEF RESEARCH REPORT published: 10 February 2021 doi: 10.3389/fimmu.2021.634797 CDK11 Promotes Cytokine-Induced Apoptosis in Pancreatic Beta Cells Independently of Glucose Concentration and Is Regulated by Inflammation in the NOD Edited by: Mouse Model Bisheng Zhou, University of Illinois at Chicago, Ester Sala 1,2, Celia Vived 1,2†, Júlia Luna 1,2†, Noemı́ Alejandra Saavedra-Ávila 1,2†, United States Upasana Sengupta 1,2†, A. Raúl Castaño 3, Sabrina Villar-Pazos 4,5, Laura Haba 6, Reviewed by: Joan Verdaguer 1,2, Ana B. Ropero 7, Thomas Stratmann 8, Javier Pizarro 9,10,11, Xingrong Du, Manuel Vázquez-Carrera 9,10,11, Angel Nadal 4,12, Jill M. Lahti 5‡ and Conchi Mora 1,2* St. Jude Children’s Research Hospital, 1 Immunology Unit, Department of Experimental Medicine, Faculty of Medicine, University of Lleida, Lleida, Spain, 2 Institut de United States Feng Xinyu, Recerca Biomèdica Lleida (IRB-LLeida), Lleida, Spain, 3 Departament of Cell Biology, Physiology and Immunology, Shanghai Institute of Biochemistry and Autonomous University of Barcelona, Barcelona, Spain, 4 Instituto de Investigación, Desarrollo e Innovación en Biotecnologı´a Cell Biology (CAS), China Sanitaria de Elche, IDiBE, Universidad Miguel Hernandez, Elche, Spain, 5 Department of Tumor Cell Biology, St. Jude Children’s Research Hospital, Memphis, TN, United States, 6 Experimental Diabetes Laboratory, Institute for Biomedical *Correspondence: Research August Pi i Sunyer (IDIBAPS), Barcelona, Spain, 7 Instituto de Bioingenierı´a, Universidad Miguel Hernández, Elche, Conchi Mora Spain, 8 Department of Cell Biology, Physiology and Immunology, Faculty of Biology, University of Barcelona, Barcelona, conchi.mora@udl.cat Spain, 9 Department of Pharmacology, Toxicology and Therapeutic Chemistry, Faculty of Pharmacy and Food Sciences and † These authors have contributed Institute of Biomedicine (IBUB), University of Barcelona, Barcelona, Spain, 10 Spanish Biomedical Research Center in equally to this work Diabetes and Associated Metabolic Diseases (CIBERDEM)—Instituto de Salud Carlos III, Madrid, Spain, 11 Pediatric ‡ Research Institute, Hospital Sant Joan de Déu, Esplugues de Llobregat, Spain, 12 Diabetes and Associated Metabolic Deceased Disorders CIBERDEM, Universidad Miguel Hernández de Elche, Elche, Spain Specialty section: This article was submitted to Background: Pancreatic islets are exposed to strong pro-apoptotic stimuli: inflammation Nutritional Immunology, and hyperglycemia, during the progression of the autoimmune diabetes (T1D). We found a section of the journal Frontiers in Immunology that the Cdk11(Cyclin Dependent Kinase 11) is downregulated by inflammation in the T1D Received: 28 November 2020 prone NOD (non-obese diabetic) mouse model. The aim of this study is to determine the Accepted: 07 January 2021 role of CDK11 in the pathogenesis of T1D and to assess the hierarchical relationship Published: 10 February 2021 between CDK11 and Cyclin D3 in beta cell viability, since Cyclin D3, a natural ligand for Citation: Sala E, Vived C, Luna J, CDK11, promotes beta cell viability and fitness in front of glucose. Saavedra-Ávila NA, Sengupta U, Methods: We studied T1D pathogenesis in NOD mice hemideficient for CDK11 (N-HTZ), Castaño AR, Villar-Pazos S, Haba L, Verdaguer J, Ropero AB, Stratmann T, and, in N-HTZ deficient for Cyclin D3 (K11HTZ-D3KO), in comparison to their respective Pizarro J, Vázquez-Carrera M, controls (N-WT and K11WT-D3KO). Moreover, we exposed pancreatic islets to either Nadal A, Lahti JM and Mora C (2021) CDK11 Promotes Cytokine-Induced pro-inflammatory cytokines in the presence of increasing glucose concentrations, or Apoptosis in Pancreatic Thapsigargin, an Endoplasmic Reticulum (ER)-stress inducing agent, and assessed Beta Cells Independently apoptotic events. The expression of key ER-stress markers (Chop, Atf4 and Bip) was of Glucose Concentration and Is Regulated by Inflammation also determined. in the NOD Mouse Model. Front. Immunol. 12:634797. Results: N-HTZ mice were significantly protected against T1D, and NS-HTZ pancreatic doi: 10.3389/fimmu.2021.634797 islets exhibited an impaired sensitivity to cytokine-induced apoptosis, regardless of Frontiers in Immunology | www.frontiersin.org 1 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 glucose concentration. However, thapsigargin-induced apoptosis was not altered. Furthermore, CDK11 hemideficiency did not attenuate the exacerbation of T1D caused by Cyclin D3 deficiency. Conclusions: This study is the first to report that CDK11 is repressed in T1D as a protection mechanism against inflammation-induced apoptosis and suggests that CDK11 lies upstream Cyclin D3 signaling. We unveil the CDK11/Cyclin D3 tandem as a new potential intervention target in T1D. Keywords: apoptosis, beta cell, CDK11, cyclin D3, inflammation, type 1 diabetes, glucose, insulin INTRODUCTION CDK11p58 and CDKp110 are activated by Caspase-3 mediated cleavage into CDK11p46, in response to pro-apoptotic stimuli, In T1D, the proinflammatory milieu has been reported to impair which amplifies the apoptotic process through phosphorylation of the functioning of beta cells and induce apoptosis (1–4). One of regulating proteins such as the eukaryotic Initiation Factor 3f the best animal models of T1D is the NOD (Non-Obese (eIF3f) (24–26). Moreover, CDK11p110 cleavage by Caspase 3 Diabetic) mouse model (5). A series of reports have focused on releases CDK11 p60 , a regulatory moiety that mediates studying the effects of the main proinflammatory cytokines alone cytochrome c release from mitochondria and apoptosis (27). or in combination on the proteome/transcriptome of Since islet beta cells undergo cytokine-induced apoptosis in T1D insulinoma/islet cells in vitro (6–9). Those reports included a (28, 29), CDK11 cleavage by Caspase-3 is expected to exacerbate discrete combination of cytokines that partially resembles the apoptosis in beta cells. Therefore, since beta cell replication has complex inflammatory niche hosting the pancreatic islet; and been reported to occur in response to the autoimmune attack (30), most studies used insulinoma tumor cell lines. Therefore, we mitotic beta cells would express CDK11p58, which, in turn, would explored the in vivo effects of inflammation on the transcriptome amplify apoptosis. of NOD endocrine cells by applying cDNA microarray Moreover, beta cells synthesize insulin upon demand; this technology (10) and found that the mRNA expression of submits them to Endoplasmic Reticulum (ER)-stress and the Cdk11 (cyclin-dependent kinase 11) was downregulated in the Unfolded Protein Response (UPR). In T1D, insidious insulitis pancreatic endocrine cells (PEC) from NOD female mice upon and hyperglycemia induce ER-stress and UPR activation, leading inflammation. CDK11 is a ubiquitous PITSLRE cyclin- to beta cell apoptosis through the upregulation of CHOP (C/EBP dependent kinase with two gene products: p58 and p130 in homologous protein), a key transcription factor responsible for mice, and p58 and p110 in humans (11). The p58 gene product ER-stress induced-apoptosis through Bcl-2 inhibition (31, 32). In originates from a differential ribosome entry through an IRES T1D, both, inflammation and ER-stress crosstalk and feed-back (Internal Ribosomal Entry Sequence) in the mRNA of Cdk11 each other positively, resulting in exacerbation of beta cell during its translation in mitosis (12). death (33). CDK11p110 binds to cyclin L, which is the main CDK11p110 In this study, we aim to explore whether there is a causal regulatory subunit, and several pre-mRNA splicing factors to relationship between the downregulation of Cdk11 in beta cells regulate pre-mRNA transcription and processing in proliferating and the onset of T1D. Moreover, we aim to determine whether the cells (13–17). CDK11p110 is expressed during all cell cycle phases simultaneous downregulation of both, Cdk11 and Cyclin D3 at the (11), while CDK11p58 is only expressed during mitosis (18). onset of T1D has a pathophysiological meaning. To this end, we CDK11p58 is required during mitosis for the maintenance of the developed CDK11-hemideficient NOD mice and assessed their sister chromatid cohesion and participates in cytokinesis and phenotype in terms of beta cell viability, autoimmune repertoire, mitosis termination (18, 19). The activity of CDK11p58 depends and autoimmune diabetes. Furthermore, we developed NOD mice on its interaction with Cyclin D3 to promote its kinase activity, that were hemideficient in CDK11 and deficient in Cyclin D3 and migrate into the nucleus and regulate the cell cycle progression assessed their phenotype in terms of T1D onset. through the G2/M phase (20, 21). The CDK11-deficiency in mice is embryonically lethal, while CDK11-hemideficient mice are viable (22). Furthermore, CDK11p58 is associated with the downregulation MATERIALS AND METHODS of the expression of Bcl-2 during apoptosis (23), and both Mice Mice were housed under specific pathogen-free (SPF) conditions. Abbreviations: CDK11, Cyclin dependent Kinase 11; T1D, Type 1 Diabetes; PEC, All animal experimentation procedures performed in this study Pancreatic Endocrine Cells; NOD, Non Obese Diabetic; qRT-PCR, quantitative were overseen and approved by the Institutional Ethical Real Time-Polymerase Chain Reaction; WT, Wild Type; HTZ, Heterozygous; Committee for Animal Experimentation of the University of UPR, Unfolded Protein response; ER, Endoplasmic Reticulum; Bip, Binding- immunoglobulin protein; Atf4, Activating Transcription Factor 4; Chop, C/EBP Lleida (CEEA) in accordance with European and U.S.A. Homologous Protein. regulations on Animal Experimentation. All results involving Frontiers in Immunology | www.frontiersin.org 2 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 animal experiments were obtained with female mice. Mice marker; Dako, Glostrup, Denmark), phycoerythrin-conjugated heterozygous for the CDK11-deficiency (HTZ) with the mixed CD45 (hematopoietic marker; BD), and 7AAD (cell death 129SvxC57BL/6 genetic background (22) were backcrossed to marker; BD) antibodies. The data were acquired using a the NOD background for 15 generations to identify the Idd FACSCanto II flow cytometer (BD). susceptibility loci (34). The null mutation of Cdk11 was detected by PCR (22) using the CPrev: CAAGAGAAGCCTGA GCAAATAG and the mp70: GAGATACTCTTTACAT TUNEL Assays Determination of late (TUNEL) apoptotic events was performed GCCAACC primers. Cyclin D3 deficient mice in the NOD using the In Situ Cell Death Detection Kit, Fluorescein (ref, background (D3KO) were genotyped by PCR as previously 11684795910, Roche-Sigma). Insulin was detected with anti- reported (10). insulin antibody (A0564, Dako), and nuclear staining was performed with Hoechst 33342 (B2261, Sigma). Quantitative Real-Time PCR (qRT-PCR) See Supplementary Methods. Isolation of Islet-Infiltrating Lymphocytes (IILS) Assessment of Pancreatic Infiltration The islet-infiltrating lymphocytes (IILs) were isolated by Immunohistochemical analysis of pancreatic infiltration was trypsinization of the previously isolated islets using 500 µl of performed on paraffin sections by hematoxylin & eosin trypsin/EDTA (Lonza, Verviers, Belgium) for 5 min at 37°C to counterstaining (Sigma-Aldrich, St. Louis, MO, USA) for the obtain single cell suspensions. infiltration studies (Leica, Wetzler, Germany). To score the islet infiltration, the following values were assigned: 0 for non- infiltrated islets, 1 for peri-insulitic lesions, 2 for intra-insular Assessment of Immune Cell Subsets by insulitis with less than 50% of the islet area infiltrated, and 3 for Flow Cytometry insulitis with more than 50% of the islet area infiltrated. A See Supplementary Methods. minimum of eight sections per mouse at four different levels (>150 islets per genotype) were examined. For each individual DNA Fragmentation Assays mouse, the following formula was applied to calculate the Handpicked islets were cultured in supplemented DMEM in the infiltration score, and then, the mean ± SEM of each genotype presence or absence of a cytokine cocktail containing IFN-g was plotted: 100U/ml, TNF-a 1000 U/ml, IL1b 100U/ml, afterwards islets (X islets of score 0  0) + (Y islets of score 1  1) + (Z islets of score 2  2) + (W islets of score 3  3) were harvested and lysed by hypotonic shock, stained with (X + Y + Z + W) Propidium Iodide (Sigma) and acquired by flow cytometry to measure nuclei fragmentation (% DSubG1) as previously reported (35). Results were calculated as percentage on Diabetes Assessment, Adoptive Transfer increment in SubG1 induction by cytokines as follows: Experiments, and Pancreatic Islet Isolation All these procedures were performed as previously ( % SubG1 events upon exposure to cytokines) − ( % SubG1 events in basal condition)  100 ( % SubG1 in basal condition) described (34). Magnetic Cell Separation of Pancreatic ER-Stress Induction Endocrine Cells (PECs: CD45- Fraction) Handpicked islets from 5–7-week-old female mice donors were Islets were isolated and disaggregated by trypsin digestion, and cultured in supplemented DMEM containing 11mM glucose, the resulting cell suspension was incubated with an anti-mouse either in the presence or absence of 1mM Thapsigargin (Sigma) CD45-phycoerythrin (PE) antibody (Becton Dickinson (BD), for the indicated time (either 24 h for ER-stress marker induction CA, USA), subsequently washed and incubated with the anti- or 96 h for subG1 measurements). PE antibody coupled to magnetic beads, and submitted to AUTOMACS magnetic negative selection according to the manufacturer’s instructions (AUTOMACS; Miltenyi Biotec, Statistical Analysis Mann-Whitney U-tests were performed to evaluate the Bergisch Gladbach, Germany). differences between pairs of groups. The log-rank test was applied to assess the difference between the diabetes Western Blot Assays cumulative incidence curves. The one-way ANOVA test was See Supplementary Methods. used to assess the differences in TUNEL studies. The two-way ANOVA test was used to analyze the differences in infiltration Islet Cell Staining score studies, and SubG1 studies respectively. All analyses The isolated islets were trypsinized, fixed and permeabilized prior to were performed using the GraphPad Prism (v5.0a) statistical the cellular staining with the anti-mouse Glut-2 (beta-cell marker; package. The threshold for statistical significance was set R&D Minneapolis, MN, USA), anti-mouse Ki67 (proliferation at 0.05. Frontiers in Immunology | www.frontiersin.org 3 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 RESULTS NOD/SCID females (0.223 ± 0.047 vs. 0.674 ± 0.09, respectively). CDK11p130 protein expression was also reduced in PECs from Cdk11 Is Downregulated in b Cells by NOD mice (Figures 1B, C). Nevertheless, CDK11p58 protein Autoimmune Insulitis in a Dose-Dependent levels were higher in NOD PECs compared to those from NOD/ Manner Without Impairing b Cell SCID, underscoring the role of inflammation in post- Proliferation transcriptional regulation of CDK11 (Figures 1B, C). At 11 weeks of age, the NOD female pancreatic islets are heavily We further aimed to determine whether the mRNA infiltrated, while those from age-matched NOD/SCID female expression of Cdk11 was inversely correlated with the severity mice are insulitis-free, because they lack lymphocytes (36, 37). of the leukocyte infiltration into the islets. Therefore, we Using the cDNA microarray technology, we observed that the performed qRT-PCR to detect Cdk11 in the PECs of the NOD/ mRNA expression levels of Cdk11 were 2.4-fold lower in the SCID islets from the 11-week-old female mice that had been PECs from the 11-week-old pre-diabetic NOD female mice adoptively transferred with increasing amounts of total compared to those from age-matched NOD/SCID female mice leukocytes from prediabetic NOD female donors 2 weeks (see Supplementary Table 1). We confirmed the microarray before the islet isolation. We observed a significant inverse results by performing quantitative real time PCR (qRT-PCR) relationship between Cdk11 expression levels in PECs and the assays (Figure 1A) and found that the PECs from the 11-week- amount of transferred leukocytes (Figure 1D). old NOD females exhibited a threefold reduction in the mRNA To address whether the downregulation of Cdk11 mRNA expression levels of Cdk11 compared to those from age-matched prior to T1D onset altered the NOD beta cell replicative activity, A B C D E FIGURE 1 | Cdk11 is downregulated by the islet infiltration at the mRNA level, without affecting the proliferative activity of b cells in NOD mice. CDK11p58 protein expression is enhanced, while CDK11p130 is downregulated, compared to NOD/SCID in NOD PECs (A). Pancreatic islet cells from either 11-week-old NOD female mice (n = 6) or 11-week-old NOD/SCID female mice (n = 10) were extracted, and the CD45- cell subset (PECs) was selected by magnetic sorting to perform a qRT- PCR analysis to assess the expression of cdk11. (B, C) Pancreatic islet cells from either 11-week-old NOD female mice (n = 16 mice; 1212 islets) or 11-week-old NOD/SCID female mice (n = 15 mice; 1586 islets) were extracted, and the CD45- cell subset (PECs) was selected by magnetic sorting to perform western blot analysis to detect both CDK11p58 and CDK11p130 proteins. (C) CDK11p58 and CDK11p130 expression values were respectively normalized to b-actin levels in PECs from either NOD (black columns) or NOD/SCID (white columns). (D) Assessment of the expression levels of cdk11 in PECs from NOD/SCID mice that were adoptively transferred with different amounts (×106) of total splenocytes (TSC (i.e., total spleen cells that were previously depleted of red blood cells) from 8-week-old NOD female donors; the numbers of recipient mice per condition 2 weeks before the islet extraction at 11 weeks of age were as follows: 0, n = 5; 1, n = 5; 3, n = 6; 5, n = 6, and for 10, n = 5). The mRNA expression of cdk11 in the PECs was measured by qRT-PCR. All comparisons were related to TSC: 0. A linear regression was performed according to the equation y=-0.036367x+1.106, resulting in p = 0.0177. (E) The percentages of the proliferating beta cells (CD45- Glut2+ Ki67+) were measured by flow cytometry. Pancreatic islets were isolated from 11-week-old NOD (n = 6) and NOD/SCID (n = 4) mice, trypsinized and stained for CD45, Glut2, and Ki67). The data are represented as the mean ± SEM in (A, E); in (D), individual values are plotted. In (A) and (E), the value of the NOD/SCID group was used as a reference. *p ≤ 0.05, **p ≤ 0.01. Frontiers in Immunology | www.frontiersin.org 4 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 we assessed the beta cell proliferation rates using Ki67 staining of revealed a differential translational regulation in CDK11 GLUT-2+ CD45- (beta) cells from 11-week-old NOD and NOD/ hemideficency favoring CDK11p130 over CDK11p58. SCID mice and analyzed these rates by flow cytometry (Figure We monitored the spontaneous onset of T1D in the N-HTZ 1E). We found no differences in the beta cell proliferation rates and N-WT littermate mice. In parallel we assessed the between the groups. spontaneous incidence of diabetes in NS-HTZ compared to that in NS-WT littermates, in case an intrinsic defect in beta CDK11 Hemideficiency Protects N-HTZ cell generation was related to CDK11 hemideficiency. Mice Against Diabetes Without Affecting We found that the N-HTZ mice exhibited a significantly the Diabetogenicity of the Autoimmune delayed kinetics of the disease and lower cumulative incidence of Repertoire diabetes (70.8% vs. 81.4% respectively) compared to the N-WT To ascertain whether the mRNA downregulation of Cdk11 at the littermates (Figure 2D). None of the NS-HTZ mice developed onset of T1D was causally related to beta cell death in vivo, we spontaneous diabetes (see Supplementary Table 2), which developed a NOD mouse model that was genetically evidenced that no major intrinsic defect in beta cell mass and/ hemideficient in CDK11 (N-HTZ) (Figure 2A), since the or function is associated to the CDK11 hemideficiency. homozygous CDK11 deficiency is embryonically lethal (22). To determine whether the protection against T1D shown by We also obtained NOD/SCID mice that were hemideficient in the N-HTZ mice was due to a milder insulitic attack compared to CDK11 (NS-HTZ) and checked protein expression levels in NS- the N-WT mice, we first scored the islet leukocyte infiltration HTZ versus NS-WT islet cells. CDK11p58 protein levels were associated with both genotypes at three different time points reduced while CDK11p130 reduction was minimal in NS-HTZ prior to the onset of diabetes (5, 8, and 10 weeks of age). compared to NS-WT mice (Figures 2B, C). This outcome Dramatic differences were found in the severity of the insulitic A B C D E F G H FIGURE 2 | CDK11 hemideficiency protects against the onset of T1D without altering the diabetogenicity of the autoimmune repertoire. (A) Pancreatic islet cells were isolated from 6-week-old NS-WT and NS-HTZ (n = 5) mice. qRT-PCR analysis was performed to assess the mRNA expression of cdk11. (B) Pancreatic islet cells were isolated from 35-week-old NS-WT and NS-HTZ mice, each lane contains protein corresponding to pooled islets from two mice (130 islets per lane). Western blot analysis was performed to detect CDK11p58, CDK11p130, cyclin D3. (C) CDK11p58, CDK11p130, and Cyclin D3 expression values were respectively normalized to b-actin levels in pancreatic islets from either NS-WT (black columns) or NS-HTZ (white columns). The cumulative incidence of spontaneous diabetes in N-WT (n = 43) and N-HTZ (n = 48) mice was assessed. (E) Pancreata from 5-, 8-, and 10-week-old mice (minimum of n = 9 mice and maximum n = 16 per experimental group) were extracted, and the leukocyte infiltration levels were scored. (F) 4- to 5-week-old NOD/SCID recipient (n = 9 recipient mice for N-WT leukocyte donors; and n = 10 recipient mice for N-HTZ leukocyte donors) mice were adoptively transferred with 10 million of total cells from pancreatic lymph nodes (PLNs) from either 9-week-old N-WT or N-HTZ donors. The cumulative incidence of adoptively transferred diabetes is plotted. (G) Percentage of T regulatory lymphocytes (CD3+CD4+CD25+FoxP3+) in the PLNs (n = 12) is plotted. (H) Percentage of T regulatory lymphocytes (CD3+CD45+CD25+FoxP3+) in the lymphocytic infiltrate of the pancreatic islets [N-WT (n = 7) and N-HTZ (n = 6)] is shown. In (G) and (H), 7-week-old N-WT and N-HTZ mice were used. The data are shown as the mean ± SEM in (A, E, G, H); as the percentages of cumulative diabetic mice at a particular time point in (D, F). *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. Frontiers in Immunology | www.frontiersin.org 5 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 attack between both genotypes independently of the age tested, inflammation in T1D (30), would enhance beta cell apoptosis. being the infiltration score significantly lower in the N-HTZ Therefore, the physiological downregulation of CDK11 should genotype compared to the N-WT (Figure 2E). have a protective effect against T1D onset. We further assessed the diabetogenicity of the N-HTZ In order to confirm the hypothesis that CDK11 leukocyte repertoire compared to that of N-WT littermates by hemideficiency protects from cytokine-induced apoptosis, we performing adoptive transfer of diabetes into NOD/SCID cultured pancreatic islets from either wild type (NS-WT) or recipients. In total, 10 million cells isolated from Pancreatic CDK11 hemideficient (NS-HTZ) NODSCID female mice in the Lymph Nodes (PLNs) from either 9-week-old N-HTZ or N-WT presence of a proinflammatory cytokine cocktail (TNFa+IL- donors were transferred into NOD/SCID recipient mice (Figure 1b+IFNg) at either low (5mM), medium (11mM) or high 2F). The onset of the adoptively transferred diabetes was (25mM) glucose concentration, respectively, for 24 h. monitored, and no difference between both genotypes was Subsequently, the induction of DNA fragmentation was observed in their diabetogenic capacity, which was supported assessed by flow cytometry as a measurement of late apoptotic by the unaltered representation of Tregs in the PLNs and events (Figures 3B–D). The induction of apoptosis due to the insulitic infiltrate from N-HTZ females compared to the N- exposure to pro-inflammatory cytokines was significantly lower WT genotype (Figures 2G, H respectively). We also evaluated in the NS-HTZ group compared to the NS-WT control group, the frequency of effector and naïve T cells in PLNs and in the independently of the glucose concentration. Therefore, an insulitic infiltrates; the in vitro proliferative activity and impairment in the pro-inflammatory trigger of apoptosis does apoptotic rates of PLN cells in front of stimulation by islet account for the protection exerted by the CDK11 hemideficiency autoantigens (Supplementary Figure 1); and no major against beta cell apoptosis. This finding highlighted the role of difference was found between both genotypes. Moreover, in CDK11 in promoting apoptosis once the proinflammatory vitro Th1 response was not impaired in PLN cells from N- milieu pervades the pancreatic islets during the progression of HTZ mice (Supplementary Figure 2). the autoimmune attack. Therefore, the target tissue itself, the beta cell compartment, Since CDK11p58 has been proven to be essential for sister and no an impaired autoimmune repertoire in the N-HTZ mice, chromatid cohesion (18, 19), we investigated whether the should account for the protection against diabetes exerted by the hemideficiency in the CDK11 resulted in impaired beta cell CDK11 hemideficiency. NOD macrophages are less efficient in proliferation too. The percentage of proliferating beta cells engulfing apoptotic cells leading to inflammation (38). The (CD45-Glut-2+Ki67+) was assessed by flow cytometry (Figure release of smaller amounts of beta cell antigens by N-HTZ 3E), but, surprisingly, no differences were found in the beta cell islets due the lower apoptosis rates, would not sustain islet proliferation rate between the genotypes, which implied that the inflammation as is fostered by the inefficient scavenger role that CDK11 plays in beta cell viability is cell- capability of N-WT macrophages, leading, hence, to a lower cycle-independent. infiltration score in N-HTZ. CDK11 Is Involved in the Upregulation of CDK11 Promotes Cytokine-Induced Beta Atf4 and Chop in Beta Cell Response to Cell Apoptosis ER-Stress, Although ER-Stress Induced We further explored the relationship between the viability of beta Apoptosis Is Not Altered by CDK11 cells and the expression levels of CDK11. We determined the Hemideficiency number of late apoptotic beta cells using the TUNEL assay in WT In order to ascertain whether CDK11 hemideficiency also and HTZ mice from both 5-week-old NOD and NOD/SCID inhibits UPR-induced apoptosis in beta cells, we cultured strains (Figure 3A) (Suplementary Figure 3). In the NOD pancreatic islets from either wild type (NS-WT) or CDK11 strain, the N-WT beta cells exhibited a dramatic increase (10.8 hemideficient (NS-HTZ) NODSCID female mice in the times) in apoptotic activity compared to those from N-HTZ presence or absence of Thapsigargin, a drug that induces ER- (173 × 10-5 ± 69 × 10-5 vs. 16 × 10-5 ± 9 × 10-5, respectively). stress. 24 h later, the islets were collected and the RNAs of all The differences in the apoptosis rates between the WT and HTZ experimental groups extracted to quantify the relative expression genotypes were abrogated in the NOD/SCID strain (3 × 10-5 ± 2 × levels of three key UPR markers: BiP (Binding-immunoglobulin 10-5 vs. 95 ± 74 × 10-5, respectively) (Figure 3A). These results protein, a chaperone) (31); Atf4 (Activating transcription Factor reveal that the protection against T1D exerted by the CDK11 4) (39); and Chop (C/EBP Homologous Protein) (32). ATF4 hemideficiency in NOD mice is caused by a negative interference leads to CHOP expression, which leads to apoptosis in with cytokine (inflammation)-induced apoptosis, since, in the terminal UPR. NOD/SCID strain, as expected, the apoptosis levels are The results obtained showed a significant reduction in the negligible regardless of CDK11 expression levels, because there upregulation of both, Chop (132.1 ± 39.61 vs. 411.6 ± 140.4) is no inflammation. (Figure 3F) and Atf4 (23.79 ± 17.51 vs. 144.2 ± 40.17) (Figure Altogether, these observations suggested a scenario in which 3G), upon ER-stress induction in the NS-HTZ compared to the the inflammation-driven cleavage of CDK11p58 and CDKp130 NS-WT islets, respectively. In NS-HTZ, the induction of both into CDK11p46 and, the induction of CDK11p58 in a mitotic transcription factors by ER-stress is severely impaired. However, effort to compensate for the initial beta cell loss induced by ER-stress-induced BiP upregulation, was not significantly Frontiers in Immunology | www.frontiersin.org 6 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 A B C D E F G H I FIGURE 3 | CDK11 hemideficiency protects beta cells against inflammation-induced apoptosis. (A) TUNEL assessment of apoptotic beta cells was performed in pancreatic sections from 5-week-old N-WT, N-HTZ, NS-WT and NS-HTZ mice (n = 7 mice per genotype; N-WT: 275 islets, N-HTZ: 237 islets, NS-WT: 201 islets, NS-HTZ 232 islets) and quantified using epifluorescence microscopy. TUNEL-positive nuclei per insulin-positive area were measured in each experimental group. (B–D) Induction of DNA fragmentation as a measurement of apoptosis induction by proinflammatory cytokines. NS-WT and NS-HTZ pancreatic islets (25 islets per condition) from 5-week-old female mice were cultured either in the presence or absence of cytokines (IFN-g + TNF-a + IL1b) (n = 7 experiments) for 24 hours at the indicated glucose concentrations, then, cells were harvested and stained to quantify DNA fragmentation by Flow Cytometry. (B) Gating strategy is presented: Gate P1 comprehends the apoptotic SubG1 events used for subsequent calculations. Both panels correspond to NS-WT islets cultured in 5mM glucose for 24 h, in the presence (bottom panel) or absence (upper panel) cytokines. Results are plotted either as percentage of increment in SubG1 events upon cytokine exposure (C) or raw data comparing basal levels of subG1 to those after cytokine exposure (D). (E) Islet cells from 6-to-8-week-old NS-WT and NS-HTZ (n = 12) mice were stained to assess the beta cell proliferative activity; the percentages of CD45- Glut2+ Ki67+ proliferating cells are plotted from one experiment that is representative of three independent experiments. (F–H) Upregulation of UPR markers upon ER-stress induction. NS-WT and NS-HTZ pancreatic islets (50 islets per conditon) from 5–7-week-old female mice were cultured for 24 h either in the presence or absence of 1 mM Thapsigargin, harvested for RNA isolation and qRT-PCR of Chop (F), Atf4 (G), and BiP (H) was performed. The graphs plot the induction of RNA levels compared to those of untreated controls. Data are means samples ± SEM (n = 4 independent experiments). (I) DNA fragmentation (%subG1) as a measurement of apoptosis induction by thapsigargin. NS-WT (n = 4 mice) and NS-HTZ (n = 3 mice) pancreatic islets (30 islets per condition) from 5-week-old female mice were cultured for 96 h either in the presence or absence of thapsigargin (1µM). ***p < 0.005, **p < 0.01, and *p < 0.05. different between the NS-HTZ genotype compared to the NS- Cyclin D3 Promotes the Viability of Beta WT islets (1191 ± 277,4 vs. 2407 ± 647,4, respectively) (see Cells Independently of CDK11 Signaling Figure 3H). Cdk11, similarly to Cyclin D3, is downregulated in beta cells This outcome evidenced that CDK11 could have a role in following an autoimmune attack in T1D. However, while the promoting UPR-induced beta cell death, since CDK11 downregulation of Cyclin D3 compromises the beta cell viability, hemideficiency would protect beta cells from apoptosis by the downregulation of CDK11 is protective against beta cell inhibiting UPR-induced CHOP expression. To further assess death. This is an intriguing finding since Cyclin D3 is a this hypothesis, we cultured pancreatic islets from both, NS-WT regulatory partner of CDK11p58 (40–44), which, in turn, is and NS-HTZ in either the presence or absence of Thapsigargin for required for cell cycle progression and is involved in apoptosis. 96 h (4 days), collected the islets, and measured %subG1 events by Therefore, in the absence of Cyclin D3, CDK11p58-promoted flow cytometry (Figure 3I). Surprisingly, no differences in the apoptosis should be impaired, and NOD mice that are deficient levels of Thasigargin-induced apoptosis were observed between in Cyclin D3 should exhibit milder T1D due to its role in both genotypes. This outcome revealed that CDK11 activating CDK11p58. However, we previously showed (10) that hemideficiency had not a significant role in inhibiting ER-stress the Cyclin D3 deficiency dramatically exacerbates T1D in NOD induced apoptosis in beta cells. mice, while its overexpression in beta cells protects them against Frontiers in Immunology | www.frontiersin.org 7 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 diabetes; furthermore, neither of these conditions, i.e., deficiency DISCUSSION or overexpression, has any effect on beta cell replication (10), which strongly suggests that the role of Cyclin D3 in beta cells is Role of CDK11 in the Beta Cell independent of CDK11p58. Compartment Thus, we speculated about the relationship between CDK11 is a pleiotropic kinase that is involved in gene CDK11 and Cyclin D3 regarding beta cell viability and expression, cell division and cell death. The function of whether a hierarchy exists among the respective signaling CDK11 and its regulation in these processes remain unclear pathways that are engaged by these two molecules in beta cells. in beta cells. Beta cells have been shown to increase their Our working hypothesis is that CDK11 may work as a decoy proliferation rates prior to the onset of T1D (30), therefore, it receptor for Cyclin D3, quenching Cyclin D3 anti-apoptotic is plausible that while the overall mRNA expression of Cdk11 is activity in beta cells. In the N-HTZ genotype, the levels of downregulated due to insulitis, the translation of CDK11p58 is free, unbound Cyclin D3 are higher than in N-WT, and hence, favored at the expense of CDK11p130 in an attempt to subdue beta cell survival is promoted. Therefore, if this was the the loss of beta cells. However, this defense mechanism may be case, we would not expect any amelioration of the detrimental to beta cell survival because CDK11p58 also signals severity of the disease in Cyclin D3 deficient mice upon for mitosis-coupled apoptosis (24) and becomes a substrate for CDK11 hemideficiency. the amplification of caspase-3, which would result in cytokine- We answered this question by obtaining NOD mice deficient induced beta cell death in T1D. Therefore, downregulation of in Cyclin D3 that were either hemideficent (K11HTZ-D3KO) or Cdk11mRNA would prevent massive CDK11p58 translation not (K11WT-D3KO) for CDK11. under inflammatory circumstances. In the spinal cord and We monitored the spontaneous incidence of diabetes in these astrocytes, the CDK11p58 protein is also upregulated during two experimental groups and observed that regardless of the inflammation. Moreover, in Schwann cells, a pro-inflammatory CDK11 hemideficiency, the Cyclin D3 deficiency had a stimulus, such as LPS, could upregulate CDK11p58, and the dominant effect on the onset of diabetes and exacerbated the partnership CDK11p58/Cyclin D3 inhibits of the cell cycle and disease compared to normal NOD mice (K11WT-D3WT) since promotes apoptosis (43). In this sense we found that the the CDK11 hemideficiency did not alleviate the T1D in the increased CDK11p58 expression levels in NOD PECs prior to Cyclin D3 deficient mice (Figure 4). the onset of diabetes did not alter beta cell replication. Altogether, this observation suggested that at least one of the Moreover, genetic CDK11 hemideficiency, (affecting mostly to signaling pathways responsible for the survival of beta cells in the CDK11p58 protein compartment, while CDK11p130 remains which Cyclin D3 is involved is independent of the CDK11 roughly untouched in endocrine cells), protected beta cells from signaling pathway. apoptosis in the NOD proinflammatory niche. Furthermore, when exposed to a powerful proinflammatory mileu NS-HTZ islets exhibited significantly lower apoptotic rates compared to those from the NS-WT control group, confirming the involvement of CDK11p58 in cytokine-induced apoptosis. Therefore, the scenario in which CDK11 would be cleaved upon inflammation promoting apoptosis with a dominant effect over the other CDK11 roles, is evidenced as a mechanism responsible for protection against beta cell apoptosis in the CDK11 hemideficient NOD mice. ER-stress and inflammation are responsible for most of the beta cell failure in T1D. Both, inflammation and ER-stress crosstalk and feed-back each other positively, resulting in exacerbation of beta cell death (33). It has been reported that interfering the UPR signaling reverses T1D in the NOD mouse model (45). Both, CDK11p58 (24) and, CHOP (32), a molecule causally related to ER-stress-induced apoptosis, have been reported to promote apoptosis by interfering with the expression of Bcl-2, a key anti-apoptotic molecule in ER-stress (46). However, we have not observed an impairment in the induction of apoptosis by ER-stress associated to the CDK11 hemideficiency, which suggests that other (not dependent on FIGURE 4 | CDK11 hemideficiency protection against T1D depends on CHOP) apoptotic cascades in terminal UPR, such as those Cyclin D3 availabilty. The cumulative diabetes incidence in K11HTZ-D3KO downstream IRE1 (47), remain unaltered, and CDK11 would (n = 19), K11WT-D3KO (n = 23) and K11WT-D3WT (n = 34) mice was be mainly involved in cytokine-induced apoptosis. assessed. The data are shown as the percentages of the cumulative diabetic The expression of the CDK11 promoter is hindered by mice at a particular time point. *p ≤ 0.05. inflammation, probably through Ets-1 transcription factor Frontiers in Immunology | www.frontiersin.org 8 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 (48–51), resulting in a protection against apoptosis in beta cells. In AUTHOR CONTRIBUTIONS the other cell types, cyclin D3 interacts with CDK11p58 to promote cell division and/or apoptosis. In adult mouse beta cells, we did not CM designed the study. ES, CV, JL, NS-A, US, LH, SV-P, JP, observe changes in replication due to either Cyclin D3 or the and CM performed the study. TS, JV, AC, AR, AN, and MV-C CDK11 hemideficiency, which suggests that the protection against contributed new reagents/analytic tools and revised diabetes is not related to proliferation. Moreover, a priori, one the manuscript. JML provided the CDK11 hemideficient would expect that the deficiency in Cyclin D3, which is a putative mouse model in the mixed 129SvxC57BL/6 genetic activator of CDK11p58, should promote a phenotype that is similar background. CM wrote the paper and supervised the study. to that observed in the CDK11 hemideficient mice, and the All authors contributed to the article and approved the deficiency of both should add up to the same direction. submitted version. However, this is not the case since both molecules have opposite effects on the onset of T1D. This observation shows that the downregulation of CDK11 exerts its protection against T1D in a Cyclin D3-dependent fashion. This outcome would be compatible FUNDING with a scenario in which CDK11p58 binds Cyclin D3 to prevent its signaling to promote beta cell survival, and, to activate CDK11p58 This work was supported by the Spanish Ministerio de pro-apoptotic activity. Therefore, the anti-apoptotic role of Cyclin Economía, Industria y Competitividad Grant SAF2017-82567-R, D3 should be more prominent in non-cycling cells, a scenario Ministerio de Ciencia y Tecnologı́ a Grant SAF2008-02536, and, when CDK11p58 is not expressed. Ministerio de Ciencia, Innovació n y Universidades Grant SAF2014-55077-R (to CM) and SAF2016-77227-R and SAF2013-45140-R (to TS and JV); the 2009 SGR-DGR grant Role of CDK11 in the Autoimmune from the Generalitat de Catalunya (to JV and CM) and the Repertoire TR265 no. A001E-12132/2009 from the University of Lleida (to The diabetogenicity of the autoimmune repertoire was not JV and CM); the Generalitat Valenciana PROMETEO/2011/080 affected in N-HTZ mice. Moreover, the representation of the grant; and the Ministerio de Economı́a y Competitividad FoxP3+CD25+CD4+ Treg cell subset in either the regional lymph BFU2011-28358 grant (to AN). NS-A and CV were recipients of nodes or the islet infiltrate in not altered in CDK11 hemideficient a predoctoral fellowship from the University of Lleida. US, ES, and mice. Therefore, the protection observed against T1D in N-HTZ JL have been granted with AGAUR predoctoral fellowships from cannot be explained by an altered diabetogenic repertoire in the Generalitat de Catalunya. CM and JV are assistant professors this genotype. of Immunology in the Serra Hunter Program at the University of Lleida and, investigators at the same institution and the Lleida Institute for Biomedical Research Fundació n Pifarré. CONCLUSION CDK11 downregulation prior to T1D onset has a protective effect on beta cell survival in an inflammatory context. This ACKNOWLEDGMENTS protective effect depends on Cyclin D3 availability. Further work We want to thank Judit Ribas for her scientific and technical needs to be performed to identify the relationship between advice, and Anaïs Panosa and Ainhoa Garcı́a for their technical CDK11-mediated proapoptotic signaling and Cyclin D3 in beta assistance. We also want to thank Santiago Rı́os for assistance in cells, to design better therapeutic approaches targeting T1D. statistical analysis. The authors want to acknowledge the IRBLleida Flow Cytometry Facility for cell s ubset characterization and apoptosis assessment experiments, and DATA AVAILABILITY STATEMENT the Animal Facility at the University of Lleida for remarkable NOD mice husbandry. The cell culture experiments were The original contributions presented in the study are included in performed in the Cell Culture Scientific and Technical Facility the article/Supplementary Material. Further inquiries can be from the University of Lleida. directed to the corresponding author. ETHICS STATEMENT SUPPLEMENTARY MATERIAL The animal study was reviewed and approved by the Institutional The Supplementary Material for this article can be found online Ethical Committee for Animal Experimentation of the at: https://www.frontiersin.org/articles/10.3389/fimmu.2021. University of Lleida (CEEA). 634797/full#supplementary-material Frontiers in Immunology | www.frontiersin.org 9 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 REFERENCES 20. Chi Y, Zhang C, Zong H, Hong Y, Kong X, Liu H, et al. Thr-370 is responsible for CDK11p58 autophosphorylation, dimerization, and kinase activity. J. Biol. 1. Mandrup-Poulsen T. The role of interleukin-1 in the pathogenesis of IDDM. Chem (2011) 286:1748–57. doi: 10.1074/jbc.M110.107367 Diabetologia (1996) 39:1005–29. doi: 10.1007/BF00400649 21. Zhang S, Cai M, Zhan S, Xu S, Chen S, Chen X, et al. Interaction of 2. Hoorens A, Pipeleers D. Nicotinamide protects human beta cells against P58PITSRE, a G2/M-specific protein kinase with cyclin D3. J Biol Chem chemically-induced necrosis, but not against cytokine-induced apoptosis. (2002) 277(38):35314–22. doi: 10.1074/jbc.M202179200 Diabetologia (1999) 42:55–9. doi: 10.1007/s001250051113 22. Li T, Inoue A, Lahti JM, Kidd VJ. Failure to proliferate and mitotic arrest of 3. Pavlovic D, Chen MC, Gysemans CA, Mathieu C, Eizirik DL. The role of CDK11p110/p58-Null Mutant Mice at the Blastocyst Stage of Embryonic Cell interferon regulatory factor-1 in cytokine-induced mRNA expression and cell Development. Mol Cell Biol (2004) 24(8):3188–97. doi: 10.1128/ death in murine pancreatic beta-cells. Eur Cytokine Netw (1999) 10:403–12. mcb.24.8.3188-3197.2004 4. Rabinovitch A. An update on cytokines in the pathogenesis of insulin- 23. Yun X, Wu Y, Yao L, Zong H, Hong Y, Jiang J, et al. CDK11p58 protein kinase dependent diabetes mellitus. Diabetes Metab Rev (1998) 14:129–51. activity is associated with Bcl-2 down-regulation in pro-apoptosis pathway. doi: 10.1002/(sici)1099-0895(199806)14:2<129::aid-dmr208>3.0.co;2-v Mol Cell Biochem (2007) 304:213–8. doi: 10.1007/s11010-007-9502-x 5. Anderson MS, Bluestone JA. The NOD mouse: A Model of Immune 24. Lahti JM, Xiang J, Heath LS, Campana D, Kidd VJ. PITSLRE protein kinase Dysregulation. Annu Rev Immunol (2005) 23:447–85. doi: 10.1146/ activity is associated with apoptosis. Mol Cell Biol (1995) 15:1–11. annurev.immunol.23.021704.115643 doi: 10.1128/MCB.15.1.1 6. D’Hertog W, Overberg L, Lage K, Ferreira GB, Maris M, Gysemans C, et al. 25. Beyaert R, Kidd VJ, Cornelis S, Van de Craen M, Denecker G, Lahti JM, et al. Proteomics analysis of cytokine-induced dysfunction and death in insulin- Cleavage of PITSLRE kinases by ICE/CASP-1 and CPP32/CASP-3 during producing INS-1E cells: new insights into the pathways involved. Mol Cell apoptosis induced by tumor necrosis factor. J Biol Chem (1997) 272 Proteomics (2007) 6:2180–99. doi: 10.1074/mcp.M700085-MCP200 (18):11694–7. doi: 10.1074/jbc.272.18.11694 7. Cardozo AK, Kruhøffer M, Leeman R, Orntoft T, Eizirik DL. Identification of 26. Shi J, Hershey JWB, Nelson MA. Phosphorylation of the eukaryotic initiation novel cytokine-induced genes in pancreatic beta-cells by high-density factor 3f by cyclin-dependent kinase 11 during apoptosis. FEBS Lett (2009) oligonucleotide arrays. Diabetes (2001) 50:909–20. doi: 10.2337/ 583:971–7. doi: 10.1016/j.febslet.2009.02.028 diabetes.50.5.909 27. Feng Y, Ariza ME, Goulet AC, Shi J, Nelson MA. Death-signal-induced 8. Wägner AM, Cloos P, Bergholdt R, Boissy P, Andersen TL, Henriksen DB, relocalization of cyclin-dependent kinase 11 to mitocondria. Biochem J (2005) et al. Post-translational protein modifications in type 1 diabetes: a role for the 392:65–73. doi: 10.1042/BJ20050195 repair enzyme protein-L-isoaspartate (D-aspartate) O-methyltransferase? 28. O’Brien BA, Harmon BV, Cameron DP, Allan DJ. Apoptosis is the mode of Diabetologia (2007) 50:676–81. doi: 10.1007/s00125-006-0556-1 beta cell death responsible for the development of IDDM in the nonobese 9. Kacheva S, Lenzen S, Gurgul-Convey E. Differential effects of diabetic (NOD) mouse. Diabetes (1997) 46(5):750–7. doi: 10.2337/ proinflammatory cytokines on cell death and ER stress in insulin-secreting diab.46.5.750 INS1E cells and the involvement of nitric oxide. Cytokine (2011) 55:195–201. 29. Eizirik DL, Mandrup-Poulsen T. A choice of death- the signal transduction of doi: 10.1016/j.cyto.2011.04.002 immune-mediated beta-cell apoptosis. Diabetologia (2001) 44:2115–33. 10. Saavedra-Á vila NA, Sengupta U, Sá nchez B, Sala E, Haba L, Stratmann T, et al. doi: 10.1007/s001250100021 Cyclin D3 promotes pancreaticb -cell fitness and viability in a cell cycle- 30. Sreenan S, Pick AJ, Levisetti M, Baldwin AC, Pugh W, Polonsky KS. Increased independent manner and is targeted in autoimmune diabetes. Proc. Natl. b-cell proliferation and reduced mass before diabetes onset in the Non Obese Acad. Sci. USA (2014) 111:E3405–14. doi: 10.1073/pnas.1323236111 Diabetic mouse. Diabetes (1999) 48:989–96. doi: 10.2337/diabetes.48.5.989 11. Xiang J, Lahti JM, Grenet J, Easton J, Kidd VJ. Molecular Cloning and 31. Marhfour I, Lopez XM, Lefkaditis D, Salmon I, Allagnat F, Richardson SJ, Expression of Alternatively Spliced PITSLRE Protein Kinase Isoforms. J Biol et al. Expression of endoplasmic reticulum stress markers in the islets of Chem (1994) 269(22):15786–94. doi: 10.1016/S0021-9258(17)40749-6 patients with type 1 diabetes. Diabetologia (2012) 55:2417–20. doi: 10.1007/ 12. Cornelis S, Bruynooghe Y, Denecker G, Huffel SV, Tinton S, Beyaert R. s00125-012-2604-3 Identification and Characterization of a Novel Cell Cycle-Regulated Internal 32. Sano R, Reed JC. ER stress-induced cell death mechanisms. BBA (2013) Ribosome Entry Site. Mol Cell (2000) 5:597–605. doi: 10.1016/s1097-2765(00) 1833:3460–70. doi: 10.1016/j.bbamcr.2013.06.028 80239-7 33. Meyerovich K, Ortis F, Allagnat F, Cardozo AK. Endoplasmic reticulum stress 13. Trembley JH, Hu D, Hsu LC, Yeung CY, Slaughter C, Lahti JM, et al. PITSLRE and the unfolded protein response in pancreatic islet inflammation. J Mol p110 Protein Kinases Associate with Transcription Complexes and Affect Endocrinol (2016) 57:R1–R17. doi: 10.1530/JME-15-0306 Their Activity. J Biol Chem (2002) 277(4):2589–96. doi: 10.1074/ 34. Marzo N, Ortega S, Stratmann T, Garcı́a A, Rı́os M, Gimé nez A, et al. Cyclin- jbc.M109755200 dependent kinase 4 hyperactivity promotes autoreactivity in the immune 14. Hu D, Mayeda A, Trembley JH, Lahti JM, Kidd VJ. CDK11 Complexes system but protects pancreatic cell mass from autoimmune destruction in the Promote Pre-mRNA Splicing. J Biol Chem (2003) 278(10):8623–9. nonobese diabetic mouse model. J Immunol (2008) 180:1189–98. doi: 10.4049/ doi: 10.1074/jbc.M210057200 jimmunol.180.2.1189 15. Trembley JH, Hu D, Slaughter CA, Lahti JM, Kidd VJ. Casein Kinase 2 35. Allison J, Thomas HE, Catterall T, Kay TWH, Strasser A. Transgenic Interacts with Cyclin-dependent Kinase 11 (CDK11) in Vivo and Expression of Dominant-Negative Fas-Associated Death Domain Protein in Phosphorylates Both the RNA Polymerase II Carboxyl-terminal Domain beta Cells Protects against Fas Ligand-Induced Apoptosis and Reduces and CDK11 in Vitro. J Biol Chem (2003) 278(4):2265–70. doi: 10.1074/ Spontaneous Diabetes in Nonobese Diabetic Mice. J Immunol (2005) 2005 jbc.M207518200 (175):293–301. doi: 10.4049/jimmunol.175.1.293 16. Sachs NA, Vaillancourt RR. Cyclin-dependent kinase 11p110 activity in the 36. Bosma GC, Custer RP, Bosma MJ. A severe combined immunodeficiency absence of CK2. Biochim Biophys Acta (2003) 1624:98–108. doi: 10.1016/ mutation in the mouse. Nature (1983) 301:527–30. doi: 10.1038/301527a0 j.bbagen.2003.10.001 37. Blunt T, Gell D, Fox M, Taccioli GE, Lehman AR, Jackson SP, et al. 17. Yang L, Li N, Wang C, Yu Y, Yuan L, Zhang M, et al. Cyclin L2, a Novel RNA Identification of a nonsense mutation in the carboxyl-terminal region of Polymerase II-associated Cyclin, Is Involved in Pre-mRNA Splicing and DNA-dependent protein kinase catalytic subunit in the scid mouse. Proc Induces Apoptosis of Human Hepatocellular Carcinoma Cells. J Biol Chem Natl Acad Sci USA (1996) 93(19):10285–90. doi: 10.1073/pnas.93. (2004) 279(12):11639–48. doi: 10.1074/jbc.M312895200 19.10285 18. Wilker EW, van Vugt MATM, Artim SC, Huang PH, Petersen CP, Reinhardt 38. O’Brien BA, Huang Y, Geng X, Dutz JP, Finegood DT. Phagocytosis of HC, et al. 14-3-3s controls mitotic translation to facilitate cytokinesis. Nature apoptotic cells from NOD mice is reduced. Diabetes (2002) 51:2481–8. (2007) 446:329–32. doi: 10.1038/nature05584 doi: 10.2337/diabetes.51.8.2481 19. Hu D, Valentine M, Kidd VJ, Lahti JM. CDK11p58 is required for the 39. Cnop M, Foufelle F, Velloso LA. Endoplasmic reticulum stress, obesity and maintenance of sister chromatid cohesion. J Cell Sci (2007) 120:2424–34. diabetes. Trends Mol Med (2012) 18:59–68. doi: 10.1016/j.molmed. doi: 10.1242/jcs.007963 2011.07.010 Frontiers in Immunology | www.frontiersin.org 10 February 2021 | Volume 12 | Article 634797 Sala et al. Antagonism Between CDK11 and Cyclin D3 40. Zong H, Chi Y, Wang Y, Yang Y, Zhang L, Chen H, et al. Cyclin D3/ 48. Feng Y, Goulet AC, Nelson MA. Identification and characterization of the CDK11p58 complex is involved in the repression of androgen receptor. Mol human Cdc2l2 gene promoter. Gene (2004) 330(2004):75–84. doi: 10.1016/ Cell Biol (2007) 27:7125–42. doi: 10.1128/MCB.01753-06 j.gene.2004.01.006 41. Wang Y, Zong H, Chi Y, Hong Y, Yang Y, Zou W, et al. Repression of 49. Kahle A, Feng Y, Nelson MA. Isolation and characterization of the human Estrogen Receptor alpha by cdk11p58 through promoting its ubiquitin– Cdc2L1 gene promoter. Gene (2005) 344(2005):53–60. doi: 10.1016/ proteasome degradation. J Biochem (2009) 145:331–43. doi: 10.1093/jb/ j.gene.2004.10.020 mvn177 50. Li L, Miao X, Ni R, Wang L, Gu X, Yan L, et al Epithelial-specific ETS-1 (ESE1/ 42. Chi Y, Hong Y, Zong H, Wang Y, Zou W, Yang J, et al. CDK11p58 represses ELF3) regulates apoptosis of intestinal epithelial cells in ulcerative colitis via vitamin D receptor-mediated transcriptional activation through promoting its accelerating NF-kB activation. Immunol Res 62:198–212. doi: 10.1007/s12026- ubiquitin–proteasome degradation. Biochim Biophys Res Commun (2009) 015-8651-3 386:493–8. doi: 10.1016/j.bbrc.2009.06.061 51. Oettgen P. Regulation of vascular inflammation and remodeling by ETS factors. 43. Ji Y, Xiao F, Sun L, Qin J, Shi S, Yang J, et al. Increased expression of Circ Res (2006) 24(99):1159–66. doi: 10.1161/01.RES.0000251056.85990.db CDK11p58 and cyclin D3 following spinal cord injury in rats. Mol Cell Biochem (2008) 309:49–60. doi: 10.1007/s11010-007-9642-z Conflict of Interest: The authors declare that the research was conducted in the 44. Duan Y, He X, Yang H, Ji Y, Tao T, Chen J, et al. Cyclin D3/Cdk11p58 complex absence of any commercial or financial relationships that could be construed as a involved in Schwann cells proliferation repression caused by lipopolysaccharide. potential conflict of interest. Inflammation (2010) 33:189–99. doi: 10.1007/s10753-009-9173-8 45. Morita S, Villalta SA, Feldman H, Register AC, Rosenthal W, Hoffmann- The reviewer XD declared a shared affiliation, with no collaboration, with one of Petersen IT, et al. Targeting ABL-IRE1a signaling spares ER-stressed the authors, JL, to the handling editor at the time of review. pancreatic beta cells to reverse autoimmune diabetes Cell. Metab (2017) 25:883–97. doi: 10.1016/j.cmet.2017.03.018 Copyright © 2021 Sala, Vived, Luna, Saavedra-Á vila, Sengupta, Castaño, 46. Lam M, Dubyak G, Chen L, Nunez G, Miesfeld RL, Distelhorst CW. Evidence Villar-Pazos, Haba, Verdaguer, Ropero, Stratmann, Pizarro, Vázquez-Carrera, that BCL-2 represses apoptosis by regulating endoplasmic reticulum- Nadal, Lahti and Mora. This is an open-access article distributed under the terms associated Ca 2+ fluxes. Proc Natl Acad Sci USA (1994) 91:6569–73. of the Creative Commons Attribution License (CC BY). The use, distribution or doi: 10.1073/pnas.91.14.6569 reproduction in other forums is permitted, provided the original author(s) and the 47. Lu J, Shen H, Li Q, Xiong F, Xie R, Yuan M, et al. KCNH6 protects pancreatic copyright owner(s) are credited and that the original publication in this journal is b-cells from endoplasmic reticulum stress and apoptosis. FASEB J (2020) cited, in accordance with accepted academic practice. No use, distribution or 34:15015–28. doi: 10.1096/fj.202001218R reproduction is permitted which does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 11 February 2021 | Volume 12 | Article 634797