Cellular and Molecular Life Sciences (2022) 79:468 https://doi.org/10.1007/s00018-022-04481-1 Cellular and Molecular Life Sciences ORIGINAL ARTICLE CD69‑oxLDL ligand engagement induces Programmed Cell Death 1 (PD‑1) expression in human CD4 + T lymphocytes María Jiménez‑Fernández1,2 · Cristina Rodríguez‑Sinovas3,4 · Laia Cañes4,5 · Carme Ballester‑Servera4,5 · Alicia Vara1 · Silvia Requena1 · Hortensia de la Fuente1,4 · José Martínez‑González4,5 · Francisco Sánchez‑Madrid1,2,4 Received: 18 February 2022 / Revised: 28 June 2022 / Accepted: 12 July 2022 / Published online: 5 August 2022 © The Author(s) 2022 Abstract The mechanisms that control the inflammatory–immune response play a key role in tissue remodelling in cardiovascular diseases. T cell activation receptor CD69 binds to oxidized low-density lipoprotein (oxLDL), inducing the expression of anti-inflammatory NR4A nuclear receptors and modulating inflammation in atherosclerosis. To understand the downstream T cell responses triggered by the CD69-oxLDL binding, we incubated CD69-expressing Jurkat T cells with oxLDL. RNA sequencing revealed a differential gene expression profile dependent on the presence of CD69 and the degree of LDL oxida- tion. CD69-oxLDL binding induced the expression of NR4A receptors (NR4A1 and NR4A3), but also of PD-1. These results were confirmed using oxLDL and a monoclonal antibody against CD69 in CD69-expressing Jurkat and primary CD4 + lym- phocytes. CD69-mediated induction of PD-1 and NR4A3 was dependent on NFAT activation. Silencing NR4A3 slightly increased PD-1 levels, suggesting a potential regulation of PD-1 by this receptor. Moreover, expression of PD-1, CD69 and NR4A3 was increased in human arteries with chronic inflammation compared to healthy controls, with a strong correlation between PD-1 and CD69 mRNA expression (r = 0.655 P < 0.0001). Moreover, PD-1 was expressed in areas enriched in CD3 infiltrating T cells. Our results underscore a novel mechanism of PD-1 induction independent of TCR signalling that might contribute to the role of CD69 in the modulation of inflammation and vascular remodelling in cardiovascular diseases. Keywords CD69 leucocyte activation receptor · oxLDL · Cardiovascular disease · Programmed Death 1 (PD-1) · Anti- inflammatory response · Inflamed aorta José Martínez-González and Francisco Sánchez-Madrid contributed equally to this work. Introduction * José Martínez‑González The early lymphocyte activation antigen CD69 is a type II jose.martinez@iibb.csic.es C-lectin membrane receptor readily induced upon leuko- * Francisco Sánchez‑Madrid cyte stimulation. Initial in vitro studies of CD69 suggested fsmadrid@salud.madrid.org a pro-inflammatory role for this molecule [1]. However, 1 growing evidence from in vivo analyses in mice lacking Hospital Universitario de la Princesa, Universidad Autónoma de Madrid (UAM), Instituto de Investigación CD69 highlighted its role as a molecular brake to control Sanitaria del Hospital Universitario de La Princesa (IIS-IP), exacerbated inflammatory responses [2]. Animal models c/ Diego de León, 62, 28006 Madrid, Spain including collagen-induced arthritis, autoimmune experi- 2 Centro Nacional de Investigaciones Cardiovasculares mental myocarditis, allergic asthma and colitis, among oth- (CNIC), Madrid, Spain ers, clearly established that CD69 exerts an immune regula- 3 Institut de Recerca Hospital de la Santa Creu i Sant Pau tory and protective role in inflammation [2–4]. CD69 acts (IRHSCSP), IIB-Sant Pau, Barcelona, Spain through different mechanisms, mainly inhibiting Th17 cell 4 CIBER de Enfermedades Cardiovasculares (CIBERCV), differentiation and regulating the Th17/Treg cell balance. Madrid, Spain Hence, expansion of Th17 cell activity was observed in the 5 Instituto de Investigaciones Biomédicas de Barcelona absence of CD69; conversely, the suppressive function of - Consejo Superior de Investigaciones Científicas CD4 + CD25 + FoxP3 + Tregs was diminished in CD69- (IIBB-CSIC), IIB-Sant Pau, C/ Rosselló, 161, deficient mice, revealing the importance of CD69 expression 08036 Barcelona, Spain 13 Vol.:(0123456789) 468 Page 2 of 13 M. Jiménez‑Fernández et al. on this T cell subset [5]. The mechanisms by which CD69 purified by density gradient with Pancoll human (PAN-Bio- controls Treg activity include upregulation of TGF-β [4, 6], tech) from healthy donors and CD4+ T lymphocytes iso- elevated production of the cytokine IL-10 and expression of lated using the EasySep™ Human CD4+ T Cell Isolation GITR, CD38, ICOS and CTLA-4 [7]. CD69 is required for Kit (Stemcell). the expansion of Tregs in the thymus (tTregs) through the regulation of BIC/microRNA 155 and its target, suppressor Low‑density lipoproteins (LDL) isolation of cytokine signalling 1 (SOCS-1) [8]. and oxidation. Understanding the pathophysiologic function of CD69 has been challenging by the absence of known ligands. Native LDL (nLDL) were isolated from pooled plasma Recently, CD69 has been identified as an oxidized low- of healthy blood donors of the Barcelona area [14]. The density lipoprotein (oxLDL) receptor in human CD4 + T procedure was approved by the Ethics Committee of the lymphocytes [9]. OxLDL-CD69 binding triggers a protec- Hospital de la Santa Creu i Sant Pau (Barcelona, Spain) tive effect, enhancing the anti-inflammatory transcription and was conducted in accordance with the Declaration factors NR4A1 and NR4A3 (NOR-1), downregulating pro- of Helsinki. Briefly, pooled plasma was centrifuged inflammatory cytokines and promoting Treg differentiation, (80,000×g for 30 min at 4 °C) to remove chylomicrons. thereby modulating atherosclerosis [9]. LDL (d = 1.019–1.063 g/mL) were isolated by potassium An increasingly number of immunoregulatory molecules bromide density gradient ultracentifugation using a Beck- such as CTLA-4, PD-1, ICOS, LAG-3 and TIGIT inhibit the man Coulter Optima™ L-100 XP ultracentrifuge and a activity and proliferation of T lymphocytes [10, 11]. The Beckman 50.2 Ti rotor (Beckman Coulter) at 36,000 rpm inhibitory receptor programmed cell death-1 (PD-1, CD279) for 18 h at 4 °C (gmax = 156,000). The LDL fraction was is a key therapeutic target to treat immune and inflammatory dialysed four times against 200 volumes of PBS for 24 h. disorders as well as to stimulate the anti-tumour immune All solutions were deoxygenated by N ­ 2 bubbling. LDL response [12]. PD-1 and its ligands PD-L1 (B7-H1, CD274) were sterilized by filtration through a low-protein-binding and PD-L2 (B7-DC, CD273) belong to the B7-CD28 family non-pyrogenic filter (Millex-GV, Millipore). Oxidized and are highly regulated during normal immune responses LDL (oxLDL) with different degrees of oxidation were [13]. prepared by exposing nLDL to 10 µM ­CuSO4 at 37 °C for Here, we report that the binding of oxLDL to CD69 increasing times. LDL were stored at 4 °C and protected induces the expression of PD-1 through the activation of from exposure to light. The content of protein (BCA pro- nuclear factor of activated T cells 1 (NFAT) in human T tein assay™, Pierce) and cholesterol (Cholesterol Assay cells. Analysis of human inflammatory artery samples Kit™, Abcam) was determined by colorimetric assays. showed a high expression of CD69 and PD-1 and strong LDL were free from contamination from other lipopro- correlation between them. Our data demonstrate that PD-1 teins (as determined by electrophoresis) or endotoxin (as upregulation is a downstream mechanism involved in the determined by a Limulus assay) (Chromogenic Endotoxin negative regulation of the inflammatory response upon Quan kit™, Pierce). Thiobarbituric acid-reactive sub- CD69–ligand engagement. stances (TBARS) content of LDL was used as an indirect evaluation of lipid peroxidation [14]. In the present study, we used nLDL (TBARS below 1 nmol malonaldehyde (MDA)/mg of LDL protein), moderately oxidized LDL Materials and methods (moxLDL; TBARS around 10 nmol MDA/mg of LDL pro- tein) and highly oxidized LDL (hoxLDL; TBARS between Cell cultures 25 and 50 nmol MDA/mg of LDL protein). Human T lymphocyte Jurkat cell line stably transfected or not with CD69 (JKCD69 and JKwt, respectively) was used mRNA extraction and RNA‑seq analysis [9]. The cells were maintained with RPMI 1640 + Glu- taMAX™ (Gibco) supplemented with 10% of heat-inacti- For experiments aimed to analyse mRNA levels, cells were vated foetal bovine serum (FBS) (GE Healthcare Hyclone,), incubated for 45 min, 4 and 8 h under the following condi- 0.1 U/ml penicillin and streptomycin (Lonza) and the selec- tions: without LDL, with nLDL or with oxLDL. Then, the tion antibiotic geneticin G418 (InvivoGen) at 0.5 mg/ml. cells were collected, rinsed once with phosphate-buffered Both cell lines were incubated at 1·106 cells/ml with lipo- saline (PBS) and preserved in QIAzol Lysis Reagent (Qia- proteins in their native or oxidized form at a concentra- gen) at − 80 °C until they were processed. Total RNA was tion of 50 µg/ml in 2% FBS medium the time indicated in isolated using the miRNeasy Mini Kit (Qiagen), and DNase each experiment. Peripheral blood mononuclear cells were digestion was performed as recommended. The cDNA was 13 CD69‑oxLDL ligand engagement induces Programmed Cell Death 1 (PD‑1) expression in human CD4 … Page 3 of 13 468 obtained by the GoScript™ Reverse Transcriptase Kit (Pro- using Minitab Statistical Software (Minitab, LLC, Penn- mega) according to the manufacturer’s protocol. sylvania, USA). On the other hand, the gene expression of For the RNA-sequencing, JKwt and JKCD69 cells were selected genes was compared in JKCD69 incubated with incubated for 4 h with nLDL moxLDL and hoxLDL as an moxLDL and hoxLDL to the control samples (JKwt sam- exploratory study. RNA was quantified by spectrophotom- ples and JKCD69 resting and nLDL). etry (ND-1000; NanoDrop Technologies), and RNA qual- ity was determined by microcapillary electrophoresis on Quantification of gene expression an Agilent 2001 Bioanalyzer (Agilent Technologies) with RNA 6000 Nanochips. The samples were sequenced and Total RNA from human tissues was isolated using processed by the Bioinformatics Unit and the Genomics TRIsure™ reagent (Bioline) and reverse transcribed into Unit from the Centro Nacional de Investigaciones Cardio- cDNA using the high-capacity cDNA Reverse Transcrip- vasculares Carlos III (CNIC), Madrid, Spain. 200 ng of tion Kit (Applied Biosystems). Gene expression for NR4A total RNA was used to generate barcoded RNA-seq librar- receptors in human samples and cell culture studies was ies using the NEBNext Ultra II Directional RNA Library evaluated by real-time PCR using the SensiFAST probe Prep Kit (New England Biolabs). Briefly, poly-A + RNA Hi-ROX mix (Bioline), the ABI PRISM 7900HT sequence was purified using poly-T oligo-attached magnetic beads detection system (Applied Biosystems) and specific Taqman followed by fragmentation and then first and second cDNA probes (Applied Biosystems) for NR4A1, (Hs00172437_m1) strand synthesis. Second strand was synthesized with U and NR4A3 (Hs00175077_m1). Alternatively, mRNA levels ribonucleotide instead of T. Next, cDNA 3′ ends were were assessed with the ­GoTaq® qPCR Master Mix system adenylated and the adapters were ligated, second strand (Promega) for SYBR Green real-time PCR analysis. The were fragmented by USER enzyme that cut the uracils to specific primers pairs (PD-1 forward CTC CAG GCA TGC preserve the directionality of the original RNA. Then the AGA TCC, reverse GGC CTG TCT GGG GAG TCT A; libraries were amplified by PCR. Directional RNA librar- CD69 forward CAA GTT CCT GTC CTG TGT GC and ies were determined using the Agilent 2100 Bioanalyzer reverse GAG AAT GTG TAT TGG CCT GGA; and GAPDH high-sensitivity DNA chip. Libraries were sequenced, Sin- forward AGC CAC ATC GCT CAG ACA C and reverse gle-End 1 × 60, on a HiSeq 2500 (Illumina) and processed GCC CAA TAC GAC CAA ATC C) were designed using with RTA v1.18.66.3. FastQ files for each sample were the Roche Life Science web (Penzberg, Germany) and syn- obtained using bcl2fastq v2.20.0.422 software (Illumina). thesized by Metabion (Steinkirchen, Baviera, Germany). FastQ files were pre-processed by FastQC (Babraham Detection of amplified products was performed with the Bioinformatics—FastQC A Quality Control tool for High CFX384 Touch™ Real-Time PCR detection system (Bio- Throughput, available online at: https://​www.​bioin​forma​ Rad). Data were analysed with the 3.1 version of the CFX tics.​babra​ham.​ac.​uk/​proje​cts/​fastqc/), to assess reads qual- Manager software (Bio-Rad). The PCR protocol consisted ity and Cutadapt to trim sequencing reads [15]. Result- of a first denaturation step of 10 min at 95 °C followed by ing reads were mapped against reference transcriptome 40 cycles of 10 s at 95 °C and at 60 s of annealing/exten- GRCh38.91 and quantified using RSEM [16]. Expected sion step at 60 °C and a 10 s at 95 °C before the melting expression counts calculated with RSEM were then pro- curve was achieved. The real-time qPCRs were performed in cessed with an analysis pipeline that used the Bioconduc- triplicate for all the targets. The relative mRNA levels were tor package EdgeR for normalization (using TMM method) determined using the comparative threshold (CT) method and expression ratio calculation, taking only into account after checking primers efficiency. Normalization was accom- those genes expressed with at least 1 count per million plished to β-ACTIN (Hs99999903_m1; for human tissue (CPM) in at least one sample [17]. The gene expression samples) or GAPDH levels. The relative mRNA levels were was normalized of every sample relative to its control for expressed as folds of change over untreated samples. each cell line (LDL vs. untreated) and then compared the changes detected with increasing LDL oxidation degrees Anti‑CD69 antibody assays in JKCD69 respect to JKwt. As an exploratory study, genes were selected as follows. On the one hand, genes candi- Monoclonal antibody TP1/55 anti-CD69 employed was gen- dates were selected using a threshold of fold change > 1.8 erated at the Hospital Universitario La Princesa (Madrid, in JKCD69 cells treated with hoxLDL. Heatmap and clus- Spain) as it has been previously described [1]. The JKwt tering for those genes were accomplished using Galaxy and JKCD69 cells were cultured in supplemented medium web platform and the public server at usegalaxy.org. Origi- at 1·106 cells/ml with the antibody (10 μg/ml) and the nal values were Log (x + 1)-transformed, columns were crosslinker AffiniPure Fab Fragment Goat Anti-Mouse clustered using Euclidean method and scaling the data by IgG, Fcγ Subclass 2b Specific (20 μg/ml, Jackson Immu- row. Principal component analysis (PCA) was performed noResearch) for three days. As isotype control, the mouse 13 468 Page 4 of 13 M. Jiménez‑Fernández et al. IgG2b kappa isotype control antibody (10 μg/ml, Thermo was approved by the Ethics Committee of the Hospital de la Fisher Scientific/Life) was engaged and crosslinker was also Santa Creu i Sant Pau (Barcelona, Spain) and was conducted added to this condition. For human primary CD4+ T cells in accordance with the Declaration of Helsinki. Samples experiments, cells were activated with phorbol 12-myristate were split and processed for conventional histological stain- 13-acetate (PMA) (50 ng/ml, Sigma-Aldrich) and ionomycin ing/immunohistochemistry or frozen in liquid nitrogen and (Io) (1 μg/ml, Sigma-Aldrich) for 3 h to induce the CD69 stored at − 80 °C for RNA and protein extraction. The speci- expression prior the addition of anti-CD69 monoclonal anti- mens for conventional staining/immunohistochemistry were body and crosslinker as mentioned before. immersed in fixative solution (4% paraformaldehyde/0.1 M phosphate-buffered saline, pH 7.4). After overnight treat- Analysis of protein surface expression by flow ment, they were sectioned into blocks and embedded in par- cytometry affin. Vessels were cut into 4-µm-thick sections that were mounted on coated glass microscope slides (FLEX IHC, The protein levels were measured after different time Dako, Santa Clara, CA, USA) and stored at room tempera- points of incubation by flow cytometry in a BD FACS ture until tested. The presence of inflammation in the aortic Canto II Flow Cytometer and analysed with FlowJo soft- samples from patients selected for this study was confirmed ware (FlowJo, LLC). Cells were rinsed twice with PBS by hematoxylin and eosin staining. Healthy control aortas (1X) and resuspended with PBS and Ghost Dye Red 780 from multi-organ donors were macroscopically normal and (1:1000 dilution, Tonbo Biosciences) for 30 min on ice. devoid of inflammatory lesions as determined by hematoxy- After that, cells were rinsed once with PBS-0.2% Bovine lin and eosin staining. Serum Albumin (BSA) (PanReac AppliChem) and resus- pended in 50 μl of this buffer with the PE anti-human CD279 Western blot analysis (PD-1) (EH12.2H7, 1:100 dilution, BioLegend) antibody for 30 min on ice. The analysis strategy was to compare the Human specimen samples for western blotting were snap- mean fluorescence intensity between treated and untreated frozen in liquid nitrogen and stored at -80ºC. Human pro- live cells for oxLDL experiments. For anti-CD69 monoclo- tein extracts were obtained using ice-cold lysis buffer as nal antibody engagement experiments, samples were incu- described [19]. Protein content of lysates was determined bated with mouse serum (1:100 dilution, Sigma-Aldrich) for by the bicinchoninic acid (BCA) protein assay (Pierce) using 15 min before adding the same volume with flow cytometry bovine serum albumin as standard. Protein lysates (15 µg/ antibodies. Percentage of positive cells for PD-1 were rep- lane) were resolved using sodium dodecyl sulphate–polyac- resented at different times. rilamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride membranes (Immobilon, Merck- NR4A3 interfering by siRNA and cyclosporine Millipore; Burlington MA, USA, IPVH00010). Blots were A assays blocked with 5% non-fat dry milk in TBST buffer (100 mM sodium chloride, 10 mM Tris–HCl pH 7.5, 0.05% Tween-20) Gene knockdown of NR4A3 was performed using ON- at room temperature for 1 h. Membranes were then incu- TARGETplus SMARTpool small interfering RNA (siRNA) bated overnight with mouse monoclonal antibodies against: and ON-TARGETplus Control Pool Non-targeting Pool as PD-1 (ab52587, Abcam, Cambridge, UK), CD69 (sc- scramble (Horizon) by electroporation. Cells were washed 373799, Santa Cruz Biotechnology, Inc., Dallas, TX, USA), once with Hanks’ Buffered Saline Solution (HBSS, Lonza) NOR-1 (H00008013-M06, Abnova, Taipei, Taiwan) or CD3 and once with Opti-MEM reduced serum medium (Gibco) (M7254, Dako). Appropriate horseradish peroxidase-con- before electroporate. After 48 h, specific knockdown was jugated secondary antibodies (Dako) and the Luminata™ confirmed by RT-qPCR of mRNA levels and experiments Western HRP Substrate (Immobilon, Merck-Millipore) were performed. Cyclosporine A (CsA) (5 μg/ml, SML1018 were used to detect bound antibodies. The size of detected from Sigma-Aldrich, Burlington, MA, USA), PMA (50 ng/ proteins was estimated using protein molecular mass stand- ml, Sigma-Aldrich) and Io (1 μg/ml, Sigma-Aldrich) were ards (Hyperpage Prestained Protein Marker; Bioline, Paris, employed when indicated. France). ß-actin (A5441, Sigma-Aldrich) was used to verify equal loading of protein on each lane. Human artery sampling and preservation Abdominal aortic samples were collected from patients undergoing open repair surgery for abdominal aortic aneu- rysm at the Hospital de la Santa Creu i Sant Pau (Barce- lona, Spain) and from multi-organ donors [18]. The study 13 CD69‑oxLDL ligand engagement induces Programmed Cell Death 1 (PD‑1) expression in human CD4 … Page 5 of 13 468 Immunohistochemical analysis were incubated in the presence or absence of nLDL, and copper-oxidized LDL at two different degrees of oxidation, Tissue sections (5 µm thick) were deparaffinised in xylene, moxLDL and hoxLDL. To focus in those genes modified rehydrated in graded ethanol solutions and subjected to heat- by hoxLDL in JKCD69 T cells, we used a cut-off of fold induced antigen retrieval either in 10 mM citrate buffer at change > 1.8 and we identified 577 upregulated and 238 pH 6 (for CD69 and NOR-1 detection) or in Tris–EDTA downregulated genes in this condition. Principal component buffer pH 9.0 (for PD-1 detection). Then, slides were treated analysis of these genes in all of the samples was represented with 3% hydrogen peroxide in methanol for 30 min to block (Fig. 1A). Statistical analysis showed clear differences in endogenous peroxidase activity. Subsequently, slides were gene expression profile depending on the degree of LDL incubated with 10% normal serum and endogenous avidin oxidation and the presence or absence of CD69 (Fig. 1A). and biotin blocked using a commercial kit (Vector Laborato- Native LDL-treated cells exhibited similar expression pro- ries, Burlingame, CA, USA). Then samples were incubated files to those of untreated cells (Fig. 1A). The differences overnight at 4 °C with the following primary antibodies displayed in the PCA analysis are reflected on the heatmap against: PD-1 (ab237728, Abcam) CD69 (sc-373799, Santa clustering (Fig. 1B). Heatmap suggests that JKCD69 cells Cruz Biotechnology, Inc.) or NOR-1(H00008013-M06, incubated with hoxLDL and moxLDL displayed a differ- Abnova). Samples were washed and then incubated with the ential behaviour compared to JKwt with the same LDLs appropriate biotinylated secondary antibodies for 1 h (Vec- (Fig. 1B). Interestingly, we identified PD-1 as one of the tor Laboratories, Burlingame, CA, USA). After washing genes with higher levels of induction in JKCD69 vs. JKwt three times in PBS, the avidin–biotin peroxidase complex in response to oxLDL (Table 1; Supplementary Data 1). As (ABC, Vector Laboratories) was added for 30 min. Colour an internal control, we identified increased expression of was developed by incubation with 3,3′-diaminobenzidine NR4A receptors, especially NR4A3, in JKCD69 treated with (DAB), and then, sections were counterstained with haema- hoxLDL compared to JKwt treated with hoxLDL or JKCD69 toxylin, dehydrated, cleared and mounted. Negative controls, cells treated or not with nLDL (Table 1). in which the primary antibody was omitted, were included to test for non-specific binding of the secondary antibody. OxLDL‑CD69 binding increases PD‑1 expression Statistical analyses To confirm that oxLDL binding to CD69 upregulates PD-1 in T cells, the expression of PD-1 was assessed by real-time Data were analysed with GraphPad Prism 7.04 (GraphPad PCR (RT-PCR) in JKCD69 and JKwt cells treated with hox- Software, San Diego, CA, USA). Significance differences LDL. PD-1 mRNA levels were significantly increased in were determined by two-way ANOVA when both cell lines JKCD69 after 4 h of incubation with oxLDL compared with were used with different treatments and one-way ANOVA JKwt cells, while nLDL did not induce PD-1 expression in to compare more than two means with one cell line. Tukey either cell line (Fig. 1C). Flow cytometry analysis showed or Bonferroni post hoc tests were used to perform multiple that oxLDL increased PD-1 surface expression in JKCD69 comparisons as indicated. To compare two means in interfer- compared to JKwt at 48 h, while no significant changes were ing experiments, unpaired t test and paired t test were used elicited by nLDL (Fig. 1D, E). Jurkat cells express very low when appropriate. Differences from human samples experi- or negligible levels of CD69 (Supplementary Fig. 1) which ments were analysed with Mann–Whitney test and Pearson is variable and dependent on the culture conditions and this product–moment correlation was applied. Differences were could explain the low levels of PD-1 induction (Fig. 1D). considered significant at P < 0.05. In all cases number of To evaluate the specificity of CD69, we assessed whether samples indicated at figure legends correspond to independ- the engagement of CD69 with specific antibodies triggers ent experiments. a similar response than oxLDL. JKwt and JKCD69 cells were cultured in the presence or absence of a CD69 anti- body for 4 h, and mRNA levels were analysed by RT-PCR. Results In JKCD69 cells, the antibody significantly increased PD-1 mRNA levels at 4 h (Fig. 2A). It also increased the OxLDL‑CD69 binding induces a differential amount of PD-1 detected on the cell membrane, with a peak transcriptional response in human CD4 + T expression at 48 h (Fig. 2B), similar to the induction effect lymphocytes observed with oxLDL. Moreover, the addition of anti-CD69 antibody after 3 h of stimulation with PMA/Io increased To identify genes differentially expressed in response to PD-1 expression in primary human CD4 + T cells from the binding of oxLDL to CD69, JKCD69 and JKwt cells healthy donors (Fig. 2C). The CD69 antibody also induced 13 468 Page 6 of 13 M. Jiménez‑Fernández et al. Fig. 1  OxLDL-CD69 binding induces a differential transcriptional JKwt treated with hoxLDL. Data are mean ± SD (n = 3) and were response increasing the PD-1 expression. A Principal component analysed with two-way ANOVA (Tukey post hoc test). *Indicates analysis (PCA) performed using the up- and downregulated genes significant differences (P < 0.05) with the treatment for each cell line from the RNA-Seq data of each incubation condition. The proportion (native vs. oxLDL), # JKCD69 vs. JKwt in the same condition, ◦ sig- of variability explained by each principal component (PC1 and PC2) nificant differences respect the previous time (45’ vs. 4 h, 4 h vs. 8 h). are included in the axis. Blue samples correspond to JKwt and red D Mean fluorescence intensity of PD-1 in both cell lines at different samples to JKCD69. (nLDL: native LDL; moxLDL: moderately oxi- time points. Data are mean ± SD (n = 3) and were analysed with two- dized LDL; hoxLDL. highly oxidized LDL). B The oxidation degree way ANOVA (Bonferroni post hoc test). *Indicates significant differ- of oxLDL induces a differential transcriptomic response. Heatmap ences (P < 0.05) in the treatment for each cell line (native vs oxLDL), based on the fold change of 577 upregulated and 242 downregulated # JKCD69 vs. JKwt in the same condition, ◦ significant differences genes in JKCD69 with the higher level of oxLDL. Data are shown respect the time. E Histograms of a representative experiment of the as Ln (x + 1) and are clustered using correlation distance and aver- induction of PD-1 in JKCD69 and JKwt after 48 h incubated with age linkage. Up- and downregulated genes are shown in red and blue LDLs respectively. C Relative expression of PD-1 mRNA in JKCD69 and the expression of NR4A1 and specially NR4A3 (Fig. 2D). transcriptional regulation of PD-1 [23]. Thus, we probed No significant effects on gene expression were observed in whether NFAT activation was responsible for the induc- T cells devoid of CD69 (JKwt). tion of PD-1 through CD69 binding. Treatment with CsA, which impairs the dephosphorylation of NFAT by cal- PD‑1 induction through CD69 engagement cineurin and its recruitment to the nucleus, prevented the is regulated by NFAT induction of PD-1 and the NR4A3 mRNA expression trig- gered by CD69 activation or the combination of PMA and CD69 crosslinking triggers a mobilization of ­C a 2+ ionomycin (Fig. 3A, B) [24]. However, CsA did not affect extracellular influx as well as the activation of NFAT NR4A1 mRNA expression induced by CD69 (Fig. 3C). In proteins [20–22]. Moreover, NFATc1 is involved in the order to determine the contribution of NR4A3 to the regu- lation of PD-1 expression, we inhibited NR4A3 expression 13 CD69‑oxLDL ligand engagement induces Programmed Cell Death 1 (PD‑1) expression in human CD4 … Page 7 of 13 468 Table 1  Potential genes of interest differentially expressed in Discussion JKCD69 vs. JKwt in response to the CD69-oxLDL binding Official gene symbol Fold change In this work, we identify a novel TCR-independent induction mechanism of PD-1 through the engagement of CD69 and PDCD1 4.91 oxLDL. RNA-Seq analysis of JKwt and JKCD69 cultured ENTPD1 3.33 with LDL oxidized to a different degree revealed a CD69- KCNMB1 3.09 dependent and LDL oxidation-dependent response. Cells ENPP1 2.64 left untreated or treated with native LDL displayed simi- NR4A3 2.41 lar expression profiles, while oxLDL produced differences TLR6 1.93 in gene expression that are more pronounced in JKCD69. NR4A1 1.82 These results underscore the specificity of CD69 for oxLDL PDGFC 0.65 and identified specific responses triggered by CD69-oxLDL TNFRSF4 0.64 binding. Differences elicited by lipoproteins based on their RETN 0.63 oxidation degree are in line with previous studies [25–27]. IL26 0.51 Interestingly, PD-1 was found to be one of the most over- CCR8 0.35 represented mRNA in JKCD69 cells treated with oxLDL. As expected, oxLDL markedly elevated the expression of NR4A3 and also NR4A1 [9]. In vitro assays confirmed the using a specific siRNA (Fig. 3D). Interestingly, silenc- mRNA and protein induction of PD-1 in JKCD69 in the ing NR4A3 slightly upregulated PD-1 expression upon presence of highly oxLDL. The specificity of the signal and induction with both anti-CD69 and PMA/Io stimulation the dependence of NR4A and PD-1 induction through CD69 in JKCD69 cells (Fig. 3E). in T cells was confirmed using an anti-CD69 monoclonal antibody. Interestingly, this finding was corroborated with PD‑1 and CD69 are highly expressed in inflamed primary human CD4 + T cells. human arteries PD-1 expression is transiently induced by activating stimuli such as PMA and ionomycin, concanavalin A, anti- To assess whether CD69 and PD-1 upregulation is associ- CD3/CD28 antibodies or by antigen–TCR engagement and ated with chronic vascular inflammatory disease, CD69, cytokine receptor engagement [28, 29]. Extracellular influx PD-1 and NR4A3 mRNA expression was analysed by RT- of ­Ca2+ induces the expression of NFAT regulated genes PCR in inflamed abdominal aortic samples from aneu- through calcineurin activation, including PD-1 [23]. Our rysm patients undergoing open repair surgery (n = 49) data highlight the role of NFAT as a transcriptional regulator and healthy abdominal aorta from multi-organ donors involved in the expression of PD-1 and NR4A3 induced by (n = 11). The clinical and demographic data of the patients CD69. However, the expression of NR4A1 was not affected and donors are collected in Table 2. CD69 and PD-1 were following NFAT inactivation. These results are in accord- highly expressed in inflamed aortas compared with donor ance with previous studies that evaluated the sensitivity of samples, and their mRNA levels showed a strong posi- the NR4A subfamily of nuclear receptors to NFAT pathway tive correlation (r = 0.655 p < 0.0001) (Fig. 4A). In addi- inhibitors in mice [30]. However, our data also revealed a tion, CD69 and PD-1 proteins levels were also increased slightly increased expression of PD-1 in NR4A3-silenced in diseased aortas, as determined by western blotting cells treated with anti-CD69 or with PMA/Io stimulation. (Fig. 4B). NR4A3 expression (mRNA and protein) was Accordingly, recent studies have confirmed a relationship also upregulated in inflamed aortas, as expected (Fig. 4B, between the expression of NR4A receptors and PD-1 in C). Furthermore, immunohistochemical analysis of con- T cells [31, 32]. Hence, the three NR4A proteins regulate secutive aortic sections revealed high CD69 and PD-1 the accessibility of an enhancer located at ~ 23 kb 5′ of the immunostaining in areas enriched in inflammatory cells Pdcd1 transcription start site (TSS) in a number of murine (Fig. 4D, left and middle panels), and a similar pattern models for exhausted CD8 + T cells [32]. The upregula- was observed for NR4A3 (Fig. 4D, right panels). PD-1 tion of hoxLDL-induced NR4A3 might be functioning as immunostaining was localized in areas of inflamed arter- a negative feedback loop to counter-regulate the expression ies enriched in CD3 + cells (Supplementary Fig. 2). of PD-1 induced by CD69. The PD-1/PD-L1 pathway contribution in cardiovascular diseases has been thoroughly studied. Pd-1 and Ldlr dou- ble knockout mice developed larger atherosclerotic lesions with higher CD4 + and CD8 + T cells and macrophages 13 468 Page 8 of 13 M. Jiménez‑Fernández et al. Fig. 2  CD69 engagement induces the PD-1 mRNA and protein control), # JKCD69 vs. JKwt in the same condition, ◦ significant dif- expression. A Relative expression of PD-1 mRNA induced by anti- ferences respect the previous time (24 h vs. 48 h, 48 h vs. 72 h). C CD69 TP1/55 antibody in JKCD69 after 4 h of stimulation. JKwt Left, percentage of primary human CD4 + T cells expressing PD-1 cells response was assessed as a control. An isotype control antibody after 48 h of stimulation. Data are mean ± SD (n = 5), and differ- was also used. Data are mean ± SD (n = 3) and **P < 0.01 differ- ences were determined with one-way ANOVA (Tukey post hoc test). ences were determined with two-way ANOVA (Tukey post hoc test). **P < 0.01. Right, histogram of a representative experiment after 48 h B Percentage of positive cells for PD-1 in both cell lines at different of treatment. D Relative expression of NR4A1 and NR4A3 mRNA time points after engagement of the CD69 antibody TP1/55. Data induced by anti-CD69 TP1/55 antibody in JKCD69 after 4 h. Data are mean ± SD (n = 3), and differences were analysed with two-way are mean ± SD (n = 3), and differences were analysed with two-way ANOVA (Tukey post hoc test). *Indicates significant differences ANOVA (Bonferroni post hoc test). **P < 0.01. ***P < 0.001 (P < 0.05) with the treatment for each cell line (antibody vs. isotype 13 CD69‑oxLDL ligand engagement induces Programmed Cell Death 1 (PD‑1) expression in human CD4 … Page 9 of 13 468 Fig. 3  NFAT regulates PD-1 expression induced by CD69 engage- of NR4A1 is independent of NFAT blockade. Relative expression of ment in T cells. A Cyclosporine A reduces the expression of NR4A1 in JKCD69 with anti-CD69 antibody, with or without CsA PD-1 + in T cells. JKCD69 cells were incubated with crosslinked and with CsA in resting cells incubated for 4 h. Data are mean ± SD anti-CD69 with or without cyclosporine A (CsA), or phorbol (n = 3) and were analysed with one-way ANOVA (Tukey post hoc 12-myristate 13-acetate and ionomycin (PMA/Io) stimulated with or test). *P < 0.05, **P < 0.01. D Relative expression levels of NR4A3 without CsA for 48 h. Data are mean ± SD (n = 3), and differences mRNA after siRNA knockdown in resting and anti-CD69 stimula- were determined by unpaired t test. * P < 0.05. B Cyclosporine A tion conditions. Data showed are mean ± SD (n = 3), and differences inhibits CD69-mediated NR4A3 induction. Relative expression of were determined by unpaired t test. *P < 0.05. ** P<0.01. E siRNA NR4A3 in JKCD69 incubated with anti-CD69 monoclonal antibody, interfering of NR4A3 expression enhances PD-1 + cell population. with or without CsA and with CsA in resting cells for 4 h. Data are JKCD69 were stimulated with anti-CD69 or PMA/Io for 48 h. Data mean ± SD (n = 3), and differences were analysed with one-way are mean ± SD (n = 3), and differences were determined by paired t ANOVA (Tukey post hoc test). * P < 0.05, ** P < 0.01. C Induction test. *P < 0.05 cell numbers [29]. PD-1 blockade increased atherosclerotic lesion inflammation in hypercholesterolemic Ldlr−/− mice, Table 2  Clinical features of patients and controls suggesting that PD-1 acts as negative regulator of inflam- Clinical parameters Controls (n = 11) Patients (n = 49) mation in atherosclerosis [29]. Accordingly, cardiovascular complications from immune checkpoint inhibitors therapies Age (years ± SEM) 61 ± 5.4 71.3 ± 0.66 as PD-1 and PD-L1 inhibitors are an emerging concern that Males (%) 91 100 merits further investigation [33]. Our analysis from inflamed Smoking (%)a 55 81.6 abdominal aortic samples revealed a strong correlation Hypertension (%) 64 67.3 between PD-1 and CD69 mRNA expression. Moreover, Diabetes (%) 36 10.2 more than half of the patients analysed had hyperlipidemia. Hyperlipidemia (%) 27 57.1 Immunohistochemical analyses revealed strong expression Ischaemic cardiomyopathy (%) 0 18.4 of PD-1, CD69 and NR4A3 in areas enriched in inflamma- SEM standard error of the mean tory cells, in which the regulation of PD-1 through CD69 a Current and ex-smokers 13 468 Page 10 of 13 M. Jiménez‑Fernández et al. 13 CD69‑oxLDL ligand engagement induces Programmed Cell Death 1 (PD‑1) expression in human CD4 … Page 11 of 13 468 ◂Fig. 4  PD-1 and CD69 are highly expressed in inflamed human arter- S2017/BMD-3671-INFLAMUNE-CM from the Comunidad de ies. A PD-1 and CD69 mRNA levels were measured in inflamed Madrid, a grant from the Ramón Areces Foundation “Ciencias de abdominal aortic samples (IAA) from patients undergoing open la Vida y la Salud”, “La Caixa” Banking Foundation (HR17-00016) repair surgery for abdominal aortic aneurysm (n = 49) and healthy to FSM; grants PDC2021-121719-I00 and PDI-2020-120412RB- abdominal aortic samples (HAA) from donors (n = 11). Data are I00 to FSM, and RTI2018-094727-B-100 to JMG funded by MCIN/ expressed as mean ± SD *P < 0.05 (Mann–Whitney test). (right panel) AEI/10.13039/501100011033 and by “ERDF A way of making Statistically significant positive correlation between CD69 mRNA Europe”; the Agència de Gestió d'Ajuts Universitaris i de Recerca levels and those corresponding to PD-1 in inflamed aortic samples (AGAUR) (2017-SGR-00333) to JMG; and a grant from the Insti- from patients. The Pearson Product Moment Correlation was applied. tuto de Salud Carlos III (PI18/0919) to CR. M. Jiménez-Fernández B CD69, PD-1 and NR4A3 protein levels were evaluated by west- is supported by a FPI-Severo Ochoa-CNIC (PRE2019-087941); C. ern blot in aortic lysates from IAA and HAA samples. Representa- Ballester-Servera is supported by a FPU fellowship (Ministerio de tive immunoblots are shown. C NR4A3 mRNA levels assessed in the Universidades). samples described in A). Data are expressed as mean ± SD. *P < 0.05 (Mann–Whitney test). D Immunohistochemical analysis of CD69, Data availability The data underlying this article are available in the PD-1 and NR4A3 in haematoxylin counterstained sections from IAA article and in its online Supplementary material. and HAA samples. The indicated areas are shown at high magnifi- cation (middle panels). Bar: 100 µm (upper and lower panels) and 50 µm (middle panels). Negative controls in which the primary anti- Declarations body was omitted are shown on the right Conflict of interest The authors have no conflict of interest to declare. and NR4A3 might contribute to this chronic vascular inflam- Ethical approval Written consent was obtained from all participating subjects. The procedure was approved by the Ethics Committee of matory disease. the Hospital de la Santa Creu i Sant Pau (Barcelona, Spain) and was The role of CD69 has also been studied in tumour conducted in accordance with the Declaration of Helsinki. immunology where it has been identified as a negative regulator of the anti-tumour responses. Cd69-deficient Consent for publication Consent to publish has been received from all participants. mice show a reduced tumour growth and the downreg- ulation of CD69 by monoclonal antibodies enhanced the anti-tumour immunity [34, 35]. In addition, CD69 Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- exerted a role in controlling immune inflammatory and tion, distribution and reproduction in any medium or format, as long autoimmune responses [2, 36]. Conceivably, the regula- as you give appropriate credit to the original author(s) and the source, tory role of CD69 in tumour and inflammatory responses provide a link to the Creative Commons licence, and indicate if changes could be mediated, at least in part, by modulation of PD-1 were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated expression. otherwise in a credit line to the material. If material is not included in Our study identifies a new function of CD69 by pro- the article's Creative Commons licence and your intended use is not moting PD-1 expression in CD4 + T lymphocytes after the permitted by statutory regulation or exceeds the permitted use, you will engagement with oxLDL that might be responsible for the need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. exacerbated activation state found in the absence of CD69 in the atherosclerosis model. The discovery of this regula- tion underscores the relevant role of these molecules in inflammatory cardiovascular diseases and provides new References knowledge for developing therapeutic strategies. 1. Cebrián M, Yagüe E, Rincón M, López-botet M, De Landázuri MO, Sánchez-madrid F (1988) Triggering of t cell proliferation Supplementary Information The online version contains supplemen- through aim, an activation inducer molecule expressed on acti- tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 00018-0​ 22-0​ 4481-1. vated human lymphocytes. J Exp Med 168:1621–1637. https://​ doi.​org/​10.​1084/​jem.​168.5.​1621 Acknowledgements We would like to thank Dr. Ana Dopazo and the 2. González-Amaro R, Cortés JR, Sánchez-Madrid F, Martín P Genomic Unit of CNIC for their technical assistance with RNA-seq, (2013) Is CD69 an effective brake to control inflammatory dis- Jorge Díaz-Rullo for statistical support and Dr. Miguel Vicente Man- eases? Trends Mol Med 19:625–632. https://​doi.​org/​10.​1016/j.​ zanares for editing the manuscript. molmed.​2013.​07.​006 3. Radulovic K, Manta C, Rossini V, Holzmann K, Kestler HA, Author contributions MJF, CRS, HF, JMG and FSM designed the Wegenka UM, Nakayama T, Niess JH (2012) CD69 regulates research; MJF, CRS, LC, CBS, AV and SR performed the experiments; Type I IFN-induced tolerogenic signals to mucosal CD4 T cells MJF, CRS, HF, JMG and FSM analysed and interpreted the results. that attenuate their colitogenic potential. J Immunol 188:2001– MJF drafted the manuscript with feedback from all the authors, who 2013. https://​doi.​org/​10.​4049/​jimmu​nol.​11007​65 approved the submitted version. 4. Sancho D, Gómez M, Viedma F, Esplugues E, Gordón-Alonso M, Angeles García-López M, de la Fuente H, Martínez-A C, Lauzurica P, Sánchez-Madrid F (2003) CD69 downregulates Funding Open Access funding provided thanks to the CRUE-CSIC autoimmune reactivity through active transforming growth agreement with Springer Nature. This work was supported by grant 13 468 Page 12 of 13 M. Jiménez‑Fernández et al. factor-β production in collagen-induced arthritis. J Clin Invest tyrosine hydroxylase for abdominal aortic aneurysm: impact on 112:872–882. https://​doi.​org/​10.​1172/​JCI20​03191​12 inflammation, oxidative stress, and vascular remodeling. Hyper- 5. Cortés JR, Sánchez-Díaz R, Bovolenta ER, Barreiro O, Lasarte tension 78:681–692. https://​d oi.​o rg/​1 0.​1 161/​H YPER​T ENSI​ S, Matesanz-Marín A, Toribio ML, Sánchez-Madrid F, Martín ONAHA.​121.​17517 P (2014) Maintenance of immune tolerance by Foxp3+ regula- 19. Clemente C, Rius C, Alonso-Herranz L, Martín-Alonso M, Pol- tory T cells requires CD69 expression. J Autoimmun 55:51–62. lán Á, Camafeita E, Martínez F, Mota RA, Núñez V, Rodríguez https://​doi.​org/​10.​1016/j.​jaut.​2014.​05.​007 C, Seiki M, Martínez-González J, Andrés V, Ricote M, Arroyo 6. Radstake TRDJ, van Bon L, Broen J, Wenink M, Santegoets AG (2018) MT4-MMP deficiency increases patrolling monocyte K, Deng Y, Hussaini A, Simms R, Cruikshank WW, Lafyatis recruitment to early lesions and accelerates atherosclerosis. Nat R (2009) Increased frequency and compromised function of Commun. https://​doi.​org/​10.​1038/​s41467-​018-​03351-4 T regulatory cells in systemic sclerosis (SSc) is related to a 20. Bikah G, Pogue-Caley RR, McHeyzer-Williams LJ, McHeyzer- diminished CD69 and TGFβ expression. PLoS One 4:e5981. Williams MG (2000) Regulating T helper cell immunity through https://​doi.​org/​10.​1371/​journ​al.​pone.​00059​81 antigen responsiveness and calcium entry. Nat Immunol 1:402– 7. Yu L, Yang F, Zhang F, Guo D, Li L, Wang X, Liang T, Wang 412. https://​doi.​org/​10.​1038/​80841 J, Cai Z, Jin H (2018) CD69 enhances immunosuppressive 21. D’Ambrosio D, Trotta R, Vacca A, Frati L, Santoni A, Gulino A, function of regulatory T-cells and attenuates colitis by prompt- Testi R (1993) Transcriptional regulation of interleukin-2 gene ing IL-10 production. Cell Death Dis. https://​doi.​org/​10.​1038/​ expression by CD69-generated signals. Eur J Immunol 23:2993– s41419-​018-​0927-9 2997. https://​doi.​org/​10.​1002/​eji.​18302​31140 8. Sánchez-Díaz R, Blanco-Dominguez R, Lasarte S, Tsilingiri K, 22. Rao A, Luo C, Hogan PG (1997) Transcription factors of the Martín-Gayo E, Linillos-Pradillo B, de la Fuente H, Sánchez- NFAT family: regulation and function. Annu Rev Immunol Madrid F, Nakagawa R, Toribio ML, Martín P (2017) Thymus- 15:707–747. https://​doi.​org/​10.​1146/​annur​ev.​immun​ol.​15.1.​707 derived regulatory T cell development is regulated by C-Type 23. Oestreich KJ, Yoon H, Ahmed R, Boss JM (2008) NFATc1 lectin-mediated BIC/microRNA 155 expression. Mol Cell Biol regulates PD-1 expression upon T cell activation. J Immunol 37:1–18. https://​doi.​org/​10.​1128/​mcb.​00341-​16 181:4832–4839. https://d​ oi.o​ rg/1​ 0.4​ 049/J​ IMMUN ​ OL.1​ 81.7.4​ 832 9. Tsilingiri K, De La Fuente H, Relaño M, Sánchez-Díaz R, Rod- 24. Fruman DA, Klee CB, Bierer BE, Burakoff SJ (1992) Calcineurin ríguez C, Crespo J, Sánchez-Cabo F, Dopazo A, Alonso-Lebrero phosphatase activity in T lymphocytes is inhibited by FK 506 and JL, Vara A, Vázquez J, Casasnovas JM, Alfonso F, Ibáñez B, cyclosporin A. Proc Natl Acad Sci 89:3686–3690. https://​doi.​org/​ Fuster V, Martínez-González J, Martín P, Sánchez-Madrid F 10.​1073/​pnas.​89.9.​3686 (2019) Oxidized low-density lipoprotein receptor in lympho- 25. DamiaÂn-Zamacona S, Toledo-Ibelles P, Ibarra-Abundis M, cytes prevents atherosclerosis and predicts subclinical disease. Uribe-Figueroa L, HernaÂndez-Lemus E, Macedo-Alcibia KP, Circulation 139:243–255. https://​d oi.​o rg/​1 0.​1 161/​C IRCU​ Delgado-Coello B, Mas-Oliva J, Reyes-Grajeda JP (2016) Early LATIO​NAHA.​118.​034326 transcriptomic response to ldl and oxldl in human vascular smooth 10. Wherry EJ (2011) T cell exhaustion. Nat Immunol 12:492–499. muscle cells. PLoS One. https://​doi.​org/​10.​1371/​journ​al.​pone.​ https://​doi.​org/​10.​1038/​ni.​2035 01639​24 11. Martín P, Blanco-Domínguez R, Sánchez-Díaz R (2021) Novel 26. Lara-Guzmán OJ, Gil-Izquierdo Á, Medina S, Osorio E, Álvarez- human immunomodulatory T cell receptors and their double- Quintero R, Zuluaga N, Oger C, Galano J-M, Durand T, Muñoz- edged potential in autoimmunity, cardiovascular disease and Durango K (2018) Oxidized LDL triggers changes in oxidative cancer. Cell Mol Immunol 18:919–935. https://​doi.​org/​10.​1038/​ stress and inflammatory biomarkers in human macrophages. s41423-​020-​00586-4 Redox Biol 15:1–11. https://​doi.​org/​10.​1016/j.​redox.​2017.​11.​017 12. Qin W, Hu L, Zhang X, Jiang S, Li J, Zhang Z, Wang X (2019) 27. Hermansson A, Ketelhuth DFJ, Strodthoff D, Wurm M, Hansson The diverse function of PD-1/PD-L pathway beyond cancer. EM, Nicoletti A, Paulsson-Berne G, Hansson GK (2010) Inhibi- Front Immunol. https://​doi.​org/​10.​3389/​fimmu.​2019.​02298 tion of T cell response to native low-density lipoprotein reduces 13. Sharpe AH, Wherry EJ, Ahmed R, Freeman GJ (2007) The atherosclerosis. J Exp Med 207:1081–1093. https://​doi.​org/​10.​ function of programmed cell death 1 and its ligands in regu- 1084/​jem.​20092​243 lating autoimmunity and infection. Nat Immunol 8:239–245. 28. Gianchecchi E, Fierabracci A (2018) Inhibitory receptors and https://​doi.​org/​10.​1038/​ni1443 pathways of lymphocytes: the role of PD-1 in Treg development 14. Calvayrac O, Rodríguez-Calvo R, Alonso J, Orbe J, Martín- and their involvement in autoimmunity onset and cancer progres- Ventura JL, Guadall A, Gentile M, Juan-Babot O, Egido J, Beloqui sion. Front Immunol 9:1–12. https://d​ oi.o​ rg/1​ 0.3​ 389/fi ​ mmu.2​ 018.​ O, Paramo JA, Rodríguez C, Martínez-González J (2011) CCL20 02374 is increased in hypercholesterolemic subjects and is upregulated 29. Bu DX, Tarrio M, Maganto-Garcia E, Stavrakis G, Tajima G, by LDL in vascular smooth muscle cells. Arterioscler Thromb Lederer J, Jarolim P, Freeman GJ, Sharpe AH, Lichtman AH Vasc Biol 31:2733–2741. https://​doi.​org/​10.​1161/​ATVBA​HA.​ (2011) Impairment of the programmed cell death-1 pathway 111.​235721 increases atherosclerotic lesion development and inflammation. 15. Martin M (2011) Cutadapt removes adapter sequences from high- Arterioscler Thromb Vasc Biol 31:1100–1107. https://​doi.​org/​10.​ throughput sequencing reads. EMBnet J 17:10–12. https://d​ oi.o​ rg/​ 1161/​ATVBA​HA.​111.​224709 10.​14806/​EJ.​17.1.​200 30. Jennings E, Elliot TAE, Thawait N, Kanabar S, Yam-Puc JC, Ono 16. Li B, Dewey CN (2011) RSEM: accurate transcript quantification M, Toellner K-M, Wraith DC, Anderson G, Bending D (2020) from RNA-Seq data with or without a reference genome. BMC Nr4a1 and Nr4a3 reporter mice are differentially sensitive to T Bioinformatics 12:1–16. https://​doi.​org/​10.​1186/​1471-​2105-​12-​ cell receptor signal strength and duration. Cell Rep 33:108328. 323/​TABLES/6 https://​doi.​org/​10.​1016/j.​celrep.​2020.​108328 17. Robinson MD, McCarthy DJ, Smyth GK (2010) edgeR: a Biocon- 31. Liu X, Wang Y, Lu H, Li J, Yan X, Xiao M, Hao J, Alekseev ductor package for differential expression analysis of digital gene A, Khong H, Chen T, Huang R, Wu J, Zhao Q, Wu Q, Xu S, expression data. Bioinformatics 26:139–140. https://​doi.​org/​10.​ Wang X, Jin W, Yu S, Wang Y, Wei L, Wang A, Zhong B, Ni 1093/​BIOIN​FORMA​TICS/​BTP616 L, Liu X, Nurieva R, Ye L, Tian Q, Bian X-W, Dong C (2019) 18. Cañes L, Alonso J, Ballester-Servera C, Varona S, Escudero JR, Genome-wide analysis identifies NR4A1 as a key mediator of Andrés V, Rodríguez C, Martínez-González J (2021) Targeting 13 CD69‑oxLDL ligand engagement induces Programmed Cell Death 1 (PD‑1) expression in human CD4 … Page 13 of 13 468 T cell dysfunction. Nature 567:525–529. https://​doi.​org/​10.​1038/​ 35. Mita Y, Kimura MY, Hayashizaki K, Koyama-Nasu R, Ito T, s41586-​019-​0979-8 Motohashi S, Okamoto Y (2018) Nakayama T Crucial role of 32. Chen J, López-Moyado IF, Seo H, Lio C-WJ, Hempleman LJ, CD69 in anti-tumor immunity through regulating the exhaustion Sekiya T, Yoshimura A, Scott-Browne JP, Rao A (2019) NR4A of tumor-infiltrating T cells. Int Immunol. https://d​ oi.o​ rg/1​ 0.1​ 093/​ transcription factors limit CAR T cell function in solid tumours. intimm/​dxy050 Nature 567:530–534. https://​doi.​org/​10.​1038/​s41586-​019-​0985-x 36. Sancho D, Gómez M, Sánchez-Madrid F (2005) CD69 is an 33. Hu JR, Florido R, Lipson EJ, Naidoo J, Ardehali R, Tocchetti immunoregulatory molecule induced following activation. Trends CG, Lyon AR, Padera RF, Johnson DB, Moslehi J (2019) Cardio- Immunol 26:136–140. https://​doi.​org/​10.​1016/j.​it.​2004.​12.​006 vascular toxicities associated with immune checkpoint inhibitors. Cardiovasc Res 115:854–868. https://​doi.​org/​10.​1093/​cvr/​cvz026 Publisher's Note Springer Nature remains neutral with regard to 34. Esplugues E, Sancho D, Vega-Ramos J, Martínez-A C, Syrbe jurisdictional claims in published maps and institutional affiliations. U, Hamann A, Engel P, Sánchez-Madrid F, Lauzurica P (2003) Enhanced antitumor immunity in mice deficient in CD69. J Exp Med 197:1093–1106. https://​doi.​org/​10.​1084/​JEM.​20021​337 13