Thyroid hormones contribute to JAK/STAT pathway abnormal activation, promoting T-cell lymphoma dissemination Open Access

T-cell lymphomas (TCLs) are a heterogeneous group of clinically aggressive hematological malignancies derived from T lymphocytes, typically characterized by poor outcomes because first-line chemotherapy regimens are associated with a high failure rate and frequent relapses.^1-3 Therefore, there is a critical need to study TCL biology and molecular pathogenesis, as well as the biological factors and pathways driving TCL progression.

Disorders involving the abnormal activation of the JAK/STAT pathway were identified in mostly all T-cell leukemia/lymphoma subtypes.^4-6 Ruxolitinib (Ruxo), an US Food and Drug Administration–approved JAK 1/2 inhibitor used to treat various blood disorders,⁷ demonstrated activity across several peripheral TCL subtypes, supporting the use of JAK/STAT inhibition as a targeted therapy for these patients. However, its use has limitations because most TCL responses to Ruxo are partial and transient, and, eventually, lymphoma cells acquire resistance to JAK inhibition.^7,8 Because activating STAT and JAK mutations are present in only a portion of samples from patients with TCL, additional mechanisms for the aberrant activation of these oncogenic pathways have been suggested, involving the action of external factors present in the tumor microenvironment, such as cytokines and hormones.^5,9-14 Thyroid hormones (THs) regulate cell metabolism, growth, and other physiological processes through the binding of 3,5,3′-triiodo-l-thyronine (T3) to its nuclear receptor thyroid hormone nuclear receptor that activates the transcription of target genes containing a TH-response element.¹⁵ In addition, both T3 and T4 (L-thyroxine) actions on integrin αvβ3 trigger signaling pathways in cancer cells,¹⁶ influencing cell proliferation, apoptosis, and invasiveness in solid^17-19 and hematological malignancies.^20-22 Specifically for malignant T cells, we have shown that THs contribute to their malignant behavior by activating transcriptional programs related to cell proliferation and angiogenesis, mostly by their action on integrin αvβ3.²¹ Also, we have demonstrated that the inhibition of integrin αvβ3 with cilengitide (Cile) improves the antilymphoma activity of bexarotene (Bex),²² a synthetic rexinoid used for cutaneous TCL treatment.

Although a connection between THs and JAK/STAT signaling has been described in physiological and pathological processes,^17,23 their role in TCL progression and dissemination has never been explored fully. Herein, we characterized the molecular mechanism by which THs contribute to the abnormal activation of JAK/STAT pathway in lymphoma cells representing different TCL subtypes. Also, we explored the mechanisms of action of Cile and Bex combination using preclinical TCL in vivo models and proteomic analysis. Using an experimental metastasis formation model, we show that the inhibition of THs’ actions on the integrin αvβ3 receptor in combination with Bex could be a potential noncytotoxic therapeutic option to diminish JAK/STAT pathway activation and lymphoma dissemination.

Murine TCL cell line EL4 was obtained from the American Type Culture Collection. Human TCL cell lines CUTLL1, OCI-Ly12, and OCI-Ly13.2, were obtained from Adolfo Ferrando (Columbia University, New York, NY) or the Ontario Cancer Institute, Canada. To evaluate the effects of THs (THs = T3 + T4), the cells were treated with a combination of T4 and T3, at a final concentration of 100 nM and 1 nM, respectively (Sigma-Aldrich). For STAT3 and JAK1/2 pathway inhibition, the cells were preincubated for 90 minutes with the inhibitors, followed by the corresponding treatments. Cryptotanshinone, Ruxo, Bex, and Cile were purchased from MedChemExpress.

Animals were bred and kept at the Institute for Biomedical Research (BIOMED, Buenos Aires, Argentina) following the ARRIVE (animal research: reporting of in vivo experiments) guidelines (Kilkenny et al²⁴). All experimental protocols were approved by the institutional committee for the care and use of laboratory animals, BIOMED. The syngeneic in vivo EL4 subcutaneous and experimental metastasis formation models were developed as previously described^22,25 using 6 to 8 weeks male C57BL/6J mice. To generate solid tumors, mice were inoculated subcutaneously with 3 × 10⁵ EL4 cells, as previously described.²² After 12 days, tumors from the different groups of treatments were analyzed by fluorescence-activated cell sorting, western blot, gelatinolytic zymography, and proteomics analysis. For the experimental metastasis test, mice were inoculated through the tail vein with 3 × 10⁵ EL4 cells. After 10 days, mice were euthanized, organs were removed and fixed in 3.7% paraformaldehyde, and tumor foci were counted. In both in vivo models mice, were randomized into 4 treatment arms: vehicle (Veh; 35% polyethylene glycol 300, 5% Tween-80, and 65% dextrose 5% in water), Bex with levothyroxine (0.5 mg per mouse per day and 150 ng per mouse per day, respectively), Cile (0.5 mg per mouse per day), and Bex with levothyroxine plus Cile. For Human CUTLL1 xenograft model, we used 6- to 8-week-old female and male immunocompromised mice (NOD.Cg-Prkdc^scidIl2rg^tm1Wjl/SzJ [NSG]) that were originally obtained from The Jackson Laboratories (Bar Harbor, ME). NSG mice were inoculated through the tail vein with 5 × 10⁶ CUTLL1 cells, and after 7 days of treatment with Veh or Bex with levothyroxine plus Cile, mice were euthanized, organs were removed and fixed in 3.7% paraformaldehyde, and tumor foci were counted.

Ten segments of lymphoma explants (1 mm³) grown as organotypic cultures from the different groups of treatments were cultured in 1 mL of serum-free RPMI 1640 for 24 hours. Lymphoma explants were then collected, lyophilized, and processed for proteomic analysis. A detailed protocol description is provided in the supplemental material. Picklist choice, protein identification, quantification, and validation were performed using the MaxQuant platform²⁶ (version 2.2.0), which includes the algorithm Andromeda as previously described.²⁷ Perseus software²⁸ (version 2.0.11) was used to perform the differentially expressed protein analysis between each condition (vs Veh). The level of analysis stringency was defined as a minimum of 2 valid values in at least 1 group. A 2-sample Student t test analysis with a P value < .05 was applied. The software Funrich²⁹ (version 3.1.4) was used for the construction of Venn diagrams.

The means of the different experimental groups were analyzed for statistical significance using the GraphPad Prism 4.0 (GraphPad Software) software, using an unpaired 2-tailed Student t test or 2-way analysis of variance followed by Tukey analysis. Differences between means were considered significant if P value <.05. Results are expressed as mean ± standard error of the mean.

To study whether THs are one of the factors that contribute to the aberrant activation of the JAK/STAT pathway, we analyzed by western blot the phosphorylation levels of STATs in TCL cell lines corresponding to immature (CUTLL1 and EL4) and mature (OCI-LYy2, OCI-Ly13.2, and HuT78) subtypes after treatment with physiological levels of THs. We found that, regardless of the JAK/STAT mutational status (see the supplemental Methods), physiological levels of THs led to an increase of 16% to 38% in STAT1, STAT3, and STAT5 phosphorylation in TCL cells after 15 minutes of treatment (Figure 1A; supplemental Figure 1A). In addition, we found that preincubation with Ruxo diminished THs-induced STATs phosphorylation, verifying the upstream involvement of JAK1/2 tyrosine kinase proteins (Figure 1A; supplemental Figure 1A). Despite the lack of statistical significance in some cases, all analyzed cell lines exhibited a consistent trend toward reduced phosphorylation. To analyze the mechanisms by which THs regulate the JAK/STAT pathway, we incubated THs-treated cells with the integrin αvβ3 inhibitor, Cile. We found that Cile significantly inhibited THs-induced STATs phosphorylation, revealing the upstream participation of integrin αvβ3 receptor in these effects (Figure 1B; supplemental Figure 1B). Because the activation of STAT transcription factors (TFs) mediates the expression of genes related to lymphoma progression, we analyzed the transcriptional expression of GATA3, CCR4, CCND1, MMP2, and MMP9 upon THs treatment. Overall, THs significantly increased messenger RNA (mRNA) levels of STAT target genes after 6 hours of treatment across the cell lines analyzed. Exceptions to this trend included MMP2 in OCI-LY12 and GATA3 in EL4 cells (Figure 1C). In addition, we evaluated the effects of Ruxo and Cile on TH-mediated transcriptional regulation. In most cases, both inhibitors effectively reduced the transcriptional response to THs, despite some degree of variability across the cell lines tested. These differences may reflect the underlying biological heterogeneity of the TCL cell lines.

It has been reported that the aberrant activation of the JAK/STAT pathway is associated with TCL progression; particularly, STAT3 regulates the matrix metalloproteinases 2 and 9 (MMP2 and MMP9) expression, and their activity is associated with cancer progression, including lymphoma.^17,23 We thus evaluated by gelatin zymography the activity of MMP2 and MMP9 in TCL cell supernatants after 24 hours treatment. We found that, except for OCI-LY12 cells, all TCL cell lines exhibited a significant increase in MMP2 and MMP9 activity after 24 hours of TH treatment, as determined by gelatin zymography (Figure 1D; supplemental Figure 1C). Notably, both the STAT3 inhibitor cryptotanshinone and Cile significantly reduced this effect, demonstrating the implication not only of STAT3 but also of the TH membrane receptor in the induction of MMP activity (Figure 1D-E).

Recent findings revealed that Ruxo is effective across several TCL subtypes.⁸ However, this response was limited as with other single agents, emphasizing the need to explore new combinatorial strategies for effective and noncytotoxic TCL therapies. Bex is an RXR-selective retinoid approved by the US Food and Drug Administration for cutaneous TCL.^30,31 However, the Bex activity in noncutaneous TCL subtypes remains poorly studied. We thus evaluated Bex in vitro activity on our TCL cell lines and compared it with Ruxo. Likewise, the effects on cutaneous TCL (CTCL) cells, Bex significantly reduced cell viability by 10% to 25% and induced apoptosis by 25% to 80% in all the TCL cell lines (Figure 2A-B, respectively). We also found that the combination with Cile resulted in an improvement of Bex antilymphoma activity. Furthermore, we noted that Ruxo significantly decreases TCL cell viability, however, its effectiveness remained lower than that observed with the combination of Bex and Cile (Figure 2A). In addition, the induction of apoptosis by the Bex and Cile combination was significantly greater in all the TCL cell lines examined compared with Ruxo alone or when combined with the aforementioned drugs (Figure 2B). To further explore the underlying mechanisms, we assessed cell cycle distribution in response to the treatments. However, we observed no significant changes, suggesting that alterations in cell cycle progression do not account for the reduced viability (supplemental Figure 2).

We previously found that the pharmacologic inhibition of integrin αVβ3 with Cile improved the antineoplastic effect of Bex on the EL4 in vivo tumor growth.²² Using the same TCL syngeneic mice model, we further investigated whether this combinatory approach affects the JAK/STAT pathway in vivo. It is important to mention that Bex induces hypothyroidism and thus requires the concomitant administration of levothyroxine^30,31 (T4) to maintain tissue and circulating physiological levels of THs. When the implanted TCL tumors reached 75 mm³, mice were randomized into 4 treatment groups to receive Veh, Bex with T4 supplementation (BexT4+), Cile, and their combination (BexT4+Cile) according to the scheme shown in Figure 3A. We first analyzed tumor STAT1, 3, and 5 phosphorylation by western blot and found nonsignificant differences in tumors of the different treatments (data not shown). However, when we analyzed the EL4 tumor cell population by fluorescence-activated cell sorting, we observed that phosphorylated STAT1 and STAT5 were significantly decreased in BexT4+Cile tumors (vs Veh; Figure 3B). Moreover, we found that tumor-MMP2 and MMP9 activity was significantly inhibited by BexT4+Cile but not by each drug alone (Figure 3C; supplemental Figure 2A). To confirm that this effect was a result of the specific action of Bex and Cile on TCL cells, we analyzed the gelatinolytic activity of EL4 supernatants after 24 hours of treatment and found that MMP2 and MMP9 activities were significantly decreased when compared with Veh (Figure 3D; supplemental Figure 2B). These data suggest that the combination of Bex and Cile could be a potential noncytotoxic therapeutic option to inhibit the aberrant activation of the JAK/STAT pathway and factors such as MMP2 and MMP9, which are related to tumor progression and dissemination in patients with TCL.

Because proteomics profile analysis provides unique insights into tumor biology and could reveal protein biomarkers of lymphoma vulnerabilities, we decided to deepen our previous in vivo findings using liquid chromatography–tandem mass spectrometry. We characterized and compared the TCL tumor proteomic landscapes after 12 days of treatment (Figure 4; supplemental Figure 3A). The experimental workflow and a Venn diagram representing the total number of proteins that were identified (Veh, 2559; BexT4+, 2239; Cile, 2639; and BexT4+Cile, 2627) and the overlap among each treatment are shown in Figure 4A. We then analyzed differential protein expression to determine the proteomic profiles induced by the different treatments (vs Veh) and only proteins that were differentially upregulated and downregulated at P value <.05 were included in the study. In BexT4+ tumors, 160 proteins were identified as differentially expressed, of which 78 were upregulated and 82 were downregulated (Figure 4B-C). In Cile-treated tumors, 241 proteins were found as differentially expressed, of which 115 were upregulated and 126 downregulated. In case of BexT4+Cile-treated tumors, we identified 430 differentially expressed proteins, among them, 214 were upregulated and 216 were downregulated. Because our major goal was to strengthen the rationale for proposing the combination of Bex and Cile as a potential TCL therapy, we focused our analysis on the BexT4+Cile-induced proteome. Differentially expressed proteins with relevant functionalities were selected, and a heat map was generated (Figure 4D). We found that BexT4+Cile treatment significantly modulated the expression of proteins associated with biological processes such as cell proliferation^32,33 (branched chain amino acid transaminase 1, RNA binding motif protein, X-linked-like 1, and signal transducing adaptor family member 1), angiogenesis/migration^34,35 (fibronectin 1-Anastellin, radixin [RDX], ezrin [EZR], and moesin [MSN]), and metabolism^36-38 (succinate-CoA ligase [GDP-forming] subunit alpha, mitochondrial, enolase 1 [ENO1], and transaldolase 1), among others. In addition, we performed a TF enrichment analysis to determine the enrichment of TFs target genes among the differentially expressed proteins (Figure 4E). These TFs were ranked based on their enrichment scores and statistical significance (supplemental Figure 3C-D). Interestingly, we found that BexT4+Cile downregulates proteins associated with TFs involved in angiogenesis, metastasis, and cancer progression such as FUBP1, ENO1, YBX1, DNAJC2, HNRNPK, and KHSRP.^39-42 Based on the biological processes modulated in the BexT4+Cile proteomic profile, we evaluated the impact of the combined treatment on TCL metastatic dissemination using both syngeneic (EL4) and xenograft (CUTLL1) in vivo experimental metastasis models (Figure 5A). In the syngeneic model, BexT4+Cile treatment significantly reduced the number of TCL metastatic foci in both the liver and kidneys (Figure 5B-C) compared with Veh-treated controls. In the xenograft model using CUTLL1 cells, lungs from BexT4+Cile-treated mice exhibited a statistically significant reduction in visible metastatic foci (Figure 5D-E). In the kidneys, Veh-treated mice showed a congested and mottled surface with scattered micronodules, whereas no visible lesions were observed in BexT4+Cile-treated animals (Figure 5D). No macroscopic changes were observed in the liver (data not shown). Microscopically, BexT4+Cile treatment showed a trend toward reduced neoplastic infiltration in both lungs and kidneys, although this did not reach a statistical significance (Figure 5E; supplemental Figure 4B). No differences in liver infiltration were observed between treatment groups.

To determine the therapeutic potential of our findings for the treatment of patients with TCL we evaluated whether there is an association between THs receptor expression, JAK/STAT pathway, and clinical outcome using a public data set of samples from patients with TCL (GSE58445, n = 193). We first divided patient samples into ITGB3 or ITGAV high and low mRNA levels groups and performed a TF enrichment analysis. We found that patients with high ITGAV levels are enriched in TFs involved in angiogenesis, migration, and metastasis, such as hypoxia-inducible factor 1, early growth response 1, and activator protein 1 (Figure 6A). Also, we observed that patients with high ITGB3 levels are enriched in target genes of STAT1, 3, and 5B (Figure 6B). Moreover, we found that the high mRNA expression of ITGB3 significantly correlates with a lower overall survival rate (Figure 6C). Finally, we examined the mRNA expression levels of several proteins with relevant biological functions significantly regulated by the combination of Bex and Cile in our in vivo model. Interestingly, we observed that patients with high expression levels of BexT4+Cile–downregulated proteins with implications in cancer progression, such as RDX, MSN, and ataxin 2-like, among others, correlated with worse overall survival (Figure 6D; supplemental Figure 4).

Overall, our findings support the potential of Cile and Bex use as a promising combinatorial therapeutic approach for more effective and less cytotoxic TCL treatment.

Our work reveals that THs are important components of the tumor microenvironment that can contribute to the abnormal activation of the JAK/STAT oncogenic pathways in TCL cells by activating their membrane receptor, integrin αvβ3. We characterized the mechanisms underlying these effects, including JAK1/2 activation, STAT1, 3, and 5 phosphorylation, the upregulation of the transcription levels of STAT target genes, and the induction of metalloprotease activity (Figure 7, mechanism of action 1). Moreover, we found that the pharmacological inhibition of integrin αvβ3 in combination with Bex decreased in vivo TCL dissemination by inducing a proteomic profile that reduces biological processes such as cell proliferation, angiogenesis, metastasis, and lymphoma progression (Figure 7, mechanism of action 2).

JAK/STAT oncogenic pathway plays a central role in cell proliferation, differentiation, and survival by transmitting signals from the extracellular environment to the nucleus. In this study, we showed that THs, by acting through the integrin αvβ3, induced STATs phosphorylation and promoted the transcription of STAT-target genes associated with lymphoma progression^43-46 such as GATA3, CCR4, CCND1, MMP2, and MMP9. Additionally, we verified that, except for OCI-Ly12 cells, THs not only increase mRNA levels but also the activity of MMP2 and MMP9 in the supernatants of TCL cells, an effect that was diminished by the αvβ3 integrin pharmacological inhibitor, Cile. These metalloproteinases are cancer-associated secreted endopeptidases involved in cell migration, invasion, inflammation, and angiogenesis,⁴⁷ and were proposed as poor prognosis factors for patients with lymphoma, highlighting the relevance of Cile activity. The absence of MMP2/9 activity modulation in OCI-LY12 cells may reflect the biological heterogeneity within peripheral not otherwise specified TCL (PTCL–NOS) subtypes. OCI-LY12 cells are representative of the TBX21-driven subtype, which exhibits constitutive activation of NF-κB and JAK-STAT pathways, potentially masking further modulation of MMPs by THs. Moreover, the TCL cell lines used in this study display distinct JAK and STAT mutational statuses, which likely contribute to the heterogeneous responses to THs. Additionally, increased CCR4 expression observed in OCI-LY12 cells has been linked to angiogenesis in CTCL,⁴⁸ suggesting alternative mechanisms of microenvironment remodeling beyond MMP activity.

As highlighted above, activation of JAK/STAT signaling is the unifying feature in samples from patients with TCL,⁴ thus JAKs have emerged as potential therapeutic targets.^49,50 Although it was demonstrated in a phase 2 study, that Ruxo was active across various peripheral TCL subtypes with phosphorylated STAT expression, limitations for its use were found: (1) dose escalation is restricted because of its potential to cause anemia and thrombocytopenia through JAK2 inhibition, and (2) as many single agents, Ruxo response is transient and patients developed resistance in a mechanism that may involve activation of MAPK/phosphoinositide 3-kinase (PI3K) signaling.⁵¹ It was recently reported that the transcriptome of B16 melanoma cells was enriched in genes associated with PI3K-AKT and MAPK signaling after Cile treatment.⁵² Additionally, it was described that Cile inhibits the phosphorylation of focal adhesion kinase, Src, and Akt in glioma cells,⁵³ suggesting that integrin αvβ3 inhibition could potentially provide the benefit of reducing the abnormal activation of both JAK/STAT, MAPK, and PI3K signaling pathways.

The antitumor activity of Cile was evaluated in multiple clinical trials⁵⁴ for cancer treatment. However, cancer cells often develop resistance to single-target drugs, leading to suboptimal therapeutic outcomes. To enhance treatment efficacy, combining Cile with other therapies has proven to be a more effective approach. As an alternative for a less toxic and more effective multitargeted therapeutic option for patients with phosphorylated STAT–expressing T-cell malignancies, we demonstrated that the synthetic rexinoid Bex has antilymphoma activity in immature and mature TCL cells and that its combination with Cile has greater effects when compared with Ruxo. Bex has demonstrated efficacy in CTCL, although its activity in PTCL has been less explored. A series of clinical cases has reported on the use of Bex as monotherapy in patients with PTCL,⁵⁵ demonstrating that it is well tolerated and shows promising activity. Moreover, in 2024, the Pharmaceuticals and Medical Devices Agency of Japan approved Bex for the treatment of adult T-cell leukemia/lymphoma with skin lesions, based on results from the B-1801/B-190 trial.⁵⁶ These clinical reports, together with our findings, further support the rationale for repositioning this synthetic rexinoid for the treatment of non-CTCL and to propose a way to optimize its therapeutic effect with integrin inhibition.

Regarding the in vivo effects of this combination, our previous research demonstrated that it significantly inhibited TCL tumor growth without compromising antitumor immunity.²² Here, we deeply explored their mechanisms of action and found that the antilymphoma in vivo activity of Cile and Bex combination involves the decrease of STATs phosphorylation and tumor metalloproteinase activity. The more pronounced inhibition of MMP2 and MMP9 observed in vivo likely reflects the influence of the tumor microenvironment, in which Cile may also act on nonmalignant stromal and immune cells, contributing to the regulation of metalloproteinase activity. Importantly, we also demonstrated that combining integrin αvβ3 inhibition with Bex reduced TCL metastatic formation in the kidneys and liver. Lymphoma dissemination, the colonization of nonlymphoid organs by tumor cells, is associated with poor prognosis and has become a relevant focus in the context of relapsed and refractory lymphomas.^57,58 TCL cell dissemination is facilitated by the interaction between integrin αvβ3 and the extracellular matrix, activating key survival and migratory pathways such as focal adhesion kinase and PI3K/Akt.^59,60 By antagonizing this axis, Cile disrupts downstream signaling leading to cytoskeletal disassembly and reduced cell adhesion.^53,61 In our in vivo model, the reduced metastatic spread observed with the BexT4+ and Cile combination likely stems from this mechanism, further supporting the therapeutic value of targeting integrin αvβ3 in TCL. Notably, we observed this effect in both a syngeneic and a xenograft model. Specifically, we successfully established, to our knowledge, for the first time, an in vivo xenograft model of experimental metastasis using CUTLL1 cells in NOD-SCID mice, providing a valuable platform for studying TCL dissemination and therapeutic interventions.

Consequently, in the characterization of the proteomic profile regulated by BexT4+Cile in TCL tumors, we identified proteins and signaling pathways related to cell proliferation, immune response, angiogenesis, metastasis, and lymphoma progression. For example, the EZR-RDX-MSN proteins, which are reduced in TCL tumors treated with Cile and Bex, have been identified as crucial factors in the pathogenesis of diffuse large B-cell lymphoma, and their inhibition was proposed as a new strategy for controlling lymphoma growth.³⁵ The glycolytic enzyme α-enolase, ENO1, was found overexpressed in bone marrow biopsies from acute myeloid leukemia,⁴⁰ it also, contributes to cell proliferation and invasion in Burkitt lymphoma.³⁷ Also, elevated branched chain amino acid transaminase 1 expression levels have been identified as a contributing factor to chemotherapy resistance and poor prognosis in TCL.⁶² Interestingly, Cile and Bex combination increased Anastellin (fibronectin 1-A) expression, a small protein fragment derived from the C-terminal part of fibronectin with antiangiogenic activity.^34,63 

Additionally, the translational impact of our results was evidenced by the bioinformatics analysis on a public data set of 193 samples from patients with TCL. Patients with high levels of ITGAV and ITGB3 were enriched in genes related to TCL progression and dissemination. Moreover, patients with elevated ITGB3 levels exhibited enrichment in STAT genes, which was further correlated with poorer overall survival.

In sum, our findings suggest that blocking the actions of THs on the integrin αvβ3 receptor, in combination with Bex, could serve as a potential noncytotoxic therapy to reduce abnormal activation of the JAK/STAT pathway and limit lymphoma spread.

This work was supported by the Agencia Nacional para la Promoción Científica y Tecnológica (PICT 0758/2020 [F.C.] and PICT 3807/2018 [G.A.C.]), Fundación Fiorini (F.C.), CONICET (PIP No. 11220210100586CO [G.A.C.]), and Pontificia Universidad Católica Argentina. The authors also thank the Program for Technological Development in Tools for Health (RPT-FIOCRUZ) for providing access to the mass spectrometry platform (RPT02H).

Contribution: M.D. was responsible for investigation, visualization, writing the original draft, and reviewing and editing the manuscript; A.C. was responsible for proteomic data curation, formal analysis, visualization, and reviewing and editing the manuscript; L.A. was responsible for investigation and visualization; M.V.R. curated transcriptomic data and performed visualization; H.A.S. conducted the investigation and reviewed the manuscript; G.G., I.L.M.S., and J.R. were responsible for investigation; M.T.G.d.D. was responsible for investigation and visualization; J.A.D.A. and C.R. were responsible for investigation; L.C. contributed to conceptualization and resources and reviewed the manuscript; G.A.C. contributed to conceptualization, resources, formal analysis, funding acquisition, investigation, methodology, writing the original draft, project administration, and review and editing of the manuscript; and F.C. contributed to conceptualization, resources, formal analysis, supervision, funding acquisition, investigation, methodology, project administration, writing of the original draft, and reviewing and editing the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Florencia Cayrol, Laboratory of Neuroimmunomodulation and Molecular Oncology, Instituto de Investigaciones Biomédicas, Consejo Nacional de Investigaciones Científicas y Técnicas, Universidad Catolica Argentina, Alicia Moreau de Justo 1600, 3er piso, Ciudad Autónoma de Buenos Aires, CP: 1107 AAZ Buenos Aires, Argentina; email: florencia_cayrol@uca.edu.ar; and Graciela Alicia Cremaschi, Laboratory of Neuroimmunomodulation and Molecular Oncology, Instituto de Investigaciones Biomédicas, Consejo Nacional de Investigaciones Científicas y Técnicas, Universidad Catolica Argentina, Alicia Moreau de Justo 1600, 3er piso, Ciudad Autónoma de Buenos Aires, CP: 1107 AAZ Buenos Aires, Argentina; email: graciela_cremaschi@uca.edu.ar.