Received: 26 April 2021 Revised: 4 June 2021 Accepted: 8 June 2021 Published online: 1 July 2021 DOI: 10.1002/ctm2.474 LETTER TO EDITOR Differential miRNA plasma profiles associated with the spontaneous loss of HIV-1 control: miR-199a-3p and its potential role as a biomarker for quick screening of elite controllers To the Editor: ure 1A, Table S2). From the 23 miRNAs, 22 miRNAs were People living with HIV (PLWH) who are able to main- positively correlated with viral load (VL), seven miRNAs tain suppressed viral load (VL) for years in the absence of were positively correlated with CD4+ T-cell counts, and antiretroviral therapy (ART) are known as elite controllers 11 miRNAs were positively correlated with CD8+ T-cell (ECs). ECs represent a heterogeneous population in terms counts (Table S3). Interestingly, the spontaneous loss of of virological, immunological, and clinical outcomes, and viral control in ECs can be defined by the expression approximately 25% of ECs lose viral control overtime. The of hsa-miR-27a-3p, hsa-miR-376a-3p, and hsa-miR-199a- study of the mechanisms leading to the loss of viral control 3p (Figure 1B), as confirmed by the diagnostic accuracy in ECs is crucial for the identification of differential mark- determined by the ROC analysis (Figure 1C). Notably, ers for the design of novel eradication and immunothera- hsa-miR-27a-3p, hsa-miR-376a-3p, and hsa-miR-199a are peutic strategies. tightly connected and related to lipid metabolism (Fig- Previously, we identified virological and immunolog- ure 1D). Then, we also evaluated the plasma miRNA profile ical factors involved in the spontaneous loss of viral in TCs under the postloss condition and we found 38 miR- control,1 and we also demonstrated that proteomics and NAs differentially expressed among groups, suggesting metabolomics are powerful tools to identify potential that viremia strongly influences the plasma miRNA pro- biomarkers and therapeutic targets in ECs.2,3 Addition- file of ECs (Figure 2A, Table S4). And again, these signifi- ally, genome-wide associations and transcriptome analyses cantly expressed miRNAs among groups were also related have also been described for ECs and compared to other to relevant genes linked to lipid pathways (Figure 2B), and phenotypes of PLWH,4,5 in particular the study of specific some of them positively correlated with VL, CD4+ T-cell microRNA (miRNA) expression profiles.5 miRNAs play counts and CD8+ T-cell counts (Table S5). Moreover, of the vital roles in development, apoptosis, and oncogenesis by 23 miRNAs significantly differentially expressed under the interfering with gene expression at the post-transcriptional preloss condition and the 38 miRNAs significantly differ- level.6 In HIV/AIDS scenario, the most relevant feature entially expressed under the postloss condition, only the is that some miRNAs can modulate HIV replication by upregulation of 19 miRNAs overlapped (Figure 3A). Note- directly targeting HIV RNA or targeting messenger RNA worthy, the expression of hsa-miR-199a-3p showed an opti- (mRNA) of cell factors necessary for HIV replication.7 mal percentage of separation and an ability to differenti- Our study conducted in 18 ECs (Figure S1, Table S1), 12 ate between both groups of ECs before and after the loss individuals who experienced a loss of spontaneous viral of viral control (Figure 3B). Accordingly, the upregulation HIV-1 control (transient controllers, TCs) and six ECs of hsa-miR-199a-3p could be related to lipid dysregulation who persistently maintained viral control during the same in TCs, which in turn may potentiate the activation of a follow-up period (persistent controllers, PCs), showed an cytokine deregulation1 and in last term bias the virological upregulated plasma miRNA profile in TCs before and after control in TCs. the loss of viral control. Thus, our results confirmed differences in the expres- First, 23 miRNAs were found differentially expressed sion of some miRNAs with target sites located in viral RNA in TCs compared to PCs at the preloss time point (Fig- regions encoding viral accessory proteins, suggesting the This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2021 The Authors. Clinical and Translational Medicine published by John Wiley & Sons Australia, Ltd on behalf of Shanghai Institute of Clinical Bioinformatics Clin. Transl. Med. 2021;11:e474. wileyonlinelibrary.com/journal/ctm2 1 of 6 https://doi.org/10.1002/ctm2.474 2 of 6 LETTER TO EDITOR F I G U R E 1 miRNA profile distinguishes TCs and PCs under the preloss condition. (A) Hierarchical clustering heatmap of a set of the 23 significantly expressed miRNAs; hierarchical combined tree shows the clusterization of miRNAs. For PCs, only one measurement was assessed during the follow-up and used as the control group for comparison to preloss and postloss time points in TCs. For TCs (n = 11), one patient was excluded for lack of expression data values in the preloss condition. Indicated with black dots, next to the miRNAs list, the highly upregulated miRNAs (FC > 4). Columns represent each patient, while rows correspond to significant miRNAs. (B) Plasma miRNA profile in PLWH-ECs. hsa-miR-27a-3, hsa-miR-376a-3p, and hsa-miR-199a-3p were the most highly upregulated miRNAs in TCs before the loss of viral control compared to PCs. Statistical analysis was carried out by a nonparametric Mann–Whitney t-test. Data are represented as box and whiskers (min to max values). (C) Receiver operating characteristic curves (ROC curves) from the combination of hsa-miR-27a-3p, hsa-miR-199a-3p, and hsa-miR-376a-3p (FC > 4) in TCs under the preloss condition (AUC = 1). (D) Network analysis display of the miRTarBase v8.0 database showing the association of the significantly upregulated miRNAs and their target genes in TCs compared to PCs before the loss of viral control. Cluster hubs shown in green circles indicate miRNAs, whereas blue circles depict their target genes (in purple circles the highlighted genes involved in the lipid metabolism as commented in the text). Bottom table displays some of the genes involved with these miRNAs that link hsa-miR-27a-3p and hsa-miR-199a-3p emerging concept that upregulated host-derived miRNAs regulate different steps of HDL-C metabolism, and recent in TCs might act as antiviral defence mechanism directly studies have promoted the importance of miRNAs in con- affecting important steps during HIV infection and also trolling LDL metabolism and in the regulation of genes playing a key role in HIV immune pathogenesis. Notably, involved in very low-density lipoprotein (VLDL) secretion. hsa-miR-423-3p targets the gag gene, hsa-miR-29a/29b and Our results are in accord with these data since the LDL hsa-miR-326 target the nef gene, and hsa-miR-324-5p tar- metabolism resulted to be increased in TCs under the post- gets the vif gene6,8 (Figure 4). loss condition (Figure S2 and Figure S3).2 Additionally, One mechanism associated to the inhibitory effect some of the most representative host-derived miRNAs of HIV replication could be the interaction of nef with associated with the spontaneous loss of viral control the RISC complex, which leads to the inhibition of overtime are liver-specific miRNAs, being implicated in the translation of viral proteins and viral replication. fatty acid and cholesterol biosynthesis. Thus, our results On the other hand, Nef-containing exosomes, taken suggest that disturbance in lipoprotein levels, mostly up by macrophages, have been suggested to impair induced by upregulation of some liver-specific miRNAs, lipid cholesterol efflux, causing intracellular cholesterol may be highly associated with the immunological factors accumulation, which consequently affects the risk of behind the loss of viral control in TCs. It is known that cardiovascular diseases in PLWH.9,10 Several miRNAs can HIV infection is characterized by a high energy demand LETTER TO EDITOR 3 of 6 F I G U R E 2 miRNA profile distinguishes TCs and PCs under the postloss condition. (A) Hierarchical clustering heatmap of the set of 38 significantly expressed miRNAs; hierarchical combined tree shows the clusterization of the miRNAs. For PCs, only one measurement was assessed during the follow-up and used as the control group for comparison to preloss and postloss time points in TCs. For TCs (n = 11), one patient was excluded for lack of expression data values in the preloss condition. Indicated with black dots, next to the miRNAs list, the highly upregulated miRNAs and in red dots the downexpressed miRNA in TCs under the postloss condition. Columns represent each patient, while rows correspond to significant miRNAs. (B) Network display of the miRTarBase v8.0 database showing the association of the significantly upregulated miRNAs and their target genes in TCs compared to PCs after the loss of viral control. Cluster hubs shown in green circles indicate miRNAs, whereas blue circles depict their target genes (in purple circles are the highlighted genes involved in the lipid metabolism as commented in the text). Bottom table displays some of KEGG pathways related with these seven most highly upregulated miRNAs under the postloss condition F I G U R E 3 miRNA profile in TCs under the preloss condition differs from the postloss condition. (A) Venn’s diagram representation showing the overlapping of the significantly expressed miRNAs in TCs, from both conditions, compared to PCs. In green are the highly expressed miRNAs; in red the downregulated miRNAs. (B) Logistic regression and receiver operating characteristic curves (ROC curves) from preloss and postloss relative expression elucidated hsa-miR-199a-3p as the most reliable potential biomarker for the prediction of the spontaneous loss of viral control in ECs 4 of 6 LETTER TO EDITOR F I G U R E 4 Proposed mechanisms of effect of the differential miRNA profile associated with the spontaneous loss of HIV-1 control. Host miRNAs that have been experimentally shown to regulate HIV expression are illustrated with the target sites located in viral RNA regions encoding important viral accessory proteins. The inhibitory effect of HIV replication by hsa-miR-423-3p (gag gene), hsa-miR-324-5p (vif gene), hsa-miR-29a/29b miRNA, and hsa-miR-326 (both target the nef gene) could be a mechanism against the virus after losing viral control in TCs. Of note, hsa-miR-29a/b can bind directly to the HIV-1 mRNA (nef gene) increasing its association with proteins in the RISC complex, which leads to the inhibition of the translation of viral proteins and viral replication. Moreover, during HIV infection, the viral accessory protein Nef (via exosomes) impairs activity of cholesterol transporter ABCA1, inhibiting cholesterol efflux and also increases the biosynthesis and accumulation of cholesterol from macrophages. On the other hand, the liver-specific miRNAs hsa-miR-miR-199a-3p, hsa-miR-miR-27a-3p, and hsa-miR-376a-3p were strongly related to lipid and lipoprotein metabolism (Figure 1D). Concretely, hsa-miR-27 is one of the key regulators of the expression of ABCA1 and LDL receptor, whereas hsa-miR-199a-3p is implicated in fatty acid and cholesterol biosynthesis. These results could be associated to the increased lipid profile in TCs after the loss of viral control. In red are represented the inhibited actions and in black are represented the indirect actions to reprogram the cells to aerobic glycolytic pathways that increased LDL metabolism in TCs. Notably, the expres- consequently may increase anabolic metabolism.2 So, sion of liver-specific hsa-miR-199a-3p showed an optimal the increased lipid profile in TCs before the loss of viral percentage of separation and an ability to differentiate control could be a consequence of the antiviral defence between both groups of ECs before and after the loss of mechanism of the host against viral replication. viral control. The main limitation of this work is the small sample size. However, it must be highlighted that these patients AC K N OW L E D G M E N T S are rare, and it is difficult to have a follow-up with sequen- This study would not have been possible without the col- tially stored samples. Despite the relatively low number of laboration of all the patients and medical and nursing patients, we were able to have a tight follow-up of TCs with staff who took part in the project. These results were par- samples before and after the loss of virological control. Val- tially presented during the Virtual Conference on Retro- idation studies are needed to establish the proposed miR- viruses and Opportunistic Infections (CROI, 2021, Abstract NAs as biomarkers for the loss of viral control in ECs. number 2037). We also acknowledge the Biobanc IISPV In conclusion, our study reveals a specific host-derived (B.0000853 + B.0000854) integrated into the Biobanks and miRNA pattern in ECs that may be used as a biomarker for Biomodels Platform (PT20/00197) for its collaboration and quick screening of the virological and immunological pro- CERCA Programme (Generalitat de Catalunya), ECRIS gression in ECs, and we confirmed that viremia induces integrated in the Spanish AIDS Research Network (Annex LETTER TO EDITOR 5 of 6 S1), and BiosferTeslab, a spin-off company of the Rovira IISPV: 2019/IISPV/05 (Boosting Young Talent); GeSIDA i Virgili University (URV) and the Pere Virgili Health “III Premio para Jóvenes Investigadores 2019″; Instituto Research Institute (IISPV). We also thank the comments de Salud Carlos III (ISCIII), Grant Number: CP19/00146; and criticisms of the anonymous reviewers that greatly Universitat Rovira i Virgili, Grant Number: 2019PMF- helped to improve the manuscript. Francesc Vidal is sup- PIPF-18; Instituto de Salud Carlos III (ISCIII), Grant ported by grants from the Programa de Intensificación Number: CPII19/00025 de Investigadores (INT20/00031)-ISCIII. Ezequiel Ruiz- Mateos was supported by Consejo Superior de Investiga- AU T H O R CO N T R I B U T I O N S ciones Cientificas (CSIC). María del Carmen Gasca-Capote All authors reviewed and approved the submitted version and María Reyes Jimenez-Leon were supported by ISCIII- of the manuscript. Experimental design: Jenifer Masip, FEDER, PFIS programme, FI19/00083 and FI17/00186, Carmen Gasca-Capote, Ezequiel Ruiz-Mateos, and Anna respectively. Anna Rull is supported by IISPV through Rull. Intellectual guidance: Consuelo Viladés, Joaquim the project “2019/IISPV/05″ (Boosting Young Talent), by Peraire, Francesc Vidal, Anna Rull, and Ezequiel Ruiz- GeSIDA through the “III Premio para Jóvenes Investi- Mateos. Recruitment of participants: Jenifer Masip, Ana- gadores 2019,” and by the Instituto de Salud Carlos III Irene Malo, Lorna Leal, Carmen Rodríguez Martín, Norma (ISCIII) under grant agreement “CP19/00146″ through Rallón, Consuelo Viladés, Joaquim Peraire, Montserrat the Miguel Servet Program. Jenifer Masip is supported Olona, and Francesc Vidal. Sample procurement: Jenifer by the Universitat Rovira i Virgili under grant agree- Masip and Verónica Alba. Data collection: Ezequiel Ruiz- ment “2019PMF-PIPF-18” through the call “Martí Fran- Mateos, Ana-Irene Malo, Lorna Leal, Carmen Rodríguez quès Research Fellowship Programme.” Norma Rallón is Martín, Norma Rallón, Consuelo Viladés, Joaquim Peraire, a Miguel Servet II investigator supported by a grant from and Montserrat Olona. Data analysis and interpreta- the Instituto de Salud Carlos III (ISCIII) “CPII19/00025.” tion: Jenifer Masip, María del Carmen Gasca-Capote, María Reyes Jimenez-Leon, Alberto Perez-Gomez, Eze- E T H I C S A P P R O VA L A N D C O N S E N T T O quiel Ruiz-Mateos, and Anna Rull. Manuscript prepara- PA R T I C I PA T E tion: Jenifer Masip, María del Carmen Gasca-Capote, Eze- All research protocols were approved and carried out quiel Ruiz-Mateos, and Anna Rull. Study design, data anal- according to the recommendations of the Ethical Commit- ysis, and article development: Jenifer Masip, Anna Rull, tee for Clinical Research following the rules of Good Clin- Ezequiel Ruiz-Mateos, and Francesc Vidal. Reviewed and ical Practice from the Institut d’Investigació Sanitària Pere edited the manuscript: Francesc Vidal, Anna Rull, and Virgili (CEIm IISPV). The CEIM IISPV is an independent Ezequiel Ruiz-Mateos. committee, made up of health and non-health profession- als, which supervises the correct compliance of the ethical D A T A AVA I L A B I L I T Y S T A T E M E N T principles governing clinical trials and research projects The data that support the findings of this study are that are carried out in our region, specifically in terms of available from the corresponding author upon reasonable methodology, ethics, and laws. All participants gave writ- request. ten informed consent in agreement with the Declaration of Helsinki. Jenifer Masip1,2,3 Carmen Gasca-Capote4 F U N D I N G I N F O R M AT I O N María Reyes Jimenez-Leon4 Fondo de Investigación Sanitaria-ISCIII-FEDER: Joaquim Peraire1,2,3 PI13/0796, PI16/00503, PI16/0684, PI18/1532, Alberto Perez-Gomez4 PI19/01127, PI19/01337, PI20/00326, RD16/0025/0006, Verónica Alba1,2,3 RD16/0025/0020, INT20/00031, FI19/00083, FI17/00186, Ana-Irene Malo5 CP19/00146, CPII19/00025. European Regional Develop- Lorna Leal6 ment Fund/European Social Fund; Programa de Suport als Carmen Rodríguez Martín7 Grups de Recerca AGAUR, Grant Number: 2017SGR948; Norma Rallón8,9 Gilead Fellowship Program, Grant Number: GLD14/293; Consuelo Viladés1,2,3 SPANISH AIDS Research Network, Grant Numbers: Montserrat Olona1,2,3 RD16/0025/0006, RD16/0025/0020 (ISCIII-FEDER, Francesc Vidal1,2,3 Spain). Programa de Intensificación de Investigadores, Ezequiel Ruiz-Mateos4 Grant Number: INT20/00031; Consejo Superior de Anna Rull1,2,3 Investigaciones Cientificas (CSIC); ISCIII-FEDER, PFIS ECRIS integrated in the Spanish AIDS Research Network programme, Grant Numbers: FI19/00083, FI17/00186; (Annex S1) 6 of 6 LETTER TO EDITOR 1 Universitat Rovira i Virgili, Tarragona, Spain ORCID 2 Institut Investigació Sanitària Pere Virgili (IISPV), Alberto Perez-Gomez https://orcid.org/0000-0002-3644- Tarragona, Spain 2914 3 Hospital Universitari de Tarragona Joan XXIII, Francesc Vidal https://orcid.org/0000-0002-6692-6186 Tarragona, Spain 4 Clinic Unit of Infectious Diseases, Microbiology and REFERENCES Preventive Medicine, Institute of Biomedicine of Seville, 1. Pernas M, Tarancón-Diez L, Rodríguez-Gallego E, et al. Factors IBiS, Virgen del Rocío University Hospital/CSIC/University leading to the loss of natural elite control of HIV-1 infection. J of Seville, Seville, Spain Virol. 2018;92(5):e01805-e01817. 5 Vascular Medicine and Metabolism Unit, Research Unit 2. Tarancon-Diez L, Rodríguez-Gallego E, Rull A, et al. Immunometabolism is a key factor for the persistent spon- on Lipids and Atherosclerosis, Sant Joan University taneous elite control of HIV-1 infection. EBioMedicine. Hospital, Universitat Rovira i Virgili, IISPV, Reus, Spain 2019;42:86-96. 6 Infectious Diseases Department - HIV Unit, Hospital 3. Rodríguez-Gallego E, Tarancón-Diez L, Garciá F, et al. Pro- Clínic Barcelona, IDIBAPS, University of Barcelona, teomic profile associated with loss of spontaneous human Barcelona, Spain immunodeficiency virus type 1 elite control. J Infect Dis. 7 Centro Sanitario Sandoval, Hospital Clínico San Carlos, 2019;219(6):867-876. Instituto de Investigación Sanitaria San Carlos (IdISSC), 4. Egaña-Gorroño L, Escribà T, Boulanger N, et al. Differential MicroRNA expression profile between stimulated PBMCs from Madrid, Spain 8 HIV and Viral Hepatitis Research Laboratory, Instituto de HIV-1 infected elite controllers and viremic progressors. PLoS One. 2014;9(9):e106360. Investigación Sanitaria Fundación Jiménez Díaz, 5. Rotger M, Dang KK, Fellay J, et al. Genome-wide mRNA expres- Universidad Autónoma de Madrid (IIS-FJD, UAM), sion correlates of viral control in CD4+ T-cells from HIV-1- Madrid, Spain infected individuals. PLoS Pathog. 2010;6(2):e1000781. 9 Hospital Universitario Rey Juan Carlos, Móstoles, 6. Hariharan M, Scaria V, Pillai B, Brahmachari SK. Targets for Madrid, Spain human encoded microRNAs in HIV genes. Biochem Biophys Res Commun. 2005;337(4):1214-1218. 7. Klase Z, Houzet L, Jeang KT. MicroRNAs and HIV-1: complex Correspondence interactions. J Biol Chem. 2012;287(49):40884-40890. Francesc Vidal Marsal, Department of Internal Medicine 8. Ahluwalia JK, Khan SZ, Soni K, et al. Human cellular microRNA and Infectious Diseases, Hospital Universitari de hsa-miR-29a interferes with viral nef protein expression and Tarragona Joan XXIII, IISPV, Universitat Rovira i Virgili, HIV-1 replication. Retrovirology. 2008;5:117. c/Mallafré Guasch, 4, Tarragona, Spain 43007. 9. Aryal B, Singh AK, Rotllan N, Price N, Fernández-Hernando Email: fvidalmarsal.hj23.ics@gencat.cat C. MicroRNAs and lipid metabolism. Curr Opin Lipidol. 2017;28(3):273-280. 10. Mujawar Z, Rose H, Morrow MP, et al. Human immun- Jenifer Masip and Carmen Gasca-Capote contributed odeficiency virus impairs reverse cholesterol transport from equally to this paper and are primary co-authors. macrophages. PLoS Biol. 2006;4(11):1970-1983. Ezequiel Ruiz-Mateos and Anna Rull contributed equally to this paper and are senior co-authors. S U P P O RT I N G I N F O R M AT I O N These results were partially presented during the Virtual Additional supporting information may be found online Conference on Retroviruses and Opportunistic Infections in the Supporting Information section at the end of the (CROI 2021, abstract number 2037). article.