Non-coding RNAs (ncRNA) were initially classified as “evolutionary junk”, but their important role in the regulation of transcription (1), cell differentiation (2), development (3), metabolic reprograming (4), and many others has been demonstrated [reviewed in (5–7)]. Because they act at the RNA level and most of them are not transcribed, they were originally called “riboregulators” (8). They can be divided into two large classes: the small non-coding RNAs are commonly from ∽20 to 200 nucleotides (nt), while those longer than 200 nt and up to ∽100 kb and lacking open reading frames are long non-coding RNAs (lncRNA) (9–11). The non-coding elements comprise 80% of the human genome (10). It has been suggested that between ~21,488 and ~200,000 lncRNAs (12, 13) and a subset of these, the lincRNAs, sum up >14,279 (14, 15) in humans. RNA polymerase II is responsible for lncRNA transcription (8, 16), and its biogenesis is similar to mRNAs in the sense that they are polyadenylated, capped, and spliced (5). However, it is worth mentioning that lincRNAs are less efficiently spliced. This can be partially explained because lincRNAs have weaker splicing-related motifs and lower binding to U2 small nuclear RNA auxiliary factor 2 (U2AF2, formerly U2AF65) (17). Depending on their functions, lncRNAs can be classified as cis-acting (within 10 k upstream or downstream of the coding genes they regulate) or trans-acting [identified by expression levels, according to Pearson’s correlation coefficient |r| > 0.95 (18) reviewed in (5)]. LncRNA scaffolds, guide lncRNAs, lncRNA decoys, enhancer RNAs (eRNAs), promoter-associated lncRNAs (PALRs), precursor lncRNA [reviewed in (19)], and a heterogeneous subgroup are involved in X-chromosome inactivation, telomere regulation, and imprinting (reviewed in (6)). Depending on their position in the genes, lncRNAs can also be classified as exonic, intronic, or intergenic (20). An additional classification regarding evolution are the Transcribed Ultra-Conserved Regions lncRNAs (T-UCRs, ≥481) (21, 22). An alternative and comprehensive classification can be found at (23).

LncRNAs regulate innate immunity in host–virus interactions. During viral infection, cellular lncRNAs directly regulate viral genes and modify both viral replication and pathology through changes mediated by virus in transcriptome of the host (24–26). Also, several lncRNAs could upregulate IFN-I (IFNα and IFNβ) expression to induce an antiviral response (27–30), reviewed in (31). Although lncRNAs seem to not have protein-coding function, some of them actually encode functional small peptides (32–34) [reviewed in (35, 36)]. In this review, we will focus on lncRNAs. LncRNAs can regulate different gene modulation processes. They regulate gene expression during the regulatory layers of transcription (37), RNA processing as splicing and indirectly editing (38), and/or translation (39). LncRNAs can control post-translational modification of proteins such as phosphorylation (40), acetylation (41), and ubiquitination (42, 43). In brief, they can act as a decoy or scaffold of other molecules such as proteins (44–46), RNA,and DNA (47). Also, their subcellular location gives important clues about its function in different biological processes. In the nucleus, they guide transcription factors or complexes that modify chromatin. In the cytosol, they control stability of mRNA (48) or compete with endogenous mRNA for access to the protein expression machinery (23). In the cytoplasm, lncRNAs can bind directly to transcription factors (TFs) (49). Interestingly, several lncRNAs are located near genes that encode TFs as well (50). However, particularly long intergenic non-coding RNAs (lincRNAs) regulate both the expression of neighboring genes and distant genomic sequences (6). At the transcription level, lncRNAs regulate accessibility to chromatin in processes that involve RNA polymerase and histone modification enzymes. Furthermore, their ability to bypass chromatin-modifying complexes to impact the basal transcriptional machinery and gene promoters has been demonstrated (51). One classical example of lncRNA in a non-disease state is the X-inactivation specific transcript (XIST) in women that recruits the polycomb complex. XIST induces chromosome silencing of the sexual chromosome from which it was transcribed (52) [reviewed in (53)]. LncRNAs can also stabilize mRNAs and regulate their decay by diverse mechanisms such as (1) binding to target mRNAs, (2) sequestering miRNAs and/or ribosome-binding proteins [RBPs, these lncRNAs are also called competitive endogenous RNAs (ceRNAs), decoys, or sponges], and (3) acting as scaffolds to enhance interaction of RBP-mRNA or interacting with m⁶A machinery to modulate m⁶A levels of target mRNAs. This induces gene expression (54). In addition to the complexity of their regulation, lncRNAs have slicing variants that are expressed in a tissue- and/or cell-specific manner (54). In recent years, an increased number of academic groups and pharmaceutical companies became interested in the development of novel therapies based on lncRNA regulation. Currently, at least 61 lncRNA-related clinical trials are ongoing worldwide for cancer, cardiovascular disease, diabetes, and other conditions/diseases (https://www.clinicaltrials.gov/ and Figure 1 ).

LncRNAs are also involved in pathological processes in viral infections and cancer. HIV is the causative agent of AIDS. There are two strains of HIV, HIV-1 and HIV-2 (55). In this review, we will focus on lncRNAs encoded in HIV-1 and SARS-CoV-2 genomes as well in lncRNAs of human endogenous retroviruses (HERVs). For data of lncRNAs in other viruses, see (10). With the emergence of SARS-CoV-2 in 2019, the importance and the role that non-coding RNAs play in the infection process have been highlighted and recently investigated (56).

LncRNAs of HERVs and other host lncRNAs appear as a potential novel strategy to combat viral infections. It is important to know the mechanisms of action of lncRNAs in the development and outcome of infections for the development of more effective therapeutic strategies based on these complex gene regulation layer. In order to develop an adequate infectious process, viruses have the ability to decrease or increase the expression of lncRNAs; these changes can generate antiviral or proviral functions and completely modify the development of a disease. HIV alters the expression of important cellular lncRNAs (57), but the participation of other viruses has been just recently studied (10). Ma et al. (2021), reported 242 differentially expressed (DE) lncRNAs in early infection with HIV and might show cis-dependent regulation of B and T lymphocyte-associated protein gene (BTLA), zeta chain-associated protein kinase gene (ZAP70) and growth factor receptor-bound protein 2-related adaptor protein gene (GRAP). Some lncRNAs were predicted to act as ceRNAs to regulate expression of fos protooncogene, AP1 transcription factor subunit (FOS), fosb protooncogene, AP1 transcription factor subunit (FOSB), V-jun avian sarcoma virus 17 oncogene homolog (JUN), transcription factor 7-like 2 (TCF7L2), and hypoxia-inducible factor 1, α-subunit (HIF1A) (58). LncRNAs are crucial for viral reproduction (production of new infectious virus particles), either repressing or activating HIV-transcription. After HIV or SARS-CoV-2 infection, ncRNAs, including miRNAs, small interfering RNAs (siRNAs), and lncRNAs, are modulated by the virus, in order to achieve the successful replication (nucleic acid synthesis) cycle. Attempts have been made to compare the different datasets obtained by research groups on expression changes during infection, but it represents a great challenge because the expression levels are highly sensitive to external stimuli and it is difficult to define the basal levels that normally exist in cells, because interindividual variation exists (59). It has been identified as a deregulation response of many lncRNAs in response to either an infection process or changes in cytokines such as IFN, which end up impacting the viral replication that may occur (60). In addition, about 50% of HIV-1-positive patients receive combined antiretroviral therapy (cART), which includes a cocktail of inhibitors that target viral processes including entry, reverse transcription, integration, and/or protease-mediated cleavage (61). HIV-1 latency is a state in which viral RNA is hard to detect by conventional/clinic techniques. HIV-1 latency is one of the major barriers to eradicate HIV/AIDS because the virus remains largely undetected by the surveillance of the immune system.

In this review, we summarize the role of lncRNAs in some important viral infections such as COVID-19 and HIV-1/AIDS as well as briefly underline the importance of lncRNAs transcribed from HERVs, their functions, mechanisms, and some of their nucleic acid- and protein-binding partners. To avoid confusion in this review, lncRNA genes are written in italics and the transcripts of lncRNAs are not.

Plasma levels of lincRNA chr12:57761837-57762303 and lincRNA:chr2:165509129-165519404 were better markers for HIV-1 infection whereas LincRNA chr5:87580664-87583451, XLOC_001148, and lincRNA chr10:128586385-128592960 were better markers for HIV-2 infection (90). The presence of NEAT1 in plasma has been suggested as a biomarker of HIV-1 infection (79).

Schynkel et al. (2020) analyzed the transcriptomic change of macrophages treated with three types of IFNs (IFN-α, IFN-γ, and IFN-ϵ) and 6 h post-infection with HIV-1. They found differential expression of 474 genes. From these, 112 (24%) were classified as non-coding; within these seven DE genes, they coincided between the four conditions of IFN finding five lncRNAs, namely, AC053503.1, NRIR, C8orf3, AL359551.1, and MIR3945HG, as well as two pseudogenes, namely, neutrophil cytosolic factor 1C pseudogene (NCF1C) and guanylate binding protein 1 pseudogene 1 (GBP1P1) (87).

In elite controllers of HIV, it has been suggested that in myeloid dendritic cells, the lnc MIR4435-2 host gene (MIR4435-2HG) enhances metabolic function through epigenetic mechanisms including H3K27 enrichment at an intronic enhancer of the regulatory associated protein of mammalian target of rapamycin [mTOR] (RPTOR) locus, which is the main component of the mTOR complex 1 (mTORC1) by means of a predicted triple helix of RNA–DNA. These changes shows features of trained innate immunity (91).

RUNX1 overlapping RNA (RUNXOR) is an lncRNA and an important regulator of myeloid-derived suppressor gene. RUNXOR and RUNX are upregulated in MDSCs that expand and accumulate in human PBMCs in people living with HIV (PLHIV). RUNXOR and RUNX are associated with expression of immunosuppressive molecules such as reactive oxygen species, arginase 1, signal transducer and activator of transcription 3 (STAT3), IL-6, and inducible nitric oxide (NO) synthase (92).

A total of 242 lncRNAs were DE in HAART-naïve people living with HIV (PLHIV) patients. Seventeen of those lncRNAs were predicted to regulate TCF7L2 and HIF1A genes, which are involved in HIV transcription. Twenty DE lncRNAs probably share miRNA response elements with JUN and FOS, which are associated with both HIV-1 replication and immune activation (58).

Chemokine, CC motif, receptor 5-antisense (CCR5-AS) is an lncRNA that regulates differential expression of CCR5, the major co-receptor of HIV. High expression of CCR5-AS sequesters the RALY heterogeneous nuclear ribonuclear protein (RALY) preventing its binding to CCR5 3´-UTR. This prevents the decay of CCR5 mRNA (93).

For the complete transcription of HIV-1, two stimuli are necessary. Firstly, the activation of T cells, and secondly, the translation of the viral protein Tat. With this rationale, one therapeutic approach is that peptides capable of competing for Tat binding and stimulating a viral ncRNA response have been applied. Peptide F07 #13 can induce the silencing of transcriptional genes by inducing viral ncRNA (e.g., TAR-gag and TAR) that act with transcriptional suppressor proteins cullin 4B (CUL4B), PRC2, and SIN3 transcription regulator family member A (SIN3A), ultimately leading to ubiquitination of Tat (61). Inhibition of lncRNAs that induce latency can be achieved by post-transcriptional degradation of the lncRNA with short hairpin RNA (shRNA), antisense oligonucleotide (ASO), or siRNA. Improved design of modified siRNA and ASOs has led to a higher efficacy and stability, while off-target effects are reduced. This can be attained by lncRNA transcription inhibition of by gene editing by CRISPR-Cas or promoter blockade (94, 95). As the function of lncRNAs is through formation of secondary structures with both proteins and nucleic acids, a diverse collection of RNA decoys, nanobodies, small molecules, and aptamers to induce steric hindrance are being evaluated in ongoing efforts (96). Most of the current approaches for therapy are focused on pathogenic lncRNA inhibition by delivery of lncRNA made of synthetic RNA. For lncRNA delivery, a feasible approach is to use adeno-associated virus, herpes virus, ncRNA sponges, chemical modified ncRNAs, non-viral vectors, computed tomography-guided injections, and endoscopic ultrasound (97).

The lack of sequence conservation across species is one of the main challenges of lncRNA research. One solution could be humanized mouse models. In addition, specific markers of HIV-1 latently infected cells remain to be identified. Induction of apoptosis in virus-infected cell only, is caused by downregulation of the lncRNA heterogeneous nuclear ribonucleoprotein U gene (HNRNPU, formerly SAF) with siRNA. This leaves unaffected the bystander cells (83). The effectiveness and safety of lncRNA-mediated therapy should be demonstrated in non-human primates as well as in clinical trials. Importantly, the elimination of reservoirs of latent HIV-1 remains the major hurdle to find a complete cure for HIV-1 infection. The approaches to trying to tackle this are killing infected cells, gene therapy, “shocking and kill” HIV-1 out of latency [reviewed in (97)], and/or HIV-1 silencing, e.g., through the histone chaperones, namely, chromatin assembly factor I, subunits A and B (CHAF1A and CHAF1B) (98). Nevertheless, none of these potential solutions are likely to eradicate HIV in the short term; a combination of them could lead to a feasible functional cure in the mid or long term.

The role of lncRNAs in both the initiation and progression of viral infection has been reported. In a mouse model, RNA sequencing of influenza A virus and SARS-CoV-2-infected lung tissues also demonstrated the key roles of lncRNAs in pathogenesis of respiratory virus by inducing interferon (IFN). In this model, the pathogenesis regulation is mediated by signal transducer and activation of transcription (STAT1). In COVID-19 patients, an increasing chemokine expression has been observed in reduced or absence of type I (IFNα and IFNβ) and III (IFNλ1, IFNλ2, IFNλ3, and IFNλ4) interferons (99). Thus, the lncRNA network of COVID-19 patients showed a significant downregulation of IFN-I response (15). Furthermore, in SARS-CoV-2 infection, 527 bronchoalveolar lavage fluid (BALF) and 239 protein-coding genes (PCGs in peripheral blood mononuclear cell-PBMC) could be influenced by 162 and 106 cis-acting lncRNAs, respectively. Main pathways involved are myeloid local site activation, immune system, iron homeostasis regulation, and degranulation of neutrophils in the tissue-specific regulation of host pyrin domain (PYD) and C-terminal caspase-recruitment domain (CARD)-containing protein (PYCARD) and IFNG gene (PBMC) as well as calpain 1 (CAPN1) gene mediated by MALAT1 and NEAT1 lncRNAs (99). Another report suggests that 195 and 155 lncRNAs are down- and upregulated, respectively (100). Furthermore, another group reported that, in the GEO database (GSE147507) in more than one cell line, the DE lncRNAs were LINC00312, LINC00473, LINC00605, HLA complex group 11 (HCG11), HIF1A-antisense 2 (HIF1A-AS2), eosinophil granule ontogeny (EGOT), erythrocyte membrane protein band 4.1-like 4A-antisense 1 (EPB41L4A-AS1), LINC00115, LINC00174, LINC00265, LINC00662, LINC00842, myocardial infarction-associated transcript (MIAT), NEAT1, MALAT1, maternally expressed 3 and 9 (MEG3 and MEG9), RNA component mitochondrial RNA processing endoribonuclease (RMRP), telomerase RNA component (TERC), and zinc finger protein 674–antisense 1 (ZNF674-AS1) and decreased expressions of TP3 target 1 (TP53TG1), LINC00488, prostate androgen-regulated transcript 1 (PART1), and LINC00857 (101). Also in a murine model, a report suggests that there are 5,329 different lncRNAs (15). In the binding of lncRNA and transcription factors, those more relevant were IRF1, STAT1, and MYC1. Notably, in lungs of COVID-19 patients, expression of NEAT1 was also increased significantly (101). Interestingly, expression of MALAT, NEAT1, and MIAT was common between SARS-CoV-2-infected cells and those infected with HIV. An increased expression of MALAT1 and abnormal expression of other lncRNAs was reported in BAL fluid of COVID-19 patients (15, 99, 101). Hox antisense intergenic RNA (HOTAIR), NEAT1, MALAT1, LINC00589 (formerly tumor-suppressor lncRNA on chromosome 8p12 or TSLNC8), and cardiac autophagy inhibitory factor (CAIF) lncRNAs may regulate IL-6 expression. In contrast, XIST, cyclin-dependent kinase inhibitor 2B-antisense RNA gene (CDKN2B-AS1, formerly antisense RNA in the INK4 locus or ANRIL), repulsive guidance molecule domain family, member B-antisense 1 (RGMB-AS1), NEAT1, MAS-related G-protein coupled receptor, member D (MRGPRD, formerly Gm499), and cytochrome C oxidase subunit II (MT-CO2, formerly Cox2) lncRNAs encoded in the mitochondria genome have been implicated in the formation of inflammasome (100). Another group reported 410 differentiated lncRNA among COVID-19 patients and controls (20). Different results, particularly the role of IFN-I in COVID-19 patients, must be due to the application of different methodologies, different tested time points, and non-comparable cell lines analyzed (101). The majority of these lncRNAs have been predicted by in silico approaches except NEAT1 (upregulation) and DANCR (downregulation). LncRNAs in cytokine storm are RAS gene associated with diabetes 51-antisense 1 (RAD51-AS1, formerly TODRA), non-coding RNA activated by DNA damage (NORAD, formerly LINC00657), and GAS5 (15).

Also, in the case of autoimmunity in SARS-CoV-2-infected patients, some lncRNAs act through inflammatory pathways including TLR and NF-κB signaling (10). Approximately 15%–20% of COVID-19 patients develop severe pneumonia and acute respiratory distress syndrome (ARDS). LncRNA GATA5 is elevated and positively correlated with severe COVID-19 (11). In another study, vitamin D receptor gene (VDR) expression was lower in COVID-19 patients compared with non-infected controls. Differences were not found with Linc00511 or Linc00346 (110). Also, DANCR (in lungs) and NEAT1 lncRNAs targeting microRNAs were associated with genes that differentiate mild from severe damage in SARS-CoV-2 (111).

Genotyping of SARS-CoV-2 has been very useful in monitoring both viral evolution and transmission during the COVID-19 pandemic (112). However, in a more practical and affordable way, alternative strategies have been developed to demonstrate the presence of the virus in the samples of suspected individuals. The most common methods to identify the presence/absence of SARS-CoV-2 is reverse transcription (RT)-PCR (sensitivity = 82.2%; specificity = 100%) (113). Additional strategies include RT loop-mediated isothermal amplification (without the need for RNA extraction) (sensitivity = 87.5%; specificity = 100%) (114) and deep learning computer-aided diagnosis (CAD) based on digital chest x-ray images (overall detection accuracy = 96.31%; classification accuracy = 97.40%) (115). Transcript level combination of cytochrome p450, subfamily XXVIIB, polypeptide 1 (CYP27B1), VDR, small nucleolar RNA host gene 16 (SNHG16), structural maintenance of chromosomes 3 (SMC3, formerly chondroitin sulfate proteoglycan 6 or CSHG6), Linc00511, and Linc00346 could be a potential alternative method to COVID-19 diagnostics (sensitivity = 0.62; specificity = 0.81) (110).

LncRNAs interactions with IFN in SARS-CoV-2 is an interesting research topic (15). The clotting alteration in COVID-19 may be related to inflammatory cytokines; thus, a better understanding of the cytokine regulation by lncRNAs will be helpful to understand this clinical phenotype and develop therapeutic targets to lncRNAs or molecules mimicking their function. Forty percent of known lncRNAs were found regulating different processes in the central nervous system (CNS) (15).

It has been suggested that lncRNAs can target and bind cytokine nucleotide sequences leading to downregulation of these cytokines. IFN-α is a common antiviral therapy in critically ill COVID-19 patients. Induction of MALAT1 expression by viral infection promotes activation of interferon regulatory factor 3 (IRF3) and production of type I IFN ( Figure 5 ). Inhibition of the excessive inflammation may be a useful complement to COVID-19 therapy. Potential approaches in the clinic include oligonucleotide knockdown, RNAi knockdown, and viral gene therapy interfering with gene transcription, protein translation, and extranuclear messengers (15). Furthermore, lncRNAs are spatially correlated with TFs across the genome as, e.g., the case of FOXA2-adjacent long non protein-coding RNA (lncRNA Falcor) that can regulate forkhead box A2 (FOXA2) expression. Alteration of the regulatory feedback loop of Falcor-FoxA2 leads to altered cell migration and adhesion, thus causing goblet cell metaplasia and chronic peribronchial airway inflammation (15). Also, IL-6, transforming growth factor, beta 1 (TGF-β1, formerly TGF-β), nucleotide oligomerization domain-like receptor family, pyrin domain-containing 3 (NLRP3) inflammasome, and cholinergic signaling have been proposed as potential targets in COVID-19 (15). In the cholinergic inhibition, DANCR was found to potentially regulate inflammation. The identification of dysregulated or altered lncRNAs during SARS-CoV-2 infection could be a strategy for mitigating symptoms in patients and to limit viral genome amplification enhancing the immune response (15). The most relevant lncRNAs during COVID-19 seem to be HOTAIR, FOXF1 adjacent non-coding developmental regulatory RNA (FENDRR, formerly FOXF1-AS1), LINC01505 ( Table 2 ), and imprinted maternally expressed non-coding transcript (H19). Furthermore, FENDRR, HOTAIR, and LINC1505 potentially interact with the mRNA of spike protein (116). In macrophages, HIF-1α-stabilizing lncRNA (HISLA) regulates HIF-1α transcription factor by inhibition of its binding to EGL9 family hypoxia-inducible factor 1 protein (EGL1, formerly prolyl hydrolase domain-containing protein 2 or PHD2) ( Table 2 ) (for an interaction network of SARS-CoV-2 proteins with human proteins, see Figure 6 ). This keeps continuous activation of an aerobic state in tumor cells. In the case of the treatment of COVID-19 using the strategy of an aptamer-siRNA chimera-mediated HISLA knockdown, this approach is promising but requires further investigation (15) ( Figure 5 ).

Human endogenous retroviruses [HERVs, reviewed in (117)] are transposable elements comprising ∽10% of our genome and ranging from full-length proviruses to short gene fragments. HERVs are remnants of infectious exogenous retroviruses that became part of our DNA and now are inherited in a Mendelian manner. The HERV database [HERVd, http://herv.img.cas.cz (118)] contains 519,060 entities and 725,763 repeats of 582 different ERVs (accessed on March 3, 2022) that belong to 150 families. HERVs, working as Pathogen Associated Molecular Patterns (PAMPs), have become important parts of immune mechanisms to fight invading pathogens at least for dengue virus, herpes simplex viruses, influenza viruses, and HIV-1. Mechanisms used by ERVs are regulation of viral expression, enhancement of cellular sensing pathways, direct restriction of virion assembly, and blocking of entry receptors (117). Also, some ERVs are responsible for inserting proviruses in host genome facilitating infection and/or viral latency.

Host cells are armed with a variety of pattern recognition receptors (PRRs) that recognize nucleic acids from virus such as phospholipase A and acyltransferase 4 (PLAAT4, formerly retinoid acid-inducible gene 1or RIG-1) that recognizes 5´-triphosphate RNA, interferon induced with helicase C domain 1 (formerly melanoma differentiation-associated protein 5 or MDA5) that recognizes double-stranded RNAs (dsRNA), toll-like receptors 7 and 8 (TLR7 and TLR8) that recognize single-stranded RNA (ssRNA), toll-like receptor 3 (TLR3) that recognizes both dsRNA and poly I:C, toll-like receptor 9 that recognizes CpG DNA and DNA-immunoglobulin (Ig) complexes, and DNA-dependent activator of IRFs (DAI) that recognizes double-stranded DNA (dsDNA) [reviewed in (119)].

One classic example of ERV is a murine lncRNA called lncRNA ERV-derived positively regulating antiviral responses (lnc-Epav). This lncRNA boosts antiviral gene expression by increasing levels of RelA, a subunit of NF-κB transcription factor. The way that lnc-EPAV works in RELA promoter is competitively binding and displacing the splicing factor, proline- and glutamine-rich (SFPQ, an inhibitor of RELA gene) (47) ( Table 3 ). Also, mice deficient in lnc-EPAV shows a reduced expression of type I interferons. In humans, the role of EPAV seems to be performed by interacting to MER9a2, LTR5A, and MLT2A1 HERVs. Results were concordant with a mouse model increasing levels of RELA in SFPQ knockdown human cells (47) ( Figure 7 ).

Another example of interaction with the immune response is the long non-coding endogenous avian leukosis virus on chromosome 1-antisense 1 (lnc-ALVE1-AS1), transcribed from ALVE1 gene, that can activate or bind to the Toll-like receptor 3 (TLR3) inducing an antiviral innate immune response (123) ( Table 3 and Figure 7 ).

The lncRNAs related to HERVs have also been associated with functions in tissue growth, as in the case of V-raf murine sarcoma viral oncogene homolog B1 (BRAF)-regulated long non-coding RNA (BANCR), related to pro-migratory functions necessary in the development of the heart in humans and primates (121) ( Table 3 ). This relationship with different tissues has also been analyzed with different types of cancer. In addition, lncRNA TROJAN is known to be involved in the proliferative state of breast cancer (BrCa) cells ( Table 3 ). By reducing TROJAN expression, the proliferative potential in vitro is prevented, so this could reduce tumor volume and be a therapeutic target in the future (124). Even when these examples are in cancer, the similar cell cycle alterations with viral infection lead to hypothesize about a potential role of these HERVs in HIV/AIDS and/or COVID-19. Nevertheless, the importance of ERVs to antiviral immune response is becoming clear (47, 117); further basic and applied research is needed in this area.

LncRNAs modulate both growth and viral replication as well as type I IFN expression mainly by stability regulation and interaction with mRNAs and proteins. The most relevant transcription factors that are regulated by lncRNAs are TP53, CTCF, MYC, SOX2, EZH2, SFPQ, and SUZ12 (125). In addition, lncRNAs seem to control the effector function of genes and peptides such as STAT1, STAT3, NF-κB, and IRFs and of many TFs, whose interactions remain to be discovered. The modulation of lncRNAs might regulate these transcription factors at both protein levels and at nucleotide motifs/structures. The key targeted potential therapies for SARS-Cov2 involve interfering gene transcription messengers and protein translation (15). Even when there are several questions to be solved, e.g., whether the cytokine storm activates ERVs in response to SARS-CoV-2 infection in COVID-19 (as is true for viral lncRNAs), a better understanding of how lncRNAs regulate gene expression in a cell/tissue specific manner by interactions with mRNAs, these new therapeutic targets and biomarkers hold promise for the treatment and risk stratification of patients with HIV/AIDS and COVID-19.

In addition, a promising area of research is to catalog and interpret the polymorphisms, mutations, and indels in lncRNAs (126, 127), as well as their epitranscriptomic modifications (128–130) with ethnic and cell/tissue specificity. As an example of this, our group reported that the regulatory SNP (rSNP) rs5743417, which is common in African and Afro-Americans, causes a 33% transcription downregulation in the constitutively expressed gene human β-defensin 1 (DEFB1) in the A549 pulmonary epithelial cell line (131). DEFB1 is a primordial gene of innate immunity involved in the defense against viruses [reviewed in (132)] including HIV-1 (133). In addition, hBD-1 peptide has other functions as well [reviewed in (134, 135)]. Interestingly, SNP rs5743417 overlaps with GS1-24F4.2, a lincRNA gene complementary to the X Kell blood-related 5 (XKR5) mRNA. GS1-24F4.2 has a testis-specific expression that could shed light on male infertility associated with COVID-19 (136). Results like these increase the difficulty of interpretation of genetic variants at different regulation layers but at the same time open new opportunities to dissect both molecular and clinical impact of polymorphisms, whose interpretation in the light of functional genomics remained elusive in the past decades. In general, the integral impact of rSNPs could be explained (a) by the fact that SNPs could alter transcription factor binding sites (TFBS) mainly in promoters or enhancers (137), thus causing an up- or downregulation (131); (b) by the fact that SNPs can create novel TFBS (138) and thus modify homeostatic gene expression pathways; (c) by altering the normal structural and/or sponging function of lncRNAs (54); and (d) probably by indirectly introducing variations in the epitranscriptome.

With no doubt, lncRNAs research is coming of age. These novel drugs inspired by host- and viral-encoded lncRNA have the potential to reduce the burden of HIV/AIDS and COVID-19 twofold: (1) by increasing their efficacy and (2) by minimizing the side effects by modulating gene transcription instead of the more traditional approaches such as complete silencing, knockout/editing of genes, non-selective DNA methylation, or blockade/neutralization of host peptides and proteins. We expect that a new generation of lncRNAs could modulate human and viral transcription in a more specific and unprecedented way but still maintaining homeostasis below the pathogenic threshold.

AVRR: writing—original draft preparation and figures. EMP: writing and editing manuscript and tables. All authors contributed to the article and approved the submitted version.

EP has a granted patent that protects a small molecule (and its synthesis and purification) that induces DEFB1 transcription. This synthetic molecule could be applied as therapy for viral diseases including but not limited to HIV/AIDS and COVID-19. This patent is property of CIATEJ.

The remaining author declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.