Eukaryotic chromosomes have evolved specialized nucleoprotein structures termed telomeres to protect their extremities. Mammalian telomeres are constituted by telomeric DNA, telomere-binding proteins forming the shelterin protein complex, and telomeric repeat-containing RNAs, known as TERRA (de Lange, 2009). The telomeric DNA is an array of repetitive sequences that enables the recruitment of telomere binding proteins which mediate telomere homeostasis. In vertebrates, the telomeric DNA sequence consists of TTAGGG repeats (Moyzis et al., 1988) extending for 9–15 kb in humans (de Lange et al., 1990). In several organisms the actual end of telomeres contains a 3′ G-rich overhang that invades the upstream double-strand telomeric region, forming the t-loop structure (Griffith et al., 1999; Doksani et al., 2013).

In mammals, both the double-stranded and the single-stranded telomeric DNA sequences are bound by the shelterin complex which is essential for telomere function and structure. A key role of the shelterin complex is to solve the “end protection problem” that arises from the necessity of cells to discriminate the end of chromosomes from the extremities of DNA double strand breaks (DSBs), occurring due to exogenous and endogenous insults (de Lange, 2009; 2018). DSBs are quickly sensed by the DNA damage response (DDR) that triggers a cell cycle arrest and activation of DNA repair mechanisms (Harper and Elledge, 2007). The absence of shelterin proteins or unwinding of the t-loop leads to telomere dysfunction, activation of DDR pathways at chromosome ends and consequent cell cycle arrest (Sfeir and de Lange, 2012; Doksani et al., 2013). In the absence of functional check-points, instigation of DDR signaling at telomeres results in the activation of DNA repair processes with consequent deleterious outcomes, including chromosome end-to-end fusions and genomic instability (Denchi and de Lange, 2007; Sfeir and de Lange, 2012; Doksani et al., 2013; Ly et al., 2018).

Due to the inability of the DNA replication machinery to fully duplicate the extremities of linear chromosomes (Watson, 1972; Olovnikov, 1973), in the absence of maintenance mechanisms telomeres erode during each cell division, a process known as the “end replication problem” (Allsopp et al., 1992). In human cells, telomere erosion ultimately triggers an irreversible cell cycle arrest known as replicative senescence that limits the replicative capacity of cells (Smith and Pereira-Smith, 1996; Braig et al., 2005). Telomere attrition can be counteracted by the addition of telomeric repeats to the 3′ overhang of chromosomes via the ribonucleoprotein telomerase complex (Kim et al., 1994). The essential components of telomerase are the catalytic subunit telomerase reverse transcriptase (TERT) and the RNA subunit telomerase RNA component (TERC or TR) (Greider and Blackburn, 1985; Greider and Blackburn, 1989). In adult tissues, telomerase activity can be detected in stem cells (Wright et al., 1996; Cong et al., 2002). Furthermore, approximately 90% of human cancers upregulate telomerase to overcome replicative senescence and attain proliferative immortality (Kim et al., 1994). In humans, telomerase is also active during early embryogenesis while it is inactivated by the silencing of TERT in somatic cells from early development (Wright et al., 1996). Thus, in most human tissues, telomeres shorten with age (Demanelis et al., 2020). Furthermore, senescent cells with damaged telomeres accumulate during aging in various organisms (Krishnamurthy et al., 2004; Herbig et al., 2006; Jeyapalan et al., 2007). Persistent activation of DDR pathways at dysfunctional chromosome ends has been observed in aging post-mitotic cells, including neurons and adipocytes, also independently of telomere length (Fumagalli et al., 2012; Hewitt et al., 2012). In line with these findings, telomere dysfunction is considered an important contributor to organismal aging (Rossiello et al., 2022; López-Otín et al., 2023).

Increasing evidence indicates that telomere homeostasis is regulated by the telomeric repeat-containing RNA TERRA, long noncoding RNAs transcribed from chromosome ends that associate with telomeres (Feuerhahn et al., 2010). TERRA functions at chromosome ends by base-pairing with telomeric repeats, forming RNA:DNA hybrids (R-loops), and by interacting with telomere binding proteins to regulate telomeric chromatin, telomeric DNA replication, telomere mobility, telomere protection and telomere elongation (Flynn et al., 2011; de Silanes et al., 2014; Azzalin and Lingner, 2015; Chu et al., 2017a; Niehrs and Luke, 2020). TERRA also acts in trans, regulating gene expression as well as participating in extranuclear processes (Chu et al., 2017a; Bettin et al., 2019). Most studies on TERRA have been performed using cancer cell lines. These studies indicated TERRA as a key player of telomere biology and genome integrity (Cusanelli and Chartrand, 2015). Furthermore, TERRA deregulation was found to impact cancer cell survival pointing to TERRA as an attractive therapeutic target (Azzalin and Lingner, 2015). In addition to these findings, TERRA expression has been reported in normal mammalian cells, including fibroblasts, embryonic stem cells and germ cells (Azzalin et al., 2007; Schoeftner and Blasco, 2008; Reig-Viader et al., 2013; Reig-Viader et al., 2014b; de Silanes et al., 2014; Sagie et al., 2014; Chu et al., 2017a; Kordowitzki et al., 2020). Furthermore, TERRA has been detected in human post-mitotic cells (Diman et al., 2016). Studying TERRA in non-cancer settings will be important to help delineate its involvement in genetic pathologies, as well as physiological processes including aging and in age-related diseases (Maicher et al., 2012; Aguado et al., 2020).

In this minireview, we first provide an overview of the current knowledge on TERRA biogenesis in mammalian cells. We then discuss the main findings on the regulation and functions of TERRA obtained in non-cancer cells. In accordance with the topic of this special issue, we lastly report and discuss the evidence on TERRA biology in mammalian germ cells.

In mammalian cells, TERRA molecules are generated by RNA polymerase II from the telomeric C-rich strand, with transcription starting from subtelomeres and proceeding toward chromosome ends (Azzalin et al., 2007; Schoeftner and Blasco, 2008). TERRA transcripts contain a 5′ sequence derived from subtelomeres, followed by tracts of telomeric repeats at their 3′ end (UUAGGG in vertebrates) (Azzalin et al., 2007; Schoeftner and Blasco, 2008; Porro et al., 2010; Diman and Decottignies, 2018) (Figure 1A). TERRA molecules range from 100 nt to 9 kb in length in mammals, and contain a canonical 7-methylguanosine (m⁷G) 5′ cap (Porro et al., 2010). 7% of the total human TERRA population presents a 3′ polyadenylated tail (Figure 1A), which determines the stability and localization of the transcripts (Azzalin and Lingner, 2008; Schoeftner and Blasco, 2008; Porro et al., 2010). While the mechanism of TERRA polyadenylation remains to be elucidated, it has been recently observed in different cancer cell lines that polyadenylation occurs on TERRA transcripts expressed from specific telomeres, indicating that this process can be regulated in a telomere-specific manner (Savoca et al., 2023). Indeed, the hnRNP RALY and the poly(A) binding protein PABPN1 interact with TERRA in HeLa cells and regulate its stability through mechanisms that are defined by the presence of the poly(A) tail. A previous study supported the role of other hnRNPs, such as hnRNP F, in TERRA stability in mouse cells (de Silanes et al., 2014). Which RNA decay enzymes target TERRA in human and mouse cells remains to be defined. Interestingly, emerging evidence indicates that not only the 3′ end of TERRA but also its subtelomeric sequences are important for the stability of the transcripts. Indeed, it has been recently reported that the methyltransferase METTL3 catalyzes the N6-methyladenosine (m6A) modification of TERRA transcripts (Chen et al., 2022) (Figure 1A). These modifications occur within the subtelomeric sequences of TERRA and are recognized by the m6A reader YTHDC1 which stabilizes TERRA molecules. METTL3 or YTHDC1 depletion promotes TERRA degradation (Chen et al., 2022) in human cancer cells using alternative telomere lengthening mechanisms (ALT) which do not involve telomerase activity (Bryan et al., 1995). Further studies will be required to elucidate the mechanisms of TERRA posttranscriptional modifications and the processes involved in TERRA transcripts decay in non-cancer cells. These processes may well be influenced by the structure of these RNAs.

Indeed, the G-rich telomeric tract of TERRA can form stable G-quadruplex (G4) structures composed by the stacking of planar arrangements of four guanines held together by hydrogen bonds (Randall and Griffith, 2009; Collie et al., 2010; Xu et al., 2010) (Figure 1A). G4 structures of TERRA were proposed to mediate the interaction with proteins involved in telomere stability (Takahama et al., 2013; Wang et al., 2017; Mei et al., 2021).

TERRA transcription in mammals is tightly regulated by multiple mechanisms. A subset of human subtelomeres contain CpG-island promoters located upstream of TERRA transcription start sites (Nergadze et al., 2009; Porro et al., 2014a). Methylation of CpG-rich TERRA promoters is regulated by DNMT1 and DNMT3b DNA methyltransferases and represses TERRA expression (Nergadze et al., 2009; Feretzaki et al., 2019; Le Berre et al., 2019). The chromatin state of subtelomeres and telomeres is also a major determinant for the regulation of TERRA transcription. Large amount of evidence supports a model in which a decrease in repressive chromatin marks at these genomic regions leads to chromatin relaxation and increased levels of TERRA (Schoeftner and Blasco, 2008; Marion et al., 2009; Arnoult et al., 2012; Episkopou et al., 2014; Rippe and Luke, 2015; Diman and Decottignies, 2018; Barral and Déjardin, 2020; Novo, 2021). However, a study by Gauchier et al. (2019) in mouse embryonic stem cells (mESCs) suggests that the heterochromatin mark H3K9me3 at telomeres catalyzed by the histone methyltransferase SETDB1 positively correlates with TERRA levels (Gauchier et al., 2019). Differences in H3K9me3 density at subtelomeres versus telomeres and the use of different cell lines may explain the discrepancy of the results between studies (Barral and Déjardin, 2020; Udroiu and Sgura, 2020; Novo, 2021). Finally, TERRA levels are cell cycle regulated. TERRA accumulates in G1 phase, prior to the replication of telomeres, and decreases in S phase in telomerase-positive cells (Porro et al., 2010; Flynn et al., 2011; Arnoult et al., 2012; Flynn et al., 2015). The nature of the cell cycle control of TERRA levels remains to be elucidated.

Several studies indicate that different cell types express different TERRA levels and, importantly, TERRA levels markedly vary among telomeres of a given cell type suggesting that chromosome ends regulate TERRA expression by distinct mechanisms. The study of the telomere-specific regulation of TERRA is fundamental to understanding how the multiple cellular functions of this telomeric lncRNA are coordinated in cells (Rivosecchi J., Jurikova K., Cusanelli E., submitted for publication).

The study of TERRA in non-tumor human and mouse cells is contributing to advance our understanding of the mechanisms regulating TERRA expression as well as of its telomeric and extratelomeric functions (Table 1).

Furthermore, several lines of evidence indicate the involvement of TERRA transcripts in the telomere biology of ICF patients derived cells, at least in part through the formation of telomeric RNA:DNA hybrids. Intriguingly, altered metabolism of RNA:DNA hybrids structures has been described in cells from patients carrying other genetic diseases, including amyotrophic lateral sclerosis 4 (ALS4) (Grunseich et al., 2018), ataxia oculomotor apraxia type 2 (AOA2) (Becherel et al., 2015; García-Muse and Aguilera, 2019) and Aicardi–Goutières syndrome (AGS) (Lim et al., 2015). It will be interesting to investigate whether in these pathological conditions impaired regulation of R-loops structures occurs also at telomeres, suggesting an involvement of TERRA in the diseases.

The expression of TERRA in germline cells during meiosis opens the intriguing possibility that this RNA may participate in the meiotic process during gametogenesis. Alterations of telomere structure and shortening of telomeres in germ cells are associated with reduced fertility (Lopes et al., 2019; Rossiello et al., 2022) due to disrupted gametogenesis which in turn produces aneuploid gametes and miscarriages (Reig-Viader et al., 2016). It would be interesting to assess TERRA levels in these conditions to study its potential contribution to dysfunctional gametogenesis. In this regard, spermatocytes from infertile males have shown reduced number of TERRA foci and altered nuclear distribution compared to control individuals (Reig-Viader et al., 2014a), supporting the protective role that TERRA may exert at telomeres in germ cells.

Studying TERRA in mammalian models different from cancer cell lines will enable the investigation of new potential roles of these telomeric transcripts in more physiological conditions. Despite the differences in telomere biology between mice and humans, which include longer telomeres and broader expression of telomerase in mouse cells, and the distinct features of TERRA transcription regulation and genomic origins between the two species, mouse models may represent an important tool to decipher the physiological functions of TERRA during embryogenesis and in differentiated somatic cells. Generating TERRA KO models will be a difficult task given to the multiple genomic sites transcribing telomeric repeat-containing RNAs at chromosome ends and PARs. Nevertheless, the study of TERRA in the available conditional knock-out models for TERRA-interacting factors or telomere-binding proteins may help advancing our understanding on TERRA biology and to verify the potential involvement of TERRA in the phenotypes of these mice (Cheong et al., 2003; Li et al., 2019; Gaspar et al., 2022). These studies are expected to have implications for the study of TERRA biology also in humans. Indeed, several TERRA-binding proteins are conserved between mice and humans, such as the BLM helicase and the paraspeckles component NONO, a condition that could be attributed to the same G-rich telomeric sequence of TERRA in these species (Deng et al., 2009; Flynn et al., 2011; Petti et al., 2019; Viceconte et al., 2021). Along these lines, studies of TERRA in non-tumor settings will have implications also for cancer research. Cancer arises as a consequence of mutations occurring in normal cells that disrupt oncogenes and oncosuppressor genes, leading to their transformation into malignant cells (Vogelstein and Kinzler, 2015). Organisms have developed tumor suppressive mechanisms to counteract DNA mutations and cellular transformation, most of which remain to be defined (Hanahan and Weinberg, 2011). Studies conducted in normal cells can thus enable the identification of these pathways. Oxidative stress has been identified as one of the multiple factors stimulating tumorigenesis (Hayes et al., 2020). In this regard, the proposed function of TERRA in counteracting oxidative stress in skeletal muscle cells and aged muscle tissues (Diman et al., 2016; Balan et al., 2021) highlights that telomere transcription may participate in a tumor suppressive pathway to prevent oxidative damage to genomic DNA and consequent accumulation of mutations, also in aged tissues in which DNA damage accumulates (Schumacher et al., 2021). Direct evidence for this hypothesis will need to be provided.

Therefore, it will be interesting to study TERRA in aging organisms and tissues. Indeed, cancer is considered a disease of aging (The importance of aging in cancer research, 2022) and several hallmarks of aging overlap with the hallmark of cancer, including genomic instability and DNA damage (Hanahan and Weinberg, 2011; López-Otín et al., 2023). Furthermore, recent evidence indicates that aging tissues contribute to the formation of a tumor promoting microenvironment, sustaining cancer cell malignancy and resistance to therapy. This condition is instructed for a good part by normal cells surrounding the tumor creating a proinflammatory and tumor-permissive environment (Fane and Weeraratna, 2020; Lex et al., 2020). The aging process is associated with accumulation of DNA damage (Pereira and Ferreira, 2013; López-Otín et al., 2023), a condition which results in upregulation of TERRA in different cellular settings (Porro et al., 2014a; Porro et al., 2014b; Tutton et al., 2016). The fact that TERRA-containing exosomes stimulate a proinflammatory response in peripheral blood mononuclear cells suggests that extracellular function of TERRA may participate in the formation of a tumor-permissive environment (Wang et al., 2015). Thus, studies of TERRA in aging organisms and tissues may help verify this hypothesis.

TERRA may also play a cell autonomous role in promoting cellular transformation. Indeed, increased expression of TERRA from dysfunctional telomeres has been shown to contribute to DDR activation at chromosome ends (Porro et al., 2014a; Porro et al., 2014b), a condition well known to fuel genome instability (O’Sullivan and Karlseder, 2010). Human precancerous cells accumulate DDR as result of oncogene activation (Bartkova et al., 2005; Gorgoulis et al., 2005). It will be interesting to study whether this condition results in TERRA deregulation and consequent telomere dysfunction. One could speculate that during aging increased TERRA levels in normal and/or precancerous cells, may induce cell autonomous effects by sustaining telomere dysfunction as well as non-cell autonomous consequences by promoting a pro-inflammatory microenvironment, both activities may seed cancer development. For this reason, studying the mechanisms of TERRA transcription, processing and localization in non-tumor cells may be instrumental to dissect the functions of these telomeric transcripts in cancer.

The use of yeast as a model system has been critical to study the biology of TERRA and telomeric R-loops, also during senescence (Coulon and Vaurs, 2020; Niehrs and Luke, 2020; Zeinoun et al., 2023). Expanding our analyses to different eukaryotes, such as plants (Vrbsky et al., 2010; Majerová et al., 2011), protozoa (Morea et al., 2021; Saha et al., 2021), and zebrafish (Park et al., 2019; Idilli et al., 2020; El Maï et al., 2023) will enable us to broaden our knowledge on this telomeric RNAs potentially unveiling novel and conserved functions. Importantly, species-specific mechanisms regulating TERRA expression will most likely surface, as it has been observed for the different impact of DNA methylation in TERRA expression between mice and humans (Toubiana et al., 2020). Furthermore, it has been reported that TERRA may act as a positive regulator of telomerase in the budding yeast S. cerevisiae and the fission yeast S. pombe (Cusanelli et al., 2013; Moravec et al., 2016); conversely TERRA may repress telomerase in human cells (Redon et al., 2010). Thus, species-specific TERRA functions most likely exist. Studying TERRA in multiple model systems will enable us to help unveil unexpected and intriguing findings which lie ahead on the biology of this much studied and yet not completely understood RNA.