Biogenesis and Regulation of Telomerase during Development and Cancer.

THE BIOGENESIS OF HUMAN TELOMERASE
The last few years have seen tremendous progress in our understanding of telomerase biogenesis, led primarily by advances in cryogenic electron microscopy (cryo-EM) (Nguyen et al. 2018; Ghanim et al. 2021; Wan et al. 2021; Liu et al. 2022), live cell imaging at single-molecule sensitivity (Schmidt et al. 2016, 2018; Laprade et al. 2020), and nascent RNA end-sequencing (Roake et al. 2019). Additionally, the discovery of novel mutations in genes not previously associated with telomere biology disorders contributed to our knowledge of the molecular series of events that are necessary for efficient telomerase function (Dhanraj et al. 2015; Stuart et al. 2015; Tummala et al. 2015; Gable et al. 2019). We will incorporate these recent findings to provide an up-to-date view of telomerase formation and assembly. While in vitro telomerase activity only requires the telomerase reverse transcriptase (TERT) and the telomerase RNA (TR) components, in vivo the catalytically active telomerase RNP is far larger and is composed not only of TERT and TR, but also by the telomerase Cajal body protein 1 (TCAB1), and two copies of each of the proteins that comprise the dyskerin complex: dyskerin (DKC1), NOP10, NHP2, and GAR1 (Fig. 1). Moreover, recent data obtained from cryo-EM identified an H2A–H2B dimer directly bound to an essential motif of TR, which indicates these could be part of the telomerase complex, and modulate TR function (Ghanim et al. 2021; Liu et al. 2022). For ease of reading, we will describe the role of these different components in telomerase biogenesis separately and detail the sequential steps of events that culminate in the assembly of a functional telomerase complex that is recruited to DNA and able to efficiently elongate telomeres.

The Biogenesis, Structure, and Localization of hTR and Its Associated Components
The sequence of events necessary for telomerase assembly and function revolves around its RNA component—TR. Among different species, TRs serve as the platform for the RNP complex assembly and as a template for the reverse transcriptase function of TERT. The TR component of telomerase also plays a central role in telomerase accumulation and localization in vertebrates, through its different domains and association with different components of the telomerase complex. The structure and size of TR components vary significantly between different species, ranging from ~150 nt in ciliates to ~450 nt in vertebrates, and more than 1300 nt in yeast (Theimer and Feigon 2006; Podlevsky et al. 2008). Here, we provide a detailed analysis of the biogenesis and function of the human telomerase RNA (hTR) component.

hTR Structure
In humans, mature hTR molecules are 451 nt long (Fig. 1). hTR is composed of two separate lobes: the H/ACA lobe that contains the H/ACA domain and is bound by TCAB1 and two sets of the dyskerin tetramer complex, and the catalytic lobe, which includes the pseudoknot-template domains and is directly bound to TERT. These two separate lobes are connected by the conserved regions 4 and 5 (CR4/5) of hTR, where the P6 and P6.1 loops are located (Fig. 1). The CR4/5 domain of hTR directly interacts with TERT, independently from the template domain, through its P6–P6.1 hairpin region (Zhang et al. 2011). Recently, it was described that histone H2A–H2B dimers are also found bound to the CR4/5 domain of hTR (Ghanim et al. 2021; Liu et al. 2022), suggesting a role for these histones in the folding, and therefore function, of hTR. The H/ACA domain sits on hTR’s 3′ end and is configured in a “hairpin–hinge–hairpin–tail” arrangement. The “hinge” is formed by the H box consensus sequence (5′-AGAGGA-3′), which is then followed by a 5′-ACA-3′ sequence located 3 nt upstream of hTR’s 3′ end (Mitchell et al. 1999). This H/ACA box domain is also shared with small nucleolar (sno) and small Cajal body (sca) RNA molecules, which act as guide RNAs in the site-specific pseudouridylation of ribosomal RNAs and small nuclear RNAs, respectively (Borchardt et al. 2020). However, to date, no pseudouridylation targets of hTR have been reported, suggesting that the H/ACA domain of hTR could function solely as a stability factor for hTR (discussed below). The H/ACA lobe also contains a stem-loop structure that holds a 4-nt-long Cajal body box (CAB) motif that binds to TCAB1, and a biogenesis-promoting box (BIO box) motif that is involved in hTR stability and accumulation (Egan and Collins 2012; Ketele et al. 2016). On the opposite side of the molecule, at hTR’s 5′ end, sits its catalytic lobe, which represents the largest functional domain of the molecule. This region is divided into three segments, a large pseudoknot loop that directly binds to TERT, a short template region that is complementary to telomeric DNA, and the P1 stem region, which serves as a template boundary element (Fig. 1; Zhang et al. 2011). Mutations in the different regions of the catalytic domain are the most common mutations in hTR found in telomere biology syndrome patients (Revy et al. 2023).

hTR Biogenesis and Assembly of the Telomerase Ribonucleoprotein
While the majority of snoRNAs and scaRNAs are contained within introns of mRNAs and transcribed along with their host genes, hTR is unique in that it is transcribed individually from a dedicated promoter (Feng et al. 1995). However, similarly to snoRNAs and scaRNAs, the dyskerin complex associates with hTR cotranscriptionally and is necessary for its stability (Darzacq et al. 2006). Each of the “hairpins” in the H/ACA domain of hTR initially associates with a heterotetramer composed of dyskerin, NOP10, NHP2, and NAF1. While dyskerin and NOP10 bind directly to hTR, NHP2, and NAF1 bind to dyskerin itself (Egan and Collins 2012; Qin and Autexier 2021). The binding of dyskerin to NOP10, NHP2, and NAF1 happens before its association with hTR. As NAF1 has been shown to bind to nascent H/ACA sno-RNAs (Fatica et al. 2002), this can help explain the cotranscriptional binding of the dyskerin complex to hTR. Binding to the dyskerin complex is facilitated by the BIO box region in hTR and is essential for hTR stability. Indeed, pathogenic mutations in dyskerin, NOP10, NHP2, and NAF1 reduce hTR levels and have been identified in telomere biology syndrome patients, further illustrating the vital role of this complex for telomerase function (Revy et al. 2023).
At a later stage of the maturation process, and after binding of hTR to the dyskerin complex, NAF1 is substituted by GAR1 (Leulliot et al. 2007), an event that takes place in Cajal bodies (CBs) (Darzacq et al. 2006), nuclear compartments that are scaffolded by coilin and play a central role in the biogenesis of snRNAs and snoRNAs (Neugebauer 2017). The recruitment of hTR to CBs is performed by TCAB1, which binds to the CAB domain of hTR and is a core component of the telomerase complex (Venteicher et al. 2009). A recent high-resolution cryo-EM structure of the H/ACA domain of hTR revealed that TCAB1 directly interacts with the 3′ end of NHP2 (Ghanim et al. 2024). This structure of the H/ACA domain of hTR meticulously mapped the interactions between hTR and the two dyskerin heterotetramers and showed that these interact extensively with one another via the two DKC1 subunits (Ghanim et al. 2024). Additional proteins are required for proper hTR biogenesis but are associated with the telomerase complex only transiently. These include SHQ1, which binds to the RNA-binding region of dyskerin (Walbott et al. 2011) before its binding to NAF1 and hTR (Grozdanov et al. 2009). This binding, which happens in the cytoplasm, prevents the premature association of dyskerin with hTR and prevents nonspecific binding of RNAs to dyskerin. In this multistep process of hTR biogenesis, SHQ1 is then removed (upon nuclear import) from dyskerin by the AAA+ ATPases pontin and reptin (Machado-Pinilla et al. 2012), which are themselves essential for hTR accumulation and telomerase activity (Venteicher et al. 2008). This sequential step of events necessary for telomerase RNP biogenesis is additionally controlled by proteins that regulate hTR localization and cellular trafficking and will be discussed in more detail in the following section.

hTR Localization and Trafficking
Multiple lines of evidence establish correct subcellular localization as a central regulatory mechanism for telomerase biogenesis and activity in human cells (Fig. 2, middle column). Telomere elongation by telomerase requires this complex to eventually associate with DNA. However, before that, correct telomerase assembly involves dynamic nuclear trafficking in which many steps revolve around hTR maturation and are dependent on the different proteins that bind to hTR at different stages of its biogenesis. While at any given time, most hTR molecules are found freely diffusing around the cellular nucleus (Schmidt et al. 2016, 2018; Laprade et al. 2020), telomerase associates with CBs in human cells (similarly to other scaRNPs). This association is dependent on the binding of TCAB1 to the CAB box region of hTR (Venteicher et al. 2009; Laprade et al. 2020). Replacement of NAF1 with GAR1 in the telomerase complex most likely happens immediately before hTR association to TCAB1 and consequent translocation to CBs (Darzacq et al. 2006), and TCAB1 has been shown to directly interact with GAR1 (Ghanim et al. 2021), highlighting a highly coordinated assembly/trafficking process. TCAB1 folding and action is in turn dependent on the TRiC chaperonin (Freund et al. 2014), and loss of either TriC or TCAB1 leads to telomerase mislocalization and reduced catalytic activity, without reducing hTR levels (Venteicher et al. 2009; Freund et al. 2014; Chen et al. 2018). Indeed, while in wild-type (WT) cells hTR is transiently associated with the nucleolus, deletion of TCAB1 causes a significant accumulation of hTR in nucleoli (Fig. 2, left column; Venteicher et al. 2009; Klump et al. 2023). Accordingly, it has been recently demonstrated that TCAB1 binding to hTR prevents its trafficking into nucleoli (Klump et al. 2023), which indicates that hTR molecules found in nucleoli might represent immature precursors (Nguyen et al. 2015).
Recent evidence also shows that the recruitment of hTR to telomeres does not depend on its localization to CBs (Laprade et al. 2020), which is substantiated by the fact that the ablation of CBs (through depletion of coilin) does not impair the trafficking of hTR to telomeres or telomerase function (Stern et al. 2012; Chen et al. 2015). However, TCAB1 has also been shown to control telomerase catalytic activity by promoting the proper folding of hTR’s CR4/5 region, which increases its binding to TERT (Chen et al. 2018). Interestingly, hTR association with TERT reduces its residence time in CBs, which is proposed to facilitate its sequential recruitment to telomeres (Laprade et al. 2020). While the molecular events leading to telomerase recruitment to telomeres will be discussed in a separate section below, these results show the dynamic process of hTR trafficking in the nucleus, and how proper hTR localization determines telomerase efficiency in human cells, which is corroborated by mutations in TCAB1 leading to mislocalized telomerase and causing severe phenotypes in patients suffering from telomere biology disorders (Zhong et al. 2011).

Posttranscriptional Modifications Are Essential Regulators of hTR Localization, Levels, and Function
Over the last few years, different posttranscriptional modifications to both the 5′ and 3′ end of hTR have emerged as major regulators of hTR biogenesis and telomerase activity (Fig. 3). Importantly, these modifications are performed by pathways that can be modulated to restore telomerase activity and telomere homeostasis in cells harboring mutations that cause reduced levels of hTR, creating new possibilities for the clinical management of patients suffering with different telomere biology syndromes. These posttranscriptional modifications are discussed below.

The methylation status of the 5′ cap regulates hTR levels and localization.
RNA polymerase II (RNA Pol II) transcripts acquire a monomethylguanosine (MMG) modification on their 5′ end, referred to as 7-methylguanylate cap (m7G). This 5′ end capping process happens cotranscriptionally and regulates different processes on these RNA molecules, including nuclear export and degradation (Ramanathan et al. 2016). While in snRNAs formation of an MMG cap recruits the cap-binding complex (CBC), the adaptor protein PHAX, and the nuclear export protein CRM1 for cytoplasmic export, MMG-capped snoRNAs are directed by PHAX directly into CBs (Izaurralde et al. 1995; Ohno et al. 2000). Once in CBs, MMG caps of snoRNAs are hypermethylated to N2, 2, 7 trimethylguanosine (TMG) caps by trymethylguanosine synthase 1 (TGS1). TMG-capped snoRNAs are then recruited to nucleoli independently from CRM1 (Pradet-Balade et al. 2011). While the mechanisms that specifically control hTR export and localization are not fully described, it has recently been shown that depletion of TGS1 (and therefore prevention of TMG cap formation) increases hTR association to both the CBC complex and Sm chaperone proteins, which causes an increase of mature hTR levels in the nucleus and cytoplasm of cells (Fig. 2, right column; Chen et al. 2020). MMG-capped hTR remains functional and able to bind to TERT, and TGS1-depleted cells show increased telomerase activity and telomere elongation when compared to their WT counterparts (Chen et al. 2020). Although the molecular mechanisms causing hTR accumulation in TGS1-depleted cells remain obscure, these results indicate that TGS1 silencing could be a potential avenue for telomere elongation in cells that harbor mutations that impair telomerase activity due to reduced hTR levels (Batista et al. 2022). Indeed, the chemical inhibition of TGS1 with sinefungin has recently been shown to increase hTR levels and telomerase activity in human cells (Buemi et al. 2022; Galati et al. 2022), including cells from patients with telomere biology disorders (Galati et al. 2022).

Correct 3′ end processing is a major determinant of hTR accumulation and telomerase activity.
While mature hTR comprises a 451-nt-long molecule, at any given time hTR molecules exist as a pool of different transcripts that vary in length, due to 3′ end extensions (Goldfarb and Cech 2013; Roake et al. 2019). These 3′ end extended molecules are encoded by the hTR locus, and it has been suggested that they are formed by readthrough of RNA Pol II (Qin and Autexier 2021). While the regulation of hTR transcription termination in human cells remains mostly unknown (as unlike other snoRNAs, hTR is not contained in introns of host genes), it has recently been shown that the integrator complex regulates termination of hTR transcription (Rubtsova et al. 2019), as is the case for other RNA classes transcribed by RNA Pol II (Mendoza-Figueroa et al. 2020). While the majority of hTR is composed of mature, 451 nt molecules, nascent RNA end-sequencing has recently revealed that the majority of hTR is initially transcribed as longer precursors, which are then processed into 451 nt molecules (Roake et al. 2019).
Following transcription, hTR precursors are rapidly polyadenylated at their 3′ end by the noncanonical poly(A) polymerase PAPD5 (Fig. 3; Moon et al. 2015; Tseng et al. 2015; Shukla et al. 2016). PAPD5 is part of the TRAMP complex, which is composed also of the helicase MTR4 and ZCCHC7, a zinc-finger protein that binds to RNA (Schmid and Jensen 2019). The TRAMP complex facilitates the surveillance and degradation of abnormal RNAs by the RNA exosome complex, which represents the main 3′–5′ ribonuclease in human cells and acts on most types of nuclear and cytoplasmic RNAs (Chlebowski et al. 2013). Indeed, the addition of 3′ end poly(A) tails by PAPD5 quickly leads to exosome-mediated degradation of hTR (Moon et al. 2015; Tseng et al. 2015; Shukla et al. 2016). The recruitment of the RNA exosome complex is performed by the nuclear exosome targeting (NEXT) complex, which is composed of a dimer of MTR4, ZCCHC8, and RBM7 (Gerlach et al. 2022). Depletion of either ZCCHC8 or RBM7 causes increased levels of 3′ end adenylated hTR precursors (Tseng et al. 2015), and mutations in ZCCHC8 have recently been found in pulmonary fibrosis patients with short telomeres (Gable et al. 2019). Cells harboring ZCCHC8 mutations showed an accumulation of 3′ end extended hTR, at the expense of mature hTR molecules, leading to reduced telomerase activity (Gable et al. 2019). These results indicate that exosome targeting is a major regulator of hTR maturation and function. Interestingly, the 3′ and 5′ end posttranscriptional modifications to hTR may act in a coordinated fashion, as the CBC complex, which binds to MMG-capped hTR molecules, also recruits the exosome RNA decay machinery through NEXT (Tseng et al. 2015). More research is necessary to clarify how these processes are coordinated.
Working in opposition to PAPD5, poly(A)-specific ribonuclease (PARN) deadenylates hTR precursors, preventing their degradation by the exosome and leading to the accumulation of mature, functional hTR molecules (Moon et al. 2015; Shukla et al. 2016; Roake et al. 2019). Moreover, it has been suggested that PARN could, in addition to deadenylating hTR precursors, also directly trim hTR 3′ end transcripts to generate mature, 451 nt hTR molecules (Tseng et al. 2018). The role of PARN in hTR biogenesis was initially found through the discovery of clinically relevant PARN mutations that caused significant telomere shortening in patients. These patients show low levels of mature hTR, resulting in reduced telomerase activity and telomere shortening (Dhanraj et al. 2015; Stuart et al. 2015; Tummala et al. 2015). The direct correlation between exosome-mediated decay of 3′ poly(A) tailed hTR precursors and telomerase activity opens the possibility that the inhibition of 3′ end tailing by PAPD5 could be explored as a clinical alternative for patients harboring mutations that cause reduced accumulation of mature, functional hTR. Indeed, the genetic inhibition of PAPD5 in cells harboring mutations in DKC1 or PARN reduces the abundance of 3′ extended hTR molecules, and increases the number of deadenylated, functional hTR molecules (Boyraz et al. 2016; Shukla et al. 2016). Moreover, the genetic inhibition of PAPD5 restores hematopoietic development and output in cells with mutant DKC1 (Fok et al. 2019), rescuing, therefore, a major phenotype found in telomere biology disorders. Adding support to this hypothesis, the pharmacological inhibition of PAPD5 was recently shown to increase mature levels of hTR, increase telomerase activity, promote telomere lengthening, and improve hematopoietic development in both DKC1 and PARN mutant models (Nagpal et al. 2020; Shukla et al. 2020).
Finally, while the opposing roles of PARN and PAPD5 during hTR biogenesis are well established, it is interesting that in cells depleted of both (PARN and PADP5), the maturation kinetics of hTR is similar to WT cells while 3′ end adenylation of precursors is significantly reduced (Roake et al. 2019). This indicates the possibility of an additional 3′–5′ exonuclease able to deadenylate hTR and promote its maturation. A possible candidate is TOE1, an exonuclease that localizes to CBs and deadenylates different noncoding RNA classes (Son et al. 2018). Indeed, TOE1-deficient cells show an increase in oligoadenylated hTR precursors and impaired telomerase activity (Machado-Pinilla et al. 2012).
In conclusion, the biogenesis of hTR is a highly coordinated process, involving multiple proteins and protein complexes, with successive steps occurring sequentially at different subnuclear localizations. While central to telomerase biogenesis and activity, and despite tremendous progress in recent years, many open questions remain on the regulators of the hTR maturation process and how these can be exploited for targeted treatment of telomere biology disorders.

Regulation of hTERT Expression and Levels
While hTR is ubiquitously expressed in human cells, human TERT (hTERT) expression is restricted to specific cell types and developmental stages, often serving as a limiting factor for telomerase assembly and activity in human tissues (Bodnar et al. 1998). While high levels of telomerase are found in the germline, during embryogenesis (in the inner cell mass of the blastocyst), and in various progenitor/stem cell compartments, telomerase is quickly silenced as development progresses and cells differentiate into functional lineages (Wright et al. 1996). On the contrary, cellular reprogramming experiments demonstrated that telomerase is reactivated in induced pluripotent stem cells generated from human fibroblasts, consolidating the developmental stage as a major regulator of hTERT expression (Takahashi et al. 2007; Agarwal et al. 2010; Batista et al. 2011). While often seen as a cancer prevention mechanism, as cells with no telomerase show progressive telomere shortening that results in senescence, the molecular mechanisms regulating hTERT suppression during tissue development and formation remain poorly understood. Similarly, mechanisms regulating hTERT activation during tumorigenesis are not yet completely described, despite significant progress over the last few years (Yuan et al. 2019; Lorbeer and Hockemeyer 2020). In this section, we summarize different events that contribute to the regulation of hTERT expression, and therefore to the biogenesis of telomerase in human settings. A more comprehensive view of this subject can be found in Martin and Hockmeyer (2025).

Genetic and Epigenetic Factors that Regulate hTERT Transcription
The hTERT gene is ~40 kb long, consisting of 16 exons (15 introns) localized in the short arm of chromosome 5 (5p.15:33) (Fig. 4; Yuan et al. 2019). It has a single promoter that is rich in binding sites for various transcription factors including Myc, Klf4, and Sp1. While these have been shown to regulate hTERT transcription, their expression does not completely rescue hTERT silencing in terminally differentiated human somatic cells (Greenberg et al. 1999; Wu et al. 1999; Oh et al. 2001; Wong et al. 2010). However, hTERT is reactivated in most human cancers, and different molecular mechanisms account for this, including integration of exogenous tumor viral regulators (including Epstein–Barr virus [EBV], and cytomegalovirus [CMV]) (Bellon and Nicot 2008), and hTERT copy number variations (Gay-Bellile et al. 2017). In addition, hypermethylation of the hTERT promoter region has recently been associated with hTERT up-regulation in human cancer. This hypermethylation is found specifically in a region immediately upstream of the TERT core promoter that contains 52 different CpG sites and is now described as the TERT hypermethylated oncological region (THOR) (Lee et al. 2019). This epigenetic regulation of hTERT expression, where hypomethylated THOR is associated with hTERT repression and hypermethylated THOR is found in cancer samples that show hTERT reactivation, has been found in a variety of cancer types (Lee et al. 2019) and potentially accounts for a significant mechanism leading to telomerase formation and activity during tumorigenesis (Lee et al. 2021). Once transcribed, hTERT transiently binds to the chaperone heat shock protein 90 (Hsp90), which increases its stability and folding, facilitating its binding to hTR and modulating telomerase activity (Holt et al. 1999; Forsythe et al. 2001).

Somatic Mutations in the TERT Promoter Robustly Increase hTERT Levels and Telomerase Activity
While the mechanisms described above are commonly observed in tumor samples, the most frequently observed mechanism of hTERT reactivation in human cancer is the acquisition of somatic mutations in specific positions of the TERT proximal promoter (−57; −124 and −146, in relation to the ATG transcriptional start site). These mutations (referred to as TERT promoter mutations [TPMs]), were initially found in melanoma (Horn et al. 2013; Huang et al. 2013), but have now been found in multiple tumor types that are usually derived from cells with reduced self-renewal, including gliomas and hepatocellular carcinomas (HCCs) (Killela et al. 2013). In fact, a comprehensive analysis of 31 cancer types from The Cancer Genome Atlas (TCGA) established that TPMs represent the most common noncoding mutation in cancer and one of the most frequent mutations found in cancer overall, with a staggering number of 27% of all analyzed samples harboring a TPM (Barthel et al. 2017). Functionally, TPMs create de novo binding sites for ETS (E26 transformation specific) transcription factors, most predominantly the GA-binding protein α and β (GABPα and GABPβ) (Bell et al. 2015). This binding happens as a tetramer (with two GABPα/β being formed), and the inhibition of either GABPα or GABPβ reduces hTERTexpression in cancer cells harboring mutations in the TERT promoter (Bell et al. 2015; Mancini et al. 2018). Mutations in the TERT promoter are heterozygous, and the allele harboring the mutation is the only transcriptionally active allele, showing histone marks typically associated with active chromatin (while the allele harboring the WT promoter sequence retains marks of epigenetic silencing) (Huang et al. 2015; Stern et al. 2015). Mutations in the TERT promoter occur early during tumorigenesis (Nault and Zucman-Rossi 2016), and expression analysis shows that up-regulation of hTERT levels in cells harboring these mutations is gradual and associated with the accumulation of critically short telomeres (Chiba et al. 2017). Combined, these results indicate that TPMs occur early during carcinogenesis where they initially increase cellular life span by lengthening of critically short telomeres and then promote continued cellular proliferation (Lorbeer and Hockemeyer 2020).

Alternative Splicing Regulates hTERT Levels during Development
hTERT levels are not only regulated transcriptionally (as discussed above) but also posttranscriptionally. Transcriptome analysis of hTERT has revealed more than 20 alternatively spliced isoforms (Sæbøe-Larssen et al. 2006; Bollmann 2013) that are mostly predicted to encode catalytically inactive proteins. Interestingly, different hTERT isoforms have been associated with different cancer types (both telomerase positive and negative) and developmental stages, suggesting that they play important roles in hTERT regulation and telomerase activity. Indeed, several lines of evidence suggest that alternative splicing regulates telomerase activity during human development (Ulaner et al. 1998; Brenner et al. 1999) and cancer (Fajkus et al. 2003; Petrenko et al. 2010). The hTERT protein contains four domains (the telomerase amino-terminal [TEN] domain; the TR-binding domain [TRBD]; the reverse transcriptase domain [RT domain], and the carboxy-terminal extension [CTE]) (Fig. 4). Most alternatively spliced hTERT mRNAs differ in exons that encode the RT domain and are, therefore, predicted to generate catalytically inactive proteins. However, some of these splice variants have been shown to be translated into protein and to negatively regulate telomerase activity in human cells (Zhu et al. 2014), or even to regulate cellular processes outside telomere elongation (Bollmann 2013). Recently, targeted sequencing techniques that allow analysis of low abundance transcripts identified hTERT reads spanning the junction of exon 1 to exon 3 (indicating loss of exon 2–hTERT-ΔEx2) in human terminally differentiated cells, but not in their undifferentiated, isogenic counterparts (Penev et al. 2021). hTERT-ΔEx2 was enriched in cell types that lack telomerase activity, and reduced in pluripotent cells with high telomerase levels, including embryonic and induced pluripotent stem cells. The absence of exon 2 causes a frameshift that creates two tandem premature stop codons in exon 3, which have been predicted to trigger nonsense-mediated RNA decay (Withers et al. 2012; Hug et al. 2016) and cause rapid decay of hTERT transcripts. The forced inclusion of this exon during stem cell differentiation prevents telomerase silencing during tissue development (Penev et al. 2021). This posttranscriptional regulation of hTERT, therefore, directly modulates telomerase activity during development and synergizes with transcriptional mechanisms to promote telomerase biogenesis and function in human stem cells (Barranco 2021; Penev et al. 2021).

Telomerase Recruitment to Telomeres
An essential step for telomere elongation is the recruitment of a functional telomerase complex to telomeric DNA. However, the low abundance of telomerase and telomeres in human cells (estimated at ~240 telomerase molecules and 184 telomeres in the late S phase of the cell cycle) (Xi and Cech 2014) indicates that this process does not happen by simple diffusion. Rather, extensive research shows that telomerase is actively recruited to telomeres, where its action is regulated to promote homeostasis of telomere length. As we discuss in previous sections, the assembly of the telomerase RNP is dependent on the expression of its multiple components, happens at multiple nuclear sites, and with regulated timing. In this section, we discuss how telomerase is recruited to telomeres and how telomere length is regulated in human cells.

Localization to Cajal Bodies Increases Telomerase Recruitment to Telomeres
Multiple lines of evidence indicate that TCAB1 facilitates the recruitment of telomerase to telomeric DNA (Tycowski et al. 2009; Venteicher et al. 2009). Accordingly, mutations in TCAB1 are found in patients suffering with telomere biology syndromes (Zhong et al. 2011). TCAB1 binds to the CAB domain of hTR (Fig. 1), and if this interaction is abrogated, telomerase mislocalizes to the nucleolus (Tycowski et al. 2009; Venteicher et al. 2009). If TCAB1 is absent, hTR does not localize to CBs but is rather tightly associated with the nucleolus (Laprade et al. 2020). However, the importance of telomerase localization to CBs remains unclear, as telomerase in mice cells does not always accumulate in CBs (Tomlinson et al. 2010) and even some human tumor cells (with high levels of telomerase) retain efficient telomere maintenance after disruption of CBs (Chen et al. 2015). These results indicate that telomerase association with CBs could modulate telomerase activity by improving telomerase–telomere interaction and more efficient telomere elongation, particularly in cells with reduced levels of telomerase expression. However, more studies are required to confirm this hypothesis and to precisely pinpoint how the localization of telomerase to CBs promotes its recruitment to telomeres in human cells.

Shelterin Recruits Telomerase to Telomeres
Telomeric DNA is comprised of TTAGGG repeats extending up to 15 kb length in humans. It can be divided into two separate structures, a double-stranded DNA region that is several kilobases long, followed by a single-stranded 3′ end tail known as the G overhang (Lim and Cech 2021). Telomeres are associated with shelterin, a protein complex composed of six different proteins that bind both the double-stranded and single-stranded telomeric regions (Palm and de Lange 2008). Association with shelterin arranges telomeres into different structures, including end-capped telomeres and telomere loops (Lim and Cech 2021), structures that must be resolved for telomerase to access telomeric overhangs. Interestingly, the recruitment of telomerase to DNA ends is mediated by shelterin, which facilitates the pairing of the 3′ overhang with the hTR template within telomerase (Hockemeyer and Collins 2015).
Shelterin is comprised of Telomeric Repeat binding Factors 1 and 2 (TRF1 and TRF2), Protection of Telomeres 1 (POT1), TRF1-Interacting Nuclear Protein 2 (TIN2), Rap1 (the human ortholog of the yeast Repressor/Activator Protein 1), and TPP1. TRF1 and 2 bind double-stranded telomeric DNA, and POT1 binds to the single-stranded 3′ overhang. Rap1 works as an accessory unit of TRF2. TIN2 directly binds to TRF1 and 2, whereas TPP1 is connected to shelterin via its binding to both TIN2 and POT1 (de Lange 2018). The recruitment of telomerase to telomeres was initially shown to be reduced with the loss of TIN2-anchored TPP1 (Abreu et al. 2010). More recent data show that the recruitment of telomerase to telomeres is mediated by the binding of the TEN domain of TERT to the TEL-patch region of TPP1, which resides on the surface of its oligonucleotide-binding fold domain (Fig. 5; Nandakumar et al. 2012; Sexton et al. 2012; Zhong et al. 2012). Mutations in the TEL-patch domain of TPP1 lead to telomere shortening in telomerase-positive cells, and have been found in patients with telomere biology disorders, highlighting the importance of this interaction for telomerase recruitment (Guo et al. 2014; Kocak et al. 2014; Bertrand et al. 2024). Additionally, mutations in the amino-terminal OB-fold domain of TPP1 have also been identified in telomere biology disorder patients (Tummala et al. 2018), and this region of the protein has been shown to be essential for telomerase processivity and recruitment to telomeres (Grill et al. 2018). The critical role of TPP1 for telomerase action at telomeres has been confirmed by high-resolution live cell imaging experiments that also show that before forming a stable association with telomeres, telomerase probes telomeres thousands of times during the S phase of the cell cycle. Both of these transient interactions and stable interactions require TPP1-TERT binding (Schmidt et al. 2016, 2018). Finally, these functional studies have been confirmed by recent structural studies that defined the interaction between TPP1 and telomerase that facilitate recruitment and processivity (Liu et al. 2022; Sekne et al. 2022).
The role of shelterin in regulating telomerase and telomere length has been further solidified by recent genome-wide sequencing efforts in different types of human cancers. These have identified heterozygous mutations in POT1 as a recurrent event in tumorigenesis (Wu et al. 2020). These cancer-associated mutations in POT1 do not cause deprotection of telomeres and do not activate DDRs but rather seem to be selected for and persist over time (Kim et al. 2021). In fact, heterozygous mutations in POT1 have recently been shown to cause elongated telomeres in patients with clonal hematopoiesis and a wide range of benign and malignant neoplasms (DeBoy et al. 2023). The increased risk of cancer in these patients seems to be directly related to their increased capacity to sustain telomere length over time, leading to extended cellular longevity. These results show that while exacerbated telomere shortening is associated with increased genetic instability and DNA damage in patients suffering with telomere biology disorders such as dyskeratosis congenita (Revy et al. 2023), enhanced telomere maintenance, leading to exacerbated telomere elongation, is also detrimental for correct tissue maintenance over time.
Finally, as the protection of single-stranded telomere overhangs by POT1–TPP1 requires its binding to the central shelterin component TIN2 (Takai et al. 2011), it is not surprising that mutations in TIN2 are also found in patients with telomere biology disorders (Revy et al. 2023). These mutations do not reduce telomerase activity and do not impact TIN2 localization but disrupt TPP1-dependent recruitment of telomerase to telomeres, leading to exacerbated telomere shortening in patients (Yang et al. 2011). These results indicate that TIN2 might act as a telomerase stimulating factor, and that mutations found in patients, instead of causing direct deprotection of telomeres through defective shelterin function, lead to disease due to reduced telomerase action in patients (Frank et al. 2015). In agreement with this possibility, we know that TIN2 cooperates with TPP1–POT1 to act as a telomerase-stimulatory factor (Pike et al. 2019). Of note, mice harboring a clinically relevant TIN2 mutation showed telomere defects that were, at least partially, independent of telomerase, as they were also observed in telomerase-negative cells (Frescas and de Lange 2014). These results indicate more research is necessary to establish precisely how mutations in TIN2 cause severe phenotypes in patients, and which molecular strategies can be used to mitigate these defects.

CONCLUSIONS AND PERSPECTIVES
Research regarding telomerase biogenesis and assembly has benefited significantly from recent developments in single-molecule imaging and improvements in cryo-EM resolution. Individual telomerase molecules can now be tracked in human living cells, with their precise subcellular localization pinpointed during different stages of RNP maturation and trafficking to telomeres. Adding to this, the enormous amount of data generated by high-throughput sequencing allowed for the identification of mutations in genes not previously implicated in telomerase biogenesis, such as PARN and ZCCHC8. These results have also helped increase our knowledge on the several steps necessary for telomerase biogenesis, in particular regarding its RNA component, hTR.
Telomerase facilitates continuous cellular proliferation while, in unperturbed settings, mitigating tumorigenesis. Not surprisingly, its biogenesis and regulation are highly coordinated, involving different subcellular compartments and multiple protein complexes that ensure telomerase molecules are assembled before the decay of its RNA component, and able to eventually attach to, and elongate, telomeres. Similarly, the tight control of expression of its RT component adds an additional layer of security to prevent widespread telomere elongation in human tissues. These events happen in a multistep process that includes the assembly of the H/ACA lobe of telomerase, which is then followed by posttranscriptional modifications in both 5′ and 3′ end of hTR, and finally by its association with the RT component of telomerase. A summary of the many factors influencing telomerase biogenesis, assembly, recruitment, and function at telomeres can be found in Figure 6.
The relevance of the tight control of telomerase biogenesis becomes clear when mutations compromise this highly coordinated process. On one hand, increased levels of TERT, caused by mutations in its promoter, gene amplifications, or epigenetic mechanisms are found in most human cancers. On the other hand, mutations that compromise telomerase biogenesis, in particular mutations that compromise the stability and localization of hTR, are the most common genetic alterations found in patients with telomere biology disorders. However, the fact that telomerase biogenesis and assembly is a complex pathway opens an array of possibilities that can be exploited for clinical intervention, such as the inhibition of PAPD5 and TGS1 in patients harboring mutations that compromise hTR stability (Batista et al. 2022).
Finally, while our knowledge of telomerase biogenesis has significantly increased over the last decade, several aspects of this process remain obscure. For instance, a possible functional link between the 5′ and 3′ end posttranscriptional modifications of hTR could shed light on how these pathways act to promote its stability in different cell types. Moreover, several lines of evidence indicate that an additional pathway for the 3′ end processing of hTR must exist and could also have clinical implications. Moreover, several aspects regarding the timing and location of different steps during biogenesis and to what extent specific subcellular location sites determine telomere homeostasis remain unknown. These will likely be achieved through in-depth structural analysis of the intermediates formed during telomerase biogenesis and trafficking. Similarly, additional research is necessary to determine more precisely how telomerase interacts with and is regulated by other telomere-binding proteins, including shelterin and the CST complex.
Collectively, we hope we have demonstrated that while tremendous amounts of data have been generated since the initial discovery of telomerase, much remains to be determined on how this fascinating RNP is formed, assembled, and directed to telomeres, where it plays an essential role for cellular and organismal viability.