Unlike circular DNA found in prokaryotes, eukaryotes share minisatellite sequences at the end of linear chromosomes named telomeres. The importance of this particular structure was first demonstrated in the late 1930s by McClintock and Muller [1
]. By irradiating Drosophila melanogaster
fruit flies, they observed that mutations never occurred at the end of the DNA molecule suggesting the presence of a protective cap restricted to these regions. Since then, explorations and hypothesis on their complete function never ceased within the scientific community and beyond (i.e., general audience).
Major findings on the organization of telomeres were obtained in the Tetrahymena
ciliate model, which possess hundreds of mini telomeres, with the identification of its nucleotide composition [2
] and the deciphering of the telomerase-dependent mechanism involved in extending this hexameric sequence [3
Thanks to the increased sensibility of techniques, the human telomeric sequence composed of a 5′-TTAGGG-3′ repeated motif was later identified [6
] followed by the identification of the particular 3′ overhang and T-loops structure [7
] bound by proteins forming the sheltering complex [10
]. Altogether, the nucleoprotein structure creates the protective cap hypothesized in the 1940s, and set the molecular base of subsequent telomere studies.
The genetic information of eukaryotes is assembled in a linear genome. Hence, the first major role of DNA ends is to protect informative genetic sequences from erosion. Free DNA extremities are exposed to exonuclease and at risk for non-homologous ends joining between DNA molecules [20
]. These regions are thus considered as endangered sequences for the cell. Telomere structure limits this loss and preserves the integrity of chromosomes while allowing distinction between natural chromosome ends and damaged DNA that need to be repaired [21
From cancer research to aging studies, investigations on the regulation of the very tip of chromosomes led to paradoxical observations: (i) telomerase, the enzyme responsible for telomere elongation is not produced in somatic cells, (ii) most cancer telomerase-positive cells display rather short telomeres as telomerase upregulation is considered a late event during malignant transformation; (iii) stem cells harbors long telomeres and express telomerase; (iv) short telomeres are associated with aging and increased cancer risk, or the complete opposite; (v) telomere maintenance involves epigenetic mechanisms and chromatin changes. Altogether, the wealth of studies available in the literature raises questions on the relevance, feasibility, and caveats that one encounters when questioning telomeres dynamics. In the end, does telomere length and size matter?
In this review, we will discuss the past decades of telomere research by crossing lessons on telomere length regulation and impact in Human. We will first focus on telomere length dynamics before developing on consequences at the epigenetic level and possible implication on human health through position effect mechanisms.
Telomeropathies usually refer to diseases caused by defects in factors involved in the telomere maintenance machinery such as Dyskeratosis congenital, Revesz, Coat Plus or Hoyeraal Hreidarsson syndromes [113
]. Telomere dynamics is profoundly affected in these diseases due to haploinsufficiency of components of either the telomerase holoenzyme (DKC1, TERT, and TERC), the TCAB1 shuttling factor, SnRNP components (NHP2 and NOP10 required for telomerase activity), the TIN2 shelterin component or the regulator of telomerase length helicase, RTEL1 [116
]. However, unrelated syndromes involving changes in subtelomeres or telomeres might also be considered as telomeropathies [115
]. Indeed, despite its protective role, telomere and chromatin environments are also involved, directly or indirectly, in different human genetic diseases.
Chromosomal translocation of region encompassing whole gene loci and thus implicating modification of the chromatin environment in gene expression modulation were observed in numerous genetic diseases. In the context of these so called “position effect mutations” modification in gene expression can be the consequence of two mechanisms. First, the chromosomal abnormality can separate the coding sequence from its regulating enhancers or silencers. Second, in case of translocations, the gene can be delocalized to a region harboring a different chromatin state. In both cases, a gene normally repressed can be activated and vice versa [120
]. In the majority of these diseases, position effect mutations can lead to slightly different phenotypes from that caused by coding regions mutations. Indeed, position effect mutations can affect gene expression only in a subset of tissues or produce a milder phenotype but in most cases, the underlying mechanism is not established.
Non-physiological modification of telomere and chromosome ends integrity has also been associated to a number of diseases. Some patients diagnosed with various malformations, mental impairment, and growth retardation display formation of a ring chromosome. Such structures can occur on any autosomes and are formed by deletion on the short and long arms of the chromosomes followed by fusion at the breakage point. Consequences of ring chromosomes depend on different parameters: the chromosome implicated, the number of genes deleted, and the stability of the ring chromosomes. Most ring chromosomes are associated with loss of genetic material but in some rare cases, only telomeric regions are lost resulting in fusion of residual telomeres from both ends, forming a single telomere in the ring. This mechanism has been observed in ring 14, 17, and 20 syndromes [121
]. Patients harboring ring chromosomes with telomere-telomere fusion showed downregulation of genes of the implicated chromosomes suggesting changes in the chromatin architecture. Indeed, telomere-telomere fusion can lead to the formation of a unique long telomere, enhancing telomere position effect into the ring chromosomes linking ring syndrome clinical phenotype to TPE-sensitive genes.
One of the best characterized human genetic diseases implicating TPE is Facio-Scapulo-Humeral Dystrophy (FSHD) [90
]. A particularity of this disease is the late onset, with appearance of the first symptoms around the second life decade. The locus associated to FSHD is in the subtelomeric region of the long arm of chromosome 4 (4q35 locus) raising the possibility of TPE implication in FSHD pathophysiology. The leading candidate gene for FSHD pathogenesis is the DUX4 homeobox protein encoded by the last D4Z4 macrosatellite at the 4q35 locus. In FSHD patients presenting a diminished number of D4Z4 repeats (1–10 in FSHD1 patients versus 11 to 150 in non-affected individuals), relaxation of the D4Z4 array chromatin structure is associated with DUX4
In this disease, the impact of telomere shortening was investigated in immortalized myoblasts expressing a floxable hTERT
]. In myoblasts harboring different telomeres size, DUX4
expression was upregulated upon telomere shortening in FSHD cells, with a progressive effect as telomeres shorten, long before replicative senescence. TPE also extended to the upstream FRG2
gene, located 70 kb more centromeric to DUX4
, but was less prominent and not observed for FRG1
located 90 kb upstream of FRG2
. These observations uncovered FSHD as the first disease linked to classical TPE [90
]. Using Hi-C experiments and 3C or 3D-FISH validation, long-range interactions between the FRG1
locus were also identified in myoblasts from FSHD1 patients [125
]. Interestingly, this long distance loop is lost upon telomere shortening in FSHD cells but not in control cells and correlates with SORBS2
]. This suggests that telomere shortening on a D4Z4-contracted allele induces modification of the 4q35 folding with a more relaxed chromatin state. The SORBS2
gene is silenced in normal cells by TPE-OLD mechanisms while this telomere-dependent mechanism is impaired in FSHD myoblasts with cooperation between the telomere and the repetitive macrosatellite array [125
Another example of rare a disease involving telomeric changes is the immunodeficiency, centromeric region instability, and facial anomalies syndrome (ICF). ICF arises due to mutation in four different genes, DNMT3B
(ICF3), and HELLS
]. Interestingly, while hypomethylation of subtelomeric regions, increased TERRA levels, telomere shortening and rapid senescence have been clearly evidenced in ICF1 [131
], these hallmarks remained unchanged in ICF2-4 [133
]. This again highlights the cross talks between the chromatin structure of telomeres and subtelomeres and further indicates that subtelomeric heterochromatin is regulated by specialized factors, the majority of which remain to be identified.
Furthermore, in agreement for a role of telomere subnuclear positioning in telomere regulation, telomeric defects are also observed in premature aging syndrome such as Hutchinson Gilford progeria linked to mutation in the LMNA
gene encoding A-type Lamins [134
]. In this disease and corresponding animal models, telomeres are shortened and mislocalized [135
]. This is accompanied with ROS accumulation, an increased senescence and DNA damage response highlighting the existence of functional links between the different hallmarks of aging [138
Telomere length is thus at the corner of different biological mechanisms but also associated to a broad range of diseases, from rare genetic syndromes to common ones. As we previously described, telomere shortening occurs normally during aging. This information has been broadly disseminated in the general population with an increasing number of newspapers mentioning the links between aging and telomeres length. As a consequence, numerous private companies offer telomere length measurement with the promise to “help you stay younger longer”, not mentioning cures or food supplement protecting telomeres.
A study published in 2009 aimed at establishing correlations between telomere length, causes of death and years of healthy life (YHL) by applying different statistical analysis on large population cohorts to prove correlation between these three factors [139
]. The only positive and significant correlation was made between telomere length and YHL, suggesting that people with longer telomere could appreciate longer healthy aging than others. In agreement, telomere shortening has been associated with a number of age-related diseases such as cardiovascular diseases, dyslipidemia [140
], hypertension [141
], atherosclerosis [142
], stroke [143
], cancers [144
], or diabetes [145
]. In the same line, a number of studies showed the benefit of a healthy and sportive life in order to support the interest of telomere measurement and the protective effect of long telomeres in healthy living however the causative relationships remain to be deciphered in depth.
If one can criticize the statistical tests power, we now know the extreme variability of telomere length among individuals. Such studies validated telomere length as a good biomarker for healthy aging but interpretation must be taken with caution given the number of additional factors to be considered.
Topics on aging never get old and telomere biology is no exception. Telomere epidemiology studies already revealed that we are not “telomerically” equal and discoveries on telomere length regulation, heritability, dynamics and association with aging, raised interest on the topic to a general audience, from journalists to insurance companies. This peculiar part of our DNA has been at the center of interest for the past 80 years. In this context, it falls under the responsibility of the scientific community to convey a balanced message regarding the implication of telomere homeostasis in diseases and wellness, considering that numerous additional factors must be considered. Indeed, in human, the clear-cut influence of telomeres, beyond the well-described effect of critical shortening, remains unclear and record on telomere shortening rate at specific telomere-ends are missing, leaving the former observation to potential biased conclusions. Moreover, the recent identification of long non-coding RNA (TERRAs) produced by telomeres and the long-distance impacts of telomeres on gene regulation represents a new challenge. Thanks to the rise of high-resolution methods (TeSLA, STELA, HiC, molecular combing) and accessibility to affordable deep sequencing platforms (NGS, nanopore technology, PacBio), future research will help deciphering subtelomere complexity, likely uncover novel pathways linked to telomere biology and opens new research opportunities to understand the genome wide impact of telomere in physiological and pathological situations or adaptation to environmental cues. In this context, if all agree that telomere matters, questions on their influence, weakness and strength linger.