Patterns of telomere length with age in African mole-rats: New insights from quantitative fluorescence in situ hybridisation (qFISH)

Introduction

In most eukaryotes, telomere length reduction with age is a direct outcome of incomplete replication of chromosome ends during replication; the ‘end replication problem’ (Hayflick & Moorhead, 1961; Lansdorp et al., 1996; Zakian, 2012). Once telomeres reach a critically short length, cellular signals initiate the process of cell cycle arrest and subsequent senescence, known as the ‘replicative theory of ageing/senescence’. Studies of telomere length dynamics in humans and birds have supported this theory, finding a strong correlation between telomere length and lifespan, with a decrease in telomere length with age (Vleck, Haussmann & Vleck, 2003; Harley, Futcher & Greider, 1990; Harley, 1995; Haussmann & Vleck, 2002; Haussmann, Vleck & Nisbet, 2003; Heidinger et al., 2012). Telomere-dependent replicative senescence is also seen as an anti-cancer mechanism in humans, as progressive telomere shortening ultimately leads to permanent arrest of cell proliferation. While in some species there is good evidence for a link between telomere length and life history, it is less clear whether there is a direct causal role between them.

It is thought that long-lived species have evolved a suppression of telomerase activity to mediate replicative senescence. It is argued that there has been selection in larger animals (which tend to live longer) to have evolved suppressed telomerase activity to alleviate an increased cancer risk arising from their large number of cells and increased lifespan, which provides increased opportunities for cancer-inducing mutations to occur (‘Peto’s paradox’). This hypothesis has been suggested to apply generally in mammals (Gomes et al., 2011), including rodent species (Seluanov et al., 2007), where telomerase activity has been shown to correlate with body mass.

Naked mole-rats (NMRs) may be an anomaly to Peto’s hypothesis—they are a small rodent with extreme longevity (potentially living up to 37 years), but with high levels of telomerase activity (Tian et al., 2018). Alternative telomere-independent mechanisms appear to have evolved in NMR to suppress cancer-inducing mutations (Gorbunova et al., 2014; Tian, Seluanov & Gorbunova, 2017) and to resist the normal signs of senescence (Buffenstein, 2005, 2008) over their lifespan. This article examines NMRs and two other species of African mole-rats in the family Bathyergidae with different longevities to compare age-related dynamics in telomere length: Fukomys damarensis (or Damaraland mole-rat, DMR) and Cryptomys hottentotus mahali (or Mahali mole-rat, MMR) with estimated lifespans of 11 and 15 years respectively (Weigl, 2005). We apply the technique of quantitative peptide nucleic acid-FISH (PNA-QFISH) to (i) identify terminal and/or non-terminal telomeric repeats in the NMR, DMR and MMR, and (ii) quantify changes in telomere size with animal age.

Materials and Methods

Results

Terminal telomeric repeats were successfully detected and visualised in the DMR, MMR and NMR (Fig. 1). Two homologous chromosome pairs in the DMR clearly share interstitial telomeric repeat sequences (ITSs) along their chromosome arms, illustrated by sections of diffuse qFISH staining. Additionally, a single homologous chromosome pair appears to have a shorter tract of these repeat sequences (fainter staining area seen in some cells; Fig. 1B). The NMR potentially had ITSs on one chromosome pair (Fig. 1D), although the region of increased fluorescence was faint, but noticable. In contrast, MMR chromosomes appeared to display a strictly terminal telomeric morphology, with no apparent ITSs observed on any chromosomes (Fig. 1C).

The karyotype for the DMR samples from our collection site (2n = 78) was consistent with previous studies (Nevo et al., 1986). We confirmed that the MMR had the same chromosome number as other mole-rats in the Cryptomys hottentotus clade (2n = 54 for females, and 53 for males; Fig. S1) including Cryptomys h. hottentotus, Cryptomys h. natalensis and Cryptomys h. pretoriae (Nevo et al., 1986; Faulkes et al., 2004; Deuve et al., 2008).

In DMRs, although there was a significant difference in total fluorescence among individuals within age classes (F_24,1371 = 27.05, p < 0.001), overall there was also a significant reduction in total telomere fluorescence in the bone marrow cells of old DMRs compared with young animals (F_2,1371 = 10.14, p < 0.001; Fig. 2A). For the MMR, significant variation in telomere fluorescence of bone marrow cells was found between individual MMRs within each group (F_24,1371 = 27.05, p < 0.001), and among groups of young, middle-aged and old individuals (F_2,1371 = 10.14, p < 0.001; Fig. 2B). A post-hoc Tukey test revealed a small, but significant increase in telomere fluorescence in old vs middle-aged and young MMRs (p < 0.001 and P < 0.05 respectively; Fig. 2B). Finally, as with the other two species, NMRs had significant differences in telomere fluorescence within age classes (F_6,441 = 12.01, p < 0.001), and between classes (F_2,441 = 23.90, p < 0.001). Total telomere fluorescence did not decline with age—old vs young animals were not significantly different (p = 0.21), while middle aged animals had a significant increase in fluorescence compared to young and old age classes (p < 0.001). Individual animal sample means with standard errors plotted against age in years are shown on Fig. S2.

Discussion

The use of qFISH for calculating telomere length is well established as a research tool (e.g. in mouse strains: Zijlmans et al., 1997), comparable with telomere length measurement method using Southern blotting/in situ hybridisation techniques and in some applications may be more precise than qPCR methods (Hultdin et al., 1998; Gutierrez-Rodrigues et al., 2014). Apart from mammals, qFISH has also been employed as a single technique estimator of telomere length in diverse vertebrate taxa such as fish (Panasiak, Dobosz & Ocalewicz, 2020) and reptiles (Olsson et al., 2020). PNA-qFISH clearly identified terminal telomeric repeats in the DMR, MMR and NMR and we report for the first time ITSs in the DMR and tentatively in the NMR (Figs. 1A and 1D). Chromosomes containing ITSs as observed in the DMR and potentially the NMR may indicate active karyotype evolution, and potentially mark regions of fusion of ancestral chromosomes or sites of double-strand break repair in unstable areas of the genome (Meyne et al., 1990; Azzalin, Nergadze & Giulotto, 2001; Ruiz-Herrera et al., 2008). Such molecular instability can lead to further breakage or fission and chromosome fusions. It is well known that the DMR karyotype is among one of the most rearranged karyotypes amongst members of the family Bathyergidae (Deuve et al., 2008). Variation in chromosome number to produce chromosomal races within the DMR has resulted in 2n varying from 74 to 78 (Nevo et al., 1986) and also 2n = 80 individuals can occur in different geographic areas (Deuve et al., 2008). This chromosomal variation, coupled with the observation of ITSs, suggest that the DMR genome has ongoing, active karyotype evolution, in contrast to the fixed karyotypes seen in the MMR and others in the Cryptomys hottentotus clade (Deuve et al., 2008). Future work could involve using G and C-banding techniques and chromosome painting using whole chromosome paints to observe association between heterochromatic regions and ITS and terminal telomeric repeats (Yang et al., 2004). This may allow more detailed insights into the various types of past chromosomal rearrangements that have occurred in the DMR and other species in the genus Fukomys, where there is extensive karyotypic variation among species (Van Daele et al., 2004).

The presence of double telomere signals at the end of some chromatids in the MMR (Fig. 1C) is a feature that has been reported previously in other mammals and may arise through telomere—nuclear envelope interactions. Such interactions are proposed to protect against recombination of telomeric repeats during mitosis and are considered essential for correct meiotic pairing of homologous chromosomes during meiosis (Schober et al., 2009; Chikashige et al., 2006; Ding et al., 2007).

In all three species we found significant intra-specific differences in total telomere fluorescence among individuals. Variance in telomere size among individuals has been widely reported in other species, and some studies have shown associations between early-life telomere length and individual life history for example European badgers Meles meles (Van Lieshout et al., 2019), and meerkats (Cram et al., 2017). The significance of individual variation in telomere sizes in mole-rats identified in our study cannot be addressed further without additional longitudinal studies using PCR-based techniques to quantify telomere size. We also found interesting species differences in the variation in telomere fluorescence with age, although a limitation of the study was being able to accurately age the wild caught DMRs and MMRs. Our results revealed a significant reduction in telomere fluorescence and by implication a reduction in telomere size with age class when comparing young and old DMRs (Fig. 2A), suggesting that telomere shortening is a cause or biomarker for cellular senescence in this species. In contrast, both MMRs and NMRs did not show an age-related decline in total telomere fluorescence (Figs. 2B and 2C). Moreover, in both species we observed an increase in telomere fluorescence with age compared to the young age category, although our sample size was small for the NMR. These observations do not support the hypothesis of cellular senescence in these species and support other indirect evidence for telomere maintenance with age published for the NMR. Seluanov et al. (2007) found that telomerase activity in NMRs was high, and comparable with other small (but short-lived) rodents like mice and gerbils that have a high incidence of cancer. Furthermore Kim et al. (2011) showed stable levels of TERT expression (that functions to maintain telomeres) regardless of age (comparing a new-born, 4 and 20 year old NMRs). Recently, Adwan Shekhidem et al. (2019) used a different (qPCR-based) approach to show that telomere length in the NMR did not shorten with age, instead showing a mild elongation. The telomeric fluorescence signal in the NMR here suggest elevated activity of telomere elongating processes in middle aged animals. Conversely, Adwan Shekhidem et al. (2019) found that telomere length in the cancer resistant, long-lived blind mole-rat Spalax declined with age, in a similar fashion to short-lived rodents, similar to our observations here for DMR. Collectively, these results highlight some fascinating variation in telomere dynamics within the African mole-rats, and among other rodents, revealing the potential for evolutionary gains and losses of mechanisms associated with telomere maintenance, cancer resistance, longevity and senescence.

Supplemental Information