P66SHC deletion improves fertility and progeric phenotype of late‐generation TERC‐deficient mice but not their short lifespan

Summary Oxidative stress and telomere attrition are considered the driving factors of aging. As oxidative damage to telomeric DNA favors the erosion of chromosome ends and, in turn, telomere shortening increases the sensitivity to pro‐oxidants, these two factors may trigger a detrimental vicious cycle. To check whether limiting oxidative stress slows down telomere shortening and related progeria, we have investigated the effect of p66SHC deletion, which has been shown to reduce oxidative stress and mitochondrial apoptosis, on late‐generation TERC (telomerase RNA component)‐deficient mice having short telomeres and reduced lifespan. Double mutant (TERC −/− p66SHC −/−) mice were generated, and their telomere length, fertility, and lifespan investigated in different generations. Results revealed that p66SHC deletion partially rescues sterility and weight loss, as well as organ atrophy, of TERC‐deficient mice, but not their short lifespan and telomere erosion. Therefore, our data suggest that p66SHC‐mediated oxidative stress and telomere shortening synergize in some tissues (including testes) to accelerate aging; however, early mortality of late‐generation mice seems to be independent of any link between p66SHC‐mediated oxidative stress and telomere attrition.


Introduction
Telomeres are conserved repetitive DNA sequences at the ends of chromatids that protect from chromosomal rearrangements and loss of genetic information (Blackburn et al., 2006). Shortening of telomeres occurs in somatic cells during aging (Daniali et al., 2013) due to DNA damage and incomplete end processing (von Zglinicki et al., 2000). Above a threshold, telomere erosion results in the arrest of cellular proliferation and the dysfunction of renewable tissues. For this reason, telomerase activity and recombination-based processes that maintain telomere length are critical for tissue homeostasis (Sahin & Depinho, 2010). In fact, mutations in either the gene encoding the telomerase catalytic subunit (telomerase reverse transcriptase, TERT), or the telomerase RNA gene (TERC) are found in patients with dyskeratosis congenita who have short telomeres and show accelerated aging and reduced lifespan (Kirwan & Dokal, 2009). Consistently, late-generation (G3 and beyond) TERT (Rudolph et al., 1999) or TERC-deficient mice (TERT À/À and TERC À/À , respectively) are short living and show reduced fertility, early alopecia, kyphosis, anemia, and lymphopenia (Wong et al., 2003). Interestingly, TERC À/À mice present altered mitochondrial functions, including increased production of reactive oxygen species (ROS) from the electron transfer chain (ETC) (Passos et al., 2010;Sahin et al., 2011). Thereby, telomeres have been suggested to play a role in controlling mitochondrial ROS accumulation and oxidative stress during aging (Sahin & Depinho, 2010).
As telomere DNA is particularly sensitive to oxidative damage (Oikawa & Kawanishi, 1999) and single-strand breaks or other stress-induced lesions at telomeres are inefficiently repaired (Petersen et al., 1998), oxidative stress is considered to accelerate telomere erosion (von Zglinicki, 2002;Houben et al., 2008). Consistently, mice expressing a variant of the mitochondrial uncoupling protein 2 that increases ROS production by ETC, and therefore, oxidative stress shows short telomeres (Salpea et al., 2010). Moreover, it has been shown that treating mouse embryos with chemicals able to induce mitochondrial dysfunction and ROS accumulation, led to telomere loss and chromosomal instability, which was prevented by a concomitant treatment with N-acetylcysteine, a compound that improves ROS scavenging (Liu et al., 2002). However, other mutations in genes involved in ROS metabolism, such as SOD2 haploinsufficiency, do not cooperate with telomere dysfunction (Guachalla et al., 2009), and the treatment with N-acetylcysteine does not rescue progeric phenotypes of late-generation TERT-deficient mice (Sahin et al., 2011).
To investigate the link between endogenously generated oxidative stress, telomeres, and aging, we have investigated the effect of p66SHC deletion on late-generation TERC-deficient mice, by generating and studying double mutants TERC À/À P66SHC À/À mice.
These data clearly suggest that p66SHC deletion improves fertility of late-generation TERC À/À mice.
Notably, the frequency of apoptotic cells as evaluated by IHC analysis with anti-activated caspase-3 antibody appeared significantly reduced in the testis from p66SHC À/À background (Fig. S2). Aplastic anemia has been also associated to telomere shortening in both humans and mouse models (Siegl-Cachedenier et al., 2007); however, blot tests and bone marrow examination did not evidence severe aplastic anemia either in TERC À/À p66SHC +/+ or TERC À/À p66SHC À/À mice (Fig. 5B). No clear signs of muscle degeneration but rare loss of the striations, vacuolization, and infiltrations in the G3 TERC À/À p66SHC +/+ , as shown in the Fig. S3, were observed. Abnormal glycaemia was not revealed in all generations as well. Lung emphysema was observed in late-generation TERC À/À regardless the p66SHC genotype, whereas lung fibrosis was markedly evident in p66SHC À/À mice (Fig. 5C).
P66SHC does not affect the lifespan of late-generation TERC À/ À mice G3 C57Bl/6 TERC À/À mice were shown to be short living (Rudolph et al., 1999). The comparison of survival curves of WT and G3 TERC À/À p66SHC +/ + mice confirmed this observation (Fig. 6A). In particular, mortality risk ( Fig. 6A, right upper panel) was comparable for the first 2 years, whereas starting from the third year it increases much more for TERC À/À mice than for WT; G3 TERC À/À p66SHC +/+ males showed the shortest survival. The survival rate of G3 TERC À/À p66SHC À/À was identical to that of G3 TERC À/À p66SHC +/+ mice (Fig. 6B). They showed the same reduction in survival compared to the WT and p66SHC À/À (TERC +/+ ) mice. We could also determine the lifespan of G6 TERC À/À p66SHC À/À mice, revealing a further shortening of lifespan with respect to the G3 (Fig. 6C). Necroscopic examination of spontaneously dead G3 and G6 mice revealed the presence of visible tumor masses only in few cases (12 of 100 mice), equally distributed with respect to p66SHC mutation, rare enlargement of bladder and fecal prolapse, no splenomegaly or enlarged lymph nodes, and no hemorrhagic tissues or ascites.
In particular, the average body weight of G6 TERC À/À p66SHC À/À mice was 18.5 AE 0.4 g at 3 months of age and 23.1 AE 1.8 g at 1 year of age. They showed tremor and lethargy because 6-8 months of age and frequently hair loss around the nose and the limb, no abnormalities in the blood formula but a mild anemia, normal glycaemia and lipidic profile, and small lungs and thymus at death.
Oxidative stress shortens telomeres (von Zglinicki, 2002), and the deletion of p66SHC reduces oxidative stress (Trinei et al., 2002 and Fig. 3A and B). However, in the double mutants (TERC À/À p66SHC À/À ) mice, we do not observe differences of telomere length with respect to the TERC À/À p66SHC +/+ (Fig. 3 C and D). This result indicates that the effect of reducing oxidative damage, by deleting p66SHC, on telomere shortening is less important with respect to other reasons of telomere erosion such as the incomplete replication of chromosomal ends.
The function of p66SHC appears instead relevant for the downstream consequences of dysfunctional chromosome ends, as p66SHC deletion improved fertility and reduce weight and shrinkage of different organs including testis of late-generation TERC À/À mice (Fig. 4). Short telomeres of late-generation TERC À/À mice induce senescence and apoptosis (Wong et al., 2003) that determine the detrimental effects of telomere erosion ultimately. P66SHC triggers mitochondrial apoptosis upon a variety of stresses (Migliaccio et al., 2006). At mechanistic level, p66SHC is plausible to mediate cell loss following telomere erosion rather than having a role in telomere shortening.
Regardless of the recovered fertility, weight, and healthier appearance, the p66SHC deletion does not improve longevity of G3 TERC À/À mice Oxidative stress and telomere length in TERC À/À p66SHC +/+ and TERC À/À p66SHC À/À mice. (A) Levels of 8-OH-dG per lg of DNA in TERC À/À p66SHC +/+ (white bars) and TERC À/À p66SHC À/À (black bars) spleen (SP), liver (LI), lung (LU), and testis (TE) measured by competitive enzyme immunoassay. Average values AESD from n = 5 individuals for the G0 group and n = 3 individuals for the G3 and G5 groups are reported: * for P < 0.05 and ** for P < 0.01. (B) 8-isoprostane IHC representative images of testis, lung, and liver from G0 and G3 TERC À/À p66SHC +/+ and TERC À/À p66SHC À/À mice. Level of background signal is shown in the first three unstained slices. (C and D) Telomere length (expressed as a percentage versus the telomere length of C57Bl/6J WT mice) in spleen (SP), liver (LI), lung (LU), heart (HE), testis (TE) and skin (SK) from TERC À/À p66SHC +/+ and TERC À/À p66SHC À/À mice of different generation as evaluated by qPCR (C) where the average values AESD from n = 3 individuals per group, measured in quadruplicates, are reported or Q-FISH (D) where the average values AESD from n = 3 individuals per group, each sample evaluated from two different sections for a total of around 40 nuclei, are reported. * for P < 0.05, ** for P < 0.01. (E) Representative FISH images (Cy3 alone grayscale and DAPI-Cy3 merged color) obtained from the lung of different generation TERC À/À p66SHC +/+ and TERC À/À p66SHC À/À mice. (Fig. 6B). Indeed, lung emphysema and fibrosis are frequent in both G3 and later generations of TERC À/À independently by the p66SHC deletion, although we could not establish that lung dysfunction is the cause of death of these mice. Histiocytic sarcoma was reported to be the most frequent tumor lesion in C57Bl/6 mice, including p66SHC À/À (Ramsey et al., 2014) and TERC À/À (Khoo et al., 2007) models. In any case, the early mortality of TERC À/À mice is not determined by p66SHC suggesting that oxidative stress and telomere attrition do not cooperate to determine lifespan. In this view, the unification of mitochondrial (free radicals) and nuclear (telomere) theories of aging (Sahin & DePinho, 2012) appears uncertain since the peculiar reasons of death of laboratory C57BL/6 mice. From an evolutionary point of view, short telomeres and repressed telomerase (as in human somatic tissues) have been suggested to coevolve with homeothermy and replicative aging in mammals (Gomes et al., 2011), while p66SHC with the metabolic adaptation to harsh energetic environments (i.e., food deprivation and cold temperature) . In this view, the relative healthier aging of TERC À/ À p66SHC À/À compared to TERC À/À p66SHC +/+ mice suggests that the cost to pay for having a plastic metabolism (as ensured by p66SHC) is to amplify the negative effect of telomere erosion on body size and fertility.
Finally, results from the study of p66SHC/TERC support the notion that genetic factors control the response to telomere erosion of specific organs. Thus, the predictive value of telomere length on organ aging depends on genetic background. Then, p66SHC expression and activity is induced in several tissues by obesogenic diets  or hyperglycemia (Albiero et al., 2014) that are known to affect telomere length as well (Laimer et al., 2015). The interaction of aging and metabolic pathways, including p66SHC, SIRT1, or AMPK/mTOR/S6K (Fadini et al., 2011), with telomere erosion in the presence of diabetes or obesity is a promising field to reveal how metabolic disorders impact on aging.
Experiments have been carried out in accordance with institutional guidelines and the Italian law, which enforces EU directives regarding the protection of animals used for experimental and other scientific purpose.
for 5 min and incubated with 200 nM solution of Cy3 conjugated PNA TelC probe (PNA Bio) in 60% formamide, 10 mM Tris-HCl pH 7.5, and 0.2 lg mL À1 salmon sperm first for 10 min at 85 o C and then for 2 h at RT. Then, the probe was washes twice with 2XSSC, 0.1% Tween-20 solution for 10 min at 58 o C. DAPI staining was used for the staining of the nucleus. Images were collected through an Olympus BX51 upright fluorescent microscope equipped with a cool snap EZ CCD camera (Photometrics Tucson, AZ, USA) and processed by Metamorph (Molecular Device Sunnyvale, CA, USA) and ImageJ software. The Cy3 fluorescence intensity of at least 40 nuclei observed in at least two different slices per experimental group was compared (Meeker et al., 2002).

8-OH-dG and 8-isoprostane analysis
8-OH-dG was measured using the commercially available immunoassay kit (Abcam Cambridge, UK) on genomic DNA. One microgram of total genomic DNA was denatured 5 min at 95°C and digested with nuclease P1 (Sigma-Aldrich Srl), then dephosphorylated by incubation with Antarctic phosphatase (New England Biolabs; EuroClone S.P.A. Pero, Italy) and loaded in duplicate on the anti-8-OH-dG antibody-coated plate. The amount of 8-OH-dG in the samples was quantified by measuring competition with labeled 8-OH-dG for antibody binding.

Tissue analysis
P66SHC mediates telomere loss-made sterility, M. Giorgio et al. was performed on fresh tissues that were rapidly washed in PBS and incubated in 4% formaldehyde-buffered solution for 16 h at room temperature. Fixed samples were then dehydrated by increasing concentrations of ethanol and included in paraffin using the Leica ASP300 tissue processor (Leica Biosystems -Leica Microsystems Srl Milan, Italy). Then, 4 lm sections were cut from paraffin blocks and stained with hematoxylin-eosin. Anti-active caspase 3 antibody (Abcam) was used to reveal apoptotic cells by IHC.

Statistical analysis
Data were analyzed using Student's t-test and Fisher's exact test. Survival was estimated with Kaplan-Meier estimators and differences tested with the log rank test.

Supporting Information
Additional Supporting Information may be found in the online version of this article at the publisher's web-site.

Fig. S1
Representative FISH images of sections of different organs from G0, G3 and G5 TERC À/À P66SHC +/+ and TERC À/À p66SHC À/À mice, stained with the Cy3 conjugated TelC PNA probe (grayscale panels), the colored panels show Cy3 and DAPI fluorescence pattern in a stained section of WT liver as control.