Senescence, regulators of alternative splicing and effects of trametinib treatment in progeroid syndromes

Progeroid syndromes such as Hutchinson Gilford Progeroid syndrome (HGPS), Werner syndrome (WS) and Cockayne syndrome (CS), result in severely reduced lifespans and premature ageing. Normal senescent cells show splicing factor dysregulation, which has not yet been investigated in syndromic senescent cells. We sought to investigate the senescence characteristics and splicing factor expression profiles of progeroid dermal fibroblasts. Natural cellular senescence can be reversed by application of the senomorphic drug, trametinib, so we also investigated its ability to reverse senescence characteristics in syndromic cells. We found that progeroid cultures had a higher senescence burden, but did not always have differences in levels of proliferation, DNA damage repair and apoptosis. Splicing factor gene expression appeared dysregulated across the three syndromes. 10 µM trametinib reduced senescent cell load and affected other aspects of the senescence phenotype (including splicing factor expression) in HGPS and Cockayne syndromes. Werner syndrome cells did not demonstrate changes in in senescence following treatment. Splicing factor dysregulation in progeroid cells provides further evidence to support this mechanism as a hallmark of cellular ageing and highlights the use of progeroid syndrome cells in the research of ageing and age-related disease. This study suggests that senomorphic drugs such as trametinib could be a useful adjunct to therapy for progeroid diseases.


Introduction
The human ageing process is complex and occurs over decades, but insight can be gained by studying the rare monogenic conditions that result in vastly accelerated rates of ageing [1].These conditions often result from errors in the gene regulatory or DNA damage response machinery and are collectively termed progeroid syndromes.Despite often being extremely rare, the syndromes are life-changing and can be fatal [2,3].Some of the most studied progeroid syndromes are Hutchinson-Gilford Progeria syndrome (HGPS), Werner syndrome (WS) and Cockayne syndrome (CS).These are rare disorders affecting an estimated 1 in 18 million live births for HGPS and 1 in 200,000 for both WS and CS [2,[4][5][6].
HPGS is caused by mutations in the Lamin A (LMNA) gene that result in a truncated Lamin A protein termed progerin [7].The genetics of WS is also well defined, with the syndrome caused by mutations in the Werner (WRN) gene which encodes the DNA helicase, Werner protein [5].CS is more genetically Abstract Progeroid syndromes such as Hutchinson Gilford Progeroid syndrome (HGPS), Werner syndrome (WS) and Cockayne syndrome (CS), result in severely reduced lifespans and premature ageing.Normal senescent cells show splicing factor dysregulation, which has not yet been investigated in syndromic senescent cells.We sought to investigate the senescence characteristics and splicing factor expression profiles of progeroid dermal fibroblasts.Natural cellular senescence can be reversed by application of the senomorphic drug, trametinib, so we also investigated its ability to reverse senescence characteristics in syndromic cells.We found that progeroid cultures had a higher senescence burden, but did not always have differences in levels of proliferation, DNA damage repair and apoptosis.Splicing factor gene expression appeared dysregulated across the three syndromes.10 µM trametinib reduced senescent cell load and affected other aspects of the senescence phenotype (including splicing factor expression) in HGPS and Cockayne syndromes.Werner syndrome cells did not demonstrate changes in in senescence following treatment.Splicing factor dysregulation in progeroid cells provides further evidence to support this mechanism as a hallmark of cellular ageing and highlights the use of progeroid Vol:.( 1234567890) and phenotypically diverse, with several subtypes across a spectrum of distinguishing phenotypic features; CS type I (moderate disease/classical phenotype), type II (severe or early-onset), type III (mild), photosensitivity only (adult onset) and cerebraloculofacioskeletal syndrome (COFS; severe, foetal onset) [6,8].No correlation has yet been found linking the phenotypic subtypes with two known genetic mechanisms of Cockayne syndrome: CS type A and B. The two genetic causes correspond to mutations in excision repair genes ERCC8 (Cockayne syndrome A) or ERCC6 (Cockayne syndrome B) [6,8].
All of the three syndromes share common features of accelerated ageing and significantly higher risks of age-related diseases such as cardiovascular disease, cancer, diabetes and chronic kidney disease [3,9,10].However, the severity of phenotypes between HGPS, WS and CS demonstrate differences.The defects in LMNA, WRN and ERCC6/8 seen in HGPS, WS and CS, respectively, all cause major instability of the nuclear envelope and large scale unrepaired genomic damage, resulting in major curtailment of average lifespan to 14.5 years for HGPS and 54 years for WS [3,5,10,11].The severity of CS affects the observed mean lifespan across the different phenotypic subtypes.People with CS type I have a mean lifespan of 16.1 years, those with CS type II have a mean lifespan of 5 years, and those with type III have a mean lifespan of 30.3 years [8].In this study, we examine the mildest subtype, CS type III, which enables a contrast with two more severe syndromes: HGPS and WS.The CS type III subtype is generally milder with patients given an expected lifespan of 10-30 years and milder symptoms of accelerated ageing, although they remain abnormally sensitive to UV-induced DNA damage [6,8].Despite their monogenic cause, the molecular features of ageing observed in progeroid syndromes are similar to those seen during normal ageing.Assessment of potential therapeutics to slow rates of ageing (senotherapeutics) in progeroid syndromes may indicate interventions with efficacy against common, chronic ageing diseases in the wider population.
There are several theories as to how and why we age.Evidence is mounting that the physiological and functional changes that occur during the ageing process arise from the gradual failure of a series of basic health maintenance mechanisms.These mechanisms are termed the hallmarks of ageing, which are also major drivers of age-related disease [12,13].Hallmarks include genomic instability, epigenetic alterations, mitochondrial dysfunction, altered intercellular communication, deregulated proteostasis, deregulated nutrient sensing, telomere attrition, stem cell exhaustion, dysregulated regulation of alternative splicing, compromised autophagy, altered mechanical properties, disturbances to the microbiome, inflammation and cellular senescence.They are present in normal ageing in multiple species, and underpin many of the common chronic diseases of human ageing [12].Importantly, the hallmarks of ageing represent promising avenues for therapeutic targeting of the diseases of ageing.This is exemplified by observations that targeted depletion of senescent cells leads to increases in multiple healthspan and lifespan-related parameters in animal ageing/progeria models [14][15][16].In smaller studies, targeted reduction of senescence can affect the phenotype of idiopathic pulmonary fibrosis and diabetic kidney disease in humans [17,18].
More recently, dysregulation of the expression of splicing regulatory factors has been proposed as a new hallmark of ageing, since it is known to be associated with normal human ageing [13,[19][20][21].mRNA splicing is a carefully controlled mechanism by which our genes can produce many different mRNA transcripts [22].It is controlled by a portfolio of splicing factors which are themselves regulated by alternative splicing; splicing factors are proteins which act on the gene to influence a change in the splice site of a pre-mRNA transcript.In line with its recent designation as a novel hallmark, dysregulation of splicing factor expression occurs during normal ageing in multiple species, its experimental aggravation induces cellular senescence and ageing phenotypes, and its experimental amelioration alters aspects of cellular and organismal ageing in human cells and in other species [20,[23][24][25][26][27][28][29][30][31].Dysregulated splicing factor expression arises from unresolved and constitutive signalling through ERK and AKT pathways, culminating in altered activity or expression of the FOXO1 and ETV6 genes [32].The fruit fly homologues of these genes (Foxo and Aop) have also previously been demonstrated to be the genetic effectors of RAS/MEK/ERK and PI3K/ AKT signalling in relation to lifespan in Drosophila melanogaster [33].
New research into senotherapeutic compounds that modulate senescence-related pathways may Vol.: (0123456789) be an important avenue for future therapies for progeroid diseases.Compounds with senomorphic (reversal of senescence) or senolytic (lysis of senescent cells) properties may have effects on the premature ageing phenotype seen in the progeroid syndromes.Drugs with senomorphic properties like resveratrol and rapamycin affect several senescence-associated pathways, but can have pleiotropic effects [34].Some compounds that inhibit p38/MAPK have been suggested to aid in the treatment of WS, however the potential therapeutic effects of inhibiting other senescence pathways have not yet been investigated for WS [35][36][37].Only lonafarnib, a farnesyltransferase inhibitor, is approved for HGPS, but several senotherapeutic compounds like rapamycin have been identified as having potential in in vitro research [38][39][40][41][42].The action of the farnesyltransferase is able to improve the persistent farnesylation of the aberrant Lamin A protein caused by HGPS [38].
Senotherapeutic compounds can be used to target specific aspects of the senescence phenotype.For example, trametinib is a drug which specifically inhibits both isoforms of MEK (MEK1 and MEK2) and it has been approved by the FDA for the treatment of metastatic melanoma [43][44][45].The effects of trametinib on splicing factor expression have been well characterised in previous work, so we know that with the application of low dose (1-10 µM) trametinib to senescent primary human dermal fibroblasts, we are able to restore splicing factor expression and bring about a reversal of several aspects of the senescent cell phenotype [32].Drugs like trametinib may help reduce some of the senescent phenotype of progeroid diseases.
Here, we aimed firstly to determine whether disrupted splicing factor expression is a feature of the accelerated ageing phenotypes seen in progeroid syndromes, as it is for normal ageing [23].Secondly, should splicing factor profiles be disrupted in progeroid cells, we aimed to determine whether trametinib was capable of restoring correct expression of splicing factors and attenuating senescence phenotypes in progeroid cells, as it does in wildtype cells [32].This study could help elucidate novel mechanisms of senescence and identify a future point of therapeutic intervention for progeroid diseases.

Human primary cells
All cells used in this study were commercially derived, with ethical permission granted at source.Normal human dermal fibroblasts (nHDF) were purchased from Promocell, Heidelberg (catalogue number C-12302, lot number 445Z026.3).The donor was male, Caucasian and 36 years old at donation.Cells were taken from the abdomen.At the time of these experiments, nHDF cells had cumulated population doublings (cPDL) of 29.44.All three progeroid syndrome cell lines were human primary dermal fibroblasts purchased from the Coriell Institute (Camden, New Jersey, United States).Cells from an HGPS donor (catalogue number AG06917) were from a 3-year-old Caucasian male and were taken from the patient's arm.The cells have a normal karyotype (46, XY), but have a de novo single point mutation (2036 C > T) in the Lamin A (LMNA) gene.The patient displayed reduced subcutaneous tissue, thin skin, a thin beak-like nose characteristic of HGPS, thin underdeveloped nails, narrow clavicles, and growth retardation.HGPS cells had a cPDL of 6.36 for the characterisation of untreated cells, and 36.30at the time of the trametinib treatment experiments.
Cells from a CS donor (catalogue number AG07076) were donated by an 11-year-old Caucasian female with CS-type A/type III.The patient had the least severe phenotypic type of CS: CS type III.They had a phenotype of dwarfism, mental retardation, cataracts, photophobia, retinopathy and optic atrophy.Being CS type A, the cells have a mutation in the ERCC8 gene.This donor had a normal karyotype (46XX) but demonstrated compound heterozygosity for two mutations: a 649G-C transversion, and a G-to-T transversion.These mutations result in an ala205-topro (A205P) substitution, and a glu13-to-ter (E13X) substitution, respectively [47,48].CS cells had a cPDL of 11.32 for the characterisation of untreated cells, and 18.63 at the time of the trametinib treatment.

Tissue culture
All cells were grown in animal component-free conditions.Cells were cultured in DMEM 1 g/l glucose + phenol red (31885023, Gibco™), 10% human serum (H3667, Sigma Aldrich) and 1% 10,000 U/ml penicillin-10,000 µg/ml streptomycin (15140122, Gibco™).We used TryPLE™ Express (12604013, Gibco™) to detach cells.A Hirschmann haemocytometer was used to perform cell counts, which together with the cPDL numbers given by Promocell and the Coriell Institute, enabled assessment of cPDL at the time of seeding for experiments.Cells were transferred to antibiotic-free media for 48 to 72 h before seeding.Cells were seeded at approximately 30,000 cells/well in 12-well plates for staining experiments.For harvesting RNA for RT-qPCR analysis, cells were seeded at approximately 50,000 cells/well in 6-well plates.Dosing for trametinib studies was taken from our previous work, where a 10 µM dose suspended in 10% DMSO (HY-10999, Medchem-Express; J66650.AD, Thermo Scientific Alfa Aesar) resulted in attenuation of splicing factor expression and rescue of aspects of the senescent cell phenotype in wild-type human primary dermal fibroblasts [28].

Senescence-associated beta galactosidase (SAB) experiments and analysis
We used the Senescence Cells Histochemical Staining kit (CS0030, Merck) as per the manufacturer's instructions to stain senescence-associated beta galactosidase (SAB).24 h after staining, cells were imaged using Zeiss AxioCam ERC55 PrimoVert at 10 × magnification.Five images per biological replicate were later counted manually using ImageJ 1.47v software (US National Institute of Health, Bethesda, Maryland, USA) [49].
Real-time quantitative PCR (RT-qPCR) RT-qPCR was performed using TaqMan™ Universal Mastermix II (4440048, Applied Biosys-tems™) and TaqMan™ Gene Expression Assays (FAM) (4331182, TaqMan®) in three biological and three technical replicates.The assay IDs are listed above in the pre-amplification section.RT-qPCR was performed on the Quantstudio 12 K platform (Applied Biosystems™) in 5 µl reactions using 384-well plates. 1 µl of diluted/ pre-amplified cDNA was used per reaction alongside 0.25 µl of Taqman™ Gene Expression Assay (which corresponds to 900 nM primer and 250 nM probe).Cycling conditions were: 50 °C for 2 min, 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min.Fluorescence intensity was captured at the end of each cycle.Ramp speeds were 1.6 °C per second for all transitions.Gene expression was calculated by the comparative C T technique, relative to the geometric mean (untreated data) or mean (vehicle/trametinib treated data) expression level of three endogenous housekeeping genes (GUSB, IDH3B and PPIA), which had been empirically demonstrated to provide the most stable baseline for comparison within each dataset using the online RefFinder website [50,51].For assessment of gene expression, transcript levels in progeroid cells were normalised to the mean expression of each gene in wild-type control cells.For the trametinib treatment experiments, the data was normalised to the vehicle-treated controls.

Statistics
The effect size or mean ± standard error of the mean (SEM) is reported in the text, with the full statistics reported in the tables.t test statistics for SAB and immunocytochemical staining were performed using Graphpad Prism version 9.

Senescence phenotypes in HGPS, WS and CS cells
Early passage fibroblasts from all three progeroid syndromes demonstrated higher levels of senescence than similar passage wild-type fibroblasts.SAB staining was 13.5-fold, 1.6-fold and twofold higher than nHDFs for HGPS, WS and CS, respectively (p < 0.0005, p = 0.0129 and p = 0.0401) as shown in Fig. 1A-C and Table 1.Cells from donors with HGPS and WS also demonstrated lower levels of proliferation (80% and 71% lower for HGPS and WS, respectively; p = 0.0050 and p = 0.0082; Fig. 1E and F), whereas cells from the phenotypically less severe CS patient demonstrated no differences in proliferation compared to controls (p = 0.048; Fig. 1G). Figure 1D, H and L show representative images of the staining methods used to measure senescence phenotypes (SAB, Ki67 and γH2AX staining respectively) with full staining data available in Table 1.We did not detect any differences in the number of γH2AX (DNA damage repair) foci in cells from progeroid donors compared with wildtype fibroblasts (Fig. 1I-K).The expression of several genes that control DNA damage repair were lower in the progeroid cells suggesting the progeroid cells have an impaired DNA damage response (Fig. 2A-C and Table 2).CHEK1 expression was lower in all three types (HGPS: 144% lower, p < 0.0005, WS: 115% lower, p = 0.001, CS: 51% lower, p = 0.007).RB1 expression was lower in WS and CS cells: 55% lower, p = 0.013, and 73% lower, p = 0.003, respectively.ATM expression was 37% lower in WS cells, p = 0.011.TP53 expression was not significantly different between wild-type nHDFs and any progeroid cell type.Levels of apoptosis were higher in all three progeroid cell types (Fig. 2A-C and Table 2), with significantly higher levels of relative CASP3 expression (HGPS: 28% higher, p = 0.025, WS: 35% higher, p = 0.023, CS: 27% higher, p = 0.046).Relative CASP7 expression was 31% higher in HGPS cells (p < 0.0005), but was not significantly different in WS and CS cells (WS: 17% higher, p = 0.211.CS: 2% lower, p = 0.662).Dysregulation of splicing factor expression in HGPS, WS and CS cells Overall, most splicing factors were downregulated in progeroid syndromes compared to the wild type.75% of the significant effects observed in HGPS were a decrease, with 43% and 89% of effects for WS and CS, respectively (Fig. 3A-C and Table 2).Some splicing factors demonstrated altered expression in one or more progeroid cell type; HNRNPM, HNRNPA1 and PNISR expression were altered in all three cell types.HNRNPM expression was reduced by 72%, 97% and 76% for HGPS, WS and CS cells respectively (p = 0.008, 0.016 and 0.007).HNRNPA1 expression was reduced by 52%, 43% and 56% lower for HGPS, WS and CS cells respectively (p = 0.001, 0.006 and 0.001) whilst PNISR expression was increased by 74%, 93% and 52% in HGPS, WS and CS cells respectively (p = 0.007, 0.001 and 0.008).
Trametinib had no effect on senescence phenotypes in Werner syndrome cells Werner cells did not show any changes in senescence, proliferation or DNA damage repair in response to trametinib treatment, however gene expression of CASP7 and SRSF6 were decreased (Tables 3 and 5 and Fig. 5A-E).Levels of senescence (SAB staining) did not show any change in response to the treatment (Fig. 5A).Werner cells had a similar response to HPGS in terms of proliferation (an 83% decrease), but this was not significant (p = 0.1089), and γH2AX staining was unchanged (Fig. 5B-C).RB1 gene expression was decreased by 42% with treatment (p = 0.004), but ATM, CHEK1 and TP53 expression remained unchanged (Fig. 5D).

Discussion
Progeroid syndromes are life-limiting rare diseases that share many similarities with the processes that are involved in normal ageing and agerelated diseases.Senotherapeutic compounds can be used to target ageing, age-related disease and premature ageing.With few treatments available for progeroid syndromes, finding new senotherapeutics is important.Some senotherapeutic compounds like trametinib are thought to work by restoring levels of the splicing regulatory factors that control alternative splicing decisions.We first investigated if splicing factor expression profiles were altered in dermal fibroblasts from donors with Hutchinson-Gilford Progeria syndrome, Werner syndrome and Cockayne syndrome, and secondly, if a known senomorphic drug, trametinib, could impact senescence kinetics and splicing factor expression in these cell populations.We observed that cultures of early passage primary dermal fibroblast cells from individuals with some progeroid syndromes demonstrated elevated senescent cell load and altered splicing factor expression compared to that observed in early passage wildtype cells.Furthermore, we saw that trametinib was able to influence the expression levels of some splicing factors and influence some, but not all, aspects of the senescent cell phenotype in these cells.
We noted elevated senescent cell load in the more severe syndromes (HGPS and WS) compared with cells from the more moderately affected CS patient, and disease severity and senescent cell load also correlated with degree of 'rescue' upon trametinib treatment.WS and HGPS cells demonstrated fewer changes to splicing factor expression and a more moderate effect on senescent cell load; upon treatment with trametinib CS cells demonstrated a 71% decrease in senescent cell load (though this effect may be exaggerated as the percentage of SAB-stained cells in the vehicle treatment was low) compared with 33% for HGPS and no change for WS (Table 3, Figs.4A, E, 5A, E, 6A and E).This may suggest that the most severe progerias may have an increased prevalence of terminally and irreversibly senescent cells.
In line with previous studies on senescence induced in wild type cells by replicative senescence, we also observed changes to the portfolio of splicing factors expressed.Some of these splicing regulatory factors have previously been associated with ageing phenotypes in humans.HNRNPA1 and HNRNPA2B1 have previously been reported to be associated with parental lifespan in human populations [25] and with senescence in human primary cells [24,52], whereas HNRNPM and AKAP17A have previously been reported to demonstrate predictive associations with cognitive decline and loss of muscle strength [53,54].Although we did not directly address alternative splicing patterns in our progeroid cultures, previous work in a mouse model of HGPS has documented that whilst the number of alternatively spliced genes is similar to that observed in wild-type mice in young animals, the number of alternatively spliced genes in the HPGS model mice was altered relative to those seen in wild-type mice as the animals age [55].It is noteworthy that in the two cell lines (HGPS and CS) where we were able to demonstrate attenuation of some senescence phenotypes following trametinib treatment; three and six splicing factors were altered in response to treatment, compared with the situation for WS, where we detected changes in expression for a single splicing factor only.
Another interesting and surprising observation was the overall low level of γH2AX staining in our progeroid cell cultures.This is counterintuitive given that these are all syndromes of DNA damage, and elevated damage has previously been reported to be elevated in progeroid syndromes [56][57][58][59].However, it is important to note that γH2AX is more specifically a marker of the initiation of the DNA damage repair response [60,61].DNA damage repair can be inhibited in progeroid cells and other studies have reported that the intensity of γH2AX foci was low in HGPS cells [3,62].The relative lack of γH2AX foci we observe in our progeroid cell cultures may therefore reflect the  [63], affected all three progeroid cell types, and was the most affected gene of the four DNA damage repair genes studied.Our data suggests that repair is impacted in the progeroid cells compared with wild-type nHDFs.With trametinib treatment, only RB1 in WS cells and CHEK1 in CS cells were affected.Given that trametinib inhibits MEK which interacts with the pathways that govern these genes, it is likely that these two effects are a direct result of the gene regulatory network rather than an indication of trametinib markedly changing DNA damage repair response.A marker of the execution phase of apoptosis (CASP3) is elevated in all three untreated progeroid cell lines, which may provide evidence for the presence of elevated damage in these cells relative to wild-type controls.
Trametinib as a drug is often used for cancer chemotherapy in combination [64].The reduction in Ki67 is therefore unsurprising when we consider that trametinib is a known anti-neoplastic therapy [45].Trametinib commonly has a variety of side effects including gastrointestinal issues, but this is during a high dose daily treatment programme.Single doses have been enough in a senotherapeutic context to give benefit so it may be that future therapies using trametinib for age-related diseases would use a single dose model.Senescent cells can show a level of heterogeneity and there is no single definitive marker [65,66].Senescence can be induced by several means: replication (via telomere attrition), stress (such as the accumulation of mutations due to poor nuclear stability in HGPS) and/or oncogenes [67][68][69].
However, there exists a lack of clarity over whether senescence biomarkers exist that are associated with specific subtypes of senescence.Several studies suggest that senomorphic compounds may target only SASP-induced (paracrine) senescence (a form of stress-induced senescence) [70,71].It is possible therefore, that treatments such as trametinib may target only subsets of senescent cells, so that conditions where the balance of subtypes is disturbed may show differential effects on rescue.For example, cells in which senescence has arisen because of catastrophic DNA damage may act differently to those which have arisen because of SASP-induced paracrine senescence.In the former there will be an ongoing signal for senescence, whilst in the latter, once the inflammatory milieu has been normalised, senescence may be more reversible.Our results from this study are consistent with this hypothesis.An interesting question would be whether we see similar effects in later passage cells from donors with progeroid syndromes.Such cells are, in effect, prematurely senescent, and share features with wild-type cells at later passages.Whilst the progeroid cells at an early passage may be 'chronologically' young, they appear 'biologically' aged.In the current work, we have carried out our experiments on progeroid cells at relatively early passage, because the cultures become senescent much earlier than wild-type cells.The effects one might observe using later passage cells are difficult to predict, but the increase in transcriptional noise and stochastic variation that occurs during even normal cellular ageing [72], may be amplified in these cells, meaning that consistent effects on gene expression and cellular phenotypes may be harder to detect.Many senomorphic compounds cause biphasic dose responses in cells.Trametinib has been observed to exert different senotherapeutic effects at the 1-10 µM range compared to 20 µM doses in cells [28].A 10 µM dose was chosen for this study based on this research, but it is possible that with a lower dose in the range the cells may show more restoration of splicing factor expression.A 20 µM dose of trametinib may have no effect on senescence, and, as our results suggest, a 10 µM dose may be sufficient to restore the responsiveness of splicing factor expression, but a 1 µM dose may be the dosage that can produce the best response.This may be as a result of a hormetic effect.A hormetic effect is when a cell responds to a minor stressor and overcompensates to the point that the stressor causes a slight benefit to the cell overall [73].Several compounds such as resveratrol and metformin show this type of hormetic effect [74,75].Although trametinib is more specific in the mechanism that it targets compared to resveratrol or metformin, this type of dose effect is still common in compounds that target tightly controlled and highly autoregulated pathways, such as the networks that control splicing factor expression and cell fate.Autoregulation and cross-regulation have been noted in the MEK/ERK pathway that trametinib targets [32].One caveat of this work is that the cells studied were dermal fibroblasts, so other cell types from a person with a progeroid syndrome may have different characteristics and responses to trametinib.Other subtypes of Cockayne syndrome, or other progeroid syndromes, such as Bloom syndrome and Xeroderma Pigmentosum, may not be as rescuable as the syndromes examined here.Further research is needed to uncover if these findings are replicated in other progeroid syndromes.Another caveat is that, although progeroid syndromes are widely considered a reasonable model for 'normal' ageing, given that progerias are syndromic in nature, they may not necessarily reflect what happens in normal ageing when used as a model [76].It is interesting that trametinib is capable of partially rescuing the phenotype of these cells when the phenotypes are so severe.In another study, using a mouse model of HGPS, the senomorphic drug resveratrol was able to alleviate some features of the premature ageing phenotype [31].This gives more evidence in support of the idea that senomorphic and senotherapeutic drugs may help in the progeroid syndromes [39,76].
Our results are consistent with other studies that show that cells from individuals with progeroid syndromes display characteristics consistent with accelerated cellular and molecular ageing, which may be amenable to future therapeutic targeting [76].Several senomorphic drugs, such as trametinib and resveratrol show rescue of senescent cell populations in contradiction to Terzi et al.'s definition of an "irreversible state of cell cycle arrest" [26,28,77].Our body of work suggests that the most senescent and most DNA damaged cells are irreversibly senescent, but senomorphic drugs can reverse senescence in cells up to a certain point, giving more weight to the theory of stages of senescence, e.g. a reversible pre-senescent stage [68,78].
Senotherapeutic and senomorphic drugs represent an intriguing way to think about treating progeroid disease as well as age-related disease.As our population ages, we have an increased burden of age-related disease, which means that more therapies will be needed to sustain a healthy population [79].If compounds are able to address the underpinning mechanisms behind several different age-related diseases, then this could have more impact than attempting to treat each age-related disease individually.Our findings suggest that trametinib and other senotherapeutic compounds could be examined as an additional therapeutic angle for people with progeroid syndromes, and further evidences the notion that mRNA splicing factor dysregulation is a key cellular hallmark of ageing.
4.1 for Windows (GraphPad Software, San Diego, California USA, www.graph pad.com).IBM SPSS Statistics for Windows version 27.0 programme (Released 2020; IBM Corp, Armonk, NY) was used to perform t tests for the RT-qPCR data.Graphs were produced using GraphPad Prism version 9.4.1.Error bars on the graphs represent the SEM unless otherwise stated.

Fig. 2
Fig.2Forest plot of relative gene expression (arbitrary units, AU) of markers of apoptosis (CASP3 and CASP7) and DNA damage repair (ATM, CHEK1, RB1 and TP53) in progeroid cells compared against wild-type nHDFs.The mean ± standard error of the mean (SEM) are graphed.Asterisks denote a significant p value from a t test: * p < 0.05, ** p < 0.01, ***

Fig. 3
Fig.3Forest plot of relative gene expression (arbitrary units, AU) of an a priori panel of splicing factors in progeroid cells compared against wild-type nHDFs.The mean ± standard error of the mean (SEM) are graphed.Asterisks denote a significant p value from a t test: * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0005.n = 3 for all experimental groups.Dark col-

Fig. 4
Fig.4 Senescence characteristics of HGPS cells (blue squares) treated with trametinib 10 µM compared to vehicle controls.The mean ± standard error of the mean (SEM) are graphed.Asterisks denote a significant p value from a t test: * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0005.n = 3 for all experimental groups.For gene expression graphs (D-E), a dark blue colouration denotes significance for a difference in trametinib-treated HGPS cells versus vehicle-treated HGPS cells, while non-significance is shown as light blue coloura-

Fig. 5
Fig. 5 Senescence characteristics of Werner cells (pink hexagons) treated with trametinib 10 µM compared to vehicle controls.The mean ± standard error of the mean (SEM) are graphed.Asterisks denote a significant p value from a t test: * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0005.n = 3 for all experimental groups.For gene expression graphs (D-E), a dark pink colouration denotes significance for a difference in trametinib-treated Werner cells versus vehicle-treated Werner cells, while non-significance is shown as light pink coloura-

Fig. 6
Fig. 6 Senescence characteristics of Cockayne cells (green diamonds) treated with trametinib 10 µM compared to vehicle controls.The mean ± standard error of the mean (SEM) are graphed.Asterisks denote a significant p value from a t test: * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0005.n = 3 for all experimental groups.For gene expression graphs (D-E), a dark green colouration denotes significance for a difference in trametinib-treated Cockayne cells versus vehicle-treated Cockayne cells, while non-significance is shown as light green

Table 1
Wild-type nHDFs are compared against HGPS cells, Werner cells, and Cockayne cells for staining for senescence-associated beta galactosidase (SAB), Ki67 or γH2AXThe mean ± standard error of the mean (SEM) and t test p values are reported.Significant p values > 0.05 are emboldened.

Table 2
Wild-type nHDFs are compared against HGPS cells, Werner cells, and Cockayne cells for RT-qPCR analysis The mean ± standard error of the mean (SEM) and t test p values are reported.Significant p values > 0.05 are emboldened.n = 3 for all experimental groups.All results are logged and normalised to the corresponding natural log of the wild-type nHDFs resulting in a mean of 0.0000 for all nHDFs Vol.: (0123456789)

Table 3
Effects of trametinib treatment on staining for senescence-associated beta galactosidase (SAB), Ki67 or γH2AX in HGPS, Werner and Cockayne syndrome cells Progeroid cell cultures are treated with a DMSO control or 10 µM trametinib.The mean ± standard error of the mean (SEM) and t test p

Table 4
Gene expression in HGPS cells treated with a DMSO control or 10 µM trametinib Vehicle-treated cells are compared against treated cells for RT-qPCR analysis.The mean ± standard error of the mean (SEM) and t test p values are reported.Significant p values > 0.05 are emboldened.n = 3 for all experimental groups.All results are logged and normalised to the DMSO treated control resulting in a mean of 0.0000 for these cells Vol.: (0123456789)

Table 5
Gene expression in Werner cells treated with a DMSO control or 10 µM trametinib Vehicle-treated cells are compared against treated cells for RT-qPCR analysis.The mean ± standard error of the mean (SEM) and t test p values are reported.Significant p values > 0.05 are emboldened.n = 3 for all experimental groups.All results are logged and normalised to the DMSO treated control resulting in a mean of 0.0000 for these cells

Table 6
Gene expression in Cockayne cells treated with a DMSO control or 10 µM trametinib Vehicle-treated cells are compared against treated cells for RT-qPCR analysis.The mean ± standard error of the mean (SEM) and t test p values are reported.Significant p values > 0.05 are emboldened.n = 3 for all experimental groups.All results are logged and normalised to the DMSO treated control resulting in a mean of 0.0000 for these cells